Upload 936 IFPRI-matched files to data/part_2

data/part_2/0023425944.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"7a9314ec8cfccc58e036fb8eb4ea6f90","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/92c12624-9fe1-48e6-8472-bce45e557d67/retrieve","description":"In this chapter, teff production practices are described, along with their relationships with the level of land productivity of teff producers in Ethiopia. Data are used on plots of teff from a survey of 1,200 stratified randomly selected households conducted at the end of 2012. We selected this dataset as it provides high-quality data, relevant information on teff practices, and a large sample. The relationships between farming practices and teff yield are assessed through a descriptive analysis of the data and the estimation of a production function to determine teff productivity. Although input levels and characteristics of the plot and household are included as right-hand variables, the focus is on the technology employed and producer organizations used. The analysis sheds light on the relationships between farming practices and yield across plots as well as the implications for policies to improve land productivity for teff producers.","id":"-1398310017"},"keywords":[],"sieverID":"6e7d99b5-e215-4b7d-a583-6210cc33886b","pagecount":"31","content":"ne of the challenges in Africa south of the Sahara is the struggle to develop sustainable agricultural production systems to cope with increasing population pressure and the associated growth in the demand for food. The Ethiopian government wants to double the production of teff to aid in the alleviation of food insecurity among the general population, which will in the process enhance the well-being of individual farm households that grow, consume, and sell teff. The increase in Ethiopian teff production over the past decade resulted primarily from an increase in the area cultivated (Demeke and Marcantonio 2013). However, since there are limits to the amount of arable land available, the desired increases in teff production will have to be met largely through increases in agricultural productivity.Understanding the relationships between farming practices and yield increases will help in the design of programs to promote the continued and necessary increase in teff yield. For example, if the productivity difference between individual farmers is due to differences in practices employed, then one policy option is to promote the adoption of yield-enhancing technologies (Ali and Byerlee 1991). This promotion could be through modern input methods or even subsidies. This assumes that the farmers are aware of the benefits but do not use the technologies because of credit constraints or being unable to handle the uncertainty of switching practices. However, if the farmers are not using the current technology properly or are not aware of the yieldenhancing practices, there is a role for extension efforts to develop the skills and knowledge of the farmer. Either publicly funded programs or farm organizations could provide the extension education.Information about yield-enhancing technologies, such as seed variety, is generally provided to farmers in developing countries through publicly funded agricultural extension programs (Townsend et al. 2013). Producer organizations operate as an alternative mechanism to provide technical information and may also serve to link farmers with markets for inputs and outputs.For example, agricultural producer cooperatives provide farmers with access to inputs and output markets, whereas producer community groups provide opportunities to discuss best management practices.In this chapter, teff production practices are described, along with their relationships with the level of land productivity of teff producers in Ethiopia. Data are used on plots of teff from a survey of 1,200 stratified randomly selected households conducted at the end of 2012. We selected this dataset as it provides high-quality data, relevant information on teff practices, and a large sample. The relationships between farming practices and teff yield are assessed through a descriptive analysis of the data and the estimation of a production function to determine teff productivity. Although input levels and characteristics of the plot and household are included as right-hand variables, the focus is on the technology employed and producer organizations used. The analysis sheds light on the relationships between farming practices and yield across plots as well as the implications for policies to improve land productivity for teff producers.Two approaches are used to examine the links between farming practices and productivity differences. First, descriptive statistics and nonparametric tests are used as a crude measure for the relationship between farming practices and the cross-sectional variations in the level and distribution of land productivity with a focus on the role of technology adoption and extension activities. Since teff yield data has a skewed distribution according to the results of the Shapiro-Francia W' test (Shapiro and Francia 1972;Royston 1983), nonparametric tests are used-that is, the Wilcoxon-Mann-Whitney and the Kruskal Wallis tests-to determine if differences in yield across factors is statistically significant in the absence of normality. To test for a possible first-degree conditional stochastic dominance in the distribution of yields between two groups (for example, cooperative members versus nonmembers), a Kolmogorov-Smirnov (KS) test is used.Second, the following Cobb-Douglas production function is estimated by ordinary least squares (OLS),Farming PracticTwo approaches are used to examine the links between farming practices and produc First, descriptive statistics and nonparametric tests are used as a crude measure for th between farming practices and the cross-sectional variations in the level and distribut productivity with a focus on the role of technology adoption and extension activities. data has a skewed distribution according to the results of the Shapiro-Francia W' test Francia 1972;Royston 1983), nonparametric tests are used-that is, the Wilcoxon-Ma the Kruskal Wallis tests-to determine if differences in yield across factors is statistica the absence of normality. To test for a possible first-degree conditional stochastic dom distribution of yields between two groups (for example, cooperative members versus Kolmogorov-Smirnov (KS) test is used.Second, the following Cobb-Douglas production function is estimated by ordinary leas \uD835\uDC66\uD835\uDC66 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 = \uD835\uDEFD\uD835\uDEFD 0 + \uD835\uDEFD\uD835\uDEFD \uD835\uDC56\uD835\uDC56 + \uD835\uDEFD\uD835\uDEFD\uD835\uDEFD\uD835\uDEFD \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 + \uD835\uDEFF\uD835\uDEFF\uD835\uDEFF\uD835\uDEFF \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 + \uD835\uDEFE\uD835\uDEFE\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 + \uD835\uDEFC\uD835\uDEFC\uD835\uDC47\uD835\uDC47\uD835\uDC47\uD835\uDC47\uD835\uDC47\uD835\uDC47ℎ \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 + \uD835\uDF0F\uD835\uDF0F\uD835\uDC38\uD835\uDC38\uD835\uDEFD\uD835\uDEFD\uD835\uDC43\uD835\uDC43\uD835\uDC47\uD835\uDC47\uD835\uDC38\uD835\uDC38\uD835\uDC38\uD835\uDC38\uD835\uDC38\uD835\uDC38\uD835\uDC43\uD835\uDC43\uD835\uDC38\uD835\uDC38 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 + \uD835\uDF01\uD835\uDF01\uD835\uDF01\uD835\uDF01\uD835\uDC47\uD835\uDC47 [1] where i indexes household, j indexes village/district/region, and p indexes plot; \uD835\uDC66\uD835\uDC66 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 is total output for plot p from household i located in region j; \uD835\uDEFD\uD835\uDEFD \uD835\uDC56\uD835\uDC56 is a zone or kebele or ho effects that captures possible heterogeneity at the specified spatial level; \uD835\uDEFD\uD835\uDEFD \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 is a vec logarithm of variable inputs, \uD835\uDEFF\uD835\uDEFF \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 is a vector of household characteristics; \uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43\uD835\uDC43 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 is a characteristics; \uD835\uDC47\uD835\uDC47\uD835\uDC47\uD835\uDC47\uD835\uDC47\uD835\uDC47ℎ \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 is a vector of technological practices adopted; \uD835\uDC38\uD835\uDC38\uD835\uDEFD\uD835\uDEFD\uD835\uDC43\uD835\uDC43\uD835\uDC47\uD835\uDC47\uD835\uDC38\uD835\uDC38\uD835\uDC38\uD835\uDC38\uD835\uDC38\uD835\uDC38\uD835\uDC43\uD835\uDC43\uD835\uDC38\uD835\uDC38 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 i where i indexes household, j indexes village/district/region, and p indexes plot; y ijp is the logarithm of total output for plot p from household i located in region j; β i is a zone or kebele or household fixed effect that captures possible heterogeneity at the specified spatial level; x ijp is a vector of quantity of logarithm of variable inputs; H ijp is a vector of household characteristics; Plot ijp is a vector of plot characteristics; Tech ijp is a vector of technological practices adopted; Extension ijp is a vector of extension activities; Weather ijp is a vector of weather shock factors; and ε ijp is an error term that summarizes the effects of unobserved plot quality variation and plot-specific production shocks on yields. β, δ, α, τ, and ζ are parameters to be estimated. Five versions of (1) are estimated: a base model without any fixed effects (β i = 0), plus fixed effects for zone, district (woreda), village (kebele), and households.1The data for the study come from a survey of 1,200 households conducted by the Economic Development Research Institute (EDRI) and International Food Policy Research Institute (IFPRI) (see Chapter 11). The main purpose of the survey was to understand the teff value chains to make recommendations to policy makers to improve the performance of value chains for farmers, wholesalers, retailers, and consumers. The survey was conducted in 2012 in sixty villages (kebele), in twenty districts (woreda), in five major teff-producing zones (regions) of Ethiopia (East Gojjam, West Gojjam, East Shewa, West Shewa, and South West Shewa). The five zones represent 38 percent of national teff area and 42 percent of the commercial surplus (Ethiopia, CSA 2012). The average (and the coefficient of variation of) altitude in meters of the five zones are 2,323 (10 percent) for East Gojjam; 2,200 (5 percent) for West Gojjam; 1,784 (24 percent) for East Shewa; 2,099 (13 percent) for West Shewa; and 2,188 (5 percent) for South West Shewa.The detailed cross-sectional data contain household-and plot-level information on teff production. The number of teff plots per household range from 1 to 6 in the dataset with an average of 1.2. In addition to teff production on each plot, the data include information on six sets of variables identified in equation ( 1): (1) inputs (for example, amount of seed, labor use, number of oxen, DAP and urea fertilizer applied, and amount of herbicide applied); (2) household characteristics (for example, age, gender, and the level of education of the household head, and distance from the nearest marketplace); (3) plot characteristics (for example, plot size, soil color, slope); (4) technological practices (for example, seed variety and color, crop rotation, mobile phone ownership); (5) extension activities (for example, cooperative membership, the number of community meetings attended, the number of extension agent visits, and being a model farmer); and (6) weather shocks (for example, rainfall [timing and extent], frost).2Before examining the relationships between farming practices and productivity in this section, a descriptive overview is provided of (1) the production practices of Ethiopian teff farmers, and (2) teff output and inputs. First, the statistics are presented for several parts of the production process-that is, land preparation, seed use, modern input use and teff cultivation, and harvest and postharvest practices. Then the labor allocations patterns are explored. Second, summary statistics are provided for yield and the variety of explanatory variables used in the analysis to examine yield variations.Most of the farmers grow teff on fragmented lands and as part of a rotation. The average number of parcels managed in the sample used is approximately four, with an average lot size of 0.48 hectares for teff. Farmers do not often plant teff two years in a row on the same plots. For the plots in the survey, in only 34 percent of the cases was teff also planted in the year prior to the survey (Table 9.1).Plowing is one of the important land preparation practices, as the soil needs to be loose to allow the very small teff seeds to germinate. The average plot in the study sample was tilled 4.4 times (Table 9.1). In 95 percent of the cases, plowing was done by animals. In 4 percent of the cases, it was done by hand. Mechanized plowing is almost nonexistent, as it was mentioned for only 0.4 percent of the plots.3 When animals were used, these were usually owned by the farmer himself (84 percent of the cases). Renting-in animals for plowing of teff plots in these zones is rare (4 percent).Nutrient requirements can be provided through either organic fertilizer (that is, manure) or inorganic fertilizer. Manure use on plots for teff production is limited. Farmers stated that 89 percent of the teff plots never received manure, while only 6 percent of the plots received it every year and 8 percent in the year prior to the survey. Only 5 percent of the plots used another organic input instead of manure. These results overall indicate that the use of organic fertilizer in teff production is low.A wide variety of teff seeds are used. The four main color-based grades of teff are magna (very white), nech (white), sergegna (mix between white and red), key (brown/red) (see Chapter 3). In terms of market values, magna ranks first with key the least valuable. White teff seeds were used on 52 percent of the plots (Table 9.2) compared to 17 percent of magna seeds (very white). Sergegna (mix) and red seeds made up 11 percent and 20 percent of the plots, respectively. The high prevalence of white and magna seeds indicate the importance of consumer preferences as more white seeds are disproportionately sold than red or mixed seeds. Red or mixed seeds are often retained for home consumption (Minten et al. 2013).Teff seeds can broadly be classified into traditional and improved varieties. In most of the cases, farmers rely on traditional seeds-as stated by the sample farmers. Of the plots, 30 percent were reported to be planted with improved seeds; 19 percent were planted with the improved Quncho (\"top most\") variety. Officially released in 2006, Quncho was developed by the Debre Zeit Agricultural Research Center (DZARC) from two parent materials: one high-yielding but pale color (Dukem, DZ-01-974) and the other a lower yielding but possessing a popular very white seed color (Magna, DZ-01-196) (Tefera, Belay, and Sorrells 2001). Yield performance on experimental trials with Quncho was found to be significantly higher than national average yields for teff (Assefa, Chanyalew, and Metaferia 2013). For further details, see Chapter 3. This survey was done in the most productive areas of Ethiopia (representing approximately 40 percent of national production), possibly explaining the relatively high adoption of improved seeds in this dataset.Farmers obtained seeds from a variety of different sources including their own harvest, other farmers, cooperatives, private traders, and the Bureau of Agriculture. A full 78 percent of the seeds were obtained from their own harvest, and only 17 percent were purchased (5 percent were received for free or bartered). If the seeds were not from their own harvest, other farmers supplied seed for approximately two-thirds of the cases. Formal outlets for distribution are therefore relatively less important (that is, cooperatives 19 percent, private traders 9 percent, and Bureau of Agriculture 6 percent of the approximately one-fifth of the seeds purchased). The government has in recent years started a program aimed at encouraging farmers to use row planting rather than broadcast seed. According to the Agricultural Transformation Agency (ATA 2013), row planting at a reduced seed rate is believed to have important positive effects on teff productivity as compared to sowing with broadcasting (ATA 2013). However, at the time of the survey, the row-planting practice had not been promoted, so the majority of the farmers practiced broadcasting for sowing. The quantities of seeds used were therefore high-that is, on average 44 kilograms per hectare, as compared to 5 kilograms per hectare seeding rate used with row planting (ATA 2013).In terms of timing, 85 percent of the plots were sown at the normal time (typically July), 7 percent early, and 9 percent late (mostly because of late access to fertilizer [3 percent] or because of flooding or late rains [3 percent]), indicating that the survey year experienced typical growing conditions (Table 9.2).Fertilizers and herbicides were used on a large number of the plots in the survey. The application of inorganic fertilizer is particularly widespread and is the main source of additional nutrients given the low use of organic fertilizer noted earlier. In the year prior to the survey, 89 percent of the plots received DAP and 79 percent received urea. DAP was mostly used once (at the time of planting), while urea was applied once (49 percent), twice (43 percent), or even more (8 percent). Although herbicides are often used (65 percent) to help reduce weeding efforts, manual weeding is, on average, still done once per season. The use of pesticides or other sprays is less prevalent as these were only applied on 16 percent of the teff plots. The finding that inorganic fertilizers and herbicides are often used in teff production, whereas insecticides and fungicides are not commonly used, is consistent with the analysis of Gorfu and Ahmed (n.d.).As with most agricultural products, teff is subject to disease and weather shocks. However, it is believed that teff is more resistant to diseases than a number of other crops, which might partly explain its success in Ethiopia (Assefa, Chanyalew, and Metaferia 2013). During the cultivation period, 84 percent of the plots did not suffer from any insects, animal damage, or pests. If there were such problems, they were mostly related to cutworm, root rot, and ball worm (Table 9.3). Questions were also asked about the type of disease that affected teff on a plot. Farmers reported that 85 percent of the plots had not suffered from any diseases. Leaf rust was the most common Source: authors' calculation using data from the 2012 teff value-chain survey conducted by the Economic development research institute (Edri) and international Food policy research institute (iFpri).disease, with 7 percent of the plots reported to have been affected by this disease. Lodging is an important problem in teff cultivation and significant efforts are made in teff breeding to develop varieties with stronger stems (Assefa, Chanyalew, and Metaferia 2013). Farmers reported that 10 percent of the plots were affected by lodging during the survey year. Finally, the survey year was viewed as a \"normal\" cropping year in terms of the moisture content for 83 percent of the plots, with the other 8 percent of the plots suffering from drought and 9 percent from excess moisture. The lack of growing stresses in the survey year implies that the differences noted for yields in the subsequent analysis is due primarily to differences in production systems rather than the growing conditions unique to the year of the survey. Figure 9.1 illustrates the week in which teff is planted and harvested in the five zones. Almost all of the sowing happens over a period of one month at the beginning of the rainy season (the meher), stretching from the second week of July to the first week of August. However, the harvest period is much more varied and lasts from the beginning of October to the end of January. The average length of the growing season for teff is 17 weeks (Table 9.3). However, significant variation is noted with the variation correlated with elevation (Figure 9.2). Teff that is grown in relatively lower altitudes (<2,100 meters) takes 15 weeks to mature on average as opposed to 18 weeks for teff grown at higher altitudes (>2,100 meters).After teff is harvested, a threshing process is needed to separate teff grains from the straw. The average time between harvesting and threshing is about 40 days during which period the teff plants are left to dry. In almost all cases threshing is done on a dry threshing floor that is \"polished with\" cow dung and using animals (97 percent) (Table 9.3). However, some farmers thresh the teff themselves by using sticks, and a very small proportion of farmers use mechanized threshers (0.2 percent). Almost 60 percent of the farmers create their own threshing floor while 40 percent share a threshing floor with other farmers. The straw from teff is mainly used for livestock fodder, and few farmers sell the straw.Currently little is known about the use of labor in teff production practices, but the survey provides some insights into the amount of labor used for different activities, the source of that labor, and the gender of the workers (Table 9.4). In total, about 141 person-days are spent on the production of 1 hectare of teff. The most labor-intensive activities are weeding (32 person-days), harvesting (28 person-days), and tilling (21 person-days). Postharvest activities (for example, gathering and piling), threshing and winnowing each count for about 22 person-days. Family labor makes up 63 percent of the total labor used, with hired labor accounting for 14 percent, and labor provided through a reciprocal exchange among farming families comprising the remaining 22 percent. The share of hired labor is especially high during the harvesting period, when it makes up 32 percent of total labor use. If labor supply is broken down by gender, male labor dominates teff production. Males perform 82 percent of the all the work during teff production, with tilling and harvesting almost exclusively undertaken by men. Women, however, are relatively more involved in weeding activities and make up 33 percent of all labor used for this activity.Teff Yield and Inputs Table 9.5 provides summary statistics for yield and the variety of explanatory variables used in the analysis to explain yield, with each given by zone. Average yield per hectare is approximately 11 quintals or 1,100 kilograms, which is lower than the 2010/2011 national average yield of 1,260 kilograms per hectare but higher than the 2004/2005 average national yield of 950 kilograms per hectare (Demeke and Marcantonio 2013). The distribution of yields across all plots is illustrated in Figure 9.3 and indicates that some plots had yields above 30 quintals per hectare. Figure 9.3 further illustrates that teff yield is lower than 1,000 kilograms per hectare for 50 percent of the sample plots, and only 10 percent of the plots have yields of 2,000 kilograms per hectare or higher. There are significant regional variations in plot productivity with average yield highest in East Gojjam (14.9 quintals per hectare) and lowest in the West Shewa and South West Shewa zones (8.9 and 8 quintals per hectare respectively) (see Table 9.5).There are considerable differences in input use across the five zones (Table 9.5) and by yield category (Table 9.6). The average input tends to increase with yield category, particularly for yields above 14 quintals per hectare. For example, average labor use is similar at 124 person-days per hectare for the two lowest groupings with yields less than 10 quintals per hectare (that is, the first-and second-yield quartiles) and then increases to over 185 person-days per hectare used on plots with yields above 14 quintals per hectare (that is, the fourth quartile). The effect of labor on yield is similar to the effects from the other variable inputs on teff productivity.Gender and age of the operator are similar across zones, with more than 70 percent of the plots operated by males, 28 percent by both males and females, and 2 percent by females. The average age of the operator is approximately 45 years. Education varies between regions FarmiNg praCTiCES aNd prOduCTiViTy 217 with average years of formal schooling higher in the Shewa zones. For plots with yields less than 6 quintals per hectare (that is, the first-yield quartile), the average formal level of education of the operator is 1.74 years (Table 9.6). Education level increases steadily with yield category. For operators managing plots with yields above 14 quintals per hectare, for example, the education level is 2.41 years.The difference in yield appears to be inversely related to plot size (Table 9.6). For all plots in the sample, the average teff plot size allocated to teff production is 0.48 hectares (1.2 acres) and ranges between 0.03 and 5 hectares (12.35 acres). The average plot size tends to be larger in Shewa than in Gojjam (around 0.6 hectares versus 0.3 hectares), but yields are approximately 50 percent higher in the latter region. In general, small plots tend to have higher average land productivity, suggesting there is an inverse relationship in the average land productivity across plot size categories. This result is consistent with other studies that found an inverse relationship between plot size and productivity (for example, Assunção and Braido 2007;Barrett 1996). While smaller plots may have higher yields, there also appears to be higher variability in land productivity for smaller plots than larger ones (Figure 9.4).The relationship between proportion of soil type (that is, color) and land productivity is statistically significant across zones (Table 9.5) and teff yield quartiles (Table 9.6). For example, plots with black soil are more prevalent in East Gojjam, West Shewa, and South West Shewa, whereas brown soil plots are more prevalent in West Gojjam and East Shewa, and the differences are statistically significant. In addition, the majority of teff plots are on flat field (meda); 88 percent of the plots in the fourth-yield quartile are on flat field, compared with 83 percent for the first quartile.The distance of the plot to alternative site measures is generally inversely related to plot productivity. Plots closer to an all-weather road tend to have higher yields. For example, the distance from all-weather road for plots in the fourth-quartile yield category is approximately 0.78 hours (45 minutes), compared with 1.08 hours (65 minutes) for plots in the first-yield quartile. Similarly, average distance to market and to the cooperatives tends to be inversely associated with plot yield.As noted in the previous section across the sample plots, approximately 20 percent are seeded with Quncho, 10 percent with other improved varieties, and 70 percent with traditional seed varieties. Table 9.6 shows the proportion of seed variety by yield quartile and reveals the presence of considerable differences in the proportion of seed variety by yield quartiles. For example, the proportion of Quncho seed variety is approximately (text continued on page 223)  34 percent for the fourth-yield quartile, compared with 10 percent for the first-yield quartile. The average yield from Quncho for all plots is 35 percent higher than with the traditional variety and the difference is statistically significant at the 1 percent significance level based on Somers' D nonparametric test. It is also found that the relative teff yield of Quncho varies across regions with the increase in yield compared with the yield from traditional seeds ranging from 46 percent for East Shewa to 8 percent for West Shewa.The Wilcoxon-Mann-Whitney nonparametric test suggests that in all zones, except for West Shewa, there is a statistically significant difference in the underlying distributions of teff yield between Quncho and traditional seed varieties. For the pooled sample, the Kolmogorov-Smirnov (KS) test suggests that the distribution of Quncho's productivity dominates the yield distribution of both the traditional and other improved seeds. Assefa, Chanyalew, and Metaferia (2013) identify especially innovative approaches toward technology dissemination as a major reason for the success of Quncho in widespread adoption. The fundamental features of the extension approach involved the following: (1) use of technology as a package (not only was the seed variety extended but other management practices too); (2) use of large farm holders' fields for demonstration purposes;(3) coordinated multistakeholder partnership extension methods (including farmers, farmers' associations, private seed growers, governmental institutions, and NGOs); (4) revolving seed loans; and (5) regular training, follow-up, and supervision. Although these factors have contributed to widespread adoption, the success of Quncho seems also to be linked closely to the prior absence of improved teff cultivars. Quncho was, in some sense, the variety that could reinvigorate teff research. It is possible that adoption rates were so large because the market conditions were very strong and there was such a dearth of previous improved varieties.Teff seed color and seed varieties are correlated. For example, all Quncho seed varieties are reported as very white (magna, 51 percent) or white (nech, 49 percent). Similarly, the majority of other improved seeds (94 percent) are very white (magna, 18 percent) or white (nech, 75 percent). For all varieties, very white and white color seeds tend to have higher productivity.Cooperative members manage at least 60 percent of the plots across all zones with approximately three-quarters of the plots run by cooperative members in East Gojjam and South West Shewa (Table 9.5). While yields for cooperative members are 8 percent less than that generated by nonmembers in East Gojjam, yields are generally higher for members and up to 41 percent higher in West Gojjam (8.4 versus 11.9 quintal per hectare respectively). Cooperative membership is higher at 72 percent for plots in the fourth-yield quartile, compared with plots in the first quartile at 62 percent (Table 9.6). For the pooled sample, Somers' D nonparametric test indicates cooperative members outperform nonmembers by approximately 10 percent, and the Kolmogorov-Smirnov (KS) test suggests that the yield distribution from cooperative members dominates the yield distribution of nonmembers at a 1 percent significance level.Extension efforts are also associated with higher yields as shown in Table 9.6. Farmers' participation in a community meeting to discuss teffrelated issues results in a higher average teff yield. Further descriptive analysis indicates that relative to nonparticipants, teff productivity for participants was higher by approximately 6 percent in East Gojjam, 15 percent in West Gojjam, 17 percent in East Shewa, 7 percent in West Shewa, and 11 percent in South West Shewa. For the pooled sample, the yield on plots for those who attended a community meeting on teff was approximately 13 percent higher than for nonparticipants, indicating that the difference is statistically significant according to the Somers' D test. The Kolmogorov-Smirnov test also suggests that the yield distribution for participants dominates the distribution of productivity for nonparticipants. Average yield is also higher for plots run by farmers who had one or more extension agent visits, especially in West Gojjam. Perhaps related to extension efforts, the average productivity is found to be higher for mobile phone owners in all five zones and the differences are statistically significant for four of the five zones.In summary, this descriptive analysis reveals (1) plots with Quncho seed variety are more productive than traditional and other improved seeds, and the magnitude varies by zone; (2) activities to enhance understanding and market access (for example, cooperative membership and participation in community meetings) increase yield; and (3) proximity to an all-weather road and the nearest agricultural cooperative has a positive effect on land productivity. However, attributing the cross-sectional differences in productivity across plots to a single factor may hide the effect of other observable and unobservable heterogeneities. For example, when examining the effect of the adoption of Quncho on plot productivity, it is important to control for differences in household-and plot-specific observable (for example, education) and unobserved factors (for example, soil quality) that may lead to higher productivity. In the next section, a teff production function is estimated that allows for the control of multiple household-and plot-specific differences simultaneously.The estimated coefficients for five specifications of a Cobb-Douglas production function are listed in Table 9.7. The five estimations of equation ( 1) (described at the beginning of this chapter) represent the base model along with four fixed-effect versions: zone, woreda (district), kebele (village), and household. The vast majority of the estimated coefficients tend to be significant across the estimated models and tend to vary little in absolute terms, suggesting the estimation results are robust. Across all models, seed and labor have a statistically significant positive association with yield. The output elasticity averages approximately 0.2 for seed input across the five models. The amount of urea fertilizer applied and the number of oxen per household also tend to have a statistically significant association with yield. The output elasticity for these two inputs, however, is approximately one-tenth the elasticity for seed and labor. DAP is not shown to be significantly associated with yields. As a large number of farmers use significant amounts of DAP and urea, it is possible that estimates are done on those parts of the domain of the production function where marginal associations are relatively (text continued on page 230)   Source: authors' estimation using data from the 2012 teff value-chain survey conducted by the Economic development research institute (Edri) and international Food policy research institute (iFpri). Note: t statistics in parentheses; * p < 0.10, ** p < 0.05, *** p < 0.01. The number of observations for the household fixed effect is higher than for the other model. This is because of missing values for some of the farmers' (for example, school) and farms' (for example, distance) characteristics.small. The government has recently started to replace DAP with Nitrogenphosphoric fertilizer containing sulphur (NPS), given the high deficiency of sulphur (S) in Ethiopian soils, possibly indicating its lower performance.In addition, operator age and household size are the socioeconomic variables that have a statistically significant association with yield. While the summary statistics indicated little difference in operator age across regions (Table 9.5), younger farm managers produce higher yielding plots. In contrast, the regression analysis suggests operator's education is unrelated to teff productivity, although the summary statistics suggested otherwise. Any increases in household size are associated with a corresponding increase in teff output levels. This may be related to the labor available for production activities, which was shown to increase yield levels.The log of plot size has a positive effect on teff output. However, increases in (logarithmic) plot size are associated with a decrease in teff yield per hectare. The average elasticity value of approximately 0.6 is statistically significant across models. This effect translates to an elasticity of -0.4 of plot size on yield per hectare, suggesting a 10 percent increase in farm size is associated with a reduction of teff yields by 4 percent.4 A number of reasons for the inverse relationship between farm size and land productivity exist in the literature, which include labor market imperfections and opportunity costs of labor (Sen 1962), measurement errors and unobserved land quality, credit market imperfections (Assunção and Ghatak 2003), and land distribution policies and uneven off-farm work opportunities (Benjamin and Brandt 1997).Other plot characteristics related to soil type and typography generally have no link with teff productivity. While the gentle-sloped (dagetama) and hilly (gedel) plots have lower yields than level plots, the results are generally not statistically significant, except for plots on gentle slopes. In contrast, the distance measures between the farm plot and various locations tend to have a mixed relationship with yield. For example, for the kebele fixed-effect regression, a 1-minute increase in the time it takes to a travel to the nearest marketplace is associated with 0.044 percent increases in teff yield on average. Plots owned by households that are farther away from the nearest agricultural cooperative are found to have lower productivity for the kebele fixed-effect model. The result on proximity to a marketplace contradicts the findings that relate to proximity to agricultural cooperatives, although the coefficient of access to market is significant for the kebele fixed-effect model only. The closer a plot is to the nearest marketplace, the lower the rate of productivity of the plot.Proximity to a marketplace may allow farmers to participate in leisure and off-farm income-generating activities, which may provide incentive to spend less time and efforts on the farm. In this situation, distance from the marketplace may create incentives and fewer distractions from off-farm activities, thus allowing a farmer to spend more time on managing teff plots. Also, farm labor supply may be scarce for farms in close proximity to a marketplace as the labor force tends to migrate to urban areas to seek off-farm income opportunities. Furthermore, often proximity to a marketplace may not serve as a measure of remoteness, as \"local markets often exist in the most remote communities, but they operate in isolation from the rest of the world\" (Brooks 2012, 95). The local markets might be distant from urban centers where supplies of goods and services as well as opportunities for social interaction are concentrated.Developments in technology result in the anticipated link with productivity. For example, from descriptive statistics, plots seeded with the Quncho seed variety produce approximately 13 percent more than plots planted with traditional varieties. Improved seed varieties, other than Quncho, yield 10 percent more than traditional seeds. The differences on the productivity benefits from new technology across the fixed-effects models highlight the importance of controlling for observed and unobserved heterogeneities across plots and households when conducting quantitative impact analysis.Other practices considered included the use of a crop rotation. If teff was not grown on the plot in the previous year, the yield of teff is approximately 5 percent greater than on plots growing continuous teff. Owning a mobile phone leads to a 5 percent increase in teff yield for the owner, but it is only significant in one out of the four specifications used (that is, the woreda fixed-effect model). The positive yield impact from the adoption of these two practices is statistically significant and consistent with the results of previous studies such as Asfaw and Shiferaw (2010).While the major technological practices tend to have a significant positive relationship with yield, the role of extension activities and organization on teff productivity depends on the variable considered. For example, cooperative membership generally has an unexpected negative yet statistically insignificant link with yield. In contrast, participation in a community meeting increases teff yield by 4 percent across the fixed effects and is statistically significant at the 10 percent level for zone and woreda regressions. Similarly, one or more visits by an extension agent is associated with an increase in yield by an average of 6 percent to 13 percent, depending on the model. The difference in the results for the alternative forms of extension participation may be explained by the reason that each is organized for. Cooperatives are mainly targeted at making the sector more commercial and at providing access to input and output markets, particularly fertilizer and other improved inputs. Often agricultural cooperatives do not provide technical training to their members. In contrast, the community meetings and extension visits are focused on how the operator can improve the plot's productivity for growing teff. The effect of producer cooperative membership may be reflected in the use of fertilizer and other modern inputs and distance from agricultural cooperatives. In addition, the existence of spillover effects from cooperative members to nonmembers may explain the insignificant effect of cooperative membership on productivity. Other factors that have links with plot productivity are weather shocks, such as water logging, frost, and rainfall. The only weather variables that have a consistently statistically significant relationship with yield are frost and hailstorms. As expected, a greater frequency of these events decreases plot yield.There are significant variations in the productivity of plots growing teff across Ethiopia. This chapter has attempted to illustrate this variation as well as to examine the link between farming practices and these differences. The major associations for productivity differences appear to be the levels of input use, the management practices employed, the age of the operator, the ease of access to markets, and the level of engagement in extension efforts (see Chapter 7). While the links of these factors are consistent with prior expectations, the analysis does not, however, adequately consider the complexity of the system. For example, the adoption of high-yielding seed varieties (Quncho) may be the result of farmers participating in community meetings to discuss teff production and/or visits by extension agents. Conversely, the level of fertilizer use may instead be associated with the distance to travel in order to obtain this input. It is difficult to determine the direct and indirect effects of these factors, and therefore a structural equation model would be appropriate to allow for simultaneous estimation of the input factors and teff output.While further analysis would help direct policy efforts to achieve the desired gains in teff output, the current analysis highlights that there are effective means to enhance the productivity of teff and provide encouragement to those involved with farmers who are reliant on the production and consumption of teff for their livelihoods. The findings lead therefore to a number of policy implications. First, input use, and especially improved seed varieties, is associated with improved productivity, and it appears that there are high rates to return to investments in the development of better seed varieties. The investments to Quncho development clearly illustrate this positive benefit through calculations of the rates of return. Flaherty, Kelemework, and Kelemu (2010) estimate that Ethiopia invested in 2008 US$70 million in agricultural R&D. Assuming, generously, that US$10 million would have been spent annually on teff research, and that the Quncho development was achieved after breeding investments for five years, this leads to a cumulative investment of US$50 million. The results in this chapter show that productivity of Quncho is at least 10 percent higher than other varieties. Assuming that half of the producers in Ethiopia would adopt Quncho and that prices remain unchanged, that would lead to a sustained yearly benefit of US$80 million. If these benefits were to be spread over a 20-year period, the rate of return for that investment would amount to 160 percent. Admittedly, no costs for extension efforts are included in this calculation. However, even if the investments were quadrupled to US$200 million to accommodate this, the rate of return would still be high at 40 percent.Second, the Ethiopian government has invested heavily in the expansion of the agricultural extension system. At the end of 2010, the government had placed 45,000 extension agents in villages, compared with 2,500 and 15,000 extension agents in 1995 and 2002, respectively (Davis et al. 2010). With a target of three extension agents per kebele, Ethiopia has one of the largest extension agent-farmer ratios found in the world today (Davis et al. 2010). It seems that there is a payoff to some of these investments as shown in the higher productivity of teff that is being achieved by those farmers who have been able to participate in community meetings and had access to extension agents.Third, remoteness is found to be an important determinant of productivity because of the lack of incentives due to lower output prices but also possibly because of the costs of getting access to modern inputs at reasonable prices (Minten et al. 2013). Reducing the costs of remoteness through the construction of rural roads, as well as increasing distribution outlets of modern inputs, is likely to have an important positive impact on teff productivity.Fourth, teff is shown to be a labor-intensive crop with higher yields being achieved when farmers spend more time plowing and weeding. Given structural changes in the Ethiopian economy, which leads to increasing real rural wages, innovations that can initiate a reduction in these labor demands, might FarmiNg praCTiCES aNd prOduCTiViTy 233 satisfy the increasing demands for teff at an affordable price. Further efforts to simulate greater use of mechanization for planting, threshing, and weeding might therefore be important for the sustainability of the teff sector.","tokenCount":"6733","images":[],"tables":["-1398310017_1_1.json","-1398310017_2_1.json","-1398310017_3_1.json","-1398310017_4_1.json","-1398310017_5_1.json","-1398310017_6_1.json","-1398310017_7_1.json","-1398310017_8_1.json","-1398310017_9_1.json","-1398310017_10_1.json","-1398310017_11_1.json","-1398310017_12_1.json","-1398310017_13_1.json","-1398310017_14_1.json","-1398310017_15_1.json","-1398310017_16_1.json","-1398310017_17_1.json","-1398310017_18_1.json","-1398310017_19_1.json","-1398310017_20_1.json","-1398310017_21_1.json","-1398310017_22_1.json","-1398310017_23_1.json","-1398310017_24_1.json","-1398310017_25_1.json","-1398310017_26_1.json","-1398310017_27_1.json","-1398310017_28_1.json","-1398310017_29_1.json","-1398310017_30_1.json","-1398310017_31_1.json"]}

data/part_2/0025961772.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"1f75877d8bbcde9bb583ad0e828f85d4","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/5010b6b3-098f-4121-ae37-96761b30b765/retrieve","description":"as part of the IFPRI-Egypt Seminar Series- funded by the United States Agency for International Development (USAID) project called “Evaluating Impact and Building Capacity” (EIBC) that is implemented by IFPRI.","id":"-1989456155"},"keywords":[],"sieverID":"bcd66d45-0844-4c58-ae24-0d621b653da4","pagecount":"16","content":"Deprivations in nutrition and health are a significant contributor to multidimensional poverty of under-five children.Integrated response and policies are needed to fully address these multidimensional poverties. In addition, the main contributors to child poverty (deprivations in health, nutrition, and child protection), need to be addressed through linkages between interventions that promote early childhood development • UNICEF is finalizing with the MOHP the 3 new national counselling manuals on nutrition for health care workers at the PHC level. 1. Nutrition specific counselling skills 2. Nutrition of women in the perinatal period 3. Complementary feeding guidelines• UNICEF support to the national Breast Feeding Program is continuous and aims to accredit all trained facilities as Baby Friendly Facilities• Supporting the MOHP to create a micronutrient manual and system which includes inventory and supply chain management.Based on the nurturing care framework UNICEF has assisted the MOHP to update the \"well baby health cards\" with new messages addressing both growth and development as well as early stimulation positive parenting, family planning and other important messages.High quality Nutrition services at the PHC level  ","tokenCount":"177","images":["-1989456155_1_1.png","-1989456155_1_2.png","-1989456155_2_1.png","-1989456155_2_2.png","-1989456155_3_1.png","-1989456155_3_2.png","-1989456155_3_3.png","-1989456155_3_4.png","-1989456155_3_5.png","-1989456155_3_6.png","-1989456155_3_7.png","-1989456155_4_1.png","-1989456155_4_2.png","-1989456155_4_3.png","-1989456155_4_4.png","-1989456155_5_1.png","-1989456155_6_1.png","-1989456155_6_2.png","-1989456155_6_3.png","-1989456155_6_4.png","-1989456155_7_1.png","-1989456155_7_2.png","-1989456155_7_3.png","-1989456155_7_4.png","-1989456155_8_1.png","-1989456155_8_2.png","-1989456155_8_3.png","-1989456155_8_4.png","-1989456155_9_1.png","-1989456155_9_2.png","-1989456155_9_3.png","-1989456155_10_1.png","-1989456155_10_2.png","-1989456155_16_1.png"],"tables":["-1989456155_1_1.json","-1989456155_2_1.json","-1989456155_3_1.json","-1989456155_4_1.json","-1989456155_5_1.json","-1989456155_6_1.json","-1989456155_7_1.json","-1989456155_8_1.json","-1989456155_9_1.json","-1989456155_10_1.json","-1989456155_11_1.json","-1989456155_12_1.json","-1989456155_13_1.json","-1989456155_14_1.json","-1989456155_15_1.json","-1989456155_16_1.json"]}

data/part_2/0051941506.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"602ee7e229b1066a614ab7ffd1884224","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/0c0eed18-5509-416d-baa9-57686e41678f/retrieve","description":"\"Partnering for Impact: IFPRI-European Research Collaboration for Improved Food and Nutrition Security\" presentation by Luisa Marelli, European Commission – DG Joint Research Centre (JRC) Institute for Energy and Transport, Sustainable Transport Unit, on 25 November 2013 in Brussels, Belgium.","id":"-1759314009"},"keywords":[],"sieverID":"c63464ca-1f41-4847-a6ef-1f2911f5839a","pagecount":"14","content":"The EU27 wheat yield has been reduced in the baseline scenario 2020 from 8 t/ha in the 2011 report to 5.5 t/ha. This now matches the AG-LINK estimates for 2020.• Ethanol from wheat scenario is affected• The resulting ILUC emissions and area are moderately higher than the 2011 values (+15% and +20% respectively)2020 EU wheat yield corrected REASON IFPRI 2011 projections of EU wheat yield in 2020 were much higher than the DG-AGRI and OECD-FAO projections for the same year 7The EU27 wheat yield has been reduced in the baseline scenario 2020 from 8 t/ha in the 2011 report to 5.66 t/ha. This now matches the AG-LINK estimates for 2020.• Ethanol from wheat scenario is affected • Slight effect on the final ILUC emissions for key EU commodities such as sugar beet, cereals and rapeseed.• lLUC emissions increase from 0% to 29% compared to the IFPRI 2011 values• Slight effect on the final ILUC emissions for key EU commodities such as sugar beet, cereals and rapeseed.• lLUC emissions increase from 0% to 29% compared to the IFPRI 2011 valuesFood consumption is maintained constant in household and agrofood sector demandThe crop-price increases cause a reduction in consumption (for food) as well as an increase in crop supply. This means that part of the crop needed for the biofuel comes from reduction in the other competing sectors, principally food consumption, and this comes free of ILUC.• Compared to the 2011 report, several crops, except sugar beet and rapeseed, show an increase in GHG emissionsFood consumption is maintained constant in household and agrofood sector demandThe crop-price increases cause a reduction in consumption (for food) as well as an increase in crop supply. This means that part of the crop needed for the biofuel comes from reduction in the other competing sectors, principally food consumption, and this comes free of ILUC.• Compared to the 2011 report, several crops, except sugar beet and rapeseed, show an increase in GHG emissions ","tokenCount":"325","images":["-1759314009_1_1.png","-1759314009_1_2.png","-1759314009_2_1.png","-1759314009_2_2.png","-1759314009_3_1.png","-1759314009_3_2.png","-1759314009_3_3.png","-1759314009_3_4.png","-1759314009_3_5.png","-1759314009_4_1.png","-1759314009_4_2.png","-1759314009_4_3.png","-1759314009_5_1.png","-1759314009_5_2.png","-1759314009_5_3.png","-1759314009_5_4.png","-1759314009_5_5.png","-1759314009_5_6.png","-1759314009_5_7.png","-1759314009_5_8.png","-1759314009_5_9.png","-1759314009_6_1.png","-1759314009_6_2.png","-1759314009_6_3.png","-1759314009_7_1.png","-1759314009_7_2.png","-1759314009_7_3.png","-1759314009_7_4.png","-1759314009_7_5.png","-1759314009_8_1.png","-1759314009_8_2.png","-1759314009_8_3.png","-1759314009_9_1.png","-1759314009_9_2.png","-1759314009_9_3.png","-1759314009_9_4.png","-1759314009_9_5.png","-1759314009_10_1.png","-1759314009_10_2.png","-1759314009_10_3.png","-1759314009_11_1.png","-1759314009_11_2.png","-1759314009_11_3.png","-1759314009_11_4.png","-1759314009_11_5.png","-1759314009_12_1.png","-1759314009_12_2.png","-1759314009_12_3.png","-1759314009_12_4.png","-1759314009_13_1.png","-1759314009_13_2.png","-1759314009_13_3.png","-1759314009_13_4.png","-1759314009_14_1.png","-1759314009_14_2.png"],"tables":["-1759314009_1_1.json","-1759314009_2_1.json","-1759314009_3_1.json","-1759314009_4_1.json","-1759314009_5_1.json","-1759314009_6_1.json","-1759314009_7_1.json","-1759314009_8_1.json","-1759314009_9_1.json","-1759314009_10_1.json","-1759314009_11_1.json","-1759314009_12_1.json","-1759314009_13_1.json","-1759314009_14_1.json"]}

data/part_2/0065308555.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"d02d2e908dde2e53904e76c5afef25a2","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/1963fee9-11fb-495a-8bc4-4bc7b404d567/retrieve","description":"Safety net programs generally employ thresholds (using a set benchmark of assets, income, or a timeline) to graduate beneficiaries off program support (Browne 2013; Sabates-Wheeler and Devereux 2011). However, many participants in short-term programs fall back to preprogram vulnerability and poverty levels within a short time (Sabates-Wheeler and Devereux 2011), suggesting that threshold-based graduation is an inappropriate way to determine the capability of participants to leave the program. Thus, the concept of “sustainable graduation” has been put forward to better assess the ability of a household to withstand shocks without damaging losses (Sabates-Wheeler and Devereux 2013).","id":"-1738220413"},"keywords":[],"sieverID":"29d0ee6b-6e08-4075-8bac-f311f1cfd4c0","pagecount":"7","content":"Safety net programs generally employ thresholds (using a set benchmark of assets, income, or a timeline) to graduate beneficiaries off program support (Browne 2013;Sabates-Wheeler and Devereux 2011). However, many participants in short-term programs fall back to preprogram vulnerability and poverty levels within a short time (Sabates-Wheeler and Devereux 2011), suggesting that threshold-based graduation is an inappropriate way to determine the capability of participants to leave the program. Thus, the concept of \"sustainable graduation\" has been put forward to better assess the ability of a household to withstand shocks without damaging losses (Sabates-Wheeler and Devereux 2013).However, sustainable graduation, conceptualized as the achievement of long-term improvements in livelihoods and living conditions maintained across generations (Roelen 2015), is not applicable in chronic emergency contexts. Instead, we frame changes from emergency aid to a livelihoods program or to longer-term cash transfers or a mix of these as a transition, depending on each household's capabilities. As safety net investments are often donor-supported for a finite period, it is envisaged that emergency cash transfers can provide a transition to a range of livelihood-support programs and that the development of the Transition Readiness Index will inform this choice, leading to greater household capabilities to withstand recurrent shocks. www.pim.cgiar.org PIM NOTE• Moving from emergency cash transfers to livelihood support programs implies complex transitions for participants; and capturing these transitions requires comprehensive metrics.• This brief discusses a Transition Readiness Index we are developing with the World Food Programme for its safety net program in Somalia.• The index comprises three indicator categories -household agency, system performance, enabling conditions -to measure the likelihood of a successful transition from emergency cash transfers to livelihood programs, longer-term safety nets, or a mix.• The index will support the targeting of accompanying measures to increase transition readiness.Photo: UN Photo/Tobin JonesTransition readiness requires attention and actions to address context constraints, such as markets and infrastructure. Emergency cash transfers can be effective in smoothing consumption and protecting existing assets, but complementary support is needed to increase incomes and assets to sustain livelihoods and build resilience against fluctuations and shocks to the point where participants are ready to transition to the livelihood-building support (Sabates-Wheeler and Devereux 2013) -the X-factor (Devereux and Ulrichs 2015) or big push to \"unlock\" a poverty trap (Banerjee et al. 2015). Therefore, program managers coordinate humanitarian aid with development aid to create an enabling environment that increases livelihood opportunities for beneficiaries.In 2018, World Food Programme (WFP) Somalia and the Danish Refugee Council (DRC) introduced the Urban Safety Net in Mogadishu, offering a monthly cash transferequivalent to US$35 -to approximately 20,000 vulnerable households. These households include the urban poor, internally displaced people, single women, and people with disabilities.A systematic approach to linking basic livelihood protection (consumption smoothing) and entrepreneurial risktaking (production enhancement) in safety net programs is increasingly recognized as a key to achieving protective, preventive, promotive, and transformative objectives of social policy and to enhancing positive transitions.However, transitioning from an emergency to a productive safety net and/or livelihoods program is complex, dynamic, and risk-prone. Urban farming, small enterprises, and employment are potential alternative livelihood options. However, these require households to have sufficient nutritional, economic, and human capability to manage livelihood changes. The concept of graduation underestimates the complexity of people's experiences when changing livelihoods (Sabates-Wheeler and Devereux 2011). Therefore, predicting the right moment for such transitions remains an open methodological challenge. A more holistic framework is needed to prepare and monitor the transitions of cash transfer recipients. Against this background, we initiated work on the Transition Readiness Index.Although multiple conditions must be in place to enhance transition readiness, most safety net programs only consider households' monitoring and evaluation metrics when making transition decisions. Traditionally, emergency aid programs focus only on life-saving consumption (Doocy and Tappis 2017). Diversifying livelihood options is one of the ways that humanitarian aid can support smooth consumption after emergency support ends. The important question is, how can program managers know when a participant is ready to transition from emergency support to livelihood diversification support? Monitoring and evaluation of households does not capture the larger social system of which the household is part.The transition readiness model we propose captures the sources and magnitude of beneficiaries' material and non-material progress from program entry to exit. Consequently, it can enable program managers to determine when to transition respondents from emergency aid to livelihood diversification programs. We build on research by Soares and Orton (2017) and critical debates surrounding the \"graduation approach\". Our collaborative research effort proposes to develop a Transition Readiness Index to open pathways from traditional emergency cash transfers to diversified livelihood support in the Somalian context. The livelihood support will include capability-building instruments such as trauma counseling, vocational guidance, job placement, business incubation, and microfinance, among others. The testing and application of this index will inform the ability of households to self-select options toward transition from emergency cash transfers. Given that emergency cash transfers are provided for a limited time, options for households to diversify livelihoods is particularly relevant for conflict-prone contexts like Somalia. We distinguish three central transition domains -human agency, system performance, and enabling conditions -essential for a successful livelihood transition. Factors under each domain contribute, directly and indirectly, to the transition readiness of a household. Table 1 offers an overview of preliminary indicator domains for assessing transition readiness.We define human agency as the capacity of an individual to make choices and implement them in their respective living contexts. The aspirations of the individual (Glewwe et al. 2015), the ownership of productive assets, and the overall portfolio of livelihood strategies affect the ability of a household to undergo a livelihood transition. The more a targeted safety net program supports individuals in developing their agency, the higher the likelihood of a successful transition from emergency relief to a livelihoods program.Key assumptions underpinning the human agency indicators are:• Authentic aspirations and goals, including self-efficacy, as well as physical and mental health, drive transition readiness.• Asset stabilization and accumulation are prerequisites for transitioning from emergency cash transfers to livelihood support.• In addition to material resources (including food security), productive labor/assets, social resources, and physical and mental health are central to the transition.• The fewer the coping/distress strategies, the higher the likelihood for successful transitions.Individuals and households do not operate in isolation.They are members of communities, and these memberships influence the transition readiness of people and groups. Markets and the functioning of society (e.g., to3 Table 1. Indicator domains for assessing transition readinessHuman agency• Aspiration (goal orientation and trust in oneself)• Health (mental and physical)• Food security and nutrition • Productive assets and income • Livelihood strategiesFive main indicators: The greater the agency of the primary participant within the household, the more successful the transitions.• Household dependency ratio • Markets (food markets, input markets, job markets); personal security • Financial systems operational and accessible • Vulnerability framework (security, stability, risk) within the community context Four indicators: The higher the system performance of the household, markets, financial systems, and the larger community, the more conducive the transition environment.Enabling conditionsThree indicators: The better the service environment people can draw from, the better the transition performance.provide human safety and security) also mediate the transition readiness of a household. In other words, the more favorable the social dynamics within the household, community, and society, the greater the ability of a person to use available agency to make choices toward alternative livelihoods. Measuring the performance of the system from the perspective of the household, therefore, constitutes the second transition readiness domain.Key assumptions underpinning indicators of system performance are:• Communities are heterogeneous in resources, capabilities, and skills, and transfer payments must be developed accordingly for specific target groups.• Social structures, household dependency ratios (workers/family size ratio), gender norms, and communal dynamics influence choices and capabilities.• In areas facing conflict, the conflict sensitivity and awareness of households undermine transition readiness.Enablers and specific support services helping the transition of households constitute the third central transition domain. Often there are time-bound windows of opportunity that beneficiaries can seize to embark on a new livelihood pathway. At the same time, governments and their development partners can provide support services to enhance the transition readiness of beneficiaries. For example, Mexico's Oportunidades program combines cash transfers with nutrition, schooling, health services, training, job search support, youth inclusion, savings instruments, and micro-enterprise development, which increases the probability of households attaining and sustaining higher living standards after transitioning out of the program (Barrientos and Santibáez 2009). Indicators measure respondents' perception and their self-ascribed ability to access services supporting the transition from cash transfers to productive safety nets or other livelihood-support arrangements.Key assumptions underpinning the enabling conditions indicators are:• Respondents' self-assessment of access to support services is reliable. It is not the existence of the services but the ability to access existing services that matters to beneficiaries.• The safety net program provides complementary support services such as asset transfers, livelihoods training, savings facilities, and coaching and mentoring services.Each transition indicator includes thresholds and weights, as do the three domains. The data for thresholds and weights come from the literature and qualitative data derived from Urban Safety Net participants in Mogadishu. We then develop a multidimensional index to predict the transition readiness of participants in the program. This index captures human agency, system performance, and access to critical services from the enabling external environment.All three indicator domains intersect and are crucial for understanding transition readiness. Transition readiness assessments of the participant at the household level offer an aggregated view of transition readiness tested at various points of the emergency cash transfer program life-cycle. Data collection to compute a prototype index took place in December 2021. In future, data for computing the index could be routinely collected as part of the profiling and post-distribution monitoring. The Transition Readiness Index would then provide real-time analytics for the management of the Urban Safety Net and similar programs.The Transition Readiness Index builds on traditional monitoring and evaluation data from cash transfer programs. However, these data only partially cover transition readiness information. We therefore expanded the datasets with several essential metrics. Aspirations-based utility theories suggest that aspirations conceptualized as self-efficacy play a central role in the decision-making of the poor ( Lybbert and Wydick 2018). People who believe in their aspirations, understand emotions, and are sufficiently self-reflective, self-controlled, and execute planned behavior are likely to influence transition outcomes positively. Self-efficacy may relate to managing the transition from a psychological and social point of view, daily livelihood efficacy, and capacity to engage in remunerative activities. In Somalia, managing risks and uncertainties, including unforeseeable political developments resulting in armed conflicts, is equally essential for transitions. We have also included metrics to self-assess mental health and access to capacity development and mentoring services.The transition readiness of a person in a safety net program can be assessed from the perspective of the participant, the safety net service provider, social resources the participant draws from, and key informants overseeing the more extensive political and socioeconomic system. We measure transition readiness from the perspective of the cash transfer participant in target households; but these perceptions may vary over time. It is, therefore, necessary to understand the index as part of a larger and more comprehensive readiness measurement process.Ideally, those carrying out transition readiness assessments measure the ability of the participant to transition within, between, or out of a safety net program, guided by protocols.WFP and the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) are piloting these protocols. To measure the transition readiness of an individual within a household -based on data collected during the program -three assessment stages will be critical: context assessments, readiness analysis (i.e., computing the index), and taking a final readiness decision following the post-distribution monitoring every six months. In this stage, the safety net program computes the transition readiness, drawing on data that safety net programs collect and process as part of the baseline and monitoring exercise for cash transfer recipients. For details on this index, see previous sections of this brief.While the contextual analysis and the quantitative transition readiness measurement provide a good overview of transition readiness within a group of participants, the final decision depends on the personal responses of beneficiaries and expert judgments. Personal conversations with beneficiaries to validate quantitative assessments and the transition readiness index score are all considered.Only through such conversations will it be possible to integrate index-based assessments of clients' capabilities with judgements regarding transition readiness from the perspective of the participants. Although time consuming, such conversations would contribute to the inclusiveness of programs.The Transition Readiness Index, once piloted and refined, will offer several contributions to the Urban Safety Net and similar programs: First, better targeting of beneficiaries. Second, optimizing preparatory capacity development of beneficiaries for diversified livelihood programs. Third, strengthening evidence-based coordination of government presence to improve the external environment for transitions. And fourth, piloting a new design element to enhance evidence-based learning and adaptive management in the Urban Safety Net and similar programs.An evidence-based understanding of transition readiness would provide an additional realistic perspective on the time, age, and income cut-offs that programs now employ to graduate beneficiaries. The transition readiness approach provides a framework and indicators through which safety net implementers can monitor program changes and develop program support measures to improve transition readiness domains, predict transition outcomes, and identify appropriate moments for a transition. Moreover, we integrate the transition readiness measurement into theories of change and impact pathways within and out of emergency responses.As part of the first phase of the Urban Safety Net, we are finalizing the indicator framework with insights from programs in Mogadishu and Puntland. During the second Urban Safety Net phase in Mogadishu, ICRISAT, WFP, and the Benadir Regional Administration in Mogadishu will ground-truth the index, experimentally test measures that enhance transition readiness of vulnerable households, and pilot the index as a new design element for urban safety nets.","tokenCount":"2299","images":["-1738220413_1_1.png","-1738220413_2_2.png","-1738220413_6_2.png"],"tables":["-1738220413_1_1.json","-1738220413_2_1.json","-1738220413_3_1.json","-1738220413_4_1.json","-1738220413_5_1.json","-1738220413_6_1.json","-1738220413_7_1.json"]}

data/part_2/0082534753.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"302d4bfe7caaca5cfc25ba7f65f75d13","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/8382eaee-c673-4ff8-af6e-72be7ff1e934/retrieve","description":"Given the widespread effects that climate change is projected to have on the world’s most vulnerable people—and indeed the effects that are already underway—climate change features strongly in the Sustainable Development Goals (SDGs): specifically, SDG 13 calls for “urgent action to combat climate change and its impacts.” Among the many concerns related to climate change, its potentially serious implications for agriculture and food security have long been recognized.","id":"1878082368"},"keywords":[],"sieverID":"b43bc7af-a97b-41fc-8a17-dcc71d404313","pagecount":"10","content":"1. Climate change affects nutrition by influencing people's food security, disease levels and patterns, water and sanitation environments, and choices about how to allocate time to their livelihoods and to caregiving. In turn, people's nutrition status and diet choices affect their capacity to adapt to climate change and to mitigate climate change by reducing emissions of greenhouse gases.2. For the poorest groups, the seasonal cycles of food availability, infection, and time use remain a significant challenge to nutrition security and provide a stark indicator of the vulnerability of populations to climate risk.3. Different diets drive different production systems and have different emission and resource footprints. On average, meat-rich diets tend to have larger footprints. Dietary choices that are good for health can also be good for the planet.4. Countries are beginning to incorporate climate thinking into national nutrition plans.5. Climate change and nutrition have overlapping agendas. More collaboration between the two communities could generate a common agenda that could be more effectively pursued by both communities-separately and together.G IVEN THE WIDESPREAD EFFECTS THAT CLIMATE CHANGE IS PROJECTED TO HAVE ON THE WORLD'S MOST VULNERABLE PEOPLE-AND INDEED THE EFFECTS THAT are already underway-climate change features strongly in the Sustainable Development Goals (SDGs): specifically, SDG 13 calls for \"urgent action to combat climate change and its impacts.\" Among the many concerns related to climate change, its potentially serious implications for agriculture and food security have long been recognized (see, for example, Bohle et al. 1994).Much less well understood, however, are the links between climate change and nutrition. A clearer picture of how climate change affects nutrition, and how nutrition in turn affects climate change, offers an opportunity to use policies and programs in ways that are mutually beneficial for climate adaptation and nutrition status. This chapter assesses the impact of climate change as a risk for nutrition, but also the potential of nutrition choices to contribute to climate change mitigation. It begins by describing the nature of the linkages between climate change and nutrition, and then it looks at various aspects of these links. For example, many poor people in rural areas are vulnerable to seasonal variations in food supply, disease, and time use, and these seasonal changes point to the potential effects of climate change on nutrition. The chapter also highlights recent evidence on the different greenhouse gas emissions generated by different diet choices. It closes with recommendations on how to intertwine nutrition and climate analysis and actions more closely, both at the national and international levels.Much of the evidence on the links between climate change and health is summarized in a recent Lancet series by Watts et al. (2015). The evidence base on the interactions between climate and nutrition is much smaller and falls largely into two categories. The first, originating in the nutrition community, is conceptual, drawing out the links between climate change and food security (for example, Lake et al. 2012) and between climate change and undernutrition (examples include Tirado et al. 2010aTirado et al. , b, 2013;;Crahay et al. 2010). Much of this literature explores pathways from climate change to nutrition status, and from nutrition status to adaptive capacity. The second category concerns how food production and diet choices affect greenhouse gas emissions (examples include Tilman andClark 2014 andMcMichael et al. 2007). In this chapter we draw on both of these strands: how does climate change affect malnutrition in all its forms, and how do diet choices affect the mitigation of climate change?To answer these questions, we incorporate climate change into UNICEF's conceptual model of undernutrition, in two ways: First, we apply the model to malnutrition in all forms, not only undernutrition. Second, we build in the impacts of climate on the drivers of nutrition through feedback loops from climate adaptation and climate mitigation (Figure 6.1).To begin with, climate change affects the enabling environment for malnutrition reduction. Shifting and sometimes less predictable rainfall and temperature patterns affect political priorities, economic growth, and inequality because the poorest people are most vulnerable to the changes. A less favorable enabling environment for malnutrition reduction makes the underlying determinants of improved nutrition less effective. For example, unexpected and sometimes more severe weather changes disrupt the intermediate environments that are so important for good nutrition.In the food environment, climate influences people's food consumption by influencing local and global food availability (production, storage), quality (nutritional value and food safety), access (market policies and prices), and how the body utilizes food. Seasonal food scarcity (see Panel 6.1) and climate shocks (such as droughts) have long been shown to drive short-term malnutrition, morbidity, and, in Africa, mortality in vulnerable populations, especially women and girls.Global climate models suggest that by 2050 climate change will result in additional price increases of 5-25 percent for the most important agricultural crops-rice, wheat, maize, and soybeans-and that higher feed prices will result in higher meat prices (Nelson et al. 2009). This is because overall warming temperatures are expected to have a negative effect on global crop production, although this effect might be counteracted in part by the effects of CO 2 (Lobell et al. 2012). Without real efforts to adapt, people's production capacity and livelihoods are under serious threat. According to the Intergovernmental Panel on Climate Change (IPCC), climate change without adaptation will depress production of wheat, rice, and maize even under local temperature increases of 2° C. (high confidence).It is also estimated that food quality will be affected. For example, elevated CO 2 emissions representing likely levels in 2050 are associated with substantial declines in the zinc, iron, and protein content of wheat, rice, field peas, and soybeans (Myers et al. 2014). In addition, food safety may be compromised by a changing climate. High temperatures and extreme weather events create a more favorable environment for food-borne pathogens such as campylobacter and salmonella (Tirado et al. 2010a), which reduce sufferers' ability to absorb nutrients.In the health environment, climate plays an important role in the transmission of many human parasitic, viral, and bacterial diseases (such as malaria, dengue, and cholera, respectively). Rainfall and temperature determine the spatial and seasonal distributions of these diseases, influence year-to-year variability, including epidemics, and affect long-term trends (Kelly-Hope and Thomson 2008). Observed warming in the East African highlands, which is clearly associated with global climate processes, may already be changing local malaria transmission dynamics (Omumbo et al. 2011). The range of livestock and plant diseases is also expected to shift in association with changes in climate patterns. Climate shocks such as cyclones and floods can directly affect the delivery of health care services and people's access to such services by damaging care facilities and transport infrastructure.In the work/social environment, economic models suggest that climate change will significantly alter how people engaged in climate-exposed livelihoods such as agriculture allocate their time (Zivin and Neidell 2014;Ulrich et al. 2015). Moreover, we know that agricultural cropping patterns affect the time available for child care, which in turn can affect the nutrition status of children under 5 years of age (Paolisso et al. 2002).With regard to the living environment, there is compelling evidence that climate change is resulting in longterm drying in some regions, including parts of the fertile crescent of North Africa, with major social and political ramifications (Kelleya et al. 2015). Drying threatens the quantity and quality of water available for irrigation (food production), energy production (food processing), and human consumption (washing, cooking, and drinking). Figure 6.2 shows the main agricultural water systems on which climate change is expected to have an impact (Turral et al. 2011). We should also expect additional challenges to water availability for human use and consumption, including increasing salinity of coastal drinking water sources, which affects maternal health.How do all these pathways add up to change nutrition status? These changes lead to disrupted health behaviors and biological status and to disease, diminished productivity, and mortality. This situation undermines the ability   Seasonality is an important factor mediating climate change and nutrition status. In general, people's vulnerability to seasonality provides a good indicator of their extreme vulnerability to climate risk. This is because climate patterns play a fundamental role in shaping natural ecosystems, including the seasons. But seasonality is also a determinant of undernutrition especially among the poorest rural communities, where seasonal patterns of food consumption, micronutrient availability (Jiang et al. 2005), infectious disease (Kelly-Hope and Thomson 2008), and human behavior (Devereux et al. 2011) are most strongly observed. This vulnerability is particularly acute in regions where the rains are highly seasonal and rain rather than irrigation is the main source of water for agriculture.Here the period between planting and harvesting is widely known as the \"hungry season.\" Panel 6.1 summarizes more findings on how seasonality affects nutrition determinants and outcomes and draws out some of the programmatic implications.Seasonality can have substantial effects on people's nutrition status. The height of young children in India, for example, varies significantly by the month of their birth (Figure 6.3). Compared with children born in December, those born in the summer and monsoon months (April-September) have significantly lower height for their age.In Bangladesh, extreme flooding, driven in part by changes in global sea surface temperatures (such as El Niño), affects rice production, rice prices, and child nutrition. Aman rice production increases with the size of the annual flood until an optimum point, after which excessive flooding damages the crop (Figure 6.4). Extreme flooding also raises rice prices, leading to significant increases in underweight among children under age 5 (Figure 6.5 and Panel 6.2).Given the influence of children's month of birth on their nutrition outcomes, a greater focus on seasonality in nutrition assessment, programming, and policy would be warranted-even in the absence of climate change. The uncertainties introduced by climate change make such a focus even more important. Unless nutrition status and programming are more effectively season-proofed, they will not become more effectively climate-proofed.Foods from animals (\"animal-source foods\") such as meat, fish, poultry, milk, and eggs provide protein and a variety of essential micronutrients (such as iron, zinc, vitamin A, PANEL 6.1 TIME TO TAKE SEASONALITY MORE SERIOUSLY EMILY BIELECKI AND JERE HAAS D espite the progress in global devel- opment over the past 30 years, the nutritional status of mothers, infants, and young children still varies widely by season.There is some evidence on the effects of seasonality on nutrition. For adult women, seasonal swings of 0.7 to 3.8 kilograms of average body weight, and associated increases in the share of women with a lower body mass index during the rainy season, have been reported in both Africa and Asia (Ferro-Luzzi and Branca 1993). During the rainy season relative to the dry, postharvest period, women of reproductive age, including pregnant and lactating women, have reduced intakes of both macro-and micronutrients, increased morbidity, and increased agricultural labor demands (Prentice and Cole 1994). In the Gambia, the share of newborns who are small for their gestational age reached its peak at the end of the hunger season (30.6 percent) and progressively fell to 12.9 percent through the harvest period. The period with the highest rates of infants born small for their gestational age also coincided with the highest percentage of pregnant women with malaria, a known risk factor for low birth weight (Rayco-Solon et al. 2005). Studies from Africa and Asia have reported differences of more than 100 grams in average birth weights and approximately 1 centimeter in birth lengths in favor of the dry, postharvest season; these effects are greater than those reported in many maternal nutrition programs (Chodick et al. 2009;Rao et al. 2009).Despite the substantial nutritional impacts of seasonality, policymakers and program implementers may be relatively unaware of these effects owing to inadequate data. Urban elites that formulate much development policy may also be desensitized to seasonality, having the means to insulate themselves from it in their own lives (see the foreword by Robert Chambers in Devereux et al. 2011). The current focus on increasing the availability of repeated nationally representative surveys might also inadvertently hide seasonal disparities. Whatever the reasons, the absence of focus looks misguided. Furthermore, climate change is likely to make seasonal patterns more unpredictable and possibly more accentuated, making it even more important that we are alert to programmatic responses that can reduce any such variations.The persistent impact of seasonability on nutrition has important implications for policies and programs. In particular, it requires the following:• A greater emphasis on nutritional surveillance. New technologies have created opportunities to understand seasonality in a more meaningful way (see Panel 9.6 in Chapter 9), and they may allow researchers to gather and analyze spatial and temporal data more cost-effectively.• A greater focus on the nutritional status of infants during the first 1,000 days of life and adolescent girls. Adolescent girls are potential conduits of intergenerational nutritional status. Well-nourished girls will be in a better position to withstand seasonal shocks when they eventually become pregnant, so a life-course approach to interventions is vital.• A higher priority given to interventions that help households manage seasonal variations in consumption, income, and illness. Social protection programs are designed to protect households from undermining human capital in the face of shocks and are thus a sensible mechanism for delivering additional nutrition-specific components.In general, all nutrition policies and programs must carefully consider the potential trade-offs between maternal nutritional status, maternal work, and time available for adequate childcare-tradeoffs that will be exacerbated by seasonality and, possibly, by poorly thought-through attempts to address it. Note: Data come from three rounds of the National Family Health Survey and are for households with two or more children under age 3. Winter = December-March; summer = April-June; monsoon = July-September; autumn = October-November.Rice production and the extent of flooding in Bangladesh riboflavin, and vitamins B-6 and B-12) that, for some age groups in some environments, are difficult to obtain in adequate quantities from plant-source foods alone (Allen 2014;Dewey and Adu-Afarwal 2008;Murphy and Allen 2003). Small quantities of meat thus contribute to nutritious diets, especially for infants at risk of undernutrition (Dror and Allen 2011). Raising livestock is also an important part of the livelihood of many smallholder farmers; it can provide them with income for investing in education and healthcare and in this way can raise nutrition status indirectly (Smith et al. 2014).During the nutrition transition (Popkin 2011), however, diets around the world have evolved to include large quantities of meat. Between 1961 and 2009, the global supply of animal-source food available rose from 118 to 164 kilograms per person. Meat-mainly poultry, pork, and beef-was responsible for most of the increase, and meat consumption soared by 82 percent, from 23 to 42 kilograms per person per year, in that period (Keats and Wiggins 2014). Diets rich in red meats such as beef and lamb have been established as a risk factor for nutritionrelated noncommunicable diseases (Woodcock et al. 2007;WCRF 2007;Pan et al. 2012). This shift to diets very high in meat is also associated with high environmental costs. Diets higher in meat are associated with higher greenhouse gas emissions. Figure 6.6 shows that, compared with the global average diet for 2009, diets lower in meat generate lower emissions.Although livestock of all types contributes directly and indirectly to climate-changing greenhouse gas emissions, the impacts are greatest from ruminants, which include cows and sheep. Dung and urine deposits from animals emit methane and nitrous oxide, and ruminants generate significant additional methane emissions through enteric fermentation. The production of animal feed also leads to nitrous oxide emissions, and deforestation and other forms of land clearance for pasture or cultivation of animal feed lead to CO 2 release. Per unit of edible animal product, industrialized systems tend to generate fewer greenhouse gas emissions than extensive systems, but they give rise to other environmental and societal concerns including higher water use, higher point-source pollution, greater use of antibiotics (with associated antibiotic resistance concerns), and potentially greater associations with epidemic zoonotic disease outbreaks (Garnett 2011).Reducing livestock production of all kinds, and particularly reducing ruminant production, could help significantly reduce greenhouse gas emissions, but such a step would need to be accompanied by actions to reduce consumerR ice is central to the food security and nutrition of more than half of the world's population. As a \"strategic\" commodity in many Asian countries, it is subject to a wide range of government controls and interventions. In addition, rice production is highly sensitive to climate. Higher-than-optimal minimum and maximum temperatures have been shown to push down rice yields in both the laboratory and the field, making this crop highly vulnerable to the increased temperatures predicted to occur as a result of climate change (Welch et al. 2010). Rice production is also sensitive to drought and to extreme flooding, as data from Bangladesh show (Figure 6.4).Bangladesh is a rice-growing country, and more than 70 percent of the calories consumed by rural Bangladeshis come from rice (Torlesse et al. 2003). While seasonal river flooding is essential to the rice farming system, major floods cause substantial losses. When the rice crop fails because of excessive flooding in the Aman season or regional drought in Boro season, Bangladesh responds by importing from neighboring countries and increasing production in the following season. However, such transitions are not smooth, they interact with regional and global shocks, and they can lead to rapid increases in rice prices that may affect consumers (Golam Rabbani Mondal et al. 2010). In turn, rice prices have a direct impact on child nutrition. High rice prices following production shocks have been shown to be strongly associated with a decline in spending on nonrice foods (which tend to contain higher densities of micronutrients) and an associated increase in underweight children (Torlesse et al. 2003). Figure 6.5 shows a strong association between price of rice at the local level in Bangladesh and underweight in children under 5 years old after long-term trends have been removed.For Bangladesh the very real intersection of climate and nutrition is the subject of much analysis and action. One useful step would be to develop a better understanding of how global sea surface temperatures relate to local river flooding. This would allow Bangladesh to develop early warning systems to alert the health community to potential nutritional challenges before they occur, as has been done for malaria (Thomson et al. 2006). Such systems are best embedded in a comprehensive effort to (1) invest in interventions to bring down current disease burdens;(2) promote a comprehensive approach to climate risk management; and (3) support applied global and regional research agendas as well as targeted research on high-priority diseases and population groups (Campbell-Lendrum et al. 2015).demand. People with very low levels of income increase their demand for animal-source protein and fats when their income increases even marginally (Kearney 2010). Efforts to control production are thus unlikely to be effective unless the regulatory, fiscal, contextual, and sociocultural determinants of demand are also addressed.An important step in shaping consumption habits is for national dietary guidelines to recommend lower red meat consumption among high-consuming groups. The Health Council of the Netherlands and Sweden's National Food Agency, for example, are taking a lead in this respect (HCN 2011;Sweden, National Food Agency 2015). The Brazilian dietary guidelines also include some discussion of environmental issues and recommend moderating meat consumption to achieve both environmental and health benefits (Brazil, Ministry of Health 2014). For the development of the 2015 United States Dietary Guidelines, an advisory committee produced a report that makes recommendations for diets that are not only healthful, but also generate fewer environmental impacts; included in this is the recommendation to eat fewer animal products (US Office of Disease Prevention and Health Promotion 2015).What is the optimal amount of meat to produce and consume to maximize nutrition status while minimizing greenhouse gas emissions? Clearly there is no one-size-fitsall position; the answer will be country-and group-specific.But whenever nutrition and climate groups can come together on messaging with regard to meat consumption, the potential for moving their common agenda forward increases significantly. Nutrition and climate groups should develop country typologies stratified, for example, on current meat consumption levels, dependence on ruminants for livelihoods, and the intensity of greenhouse gas emissions of livestock systems to begin exploring the potential for these common agendas for action to be developed and pursued.To what extent do countries' nutrition plans include climate variability and change? At the time of this writing (March 2015), of the 26 member countries of the Scaling Up Nutrition (SUN) Movement with complete national nutrition strategy documents available in English, 10 explicitly mention \"climate\" or \"climate change.\" In the strategies climate change is linked with six directly affected categories: agriculture, food security, malnutrition, natural hazards (such as floods and droughts), seasonal variability, and adaptation/mitigation (Figure 6.7). An additional four countries allude to climate through references to its components (such as rainfall, temperature, and season-   ality). Twelve countries do neither. In addition, a series of eight country case studies on the nutrition sensitivity of agriculture and food policies, commissioned by the UN Standing Committee on Nutrition, found that five of the eight surveyed countries referred to climate change in some of their food, agriculture, or nutrition policies (Fanzo et al. 2013). This inclusion rate is encouraging, and it would be worthwhile for other countries-both SUN members and nonmembers-to include climate adaptation and mitigation factors in their nutrition plans. ","tokenCount":"3535","images":["1878082368_1_3.png"],"tables":["1878082368_1_1.json","1878082368_2_1.json","1878082368_3_1.json","1878082368_4_1.json","1878082368_5_1.json","1878082368_6_1.json","1878082368_7_1.json","1878082368_8_1.json","1878082368_9_1.json","1878082368_10_1.json"]}

data/part_2/0092361404.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"490fab8086a48b83ccf776215490140c","source":"gardian_index","url":"https://www.canr.msu.edu/prci/publications/Sugarcane%20production%20food%20security%20Uganda_PRCI.pdf","description":"This study investigates the relationship between farm household participation in sugarcane production and food security in the main sugarcane-producing sub-regions of Busoga, Buganda, and Bunyoro of Uganda. Analysis is based on primary data collected from 1,771 households in these regions as well as qualitative focus group discussions with cane growers. Descriptive analysis found that three different measures of food security -- Household Food Insecurity Access Score (HFIAS); Months of Adequate Household Food Provisioning (MAHFP); and Household Dietary Diversity Score (HDDS); revealed higher average values of HFIAS, MAHFP and HDDS for cane growing households compared to non-cane growers. Poisson regressions of the three food security measures (and an Ordered Probit Model of Food Insecurity Score, derived from HFIAS) found that farm households engaging in cane production in 2021 had lower levels of food insecurity and a higher number of months of food adequacy, on average, relative to households not producing cane, while controlling for other factors known to influence household food security. Maintaining the positive association between cane production and food security requires a policy environment and public sector governance to promote improved coordination between growers and millers and consideration of related income stabilisation mechanisms for sugarcane farmers. In addition, extension services should promote sugarcane production on farms with 8 or more acres only, farmer adherence to maintaining food crop cultivation, and use of productivity-enhancing technologies in cane and food crop production.","id":"1883826526"},"keywords":["Crop productivity","market participation","agricultural technology","smallholder farmers JEL Classifications: Q00","Q10","Q13","Q12"],"sieverID":"2badd30c-1731-4e26-8d3b-652eaf628e75","pagecount":"70","content":"The MSU consortium works with governments, researchers, and private sector stakeholders in Feed the Future focus countries in Africa and Asia to co-create a global program of research and institutional capacity development that will enhance the ability of local policy research organizations to conduct high-quality food security policy research and to influence food security policy more effectively while becoming increasingly self-reliant.The papers are aimed at researchers, policy makers, donor agencies, educators, and international development practitioners. Selected papers will be translated into other languages. Copies of all PRCI Research Papers and Policy Briefs are freely downloadable in pdf format from this link. Copies of all PRCI papers and briefs are also submitted to the USAID Development Experience Clearing House (DEC) at this link and to AgEcon Search at this link.The government of Uganda has supported policies over the past 20 years intended to facilitate the growth of its sugarcane industry (Mbowa et al, 2023) given that sugarcane is high-value crop that can improve growers' farm income and thus improve their food security. The sugarcane industry of Uganda has subsequently grown rapidly the past 20 years, as sugarcane production increased from 1.5 million MT in 2000 to 5.8 million MT 2020 (Mbowa et al, 2023), driven primarily by a four-fold expansion in cane area during that period (FAOSTAT, 2021). However, recent evidence from the three main sugarcane-growing areas of Uganda indicates that food insecurity and income poverty there have increased in recent years, relative to other regions of the country (UBOS 2021). Citing this evidence and anecdotal information, some have claimed that sugarcane cultivation is contributing to or driving the recent increase in food insecurity in cane growing areas (Mwavu et al., 2018). Yet, within the context of Uganda, there is no empirical evidence (to our knowledge) that uses large-scale household survey data to compare the food security status of cane producers and non-producers in the main cane growing regions of Uganda. Neither has there been research to assess whether the recent increase in food insecurity in these areas could be plausibly attributed to cane production or not.This study investigates the relationship between farm household participation in sugarcane production and household food security in the main sugarcane-producing sub-regions of Busoga, Buganda, and Bunyoro of Uganda. This paper addresses three main research questions. First, is participation in cane production in Uganda associated with better food security outcomes? Second, do differences in institutional arrangements between cane growers and millers influence the relationship between participation in cane production and household food security? Third, does women's influence in intra-household decision-making regarding crop choice, crop market participation (i.e. retention and/or sale of harvested crops), or the allocation of crop sales income influence household food security? Descriptive and econometric analysis is used to investigate the relationship between sugarcane production and household food securing, using primary data collected from 1,771 households in these regions in November/December 2021 as well as qualitative focus group discussions with cane growers.Using three different measures of food security --Household Food Insecurity Access Score (HFIAS); Months of Adequate Household Food Provisioning (MAHFP), and the Household Dietary Diversity Score (HDDS), descriptive analysis shows that cane growing households had higher average values of each measure as compared with non-cane growers. Econometric analysis finds that sugarcane growing households were 17 percent less food insecure, on average, relative to non-cane growers, as measured by the 27-point Household Food Insecurity Access Scale (HFIAS)while controlling for a variety of household and community-level factors known to influence household food security. Cane growing was also associated with one additional month of adequate household food provision (MAHFP), an improvement of 10 percent compared with non-cane growers. No significant association was found between cane production and HDDS, a measure of vii household food security. This analysis also found that households in Buganda subregion had better food security measures compared to those in Busoga and Bunyoro subregions, with Bunyoro faring the worst. The severity of food insecurity using the HFIAS was high among non-cane growers, though it declines in total household asset value and ownership of large animals. Households with less than two acres and those with less than 4 acres in non-cane and cane growing categories were severely food insecure, on average, for each of the three food security measures.Other factors positively associated with food security outcomes included a household growing more than one food crop, one or more household members with salaried employment, and higher levels of maximum adult female education in the household, household assets, and number of live animals. Factors negatively associated with food security status include household size (as measured by household Adult Equivalents), residence in Busoga and Bunyoro subregions relative to Buganda, having a member in wage employment, and female-headed household status. Results of testing whether a household female head/spouse has strong influence in intra-household decision making on crop choice or crop marketing show that is no evidence to support the expectation that a household with a female head or spouse with strong influence on crop choice or crop marketing.Policy reform and active public sector oversight of the sugarcane industry is within the mandates of the Ministry of Agriculture and the Ministry of Industry, Trade, and Commerce, and they must coordinate with each other and sector stakeholders to improve coordination between growers and millers. Following a difficult period of poor grower-miller coordination from 2018 to 2021 and a crash in 2021, regaining and maintaining the benefits of cane production for food security requires reforms to the industry's policy environment and public sector governance (Mbowa et al, 2023). Reforms are needed to promote improved coordination between growers and millers and consideration of related income stabilisation mechanisms for sugarcane farmers. In addition, extension services should promote sugarcane production on farms with 8 or more acres only, farmer adherence to maintaining food crop cultivation, and use of productivity-enhancing technologies in cane and food crop production.There are several ways in which Ugandan policy makers can help to maintain the income and food security benefits of sugarcane growing in the country. First, Ministry of Trade, Industry and Cooperatives (MTIC) and the Ministry of Agriculture, Animal Industry and Fisheries (MAAIF) should provide a policy environment and public sector oversight of institutional arrangements between millers and growers that promote higher cane productivity and profitability for growers and more reliable market assurance for both growers and millers. Such arrangements would facilitate stronger coordination of local cane supply and demand, grower access to improved inputs, market assurance for both growers and millers, and a transparent and fair process for millers and grower association representatives to negotiate a cane purchase price each season, based upon a set formula (Mbowa et al, 2023). This level of coordination and oversight can only be achieved if MTIC and  7. Share of households by HFIAS scale value, cane growing status, and region (%) ................. Table 8. Household food insecurity access scale by quintiles of total household asset value per AE Table 9. Poisson regressions of Household Food Insecurity Access Score (HFIAS), Household Dietary Diversity Score (HDDS), and Months of Adequate Household Food Provisions (MAHFP) .  15. Poisson model estimates of household food security and the gender and household position of household member making final decisions on allocation of crop sales income ................   The government of Uganda has supported policies over the past 20 years intended to facilitate the growth of its sugarcane industry (Mbowa et al, 2023) given that sugarcane is high-value crop that can improve growers' farm income and thus improve their food security. Participation in cane outgrower schemes can also improve grower access to yield-enhancing crop production technologies and management practices (Hess et al., 2016), which can improve farm productivity beyond cane and facilitate rural economic development (Burnod et al., 2015;Hall, 2017;Zaehringer, 2018 (a &b). The GoU also sees sugarcane as a key agroindustry that can help generate rural farm on non-farm employment, higher farm incomes and improve rural infrastructure (Fitawek and Hendriks, 2021).There is also evidence that crop diversification can enhance food security and dietary diversity, particularly when the level of production diversity in an area is low to begin with (Appiah-Twumasi and Asale 2022; Mengistu et al., 2021;Douyon et al., 2021;Sibhatu and Qaim, 2018;Sibhatu et al, 2015;Pellegrini and Tascotti, 2014;Njeru 2013).The sugarcane industry of Uganda has grown significantly in the past 20 years, as sugarcane production increased from 1.5 million MT in 2000 to 5.8 million MT 2020 (Mbowa et al, 2023). This expansion was driven almost entirely by an increase in area cultivated to cane from 20,000 ha in 2000 to over 81,000 ha in 2020 (FAOSTAT, 2021). However, recent evidence from the three main sugarcane-growing areas of Uganda indicates that food insecurity and income poverty there have increased in recent years, relative to other regions of the country (UBOS 2021). Citing this evidence and anecdotal information, some have claimed that sugarcane cultivation is contributing to or driving the recent increase in food insecurity in cane growing areas (Mwavu et al., 2018). Related concerns have been raised about sugarcane in other countries, noting that large scale sugarcane farmers and milling companies desiring to expand nucleus farms may acquire large pieces of land from communities and lead to adverse spillovers effects on local communities' food security (Aabø and Kring,2012;Fitawek and Hendriks, 2021;Lisk, 2013;Herrmann, 2017;Nolte and Ostermeier, 2017). In addition, Adams et al. (2019) note that contract farming in sugarcane production can exacerbate gender inequality in terms of access to land, intra-household labour allocation and participation in production and marketing decisions. Resulting intra-household decisions may directly or indirectly affect household food security and nutrition given women's important roles in agricultural production and domestic decisions around food sourcing and preparation. However, within the context of Uganda, there is no empirical evidence (to our knowledge) that uses large-scale household survey data to compare the food security status of cane producers and non-producers in the main cane growing regions of Uganda. Neither has there been research to assess whether the recent increase in food insecurity in these areas could be plausibly attributed to cane production or not.Available studies from East Africa on the relationship between commercial crop farming and food security in Kenya (Kirimi et al., u.d), Tanzania (Dancer and Sulle, 2015) and Uganda (Waibi, 2019;Mwavu et al. 2018) have used mainly descriptive statistics and single measures of household food security (Kipkorir, 2023;Kirimi et al. u.d), the findings from them on this relationship are mixed. In addition, studies that have investigated the food security status of sugarcane growing households in Uganda have not compared them with non-cane growers within the same communities. This paper aims to address this local evidence gap and the inconclusive research findings from East Africa fill this evidence gap for Uganda by investigating the relationship between cane production and household food security within Uganda's three main cane growing regions. It also contributes to the decades-long debate on the relationship between cash crop production and household food security in a developing country context, and to a much smaller one on the relationship between sugarcane production and household food security.This paper addresses three main research questions. First, is participation in cane production in Uganda associated with better food security outcomes? Second, do differences in institutional arrangements between cane growers and millers influence the relationship between participation in cane production and household food security? Third, does women's influence in intra-household decision-making regarding crop choice, crop market participation (i.e. retention and/or sale of harvested crops), or the allocation of crop sales income influence household food security? The paper posits the following hypotheses for these research questions:(i) Sugarcane growing households have better food security outcomes than non-sugarcane growing households. (ii) Cane growers that are registered and aided have better food security outcomes than cane growers who are not. (iii) Households where the female head/spouse has significant influence on intra-household decisionmaking regarding crop choice, crop market participation, and the allocation of crop sales income have better food security outcome, all else constant.This study is based on primary data collected from 1,771 cane growing and non-growing households in these regions in November/December 2021 as well as qualitative focus group discussions with cane growers and key informant interviews with large and small cane mills, cane grower associations, and relevant government officials. The study addresses the three research questions and tests the three hypotheses through descriptive and econometric analysis of this household survey data.The rest of this paper is organized as follows. Section 2 provides a literature review on sugarcane and food security. Section 3 then describes the methods and data used to address the research questions.Section 4 provides research results and discussion, followed by conclusions and policy implications in Section 5.This section provides review of literature on the relationship between sugarcane production and household food security, as this is needed to guide the analytical strategy for the paper, including the specification of multivariate regressions of different measures of household food security. Household food security is a multidimensional concept, it depends on multiple factors such as the stability of food supply, household food production, access to food through income, prevailing food prices and the availability of food in markets.There has been a long-standing debate and literature over the past thirty years addressing the quesiton of whether smallholder participation in contract farming (CF) arrangements results in positive or negative changes in household welfare in practice. An extensive review of literature by Otsuka, Nakano & Takahasi (2016) found that in most cases, CF improved farmers' income by introducing them to higher-return crops and yield-improving production technologies. However, some recent studies find no relationship between CF and household welfare or a negative one (Ragasa et al, 20180, while another recent review of CF studies argued that the evidence remains inconclusive because too few of the existing studies used methods that were appropriate for making causal claims (Bellemare, 2018).Among research focused on sugarcane and household welfare, some studies find positive effects or associations of cane production with improved household and community measures of wellbeing, while other find negative associations. For example, Martinelli et al. (2011) examined how human development indicators (HDIs) varied across municipalities with different levels of sugarcane production in Sao Paulo, Brazil, and found a statistically significant relationship between the presence of a strong sugar and ethanol industry and higher levels of economic and social development. Municipalities with a sugar mill, on average performed better on human development over the past decade than those without a sugar mill. Sugarcane is often grown next to factory/mill sites resulting in a need for significant infrastructural development, such as housing, roads, schools, and medical facilities for people involved in the production and processing of the cane. For sugarcane production to be sustainable, the sector should not only increase the employment and income potential of the farmers, but also contribute to overall well-being such as in food security and health (El Chami et al., 2020). However, some agricultural practices in sugarcane farming may cause health problems such as excess risk of respiratory diseases. For example, pre-harvest burning of sugarcane straw is significantly associated with higher rates of hospital admissions for respiratory diseases in children under five years old in Brazil (Paraiso and Gouveia, 2015). Martiniello and Azambuja (2019) claim that sugarcane contract farming schemes are associated with an increase in food insecurity among rural households in Eastern Africa. The authors attribute this in part to land being shifted from food crops into sugarcane. In addition, farmers tend to allocate most of their land to sugarcane cultivation in the hope of maximizing monetary revenues, at the expense of more traditional food crops. In Thailand, Intarapoom et al. (2018) examined the impacts of sugarcane farmland expansion on the four dimensions (food availability, access, utilization, and stability) of food security among the sugarcane-farming households. Results showed that increasing land allocation to cane production was associated with the lowest food security when compared to households that did not convert their land into cane production.In Uganda, Mwavu et al. (2018) assessed the contribution of commercial sugarcane production on household level food security among smallholder farmers in the Busoga sub-region (jinja and Mayuge districts), a major sugar-producing region in Eastern Uganda. They find that 87 percent of the respondents, and 7 in every 10 households of commercial sugarcane growers were lacking adequate and nutritious food in their households in the last 12 months prior to the study. Most households grow few food crop varieties and have inadequate income to purchase food to meet their needs or supplement what they grow. They conclude that sugarcane cultivation may be a key driver of food insecurity in Uganda --despite the perception that it offers benefits of poverty alleviation and improved human and social welfare at household and community levels (Mwavu et al., 2018). Similarly, Lwanga et al. (2015) also conducted a cross-sectional study on households in Nabitambala parish, Eastern Uganda, and find that only 12 percent of households were food secure while 49.7 percent were severely food insecure. By contrast, Ahmed et al. (2019) find that despite the lack of a stable market, sugarcane smallholders in Ghana have lower levels of multi-dimensional poverty and higher levels of income than the control group, and income obtained through sugarcane cultivation is higher than that of food crop farming. However, the higher levels of objective wellbeing do not translate into higher levels of subjective well-being such as satisfaction of life and happiness.Yet, it is important to note that because the two Uganda studies above interviewed only sugarcane growers, they were not able to assess whether cane growers in these areas have better, worse, or similar food security outcomes to non-growers. The answer to that question is important because food insecurity among some cane growers does not necessarily mean that participation in cane production has caused that food insecurity; the root cause of a household's food insecurity could be factors common to other households in the community and/or specific to the household in question.Several papers have discussed the growing competition between sugarcane and food crops on land use that is threatening world food production and consequently, food security. Other social and environmental impacts include the harmful impact on biodiversity and negative environmental externalities such as air and water quality and quantity, pollution, all of which affect food utilization, another pillar of food security. Notably, given that agrobiodiversity forms the basis for sustained household food production and food security, the commercial monoculture of sugarcane cultivation in Eastern Uganda impedes possible advancements in food security for the region (Mwavu et al., 2016). Therefore, the loss of biodiversity threatens agricultural production and food security and practices that conserve, sustainably use natural resources and enhance biodiversity are necessary at all levels in farming systems as they important for food production, livelihood security, health and the maintenance of ecosystems (Thrupp, 2000).Contract farming and other institutional arrangements vary greatly as they have different underlying structures, terms, and conditions. This implies that the differences in institutional arrangements between millers and growers will have various implications on the welfare and household food security of cane producers in different areas. Additionally, as outgrower schemes are used as a mechanism to commercialise small-scale farming, the impact of sugarcane farming on farmers' incomes depends on multiple factors. These include the income generating potential of the land, size of the farm, practices adopted by the farm and other institutional, local, and social contexts (Herrmann et al. 2018;Wendimu et al. 2016;Aleme 2019).For example, Herrmann et al. (2018) in Malawi compares food security measures of outgrowers and non outgrowers and find that outgrowers earn significantly higher incomes and allocate more land to food crops. In areas with compulsory participation in sugarcane outgrower schemes in Ethiopia, Wendimu et al. (2016) find that participation in outgrower schemes has a significant short and long run negative effect on the income and, a significant long-run negative impact on asset stocks of outgrowers whose land had a high potential for income generation prior to participation in sugarcane schemes. In addition, food security in outgrower villages declined over time but improved in non outgrower villages mainly due to less land allocation to food crops. Notably, crop diversity improves welfare and food and nutrition management in rural households (Tesfaye and Tirivayi, 2020). Therefore, the opportunity cost of the land was too high to have a positive impact on welfare. Similarly, Aleme (2019) used analysis of a computable general equilibrium model and concluded that there is a strong trade-off between sugarcane plantation and household welfare, represented as income and expenditure, in Ethiopia. On the other hand, Dam Lam et al. (2017) find that sugarcane farmers in Ethiopia have a lower prevalence of undernourishment and poverty levels.By contrast, cane outgrowers in Zambia have access to better water facilities, electricity, and more income earnings than non-participants, and 74 percent of cane growers were food secure compared with 47 percent of non-growers (Bubala et al. (2018). On other hand, cane growers have more debt compared to the non-cane growers. Overall, Bubala et al (2018) find that sugarcane growers participating in the outgrowers scheme were far better off than non-outgrowers and non-cane growers as it ensured improved livelihood and food security. The stark differences across studies in the relationship between participation in cane production and household food security raises important questions about whether differences in institutional arrangements and other local factors condition this relationship.Integration of smallholders in outgrower cane schemes has been advanced as a strategy for poverty reduction, but outgrower cane farmers are not homogenous (Hall et al., 2015). They have distinct resource endowments in terms of land control, labour conditions, financial resources and social resources which shape their participation in cane production. In addition, women experience a gender gap with respect to access to productive assets, like land; agricultural inputs, family labour; and services such as credit and extension, which together can result in women having lower crop productivity. These gender inequalities are underpinned by harmful gender norms which restrict women's livelihood roles and economic opportunities within agrifood systems. Women in most Sub-Saharan countries predominantly participate in nonfarm activities such as small-scale trading due to these factors. For example, in Kakamega, Kenya, sugarcane is a major cash crop, and most farmers are outgrowers. Outgrowers must own land but based on Abaluya cultural norms, most women are unable to become outgrowers due to lack of control and ownership of land (Loison 2019). (Loison 2019). In Zambia, land ownership is crucial in determining smallholder cultivation of sugarcane (Manda et al., 2020). Each household is required to hold a maximum of 4 hectares of land in the sugarcane catchment area. This implies that the landless, land scarce and marginal landowning households are excluded, including the poor who cannot afford to purchase land in the scheme catchment area. These are mainly women, the aged, widows and youths.In Jinja, eastern Uganda, a study on smallholder sugar producing households found that most cane contract farmers are men, so most of the income is paid directly to men (Ambler et al. 2021). Additionally, women spend more time working on non-cane agriculture and 4-5 times more time on household management and chores (ibid). An intervention in this area encouraged couples to register a block of cane in the wife's name, effectively transferring an asset (and potential income) from the husband to the wife. The study found that 70 percent of invited households accepted the offer to register cane blocks in the wife's name and acceptance was even higher among households randomly selected to attend a couple's workshop focused on gender equity and balance within the household. The study also finds that low socioeconomic status and household gender norms that prevent women's economic participation in the sugarcane value chain acted as barriers to the household's acceptance of the contract intervention. For this study, we explore whether final decision on what to plant on a parcel and harvested crop allocations/decisions by gender matter for food security.This study used primary data (qualitative and quantitative from Uganda's sugarcane-growing subregions collected by the Economic Policy Research Centre (EPRC), Uganda in the context of the Innovations Lab for Food Security Policy, Research, Capacity, and Influence (PRCI) project. This section provides detailed information on the description of the study areas, data collected and data analysis methods.Sugarcane production in the sub-regions of Buganda1 , Busoga2 and Bunyoro3 is dominated by three (3) historical mills with nucleus farms. These also relay on outgrowers to fill the supply gap. Agriculture in Uganda contributes about 23 percent of GDP (UBOS 2022) and employs over 70 percent of the labour force excluding subsistence farming (UBOS 2021). Crops grown vary across the three sub regions. Maize, sweat potatoes, groundnuts are the main crops cultivated in Busoga; maize, cooking banana in Buganda while in Bunyoro, maize, beans, cassava, nuts (UBOS 2021). In all the three sub regions sugarcane growing is the predominant cash crop grown by majority of small holder farmers.The EPRC-PRCI project selected three sub regions (Buganda, Busoga, and Bunyoro) in Uganda for several reasons. First these, are the regions with historical districts that started sugarcane growing in Uganda with the three largest nucleus farm estates. Second, these regions also have well established large mills, recognised, and organised out-grower-miller arrangements with massive expansion plans. Busoga and Bunyoro sub regions also had increasing income poverty (UBOS, 2021), partly attributed to sugarcane growing by households. Furthermore, these have the number of households that potentially are food secure affected from sugarcane growing (directly or indirectly). Lastly, no comprehensive study on sugarcane growing effects on food security has been conducted in these sub regions with national representation.The study used a three-stage sampling design. In the first stage, sub-counties were randomly selected in each of the 16 districts who were major sugarcane growers. From the selected sub-counties, a total of 120 villages were randomly selected using probabilities proportional to size (PPS). Next, for all the selected villages, a listing of all the farming households were conducted where information on whether a household grows cane, and cane production arrangements were collected. The listing also captured information on whether cane growing households have cane plots that are owned by women, sex of the household head, and key decision-making indicators by gender to allow for sampling that supports gender analysis.Sampling within a village first involved stratification of farmers into two categories: cane growers and non-cane growers. Next, cane growers were also categorized by their institutional arrangement with a nearby mill: registered and aided 4 , registered and not aided, and unregistered and unaided growers. The listing and stratification were followed by random selection of 20 households (15 sugarcane growers, and 5 non-sugarcane growers) from each of the selected villages.The primary data were collected using a combination of qualitative and quantitative methods. From a quantitative approach, the data is from a cross-sectional survey of farm households and communities in Uganda collected in December 2021. The survey targeted Uganda's main sugarcane growing sub-regions-Busoga, Buganda and Bunyoro, and collected quantitative data using a semistructured questionnaire at the parcel-, household-, and community-levels. It used the 2014 Uganda Population and Housing Census sampling frame to select a representative sample of cane and noncane growing households from these sub-regions.Within each sub-region, data was collected from the 16 sugar growing districts that have at least one milling factory, as cane growing farmers are nearly always concentrated near a milling factory. The reason for this is because harvested sugarcane is both highly perishable and has a low value to weight ratio, which means in practice that cane production is only profitable for farmers if it is grown relatively close to a mill. Detailed information on household demographics, food security, land management and community characteristics were collected for both cane and non-cane growers. Out of 2,400 households listed, 1,800 were surveyed and 1,771 had complete household information from 72 communities from which analysis was done. Households were classified as cane growers where the head was currently growing cane and non-cane grower where household heads did not grow cane. A binary variable (1 for cane growers and 0 for non-cane growers) was used for this categorisation. Note that analysis is at the household level unless otherwise stated. In addition, weights were applied to ensure that data is nationally representative.For the qualitative approach, key informant interviews were conducted with community leaders, district agricultural officers, and two community barazas were conducted to share on food security vs sugarcane production. Focus Group Discussions (FGDs) were also conducted with separate groups of males and females on the issues related to sugarcane production, challenges, opportunities, and implications for food security in the community and households. Twenty-one FGDs and 19 KIIs were conducted.Food security is a complex issue that cannot be measured by one indicator alone, which explains why a large number of food security measures have been developed, though each have their own strengths and weaknesses (Manikas et al., 2023;Kolog et al., 2023;Becquey et al, 2010). In this analysis, we used the definition of food security from the Food and Agriculture Organization of the to the mill. Aided means that a mill and grower agree that the mill will provide the grower with inputs such as inorganic fertilizers and planting material, typically on credit, on the condition the grower commits to deliver the agreed area of harvested cane to the mill at harvest, whereby the mill the cost of inputs provided on credit by deducting their cost from the gross value of the cane the grower delivers to the mill.United Nations (FAO), in which food security is divided into four dimensions: physical availability of food, economic and physical access to food, food utilization and the stability of these three dimensions over time (FAO, 2008). In this study, household food security indicators (availability, access, utilisation, and stability) were measured using three food security indicators of the months of adequate household food provision (MAHFP), household food insecurity access score/scale (HFIAS) and the household dietary diversity score (HDDS) as summarised in Table 1. The food items in Uganda commonly consumed within each food group are shown in Appendix Table A1. The HFIAS is a continuous measure of the degree of food insecurity (access) in the household on the past 30 days. It also reflects the three universal domains of household food insecurity, insufficient quantity and insufficient quality of food supplies. This indicator captures the household's perception about their diet regardless of its nutritional composition. The HFIAS value ranges from 0-27 for the nine-food insecurity related conditions. At a household level, a high HFIAS shows that a household is very food insecure, while a low score shows that a household is less food insecure. HFIAS is also measured on a scale of 0-3. Coates, J., Swindale, A, and P. Bilinsky (2007). Household Food Insecurity Access Scale (HFIAS) for Measurement of Household Food Access: Indicator Guide (v. 3). Washington, D.C.: FHI 360/FANTA. Source: Kolog, Asem and Mensah-Bonsu (2023) To assess the utilisation dimension, we used the household dietary diversity score (HDDS), which counts the number of different food groups the household consumed up to a maximum of 12 food groups (Swindale and Bilinsky, 2006). It is based on survey respondent estimates of household consumption during the past 7 days. HDDS was designed to serve as an indicator of food access (Swindale and Bilinsky, 2006), yet in some cases has been found to be correlated with micronutrient deficiency (Hatløy, et al., 2000). However, food utilization is described not only by access to micronutrients but also by how the body makes use of them (FAO, 2008). This is strongly influenced by the health status of household members, especially the status of the digestive system. We do not account for this variable as our data did not collect it.The month of adequate household good provisioning (MAHFP) indicator was used to assess the stability dimension of food security because the MAHFP reflects the stability of a minimum food supply throughout the year (Bilinsky and Swindale, 2010;Coates, 2013). The MAHFP counts the number of months in the last year in which the household had enough food available (Bilinsky and Swindale, 2010). It thus ranges from 0 to 12. This indicator is relatively subjective because it is up to the respondent to decide how much food he or she considers as enough. Lastly, the Household Food Insecurity Access Scale (HFIAS) was used to assess the access to food and availability.In a binary case, the programme is one if the household is a sugarcane grower and is zero otherwise. Other variables range from household institutional and locational factors such as age of household head, gender of household head, marital status of household head, education of household head and spouse, household size, farm size, access to extension services, access to markets, wealth status, crop diversification, allocation of crop harvest proceeds, household non-farm income (wages and salaries) among others (Figure 1). There are two pathways through which sugarcane growing contributes to household food security. First, sugarcane growing households may decide to allocate their land to sugarcane production to obtain cash income. The household income generated through cane sales can be used to purchase food items. It is that proportion of household income from cane sales spent on food that enhances household food security status. Second, if the income is not adequate to meet food obligations as the case has been when farmers failed to sell cane for years, households may decide to reduce food expenses to maintain a certain standard of food status in a household. In the long run, if there is a decline in income from cane which could be due to a reduction in prices at which cane is sold or a complete failure by millers to buy the farmers cane, then sugarcane food intake may be impaired. In analysing the effect of sugarcane production on household food security, the study used a Poisson regression model specification given that the three food security outcome variables used are count variables. The model estimates the impact of these predictors on the expected count or rate of the event of interest. The Poisson regression model assumes equi-dispersion -that the mean and variance of the count variable are equal. However, if this assumption is violated and overdispersion occurs, Poisson estimates can be biased, though alternative models like the negative binomial regression can be used.The Poisson regression model is a generalised linear model (GLM) that meets the classical assumptions with only one exception, the distribution. The dependent variable assumes the Poisson distribution; regardless of whether the distribution is maintained or not, asymptotically normal and consistent estimators of k B are obtained.The GLM is written as:Where 0  is the intersection term,  is a vector of coefficients, ( ). ( )In particular, the expected value is expressed as an exponential function (equation 2a), and the mean is equal to the variance (equation 3):( ) ( )The probability density function (pdf) of the Poisson distribution is given by: ( )Where ( ) fy is the probability that the variable y takes non-negative integer values ( ) Where y is the HFIAS/HDDS/MAHFP, a count dependent variable; 0  is the intercept; 12 , ,..., k    are vectors of unknown parameters to be estimated; k x is a vector of explanatory variables of household i ; i u is a robust standard error term. Explanatory variables include demographic, socioeconomic (household is cane grower or not), and social characteristics (household size, age, sex, level of education of head and spouse, land area, marital status of head, access to market, annual household income among others). The maximum likelihood method (MLE) estimates the model parameter vector.Put more simply, we estimate our Poisson model as below with variables as explained above with inclusion of a binary variable of whether a household is a cane grower or not as an explanatory variable.Determinants of household food security among households in sugarcane sub regions are derived by employing an ordinal/ordered Probit, as one of the measures of food security is also derived as a categorical and ordinal. The multinomial probit or logit in this case would not be ideal as it does not account for the dependent variable's ordinal nature despite the outcome being discrete. The ordered probit is the most widely used in several studies for ordered response data and it assumes a normally distributed error term.The ordered probit model, as formulated by (Greene, 2002) is modelled on an unobservable latent random variable as follows;,Where ( )( ) WhereThe probability can be written as:Where J is the categories of responses to food security and ( )Where ( ) .  is the standard normal density function. Therefore, the empirical model for the analysis of this objective is specified as follows:The dependent variable, given as FS is the household's food security status proxied by HFIAS. i characterizes the i th household, ( )represents the four categories of the dependent variable indicated as; if household falls within severely food insecure, moderately food insecure, mildly food insecure or food secure categories for HFIAS, , , ,    are estimated parameters; W and Z are socioeconomic characteristics, and institutional and location characteristics of the respective household expected to influence their food security status.Following Hyodo and Hasegawa ( 2021), the Chi-square test was used to analyze whether food security status of households was affected by some independent variables of the study. This was to test the general null hypothesis that food security status of households is independent of the categorical variables of interest. The general alternative hypothesis states that food security status of households is not independent of the categorical variable of interest. This checks the robustness of the ordered probit model used in the study. The Chi-square statistics are calculated as below:( ) To analyze if the differences in institutional arrangements between millers and growers lead to differences in the household food security of cane producers, we use Poisson regression and control for only cane growers and introduce miller-outgrower arrangements as one of the explanatory variables. We modify equation 7a and include out-grower models and estimate equation 1 c.Where y are the various measures of food security.Lastly, the study also set out to answer the question of whether women's influence in intrahousehold decision-making in crop choice, crop marketing, and allocation of crop sales income influences household food security. To answer this, the survey was designed to gathered genderdifferentiated information on land ownership and decision making at the plot level, for all plots. For example, for each plot, the survey asked, \"who makes the final decision on what to plant on the parcel?\" and 'Who decides how to allocate crop production harvested on this parcel? e.g. whether to sell, consume etc)? Collating all plots owned by an individual within a household, we construct a decision-making variable at plot level first which we collapse to household level. These are mutually exclusive categories.,Different impact estimation procedures may lead to slightly different impact estimates, especially when cross-sectional data is used for impact assessment. Because this study uses cross-sectional data, sensitivity analysis of our impact estimates was conducted using a propensity score matching (PSM) approach, using different matching techniques (nearest neighbour, and Kernel regression).Additional matching techniques such as the nearest neighbour matching (NNMatch), inverse probability weighting (IPW) was also used for robustness. In the absence of randomized treatment of an intervention or program, estimating the impact of program participation (or treatment) can be difficult, as factors typically unobserved within household surveys -such as cultural norms, religious beliefs, and other unmeasured explanatory variables --may simultaneously affect participation in the program and the outcome. In this case, sugarcane production may be correlated with unobserved household-specific factors also correlated with household food security. Due to a lack of a suitable instrumental variable, the following analysis proceeds on the assumption that a dummy variable indicator of household participation in cane production is exogenous within our regression models.This assumption does not appear to be strong given the range of household and community-level controls in our regression models that are known to influence a farmer's decision to grow cane or not -such as their total landholding and total asset value --as well as their household food security status during the recall period.In this section, first we examine the social-economic characteristics of the cane and non-cane growing households in the survey sample, the institutional and community characteristics of villages in which they reside, and food security status of cane and non-cane growing households. We then analyze the relationship between the various food security indicators and participation in sugarcane production. Finally, we Poisson and ordered Probit regressions to assess various specifications of factors associated with household food security and then address the key questions and hypotheses to be tested.We next compare socioeconomic characteristics of cane growers and non-growers to see if there are any systematic differences between them. The average age of household heads in the sample was 48 years for both groups (Table 2), though cane growers were more likely to be male headed (9 out of 10) compared to 7 out of 10 among non-cane growers. Average family size is slightly larger for cane growers. About 46% and 49% of the cane and non-cane grower households' heads, respectively, were employed for salary or wage. On average, cane growers had about 7 times more land than the non-cane growers, and cane growers are more educated and likely to be literate. In addition, maximum adult (18+) female education in the household was 8.3 years while for non-cane growers had 7 years. Cane growers have considerably more asset wealth, as the average value of their household assets is over three times that of non-growers (Ugx3.3m vs Ugx0.92 million respectively). Among cane growers, decisions on how to allocate crop production harvested such as sell or consume are primarily made by the male head (49%) with fewer decisions made by the female head/spouse or jointly done. Non-grower households are less likely to have a male head or spouse (40 percent) make decisions about the allocation of crop sales income allocation decisions. In short, patriarchy is more entrenched in cane grower households. Give the potential role of education in shaping household incomes and decision making on food security, sub regional insights on maximum adult education attainment show that Buganda and Busoga subregions have similar levels of maximum adult female education levels, while that of Bunyoro is a bit lower (6.8 years) (Figure 2). Likewise, maximum adult male education levels were higher on average in Buganda and Busoga than in Bunyoro. On average, both cane and non-cane growers cultivate 3 food crops. About 24.9% and 19.4% of cane grower and non-cane grower households respectively received extension support for other crops grown. Access to market on average took 23 minutes to walk within 2 miles. Most households (over 70%) in our study are from Busoga subregion. There were no significant differences by cane production status in distance to districts or food markets, though minor differences in household shocks for disease and pests and income related shocks.  The mean values for HDDS and MAHFP were higher in cane grower households than for non-cane growers while the mean HFIAS score was higher in non-canegrower households (Table 4). That is the HDDS mean score for cane and non-cane growers is 6.6 and 5.97 respectively with the same median score of 6, on a scale of 0 to 12 food groups. The MAHFP shows that cane grower households had on average 10.3months of adequate food while non-cane grower households had 9.1 months of adequate food provisions. For HFIAS score, the lower the score (0-27) the less food insecure is a household, implying that a HFIAS score of 6.14 and 7.84 for cane and non-cane grower households implies that the latter households' access to food is burdensome. Using the HFIAS scale, about 27% of the sampled cane grower households were food secure. In the same category, sampled households who were mildly, moderately, and severely food insecure represented about 9%, 43% and 21% respectively as measured by HFIAS scale. In sampled non-cane grower households, 24% were food secure while 10%, 36% and 30% were mildly, moderately, and severely food insecure respectively. Non-cane growers were the majority in the moderately and severely food insecure categories compared to cane growers. Using the HFIAS scale, 4 out of 10 households are moderately food insecure irrespective of cane growing status. Significant differences are observed in all food security indicators between cane and non-cane grower households. Clearly food insecurity (using the three measures) is still a major challenge faced by majority of households in Uganda and more especially among non-cane growers. The descriptive analysis above has compared household food security and household characteristics between households that currently grow cane (2021) and those that do not and finds that cane growers have higher food security measures (on average) compared with non-growers. However, if cane production \"causes\" food insecurity, on average, then the mean of average food security of non-cane growers could potentially be pulled down (perhaps below that of current cane growers) due to low food security levels of past cane growers that have not yet recovered from financial losses in a prior year related to cane production. Ideally, if observations of food security and cane participation for the sample households had been observed in both 2021 and a prior year, it would theoretically have been possible to test whether there is any causal effect of cane participation on household food security using a difference-in-differences approach. Because this study has only cross-sectional data 2021 to use, a different approach is needed. We thus divide the subgroup of households that do not currently grow cane into those that are past cane growers and those that have never grown cane. Interestingly, past cane growers have better or similar food security measures than those that have never grown cane (grey bars in Figure 3). This does not imply that some past cane growers are not food insecure and that that status may be due in part to financial losses from cane production in a prior year. However, it does show that the subsample of farmers that have never grown cane have lower food security measures than both current and past cane growers, on average. Next, the subgroups of current cane growers and past cane growers are divided into two subgroups based on their current (or past) registration status with a mill. For example, if a current grower is registered (contracted to deliver cane to a mill) or registered with aid (registered and provided some inputs by the mill, possibly on credit), they are termed \"current cane with miller\" in Figure 4.Current cane growers that are not currently registered or registered with aid are classified as being \"without a miller. This can also be done with past cane growers, where their \"with\" or \"without \"miller status refers to their relationship (or not) with a mill in the last year that they grew cane.We find that both current and past cane grower households who have (had) arrangements with a mill in 2021 (or in last year they grew cane) have better food security levels for HFIAS (i.e. lower value) and MAFHP and HDDS (higher value) than both households that have never grown cane and past cane growers who did not have arrangements with a mill. It is not clear whether causality runs from inherent grower characteristics that enable them to achieve higher income and food security, if becoming registered or registered with aid with a mill causes better food security outcomes, or if some of both occur. More cane growing households indicated in December 2021 that they had had adequate food provision over the prior 12 months as compared with non-cane households. In addition, a histogram of MAHFP, which shows the density of this variable, shows visually that current cane growers are more likely to have higher number of months of adequate food provision compared with non-growers in 2021 (Figure 5).Figure 6 further shows that about 24 percent of the sampled households reported February, March, and January of 2021 as the months in which they had inadequacy in food provisioning. These months correspond with the second season cropping period when Uganda's climate variations often result in droughts that in turn can result in food price spikes as food supplies dwindle. The seasonality of food prices and availability highlight the need for households to have enough income and/or savings to ensure access to food even during the months each year with the most limited food supplies and highest food prices.Figure 7 shows that while February 2021 was the commonly cited month of inadequate household food provisions irrespective of cane status, for the other months inadequacy differed among cane and non-cane growers. While among non-cane growers, seasonality in cropping periods partly explained the food inadequacies in the various months with a clear patten, among cane growing households, the pattern beyond seasons was partly driven by sales of harvest to mills and payment for sugarcane harvests. At the time of the survey, at least a good portion of households had sold Histograms of HDDS shown in Figure 9 show the distribution of HDDS values for cane growers and non-growers. This indicates that HDDS of cane growers is concentrated among slightly higher values of HDDS compared with non-growers, which helps to explain why mean HDDS is higher for growers than non-growers.Similarly, histograms of the HFIAS scores in Figure 10 show the distribution of HFIAS among cane growers and non-growers across the variable's range from 0-27. The higher the score the more food insecure a household is with respect to accessing its required food needs. The large density around zero indicates a significant share of farmers that indicated having no aspects of household food insecurity within the past week. Based on the HFIAS, approximately 3 of every 10 cane growers were classified as being \"severely food insecure\" in the past 30 days, 6 of 10 \"moderately food insecure\"; 2 of 10 were \"Mildly food insecure\" and 4 in ten were \"food secure\" (Figure 11). A similar pattern is observed for non-cane growers yet they are more likely to be severely food insecure (30 percent) than cane growers (21 percent) (Table 6). Household food security also varies by gender of the head, as those with female heads are more likely to be severely food insecure (39%) than those with male heads (23%) (Appendix Table A2). Yet, female headed cane growing households are less likely to be severely food insecure (33%) compared with female headed non-grower household (41%). A similar pattern is seen for male headed cane and non-cane growing households  Not surprisingly, food insecurity has a strong negative relationship with household total landholding, as the mean and median of landholding declines significantly as one moves from the sample of food secure households (mean landholding of (12.8 acres) to those that are mildly food secure (7.8), then from mildly (7.8) to moderately food insecure (6.9), and again from moderately to severely food insecure (3.8) (Table 7). It is also clear that current cane growers have much higher mean and median landholding (14.8 and 6.6 acres, respectively) than both past (3.5 and 2 acres) and \"never\" cane growers (2.5 and 1.75 acres). Because past cane growers have better mean and median food security levels than never cane growers (Figure 4) yet similar median landholding, this suggests that past cane growers likely have access to relatively high return non-farm sources of income, which may help explain why their food security status is better than that of never cane growers. On average, February is the month in which the highest share of households (24% across the 3 regions) reporting that they had inadequate food provision. The share of household with inadequate quantities of food rises considerably from November to February, then begins to decline in March as farm households being to harvest food from their major season, then falls to a share of only 5 percent of households from June until November. The trend corresponds to Uganda's climatologies and the bimodal cropping seasons (major season is in March-May and a shorter rainfall season in September-November). Farm households in Buganda are more likely to be food secure (41 percent) than those in Busoga (23.8) or Bunyoro (15 percent) (Table 7). In each region, cane growers are more likely to be food secure (and les slikely to be severely food insecure) that non-cane growers. The most significant regional difference in food security in 2021 is that moderate and severe food insecurity is much more common in Busoga and Bunyoro than in Buganda. For example, about 68 percent of households in Busoga were moderately or severely food insecure compared with 65 percent in Bunyoro yet only 44 percent in Buganda. This is likely due to the fact that Bugandan households in our sample have considerably more asset wealth than the other regions. In fact, median Bugandan total household value of assets per AE -including livestock, household, transport assets -was 518,000 Ush/AE, which is nearly double that of Bunyoro (234,000 Ush/AE) and Busoga (288,000 Ush/AEAE). This is likely explained by the fact that rural households in Buganda have access to more remunerative activities in both crop production (coffee and banana, in addition to cane) and nonfarm own business and employment -given their proximity to Kampala. Part of the relatively high food insecurity in Bunyoro observed in Dec 2021 may be related to the fact that the 2 large mills in Bunyoro harvest farmers' cane for them, but then often do not pay the farmers for up to 2 months. Box 1 highlights stakeholers views on the status of sugarcne growing and food secueiry in Bunyoro subregion, Masindi district in particular.Majority of the people were facing food uncertainties in Masindi sugarcane growing communities. This is because many people have taken on cane growing as a buffer for earning money. This means they left little or no land at all for food crops. A participant said: \"We have an issue of food security since most of the land is being utilised for cane growing. I started with one hectare and educated my children but as per now, Kinyara has taken long to harvest my cane and some is drying in the field.\"Another perspective was that those who are growing cane and have no food insecurity issues are the rich. A participant said: \"When you look at people involved in cane growing, these are people who are well-off. Almost 75% of the farmers are hiring land from the poor land owners.\"Even those not growing cane are facing challenges because they hire out their land to outsiders to grow cane on them. This means their land is locked out for longer periods leaving them with no room to grow food on their own. Participants added that: \"A poor landowner has rented out land for a period of 6 years to an investor and remains with a very small portion for growing crops but that land is held for 10 years not the 6years agreed upon. The harvest is agreed upon for 18 months but it might go to 34 months which you have to multiply by four harvests. This is affecting landlords since your land is being held for a long period of time than anticipated and he doesn't pay you for the longer period but for only the harvest.\"\"You find a home surrounded by cane which does not belong to that household. We are also selling our own land once we get problems. Once done there is no food for the home. People are now cutting forest and digging in food reserves hence we have a problem of food. You find a woman waking up early in the morning to go and weed in Kinyara and a man gets a Panga to go and cut cane in order to get Posho, hence we have a serious food security problem.\"It was also revealed that people new to sugarcane are dedicating all their land to cane without leaving some for other foods. A participant said: \"We farmers who started growing cane earlier have more knowledge than others and we don't have any food insecurity. Here for example, you cut your cassava and plant like 100 stems. These you will eat like for four months once harvested. How does a person complain of food security when he is s cane grower, I call that poor planning.\"Others said food insecurity was a much wider problem than people were willing to admit, and it goes beyond cane growing communities. A participant said that: \"There is no food completely when you go deep down in the villages and that is poor planning. Even outside cane growing area like in Kampala, we have people who feed on beans and posho from Monday to Monday. Food security here is hard, I used to feed my family on fish twice in a week and I cannot access fish because the government interfered and can now only afford small fish\". Another participant emphasised that: \"People do not follow advice. You are supposed to leave some land for food but they don't. The population has increased and the land acreage for sugarcane growing has increased and as it increases, the land acreage of food crops is reducing. Some 80% of the land being used for cane growing is managed by external people and a few people who come from within and the few rich people who own these farms. Some tenants who came from West Nile went back to West Nile hence migrating to their homes.\"Source: Community Baraza held in Masindi District on \"Sugarcane production Vs Food Security: Is there cause for worry\", December, 2021Household wealth is expected to have a high positive correlation with household food security, as it provides the ability to achieve food security through the market purchase pathway. This strong positive correlation is apparent from shares of household by wealth quintile and by food insecurity access scale category. For example, among households categorized as food secure by the HFIAS measure, only 7 percent are from the lowest (first) asset wealth quintile compared with 33.7 percent in the top (fifth) quintile (Table 8, upper panel). Likewise, 42.6 percent of households categorized as severely food insecure are in the lowest quintile compared with only 6.8 percent of households in the top quintile. Shares of household distributed across the four food insecurity access scale categories computed by row show the same pattern, as only 9.7% of households in the first (lowest) wealth quintile are food secure compared with 55.8 percent that are several food insecure. However, the vulnerability of rural Ugandans in these regions to various kinds of unexpected shocks (weather, insects, crop disease, illness, etc) appears to affect at least the temporary food security of even those in the top wealth quintile. For example, about 26 percent of households in the top wealth quintile reported moderate to severe food insecurity during the month prior to interview in Dec 2021. -------share of households by row (%) --------------share of households by column (%) -------This was re-echoed by a female cane grower \"…People who have small pieces of land are growing sugarcane and they surely don't have food. Most people with limited land like two acres have used all the land for sugarcane growing, they even hired out most of their land to other people still to grow sugarcane and they abandoned food production. Me as a person, I left some land for different types of food like Beans, Cassava, Potatoes so me am very safe concerning food, but most people in my village don't have food (by Female FGD participant, Imanyiro Mayuge)In looking at specific wealth indicators such as household ownership of livestock (Figure 11) and value of assets (Figure 12) by HFIAS, cane growing status and subregion levels. For instance, we note that Buganda followed by and past cane growers and food seuecure and midley food insecure houdholds owenered a higher share of ltropical livestock than in Bunoro, non cane growing and severely foo insecure houeholdshad more live animals (Figure 11). Food secure, current cane growers and those in Buganda sub regions had household assets of much higher value (Figure 12). Poisson correlates of household food security Table 9 presents results of the Poisson model of household food security regressed on various household and community-level factors known to influence household food security. Three different dependent variable measures of food security are used (separately) 0 HFIAS, HDDS and MAHFP. Before proceeding with regression analysis, multicollinearity among explanatory variables was tested by first estimating an ordinary least squares (OLS) model and all variables with a very high VIF were excluded from the Poisson model specification -such as marital status of the household head -and others with VIF <14, such as total land size and household age were maintained in the model specification. Furthermore, choice of the Poisson model was selected by econometric validity of the results obtained with the Negative Binomial regression model method, which is used to test whether the dispersion of the dependent variable would be better fit by a Poisson or a Negative Binomial model. 55 Application of the Poisson model was verified by the negative binomial model alpha of 0.971 with robust clustered standard errors of 0.1465 which indicates that the specification was not significantly different from zero. This implies that a Poisson produced a better fit for the data.  Results from the Poisson regression of HFIAS show that household participation in cane production is associated with a statistically significant 1.181 reduction in HFIAS (p<0.1), which is equivalent to a 17 percent reduction in food insecurity, relative to the sample mean of HFIAS (Table 9, Col.2). Other factors with a statistically significant and negative association with HFIAS (lower food insecurity) include male headed households, age of the household head, log of the annual income earned from a salary, maximum adult female education in a household (p<0.05), number of live large animals, household value of assets, and the number of food crops grown (p<0.05). All these factors are associated with lower household food insecurity.Factors that have a statistically significant positive association with food HFIAS (higher food insecurity) include spatial dummies for Busoga and Bunyoro subregions, as compared with Buganda, the base category (p<0.05). In addition, households with one or more members in wage employment, following an indigenous religion (as compared with being Catholic, the base category), household shocks pertaining to crop pets and diseases, and the distance to the nearest district town (p<0.05) all have a statistically significant and negative association with household food insecurity. If the binary household cane production variable is interacted with select variables, it does not change their association with HFIAS (Table 9, Col. 2).A Poisson regression with HDDS as the household food security measure (Table 9-Col (3)) finds no statistically significant association of cane production on household diet diversity (i.e. food security). Yet, the value of annual income from salaried employment, household maximum adult female education, number of live small animals owned, and total household value of assets all have a statistically significant and positive association with HDDS (i.e. better food security). There is a statistically significant and negative association between residence in Bunyoro subregion, indigenous religion, natural shocks (drought, floods, mudslides) and the death of a household member. (Table 9-Col. (4) provides estimates with interaction terms, the results of which are similar to those in Col (3) though for cane growers, distance to the district is associated with better food security while being a cane grower in Buganda is associated with lower food security.A Poisson regression of MAHFP (Table 9-Col ( 5)) shows that growing cane is associated with 1.06 additional months of adequate food provisions (p<0.05). This suggests that one benefit of growing cane enables households to better manage seasonality of food production, likely through higher household income. Other factors with a statistically significant association with MAHFP include male headship, maximum adult female education (years), the number of large animals and number of crops grown. Busoga and Bunyoro in comparison to Buganda are expected to have 1.65 and 1.06 lower MAHFP, while an additional household member (in AE terms, this would be a 20+ year old male) is associated with a 0.18 reduction in MAHFP (p<0.05).Ordered probit analysis of HFIAS is implemented to provide additional insight into correlations between cane participation and other factors and food insecurity and the degree of it. The HFIAS dependent variable is found to be ordered and the categories were significant (p < 0.001) (Table 10).The threshold value indicating the food insecurity categories; 1  , 2  , and 3  (cut1, cut2, and cut3) indicated that the categories are ranked in an ordered manner. The dependent variable is the food insecurity prevalence levels where the HFIAS values from 0 to 27 are categorized into four outcomes including: 0=food secure; 1=mildly food insecure, 2=Moderately food insecure, and 3=severely food insecure. The predicted probabilities of Y = 1 or the marginal effects were estimated, which measured changes in the probability of a food insecurity (access) outcome with respect to a one-unit change in an explanatory variable. Marginal effects of the respective models presented in Cols ( 2), ( 3), (4), and ( 5) are discussed since the coefficients of the ordered probit model (Col. ( 1)) do not show the magnitude of the effect of the independent variables (HFIAS)6 .Sugarcane growing households were likely to be classified as food secure and mildly food insecure levels and less likely to be classified as moderately and severely food insecure (Table 9). While the signs of marginal effects on the cane participation dummy are largely as expected, none have a statistically significant association with any of the HFIAS categories. Household asset value, growing more than one food crop, earning a non-farm salary, and a higher adult female education (years) all have a positive and statistically significant association with HFIAS categories. For example, an increase in household asset value is associated with a higher probability of being food secure or only mildly food insecure and is associated with a lower probability of being moderately or severely food insecure. It is not clear what the positive association of HFIAS with years of adult female education is capturing, but it could be an income source not included in the model. Higher levels of household asset value can improve food security though facilitating better credit access as well as potential sale of an asset if needed to make up for any shortfall in the household's expected farm and non-farm income during any given period.Age of head is associated with better food security, which may be capturing the likely positive association of experience and wealth with food security. As in earlier analysis, households located in Busoga and Bunyoro are less likely to be food secure and mildly food insecure, relative to household from Buganda (base category) and more likely to be in the moderately and severely food insecure categories. An additional household member (adult equivalent) is associated with lower food security. This is consistent with similar studies and not surprising as an additional individual may offer additional income for the household, yet it is certain that an additional member will increase the household's food and other expenditures (Kirimi, ud). Table 11 shows the detailed results of different matching techniques use to estimate the impact of sugarcane production on the three food security measures. The analysis is adjusted using all explanatory variables included in the Poisson regression model.Using the NNMatch and IPW matching methods, the average treatment on the treated (ATT) of cane production on HDDS is statistically significant and unexpectedly negative. This implies that growing cane is associated with a -0.33-point reduction in HDDS for cane growers, though this is only about a 4 percent reduction (Table 11, Col 2). Likewise, the PSM (1:1) matching method finds a statistically significant and negative association between growing cane and HDDS for both growers and non-growers (Average Treatment Effect), though the magnitude is quite small. The ATC (average treatment on the untreated) on the HFIAS of non-cane growers indicates that if they were to grow cane, their food insecurity be reduced by -0.8 and their months of adequate food provision would increase by 0.4 months. The Mahalanobis-distance kernel matching, NNMatch and PSM (1:1) estimators all find a statistically significant and positive association of cane growing with a 0.4 increase in the months of adequate food provision. That means that if a non-cane grower were to decide to grow cane, these matching methods estimate that their food insecurity would fall (HFIAS) slightly, and their months of food provision would increase by nearly half a month. None of the impact measurements using the PSM-kernel matching approach were statistically significant.While the latter results above suggest that cane production could benefit farmers that currently do not grow cane, the matching results need to be taken with caution, as the key assumption underlying matching methods is that after controlling for explanatory variables that are observed, there are no known factors that are unobserved -i.e. not included as an explanatory variable--that are expected to affect household food security, yet may also be correlated with cane production. If that is the case, then impact estimates from matching approaches can be biased just as endogeneity from omitted variables in regression analysis (a form of endogeneity) can bias program intervention dummies or other variables in a model. In this case, it seems that the assumption required for unbiased matching is strong given that we do not have a measure of expected or actual rainfall or other variables measuring agroecological potential.Graphical diagnosis assisted our observation of the distribution of propensity scores between treatment and control groups through use of Kernel density plots, cumulative density plots and a box whisker plot. The graphical representation of some of our matching model techniques show how matching successfully reduced bias between the treated and the untreated (Figure 13 and 14). Figure 14 demonstrates that the difference in the means of the propensity scores for the two groups being compared were small, where the means must be less than half a standard deviation apart.Likewise, the distributions of the covariates in both groups are nearly symmetric in the matched sample and the distributions of the covariates in both groups have nearly the same variances. The second research question of this study is whether, among cane growers, household food security varies by type of grower-miller institutional arrangement. We use descriptive analysis only as it is not feasible to include dummies of 3 of the 4 institutional arrangements into our food security regressions as such variables are likely endogenous and there is only one instrument that appears to be valid for them.Households that were not registered and not aided had better HFIAS and HDDS scores than those with some type of grower-miller arrangement, followed by farmers with using multiple arrangements (any two of the 4 institutional arrangement categories in Figure 15). Interestingly, households that were registered and aided had worse scores across our food security measurement indicators. Based on findings from recent analysis on cane productivity and profitability in Uganda (Mbowa et al, 2023b), it is likely that cane growers in Busoga are driving this result, as many were registered or registered-aided, yet few in Busoga saw benefits from those institutional arrangements in 2021 due to a near collapse in coordination between millers and growers. Households that were not registered and not aided had the lowest MAHFP score, yet the difference was not statistically significant from that of registered and aided households. Note that these figures considered only households that sold cane in 2021 but not those that were growing cane and had not yet sold. Including those households would make it less likely that any expected positive association between registration or registration with aid and household food security is found. These results should thus be interpreted with caution. Does women's involvement in sugarcane production and marketing decisions influence household food security outcomes?Households where male heads control decision-making with regard to all parcels had relatively better food security measures or a cane grower household. The HFIAS was 5.57, HDDS-6.74 and MAHFP-10.37. All parcels/plots owned by a household on which decisions made were solely by female heads/spouse had the worst food security outcome measures with HFIAS-8.45, HDDs-5.79 and MAHFP-10.1. The key insights here are that female heads/spouses are highly vulnerable to food security for both cane and non-cane growers. Also, partly, when women make decisions on parcels the productivity/income is lower due to gender gaps in access to resources/services etc beyond decision making. However, cane growers' decision making on plots by female heads had seemingly better food security outcomes to those by their fellow counterparts who do not grow cane.We next estimate a Poisson model of HFIAS that includes all the explanatory variables in prior specifications, with the addition of a dummy variable that =1 for households where a female head/spouse has the final say on cropping choices. The statistically significant partial effect of this dummy indicates that HFIAS increased by 4.1 for households that have strong women's influence on crop choice (p<0.1) (Table 12, Col.2). This is equivalent to a 55 percent increase in this measure of food insecurity. This result is not only the opposite of our expectation, but of such a large magnitude that it suggests that the female head/spouse cropping choice dummy may be capturing some of the negative association of female headedness with household food security -a common finding.The unexpected negative association between strong female head/spouse influence on crop choice and household food security may be because this dummy variable measure of women's influence has a high correlation (-0.75) with the dummy that =1 if household head is male (similar correlation but positive if a female head dummy is used). Because female headed households have lower levels of landholding and assets on average, they would be expected to have relatively lower food security outcomes compared with male headed households. It is possible that the high correlation between a male or female head dummy and the measure of women's influence implies that the negative sign on the women's influence dummy may be picking up part of the negative relationship between female headship and food security outcomes. If we instead use a decision dummy that =1 for households where a female head/spouse has the final say on allocation of crop sales income (Table 14, Col.2), the partial effect is again statistically significant though the magnitude is somewhat smaller at 2.9. This indicates that decision-making on crop sales income allocation held by a female head/spouse is associated with an increase in food insecurity. Again, this result may be confounded by very high correlation between the decisionmaking dummy and a dummy for the gender of head of household. To attempt to alleviate this challenge, we create an alternative measure of women's influence on cropping. This is the share of household area cultivated from plots where a female head/spouse has the final decision on cropping choice. The results of using this share variable are similar to those above. Further exploration of this result is beyond the scope of this paper. As noted above, this may reflect the fact that most households where a female head/spouse has the final decision on crop marketing (or crop choice or allocation of crop sales income) are female headed households -and their lower asset base on average and few household income earners typically means that their household food security outcomes are lower than those of male headed households. We try another alternative and define the dummy for female head/spouse final decision-making to -1 if a female head/spouse has that authority on one or more household plots (not all of them, as with analysis above). While the results were slightly different, there is still no evidence that households where a female head/spouse has the final say on intra-household decisions on crop choice, crop marketing, and/or allocation of sales crop income is associated with better food security outcomes.Households with cropping choices are made jointly by the head/spouse (for all plots) have 1.27point higher HFIAS (higher food insecurity), relative to the base (households where a male head makes the final decisions on crop choice). There are no statistically significant associations between the decision dummy and HDDS or MAHFP outcomes in Table 12 or Table 14. Finally, we consider decision-making on allocation of crop sales income, which has a different result in that \"Mixed decision making\" --where decisions on all plots owned by a household were not taken by either a male head/spouse, a female head/spouse, jointly by husband/wife, and by othersis associated with an HFIAS that is 2.1 points higher than those of households where male heads make this decision (Table 15). This study investigates the relationship between farm household participation in sugarcane production and food security within the main sugarcane-growing sub regions of Buganda, Busoga and Bunyoro of Uganda. The study also investigates whether women's influence in intra-household decision-making in crop production and marketing is associated with better food security outcomes.The study used three measures of food security, the HFIAS, HDDS and MAHFP and descriptive and econometric regression analysis of primary data collected from 1,171 households in these regions, including 983 cane growers and 788 non-growers.Econometric analysis finds that sugarcane growing households were 17 percent less food insecure, on average, relative to non-cane growers, as measured by the 27-point Household Food Insecurity Access Scale (HFIAS) -while controlling for a variety of household and community-level factors known to influence household food security. Cane growing was also associated with one additional month of adequate household food provision (MAHFP), an improvement of 10 percent compared with non-cane growers. No significant association was found between cane production and HDDS, a measure of household food security. This analysis also found that households in Buganda subregion had better food security measures compared to those in Busoga and Bunyoro subregions, with Bunyoro faring the worst. The severity of food insecurity using the HFIAS was high among non-cane growers, though it declines in total household asset value and ownership of large animals. Households with less than two acres and those with less than 4 acres in non-cane and cane growing categories were severely food insecure, on average, for each of the three food security measures.Other factors positively associated with food security outcomes included a household growing more than one food crop, one or more household members with salaried employment, and higher levels of maximum adult female education in the household, household assets, and number of live animals. Female education levels may promote better food security outcomes as more educated female adults may be more likely to adopt improved crop production technology and farming practices that can promote higher farm income and household food security. The value of household assets (a form of saving) increases household resilience to adverse shocks on household food production and incomes and thus also affect food security (Kirimi, ud).Factors negatively associated with food security status include household size (as measured by household Adult Equivalents), residence in Busoga and Bunyoro subregions relative to Buganda, having a member in wage employment, and female-headed household status. Larger households have higher consumption needs, often have higher dependency ratios, and are typically found to have lower food security, controlling for other factors. Other studies such as Mathenge et al (2010) found that a majority of female heads in Kenya were widows, who face great difficulties in terms of access to land and other assets and with increased care-giving responsibility within the family.Results of testing whether a household female head/spouse has strong influence in intra-household decision making on crop choice or crop marketing show that is no evidence to support the expectation that a household with a female head or spouse with strong influence on crop choice or crop marketing. That said, this result may be due to high correlation between this women's influence measure and a separate gender or the household head dummy, which must remain in the model.Policy reform and active public sector oversight of the sugarcane industry is within the mandates of the Ministry of Agriculture and the Ministry of Industry, Trade, and Commerce, and they must coordinate with each other and sector stakeholders to improve coordination between growers and millers. Following a difficult period of poor grower-miller coordination from 2018 to 2021 and a crash in 2021, regaining and maintaining the benefits of cane production for food security requires reforms to the industry's policy environment and public sector governance (Mbowa et al, 2023).Reforms are needed to promote improved coordination between growers and millers and consideration of related income stabilisation mechanisms for sugarcane farmers. In addition, extension services should promote sugarcane production on farms with 8 or more acres only, farmer adherence to maintaining food crop cultivation, and use of productivity-enhancing technologies in cane and food crop production.There are several ways in which Ugandan policy makers can help to maintain the income and food security benefits of sugarcane growing in the country. First, Ministry of Trade, Industry and Cooperatives (MTIC) and the Ministry of Agriculture, Animal Industry and Fisheries (MAAIF) should provide a policy environment and public sector oversight of institutional arrangements between millers and growers that promote higher cane productivity and profitability for growers and more reliable market assurance for both growers and millers. Such arrangements would facilitate stronger coordination of local cane supply and demand, grower access to improved inputs, market assurance for both growers and millers, and a transparent and fair process for millers and grower association representatives to negotiate a cane purchase price each season, based upon a set formula (Mbowa et al, 2023). This level of coordination and oversight can only be achieved if MTIC and MAAIF take a more active role in governance and oversight of the cane industry and grower-miller relations (ibid, 2023).Second, MAAIF extension services should promote sugarcane production on farms of no more than 8 acres, the adherence of cane growers to maintaining at least part of their cultivated area in food crops, and the adoption of productivity-enhancing crop technologies by cane producers for both cane and food crops. Productivity improvements in both cane and food crops in cane growing regions are vital to enable farm households to improve both their household incomes and food security.Appendix Table A2: Household dietary diversity score (HDDS) and food insecurity access scale (HFIAS) by gender of household head and cane grower status, % ","tokenCount":"13654","images":["1883826526_1_1.png","1883826526_24_8.png","1883826526_24_9.png","1883826526_24_10.png","1883826526_24_11.png"],"tables":["1883826526_1_1.json","1883826526_2_1.json","1883826526_3_1.json","1883826526_4_1.json","1883826526_5_1.json","1883826526_6_1.json","1883826526_7_1.json","1883826526_8_1.json","1883826526_9_1.json","1883826526_10_1.json","1883826526_11_1.json","1883826526_12_1.json","1883826526_13_1.json","1883826526_14_1.json","1883826526_15_1.json","1883826526_16_1.json","1883826526_17_1.json","1883826526_18_1.json","1883826526_19_1.json","1883826526_20_1.json","1883826526_21_1.json","1883826526_22_1.json","1883826526_23_1.json","1883826526_24_1.json","1883826526_25_1.json","1883826526_26_1.json","1883826526_27_1.json","1883826526_28_1.json","1883826526_29_1.json","1883826526_30_1.json","1883826526_31_1.json","1883826526_32_1.json","1883826526_33_1.json","1883826526_34_1.json","1883826526_35_1.json","1883826526_36_1.json","1883826526_37_1.json","1883826526_38_1.json","1883826526_39_1.json","1883826526_40_1.json","1883826526_41_1.json","1883826526_42_1.json","1883826526_43_1.json","1883826526_44_1.json","1883826526_45_1.json","1883826526_46_1.json","1883826526_47_1.json","1883826526_48_1.json","1883826526_49_1.json","1883826526_50_1.json","1883826526_51_1.json","1883826526_52_1.json","1883826526_53_1.json","1883826526_54_1.json","1883826526_55_1.json","1883826526_56_1.json","1883826526_57_1.json","1883826526_58_1.json","1883826526_59_1.json","1883826526_60_1.json","1883826526_61_1.json","1883826526_62_1.json","1883826526_63_1.json","1883826526_64_1.json","1883826526_65_1.json","1883826526_66_1.json","1883826526_67_1.json","1883826526_68_1.json","1883826526_69_1.json","1883826526_70_1.json"]}

data/part_2/0103869539.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"80e27fc172a6dfa58e114d9291522d7a","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/44a2b31a-061c-4e7e-825b-869a54633e4f/retrieve","description":"Informal vendors are a critical source of food security in African cities and play a key role in food system transformation. However, the livelihoods of these traders and the governance constraints they encounter are not well-understood outside of primate cities. This study focuses on two distinct secondary cities in Nigeria – Calabar in the South-South geopolitical zone and Minna in the Middle Belt region. Local and state officials in each city were interviewed on the legal, institutional, and oversight functions they provide within the informal food sector. This was complemented with a survey of 1,097 traders across the two cities.","id":"-2081858841"},"keywords":[],"sieverID":"05b5dfc3-c54a-431e-825d-aab18522f938","pagecount":"2","content":"Informal vendors are a critical source of food security in African cities and play a key role in food system transformation. However, the livelihoods of these traders and the governance constraints they encounter are not well-understood outside of primate cities. This study focuses on two distinct secondary cities in Nigeria -Calabar in the South-South geopolitical zone and Minna in the Middle Belt region. Local and state officials in each city were interviewed on the legal, institutional, and oversight functions they provide within the informal food sector. This was complemented with a survey of 1,097 traders across the two cities. 1    The analysis highlights two main findings:1. Informal traders report less harassment by government actors than has been observed in larger Nigerian cities. However, the enabling environment is characterized by benign neglect. Government-mandated oversight functions are not comprehensively implemented, and service delivery gaps hinder food safety. 2. There are differences in the needs of traders across cities. Policies focused on food safety and the livelihoods of food traders need to be properly nuanced at subnational level.The informal economy constitutes approximately 80 percent of Nigeria's workforce. Retail employs around 40 percent of these workers. Food retail represents a major sub-sector within informal trade and includes both stationary traders within open-air markets and itinerant hawkers who operate from various locations. As in many developing countries, traders in Nigeria work in high risk 1 A detailed discussion of this research can be found in NSSP Working Paper 59, The Enabling Environment for Informal Food Traders in Nigeria's Secondary Cities: https://www.ifpri.org/publication/enabling-environment-informal-food-tradersnigerias-secondary-cities and unfavorable conditions in which they are frequently exposed to hazardous working environments. In turn, this has negative implications for the safety of the food they sell.Most research on informal food trade and food safety in Nigeria focuses on Lagos, the country's primate city. However, this study expands our understanding by focusing on two secondary cities that play significant roles in regional economies: Calabar and Minna. The two cities vary by geopolitical location, ethnic and religious composition, and the regulatory framework for informal trade. An array of officials with distinct responsibilities at both state and local government levels regulate informal food trade through hygiene training, licensing, and other oversight.Two forms of primary data collection were used. Semi-structured interviews were conducted with two dozen state and local government area officials involved in regulatory and oversight activities relevant to informal food trade in each city. Secondly, a survey of informal food sellers in nine markets within each city was conducted -530 respondents in Calabar and 567 in Minna (Figure 1). The sample was stratified between those vendors located within markets and those trading on the pavement or roads outside markets, since the latter theoretically are more vulnerable to harassment while those inside can provide a better assessment of market infrastructure.Over half of the traders sell fresh foods (e.g. fruits, vegetables, fish, meats), about one-third sell manufactured packaged goods, and the remainder prepared snacks and meals. Women are the major traders in Calabar, while men predominate in the more heavily Muslim Minna. More than half finished secondary school or completed postsecondary schooling. Almost 80 percent of traders had at least one parent who also was a trader, suggesting such trade is a family tradition for many.In addition, the survey revealed that harassment by officials, either through forced relocations or confiscation of merchandise, is not perceived by traders to be a major challenge. Only 12 and 3 percent in Calabar and Minna, respectively, reported ever experiencing such harassment. However, in contrast to statements by officials in both cities regarding their responsibilities in regulating informal food trade, less than one percent of the sampled vendors realized that they needed a license to sell food, and 78 percent had never had an inspection of their food handling or food quality by a health officer in the previous six months. There is a large gap between the de-jure roles of local bureaucrats in regulating traders and actual oversight. Overall, traders reported little interaction with officials on food safety.Traders' efforts to vend safe food are further undermined by the poor service environment in which they operate (Table 1). Few have access to clean running water, safe storage, or health facilities. Toilet access is a concern in Minna and inadequate drainage in Calabar. The lack of access to major services is even more pronounced for itinerant vendors, who often are not entitled to use the services within the markets.Despite the recognized importance of traders to urban food security, very little is known about this relatively heterogenous constituency. We found that, instead of the draconian treatment witnessed in megacities such as Lagos, informal vendors in Calabar and Minna operate in a situation of benign neglect by the government. Regular sensitization of traders to food preparation practices is critical, as is the investment by local governments of collected revenues into relevant infrastructure that will enable proper food safety practices.Recognizing the subnational variation among food vendors is critical to ensure that policy responses for food safety, gender inclusion, taxation, market siting, and other issues are supportive of traders' livelihoods and of food safety for the public. More importantly, as urbanization proceeds apace in Nigeria and informal food trade continues to be essential to the food security of the urban poor, the subnational bureaucratic institutions that regulate the safety of informal food trade need to be provided sufficient fiscal and human resources to exercise their oversight functions. ","tokenCount":"900","images":["-2081858841_1_1.png","-2081858841_1_2.png"],"tables":["-2081858841_1_1.json","-2081858841_2_1.json"]}

data/part_2/0111960796.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"fe6ea9e07044cfd3ec3b96110acba961","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/a91bf921-ec51-4acb-b0c6-2d34f4bc3ab3/retrieve","description":"Nigeria and many other sub-Saharan countries are rinding that adoption of new high-yielding varieties is severely constrained by existing farm input supply and food-marketing systems. Like other sub-Saharan countries, Nigeria imports most of her modern inputs because of the nearly complete absence of domestic manufacturing capabilities. Availability of these inputs to farmers depends on procurement and internal distribution efficiency. Unlike many other African countries, Nigeria in recent times has not had to depend on price taxation of crops to secure revenue. Yet output-marketing systems continue to put severe constraints on output growth.","id":"1824299122"},"keywords":[],"sieverID":"4cf8517e-4ba1-479a-8eb7-ab8b9d6d287a","pagecount":"14","content":"Nigeria and many other sub-Saharan countries are rinding that adoption of new high-yielding varieties is severely constrained by existing farm input supply and food-marketing systems. Like other sub-Saharan countries, Nigeria imports most of her modern inputs because of the nearly complete absence of domestic manufacturing capabilities. Availability of these inputs to farmers depends on procurement and internal distribution efficiency. Unlike many other African countries, Nigeria in recent times has not had to depend on price taxation of crops to secure revenue. Yet output-marketing systems continue to put severe constraints on output growth.Annual imports of inputs have varied considerably. Over 1970-78, the coefficients of variation of physical imports of ammonium sulphate, single superphosphate, and NPK were 31 percent, 29 percent, and 29 percent, respectively. The comparable instability indexes for CIF prices of fertilizers are 15 percent for 1960-70 and 44 percent for 1970-80. Nigeria receives most of her farm input imports from a few countries; as a result, instability in these countries is easily transmitted to Nigeria.Variability in farm input imports not only is a disincentive to production but creates output fluctuations. It may be due to several factors: fluctuations in foreign exchange earnings; instability of national government's input policies; inefficiencies in import arrangements for farm imports, whether public or private; and fluctuations in supply conditions in countries that export to Nigeria. Instability in supplies of farm inputs has undesirable consequences for input distribution and marketing. These include (1) poor markets for public and private investments in marketing infrastructures, such as warehouses for fertilizers, pesticides, and farm machinery; (2) the continuing underdevelopment of farm input, marketing, and credit institutions; (3) less investment in market promotion, resulting in delays in realization of marketing economies of scale and slow growth of markets; and (4) frequent and sudden revisions of tariff and nontariff restrictions of farm input trade.In 1976, the federal government centralized procurement in the Fertilizer Procurement Unit of the Federal Ministry of Agriculture. This step was taken to relieve state governments of the financial burden of fertilizer imports and subsidies, to end the chaos resulting from separate orders by state governments, and to reap the quantity discounts available on large orders. Initially, the federal government subsidized 75 percent of the costs of fertilizer importation, port clearance, and transportation, with farmers paying the remainder. States bore the costs of internal transportation. Federal government subsidies were later reduced to 50 percent, while states were to pay 25 percent of the subsidy plus the costs of internal distribution. In May 1983, the federal government reduced overall public subsidies on fertilizers for 1984 to 50 percent-35 percent federal and 15 percent state. Now, when fertilizer shipments arrive, contracts are awarded for inland transportation from the southern areas. In most states, the fertilizer is delivered to the fertilizer depot at the state capital, though in a few cases the fertilizer is delivered to four or five zonal depots. The state government contracts transport from the state capital to stores in the local government headquarters. From there it is hauled to village and district stores in state government and local government trucks. Most states distribute directly from state central warehouses to local government areas, using a variety of government, parastatal, and private agents to distribute supplies to farmers.The system has several drawbacks. State projections of requirements are often low, since they are usually based on the previous season's figure adjusted by an annual increment. Fertilizer imports sometimes arrive many months after recommended application dates. Table 13.1 shows that even in northern projects assisted by the World Bank, the percentage of total deliveries arriving on time is low.Because of untimely deliveries, fertilizer applications are well below recommended dosage per hectare and therefore insufficient, ineffective, and wasteful (Falusi and Adubifa 1975). Enormous storage and warehousing bottlenecks and high costs are incurred when fertilizers arrive long after recommended application dates and thus must be stored for the next season. In Lagos State, for example, closing stocks were 90 percent of average total stocks handled during 1980-81. Farmers incur losses from failure to realize the full yield potential of new varieties that require fertilizers for maximum yields. Since farmers are unsure of the fertilizer supply situation for the next season, they tend to ration their current supply.The fertilizer subsidy scheme during 1976-82 introduced inefficiencies into the procurement and distribution system. With demand growing and the quantity of imports and distribution limited by the size of the subsidy budget, farmers frequently pay prices higher than the subsidized price-and possibly as high as it would have been without subsidies. The policy of uniform subsidized prices throughout the country has discouraged private firms and cooperatives from participating in distribution beyond the state capitals, except under contract. Participation without refunds from government for transportation and other costs would result in heavy private losses. In the agricultural development projects assisted by the World Bank, for example, fertilizer is hauled to distant farm service centers at great financial cost to the project (Nigeria 1978). Finally, centralization of control has meant that no fertilizer is imported by the private sector except on contract for the government. Consequently, no fertilizer is available from the private sector, even at the unsubsidized price.Storage facilities at all levels have generally been inadequate, especially in years of abnormally high imports, such as 1981. This has caused either spoilage or large financial outlays for rental of temporary storage facilities. Transportation facilities of state ministries have been grossly inadequate in many states, and communication between state capitals and remote villages has been poor.Other problems with fertilizer distribution include late release of funds by government; poor fertilizer packaging, especially from the Kaduna superphosphate factory; inadequate or practically nonexistent soil-testing programs; ineffective extension coverage; unsuitable fertilizers for particular ecological zones; lack of adequate warning from the ports about fertilizer delivery schedules; and failure of many contractors to accompany fertilizer consignments transported by rail, resulting in losses from improper handling.Pesticides have long been used in Nigerian agriculture, especially for cocoa, cotton, and groundnuts. Importation and distribution have remained largely in private hands. Problems include an inadequate network of depots in the main producing areas, small markets for particular pesticides, and market entry barriers arising from market dominance by a few multinational firms. Other difficulties include price collusion and stocking and inventory control problems that result from inefficiencies of the pesticides subsidy policy.Cocoa pesticides were the first pesticides to be used on a large scale in Nigeria. Their use was subsidized beginning in 1959, yielding substantial benefits to farmers, given significant own-price elasticities of demand for this input (Idachaba 1976;Idachaba and Olayide 1976). However, the scheme encountered administrative and operational problems.Announcements of subsidized prices payable by farmers effective April 1 each year were often delayed (Idachaba 1981). The decisionmaking process on subsidies involved three major steps. First, the Western Region Ministry of Agriculture and Natural Resources invited quotations on current unsubsidized prices from pesticide supply firms. During 1965 to 1975, these letters were dispatched in January, only three months before the subsidy year began. The second step was the interministerial meeting on the subsidy scheme, typically held in March. The third step was executive council approval, usually given during June. In only 1 of 11 years was executive council approval given before the April 1, budget year and in three of the last four years approval was not given until July. Because of these delays, supply and distribution firms were unwilling to bear substantial risks on contracts for forward delivery and placed inadequate orders for overseas supplies. There also was diversion of pesticides from high-subsidy to low-subsidy states when price differentials exceeded transportation and marketing costs.Finally, the effectiveness of pesticide subsidies was severely limited by the high cost of sprayers (Dlamini 1975). The first subsidies on sprayers were granted only recently, over 30 years after the pesticide subsidy program began.Human labor supplies over 98 percent of farm power in Nigerian agriculture. Substitution of cheaper sources of farm power is urgent to expand cropped area and to intensify production. In addition, the relative prices of machine and human power have changed as a result of minimum wage laws and the inflationary consequences of the oil boom, making labor-saving devices relatively more attractive.The supply and distribution of farm implements is plagued by many problems including lack of spare parts, supply of inappropriate farm machinery by government-appointed contractors (many of whom know little about farm machinery), grossly inadequate repair and maintenance capability, and poor execution of government tractor hiring and other subsidy schemes.Inadequate credit has curtailed the growth of farm input markets, particularly farm machinery, and consequently has curtailed the capitalization of Nigerian agriculture. There are two major instruments for supplying farm credit. First, there is the National Agricultural and Cooperative Bank (NACB), established by the federal government to disburse farm credit at subsidized rates to the agricultural sector. Second, there are directives from the central bank to the commercial banks on minimum lending quotas and central bank guarantees for 75 percent of loans by commercial banks to the agricultural sector. Credit generally has not been available to small-scale farmers, who produce the bulk of marketed food supplies. Spatial coverage and lending strategies of the NACB have been poor, and credit has been monopolized by absentee farmers, retired civil servants, and soldiers.Most farm credit, therefore, comes from informal sources, both noninstitutional and institutional. Noninstitutional sources include friends, relatives, neighbors, licensed buying agents (produce merchants), moneylenders, and traders. Institutional sources include cooperative societies, farmers unions, esusu (credit) societies, and organized age groups.Four features of these informal sources of credit are pertinent. First, loan processing time is very short, as assessment is based less on the viability of the proposal than on personal relationships and knowledge of the borrower's background. Second, interest rates, not often explicit, are high enough to cover risks of loan defaults, especially as no marketable collaterals are issued. (For example, moneylenders simply have borrowers sign for amounts higher than they actually borrow, to cover interest. Buying agents contract to buy all of a farmer's future crop at a predetermined price, from which the loan is deducted, or take the future crop as collateral, in which case the lender exercises usufruct rights over the crop until the loan is repaid. The implied interest rates are quite high.) Third, there is little or no loan supervision. Fourth, debt-servicing charges are sometimes high. (Moneylenders, for example, often hire people to harass potential loan defaulters.)Produce marketing cooperative societies are constrained in their credit activities by limited funds from the commercial banks, by limited grassroots membership, and by management and manpower deficiencies, which make membership loyalty difficult to sustain. Informal credit is typically small scale and is inadequate for market infrastructural investments. In fact, credit from the licensed buying agent tends to entrench the produce and input marketing system.The National Seed Service (NSS) was launched in 1976 to coordinate seed multiplication efforts throughout the country and to produce foundation seed for distribution to certified seed production agencies. The NSS works with research organizations and promotes development of manpower resources, seed certification and testing, and interstate marketing. Supply and distribution have been hampered by slow development of seed multiplication programs by the states; by inadequate allocation of investment resources; by inadequate drying, processing, and storage facilities; and by rapid turnover of trained staff.In the states, supply and distribution from state seed multiplication units have been hampered by uncertainty in funding of these efforts; by multiple ministerial responsibilities of project managers of the units; by lack of seed drying, testing, and storage facilities; and by lack of training facilities for lowlevel employees.Seed marketing through noncommercial channels, such as the extension arms of state ministries of agriculture, has failed to supply farmers the right quantities of seeds at the times they need them most (Idachaba 1980b).From colonial days to the oil boom of 1974, the economy was characterized by a dual food-marketing system. One arm was the relatively well-organized commodity export system, whereby state produce-marketing boards (and, later, regional produce-marketing boards) bulked farm produce for both the domestic and export markets. In 1976, given the inflow of revenues, the regional marketing boards were replaced by commodity boards for cocoa, groundnuts, cotton, palm produce, rubber, and grains. These boards have been relatively effective only in those commodities for which private outlets were nonexistent or strictly controlled. Cocoa and cotton are examples.The old marketing boards were fairly efficient in getting the agricultural surplus to ports for export, despite notable abuses by some licensed buying agents, responsible for procuring scheduled crops. The boards and the Central Nigerian Produce Marketing Company had a network of storage and warehousing facilities and maintained a system of grades and quality differences in conformity with world market standards.A guaranteed minimum producer price scheme was launched in 1976-77, by which the government, through the new commodity boards, serves as buyer of last resort. Actual purchases have been small, because announced prices have been consistently below domestic market prices. In fact, there does not appear to be any firm basis for determining levels of guaranteed producer prices. There was probably no alternative, since the commodity boards had neither the storage capacity nor the procurement facilities to purchase all that was offered, especially commodities produced across many agroclimatic zones.The traditional food-marketing system comprises thousands of private sector itinerant traders, assemblers, transporters, processors, commission agents, and space arbitragers. This system services the needs of consumers in both villages and major urban centers. There is evidence that the system has been relatively efficient in integrating the markets of various locations, in providing storage facilities, and in equilibrating supply and demand (Hays 1975).However, it has worked efficiently only within the limited context of segmented regions, rather than as a truly integrated national market. Rural feeder roads are not all-weather roads, which has made most of rural Nigeria inaccessible in the rainy season. Few transporters and assemblers reach remote sources of food supply or demand. Market information services are rudimentary. There is little reliable information on crop outlook, supply and demand balances, prices in distant markets, stocks and inventory in major supply sources and markets, or new technology and methods of application. Areas of surplus coexist with areas of need because of grossly inadequate market information. Most rural markets do not have concrete floors and even fewer have lockup stalls. Virtually none have storage facilities. Urban markets, usually multipurpose, are generally congested and lack sanitation, storage, and parking facilities.Cooperatives are involved in food marketing, although they are more prominent in marketing export crops such as cocoa, cotton, and tobacco. Serious cooperative involvement in farm input and food marketing in Nigeria is hampered by administrative, financial, and management difficulties (Idachaba 1982).The 1973-74 drought dramatized the need for a strategic grain reserve for national food security. This led to a program aimed at building 250 thousand tons of grain reserves by 1980, an amount roughly equal to 2.5 percent of annual production.Although traditional storage (in rhumbu-granaries made of earth or woven straw) for millet and sorghum in the northern states has generally been efficient with respect to cost and product quality, efforts to increase food production will strain these facilities. Average storage capacity per rural household needs to be raised, and insect and disease control need to be intensified. Modern storage technologies have not been developed for roots and tubers, and greater production is expected to result in major storage and marketing problems. Similarly, storage of maize in the south poses more problems than in the north because of higher moisture. Efforts have been made to develop improved cribs, but storage losses are still substantial.The sharpest increases in agricultural production in Nigeria up to 1970, as in most of sub-Saharan Africa, occurred in export crops, mainly through expansion of smallholder cultivation onto virgin land. This was feasible during most of the 1940-70 period, given farm labor costs, off-farm opportunity costs of farm labor, the relative prices of labor substitutes such as machines, and relative farm producer prices.Major changes in the technology of smallholder agriculture also occurred, including pesticides and new seed varieties in cocoa production; new seed varieties and insecticides in cotton production; and use of seed-dressing chemicals, new seed varieties, and phosphatic fertilizers in groundnut production. Export prospects were good, especially with the tight world market for oilseeds in the 1940s and 1950s. The real constraint in exports was on the production side-supply constraints for purchased inputs were more important than output marketing or labor constraints.In the postcolonial era, there have been three constraints on domestic food production. One, the swelling urban sector has caused soaring demand for marketed food and raised the cost of agricultural labor supply. Two, smallholder agriculture has been handicapped by disorganization in the food-marketing system. Three, domestic food production does not attract international capital and management as export crops do. Although production of food has risen, domestic food prices have soared and food imports have increased. To improve food marketing, Nigeria introduced marketing parastatals and guaranteed producer price schemes. On the input side, the government subsidized inputs and became directly involved in input marketing.The Funtua agricultural development project assisted by the World Bank is an example both of the power of improved input supply and the constraint imposed by a poor marketing system. Maize was introduced as a new crop. With new fertilizer-responsive seed varieties and increased availability of fertilizer, maize production rose from 300 tons in 1975-76 to 57,300 tons in 1979-80. The bumper crop in the latter year threatened a market collapse. The Nigerian Grains Board, buyer of last resort at the guaranteed minimum producer price, failed to purchase maize, and Funtua ADP management had to purchase directly. The same experience was repeated at the Ayangba ADP.During drought, the imperfect food market and farm credit systems combined to force farmers to sell short: they were induced by unaccustomed high farm gate nominal producer prices to part with farm produce in response to unusually active farm gate competitive bidding, only to buy produce back later at even higher prices. Rural areas were thus depleted of marketable surplusesresulting in a paradoxical situation, in which food prices in some terminal markets were lower than in rural primary markets.Thus, since 1970, the shift in public emphasis from export crops to food production had been associated with a shift in the relative importance of different constraints. On the input side, labor has become a more important constraint than purchased inputs, although the problems with fertilizer deliveries show that input delivery systems remain a great problem. On the output side, market institutions to handle surges in food production have been so inadequate as to pose a threat to food growth strategy.Naturally, constraints of input supply and output marketing differ among states and regions, depending on the levels of development of rural physical infrastructures, degrees of ruralization and urbanization, and the degree of development of the rural institutions. Marketing constraints in regions with higher degrees of urbanization appear less important than farm input supply constraints. The opposite is true in regions with more rural areas, emphasizing the importance of infrastructure.Whether more emphasis should be placed on input or output subsidies is an empirical question. An earlier study demonstrates that adjustment lags are shorter with input prices than with output prices and that input subsidies are to be preferred, given reasonably assured markets for crops (Idachaba 1976). However, in the case of fertilizer, central procurement and the large subsidies that constrain the quantity imported have restricted total quantities on the market, thus hampering growth of usage (Idachaba forthcoming). Farm input subsidies combined with central procurement and distribution introduce disincentives for private sector participation in input distribution. A crucial issue for policy is how to pursue pricing objectives in these areas while using the private sector to maximum advantage. Note: Adjustment of the exchange rate by one-third and keeping farm prices the same would raise the subsidies 4 to 6 perenctage points for fertilizer and 10 to 15 percentage points for pesticides. The subsidy rates for mixed fertilizer (15-15-15) and for Perenox pesticide are similar to those for the inputs shown. \"Nl = $1.54.Despite an overvalued exchange rate, which reduces the real value of nominal subsidies, imported inputs have been heavily subsidized in Nigeria in real terms in recent years, as shown in table 13.2. This has resulted in the substitution of imported inputs for other inputs on a much larger scale than would otherwise have been the case. Farm input subsidies in Nigeria need to be analyzed within the context of income and resource transfers between sectors, oil-based distortions, and the theory of second best (Idachaba 1974). In this context, subsidies have a useful role in overcoming fundamental distortions of incentives between agriculture and nonagriculture, distortions simply beyond the capacity of agriculturally oriented policy to change.In the colonial economy, crop taxation provided a significant part of regional government revenues. This provided incentives for local governments to boost crop production-typically, through subsidies. Over the years, the need to set priorities for subsidies among crops has become much greater. Subsidy priorities should be determined according to the crops that the inputs are typically used on; crops with high priority are those that contribute significantly to the country's food self-sufficiency objective, that minimize the country's food import dependence, that contribute substantially to the country's caloric and protein intake, or that contribute substantially to the country's export earnings capability. Yet fanners are often able to use subsidized inputs on other crops, defeating the policy objective. Furthermore, a policy to subsidize an input may require rationing, if foreign exchange constraints limit imports.Availability of appropriate fertilizers is the key to realization of the yield potential of new seed varieties. Improved maize seed and fertilizer are close complements, for example. Yet seed and fertilizer deliveries have not been synchronized. Spraying against cocoa pests was not readily adopted because there was no matching subsidy on sprayers-and credit was not available. Defects in the supply and distribution of pesticides have resulted in heavy losses of the productivity gains from fertilizers. Accelerated application of fertilizers encourages weed growth and raises the possibility of mass pest infestation. An integrated approach to farm input supply and distribution is necessary, but policy responsibilities for different inputs are often formulated and executed in isolated fashion in different government agencies.The desirable role of parastatals in farm input supply depends on the size and density of markets. Another factor is the inability of private firms to recoup their investments in input research and infrastructure. Therefore, in the absence of public involvement in these functions, there is underinvestment in agricultural extension and input supply and distribution infrastructure.As the market for farm inputs grows, market size and market differentiation increase, and the divergence between private and social returns to agricultural extension decreases. Thus in the transition from traditional to modern agriculture, there is a tendency to rely more on the public sector for farm input distribution. However, the ability of a conventional line ministry to handle input distribution is seriously constrained. It is often unable to pay wages that would attract staff with required skills. Usually it does not operate as a selfaccounting unit with ability to attract outside sources of finance and to engage in wholesale purchase of large quantities of required inputs. With quantities of inputs thus limited, officials often use nonmarket criteria for allocating available inputs among farmers. These officials become so preoccupied with the logistics of distribution that they neglect technical aspects of new inputs and do not utilize the most effective methods of disseminating new technology.Parastatals, on the other hand, have considerable potential for an expanded role in distribution of farm inputs to small farmers. Being removed from regular civil service methods and procedures, the new generation of parastatals, such as the Kano State Supply Company, have the potential for making the quick decisions required for timely operations. They are also in a position to pay slightly higher salaries. However, performance of the new parastatals has been constrained in several ways. Officials of the supervising ministry find it difficult to divorce themselves from habitual procedures and processes. Boards of parastatals are filled with politicians, who see their main duty as overseeing contract awards for specific projects rather than the development of policies. Finally, there is a tendency for parastatals to rely permanently on government subsidies.With the projected commercialization of Nigerian agriculture, private sector firms are expected to handle farm input distribution in the long run. Although such firms have better decisionmaking processes, there are still some drawbacks. Market structures for some inputs, such as farm machinery, pesticides, and fertilizers, are dominated by a few multinationals. Such structures affect market conduct and performance, especially with regard to price collusion. Firms also tend to concentrate input distribution in areas that promise quick return, even if large sections of the rural sector, including whole regions, are left behind in the development process. Finally, the inadequacy of private credit facilities at the grassroots level has meant that only the rich and well-to-do can patronize these firms, thus widening inequalities of access to income-earning opportunities.A promising avenue of reform is the creation of parastatals, discussed above. These would facilitate the performance of private firms but would not engage directly in farm input supply and food marketing. Such institutions would provide services such as market information and coordination of pricing policies to minimize policy-induced distortions within the national economy.The wider the applications of an input, the greater the coordinating role of the facilitating institutions.A second approach to reform is structural. It includes dramatically increased resource allocations for rural roads, on-farm storage, marketing infrastructure in rural and urban markets, rural processing facilities, and structural changes in the provision of credit. Removal of barriers to entry will accelerate the flow of marketing services, which will help to improve market structure, conduct, and performance. This approach to marketing reform requires a gestation period, involves less drama, and does not command the same enthusiasm from politicians and policymakers as, say, the announcement and inauguration of a board of directors for a new parastatal!Despite efforts by the government to establish farmers' cooperatives and farmers' unions, there is limited involvement of agricultural cooperatives in farm input distribution. At times the required investment has been too great, as in the case of large farm machinery. Or the market for new farm inputs has not been large enough to absorb initial investment outlays over a reasonable area, as was the case with the Cooperative Supply Association and cocoa pesticides. Many cooperatives cannot provide the credit for seasonal farm inputs in rainfed agriculture and, consequently, tend to favor consumer items with steady yearround demand. Finally, transportation and other marketing costs of inputs are not reimbursed by either purchasers or the government.The accelerated cooperative input distribution (ACID) program proposes cooperatives as the primary outlet at the village level; a revitalized network of farmers' credit and marketing cooperative societies to provide the credit facilities to finance small-scale farmers' input purchases; a modified training, management, and visiting system, which insures two-way communication; and a unit to coordinate cooperatives engaged in farm input distribution (Idachaba 1982).The Nigerian experience highlights certain development problems in input and output marketing. Direct farm input procurement and distribution by the government has not been successful. Efforts to provide physical, social, and institutional infrastructure that would help private sector firms and cooperatives to perform better would have been more productive. Government-sponsored parastatals can play an important part in facilitating such efforts, but more research is required to determine the role such institutions should play. It is likely that this role will be larger in remote areas.Even though input subsidies may have been necessary in the early stages of transforming traditional agriculture, they can be a serious bottleneck to development, especially those provided by centralized procurement and distribution branches of a government ministry. If farm input subsidies are needed, procurement and distribution should be left to private firms, parastatals, and cooperatives, with the relevant ministry overseeing the public interest.Empirical work needs to be done on the criteria for determining priorities among input and output subsidies. Issues concerning decisionmaking by government hierarchies are also important, as is work on the proper sequence of applying policy instruments. Empirical evidence is needed on the bureaucratic processes underlying execution of public policy to identify major bottlenecks, implementation lags, and distribution of benefits. Cost effectiveness of interventions must be taken into account.More research is needed on the conceptual framework and specific methodology for fixing guaranteed minimum producer price levels. It serves little purpose to set guaranteed producer prices that are significantly above domestic market prices when the logistics of purchase and warehousing capacity are not in place. This will only cause farmers to lose confidence in government price incentive schemes.","tokenCount":"4770","images":["1824299122_1_1.png","1824299122_2_1.png","1824299122_3_1.png","1824299122_4_1.png","1824299122_5_1.png","1824299122_6_1.png","1824299122_7_1.png","1824299122_8_1.png","1824299122_9_1.png","1824299122_10_1.png","1824299122_11_1.png","1824299122_12_1.png","1824299122_13_1.png","1824299122_14_1.png"],"tables":["1824299122_1_1.json","1824299122_2_1.json","1824299122_3_1.json","1824299122_4_1.json","1824299122_5_1.json","1824299122_6_1.json","1824299122_7_1.json","1824299122_8_1.json","1824299122_9_1.json","1824299122_10_1.json","1824299122_11_1.json","1824299122_12_1.json","1824299122_13_1.json","1824299122_14_1.json"]}

data/part_2/0136612928.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"ec34c5dc7afda4425e2e3a9c2cb17297","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/7a289003-3697-4c45-abc5-eeb883be8536/retrieve","description":"","id":"-721069691"},"keywords":[],"sieverID":"676e832b-dc9f-43da-b8c4-5ba12cfa2b18","pagecount":"6","content":"Growth that is shared-so-called inclusive growth-is now widely embraced as the central economic goal for developing countries. But definitions and empirical characterizations of inclusive growth vary widely. In this brief, inclusive growth is defined as growth that builds a middle class. Macroeconomic policies can shape the environment and incentives for inclusive growth in three important areas: fiscal discipline, the more rule-based the better; a \"fair\" fiscal policy with respect to revenues and expenditures; and a business-friendly exchange rate. Although these policies are not underlying causes of growth, they are conducive to growth. The analysis here relies heavily on the experiences of the mostly middle-income countries in Latin America.In the past several decades, pro-poor growth emerged as a gentle counterpoint to a singular concern with growth alone (measured in terms of increases in per capita income), while implicitly recognizing that growth, if not always sufficient for poverty reduction, is almost certainly necessary. Inclusive growth includes and extends pro-poor growth, on the grounds that growth that is good for the large majority of people in developing countries is more likely to be economically and politically sustainable. Sustained growth matters because many low-and middleincome countries that have had long growth episodes-of 8 to 10 years-have subsequently suffered prolonged growth collapses before achieving real gains in human development and general well-being.Is there a meaningful distinction between macroeconomic policies conducive to pro-poor growth and those conducive to inclusive growth? Sound fiscal and monetary policies that are propoor are also likely to be good for the middle class, but whether that commonality extends to medium-term tax, expenditure, and transfer policies is less obvious. In the case of macroeconomic shocks, middle-class small business owners or semi-skilled workers may face greater relative losses of permanent income than poor subsistence farmers.In the end, the possible tensions or trade-offs between strictly pro-poor and more inclusive \"middle-class\" growth policies cannot be generalized. They must be assessed policy by policy in each country and are likely to change over time as circumstances change. Policymakers in developing countries (and their international supporters and advisers) should more systematically consider weighted welfare outcomes when selecting and fine-tuning macroeconomic policies, rather than relying solely on unweighted growth outcomes or overly weighted poverty outcomes. Where there are no trade-offs, all the better. The medium-term benefits of good macroeconomic policy for building a middle class argue all the more for what are sometimes painful macroeconomic decisions in the short run.Inclusive growth implies an increase in both the proportion of people in the middle class (implying that some people exit poverty) and the proportion of total income they command (implying gains at the \"expense\" of either the initially poor or the initially rich). The middle class is here defined as including people at or above the equivalent of US$10 a day in 2005 and at or below the 90th percentile of the income distribution in their own country. This definition implies some absolute and global threshold below which people are too poor to be middle class in any society, and some relative and local threshold above which people are, at least in their own society, \"rich.\" With this definition of middle class, an increase in that group's size and economic power is likely to signal that the underlying growth is based on wealth creation and productivity gains in private activities and is thus self-sustaining and transformative as opposed to being driven largely by exploitation of natural resources, remittances, or infusions of external aid. Figure 1 shows the economic  command of the middle class, so defined, for selected countries in 1990 and 2004. There is no obvious association between the change in the size of the middle class and change in the Gini coefficient over that period. Middle-class growth is associated with a rising Gini coefficient in China and a declining one in Brazil and India. The same is true for other measures of inequality (not shown). Latin America has for decades been the region with the highest inequality in the world. Latin America also has a history-until about 1990 in most countries-of high inflation, high public debt, volatile monetary policy, and, in part because of inflation, overvalued exchange rates. But in the past few years Brazil and Mexico have experienced substantial declines in poverty (using the US$2 a day poverty line), notable declines in income inequality (measured by the Gini coefficient), and a doubling of the proportion of people and of income in the middle class (Figure 1). This sharply contrasts with middle-class growth reversals in Argentina, Ecuador, and Venezuela. One reason may be that the latter countries have hewed less closely than Brazil and Mexico to standard International Monetary Fund/World Bank macroeconomic policies. Ecuador and Venezuela, with their dependence on oil exports, may have been more vulnerable to currency appreciation, which tends to be unfriendly to increasing employment and small business development. Middle-class growth in Brazil and Mexico, where macroeconomic policies have markedly improved in the past decade, suggests that eventually-with a long lag-better macroeconomic policy (combined with a benign external environment and a commodity boom) can contribute to inclusive growth.The discussion here assumes that developing countries will continue the trend of the past two decades of maintaining or increasing their openness (though more cautiously with respect to capital) in an effort to fully exploit the potential benefits of integration into the global economy.But because more open economies are more vulnerable to global financial and other shocks, and because the integration process produces losers (at least in the short run) as well as winners, maintaining good macroeconomic policy in an open economy can be politically difficult. The challenge is even more complicated where the middle class is relatively small and has little command over total income (and where it is heavily made up of households dependent on state and state-protected sectors). It is, for example, the secure middle class in mature market economies that is most likely to support policies that favor openness, maintain price stability, and help ensure a competitive exchange rate. In contrast, the poor and near-poor (living on less than US$10 a day) are at more risk of losing out with integration because they generally lack sufficient education or financial assets to exploit global good times and are vulnerable in global bad times. They also have sufficient political voice to generate self-defeating populist pressures or, in immature democracies, to support short-term patronage arrangements that betray their longterm economic interests.Countries with purchasing power parity income of less than US$1,500 or so per capita have virtually no middle class by the definition used here because daily income per capita at the 90th percentile is below US$10. That is the case for India (shown in the figure) and most countries of Sub-Saharan Africa. Many such countries are highly dependent on aid, which can account for as much as 40 percent of all government spending. The discussion of macroeconomic policies in the next section does apply to them, but the tradeoffs may in some cases be more difficult. Heavy aid inflows can, for example, complicate efforts to limit real exchange rate appreciation that could undermine expansion of small business. Donors as well as country policymakers should ensure that aid flows include adequate support for key investments in power, ports, roads, and other infrastructure. By reducing costs, these investments can help avoid pressure on the exchange rate, allowing for expansion of small business and increasing competitiveness in manufacturing, agro-industry, and services for export.Fiscal Discipline-the More Rule-Based the Better Developing countries, especially those with a bad history of inflation and poor debt management, need to accumulate a credible record of good fiscal management if they are to ensure growth that is inclusive. Most emerging markets and lowincome countries have dramatically improved their macroeconomic management since the early 1990s. They are accumulating \"good\" history. To lock in good history now requires institutionalizing a budget process that is transparent and rulebased, thereby ensuring that habits and citizens' expectations, as well as legislation and regulatory systems, support fiscal policy conducive to inclusive growth. Examples of good rules are legal ceilings on indebtedness relative to gross domestic product (GDP); a truly independent source of published estimates of revenue and expenditure; rules to lock in additional fiscal effort during booms; and for natural resource-rich countries, fiscal contingency funds that set aside unexpected revenue. Countries where the middle class is large and growing are more likely to have the political support for adherence to such rules, in what could be a virtuous cycle of inclusive growth and good rule-based fiscal policy.With the exception of Chile, most countries in Latin America have run fiscal deficits for years and still do (Table 1). Past fiscal laxity meant governments either printed money, fueling inflation, or issued large amounts of debt, driving interest rates to onerous levels. The resulting inflation hurt poor people because of their limited capacity to protect their earnings, for example, through indexed savings. High interest rates also undermined the growth of a middle class by limiting the expansion of creditworthy small firms (which generally have no alternative to the local market for their financing needs) and thus of private investment and of jobs for the unskilled and semi-skilled. Fiscal indiscipline is no longer the rule in Latin America. Average inflation fell from close to 600 percent in 1990 to just over 7 percent between 2000 and 2006. But past high borrowing means that debt service is still high. This debt must be financed, reducing the scope for new public expenditures (Table 2). In 2003 Brazil was spending 10 percent of its GDP on interest on its public debt. To the extent the debt stock must be rolled over (which depends on the extent to which overall spending can be reduced to pay off debt), public borrowing will keep interest rates higher than otherwise, crowding out private investment and job creation. Real interest rates were very high in Latin America in the 1990s, reaching more than 10 percent on average for most countries, compared with 6 percent on average in Southeast Asia and about 5.6 percent in the United States. Since 2001 interest rates have fallen against a backdrop of fairly low inflation in most Latin American countries, but they remain well above those in other regions. Of course some public debt (to finance small deficits) is reasonable, especially when economic growth ensures that the ratio of debt to GDP does not continually rise above a safe range. But emerging market economies with a history of inflation and volatility (including some outside of Latin America such as the Philippines, Thailand, and Turkey) should probably meet a tougher standard of net public debt to GDP than the standard for developed countries-the IMF suggests no more than 30 percent for emerging markets. History hurts in another way. Given high existing debt, Latin American and other developing-country governments determined to avoid new bouts of inflation have had to maintain tight fiscal policies in the past decade in several cases even in the presence of primary surpluses as high as 4 and 5 percent of GDP-that is, fiscal surpluses net of interest payments (Table 3). This situation has reduced the fiscal space for public investment in roads, schools, health care, police training, and so on-services on which the poor rely heavily. In an unhappy combination, past high public borrowing in Latin America may be contributing to the crowding out of private investment while high primary surpluses to finance debt service on current and past borrowing may be reducing public investment compared with countries in East Asia (Table 4).  Another consequence of past fiscal indiscipline is the inability to implement countercyclical fiscal spending during economic downturns. During recessions in developed countries, governments increase spending on unemployment, food stamps, and other safety net programs. The resulting increases in public spending protect the poor and help insulate the middle class, while helping to stimulate a sluggish economy. Such countercyclical measures, however, rely on the confidence of (domestic and external) market creditors in the government's willingness and ability to honor the new debt and on the local financial sector's ability to absorb new debt. Except for Chile, countries in Latin America have not been tested on this score since the 2001 debt crisis in Argentina.In short, Latin American countries are still paying for fiscal indiscipline that mostly ended more than a decade ago. With the recent global economic boom, most have grown fast enough and kept overall fiscal deficits low enough to get ahead of the destructive debt dynamic in which the burden of past debt undermines aggregate growth. But continued progress relies heavily on more years of very tight fiscal policy (unless growth rates jump to Asian levels) and perhaps too heavily on a continuation of an unusually benign external environment, particularly for commodity producers.Fiscal probity also helps limit the volatility that hurts the poor and the productive middle class. The poor and middle class gain less during booms and are the first to lose jobs during busts (those who already have real and financial assets gain most). When volatility leads to financial crisis, it also involves inequitable wealth transfers that create enduring adverse distributional effects. Evidence from the financial crises of the late 1990s in Asia and Latin America shows that many poor and middle-income households did not recover assets they liquidated during severe downturns.Volatility is the outcome of many factors, including fluctuations in commodity prices and foreign capital inflows over which governments in developing countries have limited control. Latin America's past patterns of stop-and-go spending (driven sometimes by periods of populist governments) have been a factor too, however. By accommodating this fiscal indiscipline, monetary policy further undermined investor confidence, raising interest rates and limiting job creation. The region's stronger fiscal position today, along with more flexible exchange rates and improved financial regulation and supervision, bode wellbut the recent calm may also rely mostly on ample global liquidity (itself at risk at this writing in August 2007) and buoyant export markets.Inclusive growth requires not only keeping aggregate spending in line with aggregate revenues, but also adhering to generally progressive tax systems and expenditures. The experience in Latin America is discouraging. The value-added tax, which is generally regressive, accounts for 60 percent of total revenue in Latin America, compared with 30 percent in Europe. More progressive and higher overall taxation in Europe reduce income inequality, and probably the burden on the middle class, much more than in Latin America, where loopholes and exemptions tend to reduce the tax burden on the rich, and tax evasion is rampant. Finally, high payroll taxes discourage job creation, hurting the poor and middle-income groups more than the rich, whose income comes relatively more from capital. Revenue generation averages just 18 percent, well below what might be expected given average per capita income. Low revenue generation combined with admirable fiscal discipline constrains public investments and expenditures that could otherwise be deployed to reduce inequality and induce more inclusive growth. Equally to the point, more visibly fair tax systems would not only encourage inclusive growth, but also make higher ratios of taxes to GDP more politically acceptable, including to the rich who now easily justify evasion (more efficient public spending and less corruption would also have this effect). In Argentina the effective average tax rate for the top 10 percent of households was estimated at 8 percent in the late 1990s. (In Africa the problem is heavy reliance on trade and other indirect taxes; relatively high taxes on imports raise input costs for businesses and keep consumer prices higher than otherwise.)Greater spending-on health, education, and public infrastructure-as long as it is minimally efficient, is one key to more inclusive growth. Experience in Latin America also shows that the most inefficient, non-inclusive spending occurs in poorly designed and politically driven pension programs. In Latin America, the richest quintile of the population receives on average about 60 percent of net pension benefits (full benefit amount received minus total contributions), whereas the poorest quintile receives only 3 percent.A competitive exchange rate is helpful to inclusive growth because success in manufactured exports is almost always associated with investment in new enterprises and creation of jobs for the semiskilled-in Japan and then Korea and Taiwan in the 1950s and 1960s, and more recently in China, Mauritius, and Vietnam. When Latin American countries monetized their high fiscal deficits, the results were inflation, persistently overvalued exchange rates throughout the 1970s and even in the 1980s, and excessive borrowing. Countries then attempted to protect local industries via tariffs and other barriers, reducing competitiveness. Over the past two decades, Chile, with a longer history of fiscal rectitude, has been best able to manage its exchange rate to limit appreciation.Fiscal discipline does not guarantee a competitive exchange rate. Governments can get away with high deficits while avoiding currency appreciation if, as in India until recently, capital markets are closed, private savings to finance public debt can be captured, growth prospects are especially good, and people have confidence in the currency. But in most developing countries, maintaining a competitive exchange rate is likely to help ensure inclusive growth. The increase in the size of the middle class in urban China (Figure 1) is the outcome of multiple factors, including the country's undervalued exchange rate. In Brazil and Mexico, the slaying of inflation in the early 1990s has made it easier to avoid overvaluation, which hurt exports for the two prior decades.The middle class in all economies depends on a stable macroeconomic environment. Economic volatility-due to high fiscal deficits, poor monetary policy, unsustainable public borrowing, undervalued exchange rates that temporarily make imports cheap, and inflation-is bad for the incipient middle class. The experience of mature Western economies suggests that poor people benefit when an economically strong middle class insists on accountable government and supports universal and adequate public services, by paying taxes. That experience suggests that inclusive growth as defined here will benefit poor people both directly and indirectly, by helping them escape poverty. Perhaps it is not a coincidence that the two countries in Latin America that have sustained cash transfer programs for the very poor are two where the ranks and economic weight of the middle class have doubled. It is hard to imagine that this would have been possible without more than a decade of sustained, tough fiscal and other macroeconomic policies.","tokenCount":"3042","images":["-721069691_1_1.png","-721069691_1_2.png","-721069691_6_1.png","-721069691_6_2.png","-721069691_6_3.png","-721069691_6_4.png","-721069691_6_5.png","-721069691_6_6.png"],"tables":["-721069691_1_1.json","-721069691_2_1.json","-721069691_3_1.json","-721069691_4_1.json","-721069691_5_1.json","-721069691_6_1.json"]}

data/part_2/0139280594.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"ac44ff372ee330ef5cbd27ed139aaa21","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/c5aa46a4-f3c3-4a06-ba74-43c93156bb34/retrieve","description":"as part of the IFPRI-Egypt Seminar Series- funded by the United States Agency for International Development (USAID) project called “Evaluating Impact and Building Capacity” (EIBC) that is implemented by IFPRI.","id":"1784896329"},"keywords":[],"sieverID":"8f3151d8-6cf9-430c-b9bf-e4e1649030c2","pagecount":"24","content":"International Food Policy Research Institute (IFPRI) Dec. 2018Why Egypt• Allows Egypt to connect established and emerging markets.approx. 100 Million, 50% of which are below the age of 30.• Almost 1 million new consumers are added to the Egyptian market each year.Egypt is the second largest signatory to trade agreements in the world. GAFTA, COMESA, AGADIR, QIZ, PARTNERSHIP AGREEMENT EU,…etc.Trade Agreements   As per Article 2 of the NFSA Law, NFSA's mandate is \"achieving the requirements of food safety in a way as to preserve human health and safety\".The National Food Safety Agency of Egypt ( NFSA) is an open and independent authority to protect consumers' health and consumers' interests by ensuring that food consumed, distributed, marketed or produced in Egypt meets the highest standards of food safety and hygiene. The new regulations will reduce the waiting period for obtaining industrial licenses to establish new facilities from 600 days to between 7 and 30 days.Investors in Egypt have long complained of lengthy waiting times for obtaining approvals, with the World Bank ranking Egypt number 122 of 190 countries on its 2017 Doing Business Index, partly because of difficulties obtaining permits and licenses According to the new law, the waiting period for 80 percent of industries will be reduced to one week or less, while the remaining 20 percent will require about one month due to their higher risks to health, environment, safety, or security, or a combination of these factors Established in 1958, the CFI is a non-profit organization that acts as the official representative of the Egyptian F&B industry.The CFI has a member base of 7700 companies and has a platform that represents the industry, solving and developing every aspect of the business environment that may affect the industry's competitiveness on a domestic and global scale.FEC is to be a lean institution acting as a think tank, catalyst and facilitator for all actions and policy advocacy needed to continuously and dynamically enhance export objectives of the Egyptian food industry at the macro level as well as realize of the aspirations of the Egyptian food exporters.International Food Policy Research Institute (IFPRI) Dec. 2018Trends  Now that more women are entering the workforce, they have less time to complete certain tasks such as cooking which has led more households to begin purchasing frozen semi-prepared food products in an effort to cut preparation times. There has also been a growth in the consumption of packaged confectionery (baked specifically) products which could be an indicator that local consumers are looking for an on-the-go alternative to traditional meals, in order to save time. In addition, Egypt has seen an increase in the consumption of packaged milk, growing by approximately 10% in 2016. This could be attributed to an increase in awareness over health hazards related to loose milk. Despite that, the majority of milk consumers in Egypt continue to consume loose unpackaged milk.To become one of the largest 20 economies worldwide.Economic reform Private sector development.Thank you","tokenCount":"491","images":["1784896329_1_1.png","1784896329_1_2.png","1784896329_1_3.png","1784896329_3_1.png","1784896329_8_1.png","1784896329_10_1.png","1784896329_13_1.png","1784896329_15_1.png","1784896329_16_1.png","1784896329_17_1.png","1784896329_18_1.png","1784896329_18_2.png","1784896329_19_1.png","1784896329_20_1.png","1784896329_20_2.png","1784896329_20_3.png","1784896329_20_4.png","1784896329_20_5.png","1784896329_20_6.png","1784896329_20_7.png","1784896329_20_8.png","1784896329_20_9.png","1784896329_22_1.png","1784896329_23_1.png"],"tables":["1784896329_1_1.json","1784896329_2_1.json","1784896329_3_1.json","1784896329_4_1.json","1784896329_5_1.json","1784896329_6_1.json","1784896329_7_1.json","1784896329_8_1.json","1784896329_9_1.json","1784896329_10_1.json","1784896329_11_1.json","1784896329_12_1.json","1784896329_13_1.json","1784896329_14_1.json","1784896329_15_1.json","1784896329_16_1.json","1784896329_17_1.json","1784896329_18_1.json","1784896329_19_1.json","1784896329_20_1.json","1784896329_21_1.json","1784896329_22_1.json","1784896329_23_1.json","1784896329_24_1.json"]}

data/part_2/0152136964.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"ce418462eddcea8cfb5cd0a2ee92959b","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/56b15199-49b0-428d-a4b0-a24a81d50511/retrieve","description":"Talhelm et al. (2014) provided an original rice theory to explain large psychological differences across countries and even within countries and their impact on innovation. However, their findings are subject to the problems of sample bias, measurement error, and model misspecification. After correcting these problems, most findings in the original paper no longer hold. The authors of this paper collected data on collectivism from other sources and linked them with rice areas but failed to find any relationship as predicted by the rice theory. The role of rice farming in shaping cultural psychology and innovations seems to be much more muted.","id":"399775536"},"keywords":["rice theory","individualism","innovation v"],"sieverID":"032298ce-44f3-4663-81f0-82fe75ae3ed1","pagecount":"24","content":"established in 1975, provides evidence-based policy solutions to sustainably end hunger and malnutrition and reduce poverty. The institute conducts research, communicates results, optimizes partnerships, and builds capacity to ensure sustainable food production, promote healthy food systems, improve markets and trade, transform agriculture, build resilience, and strengthen institutions and governance. Gender is considered in all of the institute's work. IFPRI collaborates with partners around the world, including development implementers, public institutions, the private sector, and farmers' organizations, to ensure that local, national, regional, and global food policies are based on evidence. IFPRI is a member of the CGIAR Consortium.Large psychological differences in individualism, analytic thinking, and loyalty/nepotism across countries and even within countries have been noted in the real world and literature for a long time (O'Neill 1973;Kim 1994;Triandis 1995;Vandello and Cohen 1999). Talhelm et al. (2014) offered an original rice theory to explain such a difference. Their story is intuitively appealing. Rice faming is both labor and water intensive. During the peak seasons, farmers adjust their schedules of transplanting and harvest so that they can exchange labor to ease the problem of seasonal labor shortage. Paddy rice requires standing water. Therefore, people in rice regions must collaborate to build elaborate irrigation systems, coordinate water use, and share the cost of construction and maintenance. All these activities involve comprehensive collaboration and coordination, encouraging farmers to form tight reciprocal relationship and avoid conflict. The rice theory hypothesizes that after thousands of years of farming practice, people living in rice-cultivating regions develop a higher degree of interdependence and collectivism than those living in wheat areas.However, it is difficult to test the theory across countries because of inherent differences in many factors, such as language, religion, politics, and climate. The authors chose to test it within China by taking advantage of the homogeneity among the Han population. They surveyed more than 1,000 Han Chinese college students and found that rice-growing southern China exhibits lower individualism and more holistic thinking than the wheat-growing north. By using secondary data at the provincial level, they also showed that people in rice-cultivating areas of China are subject to a lower divorce rate and hold fewer patents for inventions compared to wheat-cultivating areas. Because rice culture emphasizes avoiding conflict and valuing relationships, the lower divorce rate in rice-cultivating provinces provides strong evidence in support of the rice theory. As individualism and analytical thinking are normally tied to creativity, the smaller number of successful invention patents in rice areas in comparison to wheat areas offers additional support to the rice theory. Drawing on the findings of the paper, Henrich (2014) speculated that wheat farming in Europe may have contributed to the industrial revolution. The rice theory is so appealing and profound that Science highlighted the article on the cover page and many media covered the story.The theory is truly original. It sheds new light on understanding the interplay among environment, culture, psychology, and innovations. Given the originality and potential far-reaching impact of Talhelm et al. (2014), it is critical to replicate and validate the findings of the paper. This is the main objective of this paper. We found that Talhelm et al.'s analyses were subject to sample bias, measurement errors, and model misspecifications. After correcting these problems, most findings in the original paper no longer hold. We further collected data on collectivism from other sources and linked them with rice areas but failed to find any relationship as predicted by the rice theory. The role of rice farming in shaping cultural psychology and innovations seems to be much more muted.Talhelm et al. conducted three experiments to measure cultural thought, implicit individualism, and loyalty/nepotism in six sites. Ideally, the sampling size in each province should be proportional to the total population in the province. However, as shown in Table 2.1, the sample distribution is highly uneven. In the experiment on holistic thought, there are a total of 1,019 observations scattered across 28 provinces. Among the sample, Guangdong province commands the largest number of observations (193, or 18.94 percent of the total), more than double its population share. Fujian province accounts for 14.62 percent of the total sample size, compared to its 2.88 percent of population share. In comparison, only 2 individuals were surveyed in Beijing, constituting merely 0.20 percent of the sample, much lower than its population share of 1.53 percent. The experiment on implicit individualism covers 515 subjects. Fujian province dominates the sample with 142 observations, constituting 27.57 percent of the total sample size. By contrast, only 1 and 2 students were surveyed in Qinghai and Shanghai, respectively. The small sample size makes it impossible to draw a strong inference in these provinces. The sample size for the loyalty nepotism experiment is much smaller at 166. There is only 1 observation in three provinces (Ningxia, Shanghai, and Chongqing). Since the share of the cultivated rice paddy area is measured at the province level, the results on the relationship between rice area and the three outcome variables hinges crucially on spatial variations across provinces. The uneven distribution of sample sizes across provinces, in particular the extremely small number of observations in some provinces, may result in spurious estimates. To check if it is the case, we first replicated the results of Talhelm et al. (2014) based on their original sample and then dropped provinces with more than 100 observations in regressions on holistic thought and implicit individualism and provinces with fewer than 5 observations in regressions on loyalty/nepotism.Table 2.2 presents the regression estimates. Regressions R1, R5, and R11 successfully replicate the original regressions for the three outcome variables. The three outcome variables are holistic thought, implicit individualism, and loyalty/nepotism, taken from Talhelm et al. (2014). Holistic thought is measured by a triad task in which participants pair two out of three items through either abstract or functional relationship; individualism is measured as the difference in the circles of friends and self in a self-drawn social network; the variable of loyalty/nepotism measures whether people draw a sharp distinction between how they treat friends and how they treat strangers. The rice variable is defined as the percentage of cultivated land devoted to rice paddies. It is significant in all three regressions with the expected sign. Regressions R3, R7, and R13 repeat the corresponding regressions in regressions R1, R5, and R11, except they drop either the oversampled provinces (more than 100 observations) or the undersampled provinces (fewer than 5 observations). In the regression on holistic thought (R3), the rice variable remains highly significant after dropping the oversampled provinces. With respect to implicit individualism, the significance level for the rice coefficient declines from 1 percent in regression R5 to 10 percent in R7 after removing Fujian from the sample. The result on loyalty/nepotism is more sensitive. After excluding the provinces with fewer than 5 observations, the rice coefficient becomes insignificant, as shown in regression R13. In short, the uneven sampling size distribution does affect the regression results. One key objective of the Talhelm et al. (2014) study is to dismiss two other popular competing theoriesthe modernization hypothesis and pathogen prevalence hypothesis. The authors used per capita gross domestic product (GDP) to measure the impact of modernization and disease rates in earlier years to measure the precontemporary disease prevalence. The paper tested the three hypotheses independently by running three separate regressions. In principle, a more rigorous test should include all three key variables of interest in the same regression and examine their relative importance to the outcome variables. As a matter of fact, the paper later included both rice and per capita GDP when testing the impact of rice faming on innovations.In Table 2.2, we added per capita GDP and pathogens in all the regressions based on both the original sample and the restricted samples. The results are shown in the columns marked with even numbers after the capital letter R. As shown in regression R2, after controlling for the two variables, the rice variable remains significant in the regression on holistic thought based on the whole sample. However, the significance level of the rice coefficient drops to 10 percent. Once the provinces with more than 100 observations are dropped from the sample, the rice variable in the augmented regression (R4) loses its significance. Regardless of the whole sample or restricted sample being used, the rice variable in augmented regressions on implicit individualism (R6 and R8) is insignificant. Also insignificant is the rice coefficient for the outcome variable of loyalty and nepotism as shown in regressions R12 and R14. When evaluating the three competing hypotheses simultaneously, the rice story does not exhibit any stronger explanatory power on the psychological differences than the modernization and pathogens mechanisms. In fact, none of them matters to the observed psychological differences.The experiment on implicit individualism was likely subject to measurement errors. The experiment was conducted as follows, according to Talhelm et al. (2014, 606): \"Researchers measure how participants draw a diagram of their social network, with circles to represent the self and friends. Researchers measure how large participants draw the self versus how large they draw their friends to get an implicit measure of individualism (or self-inflation).\"The difference between the width of self circles and the width of friend circles was used as a measure of implicit individualism. Since the experiments were undertaken in different times and different locations, there might exist systematic differences in survey instruments (pencil and paper) across sites. For example, the width of self circles among the 193 subjects surveyed in the winter of 2011 in Beijing averaged at 1.97. The experiment on 318 observations in other places revealed an average of 1.26. The two mean values are statistically different at the 1 percent level.Even for subjects originally from the same province, scores varied greatly if measured in different places and times. For instance, the subjects from Heilongjiang province measured in the winter of 2011 in Beijing (sample size of 9) reported an average width of self circles of 1.27. In comparison, the average width of self circles among the subjects from Heilongjiang measured outside Beijing was as high as 1.97. The two samples exhibit a large difference (with a p value of .000). The two experiments (sample sizes of 19 and 10) conducted in Hubei province at different times scored 1.29 and 2.12, respectively. The p value testing the mean difference between the two samples is .002. The average width of self circles reported in two separate experiments (sample sizes of 18 and 13) in Hunan province is 1.29 and 2.12, and the difference is statistically significant at 0.2 percent. If there were systematic measurement errors on the width of self circles across space and over time, the variable of implicit individualism (the difference between the width of self circles and the width of friend circles) would automatically have embedded measurement errors, making it incomparable across sites. One way to remedy the problem is to use relative difference between the width of self and friend circles, that is, the ratio of the difference between the width of self and friend circles to the width of self circles.Regressions R9 and R10 in Table 2.2 report the estimates on the new measure of implicit individualism. No matter whether per capita GDP and pathogens are included or not, the rice variable plays an insignificant role in determining implicit individualism measured in a new way. Correcting the measurement errors makes a big difference to the main results.It has been found in the literature that analytical thinking skills are more instrumental for facilitating innovation than is holistic thinking (Witkin et al. 1977). According to the rice theory, people in wheatcultivating areas are more skilled in analytical thinking, whereas those in rice-producing regions are more inclined toward holistic thinking. Therefore, people in wheat-cultivating areas are expected to show stronger creativity than their counterparts in rice-cultivating areas. Based on cross-province patents for inventions in 1995, Talhelm et al. (2014) showed that there is indeed a negative association between the number of invention patents and the share of land devoted to rice paddy.However, the analysis is subject to several flaws. First, the results do not hold for later years. Figure 5.1 plots the accumulative growth rate of invention patents in the rice-and wheat-cultivating areas by year, with the initial year, 1995, as 1. It is apparent from the figure that the number of invention patents in the rice-producing areas has grown more rapidly than that in the wheat areas, contradictory to the prediction of the rice theory. Panel A in Table 5.1 presents the regression estimates for granted invention patents per capita in different years. The first column replicates the results for 1995, as reported in the original paper. Indeed the rice variable is negative and significant, consistent with the prediction of the rice theory. However, when repeating the exercise for 1999, 2005, and 2009, the coefficient for rice is no longer significant for 2005 and 2009. In 2009, the rice variable even turns positive, albeit insignificant. China's patent law was not put into effect, and the State Intellectual Protection Office did not accept patent filings, until 1985. Since it took on average 3 to 6 years to process invention applications, the number of approved invention patents in the first 10 years is rather low compared to later periods. For example, the number of granted invention patents in the period from 1985 to 1995 accounts for about 5 percent total granted invention patents during the period from 1985 to 2009. Specifically, the number of approved invention patents in 1995 amounts to merely 0.6 percent of total patents. So it is less accurate to use the patent number in 1995 as a measure of innovation capacity than numbers in later years.Although the regressions have controlled for per capita GDP, the cross-province analyses are subject to the problem of omitted factors, such as education level and innovation promotion policies. To partly address this concern, we further conducted a regression on a panel dataset covering the same 27 provinces from 1992 to 2009. In the panel regression, we included year fixed effects to control for some systematic shift in macro policies. Because the rice variable does not change over time, we could not directly include province fixed effects. Instead, we assumed a random effect across provinces in the regression. The rice share in the panel regression is not statistically significant.Second, Talhelm et al. (2014) use only invention patents. Although invention patents may embody more technological innovations, doubtless the patents for utility model and design are also the children of creativity. In 2009, invention patents accounted for only 22 percent of total patents, while patents for utility model and design constituted 42 percent and 37 percent, respectively. So invention patents are just a small portion of total patents. It is import to check if the results in panel A are robust to the inclusion of patents for utility model and design. We repeated the previous exercises in panel A by replacing invention patents per capita with total number of patents per capita as a dependent variable. Regressions R6 through R10 in panel A of Table 5.1 display the regression results on the new outcome variable. The rice variable in regression R6 in 1995 is -0.64 and is statistically significant. In 1999, although it is still negative, it loses significance. For the cross-province regressions on 2005 and 2009, the rice coefficient turns positive and significant. The rice variable in the regression on the panel of 1992-2009 is also positive and marginally significant. The positive rice coefficient in regressions R8 through R10 in both panels contradict the rice hypothesis.Third, there is a long time lag, up to six years from submitting a patent, for receiving an approval. In the original paper, the authors control for per capita GDP in the same year as the patents granted. However, the number of granted patents actually represents the innovation level several years back at the time of submission, not the level at the year of approval. The observed positive association between GDP and number of patents may run the other way around-it is the high innovations that contribute to GDP growth. Due to the potential reverse causality, it is problematic to use both patents granted and GDP at the year of approval. Having access to all the patent entries, we were able to compute the number of patents based on the year of submission, and we used it as a dependent variable. We repeated all the regressions in panel A, except for using the new patent variable, and presented the estimates in panel B. The results in panel B resemble those in Panel A. The rice variable does not have a consistent sign across regressions, largely rejecting the rice hypothesis.Fourth, applicants for patents may not come from the provinces where the patents are filed. In China, universities and research organizations filed about 30 percent of invention patents, and many researchers in these institutions came from elsewhere. So the number of granted invention patents in a province may not truly reflect the innovation capacity of the province's indigenous people. As a robustness check, we excluded the number of invention patents filed by universities and research organizations in the outcome variable and reported the regression results in panel B (R11-R15). None of the coefficients for rice is significant in the five regressions. It is even positive in three of the five regressions, in contrast to the prediction of the rice theory.It has been shown in the literature that individualistic countries exhibit higher divorce rates (Lester 1995). Since the rice theory predicts higher individualism in wheat regions than in rice regions, the rice theory also predicts a higher divorce rate in wheat-producing regions because of their strong individualism. The rice paper provides some supportive evidence using data from 27 provinces in 1996. 1 In the paper, the divorce rate is measured as the ratio of divorces to new marriages in each year and province. If the rice theory holds, we would expect to see the same pattern in later years as cultural preferences change slowly.To check this out, we first plotted divorce rates in the rice and wheat regions from 1995 to 2012 in Figure 6.1. In the figure, we normalized the divorce rate by the initial year value. As shown in the figure, the divorce rate in the rice-cultivating areas has grown much faster than that in the wheat-growing areas, contrasting the prediction of the rice hypothesis.  2014), divorce rate is defined as the ratio of the number of divorces to the number of marriages. We normalized the rate by the initial value in 1995.The evidence in the figure is just suggestive as it does not control for other confounding factors. Table 6.1 presents the regressions of divorce rate on the rice variable by controlling for per capita GDP. Regressions R1, R2, and R4 replicate the results in the original paper. The significant rice variable in the regressions on 1995 and 1999 confirms the findings in the original paper. But the coefficient in the regression on 2009 is not significant, unlike what is claimed in the paper. To check the robustness of the results for later years, we run regressions on 2005 and 2012. The rice variable is significant in regression R3 (2005), but not in R5 (2012). It seems that over time the association between rice farming and divorce rate weakens. When pooling all the years together and running a panel regression in R6, the coefficient for rice is no longer significant.Many divorced people get remarried. Thus, using the ratio of divorces to marriages as a measure of divorce rate likely underestimates the true divorce rate. The official divorce rate in China is defined as the ratio of divorces to total population. We repeated the above analyses using the official divorce rate and showed the results in regressions R7 through R12. The basic patterns remain. The negative association between rice area share and divorce rate is significant only in earlier years (1995 and 1999), not in later years. Compared to regression R3, the rice variable is no longer significant in 2005 (R9) when using the official divorce rate. Above all, the legacy of rice farming on divorces fades over time. In the previous sections, we have tried to replicate and verify the main results using data from the original paper. There are alternative ways to measure collectivism and individualism in the literature. It is worth testing the rice theory using alternative measures. Van de Vliert et al. (2013) surveyed 1,663 subjects in 15 Chinese provinces. Their sample size in each province largely corresponded to its population size.Based on a 14-item measure of psychological attributes associated with collectivism, they constructed a composite indicator of collectivism and computed average collectivism scores at the provincial level. They further imputed the collectivism scores for the other 19 provinces in China. They found that the wheat-producing Heilongjiang Province exhibits the highest degree of collectivism, while the riceproducing Guangdong Province scores the lowest in collectivism.Using their collectivism measure in the 15 provinces, we conducted three regressions and reported the estimates in Table 7.1. The first regression (R1) includes only rice as an independent variable. The negative and significant coefficient for the rice variable indicates that rice farming is associated with a lower degree of collectivism, confronting the rice theory. In the second regression, we added per capita GDP. The rice variable remains significantly negative. In the third regression, we further controlled for pathogen. The rice variable is still negative. Because of the small sample size, the results should be read with caution. To ameliorate the concerns on the small sample, we repeated the three regressions by adding provinces with imputed collectivism scores into our sample. The rice variable is now statistically significant in all three regressions. The negative rice coefficient across the three regions directly challenges the rice hypothesis. It has been shown in the psychological literature that family structures have something to do with collectivism (Vandello and Cohen 1999). In regions with strong collectivism, extended families with more than two generations are more prevalent. Vandello and Cohen (1999) and Yamawaki (2012) have used the share of extended families with three or more generations as one of the indicators in constructing a composite measure of collectivism in the United States and Japan, respectively.The China Population Census in 2010 includes detailed information about family structure. We used the share of extended families with more than two generations in total families at the provincial or county level obtained from the 2010 census as a measure of collectivism. For the provincial-level regressions, we followed the same specifications as in Table 7.1. For the county-level regressions, we used hierarchical linear modeling regressions because the rice and pathogen variables are at the provincial level. Table 7.2 reports the regression results based on provincial and county-level data, respectively. First, in regressions R1 and R4, we included only the rice variable. The rice coefficient is not significant. Second, we added per capita GDP in regressions R2 and R5. The rice variable remains insignificant. Last, we further included pathogen. The rice coefficient is largely the same. The per capita GDP variable, a measure of modernization, is negative and statistically significant whenever it is included in regressions. The finding lends more support to the modernization hypothesis rather than the rice hypothesis. The innovative rice theory proposed by Talhelm et al. (2014) has attracted wide media and academic attention. This study attempted to test the robustness of the popular theory. We first used the original dataset from the paper to replicate and verify its main results. We found that the findings are subject to sample bias, measurement errors, and model misspecifications. After correcting these problems, their main findings no longer hold. We further used alternative collectivism measures as an outcome variable and again failed to validate the results. Overall, we found that the rice theory is not as robust as claimed by the original article.The rice theory also offers a few clear testable predictions on the trends of patents and divorce rates between rice and wheat provinces. The rapid transformation in China in the past several decades provides a ground to directly test the predictions. In contrast to the predictions of the rice theory that collectivism and holistic thinking embedded in rice culture inhibit creativity and innovation, it turns out that the rice-cultivating areas, such as the Pearl River delta and Yangtze River delta regions, have witnessed a much faster growth in securing patents than the wheat-producing areas. Moreover, despite the supposed culture of valuing cooperation and collectivism, the rice provinces have experienced a more dramatic spike in divorce rate than wheat regions in the past two decades.The legacy of rice farming on cultural preferences, if there is any, has not shown itself to be as robust as desirable. China is affected by many cultural winds, both modern and historical, and the influence of rice farming is only one. Although the findings presented in Talhelm et al. (2014) seem to be tenuous, the ingenuity of the hypothesis will, it is hoped, continue to spark investigation into causal factors for cultural differences.","tokenCount":"4172","images":["399775536_1_1.png","399775536_13_1.png","399775536_16_1.png"],"tables":["399775536_1_1.json","399775536_2_1.json","399775536_3_1.json","399775536_4_1.json","399775536_5_1.json","399775536_6_1.json","399775536_7_1.json","399775536_8_1.json","399775536_9_1.json","399775536_10_1.json","399775536_11_1.json","399775536_12_1.json","399775536_13_1.json","399775536_14_1.json","399775536_15_1.json","399775536_16_1.json","399775536_17_1.json","399775536_18_1.json","399775536_19_1.json","399775536_20_1.json","399775536_21_1.json","399775536_22_1.json","399775536_23_1.json","399775536_24_1.json"]}

data/part_2/0200819951.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"56118f1418e09fe6a5dd12c4fbd93001","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/9787a69c-1cad-4afb-bab6-6bff617a6504/retrieve","description":"Effective agricultural extension and advisory services are a key component of efforts to achieve sustainable agricultural production, resilient livelihoods, and inclusive economic growth. These are all necessary elements for accelerating Rwanda’s agricultural transformation. Both extension and information and communication technologies (ICT) are important elements in Rwanda’s Strategic Plan for Agriculture Transformation. This paper examines the capacities of public and private agricultural extension agents in Rwanda and their readiness to use ICT in their work—that is, to be digitally equipped—and provides recommendations for enhancing agricultural extension capacities through expanding and effectively using ICT.\nTo examine capacities and readiness, we use a representative survey of 500 public and private extension agents in Rwanda, augmented by qualitative data from a literature review and key informant interviews. To assess agents’ ‘digital readiness,’ we create two indices focused on their digital experiences and attitudes toward digital modernization.","id":"-1106439090"},"keywords":[],"sieverID":"af05cfdd-2c6e-4bf8-bbd4-bc67cd84aeda","pagecount":"17","content":"Effective agricultural extension and advisory services are a key component of efforts to achieve sustainable agricultural production, resilient livelihoods, and inclusive economic growth. These are all necessary elements for accelerating Rwanda's agricultural transformation. Both extension and information and communication technologies (ICT) are important elements in Rwanda's Strategic Plan for Agriculture Transformation. This paper examines the capacities of public and private agricultural extension agents in Rwanda and their readiness to use ICT in their work-that is, to be digitally equipped-and provides recommendations for enhancing agricultural extension capacities through expanding and effectively using ICT.To examine capacities and readiness, we use a representative survey of 500 public and private extension agents in Rwanda, augmented by qualitative data from a literature review and key informant interviews. To assess agents' 'digital readiness,' we create two indices focused on their digital experiences and attitudes toward digital modernization.We find notable differences between public and private sector extension agents in terms of age, sex, educational background, workload, and digital assets. Most frontline extension staff, including farmer promoters and farmer field school facilitators, do not have access to digital assets and do not use any digital tools in their professional or personal lives. However, other public sector and private sector agricultural extension agents not working directly with farmers have much more digital experience.We find that many factors are associated with a readiness to employ digital tools in extension, including age, sex, sector, education, digital asset access, and number of trainings received. These findings suggest that increasing the availability of digital tools among extension agents and continuing to provide technical, functional, and digital toolcentered training will facilitate the integration of such tools into extension work.The results show that technical and functional trainings increase the likelihood of extension staff being ready to integrate digital tools in their work. Thus, keeping staff engaged and informed on the best agricultural extension practices, in general, can also promote digital readiness. Additional focus should be placed on providing community-level workers with the digital assets (e.g., smartphones), other tools, and knowledge to succeed in their work, given that they now do not have much access to these important technologies and information.Overall, we find that different digital tools are already being used by public and private sector extension agents in their work. This suggests that they could be open to new applications or types of tools being integrated into the extension services they provide farmers.Agricultural extension and advisory services, both public and private, include the \"entire set of organizations that support and facilitate people engaged in agricultural production to solve problems and obtain information, skills, and technologies to improve their lives and wellbeing\" (Birner et al. 2009, 342). Extension and advisory services are thus key to sustainable agricultural production, resilient livelihoods, and inclusive economic growth.Scholars have called for \"the new extensionist\" (Davis and Sulaiman 2014), emphasizing the functional competencies of extension staff, such as communication, which are needed on top of technical skills and the right mindset to help farmers cope with complex challenges to their livelihoods and welfare, such as climate change and food and nutrition security. Functional capacities have not been well documented in the academic literature on agricultural extension, especially with respect to the use of digital information and communication technology (ICT) tools (Strong et al. 2014).Digitalization for agriculture and digital tools are considered a game changer for the transformation of African agriculture (CTA 2019). This is because a large amount of data can be captured and tailored cost-effectively for presentation to individual farmers using special applications and tools (CTA 2019). \"E-extension\" is the use of electronic tools to enhance traditional extension methods (James and Raj 2021). Digital extension services offer many benefits, including wider reach, customization, and two-way communication (James 2023). However, more efforts are needed to equip public and private agricultural extension staff and other actors to use digital tools in achieving solutions to agricultural development challenges (CTA 2019).It is important to distinguish between the wider area of \"digital extension,\" which is rather holistic and considers other actors and the policy environment, from the basic use of ICT services and tools by extension officers to support their communication with farmers. Following Spielman et al. (2021, 3), we define ICTs somewhat narrowly as \"technologies related to mobile phones, services, and networks; portable devices; web-based portals, tools, and applications; and the data and information shared through these products and services via technologies as varied as interactive voice response (IVR) systems and satellite imagery.\"ICTs are also important for the implementation of the fourth Strategic Plan for Agriculture Transformation (PSTA IV) of the Government of Rwanda. This plan emphasizes extension and advisory services as its first priority area and states that ICTs are necessary to increase the impact of those services on agricultural transformation (MINAGRI 2018).In support of PSTA IV, the Ministry of Agriculture and Animal Resources (MINAGRI) has sought to enhance extension and advisory services in Rwanda by introducing the Customized Agriculture Extension System (CAES) (MINAGRI 2020).1 CAES calls for, among other features, ICT-supported extension services. It states that \"ICT can revolutionize agriculture in Rwanda\" (MINAGRI 2020: 34), highlighting that ICTs can make important contributions to several components of CAES, particularly service delivery, capacity strengthening, and coordination, monitoring, and reporting.Despite an enabling policy environment and Rwanda's embracing of ICTs, the country's agricultural extension services have generally not taken advantage of the potential of these tools to the benefit of farmers (MINAGRI 2020). Reforming these services to effectively employ such digital tools will require considerable investment in modifying agricultural extension training curricula and making learning modules on the tools widely availableideally through online portals and platforms. Multiple digital extension approaches, methods, and tools will need to be leveraged to augment and strengthen the current extension system. For CAES to succeed at scale, higher quality training and capacity building efforts around digital extension will be needed for all extension agents in Rwanda to effectively make use of these new tools. Pluralism in extension delivery with the private sector playing a significant role in meeting several of the multiple needs of farmers is also a key ingredient for the success of a digital agricultural extension strategy in the country.This paper looks at the capacities of agricultural extension staff in Rwanda and their readiness to use ICTs in their work-that is, to be digitally equipped. The International Food Policy Research Institute (IFPRI) conducted a capacity needs assessment of agricultural extension staff as part of the Feed the Future Developing Local Extension Capacity (DLEC) project in Rwanda. Using a telephone survey of both public and private extension agents, IFPRI collected information from them to use in assessing which of their capacities needed to be expanded or modified to enable them to use digital approaches and tools in providing advisory services to farmers more effectively.Other studies have examined the use of ICT and attitudes toward the use of digital ICT tools in extension in developing countries. Birke and colleagues (2018) developed a conceptual framework identifying factors that contribute to the uptake of ICTs in extension activities. These factors include characteristics of the individual extension agent, such as age and attitude; organizational characteristics of the extension service, such as management structure and infrastructure; and the contextual environment, including working norms and content availability. Birke et al. use the theory of planned behavior (Ajzen 1991) to describe the process through which these factors contribute to an increased recognition of the value of ICTs, leading to their adoption. Birke and colleagues (2018) also introduce an intermediate step in the process of shaping the individual behavior of extension agents with regards to ICTs. This step accounts for the perceived environment-that is, individual attitudes and social norms. For example, young male extension staff with a university education were found to be the most likely to use ICTs in their work, reflecting their training and attitudes. Strong and colleagues (2014), working in the Caribbean, found that extension staff tended to use ICTs for personal reasons and professional productivity, but used more traditional methods to interact with farmers. However, they also found that higher education levels of staff led to an increase in ICT use.Our study adds to this literature by expanding on the tools developed by Strong et al. (2014) to assess 'digital readiness' among extension agents. We draw on survey instruments developed by Venkatesh and Bala (2012) for the conduct of research on interorganizational business and information systems.We used a combination of qualitative and quantitative methods for this study:▪ A review of the existing literature on agricultural extension in Rwanda, drawing on multiple sources. These included, but were not limited to, project reports, grey literature, academic research, and government documents. This literature was examined to obtain analytical insights on the overall objectives of agricultural extension efforts in the country, the capabilities and skills of extension staff, the curricula used to train extension officers, including any digital approaches, and other related topics.▪ Key informant interviews were conducted with 42 experts on agricultural extension in Rwanda. A semi-structured interview guide was used to obtain insights from these experts on the current skills of agricultural extension staff and gaps in those skills.▪ An analysis of primary data drawn from a telephone survey of 500 public and private extension agents on the capabilities, practices, and needs of extension and advisory service professionals. The survey was conducted in 2021. Due to COVID-19 considerations, the survey was deployed as a phone survey rather than in-person.Our qualitative approaches consisted of the literature review and key informant interviews.An inductive approach was used for qualitative data analysis with the data collected being used to form the structure of the analysis without following a predetermined framework. Data were processed manually using thematic content analysis based on a focus-by-question approach. The approach analyzed responses to individual items in the interview guide, identified themes in the responses, and explored consistencies and differences in those themes. The responses were then summarized and parallels drawn.We carried out a literature review of the existing information on agricultural extension in Rwanda to better understand the current context and the best recommended practices for agricultural extension training, the role of youth, and the level of integration of digital technologies in extension service provision. This literature included strategy documents for various sectors, government and project reports, and research literature.For the key informant interviews, a semi-structured interview guide of 16 questions was reviewed by a panel of experts for focus and content validity. This was administered to 42 key informants through face-to-face or online interviews. To identify key informants, purposive sampling was initially used. Snowball sampling then was employed whereby the key informants first interviewed were asked to recommended additional knowledgeable individuals the research team might interview. The key informants came from public institutions that provide extension services (6), nongovernmental organizations (11), educational institutions (4), local civil society organizations (6), financial institutions (4), private extension service providers (3), digital extension providers (3), and private companies (5). Each interview took approximately 30 minutes to complete.The telephone survey of 500 agricultural extension agents (EAs) was conducted in February and March 2021 across all districts of Rwanda. 'Agricultural extension agents' were defined as all individuals who work as agronomists, agricultural extension agents, crop advisers, livestock advisers, and any other individuals who inform and promote good agricultural practices in the public, private, and non-profit sectors in Rwanda. The survey questionnaire contained modules on demographics, workload and time use, training and integration of functional skills, training of technical skills, extension activities, attitudes on technology, information and communication technologies and extension, and salary and benefits.The literature review and key informant interviews suggested that public sector EAs, who are comprised mostly of farmer promoters (FP) and farmer field school facilitators (FFSF), differ notably from EAs working in the private sector or with civil society organizations, all of whom are referred to hereafter as 'Private sector EAs.' For this reason, the sample was stratified by the two sectors, with two-thirds of the sample comprised of public sector EAs (n=333) and one-third of Private sector EAs (n=167).However, among public sector EAs, considerable variation in EA roles exists. Eighty-six percent of EAs sampled from the public sector were FPs, who are volunteer community leaders who received training to serve as farmer-to-farmer extension agents in their own villages, or FFSFs, who facilitate group-based learning (RAB 2021). The other public EAs work in much different roles, including as cell development officers, district veterinarians, and sector agronomists. Given this variability among public EAs, we disaggregated public sector EAs into sub-groups-FPs and FFSFs, and Other public sector EAs. Descriptive statistics show that age, sex, educational attainment, years of employment, and workload differ between the three EA groups (Table 2.1). Source: Authors' calculations using the IFPRI Rwanda Agricultural Extension Agents Phone Survey, 2021. Note: FP = \"farmer promoters\"; FFSF = \"farmer field school facilitators\"; EA = \"extension agent\".The average age of EAs in the private sector is much younger than among FPs and FFSFs. The private sector is much more likely to have female extension agents. Private sector and Other public sectors EAs are likely to have received considerably more education than FPs and FFSFs-notably at the secondary level and above. Positively correlated with age, years of employment as an EA is lower in the private sector relative to the public sector. Finally, the number of topics that the EAs received training on in the last two years, both on technical skills, such as plot preparation, fertilizer application, and the like, and on functional skills, such as organizing farmer groups or establishing a demonstration plot, are similar across all three groups.The access of EAs to digital tools and applications and the use of these resources for extension work also varied across the three groups of EAs (Figure 2.1). The figures show similarly high levels of digital accessibility and usage by EAs in the Other public sector and Private sector EA groups, but extremely low levels for FPs and FFSFs. More than threequarters of EAs in the Other public sector and Private sector EA groups have access to a smartphone and a data plan. However, less than one-quarter of FPs and FFSFs do. Similarly, while more than three-quarters of EAs in the Other public sector and Private sector EA groups use Short Message Services (SMS), WhatsApp, and Google regularly for their extension work, less than one-quarter of FPs and FFSFs use these or any other digital tools. Encouragingly, among EAs in the Other public sector and Private sector groups, other agriculture digital applications are quite popular, with almost three-quarters of EAs in these groups using such applications. Unfortunately, no additional information was collected on what these 'other agricultural applications' entail. However, it is encouraging that many EAs are already using digital applications tailored to agriculture to provide advisory services to farmers. For our investigation of 'digital readiness' among EAs, our methodology expands on research tools developed by Strong et al. (2014). Our survey instrument relied on an approach developed by Venkatesh and Bala (2012) for the conduct of research on interorganizational business and information systems. For our analysis, two indices were created that focused on the digital experiences and digital modernization attitudes of survey respondents.The Digital Experience Index is comprised of 14 questions-eight ask whether the EA has taught various digital tools to others, and six ask whether the EA has recommended various digital tools to the farmers with which they work as a means of obtaining agricultural information (left panel in Annex Table 1). The Digital Modernization Attitudes Index is comprised of ten questions concerning the attitude of the EA to the use of new or digital technologies in their agricultural extension work (right panel in Annex Table 1). The two sets of questions are each used to create respective indices. Each response was weighted equally to compute the indices. For the analysis, the indices were normalized to range from zero to one. The distributions of the two indices are shown in Figure 2.2.Source: Authors' calculations using the IFPRI Rwanda Agricultural Extension Agents Phone Survey, 2021. Observations: 500.The distributions of the indices vary notably. The majority of EAs scored lower on the Digital Experience Index, with more than three-quarters of EAs scoring below 0.5. In contrast, the EAs scored much higher on the Digital Modernization Attitudes Index, with more than three-quarters scoring above 0.5. The two indices are significantly positively correlated with one another, although the correlation coefficient is relatively weak (r=0.28). This suggests some overlap in the attitudes of EAs toward digital modernization as a means to improve the effectiveness of their work, but that these are still distinct indices.Our work expands beyond that of Strong and colleagues (2014) by exploring correlates of these indices. For each index, we regress a set of individual and institutional characteristics on each index value using tobit, or censored regression, models. Given that our dependent variable is an index, we censor possible estimates at zero at the lower bound and at one at the upper bound for more accurate estimates.2 While this analysis is not causal, it does help to identify variables on which policy-relevant interventions can operate to effect change in EAs' attitudes and abilities toward the use of digital ITC tools in their agricultural extension work. Our estimation specification takes the following form:where Age and Sex denote the age (years) and sex (female=1) of the i th EA, respectively; Tenure denotes the i th EA's number of years in service as an EA; Education denotes educational qualifications as a set of educational attainment categories, a binary category of having completed TVET, and an index of the number of technical and functional trainings the EA reported receiving; Sector denotes which EA group an EA is a member of-FP or FFSF, Other public sector, or Private sector; Device is a binary variable for whether an EA owns or has access to a smart device, like a smartphone, laptop, or tablet). ε is the independent and identically distributed error term. The model is estimated with robust standard errors to account for heteroskedasticity in the error term.The literature review made clear that both public and private EAs require a wide set of skills to be effective with the farmers they serve. Qualitative responses from key informants mainly confirmed the findings from the literature review. Respondents generally perceived the need for additional training for agricultural extension staff, especially on newer topics such as entrepreneurship and climate change.While some staff were said to be using ICTs, many respondents noted the need for better digital literacy, especially among certain sub-groups of extension providers such as farmer promoters.However, several informants highlighted that poor internet availability in rural areas and the high cost of certain ICTs may prove important barriers to EAs taking up digital tools for use in their work. Learning to use new ICT tools and the need for a mindset change to adopt ICTs more readily may present significant challenges. Some respondents suggested the need for incentives for EAs to encourage their digital learning.While several informants mentioned the need for digitalizing curricula and other training materials, one noted that online training may not always be effective since there is limited supervision and many EAs will lack the skills to use online platforms effectively. Physical training was viewed as preferrable for field-based staff, especially for hands-on training like the establishment of demonstration plots. Another bottleneck with digitalization is that, while many extension materials have been produced, few are in Kinyarwanda, so they will need to be translated and updated for the Rwandan context. Moreover, the range of learning materials from many sources means that information is often scattered, is not standardized, and may not have been subjected to any quality checks.Several informants mentioned digital platforms as a promising means of linking all extension staff in the country to promote information sharing and to coordinate, standardize, and harmonize the messages that EAs convey to farmers. However, it also was observed that establishing a common platform for agricultural extension in Rwanda will not be easy, given the number of stakeholders involved.Respondents also raised the issue of professionalization in areas such as certification, regulation, formalized extension education, career paths, and a professional association for extension staff. Using the right tools to do the job, including employing ICTs, forms a critical component in developing a professional agricultural extension service.Finally, informants mentioned the need for greater engagement of youth and the private sector in agricultural extension. This is linked to the use of digital tools, as younger people and the private sector EAs were viewed as more likely to use ICTs than EAs in the public sector.Using the agricultural extension agent survey data, a number of correlates were regressed against the two 'digital readiness' indices to see which demographic characteristics and work experiences are most associated with EAs being ready to digitally modernize how they work with farmers. We expected that age would likely be a main correlate of being ready to integrate digital tools into work, as younger individuals are more likely to own or use a digital device effectively. Based on previous research, we assumed that gender would also play a role. Since women tend to have less access to household assets, including digital technologies, we expect them to be less ready to integrate these tools into their extension work. We also hypothesized that EAs who have attained a higher level of education will feel more comfortable using technology, as they likely used or were exposed to digital tools in their later school years.Years of employment and training background may also play a role in an EA's readiness to employ digital tools in their work. We hypothesize that having obtained more training, whether in the form of TVET, technical skill training sessions, or functional skill training sessions, and working as an EA for a longer time may help an EA to feel more comfortable and confident in their skills and position as an extension service provider and ready to integrate new tools into their work.The various sectors in which EAs work will likely have different requirements for the integration of digital tools and processes into the workload of an EA. This may impact the digital readiness of EAs. Finally, an EA's ownership or access to digital tools will likely influence their willingness to integrate those tools into their work. We test all of these hypotheses in the regressions.We do not include a location variable in our regressions because location is strongly correlated with public policies and civil society organization activities and, as such, is collinear with the EA group variable. In addition, the small sample size inhibits us from using location-based fixed effects in our regressions because of the loss of degrees of freedom. We also do not include the number of days EAs worked per season and the remuneration they received in our set of independent variables. These variables are highly correlated with the three EA groups-days worked per season has an r of 0.58, while remuneration received has an r of 0.46.The tobit regression analysis of the two indices is shown in Table 3.1. Age, sex, educational attainment, training received, agricultural extension group, and technology access are associated with an EA's digital readiness to change. Scoring higher on both indices is negatively correlated with age, suggesting that younger EAs are more ready to modernize and incorporate digital tools into their work. The results also suggest that male EAs are more likely to be ready for digital integration than females. Both the age and sex variables are more significant and more strongly associated with the Digital Experience Index than with the Digital Modernization Attitudes Index. This contrast suggests that these characteristics are stronger indicators of the existing relationship of EAs with digital technology (which may be related to their job role and agricultural extension group) rather than to their overall attitudes toward digital modernization in conducting agricultural extension. Additionally, certain education and training opportunities are important correlates of an EA being ready to digitally modernize. Having a secondary or post-secondary education is associated with EAs scoring higher on the Digital Experience Index. Similarly, having received more technical and functional skills training in the past two years, as measured by the Training index variable, is positively associated with both indices. In contrast, the role of technical and vocational education and training (TVET) is not significant for either index. However, this result may reflect only 15 percent of EAs having obtained any TVET certification (Annex Table 2). Being a public sector employee, rather than working in the private sector or as an FP or FFSF, is not associated with an EA being 'digitally ready'. However, EAs in the private sector score significantly higher on both indices compared to FPs and FFSFs. Finally, owning or having access to a digital device, such as a smartphone, laptop, or tablet, is correlated with a higher score on both indices. However, we find that the years of employment as an EA variable is not significantly correlated with either index.Several policy recommendations emerge from this analysis.First, we find that the vast majority of FPs and FFSFs do not have access to any digital assets and do not use any digital tools in their personal or work lives. Though not surprising given the community-level role that FPs and FFSFs play, these results suggest that if the extension system seeks to digitally equip FPs and FFSFs, as well as the other public and private sector EAs, additional focus will need to be placed on providing these communitylevel workers with the digital assets (e.g., smartphones), tools, and knowledge to succeed in their work. The other public sector and private sector EAs are well positioned to provide digital training, given that our survey results indicate that their digital experience is much more extensive than that of FPs and FFSFs. However, the survey suggests that the confidence of EAs to teach others about digital tools varies and is not consistently strong across groups. Therefore, a \"train-the-trainer\" learning approach might be useful to better enable EAs in the Other public sector and Private sector EA groups to provide training on digital tools to FPs and FFSFs.Facilitating peer learning and collaboration among EAs through digital platforms or online communities could be an effective way to share experiences, best practices, and insights on using digital tools effectively. Leveraging the underutilized TVET institutions in the countryvery few EAs in either the public or private sector had received TVET training, according to our survey-may be one way to approach this.It is also important to examine the types of digital tools already being used by the other public and private sector EAs in their work. SMS, WhatsApp, Google, and 'other agricultural applications' are the most popular digital tools used in their extension work by EAs in the Private sector and Other public sector groups. Unfortunately, we do not have information on what these 'other agricultural applications' entail, but it is encouraging to see that EAs are already using applications tailored to agriculture to provide advisory services to farmers. This suggests that they will be open to new applications or types of tools to be integrated into their extension services.We also recommend the introduction of specialized training programs in digital literacy, focusing on fundamental digital tools, software, and applications relevant to agricultural extension work. These programs should be complemented with workshops and seminars on innovative digital tools and technologies, featuring hands-on training and demonstrations by experts. Additionally, EAs could be provided with access to online databases, research papers, educational materials, and collaborative learning events with technology providers, NGOs, and private sector companies so that they can contribute to the development of customized digital solutions to optimize their effectiveness further. The study underscores the need for ongoing support and technical assistance for EAs as they adopt new digital tools to facilitate a smooth transition and to address potential challenges as they emerge.We found that female EAs are less likely than male EAs to be ready to use digital tools in their work. This could be due to a variety of factors, including restricted control of assets by women within their own households. This challenge could be overcome with comprehensive digitalization training for all EAs. Lecoutere et al. (2019) find that such challenges often stem from information asymmetries between men and women in a household. Simply including women in training focused on agricultural extension and digitalization can help to overcome this gender gap.Similarly, our results show that any trainings, whether technical or functional, increases the likelihood of an EA being ready to integrate digital tools into their work. Continuing to keep EAs engaged and informed on the best agricultural extension practices through regular training will also promote digital readiness.In sum, this study provides an overview of the demographic, educational, and work backgrounds of public and private agricultural extension agents in Rwanda. We find notable differences between public and private sector EAs in terms of age, gender, educational background, workload, and digital assets. To assess the 'digital readiness' of EAs, we assess the impacts of various factors on an EA's digital experience and their attitudes toward digital modernization. We find that many factors are associated with an EA being ready to digitize, including age, sex, whether they work in the public or private sector, education, digital asset access, and how much training they receive. These findings suggest that increasing the availability of digital tools among EAs and continuing to regularly provide technical, functional, and, increasingly, digital training will help to facilitate the integration of new digital tools into their agricultural extension work.","tokenCount":"4900","images":["-1106439090_1_2.png","-1106439090_8_1.png","-1106439090_8_2.png","-1106439090_9_1.png","-1106439090_17_1.png","-1106439090_17_3.png"],"tables":["-1106439090_1_1.json","-1106439090_2_1.json","-1106439090_3_1.json","-1106439090_4_1.json","-1106439090_5_1.json","-1106439090_6_1.json","-1106439090_7_1.json","-1106439090_8_1.json","-1106439090_9_1.json","-1106439090_10_1.json","-1106439090_11_1.json","-1106439090_12_1.json","-1106439090_13_1.json","-1106439090_14_1.json","-1106439090_15_1.json","-1106439090_16_1.json","-1106439090_17_1.json"]}

data/part_2/0220721231.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"76305135ab1b6e5e313b399e473ad5f6","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/6479b6b8-fba4-4d83-837f-76af19898639/retrieve","description":"","id":"1431978322"},"keywords":[],"sieverID":"6b334889-c689-470c-b874-41f011e090e7","pagecount":"4","content":"District Nutrition Profiles (DNPs) are available for 707 districts in India. They present trends for key nutrition and health outcomes and their cross-sectoral determinants in a district. The DNPs are based on data from the National Family Health Survey NFHS-4 (2015NFHS-4 ( -2016) ) and NFHS-5 (2019NFHS-5 ( -2021)). They are aimed primarily at district administrators, state functionaries, local leaders, and development actors working at the district-level.  Note: NA refers to data unavailable for a given round of NFHS/Census.  Note: NA refers to data unavailable for a given round of NFHS/Census.• What are the trends in infant and young child feeding (early initiation of breastfeeding, exclusive breastfeeding, timely initiation of complementary feeding, and adequate diet)? What can be done to improve infant and young child feeding?  Note: NA refers to data unavailable for a given round of NFHS/Census.• How can the district increase women's literacy, and reduce early marriage, if needed?• How does the district perform on providing drinking water and sanitation to its residents? Since sanitation and hygiene play an important role in improving nutrition outcomes, how can all aspects of sanitation be improved? • How can programs that address underlying and basic determinants (education, poverty, gender) be strengthened?• What additional data are needed on food systems, poverty or other underlying determinants? Note: NA refers to data unavailable for a given round of NFHS/Census.• How does the district perform on health and nutrition interventions along the continuum of care? Does it adequately provide both prenatal and postnatal services to women of reproductive age, pregnant women, new mothers and newborns? • How has access to health and ICDS services changed over time (food supplementation, health and nutrition education and health checkups)?","tokenCount":"281","images":["1431978322_1_1.png","1431978322_1_2.png","1431978322_1_3.png","1431978322_1_4.png","1431978322_1_5.png","1431978322_1_6.png","1431978322_1_7.png","1431978322_1_8.png","1431978322_1_9.png","1431978322_1_10.png","1431978322_1_11.png","1431978322_1_12.png","1431978322_2_1.png","1431978322_2_2.png","1431978322_2_3.png","1431978322_2_4.png","1431978322_2_5.png","1431978322_2_6.png","1431978322_2_7.png","1431978322_2_8.png","1431978322_2_9.png","1431978322_2_10.png","1431978322_2_11.png","1431978322_2_12.png","1431978322_3_1.png","1431978322_3_2.png","1431978322_3_3.png","1431978322_3_4.png","1431978322_3_5.png","1431978322_3_6.png","1431978322_3_7.png","1431978322_3_8.png","1431978322_3_9.png","1431978322_3_10.png","1431978322_3_11.png","1431978322_3_12.png","1431978322_4_1.png","1431978322_4_2.png","1431978322_4_3.png","1431978322_4_4.png","1431978322_4_5.png","1431978322_4_6.png","1431978322_4_7.png","1431978322_4_8.png","1431978322_4_9.png","1431978322_4_10.png","1431978322_4_11.png","1431978322_4_12.png"],"tables":["1431978322_1_1.json","1431978322_2_1.json","1431978322_3_1.json","1431978322_4_1.json"]}

data/part_2/0236315190.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"e96cc2a7f725874af4a63e74fc455a62","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/b494a219-4361-49ad-91b6-698965d57a70/retrieve","description":"With 7 million borrowers and US$5.4 billion in outstanding loans in 2012, the Viet Nam Bank for Social Policies (VBSP) is the largest single microcredit lender in the world. We measure the impact of VBSP lending and seek to answer the question of whether continued subsidies to the bank, which amount to about 2 percent of the value of its loans, are justified. VBSP grew particularly rapidly between 2004 and 2008, when its share of total loans in Viet Nam rose from 10 to 27 percent, and by 2008 an estimated two-fifths of its loans were ostensibly used for directly productive purposes. Using data from a panel of 1,846 rural households interviewed in 2004, 2006, and 2008 as part of the Viet Nam Household Living Standards Survey, we estimated the impact of VBSP lending on consumption and income per capita, as well as self-employment earnings. Both an intention-to-treat model with fixed effects, and a quantity-of-credit model with fixed effects and using instrumental variables, show significant or close to significant impacts of VBSP microloans on consumption and income, but our data do not have enough power to determine whether this mainly works via agricultural or nonagricultural self-employment income. Without VBSP, the rural poverty rate would have been 0.7 percentage points higher in 2008 than it actually was. The subsidy is likely justified, given the evidence and scale of the positive impact of VBSP loans on consumption spending and the concentration of benefits among poorer households in Viet Nam.","id":"1942183744"},"keywords":["Viet Nam","microcredit","rural poverty","Viet Nam Bank for Social Policy"],"sieverID":"b0e1a860-475a-4d02-8ddc-c25f515b65cc","pagecount":"33","content":"The International Food Policy Research Institute (IFPRI), established in 1975, provides evidence-based policy solutions to sustainably end hunger and malnutrition, and reduce poverty. The institute conducts research, communicates results, optimizes partnerships, and builds capacity to ensure sustainable food production, promote healthy food systems, improve markets and trade, transform agriculture, build resilience, and strengthen institutions and governance. Gender is considered in all of the institute's work. IFPRI collaborates with partners around the world, including development implementers, public institutions, the private sector, and farmers' organizations, to ensure that local, national, regional, and global food policies are based on evidence.Since the 1980s, when Grameen Bank and Bank Rakyat Indonesia showed that lending small amounts of money to poor people was feasible and productive, countries have shown strong interest in what Khandker referred to as \"fighting poverty with microcredit\" (1998). Viet Nam is no exception. In March 1995 it established the Fund for the Poor to oversee subsidized lending to poor households, although most of the operations were handled by the Vietnam Bank for Agriculture and Rural Development (VBARD), which had been established in 1988. Then in 2003, lending operations to the poor were entrusted to a separate institution, the Vietnam Bank for Social Policies (VBSP), freeing VBARD to operate on more clearly commercial lines.The government and international donors-including the World Bank in its rural finance projects of 1996, 2002, and 2008-have strongly supported VBSP. Their efforts have propelled it to become arguably the largest microfinance operation in the world, with 7.1 million borrowers and US$5.5 billion in loans outstanding in 2012, up from 3.7 million borrowers and US$0.9 billion outstanding in 2004. There are also a number of other microfinance institutions in Viet Nam, most notably the people's credit funds. Despite the extensive availability of formal microfinance, informal credit remains important, and even in 2008 the value of household borrowing from friends and relatives exceeded that of borrowing from VBSP.Given the reach of microfinance in Viet Nam, it is perhaps surprising that the most fundamental question has not yet been answered satisfactorily: what impact has it had on incomes, expenditure, and poverty? Although there have been some studies on the subject, they all have weaknesses, which is why Timberg and Binh recently reached the conclusion that \"assessments are needed to determine the impact on poverty of microfinance schemes\" (2011,6).This paper measures the impact of microcredit, and particularly credit from VBSP, on household incomes and consumption in Viet Nam, and then traces the impact on poverty. The study differs from earlier work in three respects:1. It uses a panel of rural households surveyed in 2004, 2006, and 2008, the most recent applicable data, with the advantage of coming from a period when VBSP was expanding rapidly. Previous work has, at best, used only a subset of these data. 2. It pays careful attention to technique, in particular the knotty issue of dealing with selection bias. 3. It seeks to measure the effects of credit directly on incomes and expenditure, but as a robustness check it also looks at the effects on profit and income from self-employment. The rest of the paper provides some background on microfinance in Viet Nam (Section 2), argues that existing studies of the impact of microfinance are insufficient (Section 3), sets out the data and methodology used (Section 4), and comments on the policy relevance of the study (Section 5).Our main interest is in the effect of microcredit on the poor. A relatively recent study (DFC, Mekong Economics, and World Bank 2007) focused on a group it called the \"bottom of the pyramid,\" consisting of the poorest 24 percent of the Vietnamese population (in about 2007), or 20.3 million people living in 4.6 million households. The authors argued that even this group had fairly good access to bankingreceiving 5.3 million loans and accounting for 3.0 million deposit accounts (Table 2.1). By this measure, the country's most important institutional lenders to the poor in 2007 were VBARD and VBSP, with the people's credit funds and other microfinance operations playing supporting roles. At the time, VBSP was weak at mobilizing deposits, a job that was done better by the Vietnam Postal Savings Company and the people's credit funds.  Economics, and World Bank (2007); Bezemer and Schuster (2014). Note: Refers to loans of less than 30 million Vietnamese dong (US$1,432). -= data not available; \"bottom of the pyramid\" = poorest 24 percent of the population.Table 2.1 also shows some more recent information on \"microfinance lending,\" defined as loans worth up to 30,000 Vietnamese dong (VND) (about US$1,432) in 2013. Most of these loans are probably going to relatively poor households. By this standard, VBSP clearly dominates microlending, although VBARD and the 1,130 people's credit funds also lend significant amounts. By one estimate, 94 percent of active microfinance clients were women in 2008 (Bezemer and Schuster 2014, 17), although the figures reported by the Microfinance Information Exchange put the proportion closer to a half (Table 2.4, discussed later). Finally, the figures in Table 2.1 exclude informal sources of credit, which are important.Table 2.2 is based on household-level data from the 2004, 2006, and 2008 Viet Nam Household Living Standards Surveys (VHLSS) and covers a wider range of sources of credit. In each year, 9,189 households were surveyed. The proportion of households borrowing from any source stood at 44 percent in 2004 and had declined slightly, to 40 percent, by 2008. Whereas half of all households in the poorest quartile (as measured by real spending per capita) borrowed in 2004, the proportion had dipped to 47 percent by 2008. This represents a relatively high level of loan penetration: at first sight, credit rationing appears not to be widespread, which would in turn limit the scope for wider credit access to boost production or incomes. Between 2004 and 2008 there was an important shift in the sources of household credit, as loans from VBSP (which averaged about US$400 each by 2008) displaced loans from VBARD and perhaps also loans from individual creditors and from friends and relatives. Over the past decade, VBARD has shifted its focus to lending to larger farmers and to borrowers in urban areas (who are not covered by the VHLSS panel, which includes only rural households): in 2004, 28 percent of VBARD loans were smaller than the average VBSP loan, but by 2008 this proportion had fallen to 20 percent. Almost 9 percent of households had loans from multiple sources, but only about 1 percent of households borrowed both from VBSP and VBARD.One possible explanation for the rising market share of VBSP is its attractive interest rates, which are consistently lower than those charged by VBARD or by informal lenders, and in 2008 were well below the inflation rate (Table 2.3). The high inflation rate observed in 2008 (23 percent) was unanticipated and did not have an immediate effect on reported nominal interest rates; borrowers had locked into their fixed-rate loans before the inflation gave lenders a windfall.  (2004,2006,2008). Note: Expenditure excludes purchases of durables and uses sampling and household-size weights. n.a. = not applicable; VND = Vietnamese dong.The subsidized nature of VBSP credit-an issue we return to below-may also have limited the size of its loans during the period studied, which were only one-third as large as those from the big formal lenders, most notably VBARD and other banks (Table 2.2). As a proportion of consumption or income, borrowing did not rise between 2004 and 2006, but then it rose sharply by 2008 (Table 2.2); this movement is consistent with the very rapid rise in overall credit in Viet Nam during the same period (World Bank 2014), as is evident in Figure 2.1. Table 2.3 shows the level of real expenditure per capita for those who borrow from the different sources; on average, VBSP borrowers are relatively poor-being classified as poor is, in principle, a requirement for a VBSP loan-at least compared with the average borrower from a formal-sector bank or even one of the people's credit funds (which are in effect locally based credit unions).In subsequent sections we make a distinction between informal credit (from individual lenders, friends, and relatives), formal microcredit (all loans worth less than VND 30 million from formal sources, including VBSP), and loans made by VBSP. If one excludes loans from friends and family, VBSP is far and away the dominant source of microlending in Viet Nam. Indeed the country is, by some standards, the largest microcredit provider in the world, as Table 2.4 shows, with more than 7 million active borrowers in 2012, half of whom were women, and with a gross loan portfolio that exceeds US$5 billion. Most of the growth in the number of borrowers took place before 2008, but average loan size has continued to rise and now stands at about US$770.VBSP raises its capital relatively cheaply: only 4 percent of its funding comes from individual deposits, and most of it comes either directly from the government or with government guarantees. Out of VND 129.2 trillion in liabilities as of the end of 2013, 21 percent had been advanced by the national budget, 23 percent came as loans from the State Bank of Vietnam and the national treasury, 20 percent was provided by financial institutions under a mandatory 2 percent contribution program, 23 percent was financed by government-guaranteed bonds, and the remainder came from a variety of other sources (VBSP 2016).The institution is not operationally self-sufficient-which means that financial revenues do not cover financial expenses, operating expenses, and impairment losses (CGAP 2003, 12)-a characteristic it shares with SKS Microfinance and Spandana in India, but one that sets it apart from Grameen Bank or BRAC in Bangladesh (Table 2.4). The annual shortfall is equivalent to 2 percent of assets (Table 2.4).Although VBSP pays its staff well, its operating expenses are modest, at little more than 2 percent of assets, in part because of the large number of loans per staff member. While VBSP has 692 branches and services more than 9,000 transaction points monthly with mobile units, much of its lending is done via mass organizations, including the Women's Union, which administers about half of VBSP loans (Bezemer and Schuster 2014, 17), and the Youth Union. In return for a small fee (0.165 percent of outstanding balances collected on time), these organizations form borrowing groups, often require compulsory savings, and help collect loans when they are due.VBSP's default rates are claimed to be less than 2 percent (Corpuz and Paguia 2008), but some observers believe they are much higher (DFC, Mekong Economics, and World Bank 2007), and there is some evidence that VBSP borrowers do have difficulties repaying their loans (Anh et al. 2011, Table 3.19). VBSP collects relatively few deposits but uses funds provided by the government and by donors to lend to poor households at low interest rates. This support is costly: in 2010, almost two-fifths of VBSP's expenses were covered by government subsidies (VBSP 2013.1   In 2007, VBSP typically charged an interest rate of 7.8 percent on loans (for consumption, for housing, to the poor, to groups), or 9.6 percent for working capital or investment undertaken by small and medium-size enterprises, although by 2010 this latter rate had increased to 10.8 percent. These compared with standard short-term lending rates of 11.8 percent in 2007 and 14.0 percent in 2010 (World Bank 2014). VBSP lends money under 16 distinct programs; as of the end of 2010, the largest proportions of its lending went to poor households (40 percent of the total), to disadvantaged students (29 percent), to household businesses in extremely disadvantaged areas (12 percent), for rural water and sanitation (8 percent), for job creation (5 percent), and for housing for the poor (3 percent) (VBSP 2013. Table 2.5 reports the uses to which rural households reported putting their loans: about half were used for production and investment, and about one-sixth for consumption purposes.  Surveys (2004Surveys ( , 2006Surveys ( , 2008).An estimated 90 percent of VBSP loans in 2007 went to households who were certified by the local People's Committee as being poor. Most VBSP loans have terms of less than three years. As noted above, the loans were largely funded from equity provided by the national budget, governmentguaranteed bonds, and loans from the banking system. Thus VBSP is an externally funded program that channels lending to the poorer groups in rural Viet Nam. Further details on the credit and microcredit system in Viet Nam may be found in BWTP (2008) andDFC, Mekong Economics, andWorld Bank (2007).There is a large amount of churning in the market for household loans, as Table 2.6 shows. The top panel shows the proportion of households in the rural survey panel who borrowed from each major source in each of the three years. For instance, 2.7 percent of households borrowed from VBSP in 2004 but not in 2006 or 2008. Although 33.3 percent of households did not borrow in any year, only 17.7 percent borrowed in all three years. The bottom panel of Table 2.6 illustrates this point in another way, by showing the proportion of households who dropped or took out loans from each main lending group between 2004 and 2006, and between 2006 and 2008. Of the 5.2 percent of households who borrowed from VBSP in 2004, two-thirds had dropped out by 2006, but the addition of 6.0 percent of households raised the VBSP penetration rate to 7.6 percent by 2006; in turn, half of these dropped out in the ensuing two years, but 8.6 percent of households were added to the VBSP roster. VBARD dropped more than half of its borrowers between one period and the next, and informal lending was even more transitory. While credit may have been widely available during this period, households could not easily rely on multiyear credit from a given source.  Surveys (2004Surveys ( , 2006Surveys ( , 2008)). Note: VBARD = Vietnam Bank for Agriculture and Rural Development; VBSP = Vietnam Bank for Social Policies.The first rigorous studies of the impacts of microfinance used quasi-experimental methods and typically found evidence of a positive effect of microlending on incomes. In their widely cited study, Pitt and Khandker (1998) found that an additional 100 taka of microcredit in Bangladesh raised household consumption expenditure by between 11 taka (for male borrowers) and 18 taka (for female borrowers). Khandker (2005), using more recent data, found that access to microfinance in Bangladesh contributed to poverty reduction, especially for female participants. Such results are by no means confined to Bangladesh; for instance, Boonperm, Haughton, and Khandker (2013) estimated that lending by the Thailand Village Fund raised income by 1.9 percent and spending by 3.3 percent in about 2004, with the effects mainly concentrated among poorer households.A growing number of studies of the impact of microcredit use randomized evaluations, which are seen as a good way to avoid the problem of the selection bias that may occur when one simply uses data from existing borrowers. A recent survey of six of the best such evaluations (Banerjee, Karlan, and Zinman 2015), which had been performed for a variety of microcredit programs in a diverse set of countries (Bosnia, Ethiopia, India, Mexico, Mongolia, and Morocco) and with a sophisticated array of empirical and econometric strategies, concluded that microcredit has had a modest positive economic impact. The low take-up of microloans by entrepreneurs reduces the power of such evaluations; and though the studies reviewed do not show transformative effects from microlending, they do not rule out such effects either. While the external validity of any given randomized evaluation may be open to question, the accretion of broadly similar conclusions from trials undertaken in a wide variety of contexts does add weight to an emerging consensus that microcredit contributes somewhat to raising incomes and lowering poverty.So it is surprising to find that most of the studies of the impact of microcredit in Viet Nam have concluded that the effects are large. For instance, the World Bank (2013) claimed that under the Second Rural Finance Project, 275,000 new microloans were provided, and if one includes the other components of the project, 445,000 families and small enterprises saw their incomes rise by 60-65 percent as a result of the project, and 274,000 new jobs were created. The source of these miraculous results is unclear, although they figure prominently on the World Bank webpage related to this project. Cuong and colleagues (2007) used data from a panel of households from the 2002 and 2004 VHLSS surveys to argue that \"the VBSP was quite effective\" and that participation in its loans raised household incomes and spending by 30 percent of the value of the loans. Cuong (2008) used the same data to argue that VBSP lending was poorly targeted but nonetheless reduced the head-count poverty rate by 5 percentage points, and that borrowers' incomes rose by 63-93 percent and their spending by 68-70 percent. The most intractable problem here is that the information on household borrowing collected by the 2002 survey is seriously incomplete and cannot be used to provide a baseline. In a more recent study, Cuong and Van den Berg (2011) turned to the 2004 and 2006 VHLSS data and argued that over this period informal credit reduced the poverty rate from 47 percent (the counterfactual) to 39 percent, again a remarkably large effect.Duong and Nghiem (2013) pooled households from the living standards surveys of 1993, 1998, 2002, 2004, 2006, 2008, and 2010, and regressed consumption (or income) on a dummy variable that was set to 1 if a household had a loan worth US$500 or less. Based on preliminary results, the authors claimed that having such a loan raised consumption by 50 percent and income by 8 percent. The most serious difficulty here (apart from the inadequate credit data for 2002) is that no effort was made to correct for endogenous placement.A completely different result was found by Pham and Lensink (2012), who used the rural panel component of the 2004 and 2006 VHLSS data to estimate the effect of microcredit on self-employment profit-the component of income that is presumably responsive to the effects of loans used for productive purposes. They found that microcredit had essentially no effect, although (larger) VBARD loans did have a positive impact on incomes. Their study used a fixed-effects instrumental variables (IV) model that attempted to correct for selection bias. They concluded as follows:Further research on the impact of microcredit on poverty reduction in Vietnam seems to be important. There are several issues that need more attention. First, it may be interesting to compare the impact of microcredit with that of informal finance. Second, some improvements on the methodology part seem important. In particular, if data availability allows this, the use and selection of valid instruments to correct for endogeneity issues is relevant. Moreover, more attention should be paid to control for attrition bias (2012, 21). These differing views-huge effects versus little or no effect-cannot easily be reconciled, which is why there is a need for a study that uses more data, explores robustness more thoroughly, accounts for potential time-varying unobserved heterogeneity, and examines the importance of different sources of credit.The VHLSS included a panel of households in 2004, 2006, and 2008. Attrition at the household level was around 9.5 percent between 2004 and 2006 (Baulch and Dat 2011), but 1,848 usable cases remained. The household questionnaires had a module on credit and borrowing, which asked households directly whether they had borrowed from VBARD and VBSP (and other lenders) in the last 12 months, as well as other details including the loan amount, date of borrowing, purpose of borrowing, interest rate, and repayment status.The job of assessing the impact of credit, or microcredit, on spending (or other measures of welfare) is made more difficult by the fact that credit is widely available throughout Viet Nam (Anh et al. 2011), although some authors argue that rural households still face many barriers to credit (Duong and Izumida 2002). Based on a survey of 932 households in four provinces in 1997 and 2002, Barslund and Tarp (2008) found very limited evidence of credit rationing: just 9 percent of loan applications were rejected, and less than 8 percent of households refrained from requesting credit even though they would have liked to borrow. However, Barslund and Tarp (2008) were unable to assess whether households, even if they could borrow, were able to borrow as much as they wanted, and so the possibility of credit rationing cannot be ruled out. If credit is not rationed, then the contribution of VBSP or other microcredit lenders is relatively limited: borrowers may now be able to obtain credit cheaply enough. But if credit is rationed, then there may be significant potential for microcredit to boost incomes.In most Vietnamese villages there is a choice of credit sources-from VBSP to local moneylenders to family and friends to credit funds to VBARD. Thus it would be difficult to construct a randomized controlled trial now that would allow one to measure the impact of credit in a compelling way. We are therefore constrained to using an econometric (quasi-randomized) design, but we are able to make use of the availability of panel data over three years. We use two distinct models to address our central problem, which is to measure the effect of microcredit on income and consumption.The first approach is to measure the \"intention to treat.\" In order to get a loan from VBSP, a household is supposed to be poor, as defined by the local People's Committee; we refer to this requirement as eligibility (Eij). It is also a practical necessity that there be a VBSP branch nearby (which might be a mobile unit); if a commune has a branch, we consider the area to be treated (T ij ), following the terminology (and method) used by Pham and Lensink (2012). Then our model, which uses annual data, may be written aswhere the outcome (Y) of household i in village j depends on a variety of household and community variables that are exogenous to borrowing (X), eligibility for a credit program (E), and the existence of a VBSP branch nearby (T). Our interest is in the size and significance of δ. We estimate this equation using pooled data with robust standard errors for a variety of outcomes including real consumption per capita, real consumption per household, real income per capita, and real self-employment earnings per household. This approach can also be applied to VBARD lending, but it does not work satisfactorily for informal credit or credit overall, because the value of T is 1 in essentially all such cases.A more complete quantity-of-credit model would take into account the amount of borrowing actually undertaken by a household. A fixed-effects version of this workhorse may be written aswhere C measures the amount of credit per year. A variant on this specification uses the interaction (\uD835\uDC38\uD835\uDC38 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 \uD835\uDC36\uD835\uDC36 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 ) instead of \uD835\uDC36\uD835\uDC36 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 (Khandker 2005), and we employ this alternative specification below to test the robustness of our results. We initially estimate this equation in a simple pooled version, but this exercise is unlikely to be satisfactory because of the probable correlation between \uD835\uDF00\uD835\uDF00 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 and \uD835\uDC36\uD835\uDC36 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 . There are at least three plausible reasons for this correlation: if credit outlets are located in more affluent areas, rather than randomly, there is likely to be a problem of endogenous program placement; there may be unobservable household or community characteristics that simultaneously influence loan participation and its outcomes; and selection bias may also arise if the size of the loan is associated with unobservables.Our time-fixed-effects estimates are designed to sweep away the effects of any time-invariant unobservables, although they also remove the influence of potentially interesting variables that we do observe. In this case, the identification of the effects of loans relies on those cases in which someone either starts to borrow or stops borrowing, either between 2004 and 2006 or between 2006 and 2008. There can be no assurance, a priori, that these cases are random. One way to address this problemselection bias due to time-varying unobservable effects-is to instrument the credit variable, C. We do this using interactions between the eligibility measure (E), the treatment variable (T), and household characteristics (see Haughton, Khandker and Rukumnuaykit 2014;Pitt and Khandker 1998; Armendariz de Aghion and Morduch 2005; Pham and Lensink 2012), which gives instruments of the form \uD835\uDC4B\uD835\uDC4B \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 , \uD835\uDC47\uD835\uDC47 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 , \uD835\uDC4E\uD835\uDC4E\uD835\uDC59\uD835\uDC59\uD835\uDC4E\uD835\uDC4E \uD835\uDC38\uD835\uDC38 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 . We can test for the suitability of the instruments using a Sargan-Hansen J test, and for the need for instruments using the difference-in-Sargan test.The VHLSS has collected household data every second year since 2002. However, only the data for 2004, 2006, and 2008 constitute a viable panel: the 2002 round did not collect adequate data on credit, and the panel was not continued in the 2010 survey. The period 2004-2008 also coincided with a rapid expansion of VBSP, which makes these years especially useful when trying to measure the impact of VBSP lending.Table 5.1 gives a full list of the variables that we use, which are either dependent variables (the first four items) or control variables that we use systematically in the estimations. Values are shown by year and by source of loan: VBSP, VBARD, and informal sources. The numbers are shown in bold if there is a statistically significant difference (at the 10 percent level) in the values between those who borrowed from the source in question and those who did not. Thus, for example, VBSP borrowers were far more likely to be poor than those who did not borrow from VBSP.It is clear from Table 5.1 that VBSP serves a segment of the population that is relatively poorwith incomes one-third below the average of those surveyed, and consumption levels that are one-fifth below the mean. About half of the income of VBSP borrowers comes from self-employment, mainly in agriculture. Compared with their peers, VBSP borrowers live in larger households, are younger, are twice as likely to belong to minority groups, and are concentrated in the more remote and mountainous parts of the country.Although there is some overlap, VBARD mainly serves a different group-households who are not particularly poor, who derive two-thirds of their income from self-employment, and where four-fifths of family workers work in agriculture; VBARD lending is common in the Mekong Delta but less prominent in the Red River Delta, Viet Nam's other major rice bowl.During the period in question VBSP grew rapidly, lending to 6 percent of households in 2004 and 14 percent by 2008. Much of the growth was due to an expansion in geographic coverage: there was a VBSP presence in 47 percent of all districts in 2004, but in 73 percent of districts by 2008; meanwhile, the proportion of communes with a bank branch did not change. The average size of VBSP loans also rose modestly during this time, from VND 5.1 million to VND 7.6 million; typically, VBSP borrowers augmented their VBSP loans with borrowing elsewhere, bringing their total borrowing to about 50 percent more than these amounts.This growth of VBSP lending between 2004 and 2008 was associated with a fall in the share of households borrowing from VBARD-although the average size of VBARD loans tripled during this time. VBSP loans also displaced some informal borrowing. Overall, the proportion of households with loans hardly changed.   Tables 5.2 and 5.3 display the estimation results for the intention-to-treat model, in which the dependent variable is the log of real consumption per capita (excluding spending on durable goods), and borrowing is from VBSP. Table 5.4 sets out the essential results for the quantity-of-credit model, for the cases in which the dependent variable is the log of real consumption per capita, or of real income per capita, or of unearned income per household. In all cases, we have corrected for potential heteroscedasticity. Table 5.2 reports the estimates of the intention-to-treat model, first using pooled data and then with random effects (assuming exchangeability) and household fixed effects. A Hausman test favors the fixed-effects specification (chi-squared 28 = -364). The coefficient of interest is δ, estimated to be 0.062 (s.e. = 0.028) in the fixed-effects version; this is statistically significant at the 5 percent level and shows a positive effect, meaning that the availability of VBSP credit does appear to raise consumption spending per capita. The order of magnitude is also plausible, if perhaps a little high: a 6.2 percent increase in consumption is equivalent to about VND 2 million per household, or slightly more than one-quarter of the average value of a VBSP loan (VND 7 million). In a study of microfinance in Bangladesh, Khandker (2005) found a 17-taka increase in consumption for every 100 taka lent.It is reassuring that the signs of the other coefficients in the intention-to-treat model fit well (an adjusted R 2 of 0.49 in the pooled model) and have the expected signs: households who are eligible for VBSP loans are indeed poorer; per capita spending is lower for households with many young or old members, for larger households, for those working in agriculture, and for households living in the Northern Uplands and North-Central Coast. Conversely, households have higher levels of per capita spending when their members are better educated, live in the south of Viet Nam or the Red River Delta, or own more land.Table 5.2 shows the effect of VBSP loans on per capita consumption, but there are other outcomes of interest, including consumption per household (rather than per capita), income per capita, and self-employment earnings per household. The relevant results for these outcomes are shown in Table 5.3, where we present the estimates of δ for three different specifications (pooled, random effects, and household fixed effects). The effect of the availability of VBSP microcredit on consumption is significant, or close to significant, in all specifications. In the household-fixed-effects specification, which is the most compelling version, the estimated coefficient in the income equation is comparable in magnitude to the one in the consumption equation, although not quite statistically significant, with a p-value of 0.13.The bottom rows of Table 5.3 show that the availability of VBSP credit does not have a statistically significant effect on earnings from self-employment, or on either of its two components (agricultural earnings and nonagricultural self-employment). These findings echo those of Pham and Lensink (2012), who used a similar specification and data for just 2004 and 2006. At first sight, this result is puzzling, because if credit does not boost consumption via higher self-employment earnings, then by what pathway does it work? It is certainly possible that there may be a problem with the low power of the tests, an issue emphasized by Banerjee, Karlan, and Zinman (2015), but it is worth noting that in 2008, households reported that only a quarter of VBSP loans were incurred for the purposes of production or working capital (Table 2.5). A more compelling interpretation is that much of the household borrowing from VBSP is to help protect consumption from swings in income, and perhaps also to allow for significant consumption purchases. Credit for this purpose plays a valuable role in helping poor households cope when faced with shocks or large consumption needs (such as health or educational expenses).The estimates of the relevant coefficients for the quantity-of-credit model are shown in Table 5.4; the full results are available from the authors. Three specifications are shown: with fixed effects (FE), with fixed effects using instrumental variables (FE-IV), and with random effects using instrumental variables (RE-IV). In all cases, the FE-IV estimates are preferred, and the results of the other versions are shown for robustness. Two instruments are used in every case: the proportion of other households in a district who borrow from VBSP (lender1), and this variable interacted with the proportion of household members who are older than 60. The lender1 variable is exogenous to the household and is likely to be correlated with the household's borrowing from VBSP but not with the impact that a VBSP loan would have on the household's consumption or income. The interactive term is motivated by the idea that an older household may be less interested in borrowing, even though it could use the borrowed money effectively.A Hausman test comparing the FE-IV and RE-IV specifications strongly supports the householdfixed-effects version in every case (with p < 0.001). To determine the appropriateness of using instrumental variables, we applied a battery of tests, including the conventional Cragg-Donald F test on the first-stage equation, which gave a result between 5 and 6. This means that our instruments are not particularly strong. However, none of the Sargan-Hansen J tests even came close to rejecting the null hypothesis of no correlation between the instruments and the residuals in the main regression, which means that the chosen instruments are not inappropriate.The median amount borrowed by a household from VBSP in 2008 was VND 7 million (about US$400). If this were to rise by VND 1 million, then a household's per capita consumption could be expected to rise by 4.9 percent, or by about VND 350,000, according to the FE-IV estimate of the quantity-of-credit model. The estimate is close to being statistically significant (p-value of 0.11). The effect of VBSP borrowing on real income per capita is stronger, and statistically significant. Similar results are found when the specification interacts the amount of VBSP borrowing with eligibility to borrow. VBSP is subsidized to the tune of about 2 percent of the value of its loans; if this translates directly to lower interest rates (relative to alternative sources of credit), this alone would raise consumption and income by about VND 20,000 for every additional million VND borrowed. But the effects shown here are a good deal larger than this, suggesting that at the margin, VBSP loans are being put to highly productive use.As with the intention-to-treat model, here too we are unable to pick up any effect of borrowing on self-employment earnings. And our results are only moderately robust, because none of the estimates from the FE and RE-IV specifications are statistically significant, but there does seem to be some evidence here that VBSP lending had an impact. A further approximation is introduced by our use of a linear equation, even though the credit variable is 0 for those who do not borrow from VBSP (which is true in most cases), but a positive number otherwise.If microcredit significantly reduces poverty, or raises the incomes of the poor, then there is a strong case for encouraging it, and an argument can even be made for subsidized credit of the sort provided by VBSP. But if microcredit has limited or no real effects, then the case for vigorously promoting microcredit is seriously weakened.VBSP lending does have a tangible effect on poverty. Using a poverty line of VND 4.8 million per capita in 2008, we find a poverty rate of 24.86 percent for our (rural) sample. In the absence of VBSP lending-and using the magnitudes shown in the fixed-effects estimates in Table 5.3-the rural poverty rate would have been 25.55 percent.One can only speculate on how large VBSP would have been in 2008 if it had not been subsidized (to the tune of about $70 million-see Table 2.4-or about $1 per person in Viet Nam). If it had lent half as much, then the subsidy would have bought a poverty reduction of about one-third of a percent of the population-a nonnegligible effect that would likely have been difficult to achieve at this price using other policies.The focus of much of the recent writing on microcredit in Viet Nam has been on the issue of governance and sustainability; the World Bank ( 2004) cautioned about the governance structure of the VPSP; Timberg and Binh (2011) fretted about the sustainability of subsidized microcredit, as did Khoa (2013); and Anh and Tam (2013) noted that some microcredit programs closed after subsidies ended, prompting the need to pay attention to sustainability. The government is aware of the problem: in 2012 it approved a development strategy that aims \"to increase VBSP's stability and sustainability, transform it into a self-sustaining operation, and enhance its capacity to provide state policy credit to poor and nearpoor households\" (Bezemer and Schuster 2014, 34).Impact and sustainability go hand in hand; if microcredit has a large impact, it is easier to charge interest rates that enable microcredit programs to be sustainable. It is also easier to make the caseespecially if the benefits flow mainly to poorer households-for maintaining the current arrangements whereby funds are mobilized from outside the universe of poor households and channeled to them via VBSP. This situation argues for the importance of being very clear about the extent of the impact of microcredit and other forms of rural credit, which is indeed the motivation for our study.In order to estimate the impact of VBSP lending in Viet Nam, we have had to use a quasiexperimental approach because randomized controlled trials were not possible. Fortunately, we have been able to make use of data from a panel of rural households, surveyed in 2004, 2006, and again in 2008, during a time when VBSP was expanding rapidly. We find solid, if not ironclad, evidence that VBSP lending boosts household consumption spending; this could occur indirectly if borrowing raises income and hence spending, or directly if VBSP loans serve as consumer credit and are available more cheaply or conveniently than funds from alternative sources.The impact of VBSP lending on incomes, and on self-employment income (where the effect of borrowing for productive purposes should be strongest), is not clear, although it does appear to be positive. The lack of a stronger effect is in part because most loans are not used for productive purposes, and even those that are may not generate a payback within a period as short as a year or two.While we do not find evidence of the strong effects found by some authors (Cuong et al. 2007;Cuong 2008;World Bank 2013), our results show somewhat stronger positive effects than those found by Pham and Lensink (2012), perhaps because we were able to use an additional round of panel data, increasing the power of our estimates. It would appear that VBSP lending is helpful. Our estimates suggest that the impact likely does justify the current subsidy rate of about 2-3 percent of the value of funds lent, especially because the benefits accrue disproportionately to poorer households in Viet Nam.","tokenCount":"6380","images":["1942183744_1_1.png","1942183744_10_1.png"],"tables":["1942183744_1_1.json","1942183744_2_1.json","1942183744_3_1.json","1942183744_4_1.json","1942183744_5_1.json","1942183744_6_1.json","1942183744_7_1.json","1942183744_8_1.json","1942183744_9_1.json","1942183744_10_1.json","1942183744_11_1.json","1942183744_12_1.json","1942183744_13_1.json","1942183744_14_1.json","1942183744_15_1.json","1942183744_16_1.json","1942183744_17_1.json","1942183744_18_1.json","1942183744_19_1.json","1942183744_20_1.json","1942183744_21_1.json","1942183744_22_1.json","1942183744_23_1.json","1942183744_24_1.json","1942183744_25_1.json","1942183744_26_1.json","1942183744_27_1.json","1942183744_28_1.json","1942183744_29_1.json","1942183744_30_1.json","1942183744_31_1.json","1942183744_32_1.json","1942183744_33_1.json"]}

data/part_2/0245735430.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"1fb8960d533da9262589dd62089a65ad","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/3b1eb93a-8263-4492-83de-cbab9b21dbbe/retrieve","description":"The unprecedented, large-scale, rural-to-urban migration in China has left many rural children living apart from their parents. Yet the consequences for child development of living without one or more parents due to migration are largely unknown. In this study, we examine the impact of parental migration on one measure of child development, the nutritional status of young children in rural areas. We use the interaction terms of wage growth in provincial capital cities with initial village migrant networks as instrumental variables to account for migration selection. Our results show that parental migration has no significant impact on the height of children but that it improves their weight. We provide suggestive evidence that the improvement in weight may be achieved through increased access to tap water in households with migrants. To conclude the paper, we raise concerns about the sustainability of the impact.","id":"1498571560"},"keywords":["migration","children","nutrition","rural China vi"],"sieverID":"f25322d4-c4f1-4755-a869-b262624a94f2","pagecount":"32","content":"IFPRI Discussion Papers contain preliminary material and research results. They have been peer reviewed, but have not been subject to a formal external review via IFPRI's Publications Review Committee. They are circulated in order to stimulate discussion and critical comment; any opinions expressed are those of the author(s) and do not necessarily reflect the policies or opinions of IFPRI.The migration of labor out of agriculture is a primary feature of the economic development process, and the study of migration as an economic process has a long history (Lewis 1954;Fei and Ranis 1964). The role of migration in the economy has recently regained prominence in policy discussions and in the research community because of an increased flow of labor both internationally and within large countries, such as India and China. However, Clemens (2011) notes that much research has focused on the effects of immigration while neglecting emigration, the effects of migration on those left behind. From a microeconomic perspective, the effects of emigration on source households and communities can be complex. Migrants typically continue to have economic interactions with the households and communities they leave behind (Stark and Bloom 1985), and these interactions can be particularly important when markets do not function well. For example, migrant remittances can help households overcome credit constraints that hinder investment in the human capital of the next generation (Beine, Docquier, and Rapoport 2008;Yang 2008). When migrants leave households, however, the time investment in raising children also decreases ( de Brauw and Mu 2011); moreover, as household members live apart because of migration, they cannot perfectly monitor one another's behavior. Consequently, the adults remaining in the village may not serve the best interests of the children living with them (Chen 2006). Consequently, migration can affect human capital outcomes of children in several ways, and the overall effect depends on the relative importance of these factors.Focus on children's nutrition is important because poor nutrition outcomes may have long-term consequences for children. Multiple studies have found children's health and nutritional status to have a sizable and statistically significant positive impact on educational outcomes in many developing countries, such as Pakistan (Alderman et al. 2001), Ghana (Glewwe and Jacoby 1995), Guatemala (Maluccio et al. 2009), and Kenya (Miguel and Kremer 2004). 1 Moreover, positive impacts seem to occur among both boys and girls (Glewwe and Miguel 2008).Several recent studies suggest the potential importance of migration for child nutrition among those left behind. Gibson, McKenzie, and Stillman (2011) find that in Tonga, height-for-age z scores (HAZ) are lower among children younger than 18 years old left behind by migrants to New Zealand. 2They find no impact on weight-for-age z scores (WAZ) or BMI-for-age z scores (BMIZ). 3 Carletto et al.  (2011) find a positive impact on HAZ scores among Guatemalan children left behind by emigrants to the United States. Evidence from Tajikistan also shows that children in communities with more migrants have higher HAZ-scores (Azzarri and Zezza 2011). Mansuri (2006) and Nobles (2007) study the effects of international migration from Pakistan and Mexico, respectively, on child HAZ-scores, with the former finding a positive effect and the latter a negative one. Differences in data, country contexts, child age definitions, empirical specifications, and methods all may have contributed to the differences in these results, but they also highlight the complex relationship between migration and child nutrition, a subject that deserves further investigation.China is a particularly salient place to study this issue for several reasons. First, the flow of internal migrant labor has increased rapidly since economic reforms began and in particular since labor market restrictions were loosened (Liang and Ma 2004;Fan 2008), yet partially as the result of the household registration (hukou) system in China, migrants are still largely barred, either legally or financially, from accessing education for their children in urban areas. Poor housing conditions for migrant workers in urban areas also prevent family migration (World Bank 2009); consequently, wholefamily migration is rare, and many young and school-age children stay behind in rural villages. Recent estimates suggest that as many as 23 million children (younger than age 14) are left behind in migrantsending regions while their parents work elsewhere (Guo 2009).This study builds on our previous analysis of the relationship between migration and nutritional status among children in rural China (de Brauw and Mu 2011). We previously found that children aged 7 to 12 years in migrant households are more likely to be underweight perhaps because they spend more time doing household chores than children in nonimmigrant households. We also found that younger children (aged 2 to 6 years) in migrant households are less likely to be overweight when no grandparent lives with them. In this study, we examine distribution means of HAZ, BMIZ, and WAZ scores to complement the previous findings on measures based on the tails of the BMIZ distribution-overweight and underweight status. HAZ-scores measure the cumulative investments that parents have made in children over time and are particularly sensitive to investments made while children are young (World Health Organization 1983). BMIZ and WAZ-scores fluctuate more contemporaneously and can therefore highlight short-term changes in nutrition. In addition, we focus on young children aged 0 to 5 in the initial survey years since the potential effects on young children can be long lasting when their parents migrate (Parreñas 2005); moreover, we improve the identification by constructing instrumental variables for changes in parental migration status. The instrumental variables are the interaction terms of wage growth in provincial capital cities with initial village migrant networks. Finally, we provide suggestive evidence about whether parent migration affects housing conditions and hygiene in the living environment, which can affect children's nutrition.Our results show that parental migration has no significant impact on young children's HAZscores, but it improves their WAZ-scores. We provide suggestive evidence that the improvement in shortterm nutrition status may be achieved through increased access to tap water among households with migrants.The paper proceeds as follows. Section 2 provides more background about China's rural-to-urban migration and nutritional status among children in rural areas. The third section presents a conceptual framework, and the fourth section discusses the data we use and the empirical framework for the paper. The econometric results are presented and discussed in the fifth section. The final section concludes.Accelerating rural-to-urban migration has been an important demographic feature in China during the past three decades. In this paper, we are particularly interested in the implications of migration for the nutritional status of children left behind in rural villages. In this section, we will outline the evolution of China's internal migration and its implications for rural communities, followed by a discussion about how the distribution of children's nutritional status is changing in rural China.China's labor market experienced dramatic changes during the 1990s as the number of rural residents moving to urban areas for employment grew rapidly. Estimates using the 1 percent sample from the 1990 and 2000 rounds of the Population Census and the 1995 1 percent population survey suggest that the intercountry migrant population grew from slightly more than 20 million in 1990, to 45 million in 1995, to 79 million by 2000 (Liang and Ma 2004). These figures likely underreport the true scale of migration because they do not account for migration that takes place during shorter periods of time (Cai, Park, and Zhao 2008).The expansion of migration during the past two decades has been facilitated by the relaxation of constraints on the household registration (hukou) system. The initial hukou reform, taking place in 1988, established a mechanism for rural migrants to obtain legal temporary residence in China's urban areas (Mallee 1995). After rural migrants could obtain legal temporary residence, they became better able to establish networks to facilitate the job search in distant labor markets (for example, Munshi 2005); however, legal temporary resident status does not imply access to urban health and education services or social safety nets because these benefits are still linked to household registration status (World Bank 2009). This institutional arrangement effectively discourages entire families from moving to cities. Whereas individuals with rural hukou status can now purchase nonagricultural hukou status from urban governments in many places, the system continues to work against more permanent migration flow (Fan 2008). Migrants consequently maintain close ties with their ancestral villages.The impacts of migration on rural households and individuals left behind are mixed. On one hand, remittances benefit migrant-sending households, as evidenced by their higher consumption (Du et al. 2005; de Brauw and Giles 2008), improved risk-coping ability (Giles 2006;Giles and Yoo 2007), and higher levels of investment in productive assets (Woodruff and Zenteno 2007). On the other hand, individual household members who are left behind, particularly women and seniors, may be somewhat negatively affected. For example, women who live in migrant households now perform more farm work than they would have otherwise (Mu and van de Walle 2011), and the elderly with migrant adult children are at greater risk of falling into poverty (Giles, Wang, and Zhao 2011). Even though children's educational performance is not compromised but actually improves in migrant households (Chen et al. 2009), some nutrition measures of school-age children are negatively affected (de Brauw and Mu 2011).It is important to remember that in the mid-1990s, migration rates were substantially higher among men than women for most age ranges (Rozelle et al. 1999). Since the late 1990s more women have participated in rural-to-urban migration, particularly the young and the single (Du, Park, and Wang 2005;de Brauw et al. 2008). In some provinces, the gender gap in migration rates is still increasing (Mu and van de Walle 2011). Consistent with differences in migration prevalence by gender shown elsewhere, we demonstrate later in the paper that the majority of parental migration is by fathers.China faces multiple challenges in improving nutrition among rural children. First, disparities are present in child health in rural and urban areas as well as within and among regions. For example, under nutrition in western provinces is acute, and stunting and underweight are three times more prevalent in rural than in urban areas (UNICEF 2008). Child growth is becoming more sensitive to household income, a consequence of increasing income inequality and the disappearance of subsidized food coupons in the mid-1990s (Osberg, Shao, and Xu 2009). In addition, a significant share of rural children remains malnourished. In 2002, the nationwide stunting rate was still nearly 15 percent (Svedberg 2006). With inadequate access to regular micronutrient-rich diets, many rural children also suffer from iron deficiency (Luo et al. forthcoming). At the same time, induced by increased income, the structure of the Chinese diet is shifting away from high-carbohydrate foods toward high-fat, high-energy-density foods (Du et al. 2004); consequently, overweight status has been increasing among children in both rural and urban areas (Monda and Popkin 2005;de Brauw and Mu 2011).Notwithstanding these challenges, the average measures of the nutritional status of children have clearly improved over time. Economic growth is well known to be associated with improved nutritional status (Fogel 2004), and China's experience is no exception. Child growth improved during the early reform period. In 1990, the average height of children aged two to five in seven provinces was 92.5 cm (centimeters) in rural areas, an increase of 3.8 cm from the average in 1975 (Shen et al. 1996). Improvement in children's growth has continued since then as children's average HAZ, WAZ, and weight-for-height z (WHZ) scores increased dramatically between 1990 and 2000 (Chen 2000;Osberg, Shao, and Xu 2009). These increases represent an anthropometric confirmation of significant progress in average well-being. Some improvements in child growth are directly attributable to the progress made in public service provision. A salient example is water delivery because unsafe drinking water can cause diseases such as diarrhea, which can severely affect child growth (for example, Glewwe and Miguel 2008). Efforts to establish public water systems in urban China date back to the 1950s. During the 1980s, the Chinese government launched a drinking water improvement program in rural areas (Zhang 2012). Clean water from water plants improves rural children's heights and weight-for-height z scores substantially (Zhang 2012). In the same context, Mangyo (2008) also finds that access to in-yard water sources improves child health when mothers are relatively well educated. Even though the placement of village-level water projects may be exogenous to individual decisions, household-level water access may still be affected by individual demand (Mangyo 2008). Motivated by these findings, in examining the channels through which migration may affect children's nutrition, we will explicitly investigate whether parental migration improves household access to clean water.One can hypothesize four main channels through which migration might affect nutritional investment in children: an income effect, an information effect, a time effect, and an effect from changes in the structure of intra-household bargaining and cooperation.First, from the perspective of the household, a primary reason to send out migrants is to increase overall household income, no matter how migration or the household is defined.4 An increase in household income can lead to better nutritional outcomes among children through various means. For example, higher incomes could lead to increased calorie intake if food is scarce within the household (Fogel 2004). The demand elasticity's of nutritious foods, however, tend to be higher than those of coarse grains (Bouis 1994); therefore, households may use income increases to purchase more foods rich in micronutrients (such as fruits and vegetables) and protein (such as eggs and meat) (Subramanian and Deaton 1996;Nguyen and Winters 2011). Improvements in diet might manifest themselves particularly in HAZ scores. In addition, higher income could also lead to improved housing conditions and a more hygienic environment, such as an environment with better access to clean water. Finally, health service utilization for children may increase with income. The majority of rural residents were uninsured before the New Cooperative Medical Scheme was introduced in 2003, but even since 2003, out-of-pocket medical spending still positively correlates with income (Wagstaff et al. 2009).Second, independent of income effects, migration experience may improve household health information and hygiene practices. As a result, the demand for healthy food, medical service, clean water, and better sanitation may increase in migrant households. Similar to income effects, information effects are expected to improve nutrition outcomes. We cannot separate these effects in this study, but it is worth noting that either of them can bring a positive impact as a result of migration.Third, by sending out migrants, households decrease their overall labor endowment. Because households in rural China are almost always engaged in some form of local income generation, either through farming or self-employment, child care time becomes scarcer in migrant-sending households. As a result, less time is allocated to cooking (Chen 2006; de Brauw and Mu 2011) and monitoring the eating habits of children; therefore, to the extent that it is difficult and time-consuming to ensure nutritional investments in children, the time effect of migration may compromise child nutrition, especially among young children.Fourth, when household members live apart as a consequence of migration, the equilibrium in the implicit intra-household bargain may change. For example, some argue that wives gain increased autonomy and new decision making powers as household heads after their husbands migrate (Davin 1999). Considering that the empowerment of women is generally associated with better child health outcomes (for example , Thomas 1990;Duflo 2003), one may expect that fathers' migration could positively affect child nutrition. Alternatively, the adult household members left behind may use more child labor in household chores but counter any potential negative effect on children by increasing their food intake. Such strategic behavior may result in an innocuous impact on child nutrition (Chen 2006).As outlined in this conceptual framework and suggested by mixed results from existing studies, whether parental migration affects children's health and nutrition is primarily an empirical question. In the next section we present the data and empirical framework we will use to try to measure this relationship in rural China.The data used in the analysis derive from the four waves of the China Health and Nutrition Survey (CHNS) of 1997, 2000, 2004, and 2006.5 The CHNS is a longitudinal survey covering nine provinces that vary substantially in geography, economic development, and access to public resources.6 The survey includes the demographic characteristics of each household as well as asset holdings. We further make use of community-level questionnaires that accompanied the household survey.Important for this paper, the CHNS records anthropometric measurements, height, and weight for each individual within the household. To construct HAZ, WAZ, and BMIZ scores, we use the most recent growth charts made available by the World Health Organization (WHO), which have corrected previous standard growth charts for what are now considered developed-country biases (de Onis et al. 2007). 7The CHNS was not originally conceived to study migration, so the migration status of household members in the CHNS must be built up from the household roster. Starting from 1997, if an individual who was in a previous round of the CHNS is not in the current round of the survey, a question is asked regarding why this individual does not currently reside in the household. We consider any individual who has left the home to seek employment between any two waves of the survey to be a migrant. 8We pool panel data from all three pairs of surveys-1997 and 2000, 2000 and 2004, and 2004 and 2006-keeping in the sample children younger than five years in the initial years (1997, 2000, and 2004), and survey for two consecutive rounds in our analysis. Our focus on young children is motivated by the findings that they are at greater risk for problems associated with malnutrition and are more likely to respond to nutrition interventions (WHO, 1983). With two observations for each child, we can apply panel data analysis to control for time-invariant individual, family, and community characteristics.Trends related to migration and anthropometric measures found in the literature, discussed in section 2, are also present in the CHNS. To illustrate migration trends, we graph locally weighted regressions of out-migration on age for individuals aged 16 to 75 (see Figure 4.1). Age is negatively correlated with the probability of out-migration for both men and women. Except for the oldest age cohorts, migration prevalence among individuals of any given age increases over time among both men and women. The rate of increase is slower among older individuals, implying that migrant labor flows tend to be young and getting younger, consistent with findings in other studies (for example, Liang and Ma 2004). In the 25-to 40-year-old range, migration rates tend to increase among younger workers, more likely to be married and with children, from around 10 percent for men in 1997 to between 30 and 35 percent in 2006. For women, increases in migration are less dramatic, from less than 10 percent in 1997 to greater than 20 percent in 2006. It could reasonably be expected that migrations by individuals in these cohorts have the largest chances of affecting the nutritional status of children.  In Figure 4.2, we present locally weighted regressions of HAZ, WAZ, and BMIZ for children younger than 10. On average, HAZ and WAZ tend to increase over time at most ages. In 1997, average HAZ scores are less than -1, implying a significant stunting rate.9 By 2006, average HAZ increase to around -0.7, with the scores of very young children approaching those of the healthy reference population. The age profile of average HAZ scores shows that this measurement deteriorates from birth to approximately 48 months. After that, average HAZ scores level off or improve slightly. This pattern is similar to patterns found in Guatemala (Carletto et al. 2011), where the turning point is around 30 months. WAZ scores are negatively correlated with age, a pattern common in many developing countries, such as El Salvador ( de Brauw 2011). A steady increase in WAZ scores is evident during the period between 1997 and 2006. Children younger than 3 have already had average WAZ scores equal to or greater than, the international norm in 2006. Given that WAZ scores for most children in the sample are higher than their HAZ scores, it is not surprising that BMIZ scores are actually higher than international norms on average until children reach age 5. We explore the differences in distribution of these anthropometric measures between children with at least one migrant parent and those whose parents who did not migrate at the time of the survey. Figure 4.3 shows that the middle of the distribution of HAZ scores for children whose parents did not migrate is positioned slightly right of that for children with a migrant parent, but no distinct shift in the whole distribution is observed. The distributions of WAZ and BMIZ scores are similar for these two groups of children. The conceptual framework in section 3 suggests estimating an equation for nutritional outcomes of individual child i in household h of village j at time t denoted by the nutrition outcome H of child \uD835\uDC56 in household ℎ located in village \uD835\uDC57 at time t (\uD835\uDC3B \uD835\uDC56ℎ\uD835\uDC57\uD835\uDC61 ), which is a function of parents' migration status (\uD835\uDC40 ℎ\uD835\uDC57\uD835\uDC61 ) measured by a dummy variable equal to 1 if at least one parent has migrated out at time t, and 0 otherwise; the age and gender of the child at the initial years (\uD835\uDC34 \uD835\uDC56ℎ\uD835\uDC570 ); characteristics of the household (\uD835\uDC4B ℎ\uD835\uDC57\uD835\uDC61 ) and the village (\uD835\uDC49 \uD835\uDC57\uD835\uDC61 ); province-specific year effects \uD835\uDC4C \uD835\uDC5D×\uD835\uDC61 ; and individual fixed effects (\uD835\uDC62 \uD835\uDC56 ). To allow age and gender effects to vary over time to account for the way children grow as they age in the early years, we interact \uD835\uDC34 \uD835\uDC56ℎ\uD835\uDC570 with a time trend T. With the idiosyncratic error term denoted as \uD835\uDF00 \uD835\uDC56ℎ\uD835\uDC57\uD835\uDC61 , the estimation equation can be written as follows:Our conceptual framework suggests that controlling for heterogeneity between households and villages is likely to be important in explaining nutritional outcomes among children. To this end, the rich information in the CHNS is useful because it allows us to include a large set of control variables in the regression analysis. Specifically, the \uD835\uDC4B ℎ\uD835\uDC57\uD835\uDC61 vector includes the years of education of the household head, the number of senior (aged 60 and older) and working-age (aged 16 to 59) household members by gender, household size, and household assets per capita. We include an indicator for the presence of a health clinic in the village and the share of other households in the village that have tap water in the vector of village characteristics (\uD835\uDC49 \uD835\uDC57\uD835\uDC61 ). We also include the village-level market prices of rice, flour, and pork because prices may affect both relative wealth and household diet and therefore may influence both migration decisions and the nutritional status of children. Finally, the province-time fixed effects (\uD835\uDC4C \uD835\uDC5D×\uD835\uDC61 ) control for time-varying macroeconomic conditions that differ by provinces.With this large set of control variables, we still cannot rule out that unobservable traits of a child, such as innate health, can affect both the migration decision of their parents and their growth outcomes. With a panel of children, a reasonable way of controlling for individual-level fixed effects is to difference the two periods. After differencing equation (1), we can write the following: ∆\uD835\uDC3B \uD835\uDC56ℎ\uD835\uDC57\uD835\uDC61 = \uD835\uDEFC 1 ∆\uD835\uDC40 ℎ\uD835\uDC57\uD835\uDC61 + \uD835\uDC34 \uD835\uDC56ℎ\uD835\uDC570 \uD835\uDEFC 2 + ∆\uD835\uDC4B ℎ\uD835\uDC57\uD835\uDC61 \uD835\uDEFC 3 + ∆ \uD835\uDC49 \uD835\uDC57\uD835\uDC61 \uD835\uDEFC 4 + ∆\uD835\uDC4C \uD835\uDC5D×\uD835\uDC61 + ∆\uD835\uDF00 \uD835\uDC56ℎ\uD835\uDC50\uD835\uDC61 .(2)In estimating equation ( 2), any effect of migration on children's nutritional outcomes is identified off changes in migration status among households occurring between survey rounds; however, we still must be concerned that unobservable that vary over time exist and both are important determinants of child nutritional status and affect household migration decisions. Clearly, such variables are not accounted for in equation (2). For example, income shocks resulting from crop failures could affect both migration decisions and the diets of children (or other factors that influence their nutritional status). To identify the impact of migration, we must find variables that affect migration but do not independently affect children's nutritional status, except through their effects on migration. In this context, we suggest the following identification strategy. We hypothesize that relative wage growth in provincial capital cities (∆\uD835\uDC4A \uD835\uDC5D\uD835\uDC61 ) potentially affects migration decisions by households. We further hypothesize that the strength of that signal will be affected by the size of the village migrant network or the initial size of the village migrant network, \uD835\uDC40 -ℎ\uD835\uDC570 , as measured in the first available survey. We specifically measure the migrant network measured as the share of men and women in the working-age population who had migrated in the initial year. 10 The change in migration can therefore be written as follows:∆\uD835\uDC40 ℎ\uD835\uDC57\uD835\uDC61 = \uD835\uDC53(∆\uD835\uDC4A \uD835\uDC5D\uD835\uDC61 ) + \uD835\uDC5F\uD835\uDC40 -ℎ\uD835\uDC570 +\uD835\uDC40 -ℎ\uD835\uDC570 × \uD835\uDC53(∆\uD835\uDC4A \uD835\uDC5D\uD835\uDC61 ).(3)The instrumental variables strategy then follows. We hypothesize that wage growth and the migrant network individually affect migration decisions; however, on their own, each variable also might be correlated with children's nutrition outcomes even after we account for migration. Wage growth in the provincial capital might be correlated with local wage growth, which could be reflected in changes in anthropometric measures through the income effect. Similarly, households with large migrant networks might have better access to information on nutrition or hygiene practices that affect anthropometric outcomes independently of migration; however, we argue that the interaction of the migrant network variable and provincial capital wage growth will affect only migration, and once we control for the migration network and the provincial capital wage growth directly, the variation in the interaction term will affect migration and will not affect independently children's anthropometric outcomes. We note that in equation ( 2), we implicitly control both for wage growth ∆\uD835\uDC4A \uD835\uDC5D\uD835\uDC61 , through the province-specific year effect (∆\uD835\uDC4C \uD835\uDC5D×\uD835\uDC61 ), and for \uD835\uDC40 -ℎ\uD835\uDC570 as it is included in the fixed effect and is therefore differenced out. The interaction term alone (∆\uD835\uDC4A \uD835\uDC5D\uD835\uDC61 × \uD835\uDC40 -ℎ\uD835\uDC570 ) is therefore used for statistically identifying migration.Intuitively, if we control for wage growth, migration is more likely among individuals with larger migrant networks, and therefore their children are more likely to be left behind. We confirm that this relationship exists in the data in Figure 4.4, in which we stratify villages in the sample into two groups: one with larger migrant networks in the initial year and a group with smaller migrant networks in the initial year. We define the cutoff for large migrant networks as villages in which 15 percent of workingage women or 20 percent of working-age men migrated in 1997. The graphs illustrate that in villages with larger migrant networks, migration grows faster with wage growth but that the same relationship is weaker in villages with smaller migrant networks. Moreover, this pattern holds for both men and women. This approach essentially exploits the complementarily between urban wage and migrant network in an individual's migration decision. Under the assumption that the interaction variable is independent of ∆\uD835\uDF00 \uD835\uDC56ℎ\uD835\uDC50\uD835\uDC61 in equation ( 2), our estimate of \uD835\uDEFC 1 would be unbiased. We primarily test this assumption by checking the robustness of the estimates to the inclusion of various sets of variables that might confound the relationship between migration and the interaction variable. We first examine HAZ scores as the dependent variable and estimate equations ( 1) and ( 2), respectively, treating parental migration as exogenous. The results are reported in Table 5.1. In the ordinary least squares regression, household head's years of schooling, the number of working-age women, household assets, and the percentage of other households in the village having tap water are shown to be positively correlated to children's HAZ, but these correlations cease to be significant in the first differencing estimation. Two estimations yield small coefficients with large standard errors on the measure of migration, which is an indicator variable for whether a parent of the child migrated. The results seem to suggest that parental migration status may not affect children's cumulative nutrition. But these point estimates are likely to be biased because the endogeneity of migration is not accounted for. Next, we use the interaction terms described above as instrument variables for migration, and we re-estimate equation ( 2) using an instrumental variables estimator (see Table 5.2). The instruments are gender-specific interactions between growth in wages in the provincial capital and the initial village migration rate. With the full set of control variables for age and gender of the child, household characteristics, and village characteristics, we find that the instruments are strongly correlated with changes in migration status among parents (column 1); both have z statistics between 2 and 3, and the F statistic on the joint hypothesis test that both coefficients are zero is 7.8. The under identification test and the over identification test are passed comfortably, but under standard assumptions about the error terms, the instruments would be considered weak (for example, Stock and Yogo 2007); however, in the presence of clustering, it is less clear whether an F statistic between 7 and 8 is associated with weak instruments (for example, Finlay and Magnusson 2009). In the second stage, we again find no significant relationship between the migration variable and child HAZ (see Table 5.2, column 2). As expected, the standard errors rise whereas the point estimate remains small. Consistent with results reported in Table 5.1, these results confirm that parental migration does not affect children's HAZ scores. Although migration does not affect the cumulative measure of nutrition, it may affect shorterterm measures of nutrition status, such as BMIZ and WAZ scores. The estimated results of these variables based on equation (2) with the instrument variable method are reported in Table 5.3. We find a point estimate of 0.102 on parental migration status for BMIZ, but it is not statistically different from zero. When we estimate the model using changes in WAZ scores as the dependent variable, we find statistically significant coefficients of a reasonable magnitude: 0.201. This estimate essentially suggests that the children of migrants are heavier than the children of others, ceteris paribus, at the same age among children of the same gender. Given the absolute lack of significance when using HAZ scores as the dependent variable, migration does not appear to affect height; so these estimates lend credence to the ideas that BMIs are increasing and that impact is on short-term instead of long-term nutritional measures. This line of argument would suggest that our inability to reject the null hypothesis that the coefficient estimate of the BMIZ scores is different from zero is the result of a lack of statistical power in the regression. In summary, our primary regression results indicate that parental migration has a positive effect on children's weight but has no impact on height. Next we check the robustness of our estimates to potential misspecifications and confounding factors. We will then explore the mechanisms by which this effect may have taken place.One of the primary concerns about our results is that changes in household characteristics may be endogenous to children's nutritional status. For example, the change in household assets may be correlated with unobserved shocks that could also affect changes in children's nutritional outcomes; moreover, asset changes or changes in household demographic compositions may be correlated with the migration decisions of household members. Such correlation may lead to biased estimates of migration. To rule out the possibility that the above results are driven by changes in endogenous household characteristics, we estimate the regressions without including household characteristics as control variables. The results are reported in Table 5.4.Compared to the previous estimates with household-level control variables, the coefficients on migration in Table 5.4 are smaller in all three regressions. One possible reason for such difference is that parental emigration negatively affects the number of household members, whereas the number of household members, female members in particular, seems to be positively correlated with children's nutrition outcomes.11 Albeit the change in the magnitude, the results are consistent with the previous finding that parental emigration positively affects children's weight but not their height. We are also concerned about potential panel attrition bias. The CHNS sample attrition is high for children aged zero to five, particularly between 2004 and 2006 (see Table 5.5). If the attrition is systematically correlated with children's nutrition status and parental migration decision, then the estimation based on the remaining panel would be biased. In Table 5.5 we compare these variables between children who were surveyed in two consecutive rounds with those who dropped after the initial years, and we find neither statistically significant differences in average anthropometric scores measured in the initial years nor differences in the propensity to come from a migrant household. For this set of observables, panel attrition would therefore appear to be random. Nonetheless, we re-estimated the main IV regressions using a correction for attrition as follows. First, we estimated the probability that individuals stayed in the sample, given initial characteristics of individuals (Table A.2). We subsequently use the inverse of that estimated probability as a weight in the regressions (Wooldridge 2010). The intuition is that by giving more weight to children who appear less likely to remain in the sample, we make the analysis sample more representative of the original one. Continuing to control for the full set of other explanatory variables (see Table 5.6), we find that the results are largely unchanged. We find statistically significant estimates for the effect of migration on WAZ scores but not on BMIZ or HAZ scores.12  Given the pattern of results, we want to understand how migration might have affected children's weight. As discussed in section 3, weight and height could be affected by an increase in the quantity and quality of food consumed, respectively. When we estimate the effects of migration on caloric intake in a framework outlined in equation 2, we find no evidence that either the quantity or the quality of young children's diet has changed because of the migration of their parents (see Table 5.7). 13A second potential mechanism for increased weight discussed in section 3 is through improved hygiene. To measure whether various measures of hygiene could explain our findings, we estimate equation (2) using three measures of the children's living environment as the dependent variable (Table 5.7). We find no statistically significant relationship between migration and having a flush toilet in the house or having any excreta around the house. We do, however, find that households with migrants are 7 percentage points more likely to have a water tap in the house or in the yard by the second survey relative to non-migrants. Finally, we find no relationship between child healthcare variables, namely, immunization and preventive health service, and parental migration. So it seems that the plausible mechanism we can identify is through increases in the availability of tap water. In this paper, we study the effects of parent migration on young children's nutritional status in rural China. We find that migration is positively associated with short-term measures of nutritional status. According to our within-individual differenced and instrumented results, parent migration is associated with an increase of WAZ-scores of between 0.08 and 0.2 standard deviations. The estimated coefficients on HAZ and BMIZ-scores are positive but not significantly different from zero. These basic results are robust to different specifications and to a standard correction for panel attrition. In exploring possible mechanisms through which these impacts might occur, we provide suggestive evidence that migrant households are more likely to have tap water, which is likely to reflect an income effect.Even though young children's nutritional status does not seem to be compromised by the absence of their parent(s), we are not optimistic that children left behind can benefit unambiguously from parent migration. First, our previous study ( de Brauw and Mu 2011) shows that once children grow, they will have to take up more household chores. Any advantage they may have in WAZ when young will disappear. Second, the anthropometric measures, although easy to gauge, do not capture many important dimensions of health. For example, they cannot account for any psychological impact of parental migration on children.However, it is important to note that migration seems to have a positive impact on household consumption of public goods, such as tap water. Further research is needed to analyze the mechanisms through which this could happen or more generally how migration is related to the provision and use of public goods.  ","tokenCount":"6040","images":["1498571560_1_1.png","1498571560_13_1.png","1498571560_14_1.png","1498571560_15_1.png","1498571560_17_1.png"],"tables":["1498571560_1_1.json","1498571560_2_1.json","1498571560_3_1.json","1498571560_4_1.json","1498571560_5_1.json","1498571560_6_1.json","1498571560_7_1.json","1498571560_8_1.json","1498571560_9_1.json","1498571560_10_1.json","1498571560_11_1.json","1498571560_12_1.json","1498571560_13_1.json","1498571560_14_1.json","1498571560_15_1.json","1498571560_16_1.json","1498571560_17_1.json","1498571560_18_1.json","1498571560_19_1.json","1498571560_20_1.json","1498571560_21_1.json","1498571560_22_1.json","1498571560_23_1.json","1498571560_24_1.json","1498571560_25_1.json","1498571560_26_1.json","1498571560_27_1.json","1498571560_28_1.json","1498571560_29_1.json","1498571560_30_1.json","1498571560_31_1.json","1498571560_32_1.json"]}

data/part_2/0277038877.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"9708f756310aa48abc4cde1c90b4dd17","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/75ed436e-d76b-4ed0-808d-1597416c09e0/retrieve","description":"This brief summarizes nutrition-relevant policies in Liberia. We examine i) nutrition context, objectives, indicators, budget, and activities, ii) key beneficiaries, actors, and coordination, iii) monitoring, evaluation, and accountability, and iv) whether current policies are aligned with the World Health Assembly (WHA) global targets.","id":"-1459981574"},"keywords":[],"sieverID":"1de75eea-23dd-42b6-9c95-cf169d7bf71e","pagecount":"14","content":"• To strengthen understanding of the current direction of nutrition-relevant policy in Liberia and its implications. It was developed in response to partners' request and priorities.Liberia is on track to meet some but not all of the World Health Assembly (WHA) 2025 targets. Liberia has already met two of the WHA targets for children under five years of age (U5), namely U5 wasting (3.4% in 2019 i ) and U5 overweight (4.4% in 2019 ii ). There is no recent data on exclusive breastfeeding (EBF). However, by 2013 the 2025 target had already been met (54.6% in 2013 iii ). Liberia is not on-track to achieve the WHA 2025 target on anaemia in women of reproductive age (WRA) (43.6% in 2012 and 42.6% in 2019 iv ) nor on U5 stunting (39% in 2010 and 29.8% in 2019 v ). Data is insufficient to assess Liberia's progress towards achieving the WHA target on Low Birth Weight (LBW).Seven nutrition-relevant policies currently in use or in the advanced drafting stage are included in this brief (see Table 1). They are in the areas of nutrition (n=1), health (n=2), agriculture/food security (n=2), water, sanitation, and hygiene (WASH) (n=1) and economic/social (n=1). No nutrition-relevant policies identified in the areas of education/research, environment/climate/resource management, or other cross-cutting policies (e.g. gender/family, governance, etc.) were found to be sufficiently nutritionoriented following their assessment based on the policy review's inclusion criteria and were therefore excluded from this brief. All nutrition-relevant national policies, strategies, and action plans currently in use or in the advanced drafting stage as of September 2020 were included in this brief. Inclusion criteria were the presence of a nutrition objective, a budget for nutrition, and/or a nutrition indicator. Policies were not included in our analysis when i) we did not have access to the policy documents; ii) they were released or updated after expert consultation (September 2020).We obtained potentially relevant documents from a systematic search that included preidentified websites (e.g., relevant national government ministries, United Nations agencies and nongovernmental organizations), a Google search, a reference search, and country expert consultation. Targeted consultations with regional and in-country experts were used to access documents not available online and for validation. We screened identified documents (see Annex 1) against our eligibility criteria. Seven documents met our inclusion criteria. Coding, data extraction, and content analysis for these documents was carried out with NVivo qualitative analysis software and Excel.All policies except one (NSRHP) provide some contextual information on nutrition. Across the policies identified, analyses of the nutrition situation focus predominantly on the national level. More than half of the policies (across all areas except for health) report on the regional or global context, including reference to the 2004 \"Kampala Vision 2020 Summit\", the MDGs, the Convention on the Right of the Child and the Regional Agricultural Investment Program (RAIP). Two policy (NNP), in the nutrition area, also reports on the subnational context, and is one of two policies (alongside the LASIP II), which acknowledge rural/urban and regional disparities in the Liberian nutrition context. The NNP also mentions age and gender disparities, but none of the policies present any sexdisaggregated data by age group.Across policy areas, the focus is starkly on undernutrition, particularly stunting. Only one policy (NNP) presents contextual information on micronutrient deficiencies, namely vitamin A and iron deficiency, while none of the policies present data on overweight/obesity or diet-related factors that contribute to non-communicable diseases (NCDs). Whilst the NNP, a nutrition-specific policy, presents a more holistic picture of nutrition problems than policies from nutrition-sensitive policy areas, the data available is generally either missing or outdated.Most of the policies across all sectors outline causes and/or consequences of nutrition problems. Causes include environmental/WASH factors, population growth, changes in diets and lifestyles, poor dietary breastfeeding and complementary feeding practices, food and nutrition insecurity, changes in rainfall patterns, low crop yields, depletion of natural resources (e.g. through tree cutting for firewood and charcoal), shift to rubber and planting of new trees on disputed land, lack of equitable land tenure system and access to water and pasture resources, lack of policy on land acquisition for agricultural purposes, sea level rise affecting livelihoods along coastal areas where the majority of Liberians live, low agricultural production due to weak and inappropriate technologies and resulting food insecurity, poor road networks leading to poor market integration and high commodity prices, poor access to nutritious food, poor access to health services, limited and difficult access to financial schemes for improving farmers' agricultural capacities, limited access to and control over natural resources (particularly land), disease burden, weak coverage of social protection (which can sustain the livelihoods of poor households and increase their capacities for resilience to crisis), loss of agricultural production, low dietary diversity, low educational, entrepreneurial and technical skills, poverty and negative coping strategies in response to even small shocks (resulting in low school attendance, poor health and nutrition, asset sale and increase in child labor). Consequences include morbidity, mortality, vulnerability to and risk of death from infections such as diarrhoeal diseases, malaria, pneumonia, measles and HIV/AIDS, lowered physical and mental capacities, cognitive disabilities, adverse outcomes in productivity and asset accumulation, hindered social, economic and human development and, finally, poverty, which is acknowledged as both a cause and a consequence of nutrition problems. The NNP presents information on the estimated cost of malnutrition, but the data presented is outdated and is therefore inadequate to inform a reassessment of the situation in present day Liberia. Overall, the nutrition context is generally not evidencebased and adequately referenced. The NNP mentions several references but the data presented is outdated (as the policy started to be used in 2008, most of the data refers to surveys implemented between 2000 and 2008). Across all policy areas where references are reported, these refer to statistics as well as textual information. Cited data sources for evidence on nutrition disparities, causes and consequences of nutrition problems include the As shown in Table 2, five of the seven policies, across almost all policy areas, include nutrition in their general and/or specific objectives. These objectives contain nutrition-specific (e.g., improving the nutritional status of the population) and nutrition-sensitive content (e. Table 2 shows one policy with nutrition indicators that coincide with WHA indicators (LASIP II), namely U5 stunting and wasting. The policy sets 2022 as its target date. If met, these would expectedly put Liberia on track to achieve, or even surpass, the WHA key nutrition targets by 2025. None of the policies include U5 overweight/obesity in their indicators, reflecting a tendency across policies to focus on undernutrition, often in general terms, without further data disaggregation for key target groups.Policies with nutrition objectives would be expected to include both planned nutrition activities and nutrition indicators, while policies without nutrition objectives would be expected to include neither. Yet there are several instances (see Table 2) where this is not the case. Generally, this is not necessarily due to a lack of coherence within policies but because a) policies' objectives are broad and do not explicitly link to nutrition (while their indicators or planned activities are specific enough to make this link explicit), or b) indicators and/or planned activities are to be addressed in a separate programmatic document (which is sometimes noted in the main policy document). There are, however, some cases where there is incoherence within different parts of the same policy. For instance, only two of the policies include indicators. Namely the WASHSSP, includes coverage indicators clearly linked with the listed activities, although there are no objectives that align with the challenges identified in the nutrition context, which is however covered through the inclusion of relevant planned activities. The LASIP II, includes nutrition indicators, as well as activities that are clearly aligned with the stated objectives of the policy, although the challenges identified, and the objectives, are very broad, and thus weaken the overall coherence of the policy. Both the NHSWP, in the health area, and the NSPPS, in the social protection area, show strong coherence of planned nutrition activities with the policy's objectives. The latter, however, are very broad in both policies, making alignment with the nutrition context, in itself not sufficiently detailed, less straightforward. In addition, these two policies do not provide any detailed nutrition or coverage indicators. Two policies, namely the NNP and the FAPS, respectively in the nutrition and agriculture/food security areas, show some degree of coherence between nutrition objectives and activities, as well as partial coherence with the nutrition context in the case of the FAPS, and stronger coherence in the NNP. However, neither of the policies link activities with nutrition or coverage indicators, impacting negatively on overall internal coherence. Finally, the NSRHP, in the health policy area, scores poorly in internal coherence across all process steps, with no clear alignment between the nutrition context and policy objectives, activities and indicators. 3 Not applicable (NA) indicates policies that do not have sufficiently detailed budget information to assess whether nutrition is included, while ± _is used for policies that provide sufficient budget information but with no mention of nutrition. Children, particularly under five years of age (with subgroups 0-6 months, 6-9 months and 6-35 months in the NNP) feature most often in the policies' nutrition context (n=5), followed by women (including WRA and adolescent girls) (n=2). Elderly people are mentioned in one nutrition policy, namely the NNP. Men are not specifically mentioned in any of the policies' analyses on the nutrition status of the population. Additional groups targeted by the NNP are people living with HIV and tuberculosis. Apart from the NNP and the LASIP II, which highlight regional and urban/rural disparities, no policies detail geographic disaggregation. In addition, the NNP is the only policy that reports data on nutrition status disparities with relation to socioeconomic, gender and age sub-groups, making the nutrition situation analysis of this policy comparatively more holistic.As shown in Table 3, primary beneficiaries of policies vary by area. Overall, the most frequently targeted primary beneficiaries are vulnerable groups, followed by children and women (particularly pregnant and lactating women (PLW). Nutrition and health policies target specifically for these two groups, who feature as either primary or secondary beneficiaries in these policies. Beneficiaries of policies in the agriculture/food security area cover farmers (with particular attention to sub-groups such as youth and women involved in agriculture), as well as small to large scale farmers.Vulnerable groups are a key group targeted by agriculture/food security, WASH and economic/social policies, including sub-groups which span across age groups, from and adolescents to adults, and the elderly (see Table 3).As shown in Table 3, the national government is the primary actor, featuring in all of the policies (n=7), and involved in policy development, management and coordination, financing, implementation, monitoring and evaluation. Extensive roles also assigned to civil society, NGOs, technical and financial partners (n=7), followed by communities, (n=6), the private sector (n=5) and local government (n=1). The actors involved in coordination and management are the national government (n=6) and civil society, NGOs, technical and financial partners (n=6), with minor involvement of the private sector (n=2) and communities (n=2). Implementation roles are distributed between the government (n=7), communities (n=6), private sector (n=5), NGOs, civil society, technical and financial partners (n=5). Beyond national governmental bodies, M&E roles also see the participation of civil society, NGOs, technical and financial partners (n=5) and communities (n=2), with no involvement of the private sector. which covers financing roles (n=4).The importance of multisectoral coordination is highlighted across most of the policies and policy areas (namely, in all policies except the NHSWP).Coordination mechanisms include guiding principles, multi-actor and multi-sector steering and technical committees and groups, documents and tools, a joint collaborative implementing framework for capacity building at all government levels for the implementation of focused interventions, provision of facilities and workshops, experience and information sharing, consultations and collaboration with relevant public and private institutions, awareness raising activities for the mainstreaming of nutrition, inter-ministerial partnerships, horizontal linkages for fostering alignment, cooperation between government and international agencies, government leadership to ensure coherent action, resource mobilization and resource sharing, guards against sub-standard interventions/wasted investments/duplicate efforts, and dissemination of progress reports. Although the specific challenges and strengths associated with multisectoral coordination are not always specified, commitment to establishing and sustaining effective coordination mechanisms is emphasized across most policies.  Accountability mechanisms are also mentioned in all of the policies except one, namely the FAPS. These mechanisms pertain mainly to institutional and mutual accountability between involved actors, and to some extent to social accountability. They include: guiding principles, promotion of good governance, regular consultation and prioritization at central, district and county levels, establishment of appropriate venues for discussion, equipment of county authorities to assume responsibility for delivery of nutrition services, clear lines of accountability, timelines, indicators for performance tracking, joint annual sector reviews, strengthening the coordination mechanism for mutual accountability, transparent use of resources, accurate and timely dissemination of information, publicity of instances involving abuse of the system, disclosure of the contract terms and unit costs of agencies selected to administer interventions, multi-stakeholder consultations (beneficiaries, civil society organizations, local and international NGOs, private sector), inclusive and constant communication, collective responsibility and inclusive participation, conflict resolution mechanisms, transparency in targeting approaches, publishing lists of selected recipients, creating an independent appeal process, undertaking random spot checks, independent audits for improving transparency and reducing inclusion and exclusion targeting errors, communication of emerging evidence on best practice, mechanisms for redress.This policy note is intended to inform national decisions makers, policymakers and a wider audience including implementing partners across all relevant nutrition sectors. Its analysis can help to better understand gaps and incoherence within existing policies. Furthermore, the recommendations emanating from this analysis can inform revisions of existing or the development of new nutrition-relevant policies to improve impact on nutrition in their country.Recommendation 1: Address gaps and incoherence in nutrition-relevant policy.The analysis above highlights a number of gaps and incoherencies in current nutrition-relevant policy in Liberia. Future policies or revisions could:• Ensure that nutrition context, objectives, indicators, and /or planned activities align, in terms of nutrition problems and targeting of populations (e.g., nutrition objectives target several different groups but nutrition indicators only measure progress for some of these groups). This would allow to achieve better coherence within policies, introduce well-aligned impact pathways, from broad objectives to specific indicator measures, and enable identification of gaps and challenges, leading to more effective targeting.• Improve the evidence base (including SMART data to address gaps of missing and outdated statistics), and better define nutrition concepts and indicators to allow for common understanding across actors and policy areas, as well as coherence in measurement of indicators. Only few policies highlight nutrition disparities across regions, gender, urban/rural and socioeconomic status; even if some policies targeting vulnerable populations focus on specific beneficiary groups, disaggregated nutrition indicators and targets are not clearly defined. Ideally, indicators are also disaggregated by gender, geographic area and between urban and rural settings, to capture the disparities identified in a policy's context analysis, and to ensure effective progress tracking.• Invest more in inclusion of marginalized and/or underrepresented population groups. The policies we assessed provided limited nutrition context information on adolescents, men and the elderly. The policies can benefit from more inclusive consideration these groups, as they play an important role in contributing to a child's growth, development and life chances, calling for their involvement in activities addressing children's nutrition• Invest in fighting malnutrition in all its forms in Liberia by capitalizing on shared drivers, entry points and delivery platforms. In order to curb current trends in malnutrition, namely the coexistence of multiple forms, a holistic lifecycle approach is essential to address causes and consequences of malnutrition and disease burden in the country. A rising burden of overweight/obesity and diet-related NCDs across the region calls for targeting not only key groups through interventions that will impact on a child's life course, but also on the lifecycle in terms of different age groups currently being affected by different and often coexisting burdens of malnutrition, including within the same households and communities• Ensure clear budget allocation plans for nutrition across nutrition-relevant policies and sectors. Most of the policies we assessed lacked clearly defined nutrition budgets, although budgetary information may be provided in some form in additional documents. Overcoming this limitation is crucial for meeting the WHA targets, or at least for narrowing the gap between these and the current situation.Recommendation 2: Continue to invest in strong multisectoral coordination.Strengthening multisectoral coordination and actions across sectors, ministries, and departments will be essential for achieving the WHA targets in Liberia.Multisectoral and multi-actor coordination is the basic guiding principle of governance for most of the nutrition policies included in this note. Despite the presence and the importance of multisectoral coordination highlighted in most of these policies, significant challenges for its functionality were mentioned. Leadership can be strengthened by clearly defining the roles of all actors at a higher hierarchical level with an authority over all of the contributing sectors. The application of strong vertical and horizontal coordination mechanisms would buttress the country's potential to achieve the WHA targets.Recommendation 3: Mainstream nutrition into future documents across diverse policy areas.Only some policies adequately cover nutrition by including nutrition-oriented objectives and actions. The remaining policies could improve the integration of nutrition into their nutrition context, objectives, planned activities, indicators, and budgets. To begin mainstreaming nutrition into future policies and operational documents into diverse policy areas, policymakers could refer to the gaps identified throughout this policy review. This includes missed opportunities in sectors excluded from this synthesis because the policies identified were not sufficiently nutrition-oriented (namely education/research, environment/climate/resource management, or other cross-cutting policies (e.g. gender/family, governance, etc.)). Strong multi-stakeholder engagement across the policy landscape is essential for ensuring that nutrition is integrated across sectors to create and sustain an enabling environment for tackling malnutrition.Recommendation 4: Recognize nutrition as a cross-cutting area in ongoing policy drafts/revisions.The revision of existing policies and the drafting of new ones provides an opportunity for better integration of nutrition through the alignment of activities and indicators with the nutrition issues, objectives and target groups indicated in the policies. By incorporating the above recommendations, any new or revised policy could contribute to advancing nutrition at national level.","tokenCount":"3017","images":["-1459981574_1_1.png","-1459981574_1_2.png","-1459981574_1_3.png","-1459981574_1_4.png","-1459981574_1_5.png","-1459981574_1_6.png","-1459981574_1_7.png","-1459981574_4_1.png","-1459981574_5_1.png","-1459981574_6_1.png","-1459981574_6_2.png","-1459981574_6_3.png","-1459981574_6_4.png","-1459981574_7_1.png","-1459981574_7_2.png","-1459981574_7_3.png","-1459981574_7_4.png","-1459981574_7_5.png","-1459981574_7_6.png","-1459981574_7_7.png","-1459981574_7_8.png","-1459981574_7_9.png","-1459981574_8_1.png","-1459981574_11_1.png","-1459981574_14_1.png","-1459981574_14_2.png","-1459981574_14_3.png"],"tables":["-1459981574_1_1.json","-1459981574_2_1.json","-1459981574_3_1.json","-1459981574_4_1.json","-1459981574_5_1.json","-1459981574_6_1.json","-1459981574_7_1.json","-1459981574_8_1.json","-1459981574_9_1.json","-1459981574_10_1.json","-1459981574_11_1.json","-1459981574_12_1.json","-1459981574_13_1.json","-1459981574_14_1.json"]}

data/part_2/0292949950.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"5f4ef462088287bb49852a3785d81f15","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/a8d1b399-ea69-436f-aec7-d3b45011a196/retrieve","description":"Problems caused by eating unsafe food are a major health issue in many countries. Contamination by bacteria or toxins can cause these health risks, particularly for young children whose bodies are still developing. It can be difficult to know which foods are risky because you usually cannot tell by looking if food is contaminated. The goal of this research was to see whether giving consumers in Kenya information about one of these contamination risks would make them choose to purchase different products.","id":"-1490948247"},"keywords":[],"sieverID":"1cf30546-d438-42cd-8e13-c9a251a471ac","pagecount":"4","content":"Problems caused by eating unsafe food are a major health issue in many countries. Contamination by bacteria or toxins can cause these health risks, particularly for young children whose bodies are still developing. It can be difficult to know which foods are risky because you usually cannot tell by looking if food is contaminated. The goal of this research was to see whether giving consumers in Kenya information about one of these contamination risks would make them choose to purchase different products.We focused our study on aflatoxin, a common food safety hazard in Kenya. Consuming foods very high in aflatoxin can cause liver disease and even death. Over time, lower levels of exposure to aflatoxin increase the risk of liver cancer and may have negative effects on children's immune system function and growth.For one year, we purchased different types of maize flour every two months at different locations in Kenya. We found that throughout the year, posho flour -whether from the consumer's own farm, the miller's supply, or purchased grain -was much more likely to be contaminated above the regulatory limit of 10 parts per billion than packaged unga. Extra-sifted posho flour was about as safe as packaged unga. During processing of both packaged unga and extra-sifted posho, the parts of the grain where aflatoxin tends to be concentrated are mostly removed. Unfortunately, these are also the components of the grain that contain most of the fibre, vitamins and minerals in maize.MARCH 2023 We interviewed members of 1450 households in low-income neighbourhoods of Nairobi and Athi River to get information about how consumers think about food safety risks and about the types of foods they consume. We randomly chose some of these households and provided them with information about the higher aflatoxin risk of posho flour at the end of the interview.We also gave these households a poster with the information on it as a reminder. We came back to do another interview two months later to see if households who got the information changed the type of maize flour they purchased, compared to households in the control group, who did not receive this information.The poster given to randomly selected householdsWhen the research team revisited the same households two months later, half of households who received information were now consuming safer packaged maize flour, compared to one in three people who were not given information. Those given information also showed more understanding of food safety risks. We also found some evidence that people had shared information with their neighbours.Simply providing consumers with information led them to make safer food choices. Conducting regular food safety surveillance, and then providing the public with information about relatively safer versus riskier foods, can have a big impact on food choices. This approach has the potential to help consumers avoid risky foods. By expanding the market share of food businesses providing safer options, provision of this type of information can also increase the average safety of the food supply and create incentives for businesses to invest in food safety.For more information about this study, please contact Dr. Sarah Kariuki at [email protected] or Michael Murphy at [email protected].  ","tokenCount":"521","images":["-1490948247_1_1.png","-1490948247_2_1.png","-1490948247_2_2.png","-1490948247_3_1.png"],"tables":["-1490948247_1_1.json","-1490948247_2_1.json","-1490948247_3_1.json","-1490948247_4_1.json"]}

data/part_2/0306096371.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"d51f6aadc48e4b5ac02761627293be9f","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/e57e9b85-a06d-4ec2-bfc8-31fbc6b9827b/retrieve","description":"The number of people living in rural areas of low and middle-income countries is projected to increase in the coming decades. It is in the rural areas of these countries where a large majority of the world’s extreme poor reside. The livelihoods of two to three billion rural people depend on small farms. These small farms are responsible for the production and supply of a large portion of the calories feeding low- and middle-income countries. Small farms are also preservers of crops and associated biodiversity and with the right incentives can contribute to land stewardship. Small farms are diverse, and, hence, so are their associated challenges. We categorize small farms as commercial farms, small farms in transition and subsistence-oriented farms and highlight evidence-based innovations for the sustainable transformation of each type of small farm. Broadly, small farms face high transaction costs, lack collective action, and experience coordination failure in production and marketing. Lack of market access is also a major challenge. Investments in infrastructure, including those that support access to digital technologies, can improve farmers’ access to markets and incentives as well as foster growth in the midstream segments of the value chain that provide inputs, storage, processing, and logistics to small farms. Rural Non-Farm Employment (RNFE) is increasingly the main source of income for most small farmers and provides them with a risk diversification strategy and cash, both to purchase food and for farm investments to raise productivity, expand commercial activities, and produce higher-value products. Public investments and policies that facilitate growth of the agrifood system must pay more attention to creating enabling environments for the development of RNFE and strengthening the synergy between agriculture and RNFE in rural areas.","id":"378029000"},"keywords":[],"sieverID":"c162a5a8-a6c7-431b-ade8-53ca83d7e91c","pagecount":"15","content":"By 2050, the United Nations projects that 68% of the world population will live in cities (UN DESA 2019). However, with continuous population growth, the number of people living in rural areas of many low-and middle-income countries (LMICs) will continue to rise. Two-thirds of the extreme poor live in rural areas (World Bank 2016), and the livelihoods of two to three billion rural people, often the most food insecure and vulnerable, still depend primarily on small farms (Laborde Debucquet et al. 2020;Woodhill et al. 2020).There are various estimates of the number of small farms in the world, but they all suggest that these farms are numerous. Lowder et al. (2016) used agricultural census data from 167 countries to estimate that, of the total 570 million 1 farms in the world, 475 million have less than 2 hectares (ha), dominating agriculture in most LMICs, where farm sizes continue to fall. Africa south of the Sahara has the highest rural population growth rate globally, and thus the number of small farms is expected to increase more than in other regions. Africa's share of total world rural poverty is also expected to rise from 39.6% in 2015 to 58.1% in 2050 (Thurlow et al. 2019). Transforming Africa's agriculture sector is thus a priority embodied in the Malabo Declaration on Accelerated Agricultural Growth and Transformation for Shared Prosperity and Improved Livelihoods (AU 2014). However, to meet the Malabo goals and achieve multiple SDGs in all LMICs by 2030, creating an enabling environment where small farms are included in and benefit from rapid growth and transformation of agrifood systems is urgent (Barrett et al. 2020a).Small farms not only contribute to feeding the households that operate them, but also make two broader contributions. First, small farms are important to the overall food security of LMICs. Samberg et al. (2016) noted that farms less than 5 ha are responsible for 53% of the global production of food calories for human consumption. Herrero et al. (2017) reported that, in Africa and South and Southeast Asia, small farms with less than 2 ha produce around 30% of food and make valuable contributions to micronutrient-rich food production. Ricciardi et al. (2018) estimated that farms under 2 ha globally produce 30-34% of the food supply. Nonetheless, small farm households themselves are often unable to afford a nutritious diet.Second, small farms contribute to the sustainability of agrifood systems by maintaining the genetic diversity of crops and livestock and supporting ecosystem services. Small farms have more crop diversity and harbor greater non-crop biodiversity at the farm and landscape scales than do larger farms (Ricciardi et al. 2021). Subsistence-oriented small farmers plant a greater diversity of traditional crops and maintain genetic resources by cultivating land races (Fifanou et al. 2011;McCord et al. 2015). Small fields have more edges than larger fields, creating a heterogeneous landscape and providing habitat for non-crop species (Ouin and Burel 2002). To the extent that small farms have more tree cover than larger farms, they provide aboveand below-ground carbon storage, with global benefits for climate mitigation (Ritchie and Roser 2017). Trees on farms can also improve water infiltration, a hydrological service that benefits other water users in the landscape and downstream (Anache et al. 2019).For small farms to be part of inclusive and sustainable agrifood system transformation, both innovative technology and market institutions are required to support LMICs' diverse agroecological and socioeconomic contexts. Many debates on the future of small farms focus only on farm production, rather than the whole context of farm household livelihoods, which include off-farm activities, or the agrifood system on which farms depend for buying inputs and selling outputs (Reardon et al. 2019;Giller et al. 2020). The future of small farms should instead be assessed using a holistic livelihoods and agrifood system lens.2 Who Will be Small Farmers in the Future?More than 410 million farms are very small, with less than 1 ha of land, and another 70 million are between 1 and 2 ha (Lowder et al. 2016). However, discussions of farm size often ignore land quality considerations (Eastwood et al. 2010). For example, a 5 ha farm in a rainfed zone with poor quality soil may support less production than a 1 ha farm in an irrigated zone with good soil. Thus, mere farm size ranges tell us nothing about differences in agroecological land quality, or about the socioeconomic contexts in which they operate, such as market and infrastructural conditions (FAO 2014;Graeub et al. 2016). While the product mix of small farms varies depending on this context, many are diversifying that mix, driven by urbanization, consumers' dietary preferences, technology, infrastructure development, and rural-urban links. Moreover, households that operate small farms tend to have diversified income sources, including non-farm activities, and that diversification is expected to increase over time, although at different rates among different sets of small farmers (Davis et al. 2017).Despite the strong heterogeneity across small farms, they can be categorized in ways that make our analysis more tractable. Following Vorley (2002) and Hazell (2018), and based primarily on Hazell (2020), we classify small farmers in LMICs into three groups.Commercial small farmers run their farms as businesses. While commercial agriculture is an important source of income for them, many also undertake rural non-farm employment (RNFE). Most commercial small farmers do not specialize in high-value crops or livestock, as many also produce food crops. Their product and activity mix are conditioned by agroecological circumstances, urban market proximity, rural infrastructure, and the agro-processors, logistics, exporters, and wholesale enterprise investment and density in their area. Climate change and economic transformation also condition their farm businesses and will create new challenges and opportunities even over the next 10 years. Some commercial small farms will continue to focus on today's traditional export crops-for example, cocoa in Ghana, cotton in Mali, and coffee in Ethiopia-while increasing numbers will turn to products that cater to the diversifying diets of burgeoning domestic urban markets, including fruits, vegetables, fish, poultry, edible oils, milk, and feed grains such as soy. Non-cereal products are especially labor-demanding and often offer little or no economies of scale, allowing small farms to be competitive. Over time, we expect to see greater specialization in the farming of high-value products and a movement away from the combination of cash and staple crop farming, similar to what one sees among specialized vegetable farmers in the Shandong province of China (Huang et al. 2010) or specialized poultry and pig farmers near Yangon in Myanmar (Belton et al. 2020).Small farmers in transition often depend heavily on RNFE while also maintaining small plots for home food consumption, plus some semicommercialized food or non-food products. They tend to buy a substantial share of their food. These farmers are in zones where favorable non-farm opportunities exist locally or in near-by towns. With demand growing for high-value farm products in cities, some transitional farmers will commercialize their small farms while continuing their RNFE. However, others may exit agriculture or maintain just small food plots because access to food markets in their area is uncertain, or because the RNFE labor market itself is uncertain or limited (de Janvry and Sadoulet 2006). Thus, many small farmers in this group will continue to have one foot in farming and one foot in RNFE as their major sources of income, and their number is expected to remain large over the next decade.Subsistence-oriented small farmers are marginalized for a variety of reasons, many of which will be difficult to change in the next decade, such as ethnic discrimination, sickness, age, or their farm's location in a remote area with limited agricultural potential. We expect the number of these small farms to fall with economic transformation, but it is unrealistic to expect most will disappear in the next decade. These farm households tend to undertake some RNFE or farm wage labor (usually the domain of the poorest farmers or the landless), but many of the same factors that constrain their farming also prevent them from undertaking remunerative RNFE to become transition farmers. These subsistence-oriented farmers are typically net buyers of staple foods. While market and technology development will help them improve farm productivity, the above constraints limit even this. They need social protection policies and other public support beyond what the agrifood system and rural labor market can provide.RNFE is an important income source for rural small farm households and, on average, occupies more of their working time than farming in many African and Asian LMICs (Dolislager et al. 2020). For commercialized and transition small farmers, who are often in places with favorable agroclimates and adequate infrastructure, RNFE helps fund farming by providing cash or collateral for credit to buy inputs and diversifying income risk from agriculture. This can incentivize experimentation with new production technologies and riskier products like vegetables, poultry, and fish that have higher values. Increases in local RNFE activities often lead to rising rural wages (Lanjouw and Murgai 2009), which can induce the adoption of mechanization (Wang et al. 2016). However, in less favorable agroclimatic zones or hinterland areas, where most subsistence-oriented small farmers are located, RNFE is used mainly to fund food purchases and competes with, but also compensates for, unprofitable farming (Davis et al. 2009).The future of small farms will depend on technological and institutional innovations that are now appearing in some developed and developing country contexts or have yet to be developed (Herrero et al. 2020(Herrero et al. , 2021)). Technological innovations have the potential to benefit small farms in LMICs, but ensuring their appropriateness remains a challenge. High transaction costs, lack of collective action, and failures in production and marketing coordination all introduce risks for small farms and are commonly seen as barriers to adopting modern technologies and participating in value chains. Many subsistence farmers may be too remote from markets or lack the capacity to benefit from new technologies. Transition farmers can be disincentivized from adopting new technologies if they are labor-intensive and compete with their non-farm employment. Even for commercial small farmers, the adoption of new technologies requires enabling conditions from output and input supply chains.Small farmers' adoption of new technologies and the cultivation of higher-value products thus requires that they have the proper profit incentives and market access, which are, in large part, a function of the broad market institutional context. Effective market institutions require improved infrastructure that facilitates input supply chains upstream from the farm and connects small farmers to cities downstream from their farms. Downstream from the farm, output market conditions affect small farmers' prices, risk, and transaction costs. Critical factors include urban market size and proximity; the density and quality of roads between farmers and markets; and the midstream (wholesalers, logistics firms, and processors) and downstream (retailers) accessibility to and conduct toward small farmers. Developments in these enabling conditions in LMICs are themselves local innovations, which often rapidly improve market access for small farmers, as in the examples from Ethiopia, Nigeria, and India discussed below. Changes in these conditions will continue to be the main factor affecting small farmers' technology adoption, income growth, and inclusion in agrifood system transformation in the next decade. Some emerging technologies, such as e-commerce linked to digitalization, are also promising innovative market institutions that will impact the relationship between small farmers and markets in the next few decades.The urban market now makes up the largest share of national food consumption in LMICs (Reardon et al. 2019(Reardon et al. , 2021a, b), b). Proximity to urban markets in primary and secondary cities and small towns asserts a strong influence on market conditions and the technology and product choices of small farmers (Vandercasteelen et al. 2018). Highways and rural roads connecting farmers to urban markets likewise are critical to small farmers' access to these booming urban markets, suggesting the importance of public investment in rural infrastructure (Stifel et al. 2016).The combination of growing urban markets, expanding road connections, and the development of wholesale markets provides favorable conditions for the spontaneous formation of clusters of wholesalers, cold storages, processors, and logistics enterprises that provide crucial services enabling small farmers to access urban markets. The emergence of clusters of small and medium enterprises (SMEs) offering potato cold storages in Bihar, India, is a good example; these have allowed small farmers to store their produce and wait for much higher prices in the off-season (Minten et al. 2014). In Ethiopia, the spontaneous development of a teff value chain connecting rural areas to Addis Ababa has been facilitated by the growth of midstream private SMEs utilizing public infrastructure and improvements in wholesale markets. Midstream market development also spurred the adoption of new technology and a new teff variety by small farmers (Minten et al. 2016). Many thousands of small chicken farmers in Nigeria, mostly women, benefited from the rapid growth of long north-south maize supply chains, operated by thousands of SME wholesalers and feed millers, to market their chicken and eggs in towns and secondary cities (Liverpool-Tasie et al. 2017). Spontaneous clusters of traders and input suppliers are also seen in aquaculture districts of Bangladesh and are a key determinant of small farmer technology adoption (Hu et al. 2019).The relations of supply chain firms with small farmers are a critical determinant of small farmers' participation in markets for high-value agricultural products. These firms not only buy from small farms, but also often provide resources and services that small farmers need to participate in the market, from inputs and credit allowing them to adopt new technologies that meet market requirements to services such as aggregating, sorting, and packing. This facilitation is offered through formal contract-farming arrangements with large processors and retailers (Swinnen and Kuijpers 2019), as well as through informal relationships with SME wholesalers and processors that reduce the price risk for small farms (Liverpool-Tasie et al. 2020). Relative to the \"traditional\" arrangement of spot markets, this facilitation can be broadly seen as a market institution innovation, especially in the poorer LMICs. We expect these relationships to expand over the next decade as the double-pronged food system revolution continues its rapid course, with both the proliferation of SMEs and of modern large-scale firms underpinning the growth of rural-urban supply chains (Reardon et al. 2019).Despite still being in its infancy in LMICs, e-commerce (marketing online) and e-procurement (buying intermediate inputs online) are emerging rapidly. The diffusion of Internet access, mobile phones, and computers helps the spread of \"delivery intermediaries,\" whose expansion has been particularly rapid during the COVID-19 pandemic as consumers tried to avoid in-person shopping (Reardon and Swinnen 2020). COVID-19 accelerated e-commerce growth, for example, from 30% to 70% per year in India, 10% to 20% in China, and 20% to 50% in Nigeria (Vardhan 2020). The benefits of e-commerce for small farmers will depend on three conditions. First, widespread access to e-commerce will depend on mobile phone rates and Internet costs, which currently are particularly high in Africa (Torero 2019). Second, while e-commerce can make it easier for small farmers to sell to urban markets, their costs and product quality must still be competitive with medium and large farmers and importers. Small farmers linked to e-commerce may be better able to compete in more proximate niche markets. Third, e-commerce as digitalization per se only informs a buyer of a seller and a seller of a buyer; the final transaction still relies on delivery intermediaries, roads, and logistics, and the same high transaction costs that have constrained the development of non-digitized supply chains will constrain large numbers of small farmers from participating in e-commerce.Encouragingly, there are interesting examples of e-commerce that are inclusive of small farmers with potential to spread in the future, depending on the three conditions noted above. In Indonesia, the Rumah Sayur Group, a vegetable farm co-op with 2500 farmers, sold to supermarkets, wet markets, and food-service businesses in Jakarta before the pandemic. During the pandemic, they turned to Alibaba's Lazada to sell directly to consumers and retailers. In Malaysia, Lazada connected SME flower suppliers to online florists to gain a new customer base when COVID-19-related restrictions interrupted the traditional marketing system. In Africa, Facebook and other e-platforms have helped small farmers sell directly to consumers. Examples include Koop direk von boer (buy directly from the farmer), a Facebook group of farmers created in May 2020 that attracted 46,000 members across South Africa in just 2 weeks (Masiwa 2020).Upstream from the farm, market conditions affect the input prices, risk, and transaction costs facing small farmers, just as the output market affects the profitability of adopting new farm technologies and the transition to higher-value products, as do input supply chains. Importantly, input market conditions are parallel to output market conditions, affected by many of the same policies and public investments discussed in the context of downstream factors. Again, the development of these conditions is a local innovation. Changes in these conditions can rapidly improve input market access for small farmers, spurring technology change at the farm level.Some particularly interesting market institutions and technological innovations in agricultural service markets appear to be helping small farmers. We characterize them as the development of mobile \"outsource\" services. They include a wide range of services available to farmers on a fee basis. For an individual small farmer, the outlays of capital for machines required would not be affordable given their small scale and the large lump-sum fixed cost for machinery. In the early 1880s, such on-demand operational services emerged in the United States and European countries, where large farmers dominated. Small farmer demand for mechanization and agricultural operational services has risen in recent years in LMICs, first in Asia and Latin America and, more recently, in Africa. These services, perhaps especially as they are facilitated by communications innovations, appear to provide important support to small farming technological change. In general, mobile technology can help service supply and extension reach widely dispersed small farmers (Van Campenhout et al. 2021). For example, mobile mechanization services for land preparation, harvesting, and threshing are hired by many small farmers in South and Southeast Asia (Zhang et al. 2017;Paudel et al. 2019;Diao et al. 2020;Yagura 2020;Belton et al. 2021). They are increasingly accessible for small farmers in Africa (Berhane et al. 2016;Kahan et al. 2018;Takeshima 2018;Diao et al. 2020;Cabral 2021). Mobile phones are widely used for connecting service providers and small farmers, and new digital platforms appear to have potential to reach groups of small farmers. Examples include Hello Tractor in Nigeria, TroTro Tractor in Ghana, Rent to Own in Zambia, and EM3, Trringo, and farMart in India (Birner et al. 2021;Daum et al. 2020).Moreover, other SME services are emerging in various agricultural operations traditionally carried out by small farmers themselves, such as for rice seeding and transplanting in southern China (Li et al. 2015;Gong et al. 2012); spraying, pruning, land preparation, harvesting, and marketing for mango farmers in Indonesia (Qanti et al. 2017); seed propagation, digging wells and ponds, spraying, and loading trucks for vegetable farmers in Ethiopia (Minten et al. 2020); and bee pollination services for vegetable and fruit growers throughout China (Altay News 2019). Many of these services have replaced labor-intensive farming activities with machines or specialized techniques, helping small farmers who lack the cash to invest in machines, the skills to use machines and other techniques, or simply the time to spend farming because of non-farm employment. These services also introduce small farmers to new technologies that they otherwise might have been unaware of had they not been provided as part of a package of services by SMEs, such as flower hormone use to extend the harvesting of mangoes in Indonesia (Qanti et al. 2017).New institutional innovations can also benefit small farmers through contributions to sustainable land stewardship. Market-based institutions that incentivize farmers to maintain ecosystem services and biodiversity have been used for over a decade. With payments for ecosystem services (PES), the private or public sector pays land stewards (farmers) to protect watersheds, sequester carbon through tree planting, or conserve biodiversity (Milder et al. 2010). In the case of carbon, for example, the institution providing payments receives offset credits in the voluntary or regulatory carbon market. Another scheme involves certification of agricultural commodities, such as coffee, palm oil, and cacao. Certification schemes are generally implemented by non-governmental organizations (NGOs) and rely on consumers paying a premium for production practices that conform to sustainable social and environmental goals (Brandi et al. 2015;Giovannucci and Ponte 2005;Ruysschaert and Salles 2014). Smallholder farmers have benefited from these schemes only to a modest degree due to high transaction costs, low demand for ecosystem services, and poor access to information.For carbon markets, smallholder participation is impeded by the required technical capacity, as well as the costs of monitoring and complex requirements for reporting (Brandi et al. 2015;Wells et al. 2017). With certification schemes, evidence indicates mixed success for environmental, social, and economic goals. The supply of certified products is generally larger than the demand (DeFries et al. 2017). Insecure land tenure, lack of credit, and insufficient profit to warrant the required investments hamper smallholder participation in both PES and certification schemes.With rising recognition of the importance of land stewardship for climate mitigation and conservation of biodiversity, institutions to incentivize protection of ecosystem services and sustainability goals are likely to become more widespread in the coming decades. Carbon markets, which, to date, have largely been unable to stem land clearing and greenhouse-gas-emitting practices on agricultural land, will likely be a more significant driver of farmers' decisions in the future. In combination with digital technology, institutional innovations have the potential to reduce transaction costs and enable participation by smallholders to maximize their ability to benefit from these schemes, both to boost their incomes and to contribute to society's sustainability goals. Technology and training for smallholders to access and interpret satellite data, monitor their lands, and fulfill reporting requirements are needed if they are to benefit from a growing demand for ecosystem services.This chapter has sought to imagine the future of small farms and identify promising innovations in agrifood systems to improve their prospects over the next 10 years.Because small farms are heterogeneous and dynamic, we classed them into three groups: commercial, in-transition, and subsistence-oriented small farms. Each has its Commercial small farmers are the vanguard of agrifood transformation and best prepared to take advantage of the opportunities that growing market demand for agrifood products will create. They tend to be located in more favorable agroclimates, nearer to cities and towns, and in areas better served by infrastructure and midstream SMEs that facilitate input and output markets. These same market opportunities will incentivize some transitional farmers to invest in their small farms and become commercial farmers. To enhance small commercial and transitional farmers' competitiveness to pursue these market opportunities, the following government policies and public investments are important: own set of challenges and opportunities, and policies and investments that prioritize inclusive small farm transformation must be differentiated to best target the needs of each group as agrifood systems evolve (Hazell 2020).The Future of Small Farms: Innovations for Inclusive Transformation• Increase investments in infrastructure, including rural roads connecting to secondary and tertiary cities, that can create economies of agglomeration and a critical mass of proximate services such as wholesale, logistics, and farm input provision for small farmers in the surrounding rural areas, thus reducing transaction costs. Often, mobile agricultural services are clustered in towns and fan out to serve small farms in a hub-and-spoke model (Zhang et al. 2017). Many new digital technologies applied in e-commerce, information provision, and farm service businesses also depend on good infrastructure. While initial investments need to come from governments, they will serve to lure in private investments from both large companies and SMEs. • Promote education and training programs that target rural youths to develop the skills and knowledge required to support modern agriculture and marketing. These skills are necessary for both farm management and off-farm jobs in logistics, machinery maintenance/repair services, and broader RNFE. • Facilitate co-operatives and farmer groups that can collectively pursue emerging opportunities in urban markets and modern farm technology. Local networks can also be strengthened through village-level innovation platforms to link smallholder farmers with extension and research, such as China's Science and Technology Backyard (Barrett et al. 2020a). These show promise for drawing together the wisdom of (small farmer) crowds and the knowledge of cutting-edge scientific researchers to accelerate discovery, adaptation, and diffusion (Nelson 2019;van Etten et al. 2019). • Support SMEs upstream and downstream from farms by reducing unnecessary regulations and informal restrictions that often discourage SME development.SMEs are more accessible to small farmers than larger enterprises, and small farmers value the mix of services that SMEs provide (Liverpool-Tasie et al. 2020).RNFE is the main economic activity of transitional farmers and is increasingly the main source of income for most small farmers. RNFE provides small farmers with cash, both to purchase food and for farm investments to raise productivity, expand commercial activities, and produce higher-value products. RNFE is also important for some marginalized farmers, helping them reduce their reliance on risky, low-yield agriculture. For these farmers, RNFE development will directly improve food security in a way that marginally boosting agricultural production cannot (ZEF and FAO 2020;Frelat et al. 2016). Public investments and policies that facilitate growth of the agrifood system must pay more attention to creating enabling environments for the development of RNFE and strengthening the synergy between agriculture and RNFE in rural areas. In this regard, the following actions are promising for governments to actively promote agriculture-RNFE synergies for rural development and agrifood system transformation:• Pursue policies that have broad effects across economic activities in rural areas and do not limit interventions to farming alone. RNFE and farming are complementary, and both are needed for inclusive growth in rural areas.• Develop an enabling environment-including basic infrastructure, property rights, and legal systems with enforcement mechanisms-favorable to rural businesses that encourage and facilitate inclusive RNFE (Haggblade et al. 2007). • Identify engines of regional growth through consultation with the private sector and farmers, and conduct supply chain diagnostics for prioritization of strategic interventions (Haggblade et al. 2007). Emphasize differentiated strategies and flexible institutional coalitions for implementation appropriate to diverse rural areas.This chapter emphasizes the importance of market institution innovations for achieving higher agricultural productivity and quality through small farm technology adoption and improving incomes for small farm households through participation in both farm and non-farm economic activities. In addition to the policy recommendations discussed above, some additional policy recommendations are listed here, although adapting and differentiating policies over heterogeneous contexts across LMICs requires context-specific research and consultation with stakeholders (Barrett et al. ): 2020b• Support new technologies that reduce risk and are attractive to small farmers when viewed in a holistic way, taking into account farmers' resource environment, as well as their livelihood strategies. Do not automatically assume laborintensive innovations are appropriate for small farmers, who often want to reduce, not intensify, their farm labor use (Hazell 2020). For transitional farmers who depend on RNFE, proposing new labor-intensive farming activities could fail if they cut into the time farmers have available for RNFE livelihood strategies (Moser and Barrett 2006). • Ensure that agricultural interventions to support sustainable farming practices are economically viable for farmers and provide direct economic benefits. In the longer term, farmers are most strongly motivated to adopt and maintain sustainable practices when they perceive positive outcomes of these practices for their farm or the environment (Piñeiro et al. 2020). • Scale up productive social protection programs for subsistence farmers in hinterland areas who face barriers in accessing markets and other economic opportunities. Safety net programs ease liquidity constraints and increase tolerance for risk among small farms and, when integrated with measures to increase agricultural productivity, have the potential to make significant progress toward the eradication of hunger (Wouterse et al. 2020).","tokenCount":"4612","images":[],"tables":["378029000_1_1.json","378029000_2_1.json","378029000_3_1.json","378029000_4_1.json","378029000_5_1.json","378029000_6_1.json","378029000_7_1.json","378029000_8_1.json","378029000_9_1.json","378029000_10_1.json","378029000_11_1.json","378029000_12_1.json","378029000_13_1.json","378029000_14_1.json","378029000_15_1.json"]}

data/part_2/0312763368.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"d513bc3b24f8ca888a413713edcee922","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/0a057c18-79d1-4c12-b4d7-4370b4f90c7e/retrieve","description":"Droughts, hurricanes and other environmental shocks punctuate the lives of poor and vulnerable populations in many parts of the world. The direct impacts can be horrific, but what are the longer-term effects of such shocks on households and their livelihoods? Under what circumstances, and for what types of households, will shocks push households into poverty traps from which recovery is not possible? In an effort to answer these questions, this paper analyses the asset dynamics of Ethiopian and Honduran households in the wake of severe environmental shocks. While the patterns are different across countries, both reveal worlds in which the poorest households struggle most with shocks, adopting coping strategies which are costly in terms of both short term and long term well-being. There is some evidence that shocks threaten long term poverty traps and that they tend to militate against any tendency of the poor to catch up with wealthier households. Policy implications are discussed in terms of access to markets and the design of government safety net programs.","id":"-70294948"},"keywords":[],"sieverID":"1b736cfe-fd55-4e2b-92f9-511980e821a6","pagecount":"45","content":"Droughts, hurricanes and other environmental shocks punctuate the lives of poor and vulnerable populations in many parts of the world. The direct impacts can be horrific, but what are the longer-term effects of such shocks on households and their livelihoods? Under what circumstances, and for what types of households, will shocks push households into poverty traps from which recovery is not possible? In an effort to answer these questions, this paper analyses the asset dynamics of Ethiopian and Honduran households in the wake of severe environmental shocks. While the patterns are different across countries, both reveal worlds in which the poorest households struggle most with shocks, adopting coping strategies which are costly in terms of both short term and long term well-being. There is some evidence that shocks threaten long term poverty traps and that they tend to militate against any tendency of the poor to catch up with wealthier households. Policy implications are discussed in terms of access to markets and the design of government safety net programs. viiiAto Mohammed, 55 and illiterate, resides in Bati woreda of South Wollo (Ethiopia) and heads a household of nine. He has been chronically food insecure for more than ten years when he lost his only oxen due to drought. He sold the animal to buy food at the time and has not been able to acquire another. Currently, Mohammed holds one hectare of farm land and he has no grazing land. Since he owns no oxen, he has been leasing out the land for share-cropping on a 50/50 sharing arrangement.Mohammed and his family members are engaged in various types of daily labour activities for cash and food, and the household is a regular recipient of food aid.Mohammed asserts \"oxen are the crucial productive asset that would liberate me from this insecurity trap\". On the other hand, however, he does not want to take credit from a regional credit organization to buy an ox as he does not want to be indebted and fears that the debt may be passed on to his children if he fails to repay. He fears that the ox may die due to lack of adequate feed or animal diseases for which there is no dependable animal health service in the community. He also fears that he may not be able to pay back since crop failure is frequent due to insects and droughts. 1 Michael R. Carter is from the University of Wisconsin; Peter D. Little is from the University of Kentucky; Tewodaj Mogues is a Post Doctoral Fellow at IFPRI's Development Strategy and Governance Division; and Workneh Negatu is from Addis Ababa University.The direct impacts of the droughts, hurricanes and other environmental shocks can be horrific. But what are the longer-term effects of such shocks on households and their livelihoods? Do environmental shocks leave some households in a \"poverty trap,\" with so few assets that they cannot engineer an economic recovery, as Ato Mohammed's story suggests?2 Does the fear of such traps lead forward-looking households to adopt asset protection strategies which come at the very high cost of immediately reduced consumption? Are patterns of vulnerability and access to market-and socially-based coping mechanisms such that repeated environmental shocks increase community inequality by grinding away at the meagre assets of the relatively poor?In an effort to answer these questions, this paper analyses the asset dynamics of Ethiopian and Honduran households in the wake of severe environmental shocks. The work is part of a comparative project that addresses the interrelationships between climatic shocks, markets, and asset recovery strategies among households in developing countries (see Little et al., 2002). In Ethiopia, markets are relatively weak (especially for land, labour, and capital), and non-market mechanisms are important. Factor markets are better developed in Honduras, but its inegalitarian agrarian structure may limit the effectiveness and extent of the social assets that may aid recovery in Ethiopia. 3 Data on a sample of 416 rural Ethiopian households track household assets over a seven-year period of pre-drought (1996-1998), drought (1999-2000), and recovery (2001-2003). Data on a sample of 850 rural Honduran households capture the immediate impact of Hurricane Mitch in 1998 on assets and income, as well as these households' economic position in 2001, two and half years after Mitch. While the steady grinding away of economic possibility created by a prolonged drought is clearly quite different from the acute and immediate destruction of a hurricane, analysis of these two disparate cases and countries offers a unique comparative perspective on the role of market-based, socially-based and aid-based coping and recovery strategies.The remainder of this paper is organized as follows. Section II proposes an anatomy of an environmental shock, tracing the evolution of assets through time in the face of a shock, and presents an empirical model of asset accumulation to investigate households' sensitivity to, and resilience from, shocks. Section III describes the asset and income losses households in northeastern Ethiopia suffered due to the droughts of 1999/2000 and estimates the determinants of long-term asset recovery in the wake of the droughts. The factors that influence rural Honduran households' exposure to and recovery from the 1998 hurricane are examined in Section IV. Concluding remarks are offered in the final section.Ato Mohammed's story used to introduce this paper illustrates both asset sensitivity to environmental shocks (Mohammed had to sell off productive assets to survive the shock) and lack of resilience in the wake of shocks (Mohammed has been unable to rebuild his assets and livelihood in the aftermath of the shock). The goal of this section is to think through these twin factors of sensitivity and resilience. Together with the pattern of exposure to shocks, sensitivity and resilience shape the longer term economic impacts of environmental shocks.Figure 1 presents the stylized economic anatomy of an environmental shock from a household's perspective. The x-axis measures time and the y-axis measures asset stocks and income shocks. The full economic effects of an environmental disaster can be traced through three stages: the period of the shock itself; the coping period in which households deal with the immediate losses created by the shock; and the recovery period in which households try to rebuild assets lost to the disaster and depleted through coping strategies.While the boundaries between these stages are fuzzy, distinguishing between them is a useful step in thinking through the full impacts of shocks.The time interval over which an adverse environmental event occurs could be very brief (as with a hurricane), or it could be an extended period (as in the case of a prolonged drought). Households can also be buffeted by a sequence of such events. For illustrative purposes, Figure 1 is drawn as if the shock is a brief, one time event.An environmental shock has two direct impacts. First, it may destroy assets (washing away land, killing livestock) by an amount Θ. In Figure 1, this first impact is shown by the sudden interruption of the household's asset trajectory as its assets decline from point A b to point A s . 4 The second direct impact is that it may reduce disposable household income below its normal level (crops fail or households suddenly must devote income to medical expenses). This deviation from normal disposable income levels is shown in the figure by the negative disposable income shock, ε.Anatomy of an Environmental Shock Households' reactions to the direct income and asset losses during the coping phase are structured by the markets and other institutions to which they have access. 5 Households with financial market access might borrow against future earnings to sustain their consumption standard without further asset depletion. Informal finance and insurance arrangements can play the same role, as can receipt of disaster assistance.Another coping strategy is to redirect or increase work time (reduce leisure). The effectiveness of this strategy will depend on access to and depth of labour markets.Households without access to these markets may sustain their consumption by further drawing down on their assets. While this strategy will help to smooth household consumption over time, it implies that assets will exhibit excess sensitivity to shocks (declining by more than the direct asset shock, Θ). This additional decline in household assets is shown in Figure 1 by the drop in assets from A s to A c . Note that two factors will shape the severity of this secondary asset decline. The first is the household's ability to employ the alternative coping strategies listed above. The second is changes in the prices of assets relative to the price of food and other necessities. Unfavorable asset price swings (as would be expected to happen if all households in an area respond to a drought by selling cattle) will serve to further decapitalise households in the wake of a shock.Finally, households may cope by reducing consumption. While this strategy may be a last resort for households that lack other assets or options, it may also be pursued by households reluctant to increase their future vulnerability by further depleting their stock of assets. Drèze and Sen (1989) comment on the empirical significance of such \"asset smoothing\" that necessarily destabilises consumption, while Zimmerman and Carter   (2003) provide a theoretical foundation for such behavior. As Hoddinott (2006) stresses, the costs of such a coping strategy is not only immediate hunger, but that it may also permanently reduce the growth and future capacity of younger children.The third and final stage is termed the recovery phase in Figure 1. The market and social mechanisms that broker access to employment and financial services will also shape a household's resilience and its post-shock asset accumulation trajectory. A household with good access to capital (via markets, or via informal social arrangements) can borrow against future earnings to immediately rebuild asset stocks. A key question is whether the household is able to rebuild its asset base (shown in Figure 1 as the movement from A c to r A′ ), or whether it gets trapped at a low asset level (shown by the A cto A r trajectory)The story of Ato Mohammed not only suggests sensitivity to shocks, but also the existence of an asset poverty trap, understood here, following Carter and Barrett (2006), as a minimum asset threshold below which it is not possible to engineer successful longterm asset accumulation. The dashed horizontal line in Figure 1, drawn at a poverty trap threshold of A , illustrates the idea of such a threshold. In the context of the evolution of assets and the occurrence of a shock, as depicted in Figure 1, we distinguish two possible implications of an asset poverty trap. The first pertains to the recovery period. As illustrated in the figure, a household falling below the threshold -either from a direct asset shock, or from coping strategies that further reduce its asset holdings -would be unable to accumulate assets and would be observed over time to follow a path from point A c to A r . Alternatively, if no such trap exists, we would expect that households at the bottom of the asset distribution would be able to accumulate assets and move ahead over time, e.g. to A r ', irrespective of their level of asset depletion.The second implication pertains to the coping period. As discussed earlier, we might expect to see poor households pursue asset smoothing strategies when facing the risk of falling into such traps. In contrast, better endowed households would, when faced with an income shock, have less reason to resist selling their assets to bring consumption back up to normal levels, given their relative distance from the asset threshold. And again, if no such trap exists, we would not expect to see signs of asset smoothing by low wealth households. 6 Finally, it is worth reiterating that well-positioned households which have sufficient access to coping mechanisms other than asset sales, such as resort to financial markets, labour markets, timely and adequate disaster assistance, etc., may also not be observed to smooth consumption by reductions in their asset base.The notion that some households might remain mired in a trap of persistent poverty is surprising from the perspective of some dynamic economic theory that suggests that less well-off households would be expected to have every incentive to try to save, accumulate and catch up economically with their better off fellow citizens. Barrett   and Carter (2005) discuss the forces that could offset convergence, identifying lack of access to market or socially mediated access as the key force, and Carter and Barrett   (2006) expound the theoretical underpinnings of this phenomenon.In summary, the longer term effect of an environmental shock on household productive assets will depend on both sensitivity to, and resilience from, shocks.Sensitivity and resilience in turn are likely to depend on a household's own wealth prior to and in the wake of the shock, and on its access to employment and capital, as mediated by either market or social mechanisms.This section puts forward an econometric approach for exploring the longer run economic impacts of environmental shocks in Ethiopia and Honduras. The data available for both countries includes measures of pre-shock and post-recovery assets stocks (A b and A r , respectively), as well as indicators of the magnitude of the shocks received.Information is not consistently available on intermediate asset stocks (A c ), and hence it is not possible to directly explore household coping strategies. Instead, we adopt a reduced form approach, examining the overall change in assets from the pre-shock period to the post-recovery period.In order to test for the sensitivity and resilience effects discussed above, we will build on the following model of household asset growth from the pre-shock to the recovery period:where A ri are the recovery period assets of household i, A bi is the household's pre-shock asset stock, and the function r(•) specifies the household growth rate as a function the asset (Θ i ) and income shocks (ε i ) experienced by household i, as well as of other household characteristics x i . such as age that might shape desired asset levels.Conditioning the impact of these factors on asset growth are wealth, and labour market and capital access variables.To explore differential sensitivity of assets to shocks, the impact of the shock variables are conditioned by the pre-shock wealth level of the household, denoted bi A . A highly sensitive household would be one where shocks have large and lingering effects on household productive assets. Controlling for post-shock resilience, high sensitivity of assets to shocks for a particular wealth level would signal reliance on asset sales to smooth consumption, and most likely a lack of access to market-or socially-mediated access to capital or insurance. Evidence of the insensitivity of assets to shocks would indicate either access to finance, or the existence of an asset smoothing strategy provoked by a poverty trap (most likely for lower wealth households).To explore the idea that post-shock growth (resilience) differs by wealth, the growth rate is specified to depend on the post-shock wealth of the household, si A , after also controlling for initial conditions bi A . 7 The existence of a poverty trap would be signalled by a negligible post-shock growth for the poorest households, while a convergent process would be signalled by higher rates of asset growth for households that are among the poorest in the wake of the shocks.Finally, expression (1) indicates that the asset growth is conditioned by labour market access and capital access variables, L i and K i . Both factors are likely to shape both resilience and sensitivity. Note that K i can include what is commonly referred to as social capital assets.In the study area of eastern Amhara Region (South Wollo and Oromiya Zones), unlike the swift destruction brought about by Mitch in Honduras, the drought of the late 1990s was a prolonged event with uneven consequences, and its onset was gradual.Indeed, the first signs of disaster can be traced to the poor short rains (January-April, called the Belg season) of 1998 where it is estimated that harvests were only 60 percent of normal yields in the main Belg growing areas (Government of Ethiopia,1997; 1998a;   and 1998b). That year the long rainy season (June-September, called the Meher) was near normal for all areas except in the Belg growing zones where there is also some dependence on the Meher season. Because the Meher rains of 1998 were near normal in some locations, drought and relief agencies in Ethiopia failed to see the looming disaster until the Belg season of 1999 emerged as a massive failure, resulting in 90 percent loss of crops (see Castro et al., 1999). National and regional estimates for food relief in 1999were drastically altered when it was observed that the Belg season of 1999 was going to be an almost complete disaster (Government of Ethiopia, 1998a and 2000).The 1999 Meher season was only somewhat better but not good, yielding about 40 percent of normal harvests in six of the eight study kebeles (an administrative unit comprised of four or five villages). Food aid distribution started in the region in June 1999, but was not widespread until August 1999. To make matters worse, the Belg season of 2000 was very poor (75 percent reduction of normal yields). With massive imports of food aid and the recovery of the long rains in 2000, the nutritional status of the area's population had returned to near normal by early 2001. Thus, the drought of the late 1990s was keyed by the failure or near failure of three successive short rainy seasons.The first of the crop failures was only 40 percent, but with such widespread poverty it was enough to initiate the downward spiral of extensive food insecurity and distress sales of assets (mainly livestock) that characterised the region for the better part of 30 months.The long term possibilities of asset recovery from this series of shocks are conditioned by several community and household characteristics. The study region (as indeed large parts of Ethiopia) is characterized by relatively weak labor markets and nearly absent credit markets. Land markets are severely restricted in that private ownership is prohibited, and legal constraints on land rentals were only recently relaxed.In part because insurance against crop loss is practically absent and market-based coping mechanisms are limited, food aid makes up a relatively large portion of food consumption, as indicated above. Social institutions such as burial societies and religious associations are highly prevalent in South Wollo. We will explore their importance, along with those of market and aid-based mechanisms, for enhancing a household's ability to weather shocks in the short term as well as recover in the long term. Data are from a seven round household survey conducted over three-and-half years in eight kebeles in South Wollo and Oromiya zones. The dataset also includes recall questions on livestock holdings during 1996 to 1999, which assess how households fared in terms of their assets prior to the onset of the drought.1999 was also the year of heaviest livestock asset reductions, both due to animal deaths as well as sales. Based on interviews in the course of qualitative surveys in the study area, it is estimated that 25 percent of livestock reductions in 1999 were distress sales at which the seller received less than 50 percent of the normal price of the animal sold (cattle prices, for example, dropped from an average of 625 birr in the pre-drought period to 291 birr). Price swings of this magnitude constituted a huge capital loss for those forced to sell livestock during this period. Natural causes clearly precipitated the drought disaster of 1999, which resulted in a massive humanitarian effort, but the population's vulnerability to relatively small perturbations in climatic events is 'unnatural' and highlights the extreme poverty in the area.   The bottom panel of Figure 2 displays post-drought livestock trajectories based on wealth quartiles defined according to animal holdings at the end of the drought (mid-2000). As can be seen, average holdings of the poorest quartile were essentially zero at that time. Interestingly, however, this group on average managed to add substantially to their livestock holdings over the span of the following three years. While this descriptive look at the data does not control for other factors that influenced these trajectories, it is suggestive of interesting patterns, with the highest wealth households exhibiting greater asset sensitivity to shocks, and the lower wealth households destabilizing consumption and perhaps defending their modest livestock holdings over the course of the drought.Yet, despite this asset smoothing behavior which would seem to signal the existence of a poverty trap, post-shock growth appears to be relatively robust, even for those households that were completely without livestock at the end of the drought.One possible explanation for the apparent inconsistency in these findings is that those households that had completely stocked out by mid-2000 were precisely those households that enjoyed good social capital and perhaps other assets. Confident that they could borrow animals to rebuild depleted herds, they had no reason to fear becoming mired in a poverty trap. In contrast, those households that apparently defended their livestock could be isolated households who-like Ato Mohammed-would have been trapped by low post-shock resilience and growth had they let their stocks deplete to almost nothing. Employing the asset growth model put forward before, we turn now to a more thorough analysis of the data using multiple regression methods that permit us to control for these various factors that influence observed outcomes.To estimate the determinants of the long-term rate of growth of livestock assets from A b to A r , we adopt the asset growth model (1) as follows:where the error term ω i is assumed to be normally, independently and identically distributed across observations, and an additive term, ) (, that captures factors that affect households' livestock holdings but do not operate through expanding growth of existing stock. Sources of such additional livestock assets could include social transfers or gifts, as well as non-livestock assets which can be sold and transformed into livestock assets. Unfortunately, we lack data on these other assets, and instead specify that the magnitude of this additive term depends on the household's pre-drought livestock holdings (taken as a proxy of overall household wealth).To estimate (2), we subtract pre-drought assets from both sides of (2) (so that the change in livestock assets over time becomes the left-hand side variable) and employ the following functional specifications for the growth rate and additive term:The growth rate terms in the curly brackets include basic household demographic information that would be expected to affect desired household livestock holdings (the x i ). The possibility of wealth-differentiated resilience and poverty traps is explored by including a quadratic specification of post-shock wealth level, A si . Differential sensitivity to the environmental shock, ε i , is explored by letting the coefficient on the magnitude of the shock depend on the 1997, pre-shock wealth quartile of the household,where 1 is the lowest quartile). While the available data lack a measure of the severity of the shock received by each household, we approximate it with the share of households in the community of household i that suffered crop losses from June to December 2000, the most severe drought season.Finally, we let the impact of social institutions S i , food aid F i , and labour markets L i on the growth rate depend on the household's wealth. Social institutions measured as community average membership in social organizations.8 Food aid is measured as the percentage of a community participating in food for work programs. While this community level variable bypasses individual-level endogeneity problems created when more severely shocked households choose to participate in the program, it may still suffer from the fact that programs are placed in communities where the drought was most severe. The labor market indicator is an average of off-farm earnings for all households within a village. This number was then normalized to vary between zero and one by dividing it by the highest level of average earnings among the eight villages in the sample.Table 1 gives the results of the ordinary least squares estimation of the model.The first column presents the unrestricted model that allows for wealth effects for all potential conditioning factors of livestock asset growth. The second model restricts the wealth-differentiation of the role of those mediating factors where such restrictions do not weaken the explanatory power of the model. We will concentrate on the results for the restricted model, but include the full model for completeness.Demographic factors have their expected effects on livestock growth. Post-shock resilience displays a strongly wealth-differentiated pattern. The estimated coefficient indicates that the growth rate of livestock is increasing in herd size up to a level of 25 tropical livestock units (TLUs).9 Other things equal, these estimates indicate that households that exited the drought with few livestock were strongly disadvantaged in their rate of recovery period growth.The sensitivity of assets to shocks also shows a pronounced wealth-differentiated pattern. For the lowest wealth quartile, shocks have an insignificant effect on livestock, signaling the presence of asset smoothing behavior. Asset sensitivity is then increasing with pre-shock wealth. The magnitude of the estimated coefficients are such that a 10 percentage point increase in the proportion of households in a community suffering crop  losses is estimated to decrease the asset growth rate of the wealthiest group by 23 percentage points. For a second quartile household, the decrease would be only 12% points, whereas it would be almost zero for a bottom quartile household. This greater asset-sensitivity of the wealthier groups does not imply that welfare of the initially better off group more vulnerable to shocks. On the contrary, these results are consistent with the initially better endowed households pursuing a consumption-smoothing strategy, whereas the already asset-poor may be protecting their meagre wealth in the face of adverse conditions, following the asset-smoothing strategy discussed above.The estimated model also explores the degree to which social mechanisms, food aid and labor markets bolster asset growth, controlling for shocks, etc. Community membership in social organizations (social capital) increases the rate of growth (or limits the rate of loss) of livestock over the six years, but does so primarily for households in the higher wealth groups. In contrast, there is only weak evidence that labor market access significantly affects the rate of growth of livestock. While the average effect (from the restricted model) is positive and statistically significant, the full model shows that this positive effect appears to be strongest for wealthier households. Finally, availability of food aid in the community does not appear to protect households' future assets, and in fact seems to have the opposite effect.10 In addition to factors affecting the growth rate of livestock, the econometric model also includes an additive term meant to capture other sources of livestock replenishment (including own non-livestock wealth and social transfers). As can be seen in Table 1, these additive terms are quite significant and amount to 1.7 TLUs for the poorest households and rise to over 6 TLUs for the initially wealthiest households. While somewhat puzzling, the magnitude of these coefficients suggest a major force that potentially offsets the low estimated resilience of less well-off households.In summary, the econometric results in Table 1 provide some evidence of poverty traps, including the relatively low resilience of poorer households, the apparent patterns of asset, not consumption smoothing, by the poorest; and, social capital that is less effective for the poor. At the same time, other factors point to possibly net positive livestock recovery for the poorest households, especially the additive constant term. In order to assemble these countervailing forces into a single indicator of asset recovery, The Table 2 results show that the rate of recovery of livestock assets is slower where shocks are higher, and better in an environment of better access to community social capital. This table also reveals the net effect in terms of long-term growth rates of the two forces of sensitivity and resilience: We saw above that the better endowed decumulate assets faster in the course of experiencing a series of droughts, but that this group is also relatively better equipped to rebuild assets in the wake of these shocks.These two forces combine to point to an overall faster growth of the initially poor as compared to the highest-wealth group: the former expands assets on average from 0.2TLUs eightfold to about 2.1 TLUs, whereas the highest quartile increases its initial asset base of 11 TLUs to reach 28 TLUs under a scenario in which they do not face collective shocks, and in fact lose livestock over the six-year time period where shocks are high. In all cases, the predicted livestock holdings for low quartile households is driven by the additive constant term, which by itself accounts for 1.7 TLUs.Hurricane Mitch carved a path of destruction across Honduras in 1998. Unlike the Ethiopian drought described in the previous section, the impact of the hurricane was almost instantaneous. Drawing on data collected shortly after the hurricane, Morris et al.(2001) report that poor rural households lost 30% to 40% of their crop income and measured poverty immediately increased 5.5 percentage points, rising from 69.2% of households to 74.6%. They also report that lower wealth households lost 15% to 20% of their productive assets (land, livestock and plantations), compromising their capacity to generate earnings and livelihood. Unclear, however, from these early studies is whether households were able to recover from losses of this magnitude and rebuild their assets and livelihoods. While labour and other markets are deeper and better developed in Honduras than in Ethiopia, it is less obvious that local social relationships would function as effectively to mitigate the longer term effects of an environmental disaster.The data available for this study provide a window into these longer term questions. Some 30 months after Mitch, a sample of 850 rural Honduran households (clustered in 30 municipalities spread across 6 provinces) was surveyed as part of a study on the impact of land market liberalisation and asset accumulation.12 Included in the questionnaire were a number of retrospective questions that probed the direct impacts of Mitch on household assets and income. The study also collected data on household assets in 2000/2001, giving a window into the longer term patterns of asset cycles and poverty traps.Table 3 presents descriptive statistical indicators of the impact of Hurricane Mitch. Information is provided both for the overall sample and for households broken up by asset quartiles. Quartiles were defined based on households' pre-Mitch holdings of productive assets (A b ), where productive assets are defined as the value of land, plantations, machinery and livestock. All assets were valued using 2000 price information and were converted to $US using the market exchange rate. As can be seen, wealth varies substantially averaging $650 for the poorest quartile and just over $75,000for the wealthiest quartile. Annual household income in 2000/01was six times higher for wealthier households than it was for poorer households ($996 versus $5,967). These low figures are consistent with the high poverty rates typical of rural Honduras. The variation in them is also a reflection of the sharp levels of inequality found across much of rural Honduras.As can be seen, 44% of households suffered a loss of productive assets from Hurricane Mitch. The percentage of households increases with household wealth (rising from 22% to 68% from the first to the fourth wealth quartile). This finding contradicts the notion that poorer households are more vulnerable to shocks, though it may be an artifact of the fact that poorer households had relatively little to lose. Indeed, as can be seen, among those households suffering asset losses, poorer households lost a greater percentage of their productive wealth (31%) than did wealthier households (8%). Across all wealth quartiles, losses were primarily comprised of lost plantations and land.Finally, Table 3 reports asset growth rates from mid-1998 (pre-Mitch) to early 2001. 13 These figures give can be used to get a sense of the empirical gap between ' ' r A and ' r A in Figure 1-i.e., the asset gap between a household that experienced a shock and a household that did not. Across all pre-Mitch wealth quartiles, households without asset losses show substantially higher growth than those that suffered losses. The gap is 13.8%13 Median growth rates are reported in the table. Mean growth rates are higher in all cases, but follow the same qualitative pattern. Many of the high growth rates appear to be the result of inheritances received post-Mitch.for the lowest quartile where poor households with losses had showed -5% net growth (loss) over the post-Mitch period, while poor households without losses had 8.8%growth. The gap is a smaller 5.1% for the wealthiest quartile (-2.1% versus 3% post-Mitch growth). While these growth differences seem to signal that poor households are more sensitive to shocks (and less able to recover from them), among those households that did not suffer any asset losses, poor households tended to grow faster (8.8%) than did wealthier households (3.1%). We turn now to more thoroughly explore these patterns of vulnerability and resilience.For estimation purposes, we write the asset growth equation (1) as:where ω i is a standard error term. Taking the logarithm of both sides and rearranging terms yields the following expression for the growth of assets from the pre-Mitch to the post-Mitch recovery period:where the growth rate r(•) has been replaced with the specification shown above. Similarto the specification employed to analyze the Ethiopian data in the prior section, equation (3′) indicates that growth from the pre-shock through the recovery period depends on basic household demographic characteristics (the x i ). Also paralleling the Ethiopian specification, the resilience of post-shock growth depends on the post-shock asset level, measured as the post-shock asset quartile of the household ( j si Q , j=1,…,4).Data structure and differences in the nature of shocks have also necessitated certain differences in the estimation approach between the two country studies. Firstly, the specification in the Ethiopian case study differs somewhat from the Honduran case in that it arrives at the determinants of the rate of growth departing from a level, rather than a log, growth equation. Also, in the Ethiopian case the growth equation was reformulated in a way that the dependent variable was expressed as the difference in asset levels from the pre-shock to the post-recovery period. In the Ethiopian data, a nontrivial number of households have no livestock assets at all. This would make a direct approach to the estimation of the growth rate impossible, hence the indirect approach as expressed by (2') was used. Given we are not inherently interested in asset levels but in asset growth, using standard censoring models such as a Tobit regression do not offer themselves as an alternative.Also unlike the Ethiopian data, the Honduras data contains direct household measures concerning the magnitude of asset shocks (Θ i ) in addition to the income shock measure (ε i ). These measures are normalized by the pre-shock asset holdings of the household so that they enter the regression as the percentage of assets destroyed by the shock. In regression specification (3′), the sensitivity coefficients on the asset shock variable (but not the income shock variable) is permitted to differ by the pre-shock wealth quartile of the household ( j bi Q , j=1,…,4 where j=1 is the lowest quartile). 14 Aid received (F i ) and labor (L i ) and capital market (K i ) access variables are permitted in (3′) to moderate the influence of asset shocks on asset growth. 15 The aid variable is simply the amount of external assistance received by the household (normalized by the household's pre-shock stock of assets). The labor market indicator for a household is defined as the average off-farm labor market earnings within its community (there are a total of 31 separate communities within the sample). The variable was scaled to lie between zero and one by dividing it by the highest community average earnings level within the sample. The capital access measure was derived from a set of questions designed to probe whether or not a household was on its demand curve for credit (and hence price rationed in that market) or whether it had excess demand for credit and hence was subject to some form of quantity rationing (see Boucher et al. 2005   for details). Unfortunately, this measure reflects a household's status in 2001, well after the immediate post-shock period. 0.31 0.31 higher growth for households that emerged from the shock with more assets, the Honduras estimates show more resilient growth for poorer households, although none of the estimated resilience coefficients are statistically different from zero. To be clear, this finding does not mean that growth is zero, simply that it does not vary much that predicted with the core demographic factors.Again contrasting with Ethiopia, all wealth quartiles in Honduras exhibit sensitivity to asset shocks. A coefficient of -1 would indicate that (other things equal) no recovery has taken place. A coefficient less than -1 indicates further deterioration of the asset position of the household through secondary coping strategies. As can be seen, for the wealthy quartiles, the coefficients are near -1. Apparently these households had little need to draw down further on their assets to smooth consumption in the wake of the shock. In contrast, the lower initial wealth quartiles show sharply greater sensitivity to shocks for the lower quartiles as the relevant coefficients are less than -1. Table 5, discussed below, draws out the full economic implications of these estimates.While unmediated shocks are estimated to have this devastating impact on productive assets, more robust local labor markets help offset these impacts, especially for the lowest wealth quartile households. Access to capital markets has a similar effect for second quartile households, though it has little effect for other households in the sample.Income shocks (measured as income lost due to the hurricane normalized by household pre-Mitch assets) are counter-intuitively estimated to have a positive impact on growth. While it is possible that households that suffered large income losses took reactive steps that ultimately enabled them to more quickly build up productive assets (e.g., income shortfalls may have forced households into bearing the fixed costs of migration to secure wage earnings), it also seems possible that high income losses occurred when households were involved in high value market opportunities and thus had better access to market-based coping mechanisms. In any event, efforts to further decompose this effect by looking at quartile-specific effects caused the variable to melt away into insignificance. Similarly, interactions between income losses and measured market access failed to detect any significant patterns. As an addition control for other losses suffered in the hurricane, a dummy variable was included in the regression, taking a value of one for households experiencing loss of housing. More intuitively, when households did experience housing loss, recovery of productive assets was significantly slowed by approximately 10%.In summary, patterns of sensitivity and resilience in Honduras appear quite different from those in Ethiopia. Unlike Ethiopia, there is no sign of asset smoothing in the wake of shocks. Indeed, the asset growth of less well-off households appears to be extremely sensitive to shocks received. When labor and capital market access are stronger, this sensitivity is reduced. However absent that access, poorer Honduran households appear to struggle economically in the wake of a shock.In order to draw out the implications of the regression coefficients more clearly, we again calculate predicted asset levels for a variety of stylized low and high wealth households that experienced different shocks in different market environments. Table 5 presents the results of these calculations. Initial asset levels are taken to be the mean for each quartile. For each asset level, the table contrasts the experience of a household that had no asset shock with the experience of a household that suffered a 31% asset loss (which was the mean loss level for the lower wealth quartile households that experienced losses). Other shocks were set to zero and all other household characteristics are set to mean levels for the sample.For the no shock case, the low wealth household shows higher growth than the high wealth household, though this result is driven entirely by the quartile-specific resilience coefficients of dubious significance. However, in the high shock scenario, the excess sensitivity of poor households to asset shocks completely overturns this modest convergent process. Absent good market access, a low wealth household that experienced an immediate 31% asset loss is estimated to experience further declines and a net asset growth rate of -48% from its pre-Mitch position to the time of the study 30 months later.A wealthier household that experienced an identical 31% loss is estimated to have recovered partially from the loss and exhibit a net growth rate of -14%. Were we to further take into account that wealthier households on average only lost 7.5% of their assets (not 31%), then the unequalising effect of the shock would be further magnified.When poor households are compared to where they counterfactually would have been without a shock, the impacts of the shocks stand out even more sharply. Finally, Table 5 shows that more buoyant labor and capital market access serves to offset the unequalising effect of asset shocks. Under these circumstances, lower wealth households would be able to offset much of the 31% asset loss (climbing back to a net asset change of -10%). Wealthier households are also estimated to benefit slightly from better market access, but the final recovery gap between poor and rich households is almost eliminated. These results are especially interesting in the context of a related study by Carter and Castillo (2005). In that study, Carter and Castillo find that recovery from Mitch was more rapid in communities characterized by high levels of pro-social norms of trust and altruism. Interestingly, further analysis by Carter and Castillo suggests that only a subset of households seems to actually benefit from the pro-social environment, suggesting that there may be processes of exclusion that prevent all households from benefiting from socially mediated access to insurance and capital. If correct, when merged with the results presented here, we seem to see a situation in which local social mechanisms leave poor Honduran households quite vulnerable to asset shocks. In this environment, access to supporting capital and especially labour markets seem to be especially important.In the fictive world of full and complete markets, poor households could draw on loan and insurance contracts to cope with the often disastrous asset and income losses brought by severe environmental shocks. Drawing on future earnings, households in this world could rebuild lost assets and sustain their level of current consumption without further depletion of their productive assets and future possibilities. The story of oneEthiopian household told at the beginning of this paper is a case study of how far the actually existing world can be from that fictive world. In the real world of Ato Mohammed, environmental shocks can decapitalise the poor, and trap them in impoverished position from which they cannot escape. When this happens, a humanitarian problem of disaster relief becomes a long-term development problem.In an effort to better understand the nature of environmental shocks, this paper has employed longitudinal data on assets to understand the longer term impacts of two environmental shocks, the three-year drought of the late 1990s in Ethiopia and the 1998 Hurricane Mitch in Honduras. Analysis of the Ethiopian data reveals a disturbing pattern of asset smoothing or protection among the lowest wealth households, meaning that the household tries to desperately hold on to its few assets even as income and consumption possibilities dwindle. The analysis also reveals a pattern of weak resilience amongst the poorest Ethiopian households, meaning that those who exit a shock with few assets experience difficulties in rebuilding their assets and livelihoods. Together these patterns at least hint at the generality of the experience of Ato Mohammed.Analysis of the Honduran data reveals a different, but similarly provocative pattern. Relatively wealthy households seem to be able to protect their assets in the wake of a shock, while poorer households are apparently put on a downward trajectory of further asset depletion as they cope with a shock. While these households do not exhibit the asset protection strategies of the Ethiopian poor, they too appear to be quite distant from the world of full and complete markets. While there is weak evidence that poorer Honduran households can begin to rebuild their assets, absent buoyant factor markets for labor (and to a lesser extent capital), the rebuilding process is slow, and the net effect of shocks appears to be profoundly unequalising, at least over a medium term.The analysis here has of course fallen short of fully resolving all the puzzles and complexities of even the two disasters studied here. Further research may reveal other important facets of wealth-differentiated asset recovery experiences by exploring the asset composition of varied wealth categories of households. For example, do the poor in Ethiopia hold most of the animal assets in small stock and chickens, while the better-off own more cattle and plough oxen? These types of assets differ in their \"lumpiness\", their functions in income and wealth generation, and in the extent to which they are protected in the face of shocks. Future studies should address the likelihood that different asset portfolios of the rich and poor may constrain or facilitate post-disaster recovery paths.While future research to solve remaining puzzles is always desirable, given the importance of shocks, it seems worth speculating on the policy and development implications of our findings, especially since most disasters and their impacts are treated as humanitarian not development problems. An important first step would be to build social safety ('insurance') nets that keep vulnerable households from losing their assets and sinking further into poverty. For the chronically poor, a safety net of guaranteed food needs and, in some cases, minimal cash income, can allow them to divert efforts from survival-type (often destructive) coping strategies, to more remunerative activities that might build assets and 'pull' them out of poverty. Given that social networks and institutions play an important role in keeping households from falling into poverty, externally supported safety nets, as well as any form of development policy in general, need to be cognizant of the way in which such social networks operate so as to minimize any potential negative impact of programs on existing social institutions.The estimated relevance of markets for households' ability to resort to livelihoods that do not lead to asset erosion suggests programs that go further than building safety nets. Policies that improve non-farm employment opportunities, rural market infrastructure, and availability of credit-especially in the post-disaster period-are important ways that governments and development agencies can help limit long-term asset depletion. Our findings show that market conditions do make a difference in how shocks differentially affect certain communities and regions. Policies that make markets more accessible to the chronically poor and vulnerable may mitigate the kind of widespread human suffering now associated with natural disasters.No ","tokenCount":"7814","images":["-70294948_1_1.png","-70294948_1_2.png","-70294948_1_3.png","-70294948_2_1.png","-70294948_2_2.png","-70294948_2_3.png"],"tables":["-70294948_1_1.json","-70294948_2_1.json","-70294948_3_1.json","-70294948_4_1.json","-70294948_5_1.json","-70294948_6_1.json","-70294948_7_1.json","-70294948_8_1.json","-70294948_9_1.json","-70294948_10_1.json","-70294948_11_1.json","-70294948_12_1.json","-70294948_13_1.json","-70294948_14_1.json","-70294948_15_1.json","-70294948_16_1.json","-70294948_17_1.json","-70294948_18_1.json","-70294948_19_1.json","-70294948_20_1.json","-70294948_21_1.json","-70294948_22_1.json","-70294948_23_1.json","-70294948_24_1.json","-70294948_25_1.json","-70294948_26_1.json","-70294948_27_1.json","-70294948_28_1.json","-70294948_29_1.json","-70294948_30_1.json","-70294948_31_1.json","-70294948_32_1.json","-70294948_33_1.json","-70294948_34_1.json","-70294948_35_1.json","-70294948_36_1.json","-70294948_37_1.json","-70294948_38_1.json","-70294948_39_1.json","-70294948_40_1.json","-70294948_41_1.json","-70294948_42_1.json","-70294948_43_1.json","-70294948_44_1.json","-70294948_45_1.json"]}

data/part_2/0322753219.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"ef37d8fe5564f8960969a117a98f2902","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/49058bda-f489-4d27-8a92-6a5d36aa96ba/retrieve","description":"The social protection system in Myanmar has remained at a rudimentary level for the past decade, with policies scattered and fragmented across various government departments, and serving only a fraction of the eligible population. The government allocated only 0.8 percent of its expenditure to social protection constraining its ability to expand to vulnerable groups leaving households to rely on informal forms of safety nets against idiosyncratic and covariate shocks, and life-course contingencies (Niño-Zarazúa & Tarp 2021). Only 13.8 percent of the population received any form of social protection according to the 2017 MLCS, leaving much of the poor, which is about one-third of the population, out of the scope of protection. After the military takeover in 2021, government provision of social protection faced a complete collapse with near zero allocation to the population (MAPSA 2022c). In the face of the double predicament of the COVID-19 pandemic and coup, any form of anti-poverty investment should effectively target the poor based on observable and verifiable characteristics. In this research note, we explore some promising indicators which can be used by implementing agencies to effectively target the poor. We use data from the Myanmar Household Welfare Survey (MHWS) collected over the phone during July and August of 2022. The survey was conducted among 12,000 households in 310 townships of Myanmar. The MHWS is a nationally, urban/rural and state/region representative phone survey (MAPSA 2022a). The household survey questionnaire collected information on a wide variety of topics such as household composition, occupation, education, dwelling characteristics, assets, income, and agriculture.","id":"1378899601"},"keywords":[],"sieverID":"1644556c-6a33-4ef5-93f6-97d846491acd","pagecount":"2","content":"Social protection in Myanmar is currently very limited, and scarce development partner resources should be as accurately targeted to needy populations as possible, based on observable and verifiable characteristics. This note summarizes some promising poverty targeting indicators from the Myanmar Household Welfare Survey (MHWS), conducted by phone among 12,000 households in 310 townships in July/August 2022.1. Households with walls made of rudimentary materials such as hemp/hay/bamboo, etc. 2. Household with rudimentary electricity connection such as solar, battery, water mill or nothing. 3. Household located more than 2 hours away from the nearest township center. 4. Household whose primary source of income is agricultural wage work. 5. Households with floor not made of improved materials, wood, tile, vinyl, etc. 6. Households without improved toilet facility (has open pit, hanging latrine or no facility). 7. Households without any fridge. 8. Households without any wardrobe. 9. Households without any agricultural land ownership.  Low FCS Low adult DD Furthermore, certain demographic groups are of interest because of nutritional vulnerabilities:1. Household has at least one child aged 0 to 5 years.2. Household has at least two adolescent girls aged 12 to 20 years.3. Household has at least two school aged children 5 to 14 years. 4. Household with school age children 5 to 14 years not going to school. 5. Households with highest education of adults' below primary level.As shown in Figure 1, almost all the proposed indicators predict statistically significant increased risks of poverty and food security. As an example, households who primary income source is agricultural work are 28 percent more likely to be income poor, 6 percent more likely to experience hunger, 18 percent more likely to have low FCS and 13 percent more likely to have low adult diet diversity. Housing materials, asset ownership, and education are also strong predictors of these outcomes.Table 1 reports the percentage of poor households correctly identified by different combinations of indicators to assess potential errors of inclusion (nonpoor households incorrectly identified as poor). MHWS estimates suggest that 61 percent of households were income poor in July-August 2022. If we consider households with walls made of rudimentary materials, rudimentary electricity, located more than 2 hours away from the nearest township, and primary source of income being agricultural wage work as selection criteria (core indicators), we find that nearly 53 percent of the income poor can be correctly identified as poor with at least 3 of the 4 proposed indicators (see \"Core\" column in Table 1). Adding housing characteristics can improving targeting somewhat, but adding assets (no fridge or wardrobe) allows a sizable improvement in targeting: we can identify 66 percent of the poor with any 3 of the 6 criteria and 78 percent with any 4 of the 6 criteria (Column 3).In rural and urban areas we find slightly different results (not reported). Just 3 of the 4 core indicators correctly identifies nearly 62 percent of the income poor in rural Myanmar, while adding assets identifies 67 percent of the poor with any 3 of the 6 criteria and 82 percent with any 4. In urban areas, assets are more important, identifying 63 percent of the poor with any 3 indicators and 67 percent of the poor with any 4 of the 6 criteria. Demographic variable do not assist as much in reducing errors of exclusion, but are useful for identifying households that may be nutritionally vulnerable if they are additionally economically vulnerable. Adoption of some or all these indicators will help programmers cost-effectively target safety net programs, but programs should additionally validate and adapt their indicators based local settings.","tokenCount":"594","images":["1378899601_1_1.png","1378899601_2_1.png"],"tables":["1378899601_1_1.json","1378899601_2_1.json"]}

data/part_2/0326262871.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"98b202e6fe082564a000093980ae9a31","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/336aba5c-756f-4b46-9eb9-e0b7371012cb/retrieve","description":"","id":"1121707045"},"keywords":[],"sieverID":"2df7c2fe-ebc1-4347-a76f-20d26a935432","pagecount":"29","content":"Recent studies have indicated that Africa as a whole and a number of individual countries have exhibited relatively strong trade performance in the global market (Bouët et al. 2014)  as well as in continental and major regional markets (Badiane et al. 2014). The increased competitiveness has generally translated into higher shares of regional markets in total exports by the different groupings. Faster growth in demand in continental and regional markets compared to the global market has also boosted the export performance of African countries. For instance, during the second half of the last decade, Africa's share of the global export market has risen sharply, in relative terms, for all goods and agricultural products in value terms, from 0.05 to 0.21 % and from 0.15 to 0.34 %, respectively. This is in line with the stronger competitive position of African exporters mentioned earlier.By promoting competition and specialization in production, regional tradesimilar to global trade-can contribute to food security through its impact on long-term output and productivity growth. At the same time, it can positively affect employment and incomes. Where these effects are positive, trade increases the availability of food and improves the accessibility of food to affected segments of the population. Trade also helps reduce the unit cost of supplying food to local markets, thereby lowering food prices or reducing the pace of food price increase, which in turn improves the affordability of food. Finally, trade can also help stabilize supplies in domestic food markets and reduce the associated risks to vulnerable groups.All of the above-mentioned benefits can be obtained, perhaps to a larger extent, through trading with the rest of the world. For instance, one could question why a O. Badiane ( ) • S.given country should pursue an expansion of regional trade as opposed to global trade in general for stabilizing domestic food supplies, given that world production can be expected to be more stable than regional production. Several factors, such as transport costs, foreign exchange availability, responsiveness of the import sector, and dietary preferences, may provide valid economic justification for a country's efforts to boost regional trade as part of a wider supply stabilization strategy that would also include increased trade with extra-regional markets. Regional and global trade should therefore be seen as complementary rather than as substitutes.The increase in intra-African and intra-regional trade, and the rising role of continental and regional markets as major destinations of agricultural exports by African countries suggest that cross-border trade flows will exert greater influence on the level and stability of domestic food supplies. The more countries find ways to accelerate the pace of intra-trade growth, the larger that influence is expected to be in the future. The current chapter examines the future outlook for intra-regional trade expansion and the implications for volatility of regional food markets. The chapter starts with an analysis of the potential of regional trade to contribute to stabilizing food markets, followed by an assessment of the scope for cross-border trade expansion. A regional trade simulation model is then developed and used to simulate alternative scenarios to boost trade and reduce volatility in regional markets.Variability of domestic production is a major contributor to local food price instability in low income countries. The causes of production variability are such that an entire region is less likely to be affected than individual countries. Moreover, fluctuations in national production tend to partially offset each other, so that such fluctuations are less than perfectly correlated. Food production can be expected to be more stable at regional level than at country level. In this case, expanding cross-border trade and allowing greater integration of domestic food markets would reduce supply volatility and price instability in these markets. Integrating regional markets through increased trade raises the capacity of domestic markets to absorb local price risks by: (1) enlarging the area of production and consumption and thus increasing the volume of demand and supply that can be adjusted to respond to and dampen the effects of shocks;(2) providing incentives to invest in marketing services and expand capacities and activities in the marketing sector, which raises the capacity of the private sector to respond to future shocks; and (3) lowering the size of needed carryover stocks, thereby reducing the cost of supplying markets during periods of shortage and hence decreasing the likely amplitude of price variation.A simple comparison of the cereal production variability in individual countries against the regional average is carried out to illustrate the potential for local market stabilization through greater market integration (Badiane 1988). For that purpose, a trend-corrected coefficient of variation is used as a measure of production variability at both country and regional levels. We then use a normalization procedure whereby the value of the coefficient for each country is divided by the value of the coefficient for the corresponding region. Calculations are carried out for each of the three regional economic groupings (as mentioned above), and the results are presented in Table 16.6 in the annex and plotted in Fig. 16.1a-c below. The bars in the figures represent the normalized coefficients of variation, which indicate how much more (when normalized coefficient are greater than 1) or less (when normalized coefficient are less than 1) volatile a country's production is when compared with production at the level of their respective region.Of the three regions, SADC has the highest level of aggregate volatility with a coefficient of variation of 18.58 or more than two and three times that of ECOWAS and COMESA, respectively. For the vast majority of countries, national production volatility is considerably larger than regional level volatility. The only exceptions are the Democratic Republic of Congo (DRC) in SADC and to a lesser extent Côte d'Ivoire in ECOWAS. None of the COMESA countries has a more stable production than the regional aggregate. The COMESA countries can be divided into two subgroups: (1) a relatively low volatility subgroup with normalized coefficients of less than twice the regional average, including Burundi, Comoros, DRC, Egypt, and Uganda and (2) a high volatility regional subgroup with volatility levels that are at least five times higher than the regional level, comprising Malawi, Mauritius,1 Rwanda, Sudan, Swaziland, Zambia, and Zimbabwe. Kenya and Madagascar both have moderate levels of volatility and fall between the two groups. Most countries in SADC and ECOWAS are in the moderate regional category, with only Botswana and Mauritius (in SADC), and Gambia, Liberia, Mali, and Senegal (in ECOWAS) showing volatility levels more than three times higher than the respective regional levels. The countries in the moderate-and high-volatility subgroups would benefit the most from increased regional trade in terms of greater stability of domestic supplies.The likelihood that a given country would benefit from the trade stabilization potential, as suggested by the difference between its volatility level and the regional average, will be greater if its production fluctuates more and is weakly correlated with that of the other countries in the region. Figure 16.2 presents the distribution of correlation coefficients between individual country's production levels for each regional group. For each country, the lower segment of the bar shows the percentage of correlation coefficients that are 0.65 or less or the share of countries with production fluctuations that are defined as relatively weakly correlated with the country's own production movements. The top segment represents the share of countries with highly correlated production fluctuations, with coefficients that are higher than 0.75. The middle segment is the share of moderately correlated country productions, with coefficients that are between 0.65 and 0.75. in Fig. 16.2c, only three countries have less than an 80 % share of correlation coefficients below 0.65. The combination of high volatility and weak correlation suggests that countries in this region would benefit the most from increased regional trade in terms of domestic market stabilization. They are followed by COMESA countries, where 60 % of the correlation coefficients for any given country are below 0.65. In contrast, country-level production levels in the ECOWAS region tend to fluctuate more together than the other two regions, as shown by the high share of coefficients above 0.75. The division of the region into two nearly uniform subregions, Sahelian and coastal, may be an explanation. In general, however, the patterns and distribution of production fluctuations among countries in all the three regions are such that increased trade could be expected to have a stabilizing effect on domestic agricultural and food markets. But that is only one condition; the other is that there is actual potential to increase cross-border trade, a question that will be examined in the next section.Despite the recent upward trends, the level of intra-African and intra-regional trade is still very low compared with other regions. Intra-African markets accounted only for an average 34 % of the total agricultural exports from African countries between 2007 and 2011 (Badiane et al. 2014). Among the three RECs, SADC had the highest share of intra-regional trade (42 %), and ECOWAS the lowest (6 %). COMESA's share of intra-regional trade was 20 %. Although SADC is doing much better than the other two RECs, its member countries still account for far less than half of the value of agricultural trade within the region (Badiane et al. 2014).There may be a host of factors behind the low levels of intra-regional trade. These factors may not only make trading with extra-regional partners more attractive, but they may also raise the cost of supplying regional markets from intra-regional sources. The exploitation of the regional stabilization potential, as pointed out above, would require measures to lower the barriers to and the bias against transborder trade such as to stimulate the expansion of regional supply capacities and of trade flows across borders. This supposes that there is sufficient scope for specialization in production and trade within the subregions. Often, it is assumed that neighboring developing countries would exhibit similar production and trading patterns because of the similarities in their resource bases, leaving little room for future specialization. There are, however, several factors that may lead to different specialization patterns among such countries. These factors include (1) differences in historical technological investments and thus the level and structure of accumulated production capacities and skills; (2) the economic distance to, and opportunity to trade with, distant markets; and (3) differences in dietary patterns as well as consumer preferences that affect the structure of local production. The different patterns of specialization in Senegal compared with the rest of Sahelian West Africa and in Kenya compared with other Eastern African countries well illustrate the influence of these factors.Consequently, we use a series of indicators to assess the actual degree of specialization in agricultural production and trade, and whether there is real scope for transborder trade expansion as a strategy to exploit the less-than-perfect correlation between national productions to reduce the vulnerability of domestic food markets to shocks. The first two indicators are the production and export similarity indices, which measure and rank the relative importance of the production and trading of individual agricultural products in every country. The level of importance or position of each product is then compared for all relevant pairs of countries within each subregion. 2 The indices have a maximum value of 100; an index value of 100 implies that the production or trade patterns between the considered pair of countries are completely similar. The closer the index value is to zero, the greater the degree of specialization between the two countries. Index values of around 50 and below are interpreted as indicating patterns of specialization that are compatible with higher degrees of trade expansion. The estimated indicator values for the three regional groupings, covering 150 products in total, are presented in of index values. The vast majority of country pairs fall within the 0-50 range. A value of less than 60 is conventionally interpreted as compatible with higher trade exchange between the considered pair of countries. The estimated index values therefore suggest that there exists sufficient dissimilarity in the current production and trading patterns between countries and hence a scope for transborder trade expansion in all three subregions.The third indicator, the revealed comparative advantage (RCA) index, is computed to further assess the degree of trade specialization among countries within the three regions. The RCA index compares the share of a given product in a given country's export basket with that of the same product in total world exports. A value greater than 1 indicates that the considered country is performing better than the world average; the higher the value is, the stronger the country's performance in exporting the considered product. Of the nearly 600 RCA indicators estimated for various products exported by different COMESA countries, 70 % have an index value higher than 1. ECOWAS and SADC each have a total of about 450 indicators. The share of indicators higher than 1 is about the same as in the case of COMESA: 68 % for SADC and 73 % for ECOWAS. For each regional grouping, the 20 products with the highest normalized RCA index value are presented in Table 16.1. The normalized RCA is positive for RCA indicators that are greater than 1 and negative otherwise. 3 For very high RCA indicators, the normalized value tends toward 1.All the products listed in the table have normalized RCA values above 0.98. The rankings reflect the degree of cross-country specialization within each REC. In ECOWAS, for instance, a total of 12 products, spread across 8 out of 15 member countries, account for the highest 20 indicators for the region. There are 13 products in that category in the case of COMESA, and these products come from 9 out of 19 countries. SADC has the highest number of products in that category, a total of 14, but they come from only 5 out of 15 countries. The table also illustrates the difference in degree of specialization between the three major regions. Only two of the top ranking products (carded and combed cotton, and cashew nuts in shell) are common to the ECOWAS and SADC regions. Even between COMESA and SADC, only six of the top ranking products are common to the two regions, while there are no common top ranking products between COMESA and ECOWAS. A fuller appreciation of the degree of specialization across all countries in the three regions is best obtained by looking at the RCA values for the entire set of products and countries. For instance, if countries have similar patterns of specialization, the same products would tend to rank equally high and the values of the RCA indicator for the same product would not vary significantly across countries. Similarly, if countries have similar patterns of specialization, exports would be concentrated around a few products, with substantial variation of the indicator value across products. An analysis of the variance of the RCA index is, therefore, carried out to test for either of the above-mentioned possibilities. The results of the analysis, presented in Table 16.2, show that for the entire sample of African countries, nearly two-thirds   So far, the analysis has established the existence of dissimilar patterns of specialization in production and trade of agricultural products among countries within and across the three major regions. Two final indicators, the Trade Overlap Indicator (TOI) and the Trade Expansion Indicator (TEI), are calculated to examine the potential to expand trade within the three blocks of countries based on current trade patterns.The indicators measure how much of the same product a given country or region exports and imports at the same time. The TOI measures the overall degree of overlapping trade flows for a country or region as a whole, while the TEI measures the overlapping trade flows at the individual product level for a country or region. The results are presented in Fig. 16.4 and Table 16.3. The results indicate that there is a considerable degree of overlapping trade flows: 25 % for Africa as a whole and as much as 40 % for the SADC region. Normalized TOI values, obtained by dividing country TOI values by the TOI value, for the respective regions can be found in Badiane et al. (2014). In the vast majority of cases, they are significantly less than 1. The overlapping regional trade must therefore be taking place between different importing and exporting countries. In other words, some countries are exporting (importing) the same products that are being imported (exported) by other member countries in their respective grouping, but in both cases to and from countries outside the region. By redirecting such flows, countries should be able to expand transborder trade within their groupings.The TEI indicates which products have the highest potential for increased transborder trade based on the degree of overlapping trade flows. Table 16.3 lists the 20 products with the highest TEI value for each of the three regions. The lowest TEI value for any of the products across the three regions is 0.41. RCA values  16.1, even with current production and trade patterns. The remainder of the chapter therefore analyzes the outlook for intratrade expansion and the expected impact of volatility of regional food markets over the next 15 years. This is done by simulating alternative policy scenarios to boost intra-regional trade and by comparing the resulting effect on the level and volatility of trade flows up to 2025 with outcomes simulated under a baseline scenario that assumes continuation of historical trends. Italics denote products with RCA < 1; products with high TEI but which are not being produced in the regions are included as they relate to re-export trade. There were two in the case of COMESA and SADC and six in the case of ECOWASThe preceding analysis presents evidence that African countries could use increased regional trade to enhance the resilience of domestic markets to supply shocks. The high cost of moving goods across domestic and transborder markets and outwardly biased trading infrastructure are major determinants of the level and direction of trade among African countries. A strategy to exploit the regional stabilization potential therefore has to include measures to lower the general cost of trading and remove additional barriers to cross-border trade. This section simulates the impact of such changes on regional trade flows, using IFPRI's regional Economy-wide Multimarket Model (EMM) described below.4 In this study, the original EMM was modified to differentiate between intra-and extra-regional trade sources and destinations and between informal and formal trade costs in intra-regional trade transactions. In its original version, the EMM solves for optimal levels of supply QX r c , demand QD r c and net trade (either import QM r c or export QE r c ) of different commodities c for individual member countries r of the modeled region. Supply and demand balance at the national level determines domestic output prices PX r c as stated by Eq. (16.1), while Eq. (16.2) connects domestic market prices PD r c to domestic output prices, taking into account an exogenous domestic marketing margin margD r c . The net trade of a commodity in a country is determined through mixed complementarity relationships between producer prices and potential export quantities and between consumer prices and potential import quantities. Accordingly, Eq. ( 16.3) ensures that a country will not export a commodity (QE r;c D 0) as long as the producer price of that commodity is higher than its export parity price, where pwe r c is the country's FOB price and margW r c is an exogenous trade margin accounting for the cost of moving the commodity to and from the border. If the domestic market balance constraint in Eq. ( 16.1) requires that the country exports some excess supply of a commodity (QE r;c > 0), then the producer price will be equal to the export parity price of that commodity. Additionally, Eq. (16.4) governs any country's possibility to import a commodity, where pwm r c is its CIF price. There will be no import (QM r;c D 0) as long as the import parity price of a commodity is higher than its domestic consumer price. The domestic market balance constraint requires that, if a country has to import a commodity to meet a given excess demand (QM r;c > 0), then the domestic consumer price will be equal to the import parity price of that commodity. In the version of the EMM used in this study, the net export of any commodity is modeled as an aggregate of two output varieties differentiated by their market outlets (regional and extra-regional) while assuming an imperfect transformability between the two export varieties. Similarly, the net import of any commodity is modeled as a composite of two varieties differentiated by their origins (regional and extra-regional) while assuming an imperfect substitutability between the two import varieties.In order to implement export differentiation by destination, the mixed complementarity relationship in Eq. ( 16.3) is replaced with two new equations which specify the price conditions for export to be possible to both destinations. Equation (16.5) indicates that for export to extra-regional market outlets to take place (QEZ r c > 0), suppliers should be willing to accept a price PEZ r c that is not greater than the export parity price when exporting to that destination. Similarly, Eq. ( 16.6) ensures that exporting to within-region market outlets is possible (QER r c > 0) only if suppliers are willing to receive a price PER r c that is not more than the regional market clearing price PR c adjusted downward to account for exogenous regional trade margins margR r c incurred in moving the commodity from the farm gate to regional market (see Eq. (16.17) below for the determination of PR c ). Subject to these price conditions, Eqs. (16.7)-(16.10) determine the aggregate export quantity and its optimal allocation to alternative destinations. Equation (16.7) indicates that the aggregate export of a commodity by individual countries QE r c is obtained through a constant elasticity of transformation (CET) function of the quantity QEZ r c exported to extra-regional market outlets and the quantity QER r c exported to intra-regional market outlets, where e r c , ı e r c ; and ˛e r c are the CET function exponent, share parameter, and shift parameter, respectively. Equation (16.8) is the first-order condition of an aggregate export revenue maximization problem, given the prices that suppliers can receive for the different export destinations and subject to the CET export aggregation function. The equation indicates that an increase in the ratio of intra-regional to extra-regional prices will increase the ratio of intra-regional to extra-regional export quantities (i.e., exports shift toward destinations which offer higher returns). Equation (16.9) helps identify the optimal quantities supplied to each destination; it states that aggregate export revenue at producer price of export PE r c is the sum of export sales revenues from both intraregional and extra-regional market outlets at supplier prices, while Eq. (16.10) sets the producer price of export to be the same as the domestic output price PX r c , which is determined by the supply and demand balance equation (Eq. 16.1) as earlier explained. Import differentiation by origin is implemented by following the same procedure for export differentiation by destination, as described above. Equation (16.4) is replaced by Eqs. (16.11) and (16.12). Accordingly, import from extra-regional origins will happen (QMZ r;c > 0) only if domestic consumers are willing to pay a price PMZ r c that is not smaller than the import parity price for the extra-regional variety. Furthermore, import from intra-regional origins is possible (QMR r;c > 0) only if domestic consumers are willing to pay at a price PMR r c that is not smaller than the regional market clearing price PR c adjusted upward to account for exogenous regional trade margins margR r c incurred in moving the commodity from the regional market to consumers. Under these price conditions, Eq. ( 16.13) represents aggregate import quantity QM r c as a composite of intra-and extra-regional import variety quantities QMR r c and QMZ r c , respectively, using a constant elasticity of substitution (CES) function; in the equation, the terms m r c , ı m r c , and ˛m r c stand for the CES function exponent, share parameter, and shift parameter, respectively. The optimal mix of the two varieties is defined by Eq. (16.14), which is the first-order condition of an aggregate import cost minimization problem, subject to the CES aggregation (Eq. 16.13) and given import prices from both origins. An increase in the ratio of extra-regional to intra-regional import prices will increase the ratio of intra-regional to extra-regional import quantities (i.e., imports shift away from more expensive sources). Equation (16.15) identifies the specific quantities imported from each origin. It defines the total import cost at consumer price of import PM r c as the sum of intra-regional and extra-regional import costs, while Eq. ( 16.16) sets the consumer price of import to be the same as the domestic market price PD r c , which is determined by Eqs. (16.1) and ( 16 After determining export quantities and prices by destination, and import quantities and prices by origin, the regional market clearing price PR c can now be solved. Equation ( 16.17) imposes the regional market balance constraint by equating the sum of intra-regional export supplies to the sum of intra-regional import demands, with qdstk c standing for discrepancies existing in observed aggregate intra-regional export and import quantity data in the model's base year. Thus, PR c is the price that ensures regional market balance.The model is calibrated separately for each of the three RECs. Calibration is performed such that for every member country within each REC, the same production, consumption, and net trade data are replicated as observed for different agricultural subsectors and two nonagricultural subsectors in 2007-2008. Baseline trend scenarios are then constructed such that until 2025, changes in crop yields, cultivated areas, outputs, and GDP reflect the same observed changes. Table 16.6 in the annex compares the calibrated agricultural and economy-wide GDP growth rates under the baseline scenario with the observed rates in the recent years. Although the model is calibrated to the state of national economies 7 years earlier, it closely reproduces the countries' current growth performances. Four different scenarios are simulated using the EMM. The first is the baseline scenario described above, which assumes a continuation of current trends up to 2025. It is used later as a reference to evaluate the impact of the changes under the remaining three scenarios. The latter scenarios introduce the following three different sets of changes to examine their impacts on regional trade levels: a reduction of 10 % in the overall cost of trading in every country; removal of all crossborder trade barriers-that is, a reduction of their tariff equivalent to zero; and a 10 % yield increase across the board. These changes are modeled to take place between 2008 (the base year) and 2025. The change in cross-border exports is used as an indicator of the impact on intra-regional trade. In the original data, there are large discrepancies between recorded regional exports and import levels, the value of the latter often being multiples of the former. The more conservative export figures are therefore the preferred indicator of intra-regional trade.The results for the different regions are presented in Figs. 16.5 and 16.6. Figure 16.5 presents the results of the baseline scenarios for the three regions from 2008 to 2025. Assuming the current trends to continue, intra-regional trade in both ECOWAS and SADC is expected to expand rapidly but with marked differences between crops. The aggregate volume of intra-regional trade in staples would approach 3 million tons in the case of ECOWAS and about half of that amount in the case of SADC if the growth rates in yields, cultivated areas, and nonagricultural income sustained at their current level until 2025. Cereals would see the smallest gains, while trade in roots and tubers as well as other food crops would experience much faster growth in the case of ECOWAS. This is in line with the current structure of and trends in commodity demand and trade. While the increase in demand for roots and tubers is being met almost exclusively using local sources, the fast growing demand for cereals is heavily tilted toward rice, which is supplied from outside the region. The two leading cereals that are traded regionally, maize and millet, therefore benefit less from the expansion of regional demand and have historically seen slower growth in trade than roots and tubers. In the case of SADC, the rise of Angola as a main exporter of roots and tubers starting in 2013 is a main factor in explaining the strong boost in regional trade of that commodity. Zimbabwe had been the sole exporter of roots and tubers before 2013 and exported only very modest quantities. Hence, the high rates of growth of overall regional exports can be attributed to the developments in Angola.The story is a bit different in the case of COMESA. As was already made apparent by the market share analysis earlier, the COMESA regional market has been the least dynamic of the three regional markets and the only one associated with a negative market effect. COMESA is the only region where the member countries have experienced a decline in competitiveness as a whole. The underwhelming performance is reflected in the baseline scenario. If current trends were to continue, the levels of intra-regional trade would continue to stagnate, except in the case of cereals. And even for this group of products, the decline in trade volumes would be reversed, but the reversal would not be enough to bring the trade volumes back to their initial levels. The projected evolution of the trade in cereals reflects different country dynamics and a shift in the sources of regional exports. The fall in regional trade levels at the beginning of the period is a result of a continual decline in exports from the two main traditional suppliers Egypt and Malawi. At the same time, the faster growth in several other countries, particularly Tanzania and Ethiopia, results in rising exports from these countries, starting from 2011 for Tanzania and from 2019 for Ethiopia. The result is a U-shaped pattern in COMESA cereals exports: the declining exports in some countries are eventually offset by the increasing imports in other countries. The graphs in Fig. 16.6 show the cumulated changes in intra- The results show that intra-regional trade invariably increases by a considerable margin for cereals and roots and tubers (the main food crops) in response to changes in trading costs and yields. Intra-community trade levels in ECOWAS climb by between 10 and 35 % for most products over the entire period. By 2025, when compared to baseline trends, the volume of cereal trade increases by a cumulative total of between 200,000 and 300,000 mt for individual products and the volume of overall staple trade by between 1.5 and 4.0 million tons. Cereals seem to respond better than other products in general. It also appears that removing transborder trade barriers would have the strongest impact of trade flows across the board.The COMESA region shows similar increases in overall trade in staples. Cereals trade tends to be proportionally less responsive but because of its initial higher levels, the cumulative additional volume of regional trade is much higher, ranging from 0.7 million to more than 3.0 million tons above the baseline. Also, in contrast to ECOWAS, intra-regional trade in COMESA seems to be more responsive to changes in overall trading costs and yields than to changes in cross-border barriers. This may be explained by the fact that equivalent tariffs constitute a smaller fraction of producer prices, and hence changes in barriers result in smaller changes in incentives. Trade in the SADC region also seems to respond more to changes in transborder trade barriers and yields, as in the case of ECOWAS. A 10 % increase in yields would raise trade in staples by a cumulative volume of slightly more than 3.0 million tons by 2025 compared to the baseline scenario.Under each scenario, the model-simulated quantities of intra-regional exports QER r c are used to estimate an index of future export volatility at country and regional level as follows: First, a trend-corrected coefficient of variation TCV is calculated for each country, using the following formula as in Cuddy and Della Valle (1978):where CV is the coefficient of variation and R 2 is the adjusted coefficient of determination of the linear trend regression obtained using the time series of aggregate quantities of intraregional exports of all staple food crops from 2008 to 2025.Second, an index of regional volatility TCV REC is derived for each REC as a weighted average of trend-corrected coefficients of variation of its member countries with the formulawhere TCV i and TCV j are the trend-corrected coefficients of variation in aggregate exports of staple food crops in countries i and j, n is the number of member countries in the REC, s i and s j are the shares of countries i and j in the region's overall intra-regional exports of staple food crops, and v ij is the coefficient of correlation between aggregate exports of countries i and j. Finally, the coefficients of variation at country level are normalized by dividing them by the respective regional coefficients.The historical and simulated levels of cross-border trade volatility of food staples in the various regions are reported in Table 16.4. The volatility levels simulated under historical trends are calculated based on the TradeMaps database. 5 Table 16.5 shows the comparison of the simulated volatility levels under the various alternative scenarios with historical volatility levels, with the difference expressed in absolute point changes. The figures in the two tables show that volatility levels are lower under nearly all scenarios than under historical trends. The only exception is in the case of ECOWAS, where regional cross-border trade volatility decreases with a reduction of overall trading costs, but it rises when cross-border trade barriers In the SADC case, baseline and historical trends of the trade volatility deviate a lot. The main explanation is that, unlike traditional CGE models where countries are exporters or importers from the beginning and remain as such for the length of the simulation period, our model allows countries to enter or exit the regional export market based on relative prices. Therefore, we have used historical production as opposed to trade data to calibrate the model, given that not all countries have historical trade data. The baseline volatility of trade flows is therefore not a result of calibration but rather derives from the calibrated baseline production and its induced trade flows. The SADC region, unlike other regions, has undergone a major structural change in terms of the composition and source of production and thus trade of agricultural products, with Angola, a new player, emerging as the most important trading partner and roots and tubers as the single most important traded agricultural commodity. The projected overwhelming dominance of the more stable Angola in regional production and trade under continuation of current trends is the main explanation of the drop in baseline export volatility.  are removed or when yields are increased. The magnitude of changes are, however, rather small across all three scenarios. The figures also show that when the current trend of rising volumes of intra-regional trade continues, volatility levels in all three regions are expected to decline compared to historical trends. A better comparison is therefore to contrast changes that take place under the two trade policy scenarios and the productivity (meaning increasing yields) scenario with the expected volatility levels under the baseline scenario. Furthermore, the direction and magnitude of changes in the level of intra-regional trade volatility are determined by the combined effect of changes in the level of volatility as well as changes in the share of crossborder exports in individual countries. Figure 16.7 above shows changes in volatility levels (x-axis) and shares of exports (y-axis) by individual countries under each of the scenarios when compared with the baseline. The different dots indicate the position of different countries under the three scenarios. The tilted distribution of country positions to the left of the x-axis indicates that most countries' exports would experience a lower level of volatility. The combined changes in export share and volatility for individual countries under each of the scenarios are reported in Table 16.7 and presented in Figs. 16.8,16.9,16.10 in the Annex. Only countries that have historically exported are considered. Changes in a country's production patterns resulting from the simulated policy actions lead to changes in both the volatility and the level of exports,and hence the shares in regional trade of each country. The magnitude and direction of these changes determine the contribution of individual countries to changes in the volatility level in regional food markets.The current chapter has examined the potential to use increased intra-regional trade among Africa's main regional economic communities as a means to raise the resilience of domestic food markets to shocks across their member countries. The distribution and correlation of production volatility as well as the current patterns of specialization in the production and trade of agricultural products among African countries suggest that it is indeed possible to raise cross-border trade to reduce the level of instability of local food markets. The results of the baseline scenario indicate that continuation of recent trends would sustain the expansion of intra-regional trade flows in all three regions, particularly in the ECOWAS region. The findings also reveal that it is possible to significantly boost the pace of regional trade expansion, which in turn would contribute to creating more resilient domestic food markets through modest reduction in the overall cost of trading, a similarly modest increase in crop yields, or the removal of barriers to transborder trade. More importantly, the simulation results also suggest that such policy actions to promote transborder trade would reduce volatility in regional markets and help lower the vulnerability of domestic food markets to shocks.  The images or other third party material in this chapter are included in the work's Creative Commons license, unless indicated otherwise in the credit line; if such material is not included in the work's Creative Commons license and the respective action is not permitted by statutory regulation, users will need to obtain permission from the license holder to duplicate, adapt or reproduce the material.","tokenCount":"6380","images":["1121707045_1_1.png","1121707045_12_1.png"],"tables":["1121707045_1_1.json","1121707045_2_1.json","1121707045_3_1.json","1121707045_4_1.json","1121707045_5_1.json","1121707045_6_1.json","1121707045_7_1.json","1121707045_8_1.json","1121707045_9_1.json","1121707045_10_1.json","1121707045_11_1.json","1121707045_12_1.json","1121707045_13_1.json","1121707045_14_1.json","1121707045_15_1.json","1121707045_16_1.json","1121707045_17_1.json","1121707045_18_1.json","1121707045_19_1.json","1121707045_20_1.json","1121707045_21_1.json","1121707045_22_1.json","1121707045_23_1.json","1121707045_24_1.json","1121707045_25_1.json","1121707045_26_1.json","1121707045_27_1.json","1121707045_28_1.json","1121707045_29_1.json"]}

data/part_2/0327022827.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"af0155596c28847820adefba5490bc16","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/26a8fa8c-eddf-4813-99da-d565421de41a/retrieve","description":"En las últimas décadas, el mundo ha hecho un progreso impresionante para mejorar la calidad de vida de millones de personas; sin embargo, todavía sigue inconclusa la tarea de garantizar la seguridad alimentaria a los más pobres de una manera sostenible. La explosión demográfica, la expansión urbana, la desnutrición y la mala salud persistentes, las tierras agrícolas degradadas y el agua escasa, la carencia de poder de las mujeres, la globalización acelerada y la rápida aparición de nuevas tecnologías - todos estos y muchos otros factores están influenciando este esfuerzo continuo. Este libro compila docenas de resúmenes y artículos para presentar las perspectivas de los expertos sobre estos tópicos vitales. Producidas como parte de la iniciativa \"Visión de la alimentación, la agricultura y el medio ambiente en el año 2020,\" del Instituto Internacional de Investigaciones sobre Políticas Alimentarias, las piezas coleccionadas aquí ofrecen una representación completa de los temas de política que el mundo debe abordar si ha de superar la pobreza, el hambre y la degradación ambiental; y le señalan además el camino a las acciones de política que deben ejecutarse para lograr estos objetivos.","id":"-655037674"},"keywords":[],"sieverID":"f1a1bcd0-9417-43b3-8847-94b4a40d3e46","pagecount":"4","content":"Lori Ann Thrupp fue Directora de Agricultura Sostenible del World Resources Institute, Washington, D.C., Estados Unidos, cuando fue coautora del resumen incluido en este volumen. Actualmente es científica de la vida de la Agencia de Protección Ambiental de los Estados Unidos.Stephen Vosti es Profesor Adjunto del Departamento de Economía Agrícola y de Recursos de la Universidad de California en Davis, Estados Unidos.Ellen Wilson es Vicepresidenta de Burness Communications, Bethesda, Maryland, Estados Unidos.Satya Yadav fue investigador asociado del Departamento de Economía Agrícola y Sociología Rural de la Universidad de Arkansas, Estados Unidos.Manfred Zeller fue investigador de la División de Consumo de Alimentos y Nutrición del IFPRI de 1993 a 1999. Ahora es Director del Instituto de Desarrollo Rural, Facultad de Agricultura, Universidad de Goettingen, Goettingen, Alemania.","tokenCount":"125","images":[],"tables":["-655037674_1_1.json","-655037674_2_1.json","-655037674_3_1.json","-655037674_4_1.json"]}

data/part_2/0349036645.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"88fe4e4b2ef925a20c30a56b67aee873","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/596178ab-002c-4445-86c1-a6a2530d41cd/retrieve","description":"Rakesh Kapur POLICY SEMINAR Fertilizer Availability and Affordability: Implications for agricultural productivity and food security MAY 4, 2022 - 9:30 TO 11:30AM EDT","id":"-632854844"},"keywords":[],"sieverID":"c0faecc8-0741-4f90-851b-ce6e312ebdf8","pagecount":"22","content":"India -A large Global Importer of Fertilizers and its Raw Material/ Intermediates ~ 36 Million MTPA valuing more than US$ 10 Billion.Fertiliser Production Capacity ~ 55.9 million MT Urea Imports in the range @ 25-30%.For Urea manufacturing ~ 65% requirement met by Import of RLNG (Natural Gas). Govt. Policy on Urea -Fixed Retail Price and Open Subsidy.Low MRP causes higher Urea consumption.6 P&K fertilizer consumption gets affected on account of high Import price/farmgate price.Urea to DAP farmgate price ratio at 1:5Demand destruction due to higher prices? % increase• Rising price trend in the fertiliser and raw material prices. Fertilizer Prices were flat during initial Covid period; Favourable for increase in acreage and food grain production. GOI intervened to avoid the increase of farmgate prices and raised the fertilizer subsidy on fertilisers. Fertiliser Subsidy raised by GOI in 2021-22: P&K subsidy by 210% to avoid demand destruction; thus ensuring Food grain production. Urea subsidy by 29%, hence Farmers continued to remain insulated from rise in International Prices due to price control on Urea. Fertiliser Subsidy budget in 2022-23 is expected to go up for P&K Fertilisers by USD 11.1 billion and Urea by USD 7.5 billion. Russia Ukraine war began on February 24, 2022.Closure of Odessa /Yuzhny ports impacted the supplies from Black Sea.Shortages in availability of vessels.If prohibitive sanctions imposed on Russia for longer period, it may cut off 3 million MT Ammonia from global trade.Rise in crude prices to impact the freight costs and consequently, the fertiliser prices. Primary source is Middle East region /Africa /South East Asia region @ 66% of total imports Sanctions on Russia may not immediately impact the fertiliser availability.   • Global and Regional Fertilizer Markets to readjust to alternate Supply sources. Tighter Supplies -Availability to be a constraint in the longer run.• Fresh Urea Capacities being added. Addition of around 04 Million MT domestic Urea production, thereby reducing 50% of Urea Imports.• In spite of Higher Fixed Costs of new Urea Plants and substantial increase in LNG prices, the Cost of Production of Domestic Urea continues to be lower than the landed cost of Imported Urea.• Imports from other regions to partly replace the Russia/Belarus shortfall. Natural K from Molasses being pushed as a possible substitute for Potash. Natural Gas price increase to be absorbed in Subsidy Budget. Sanctions may lead to decline of Stock to Use ratio globally.Rebuilding inventories may cause higher prices. Promoting Balanced use of fertilizers. To aggressively reduce the per acre usage of fertilizers. Thrust on Micronutrient rich Specialty Fertilizer segment. Hike in Minimum Support Price (MSP) to incentivize Farmers during this period of volatility. GOI is aiming to double the Wheat Exports (Target ~10 Million MT) to ensure better returns to Farmers.","tokenCount":"453","images":["-632854844_1_1.png","-632854844_1_2.png","-632854844_1_3.png","-632854844_1_4.png","-632854844_1_5.png","-632854844_1_6.png","-632854844_1_7.png","-632854844_1_8.png","-632854844_1_9.png","-632854844_1_10.png","-632854844_2_1.png","-632854844_2_2.png","-632854844_2_3.png","-632854844_2_4.png","-632854844_2_5.png","-632854844_2_6.png","-632854844_2_7.png","-632854844_3_1.png","-632854844_3_2.png","-632854844_3_3.png","-632854844_3_4.png","-632854844_3_5.png","-632854844_3_6.png","-632854844_3_7.png","-632854844_3_8.png","-632854844_3_9.png","-632854844_3_10.png","-632854844_3_11.png","-632854844_3_12.png","-632854844_3_13.png","-632854844_4_1.png","-632854844_4_2.png","-632854844_4_3.png","-632854844_4_4.png","-632854844_4_5.png","-632854844_4_6.png","-632854844_4_7.png","-632854844_4_8.png","-632854844_4_9.png","-632854844_4_10.png","-632854844_4_11.png","-632854844_4_12.png","-632854844_4_13.png","-632854844_5_1.png","-632854844_5_2.png","-632854844_5_3.png","-632854844_5_4.png","-632854844_5_5.png","-632854844_5_6.png","-632854844_5_7.png","-632854844_6_1.png","-632854844_6_2.png","-632854844_6_3.png","-632854844_6_4.png","-632854844_6_5.png","-632854844_6_6.png","-632854844_6_7.png","-632854844_6_8.png","-632854844_6_9.png","-632854844_7_1.png","-632854844_7_2.png","-632854844_7_3.png","-632854844_7_4.png","-632854844_7_5.png","-632854844_7_6.png","-632854844_7_7.png","-632854844_7_8.png","-632854844_7_9.png","-632854844_7_10.png","-632854844_7_11.png","-632854844_7_12.png","-632854844_7_13.png","-632854844_8_1.png","-632854844_8_2.png","-632854844_8_3.png","-632854844_8_4.png","-632854844_8_5.png","-632854844_8_6.png","-632854844_8_7.png","-632854844_8_8.png","-632854844_8_9.png","-632854844_8_10.png","-632854844_8_11.png","-632854844_8_12.png","-632854844_8_13.png","-632854844_8_14.png","-632854844_8_15.png","-632854844_8_16.png","-632854844_8_17.png","-632854844_8_18.png","-632854844_8_19.png","-632854844_8_20.png","-632854844_8_21.png","-632854844_9_1.png","-632854844_9_2.png","-632854844_9_3.png","-632854844_9_4.png","-632854844_9_5.png","-632854844_9_6.png","-632854844_9_7.png","-632854844_9_8.png","-632854844_9_9.png","-632854844_9_10.png","-632854844_10_1.png","-632854844_10_2.png","-632854844_10_3.png","-632854844_10_4.png","-632854844_10_5.png","-632854844_10_6.png","-632854844_10_7.png","-632854844_11_1.png","-632854844_11_2.png","-632854844_11_3.png","-632854844_11_4.png","-632854844_11_5.png","-632854844_11_6.png","-632854844_11_7.png","-632854844_11_8.png","-632854844_11_9.png","-632854844_11_10.png","-632854844_11_11.png","-632854844_11_12.png","-632854844_11_13.png","-632854844_12_1.png","-632854844_12_2.png","-632854844_12_3.png","-632854844_12_4.png","-632854844_12_5.png","-632854844_12_6.png","-632854844_12_7.png","-632854844_12_8.png","-632854844_12_9.png","-632854844_12_10.png","-632854844_12_11.png","-632854844_12_12.png","-632854844_12_13.png","-632854844_12_14.png","-632854844_12_15.png","-632854844_13_1.png","-632854844_13_2.png","-632854844_13_3.png","-632854844_13_4.png","-632854844_13_5.png","-632854844_13_6.png","-632854844_13_7.png","-632854844_13_8.png","-632854844_13_9.png","-632854844_13_10.png","-632854844_13_11.png","-632854844_13_12.png","-632854844_13_13.png","-632854844_14_1.png","-632854844_14_2.png","-632854844_14_3.png","-632854844_14_4.png","-632854844_14_5.png","-632854844_14_6.png","-632854844_14_7.png","-632854844_15_1.png","-632854844_15_2.png","-632854844_15_3.png","-632854844_15_4.png","-632854844_15_5.png","-632854844_15_6.png","-632854844_15_7.png","-632854844_15_8.png","-632854844_15_9.png","-632854844_15_10.png","-632854844_15_11.png","-632854844_15_12.png","-632854844_15_13.png","-632854844_15_14.png","-632854844_15_15.png","-632854844_15_16.png","-632854844_15_17.png","-632854844_15_18.png","-632854844_15_19.png","-632854844_15_20.png","-632854844_15_21.png","-632854844_15_22.png","-632854844_15_23.png","-632854844_15_24.png","-632854844_15_25.png","-632854844_15_26.png","-632854844_15_27.png","-632854844_15_28.png","-632854844_16_1.png","-632854844_16_2.png","-632854844_16_3.png","-632854844_16_4.png","-632854844_16_5.png","-632854844_16_6.png","-632854844_16_7.png","-632854844_17_1.png","-632854844_17_2.png","-632854844_17_3.png","-632854844_17_4.png","-632854844_17_5.png","-632854844_17_6.png","-632854844_17_7.png","-632854844_17_8.png","-632854844_17_9.png","-632854844_17_10.png","-632854844_17_11.png","-632854844_17_12.png","-632854844_17_13.png","-632854844_17_14.png","-632854844_18_1.png","-632854844_18_2.png","-632854844_18_3.png","-632854844_18_4.png","-632854844_18_5.png","-632854844_18_6.png","-632854844_18_7.png","-632854844_18_8.png","-632854844_19_1.png","-632854844_19_2.png","-632854844_19_3.png","-632854844_19_4.png","-632854844_19_5.png","-632854844_19_6.png","-632854844_19_7.png","-632854844_19_8.png","-632854844_19_9.png","-632854844_20_1.png","-632854844_20_2.png","-632854844_20_3.png","-632854844_20_4.png","-632854844_20_5.png","-632854844_20_6.png","-632854844_20_7.png","-632854844_21_1.png","-632854844_21_2.png","-632854844_21_3.png","-632854844_21_4.png","-632854844_21_5.png","-632854844_21_6.png","-632854844_21_7.png","-632854844_21_8.png","-632854844_22_1.png","-632854844_22_2.png","-632854844_22_3.png","-632854844_22_4.png","-632854844_22_5.png","-632854844_22_6.png","-632854844_22_7.png"],"tables":["-632854844_1_1.json","-632854844_2_1.json","-632854844_3_1.json","-632854844_4_1.json","-632854844_5_1.json","-632854844_6_1.json","-632854844_7_1.json","-632854844_8_1.json","-632854844_9_1.json","-632854844_10_1.json","-632854844_11_1.json","-632854844_12_1.json","-632854844_13_1.json","-632854844_14_1.json","-632854844_15_1.json","-632854844_16_1.json","-632854844_17_1.json","-632854844_18_1.json","-632854844_19_1.json","-632854844_20_1.json","-632854844_21_1.json","-632854844_22_1.json"]}

data/part_2/0373160491.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"81d599bf977e93e84df0f9e0a2267e12","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/7ee64aa9-7190-4cdc-94a1-8fd9f91261a8/retrieve","description":"Much of the early attention to the Russia-Ukraine conflict’s food security impacts has been concentrated on countries highly dependent on wheat imports from the Black Sea region. Given the important role that wheat products play in the diets of people in Egypt, Sudan, Yemen, Lebanon, and other countries, the interruption in Black Sea wheat trade and high prices have raised serious concerns about rising levels of food insecurity, poverty, and instability around the world. But many countries are affected by price increases across a range of commodities (some predating the war), including in fertilizers, edible oils, and maize, as well as oil, natural gas, and other energy products. How are these sharp international price increases impacting countries and people, and how can countries respond? Our recent modeling study focusing on Kenya suggests higher prices, particularly for fertilizer, will reduce GDP growth and increase poverty rates in the country, putting an estimated 1.4 million additional people below the poverty line.","id":"-1209137856"},"keywords":[],"sieverID":"43225c1b-3fcf-4822-96ba-113a61c419d5","pagecount":"5","content":"Much of the early attention to the Russia-Ukraine conflict's food security impacts has been concentrated on countries highly dependent on wheat imports from the Black Sea region. Given the important role that wheat products play in the diets of people in Egypt, Sudan, Yemen, Lebanon, and other countries, the interruption in Black Sea wheat trade and high prices have raised serious concerns about rising levels of food insecurity, poverty, and instability around the world.But many countries are affected by price increases across a range of commodities (some predating the war), including in fertilizers, edible oils, and maize, as well as oil, natural gas, and other energy products. How are these sharp international price increases impacting countries and people, and how can countries respond? Our recent modeling study focusing on Kenya suggests higher prices, particularly for fertilizer, will reduce GDP growth and increase poverty rates in the country, putting an estimated 1.4 million additional people below the poverty line.Inflation in Kenya hit 7.1% as of May 2022, mainly driven by a combination of high world commodity prices and a depreciating exchange rate (Central Bank of Kenya). The majority of maize, rice, and wheat in Kenya's domestic market -including maize and wheat flours -are produced locally, while about 45% of edible oils are imported. Fuel and other petroleum products are almost all imported. Depending on the year, about 30-50% of wheat grains processed in Kenya for final consumption or used for food processing industries come from Russia and Ukraine, and imported edible oils come mostly from Indonesia and Malaysia ( UN COMTRADE database, 2022). While almost all chemical fertilizers are imported in Kenya, a relatively small portion come from Russia -about 9.3% of the total value of imported fertilizers in 2021 (UN COMTRADE). Kenya's petroleum product imports come mainly from the Middle East.Almost all the commodities affected by recent global crises are consumed directly by households and used for other sectors' production as intermediate inputs. Using IFPRI's economywide Rural Investment and Policy Analysis (RIAPA) model for Kenya, we assess the likely impacts of the ongoing global crises on the Kenyan economy and people by taking into consideration cross-sector linkages and impacts on various supply chains.The study indicates that Kenya's gross domestic product (GDP) could be 0.8% lower in 2022 compared to a situation without the global commodity price crises. More than 50% of GDP losses are from agriculture, driven primarily by higher fertilizer prices and the associated negative productivity effect from farmers' reduced use (Figure 2). (The study assumes that a 100% increase in real fertilizer prices leads to a 15% decline in fertilizer use and that farmers who do not use chemical fertilizers are about 20% less productive than farmers who use chemical fertilizers.)Household consumption is also negatively affected by rising world prices -particularly higher fuel prices that raise market prices for many consumer goods and services. The decline in consumption is larger for households toward the lower end of the income distribution, leading to increased inequality and greater poverty.According to the most recent Kenya Integrated Household Survey Report (KNBS 2018) The increase in world prices raises the national poverty headcount rate in Kenya by 2.5 percentage points, while the rural poverty rate rises more, by 3.1 percentage points, compared with just 1.4 percentage points for the urban poverty rate (Panel A in Figure 3) -this is equivalent to an additional 1.4 million people falling below the poverty line (Panel B in Figure 3). Most of the increase in poverty is caused by the fertilizer shock, and the absolute increase in the poor population is larger in rural areas, where more poor people are concentrated.To cushion consumers and producers from the impacts rising global prices, Kenya's government has implemented several fiscal measures, including fertilizer and fuel subsidies. As of April 2022, the government had budgeted a total of KSh.5.7 billion (US$49.6 million) to subsidize 114,000 metric tons of fertilizer for crops. As the planting season was already ongoing, anecdotal evidence suggests that farmers largely made their planting and fertilizer adoption decisions before the subsidy mechanism was fully implemented.The government introduced price controls in the energy sector in 2011, and in 2020 established a stabilization fund that became operational in April 2021. By April 2022, the government had already paid KSh.49.2 billion ($425.8 million) to oil marketers to keep the fuel prices in the country at a reasonable margin. With persistently high oil prices, the current fund is expected to fall short in maintaining fuel prices at pre-Ukraine war levels. In the short run, fertilizer and fuel subsidies can partially blunt the effects of global shocks on Kenya's producers and consumers. Given that global fertilizer prices will likely remain high creating fiscal constraints, more targeted fertilizer subsidy schemes may be necessary before the next planting season. Maintaining the timely availability of fertilizers is also important. However, going forward Kenya will need to expand its domestic fertilizer production to reduce its dependence on imports. The government has pursued this strategy by allowing the private sector to import fertilizers and blend in the country for specific crops. The government should also consider encouraging private sector investments to establish fully-fledged fertilizer production in Kenya and/or the broader East African region.Given the impacts of the current crisis on the poor, social safety nets should be temporarily expanded, including existing cash transfer programs such as Orphans and Vulnerable Children (OVC) cash transfers, Persons with Severe Disability (PWSD) cash transfers, the Hunger Safety Net Program, and nutrition-sensitive cash transfers. The efficiency of these programs can also be improved, including meeting challenging problems of late disbursement of funds and delays in targeting and enrollment of beneficiaries, among other issues.The current global crisis underlines the longer-term urgency of moving toward a green economy transformation. Currently, 75% of Kenya's electricity is generated from hydro, wind, solar, and geothermal energy sources, and there are significant opportunities to expand production and foster green growth in the industry and services sectors.Expanding agricultural production and productivity, meanwhile, will require an increase of public and private investments in agriculture, improvements in the business climate, and more effective implementation of agricultural and related investment projects under the government's Vision 2030 and Big 4 Agenda.Our study indicates that rising prices of key commodities sparked by the Russia-Ukraine war are stressing Kenya's economy, with the poor bearing the brunt. A combination of short-and longer-term actions by the government, business, and international partners can help to cushion those impacts, and strengthen the country's economy against future shocks.Since this analysis was published in June 2022, Kenya has elected a new government under the leadership of President William Ruto. At the core of the new government's development agenda is the Bottom-Up Economic Plan, with the main objectives of creating jobs, eradicating hunger, and lowering the cost of living for its citizens. Several of the Plan's core elements such as the focus on agriculture and the poor, revamping of fertilizer subsidies and distribution, and the longer-term goal of moving toward a green economy transformation were recommended in our original analysis. Responding to a request from the Kenyan National Treasury and Economic Planning and in collaboration with the Kenya Institute for Public Policy Research and Analysis (KIPPRA), IFPRI staff have expanded the Kenya RIAPA model used for the Ukraine-Russia impact analysis to also analyze the expected impacts of implementing the Plan from 2023 to 2027.Analysis of the impacts of the Russia-Ukraine war and the Bottom-Up Economic Plan were conducted by rapid response teams composed of Kenyan experts and institutions and IFPRI researchers under the CGIAR Initiatives on National Policies and Strategies and Foresight and Metrics. This collaborative approach to research has greatly enhanced the relevance and uptake of the findings and appreciated by Kenyan policymakers.","tokenCount":"1281","images":[],"tables":["-1209137856_1_1.json","-1209137856_2_1.json","-1209137856_3_1.json","-1209137856_4_1.json","-1209137856_5_1.json"]}

data/part_2/0386055458.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"427e2d6fd965a3c9aacfcbff08127d16","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/247d3ae3-0b1c-4c8f-8c05-d018f0d4d2fa/retrieve","description":"This paper presents a review of three existing models of capacity building to provide an understanding of strategies and approaches that have been successful in strengthening human and institutional capacity for agricultural research in Africa. The paper documents lessons in capacity strengthening from these models that can be scaled up and out to other contexts. While the three programs reviewed emphasize building capacities by developing local institutions, a major challenge is the programs' sustainability because they continue to depend on external sources of funding. In order to take advantage of the emerging global interest in capacity strengthening, it is argued that national agricultural research systems need to prepare by identifying their strategic capacity and institutional strengthening needs, as well as providing overall leadership to enable the successful use of the capacities developed. While further innovations are needed to strengthen capacities cost-effectively, the three programs reviewed highlight the role of national leadership in absorbing the capacity created, thereby emphasizing the importance of an enabling environment that can maximize the benefits of the capacities strengthened. Unless the skills developed through these programs are complemented by effective organizational capacity, providing the motivation and incentives needed to put the newly created skills to effective use, these capacities may well be eroded.","id":"-1597186603"},"keywords":[],"sieverID":"f73ecc3b-c3f6-43d8-a7e1-fb689c5c8d13","pagecount":"26","content":"This paper presents a review of three existing models of capacity building to provide an understanding of strategies and approaches that have been successful in strengthening human and institutional capacity for agricultural research in Africa. The paper documents lessons in capacity strengthening from these models that can be scaled up and out to other contexts. While the three programs reviewed emphasize building capacities by developing local institutions, a major challenge is the programs' sustainability because they continue to depend on external sources of funding. In order to take advantage of the emerging global interest in capacity strengthening, it is argued that national agricultural research systems need to prepare by identifying their strategic capacity and institutional strengthening needs, as well as providing overall leadership to enable the successful use of the capacities developed. While further innovations are needed to strengthen capacities cost-effectively, the three programs reviewed highlight the role of national leadership in absorbing the capacity created, thereby emphasizing the importance of an enabling environment that can maximize the benefits of the capacities strengthened. Unless the skills developed through these programs are complemented by effective organizational capacity, providing the motivation and incentives needed to put the newly created skills to effective use, these capacities may well be eroded.The urgent need to develop capacity for agricultural research, technology dissemination, and adoption in Sub-Saharan Africa (SSA)-as outlined in Pillar IV of the Comprehensive African Agricultural Development Program (CAADP) under the New Partnership for Africa's Development (NEPAD)-has become increasingly clear. Recent progress made in mobilizing African leadership under CAADP will require greater focus on agricultural research capacity as a means of successfully reducing poverty and eliminating hunger (FARA 2006;CAADP 2008;NEPAD-CAADP 2010). Along with the development of national investment plans, the capacity to implement such plans needs to be strengthened and become an integral part of the agricultural research and development (R&D) processes if the goal of increasing agricultural productivity by 6 percent per year is to be fully realized. A major challenge is that existing capacity strengthening approaches focus heavily on the development of individual skills rather than the institutional and organizational capacity challenges facing African countries.It is well known that transforming traditional agricultural sectors requires adequate human organizational capacity (Mosher 1966;Timmer 2011); however, a systemwide effort to develop a framework for designing, implementing, monitoring, and evaluating capacity strengthening programs has been lacking. Capacity development efforts continue to be confronted with numerous challenges, and the approach to strengthening research capacity has been fragmented. For example, several donors fund capacity strengthening programs within institutions with little coordination, past emphasis on strengthening individual skills of researchers has not resulted in adequate organizational capacity, developing individual research capacity does not guarantee successful use of such capacity at the organizational and institutional levels, and capacity development efforts often do not emanate from agricultural development strategies and national development goals.The CAADP process provides an opportunity to develop a systematic approach to strengthening agricultural research in line with the national strategies. Existing approaches face the classic development dilemma of whether to address immediate problems, such as low productivity, degraded natural resources, lack of market and trade opportunities, and ineffective institutions for accelerated agricultural development, or to focus efforts long term, in this case on developing the capacity to design and implement future programs (Fukuyama 2004). In the context of agricultural research, both problem-solving and long-term capacity development need to be addressed simultaneously so that newly initiated research programs can be sustained. Yet research institutions, like other development agencies, are pressured to focus on short-term objectives to show quick impact. Thankfully, policymakers and development partners are increasingly recognizing the need to develop long-term capacity.This paper presents a review of three existing models of capacity building as a basis for addressing the following questions: What strategies can help overcome the conflict between short-and long-term capacity development needs? What approaches have been successful in developing individual research capacity, while building sustainable institutional capacity? What can be learned from existing models of capacity strengthening efforts under the CAADP process that can be scaled up and out to other contexts?The CAADP process is driven by the implementation of four strategic pillars (CAADP 2008):1. extending the area under sustainable land management and reliable water control systems (Pillar I); 2. improving rural infrastructure and trade-related capacities to improve market access (Pillar II); 3. increasing food supply and reducing hunger (Pillar III); and 4. agricultural research, technology dissemination, and adoption (Pillar IV).The specific objectives of Pillar IV are (NEPAD-CAADP 2010):1. to develop technologies, policies, and institutional innovations that provide solutions to poverty and resource degradation in Africa; 2. to test the adaptability of these options in a participatory and iterative fashion, from the farm to the regional scale; 3. to develop appropriate mechanisms for broad dissemination and adoption of technologies and for the implementation of sustainable and supportive policies and institutional options; and 4. to empower Africa's resource-poor farmers in sustainably managing their natural resources and systems.Along with these four pillars, two cross-cutting areas support the four pillars: the development of (1) academic, professional, and vocational capacity, and (2) strategy and knowledge management capacity. In order to implement the CAADP process, it is important that adequate capacity be developed across all the four pillars in order to jointly address national capacity strengthening needs (CAADP 2008). Under the CAADP process, continuous engagement with stakeholders and the public is seen as a way to address the challenges facing the final beneficiaries. The programming and policymaking process must be evidence-based, which requires research, analysis, and monitoring and evaluation capacity. Enhanced institutional arrangements and enabling environments for program development and implementation are needed to ensure success. The programs implemented under CAADP must be constantly assessed for learning, adaptation, and further planning through revision.At the country level, CAADP country teams engage in developing strategies for agricultural development, but these teams need to be supported with policy analysis and research to generate appropriate evidence for the strategies that are identified. Human and organizational capacity requires strengthening to be able to develop the strategy and implement policies and programs once investment opportunities are identified. In the process of developing in-country capacity for the CAADP design and implementation, it is important to recognize the various actors that constitute CAADP's resource group, including NEPAD, the African Union, the regional economic communities, and national policymakers.Nurturing relationships between country teams and various groups of experts is an important aspect of securing long-term success in implementing CAADP strategies and ensuring that national representatives are aware of CAADP's institutional implementation architecture so they can protect their interests. In the context of building agricultural research and innovation capacity, the national agricultural research institutions (NARIs) need to align their priorities with strategies developed through CAADP's country processes. These processes have significantly improved the way in which priorities are identified and aligned with local needs, but strategic approaches are needed to improve efficiency in terms of maximizing the use of available resources.Uganda's agricultural sector development strategy and investment, 2010/11-2014/15, for example, was developed as part of the CAADP process. It identifies \"enhancing production and productivity\" as a strategy and investment program, and \"agricultural research and technology development\" as a subprogram that calls for strengthening national agricultural research systems (NARSs) to generate new technologies along value chains and ensure continuity in research capacity in the pursuit of cutting-edge science (MAAIF 2010). This objective is further aligned with the Ten-Year Strategic Plan (2008-18) of Uganda's National Agricultural Research Organization (NARO), which identifies maintaining a critical mass of scientists as a key factor in improving the provision of the country's agricultural research. Additional strategies identified as requiring capacity strengthening include enhancing the priority-setting process for research, developing a center of excellence for cassava, expanding the competitive grant system for funding research, and designing mechanisms for assessing the impact of NARO's research programs. Budget allocation for this subtheme includes 16.7 billion Uganda shillings (UGX) for training and workshops, which represents 4.8 percent of this subprogram's total funding of UGX 3.44 billion for the period 2013-18. The challenge, however, is to strategically reallocate the minimal resources available to achieve capacity strengthening objectives. While additional resources could be raised through special projects funded by bilateral donor agencies (a strategy that also has inherent problems), meeting the critical objectives with limited government funding is a challenge facing many SSA countries. Assuring the quality of the capacity developed is another critical factor, as is the need to establish a system of accountability, inclusiveness, and collective responsibility for the regionwide development of competencies in agricultural research and innovation systems. Simply coordinating such processes will require additional capacity and leadership at the national level.The CAADP process emphasizes that capacity development must meet the demands of stakeholders in terms of identifying current gaps, efficient approaches, and ways of transforming existing and new capacity to enable the effective implementation of programs. Additional key elements for success include contextualizing the capacity development process, promoting mutual learning and knowledge sharing, adopting best global standards, and prioritizing institutional and organizational development. This provides initial benchmarks against which current capacity development approaches can be analyzed for their contribution to the CAADP process.The Asian experience in strengthening national research capacity may be relevant for Africa, as discussions on improving African agricultural productivity often indicate (Otsuka and Kijima 2010;Swaminathan 2003); however, several factors contrast Asia's Green Revolution experience with the capacity challenges facing Africa today. For example, whereas only two major crops (rice and wheat) contributed to the Green Revolution in Asia, African agriculture is characterized by multiple agroecological systems growing several crops within a single zone (Eicher 2006;Timmer and Akkus 2008). Nevertheless, the development of capacity for agricultural science and technology during the initial period of Green Revolution in India could be illustrative. India, for example, had only 7,000 graduates and 1,000 postgraduates in 1950, but investment in higher education in the 1960s contributed to increasing these numbers threefold. By 1971, graduates and postgraduates numbered 13,500 and 4,200, respectively (Long 1979). This significantly larger pool of scientists provided the basis for sustaining the Green Revolution into the 1970s and 1980s. While India had the minimum needed capacity to conduct adaptive research in the early 1960s, much of its capacity was developed through the process of implementing Green Revolution technologies, institutions, and policies (Lele and Goldsmith 1989). Such capacity for adaptive research is currently available in Africa, albeit at minimum levels and thinly spread across the region, in part based on Africa very large number of small countries (54 currently) combined with the large number of crops that provide the continent with staple foods and thereby preclude scale economies.The lessons for capacity strengthening from the Green Revolution can be summarized as follows: • Capacity development for agricultural research in Africa must go hand in hand with program implementation.• Capacity development at the level of professional individuals needs to be complemented by institutional and organizational capacity. • Capacity investment should be an integral part of research investments in order to maximize and sustain research to obtain the desired impacts. • In addition to developing individual capacities, institutions that generate human capacity for agricultural research, technology development and dissemination, and strategic policy analysis must be strengthened (Eicher 1994;Timmer 2011;Tamboli and Nene 2011) in order to generate sustainable capacity for agricultural research. Table 1 shows the trend in capacity levels for agricultural research in selected countries in SSA over the past 40 years. While there has been considerable progress in strengthening research capacity in several of the region's countries, the capacity developed is not adequate to meet the goal of increasing agricultural productivity by 6 percent, as agreed under CAADP (Beintema and Stads 2011). In the past decade, all the East African countries have improved their research capacity; among Southern African countries, Malawi and South Africa have reduced capacity; and in West Africa, capacity in several countries has fallen, while improvements were recorded in Gabon, Ghana, Mali, and Nigeria. In average terms, however, full-time equivalent (FTE) researchers per million employed in the labor force increased only slightly, from 63 FTEs in 1981-85 to 68 FTEs in 2008. The yearly rates of growth in FTE researchers have also declined from 4.5 percent in the 1970s to 2.8 percent in the 2000s (Beintema and Stads 2011). These trends reflect the challenge of sustainably increasing agricultural research capacity in African countries.Given these trends, some core principles should underlie ongoing efforts to generate and retain agricultural research capacity for the benefit of African agriculture:• build capacity as cost-effectively as possible in areas of research that contribute to issues and crops of relevance to African countries within the context of national priorities; • minimize the time researchers/students spend away from home by maximizing opportunities for learning and retaining the capacity developed within recipients' home countries; and • design curricula to include state-of-the-art methods and cutting-edge knowledge. The next sections review the three chosen models of capacity development in turn.The Challenge Plant-breeding skills can be considered among the most important gap in capacity to meet the CAADP goal of increasing agricultural productivity in Africa. Emerging challenges, such as population growth, high food demand, rising food prices, climate change, and the energy crisis, all require increased productivity of African crops. Increasing crop yields, producing higher quality foods, fighting biotic and abiotic factors, and maintaining the productivity gains all require sound plant-breeding capacity within NARIs.The program began in 2002 at the African Center for Crop Improvement (ACCI) of the University of Kwazulu-Natal in South Africa. 2 The program works with NARIs in East and Southern Africa to identify young researchers to work with experienced plant breeders as part of their PhD training, supported by the Alliance for Green Revolution in Africa (AGRA). ACCI currently trains plant breeders from 10 countries in East and Southern Africa: Kenya, Ethiopia, Malawi, Mozambique, Rwanda, South Africa, Tanzania, Uganda, and Zambia. ACCI has predicted that 440 breeders are needed in crop improvement research focusing on 11 crops; yet, as of 2011, ACCI had only trained 84. Although other breeding programs will add to this number-such as the PhD program run by Regional Universities Forum for Capacity Building in Agriculture (RUFORUM) and MSc-level programs in Mozambique, Tanzania, and Uganda-the demand for plant breeders in the subregion far exceeds the current supply (Laing 2011).The program follows a \"sandwich\" approach. Students undergo rigorous theoretical training and development of practical plant-breeding techniques for the first two years of the program at Kwazulu-Natal University. Subsequently they select a research problem related to their home countries and conduct plant-breeding research in collaboration with their local NARIs, at which many of the students are also employed. The research is supported and supervised by professors of the University of Kwazulu-Natal, to which the final PhD thesis is submitted for assessment. The program also brings in professors from Cornell University to teach plant-breeding courses, which has been a useful way of exposing participants to additional external expertise, to the most current research methods and knowledge of plant breeding, and to mentorships that go beyond their training. Importantly, because participants work to improve crops of national priority, the outcomes of their research are highly relevant.The program has been instrumental in developing several crop varieties that have been shared with national agricultural research programs. It strengthens researchers' capacity by imparting specific plant-breeding skills and techniques and by proving consistent mentoring. In most cases the thesis research aligns with national research priorities, as well as promoting continuity and new employment for newly trained researchers in their own countries. Increasingly, the capacity developed by the program is being absorbed and retained in the subregion. On average, eight students have been trained per year, with a 75-percent graduation rate. Most of the students' research has been published in international journals, and the students have developed, registered, and released new varieties of crops, including hybrids. The program has produced high-quality research outputs in the past six years, including (as of late-2011) 66 published research papers and books; 33 research papers accepted for review; and 37 research papers in preparation for submission to journals (Laing 2011).The program's funding for the 2002-18 period totals US$18 million. The yearly cost of educating a student is US$10,483, which compares well with some of the programs in North America, which cost about twice as much per year. ACCI probably has the largest group of plant breeders in Africa, including nine full-time breeders to conduct teaching and research. The institute continues to take in new students, in addition to the existing 44 students who will graduate in the next four to five years. ACCI intends to make some modifications in the next phase of its implementation, including reducing the curriculum to four years instead of the current five; the goal is to admit students with MSc degrees in plant breeding, thereby eliminating a year of more fundamental coursework needed by nonplant-breeders. The program is also planning to hire permanent, university-funded staff as a means of sustaining impact, while hopefully reducing costs.The Challenge Existing nationally based agricultural economics education in Africa generally lacks the ability to provide comprehensive postgraduate programs, but higher education agencies do have small numbers of well-trained faculty staff who could contribute specialized expertise to high-quality programs structured at the regional or subregional levels. Lack of ability to come together at the regional level has been a major constraint to utilizing existing teaching capacity in regional development initiatives. Additionally, strong MSc-level graduates are needed to fill teaching positions and conduct research within the higher education sector in East, Central and Southern Africa.The Collaborative Masters Program in Agricultural and Applied Economics (CMAAE) was established in 2000. It is a network of 16 faculties of agricultural economics and agribusiness in 12 countries of Eastern, Central, and Southern Africa. CMAAE's major goal is to bring together specialized capacities from a variety of geographically dispersed university departments and faculties to contribute to building future capacity for agricultural economics and policy research. The program's three specific objectives are (1) to produce high-quality MSc agricultural and applied economics graduates; (2) to upgrade the teaching and research capacity of departments currently in the program, and initiate strategic plans to scale out the program to other subregions of SSA; and (3) to strengthen a continent-wide research network to promote agricultural development.The key approach of the program is to tap the best brains in Africa as teachers at the regional level. Students receive their degrees from their participating home-based universities, but they have opportunities to learn as a collective group from regional and international experts in a variety of fields of agricultural economics and policy research. Competitive research grants for thesis research are provided through the network, and students are competitively selected to participate in the program. The research conducted by the students is evaluated by an external team in order to maintain the quality of the research. Given that CMAAE is a network of teachers and teaching programs in agricultural economics, lessons from each of the universities can be shared and applied via a standardized curriculum that focuses on challenges that are common to the region. Students select their thesis research to address the economic policy challenges of their home countries and the countries in which they are studying. While the capacities generated are not necessarily tied to local organizations, the network creates general capacities for agricultural economics, farm management, program management, project monitoring and evaluation, and policy analysis in various fields related to food, agriculture, and natural resource management.Preparatory stages of the program began in 2001 and involved a thorough analysis of the demand for postgraduate programs in agricultural economics (Obwana and Norman 2001). The planning process was guided by a steering committee, which developed a proposal in collaboration with the International Food Policy Research Institute (IFPRI). Initial funding was provided the Rockefeller Foundation and enabled the establishment of a planning secretariat and a network of collaborating institutions. Between 2002 and 2004, several consultative meetings were held to develop strategic plans, client consultations, governance structures, and operating procedures. The planning process identified a number of key features for the program, including that it (1) be demand driven, (2) provide opportunities for mid-career professionals, (3) pool and share existing human and physical resources, (4) be expanded into innovative applied programs, (5) prepare students for further education, (6) focus on research-based education, (7) complement existing programs, and (8) promote collaboration among the faculties in the region.The CMAAE program involves 20-24 months of study over a period of five semesters. Participating universities are responsible for teaching eight core courses to the enrolled postgraduate students. Courses include microeconomics, macroeconomics, production economics, econometrics, mathematics for economics, statistics, research methodology, and a course on agricultural economics issues. Universities in the network are accredited by the program based on specific criteria, including, for example, that they have at least five PhDqualified faculty staff to teach core courses. As of 2011, the program had accredited seven universities: Lilongwe University of Agricultural Sciences (Malawi), the University of Zimbabwe, the University of Pretoria, the University of Nairobi, Makerere University in Uganda, Sokoine University in Tanzania, and Egerton University in Kenya.The Achievements CMAAE admitted 394 students in its first six years, from2005 to 2010 (Table 2).The first four cohorts of 155 students have graduated, with a graduation rate of 67 percent. As of 2011, 94 students were enrolled in the program's sixth cohort, of which 43 were female. As of October 2011, it was estimated that the total enrollment would be about 105 students per year. The cost per student for the two-year program is about US$24,000, which includes tuition, boarding and lodging, and thesis-related costs. This makes the program cost-effective compared with North American and European programs, but the cost is relatively high compared with the national programs in Africa, partly due to the expenditures involved in educating the student through the shared facility.Several supporting activities help the CMAAE program to engage with its network partners:• The program strengthens the capacity of faculty members in the network's nonaccredited departments by providing competitive scholarships for PhD studies and by supporting six PhD students through a sandwich program in collaboration with Cornell University. The program also supports its alumni in being able to present their research findings in international forums. • CMAAE supports faculty research through competitive research grants of about US$15,000, eight of which were awarded in 2011. • Thesis dissemination workshops bring program graduates and the stakeholders of their research together to engage in discussion; four such workshops were held in 2011. • The program became a part of the African Economic Research Consortium (AERC)-a network of African universities and research organizations conducting economic research and analysis-in April 2010 and therefore is able to share its publications and information through AERC's website, thereby facilitating ongoing engagement with stakeholders. • In 2010, the program began an exchange program among its network partners, whereby faculty members are encouraged to spend up to one year on sabbatical at another faculty. • Institutional support is provided through funding for the purchase/supplementation/replacement of equipment, resources, computer laboratories, and reference materials needed for the program. Several factors contributed to the success of the program. First, it is demand-driven: the program began with an analysis of the needs for agricultural economics capacity in the region, which facilitated the establishment of the curriculum and ensured that it met the skill sets needed by the employers in the region. Second, the quality and the relevance of the programs is high, reflected in the competitive selection of applicants; the self-sponsorship of applicants; and the sponsorship of applicants by governments, the private sector, and nongovernmental organization (NGOs). Third, the program has secured continuous buy-in from donors, and donor bases remain diversified. Finally, the program is cost-effective compared with those offered by the European and North American Universities.Although the program has been successful in meeting its objectives and continues to be well-implemented, it faces several challenges.First, the program has a completion rate of less than 70 percent, which is less than was anticipated by the program's designers. While some level of attrition is expected in any higher education program, the current level of noncompletion largely stems from the poor support of candidates by their home institutions in finishing their theses. Some supervisors delay the correction of the thesis. Some students also take a break after the course work and can lose the necessary momentum for completion. In efforts to address this challenge, the program has instituted an award of US$500 for graduates who complete their thesis on time.Second, the comparatively high cost of the CMAAE program deters self-sponsored students. Despite being competitive compared with programs at European and North American universities, the program's cost is still high by African standards. The yearly cost of an MSc degree in applied economics and management at Cornell University, for example, is roughly US$20,000 for tuition alone, whereas the yearly cost of at the University of Pretoria is about US$9,000 for both tuition and accommodation. Third, the program continues to depend on external donors, but they would like the program to become self-sustaining in the next few years. Fourth, the participating departments have different program cycles depending on their country contexts, so harmonizing admission calendars across participating departments has been difficult. Fifth, the program offers eight core courses and 20 electives; however, the demand for some electives is so low that they are not regularly offered, making critical mass a necessity if these electives are to become viable. Finally, sustaining the shared facility where students come together to take specialized courses will be costly long term. Nevertheless, the shared facility is the cornerstone of the program and accounts about 50 percent of its cost. In the future this share could be reduced as the program moves toward the use of e-learning methods.CMAAE is currently funded by the African Capacity Building Foundation and the Bill and Melinda Gates Foundation. In addition, the German Academic Exchange Service (DAAD), the Economic Research Service (ERS) of the United States Department of Agriculture (USDA), and the Centre for Development Research (ZEF) of the University of Bonn provide targeted support to fund scholarships for students.The capacity strengthening programs implemented in Africa have been criticized for being supply-driven without adequate consultation with the local institutions that implement the program. In order to shift this situation, the Forum for Agricultural Research in Africa (FARA) initiated a new program: Strengthening the Capacity of Agricultural Research and Development in Africa (SCARDA), which was implemented in March 2008 (FARA 2007).The shift was also needed from a piecemeal approach to solving capacity problems, to strategically strengthening entire institutions. In addition to learning-by-doing approaches, SCARDA emphasizes reflecting on lessons learned and improving processes. Institutional analysis is conducted through a participatory process to identify gaps and build ownership among participating institutions. Once gaps have been identified, tailor-made capacity strengthening packages are developed as part of the action plans for organizational change management that ensure that capacity strengthening actions are embedded in change management processes at the organizational level. Some elements of the capacity strengthening package include combinations of shortterm professional training in specific technical areas, research management, monitoring and evaluation, agricultural information and communication management, and long-term postgraduate training.The SCARDA approach to capacity strengthening for agricultural research for development emerged from several stakeholder consultations. It features• a holistic approach to strengthening capacity, focusing mainly on a limited number of willing institutions from a cross-section of countries; • building on existing strengths within subregions in the identification and contracting of capacity strengthening service providers; • a demand-driven and participatory approach, identifying challenges, fostering an understanding and commitment to addressing these challenges, and agreeing to set priorities through institutional analysis; • using less \"traditional\" methods to address capacity strengthening priorities (for example, mentoring, participatory institutional analysis, tailor-made courses, and participant action plans); • integrating gender and other cross-cutting social inclusion issues into the planning and implementation of capacity strengthening activities; • learning from the lessons of past capacity strengthening initiatives; and  complementing existing and planned initiatives of the participating member organizations.The SCARDA approach is a facilitated process of strengthening whole organizations to bring about institutional change through the development and implementation of targeted and tailor-made capacity strengthening packages. The program's validation phase focused on 12 NARIs and universities in SSA, referred to as focal institutions. The SCARDA approach also allows for mentoring support at individual and organizational levels to address local and national priority gaps for capacity strengthening. Platforms are established for lesson learning at national, subregional, and regional levels. These platforms facilitate lessons on organizational and management changes and on regional spillovers in implementing innovative programs and policies for capacity strengthening.In this way, SCARDA endeavors to improve the functioning of agricultural research systems and specific demand-driven competencies in participating countries. NARIs and universities not only have better knowledge, but are also connected with national agricultural innovation systems. In addition, SCARDA aims to achieve improved capacities for NARI scientists, researchers, and extension agents who transfer the technology to the farmers working with knowledge intermediaries. This paper, however, focuses on the nine national research organizations participating in SCARDA.The SCARDA process of institutional analysis, capacity strengthening implementation, and reflection and revision is illustrated in Figure 1.The process began with scoping studies to identify focal institutions, which were then analyzed as a means of developing an integrated capacity strengthening plan. Implementation of the plan involved capacity strengthening through short-term and MSc-level courses. Based on demand, determined through needs assessment studies, the following training modules were developed and offered:1. MSc-level training in areas where the focal institutions were lacking capacity; 2. research management training courses; and 3. short professional courses to upgrade skills (such as proposal writing, integrated pest management, and participatory farmer research) to improve the abilities of researchers and technicians. Assessments of the evolving employment opportunities for agricultural graduates were conducted in all three subregions. In the CORAF/WECARD subregions, demand studies focused on Congo, Ghana, Gambia, and Mali; in the ASARECA subregions, studies were conducted in Kenya, Ethiopia, Malawi, Mozambique, and Tanzania; and in the SADC subregion, studies were conducted in Botswana, Lesotho, and Swaziland. The studies showed that many of the universities in Eastern and Southern Africa have a remarkably stable and well-qualified staff, and that the universities can provide the leadership and guidance necessary to garner the resources essential to changing the agricultural sectors. The studies confirmed that a major weakness was the poor image of agriculture among many potential students. 4. The SCARDA approach to capacity strengthening is validated. The SCARDA approach has been endorsed by all stakeholders through several workshops and meetings, and feedback indicates that the program has helped to make significant changes in the functioning and the interactions among the research institutions. Currently, SCARDA provides the basis for the CORAF/WECARD strategy for capacity strengthening and knowledge management, and ASARECA is in the preparatory phase of out-scaling the SCARDA approach in six countries.The feedback from the implementing partners indicated that the SCARDA approach has been beneficial in stimulating reform efforts at the national and organizational levels. It is highly relevant for addressing crucial bottlenecks in their organizations and offers the right mix of capacity strengthening instruments. Furthermore, the SCARDA approach ensures actual use of newly acquired knowledge and skills by the trainees through the mentoring scheme. In addition, SCARDA has raised awareness of research management issues and challenges, and has upgraded the skills of the participants to enable them to tackle these challenges. Finally, a structured and better informed process has promoted the engagement of research stakeholders in addressing the agricultural research management issues of the focal institutions.Implementation of SCARDA has faced several challenges in addition to the complexities of working across so many countries and with such a wide range of stakeholders. The main challenges include (1) an overly ambitious project design, especially with regard to expectations of what can be achieved within the project's timeframe, (2) the extensive time required to formalize the working relationships between FARA, the SROs and the lead service providers, (3) full adoption of the subsidiarity principle, (4) the unpredictable nature of funding, and (5) inadequate monitoring and evaluation.Several lessons emerge from the contemporary approaches described above. In the context of the core principles of capacity strengthening, all three programs fare well with varying degrees of success. In the process of building long-term capacity, it is important to account for capacity gaps and correct them through appropriate methods of management, maintenance, and use of existing capacities, accompanied by proper incentives and institutional changes. While creating capacity for agricultural research is inherently a long-term process because it involves research-oriented higher education and training. In the short-term, created capacity can be lost if it is not effectively used and offered appropriate incentives. To ensure effective use of capacity, strategic capacity needs to be identified, and competition needs to be nurtured within each country. The SCARDA approach to building institutional capacity among the NARIs is a promising approach in this context requiring further analysis for support.In building capacity for agricultural research and innovation in Africa current capacity challenges need to be viewed within the broader agricultural development context. For example, the CAADP process requires that countries aim to allocate 10 percent of their national budgets to agriculture and work toward achieving 6 percent agricultural growth per year. While countries are currently developing strategy papers under CAADP, several success factors can contribute to agricultural development. It is not always necessary that the technologies be developed by local researchers and innovators; numerous technologies can be borrowed and adapted locally. Capacity for such local adaptation, however, is currently lacking, and it is important that capacity is built quickly to translate existing knowledge and technologies currently used in similar agroecological systems. A major challenge for such adaptive research is the large amount of heterogeneity in local cropping systems. Despite this, large areas of maize are cultivated by smallholders in Eastern and Southern Africa, and the same can be said for rice cultivation in several West African countries. These areas could be considered homogenous zones for cultivating single commodities across larger areas. While this gives an opportunity to combine the efforts of agricultural research from several countries and to conduct regional research programs, the capacity to adapt the research outputs for national use needs to be identified and recognized. In addition, institutional support for the adoption of such technologies at the regional level requires capacity to analyze the supporting policies, such as seed, marketing, and trade policies, as well as institutional strengthening to implement such policies. The ACCI program is an example of how home grown capacity could enhance the skills immediately needed, while at the same time generating solutions to long-term agricultural development problems.The ACCI program shows how locally designed capacity development programs could reduce the problem of brain drain, which has long stifled African agricultural development. It also shows how local capacity development programs can address current research challenges on African crops. Program participants used local resources and methods. The program has also shown that time can be gained by participants remaining in contact with their home institutions and completing their theses research as part of their regular employment. In addition to agricultural research capacity, the adaptive approach to technology generation will require effective use of extension and advisory services, along with the capacity to import and distribute fertilizers and chemicals needed for agricultural production. Thus, in addition to research and innovation capacity, a range of complementary capacities will be needed at the country level.The CMAAE program builds these complementary capacities by amalgamating existing teaching capacities in the region, making it possible to provide high-quality teaching and research program to strengthen both individual and institutional capacities. The CMAAE program also shows that it is possible to mobilize local talent to build regional capacities, while gradually strengthening local faculties. This program did not rely on the center of excellence approach but instead involved 16 departments collaborating and offering their own expertise to achieve a regionwide goal. Because the CMAAE program focused on a single discipline, it was able to populate the profession in a short period of time.SCARDA's contribution toward building agricultural research management capacity through the subsidiarity principle has shown that, by imparting management skills and providing mentoring, both individual and institutional capacity can be strengthened. The SCARDA approach shows that on the job training and mentoring can help institutions to retain staff and increase their productivity. Nevertheless, all these programs heavily depend on external funding. As the African countries grow, these programs will need to become a regular part of existing research and educational institutions in participating countries, and this should be the objective of the programs moving forward.Table 3 shows various capacity development process indicators for the three programs. All three have performed well in contextualizing capacity development through their demand driven process of identifying capacity needs. They bring together participants from different countries to facilitate mutual learning and knowledge sharing. Will the backup of external technical assistance they have been able to adopt best global standards. And while all three programs aim to develop institutional capacities, all three are dependent on external funding, which makes the sustainability of the programs a major concern in the long run. The program was developed as a result of high need expressed by the African agricultural research communityThe program emerged from several levels of regional consultationsThe program is a result an expressed need from stakeholders of FARA through various consultations.The needs assessment revealed the critical need for breeders who could conduct adaptive research on various African crops.Needs assessment revealed the strengths and weaknesses of various faculties and how to bring existing capacities together to generate high-quality capacityThe needs assessment indicated the importance of subsidiarity principles and for strengthening institutional capacity in addition to individual capacity.Combined both the teaching and rigorous training by the University of Kwazulu-Natal and the practical training through mentors in the local institutions. This helps to address local problems and made the capacity develop highly relevant for the participants' country.Effectively used the African capacity although additional external support was sought to fill the gaps in teaching. The shared facility approach was efficient in increasing the quality of the program jointly.SCARDA's approach to capacity development focused on strengthening the whole organization giving emphasis on filling the gaps in the skills of the individuals. This is in line with the CAADP process needs for capacity development.ACCI participants applied their knowledge to solving problems in their own countries.While the theoretical training was common to all graduates in the applied areas of the program, the participants applied their skills to address socioeconomic problems in their countries Capacity developed under SCARDA was intended to directly influence the organization and management of the research organizations. This contextualized approach helped to focus individual attention on the participating organizations.The participants came from the research institutions in various countries who returned to their jobs to conduct their thesis research; thus the capacity developed was used effectively by host institutions.While the capacity developed is of high quality, due to high demand for the applied economics capacity, graduates have found placements that contribute to the agricultural development process in their countries.SCARDA strengthened existing capacity without adverse effects in terms of attrition.Sharing of knowledge on problems and solutions was facilitated by bringing students to Kwazulu-Natal to train in plantbreeding methods.The shared facility approach brought students from various participating countries together to achieve specific learning goals. This facilitated mutual learning among participants.SCARDA programs provided adequate opportunities for mutual learning.The program depends on donor funding to support the participation of the international students.The program continues to depend on external sources of funding, although several self-and government-sponsored students have recently been accepted into the program.Uncertainty in funding and the dependency on external resources for program implementation remain challenges for long-term planning.Source: Compiled by authors.While the three programs reviewed in this paper emphasize local production of capacities through building local institutions, they continue to depend on external sources of funding. The recent economic growth in several African countries could help such programs to become self-sustaining, but funding from the participating governments, particularly in the context of CAADP implementation, will be necessary. Further recent developments in terms of organizing global approaches to capacity development for agricultural research and innovation-such as modifications to the Consultative Group for International Agricultural Research (CGIAR) and the Capacities Montpelier Action Plan (CAPMAP 2010-20)-also call for the overhaul of capacity building approaches for international agricultural research for development. The Montpelier initiative emphasizes \"building the capacity to build capacity.\" This program is likely to support the development of regional centers of excellence in training for agricultural research and development in each of Africa's subregions. In order to take advantage of the emerging interest in capacity strengthening, it is important that NARSs prepare by identifying their strategic capacity and institutional strengthening needs, as well as providing overall leadership to enable the successful use of the capacities developed. While further innovations are needed to strengthen capacities cost-effectively, the three programs reviewed emphasize the role of national leadership. Agricultural research plans for capacity strengthening offer a forum for discussion to ensure that capacity strengthening remains high on the agricultural development agenda.","tokenCount":"6824","images":["-1597186603_1_1.png","-1597186603_1_2.png","-1597186603_1_3.png","-1597186603_17_1.png","-1597186603_26_1.png"],"tables":["-1597186603_1_1.json","-1597186603_2_1.json","-1597186603_3_1.json","-1597186603_4_1.json","-1597186603_5_1.json","-1597186603_6_1.json","-1597186603_7_1.json","-1597186603_8_1.json","-1597186603_9_1.json","-1597186603_10_1.json","-1597186603_11_1.json","-1597186603_12_1.json","-1597186603_13_1.json","-1597186603_14_1.json","-1597186603_15_1.json","-1597186603_16_1.json","-1597186603_17_1.json","-1597186603_18_1.json","-1597186603_19_1.json","-1597186603_20_1.json","-1597186603_21_1.json","-1597186603_22_1.json","-1597186603_23_1.json","-1597186603_24_1.json","-1597186603_25_1.json","-1597186603_26_1.json"]}

data/part_2/0399968600.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"8fbfef803a6da7d744f4d15a60dc9b32","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/6852e419-461f-41ce-b5ea-f9358734bf0b/retrieve","description":"This chapter challenges one of the main tenets of agricultural economics—that households behave as though they are single individuals, with production factors allocated efficiently between men and women. In many contexts this is a convenient and innocuous assumption. It can be quite restrictive, however, when investigating the causes and welfare consequences of gender differences in agriculture. In response to a growing number of econometric studies that have found strong evidence against the hypothesis that households act as if they are individuals, researchers have proposed a number of different models of the interaction that occurs between individuals within the household. Many of these models share with the standard model the assumption that the allocation of resources is Pareto efficient. These models make a variety of alternative assumptions, however, concerning the sharing rule within the household and the threat points used as fallback positions by the individuals in the household in the event that a cooperative equilibrium is not achieved.","id":"-1366807099"},"keywords":[],"sieverID":"c713c076-b75e-4b82-a807-ef3e3002635f","pagecount":"6","content":"his chapter challenges one of the main tenets of agricultural economics-that households behave as though they are single individuals, with production factors allocated efficiently between men and women. In many contexts this is a convenient and innocuous assumption. It can be quite restrictive, however, when investigating the causes and welfare consequences of gender differences in agriculture. In response to a growing number of econometric studies that have found strong evidence against the hypothesis that households act as if they are individuals, researchers have proposed a number of different models of the interaction that occurs between individuals within the household. Many of these models share with the standard model the assumption that the allocation of resources is Pareto efficient. These models make a variety of alternative assumptions, however, concerning the sharing rule within the household and the threat points used as fallback positions by the individuals in the household in the event that a cooperative equilibrium is not achieved.This conception of the household is far richer than the traditional unitary model. It opens the door, in particular, to an analysis of the distribution of resources within the household. Even this enriched vision of the household is too narrow, however, to address most of the concerns raised by observation of gender differences in agricultural production. The problem is that the assumption that intrahousehold allocations of consumption are Pareto For details, see Alderman et al. (1996).efficient minimally implies that the allocation of resources in production is allocatively efficient. This implication, in turn, implies that although the issues of gender and intrahousehold allocation may have distributional implications, they are unrelated to productive efficiency. For example, discrimination against women in the allocation of credit might weaken the bargaining position of women (and thus lower their welfare), but any credit that reaches any member of a household will be allocated efficiently across the productive activities of all of the members of the household.The chapter tests the assumption of Pareto efficiency in production using an extremely detailed farm-level agronomic data set from six villages in Burkina Faso. An important characteristic of the farming systems in these villages, as in much of Sub-Saharan Africa, is that different members of the household simultaneously cultivate the same crop on different plots. This makes Burkina Faso an unusually opportune environment in which to test whether differences in who controls the plots (the man or woman of the household) are reflected in how resources are allocated across those plots. Pareto efficiency in production implies that yields should be the same on all plots planted to the same crop within a household in a given year (controlling, of course, for plot characteristics).The data are from a four-year (1981-85) panel study, conducted by the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), of 150 poor households (with incomes of less than US$100 per capita) in six villages in three agroclimatic zones.1 Enumerators visited the sample households every 10 days to collect information on inputs (labor, manure) and outputs on each plot since the previous visit, yielding usable data on a total of 4,655 cultivated plots. The data also included plot area, location, topo-sequence, and local name for soil (one of 89 possible soil types), making possible a much finer control for land quality, another production variable, than is generally possible with developing-country data.The analysis produced several surprising findings that contradict the assumption that household production factors are allocated efficiently between men and women. First, plots controlled by women have significantly lower yields than other plots within the household planted to the same crop in the same year but controlled by men. On average, yields are about 18 percent lower on women's plots than on similar men's plots simultaneously planted to the same crop within the same household (Table 8.1). The yield difference is as much as 40 percent for sorghum and 20 percent for vegetable crops, in which women tend to specialize. Large gender differences in yields, however, do not imply that women are less efficient cultivators than men. The yield differences might reflect differences in the intensity with which inputs are applied on men's and women's plots.Indeed, plots controlled by men have higher labor inputs by both men and children than do plots controlled by women (Table 8.2). Plots controlled by women have labor inputs primarily from the women themselves. Nonhousehold labor (unpaid exchange labor) is used more intensively on plots controlled by men. Moreover, virtually all manure (fertilizer) is concentrated on the plots controlled by men. Previous studies (for example, those reviewed in Quisumbing 1996) have found differences in output per acre or per person but failed to isolate the source of these differences. This study finds that the gender yield differential is apparently caused by the difference in the intensity with which measured inputs of labor, manure, and fertilizer are applied on plots controlled by men and women rather than by differences in the efficiency with which these inputs are used.What are the implications of the existing allocation of resources across plots within the household? The analysis found that the output of women's plots, and therefore total household output, could be increased by between 10 and 20 percent by reallocating actually used factors of production between plots controlled by men and women in the same household (Table 8.3). 2These findings suggest that a richer model of household behavior-one that takes into account internal household dynamics and recognizes that individuals compete as well as cooperate-is necessary for understanding the 64 HAROLD ALDERMAN ET AL.2 Table 8.3 uses alternative specifications of the production function: a Translog specification, which is more flexible, and a Cobb-Douglas specification.  structure of agricultural production and for designing appropriate policy interventions.These dynamics have affected a number of efforts to improve agriculture in Africa. For example, an attempt in the early 1980s to increase rice production among women farmers in Cameroon (Jones 1986) failed because rice was considered a male crop and any income generated from it would have been controlled by men, even if the crop were produced by women. Consequently, few women entered into rice cultivation. Instead, they continued to grow sorghum, the product they controlled, despite its lower returns. By contrast, a project in Togo to encourage soybean production (Dankelman and Davidson 1988) succeeded precisely because it was designed to ensure that the crop would remain in the hands of women. The soybeans were not introduced as a cash crop, which would have changed their status and made them a male crop; instead, they were promoted as legumes that could be used to make sauces. As a result, women began cultivating soybeans on their own small plots.The effectiveness of interventions depends not just on how women respond to them, but on how others respond to them as well. For example, a fertilizer subsidy that increases the output from women's plots may, in the longer term, cause husbands to reduce the amount of land allocated at marriage. Thus it is not sufficient for the design of better policy to recognize that households do not act as single individuals. It is also important to recognize that changing the incentives or constraints faced by one household member may induce others to change their behavior in ways that frustrate the intention of the policy intervention. Only by using models that accurately reflect complex household dynamics will policymakers be able to design policy interventions that are effective.","tokenCount":"1226","images":[],"tables":["-1366807099_1_1.json","-1366807099_2_1.json","-1366807099_3_1.json","-1366807099_4_1.json","-1366807099_5_1.json","-1366807099_6_1.json"]}

data/part_2/0423899412.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"3e4f309348fba270314c49a145e0ef08","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/4f3c9b46-8c2b-467a-8d33-86410b61ee53/retrieve","description":"Curtis Farrar surveys the initial ten years of IFPRI, beginning with the genesis of the new institute, the early years, its entry into the CGIAR System, its growth in the CGIAR. He looks at the findings and recommendations of IFPRI's First External Program Review and finally describes IFPRI in its tenth year. The author utilized, as source material, documents of the CGIAR, principally the verbatim and summary reports of CGIAR meetings, reports of TAC meetings, and files of supporting documents in each case, in the CGIAR Secretariat Library at the World Bank. In addition, he used the official files of IFPRI, including minutes and supporting documentation for meetings of the Board of Trustees and its committees, and records and documentation for external program reviews.","id":"-407380121"},"keywords":[],"sieverID":"e5a92953-20d6-43ae-ab80-ac7b430e5d22","pagecount":"47","content":"A t the time of IFPRI's 20 th an ni ver sary, I asked Curt Far rar to un - der take a his tory of the In sti tute. This has turned out to be a consid er able task. Al though the proj ect is not yet com plete, I thought it would be use ful to make some of the con tent avail able as work on the whole con tin ues. There fore, a number of chap ters will be ap pear ing in a se ries of dis cus sion pa pers on im pact as sess ment that is be ing published by the Di rec tor Gen er al's Of fice.The first chap ter of the his tory, cov er ing the pe riod from the concep tion of IFPRI in the early 1970s to its full in te gra tion into the CGIAR, does not fit the im pact as sess ment model. It does, how ever, con tain in sights of value to cur rent IFPRI board mem bers, staff, support ers, and col labo ra tors. It there fore seemed ap pro pri ate to pub lish it as a sepa rate book let and to make it broadly avail able, es pe cially to those who will be help ing us cele brate our 25 th an ni ver sary this year. Even those who were af fili ated with IFPRI through much of this period will find here in for ma tion we did not know or have for got ten that helps ex plain IFPRI as it is to day.Per Pinstrup-Andersen Di rec tor Gen eralThe Genesis of a New Institute T he In ter na tional Food Pol icy Re search In sti tute was in cor po rated in the Dis trict of Co lum bia on March 5, 1975, and opened for busi ness on August 15 of the same year. Al though the Con sul ta tive Group on In ter na tional Ag ri cul tural Re search (CGIAR) agreed in late 1979 to sup port IFPRI fi nan cially, at which point the In sti tute was already mak ing a name for it self in the broader de vel op ment com munity, it was not un til the end of 1984 that IFPRI would find it self se curely set tled in the frame work of that unique do nor or gani za tion.The story be gins well be fore 1975. While the de tailed pro posal for what be came IFPRI was not made un til 1974, IFPRI's roots, and ar guments about its pur poses and func tion, reach back a full dec ade ear lier. The prob lem of deal ing with global food pol icy is sues was a mat ter of con cern, and even of con tro versy, both be fore and dur ing the dis cussions that led to the crea tion of the CGIAR in 1971 and in its early, forma tive years. Aside from a gen eral un ease among some for eign aid man ag ers at mix ing so cial sci ence with what they per ceived as the more ro bust and less con tro ver sial ac tivi ties of physi cal sci ence, the prin ci pal is sues were whether there was need for an ad di tional source of analy sis of global food trends given that the Food and Ag ri cul ture Or gani za tion of the United Na tions (FAO) and the U.S. De part ment of Ag ri cul ture (USDA) were al ready en gaged in this work; the ap pro priate ness of hav ing an in sti tute pos si bly in flu enced by the mar ket in terests of the United States, Can ada, and Aus tra lia study ing the food trade poli cies of Europe, and the ap pro pri ate lo ca tion of such an in stitute, with Rome, Wash ing ton, and a de vel op ing coun try as the prin cipal con tend ers. These con tro ver sies were not re solved at the found ing of IFPRI. Nor did they dis ap pear when the CGIAR added IFPRI to the re search cen ters it spon sored in 1980. Most of the origi nal is sues were put to rest in 1984, with the first ex ter nal re view of IFPRI, the point at which this pa per con cludes, when the CGIAR fi nally rec og nized IFPRI as be long ing firmly to the sys tem. Some of the same ques tions still found ech oes at meet ings of the CGIAR and its Tech ni cal Ad visory Com mit tee (TAC) as late as the mid-1990s.Pol icy Research I n the late 1960s and early 1970s con cern about the dan ger of world food short ages and dis cus sions about the need for ag ri cul tural research led to the found ing of the CGIAR as a do nor or gani za tion to pro vide re li able fund ing for in ter na tional ag ri cul tural re search cen ters de voted to the in ter ests of de vel op ing coun tries. That pe riod saw persis tent food cri ses in South Asia, drought-induced fam ines in Af rica, and un ex pected short ages in over all world sup ply. Al though bio logical re search to im prove pro duc tiv ity was the over whelm ing pri or ity in these dis cus sions, many par tici pants also saw a criti cal role for so cial sci ence. Pro po nents of policy-oriented so cial sci ence re search put forth three prin ci pal ar gu ments to jus tify a sig nifi cant in vest ment of re sources in in ter na tional level re search.First was the tech no logi cal ar gu ment, based on the need to cre ate the con di tions un der which farm ers would use the new ag ri cul tural tech nolo gies to in crease food pro duc tion in the de vel op ing world, thus re duc ing the grow ing need for im ported food that the poorer coun tries could not af ford. The suc cess in Asia in the late 1960s of the mod ern va rie ties of wheat and rice, which had blos somed from the re search of the In ter na tional Rice Re search In sti tute (IRRI) and the In ter na tional Maize and Wheat Im prove ment Cen ter (CIMMYT) and its prede cessor pro grams, proved that this ap proach could work. There was, however, a risk that gov ern ments would fail to en cour age the ap pli ca tion of more pro duc tive ag ri cul tural tech nolo gies be cause of other pri orities such as im port sub sti tu tion, industry-led de vel op ment, or constrain ing poli cies and pro ce dures un der which food pro duc tion and dis tri bu tion took place. It was nec es sary, there fore, to find ways to promote ef fec tive poli cies that would nur ture crea tion and use of more pro duc tive food tech nol ogy in the de vel op ing world. Cen ters supported by the CGIAR to do re search on com modi ties or eco logi cal zones would be ex pected to study the so cial and eco nomic ob sta cles to farm ers' use of im proved tech nolo gies. IRRI and CIMMYT were already do ing so. But the range of such stud ies would be lim ited to specific com modi ties and pro duc tion sys tems. Broader ques tions, such as price, trade, credit, tax, and pub lic in vest ment poli cies, and mechanisms for the dis tri bu tion of ag ri cul tural in puts and out put, needed to be ad dressed to en sure that new tech nolo gies could be ef fec tive.Sec ond, the ru ral de vel op ment ar gu ment was founded on the belief that de vel op ment spe cial ists needed to bet ter un der stand the whole pro cess of ru ral and ag ri cul tural de vel op ment. This un der stand ing could best be achieved by a wide range of so cial and eco nomic research, par ticu larly at the vil lage level and ex tend ing be yond pol icy issues as such and also be yond the food sec tor. In the ab sence of such un der stand ing, ef forts to im prove the food situa tion by, say, im prov ing tech nol ogy or free ing mar kets would be likely to fail be cause of unfore seen in ter ac tions with other as pects of the econ omy. This approach had links to the ideas of com mu nity de vel op ment, or in te grated ru ral de vel op ment, which saw a re sur gence in the early 1970s. The flow of think ing along this line is de scribed in Krue ger, Mi cha lo pou los, and Rut tan (1989,18,(172)(173)(174)(175)(176).Third, the in ter na tional ar gu ment was based on a per ceived need for bet ter un der stand ing of the evo lu tion of the world food situa tion and its im pli ca tions for de vel op ing coun tries. With so much un certainty and con tro versy over the world's sup ply of food, and over who or what was to blame for food prob lems, it was criti cal to have a continu ous and ob jec tive as sess ment of what sup ply and de mand were likely to be, when trou ble might strike, and which coun tries were most likely to be af fected. The food shocks of 1972 and 1974 sur prised experts and re spon si ble of fi cials alike. The per cep tion that the world was ac tu ally close to run ning out of food gen er ated a strongly felt need for in for ma tion about both the short-range and long-range status of food stocks and trade flows and for bet ter sys tems to im prove the chances of the world's man ag ing its food sup ply ef fec tively un der con di tions of scar city. Ex ist ing mecha nisms had been found want ing and needed to be strength ened, sup ple mented, or re placed. The oil cri sis of 1973 gener ated con cern that chemi cal fer til iz ers, a criti cal ele ment of the Green Revo lu tion, would be come scarce, high priced, and dif fi cult for de velop ing coun tries to ob tain. Not only stocks of food, but also sup plies and dis tri bu tion of ag ri cul tural in puts, it seemed, ur gently needed moni tor ing and in ter na tional man age ment.The tech nol ogy theme, as it bears on the gene sis of IFPRI, had its roots in ag ri cul tural de vel op ment in In dia in the 1960s, and par ticularly the ex pe ri ences of Sir John Craw ford, an Aus tra lian econo mist, ad viser to the World Bank, and chan cel lor of the Aus tra lian Na tional Uni ver sity, and David Hop per, a mem ber of the Ford Foun da tion staff in New Delhi. Both of these men would later play mul ti ple im por tant roles in es tab lish ing IFPRI. Sir John was found ing di rec tor of the Austra lian Bu reau of Ag ri cul tural Eco nom ics in 1945 and a life-long cham pion of high-quality pol icy re search. Sir John led, and Hop per par tici pated in, a study of the In dian ag ri cul tural sec tor as part of a World Bank ap praisal of In dian de vel op ment in the mid-1960s directed by Ber nard Bell. Food pro duc tion in In dia was rela tively stagnant in the face of in creas ing popu la tion pres sure. Re cur rent weatherre lated pro duc tion cri ses in that coun try had played a strong part in engen der ing pes si mism about the fu ture of food sup plies not only for South Asia, but for the rest of the de vel op ing world as well. Sir John saw poli cies that would en able In dia to take ad van tage of the new cereal tech nolo gies then com ing on line as the only sen si ble strat egy for avert ing the loom ing cri sis.The new high-yielding rice and wheat va rie ties were in put in ten sive, and steps to en cour age their use raised con flicts with the two prin ci pal approaches to food pol icy cur rent in In dia. One of these was labor-intensive, low-input ag ri cul ture, as so ci ated with in te grated com mu nity de vel opment and land re form, which was strongly rep re sented in In dian po liti cal thought and had been pushed by agen cies pro vid ing tech ni cal as sis tance to South Asia over the pre vi ous dec ade, in clud ing the Ford Foun da tion.The other, pre domi nant among the fi nance and plan ning agen cies of the In dian gov ern ment, op posed in vest ment in ag ri cul ture in fa vor of de velop ment of import-substituting in dus trial ca pac ity. Food aid, which had re cently be come a large fac tor in the In dian econ omy, not only pro vided com modi ties to miti gate the food scar city caused by the un fa vor able long-term trends and va ga ries in the weather, but also gen er ated budget re sources that could be used in pur suit of in dus tri ali za tion. It was therefore con ven ient to al low for eign do nors of food to han dle short falls in total sup ply rather than use scarce re sources to raise pro duc tion to needed lev els (Hop per 1987, 160-161). In the sec ond half of the 1960s, how ever, the In dian gov ern ment be came an tago nis tic to the use of food aid by the United States as a means of forc ing pol icy changes. In dia thus had an incentive to raise pro duc tion and re duce U.S. in flu ence on In dian economic pol icy.The In dian gov ern ment, fa vored at the time by sev eral out stand ing lead ers who well un der stood what was at stake, un der took the pol icy changes needed to fa cili tate the ini tial steps of the Green Revo lu tion. Hop per has de scribed the sup port ing roles of Craw ford, other for eign ad vis ers, and for eign aid agen cies (Hop per 1987, 158-172). The ex peri ence con firmed Sir John's be lief, ex pressed in a lec ture de liv ered in Aus tra lia in 1974, that \"re search alone will achieve lit tle un less it is part of a co or di nated na tional pol icy deal ing with all other as pects of the food prob lem, such as the sup ply of fer til iz ers and other in puts, irri ga tion, credit, price in cen tives, land re forms and off-farm pro grams\" (as sum ma rized in Evans and Miller 1987, 175).Rec og ni tion of the es sen tial role of pol icy in rais ing food pro duction through tech nol ogy im prove ment did not, how ever, lead eas ily to a char ter for an in ter na tional cen ter per form ing re search on food policy. The pol icy is sues as so ci ated with dif fu sion of im proved tech nology were con sid ered coun try or lo ca tion spe cific, be cause of physi cal as well as so cial and eco nomic fac tors. Moreo ver, na tional gov ernment de ci sions on those is sues would be highly po liti cal, and thus diffi cult for an out side agency to ad dress, even if that agency could le giti mately claim ob jec tiv ity. To be sure, there would be value in research on the over all pol icy frame work and in mak ing in ter na tional com pari sons, top ics that could be stud ied at the in ter na tional level. Some of the ini tia tives pro posed in the frame work of the tech nol ogy theme saw value in the crea tion of an in ter na tional re search cen ter, but gen er ally those ap proach ing the topic from this view point laid greater stress on strength en ing the pol icy ori en ta tion of so cial sci en tists working in the in ter na tional cen ters do ing bio logi cal and eco logi cal research, en hanc ing the ca pac ity of re search ers in de vel op ing coun tries, and es tab lish ing re gional net works to link both groups to gether.Ad vo cates of the ru ral de vel op ment con cept saw it as an ap pro priate fo cus for an in ter na tional in sti tu tion. At the same time, this approach shared many of the prob lems of the tech nol ogy ar gu ment in deal ing with con tro ver sial and location-specific is sues. Un like the other two ar gu ments, this one was iden ti fied strongly with the FAO. An ef fort made within FAO to de velop a pro posal for an in ter nal or affili ated in sti tu tion to em body this broad con cept ul ti mately failed. Suffi cient re sources were not forth com ing, sug gest ing a lack of do nor con fi dence in the ca pac ity of the FAO to man age such an en ter prise. The World Bank also had an ex ten sive ru ral de vel op ment pro gram begin ning early in the 1970s.The in ter na tional ar gu ment, which sur faced in the early 1970s, did jus tify the pos si bil ity of a new in ter na tional re search in sti tu tion. Neither poli cy mak ers and pol icy ana lysts in in di vid ual de vel op ing countries nor so cial sci en tists in re search cen ters work ing on spe cific com modi ties were well placed to deal with broad is sues among countries, nor did they have easy ac cess to data about such mat ters as in terna tional food stocks and trad ing pat terns. De vel op ing coun tries were se ri ously af fected by what was hap pen ing to world food sup plies and lacked the in for ma tion and analy sis to deal con struc tively with these is sues. There was thus an ob vi ous role for some in ter na tional en tity, free from po liti cal con trol, to ap praise the food-related ac tivi ties of international or gani za tions and na tional gov ern ments from a developing-country per spec tive and pro vide those coun tries with infor ma tion on which to base sen si ble pol icy de ci sions.The in ter na tional ar gu ment was also con tro ver sial, how ever. It seemed to en croach on the re spon si bili ties of ex ist ing in ter na tional bod ies, par ticu larly the FAO. There was no ground swell of sup port from de vel op ing coun tries for the idea of a new in ter na tional re search in sti tute con cerned with food pol icy is sues. Proba bly most im por tant, how ever, it was po ten tially em bar rass ing to developed-country govern ments who re garded their food poli cies as their own busi ness and mat ters of high na tional pri or ity. This was par ticu larly true for European gov ern ments in volved in the Com mon Ag ri cul tural Pol icy, a set of pro tec tive meas ures in tended to raise in comes in their ag ri cul tural sec tors. With its evi dent Aus tra lian, Ca na dian, and U.S.-based support, the pro posed in ter na tional re search in sti tute may well have appeared to the Euro pe ans to be a stalking-horse for their ag ri cul tural com peti tors and crit ics.These themes re ceived some at ten tion in the se ries of meet ings in 1969 and 1970 that led to the crea tion of the Con sul ta tive Group on Inter na tional Ag ri cul tural Re search in 1971. Dis cus sions of so cio economic and pol icy re search con tin ued with lit tle re sult in the first years of the CGIAR. At first, the tech nol ogy and ru ral de vel op ment approaches were domi nant. The in ter na tional con cept emerged with the threat of global food scar city in the 1970s. The TAC set up to ad vise the CGIAR, with Sir John as chair man and David Hop per as a member, placed the topic on its agenda sev eral times. In con junc tion with the CGI AR's an nual meet ing in 1973, the U.S. Agency for In ter national De vel op ment (USAID) spon sored a Socio-Economic Re search Semi nar, at which Sir John summed up the dis cus sion. There was consid er able agree ment at the semi nar that food pol icy is sues needed atten tion be yond what could be done by ex ist ing cen ters, but the idea of a sepa rate in sti tute for this pur pose failed to gather gen eral sup port. There was a con spicu ous lack of agree ment on what to do, and con sider able skep ti cism was ex pressed.There was strong in ter est in a food pol icy ini tia tive, how ever, among a few key play ers: the two Ameri can foun da tions, Ford and Rocke fel ler, and the In ter na tional De vel op ment Re search Cen tre of Can ada (IDRC). (Hop per was presi dent of the IDRC, and Craw ford was a mem ber of the IDRC board.) Other do nors, in clud ing the World Bank and USAID, were sym pa thetic. The FAO was in ter ested in play ing a cen tral role.W hen TAC met in Feb ru ary 1974, things be gan to move. TAC had be fore it the re port of the USAID semi nar, which con veyed strong in ter est in so cial sci ence re search, even though it was short on con crete pro pos als. There was also follow-up work com mis sioned by the Ford Foun da tion from Oris V. Wells, a former dep uty di rec tor general of the FAO, con tain ing spe cific in sti tu tional pro pos als, but only for the func tion of col lect ing and ana lyz ing more ac cu rate short-term in ter na tional food in tel li gence (Wells 1974). FAO con trib uted a pa per en ti tled The Pos si bili ties of In ter na tional As sis tance to De vel op ing Coun tries in Re search on So cial and Eco nomic Prob lems of Ag ri cultural and Ru ral De vel op ment. TAC learned that FAO Di rec tor Gen eral Boerma-him self an econo mist-had pro posed es tab lish ing a \"more or less autono mous\" re search en tity work ing closely with FAO to study the work ings of the de vel op ment pro cess in re la tion to ag ri culture. An FAO work ing group had been set up to in ves ti gate the pos sibil ity and make pro pos als. TAC re sponded by set ting up a small in ter nal com mit tee to study the is sue. This step was ini ti ated by the TAC chair man, Sir John Craw ford, who chaired the sub com mit tee him self. The mem bers were David Hopper, Ver non Rut tan, who had re cently be come presi dent of the Ag ri cultural De vel op ment Coun cil, and Pe ter Oram, ex ecu tive sec re tary of TAC.At the next TAC meet ing in June 1974, David Hop per, speak ing for the sub com mit tee, pro posed the crea tion of a \"World Food Pol icy In stitute\" with in de pend ent fund ing and gov ern ance. The sub com mit tee suggested that the in sti tute's head quar ters be in Rome to per mit close as so cia tion with the FAO. Half of the sub com mit tee docu ment was devoted to analy sis of de vices for achiev ing such an as so cia tion with out com pro mis ing in de pend ence. The FAO is quoted as sup port ing the proposal, pro vided that the new in sti tute would not du pli cate the planned FAO early warn ing sys tem on world food sup plies and that an an nual report on the world food situa tion, per ceived as du pli cat ing a regu lar FAO re port, would be dropped. The like li hood of fi nan cial sup port from the IDRC and the Ford Foun da tion was men tioned. Rocke fel ler joined the list of pro spec tive sup port ers later. TAC ap proved the pro posal, though not unani mously, with the changes re quested by FAO.The re spon si bili ties of the in sti tute, as pro posed by TAC for the ap proval of the CGIAR, would be re search on world food pol icy and \"dis semi na tion of the re sults of this re search to as wide a pub lic as possi ble, but pri mar ily to na tional and in ter na tional agen cies con cerned with higher ag ri cul tural pol icy de ci sions\" (TAC Sub com mit tee on World Food Pol icy 1974, 3). \"Tech ni cal as sis tance should not form any part of its du ties . . . and train ing should be con fined to semi nars, work shops and the 'in -service' work of a small number of as so ci ate research schol ars from de vel op ing coun tries\" (p. 4).Three main re search tasks were sug gested:1. To keep the cur rent global food and ag ri cul tural situa tion un der in de pend ent sur veil lance. . . .To ex am ine se lected ma jor food and ag ri cul tural pol icy and trade prob lems, par ticu larly those in volv ing sen si tive re lation ships be tween and among coun tries. . . .To iden tify and re search emerg ing and fu ture prob lems of global con cern likely to have an im por tant bear ing on food pro duc tion and utili za tion (in clud ing com pe ti tion be tween sup plies for food and feed) in the longer term.A ma jor ob jec tive of these stud ies would be to in di cate the actions needed in the next few years to gear up for bet ter resource al lo ca tion and man age ment and to im prove pro ductiv ity and food avail abil ity in the long run (p. 4).The ap proach TAC took to the in sti tute is es sen tially a de vel opment of the in ter na tional ar gu ment. The idea of broad re search on the de vel op ment of ag ri cul ture and the ru ral sec tor-the ru ral de vel opment con cept-was ex plic itly ruled out as too large to be ad dressed by a sin gle, small in sti tu tion and bet ter left to other agen cies. The tech nology con cept, though in cluded in gen eral terms, was not spelled out in the sub com mit tee re port. The only spe cific men tion of tech nol ogy was re lated to as sess ing its im pact. Re la tion ships with other CGIAR centers were not dis cussed.The re port sug gested that the pro posed in sti tute could avoid du plicat ing the work of other agen cies, such as FAO and the World Bank, \"by fo cus ing its ef fort par ticu larly on the analy sis of con tro ver sial or po liti cally sen si tive is sues, where the free dom of ac tion or ex pres sion of other agen cies de pend ent on gov ern ment sup port might be in hibited\" (p. 3). It is not hard to guess that this phrase, which was em phasized in both oral and writ ten pres en ta tions, set off alarms in the halls of some CGIAR do nor coun tries.TAC's rec om men da tion was put to the CGIAR and con sid ered at both for mal and in for mal meet ings over the en su ing months. Op po sition came prin ci pally from the Euro pean do nors, in suf fi cient num bers to make clear that no fa vor able con sen sus was pos si ble. Op po nents expressed doubt that such a small in sti tute (only 12 pro fes sional staff mem bers were planned) could make a sig nifi cant con tri bu tion in a complex field, given that other or gani za tions were al ready ac tive. They suggested that a de ci sion should await the re sults of the World Food Con fer ence, sched uled for the end of 1974. \"Not voiced at the meet ing, but sus pected to be in the minds of the rele vant do nors, was re luc tance to ex pose the ag ri cul tural poli cies of in dus trial coun tries-as well as those of the Euro pean Eco nomic Com mu nity-to criti cal re view and com ment by a CGIAR body,\" ac cord ing to War ren Baum, chair man of the CGIAR (Baum 1986, 127). Rep re sen ta tives of some do nor agen cies who had fought suc cess fully to per suade their doubt ful col leagues to sup port bio logi cal re search may have been leery of try ing to add sup port for soft so cial sci ence re search. It would be sev eral more years be fore the em pha sis on good de vel op ment poli cies be came fash ion able in the de vel op ment as sis tance com mu nity.At the con clu sion of the sec ond for mal CGIAR dis cus sion in Octo ber 1974, par tici pants con cluded that the Group should take no fur ther ac tion at this time; that it would un derstand that the \"pri vate\" spon sors might wish to con sider what ac tion to take with re spect to the pro posal in the light of the World Food Con ference; that the Group would like to be kept in formed on the think ing of the \"pri vate\" spon sors; in the event that they should de cide to es tab lish a cen ter that the Group would wish to es tab lish an ef fec tive com mu nica tions link with it; and that rec og niz ing it to be a pio neer ing ac tiv ity, the Group would be pre pared to re con sider the ques tion of spon sorship at some fu ture date (CGIAR Sec re tar iat 1974, 5).The World Food Con fer ence, held only a few weeks later, iden ti fied in ade quate food pro duc tion in de vel op ing coun tries and flaws in the in terna tional com mod ity mar kets as the cul prits of the food cri sis. It pro posed an in ter na tional sys tem of man ag ing food trade and ori ented it self with the tech no logi cal op ti mists, who thought that food needs could be overcome with im proved tech nol ogy and in vest ment. Even though con cerns about popu la tion growth, en vi ron mental deg ra da tion, and lim its to growth were promi nent at the time, the con fer ence avoided hard trade-off is sues, par ticu larly be tween the en vi ron ment and fu ture food pro duc tion. The con fer ence made or gan iza tional and pro cess rec om men da tions, leading to the crea tion of the In ter na tional Fund for Ag ri cul tural De vel opment, the World Food Coun cil, and the Com mit tee on Food Aid Poli cies and Pro grams, among other or gani za tions. It wel comed the FAO's de cision to moni tor the food situa tion more closely and added spe cific rec ommen da tions, such as eradi ca tion of the tsetse fly. Lit tle was said in for mal con fer ence docu ments about na tional pol icy is sues, al though a great deal was said in the cor ri dors (Weiss and Jor dan 1976, 139-142;World Food Coun cil 1984; Reut lin ger and Cas ti llo 1994).The lack of ac tion at the World Food Con fer ence and the nega tive response of the CGIAR threw the ini tia tive back to the three or gani za tions al ready com mit ted to the idea, the Ford Foun da tion, the Rocke fel ler Foun da tion, and the In ter na tional De vel op ment Re search Cen tre. Hav ing al ready agreed to fund a CGIAR pol icy re search en deavor for its first five years, the three in sti tu tions de cided to go ahead in de pend ently. FAO had made no prog ress with its plan to es tab lish a ru ral de vel op ment pol icy research struc ture. Be fore the end of 1974, the three in sti tu tions de cided to cre ate IFPRI with a pledge of sup port for five years. Al though it was not for mally re corded, all of those in volved as sumed that IFPRI would ul timately be ac cepted by the CGIAR. IDRC, the prin ci pal do nor, pro vided three shares against one share each from the other two foun da tions. Once the in sti tute started work, it was ac corded as so ci ate status with the CGIAR, a form of re la tion ship in tended to pro mote com mu ni ca tions between IFPRI and the sys tem (Baum 1986, 117-118).T he Ford and Rocke fel ler Foun da tions and the IDRC were able to cre ate a cen ter for food pol icy re search that looked quite dif fer ent from the one that might have emerged from nor mal CGIAR pro cesses. Had the CGIAR de cided to cre ate IFPRI, it would have se lected an imple ment ing agent to per form the task, proba bly one of the foun da tions, and es tab lished a task force of in ter ested do nors to pro vide sup port and guid ance. The CGIAR it self would have re viewed prog ress at each if its meet ings and made course cor rec tions. The most ob vi ous fruit of in de pend ent ac tion was that IFPRI was cre ated in a few weeks in Wash ing ton, D.C.-where the foun da tions really wanted to have it all along-as a Dis trict of Co lum bia tax-exempt cor po ra tion, rather than in Rome as ei ther a free stand ing in ter na tional or gani za tion or an en tity un der the FAO um brella. Ne go tia tions over ap pro pri ate status with the FAO and the Ital ian Gov ern ment might have taken years. Moreo ver, the ini tial man date was sim pler and more flexi ble than what would have re sulted from ne go tia tions among a group of do nors with dif fering views of what should be done. The pro gram of work soon be came quite dif fer ent from that en vi sioned in the TAC pro posal. Had CGIAR spon sor ship been achieved, Sir John Craw ford, as TAC chair man, would have had to main tain an arm's-length re la tion ship, but in stead he be came chair man of the ini tial board of trus tees and played a strong role in shap ing the fledg ling in sti tu tion.The early IFPRI Board was a re mark able group (see box). The sponsor ing or gani za tions were rep re sented by peo ple who had given years of thought to the role of a food pol icy re search in sti tute. Half of the trus tees were from de vel op ing coun tries, a bal ance still main tained at IFPRI. These trus tees com bined pol icy prac tice in in ter na tional or gani za tions or gov ern ments with ex cel lent aca demic quali fi ca tions. There were strong mem bers from Europe and one emi nent and policy-oriented bio logi cal sci en tist from an other cen ter. As the first di rec tor, the Board se lected Dale E. Hatha way, a U.S. citi zen who had been a pro fes sor at Michi gan State Uni ver sity and a mem ber of the Ford Foun da tion staff. A spe cial ist in agri cul tural trade eco nom ics, Hatha way was no new comer to the is sues he was about to face. IFPRI files con tain a short pa per he con trib uted to inter nal Ford Foun da tion con sid era tion of a food pol icy re search ini tia tive as early as 1970 (Hatha way 1970). He was de tailed by the Ford Foun dation to the sec re tar iat of the World Food Con fer ence and there af ter served as proj ect of fi cer for the food pol icy ini tia tive on be half of all three founding institutions.At its sec ond meet ing on May 5, 1975, the first meet ing de voted to pro gram sub stance, the IFPRI Board of Trus tees ap proved a pro spectus pre pared by Hatha way (Hatha way 1975). This docu ment pro vides a con ven ient ve hi cle for con sid er ing IFPRI as it ap peared to its founders and first man ag ers.In the pro spec tus, Hatha way de scribed three meth ods for deal ing with the grim food prob lems iden ti fied at the World Food Con fer ence:1. An in creased rate of growth in food pro duc tion in de vel op ing coun tries 2. In creased com mer cial im ports of food by food-deficit de velop ing coun tries 3. Greater con ces sion ary food aid to some de vel op ing coun tries All three ap proaches were likely to be fol lowed, and each would in volve sig nifi cant pol icy de ci sions by gov ern ments and in ter na tional or gani za tions. Na tional and in ter na tional poli cies would have ma jor im pacts on the suc cess or fail ure of each of the three.De vel op ing coun tries were short on the rele vant tech ni cal skills. FAO and USDA were sub stan tially im prov ing the qual ity and frequency of their re ports on the world food situa tion but were constrained be cause of their na tures and the re view re quire ments im posed by gov ern ments. Both might also have con flicts be tween their re port ing re spon si bili ties and their ac tion re spon si bili ties. There fore, an in de pend ent in sti tute that con ducted re search and pol icy analy sis on key is sues re lated to food pro duc tion, trade, and re lated mat ters could add sig nifi cantly to the in for ma tion avail able to na tional de cision mak ers and to in ter na tional or gani za tions.The ob jec tives of the In sti tute were:1. us ing all avail able sources of in for ma tion, to provide an ob jective analy sis of the cur rent and pro spec tive world food situa tion and the im pli ca tions of this analy sis for poli cymak ers giv ing spe cial em pha sis to the needs of de vel op ing coun tries;2. to iden tify ma jor op por tu ni ties for ex pand ing world food produc tion, giv ing em pha sis to de vel op ment ac tions and policies best suited to re duc ing con straints on pro duc tion, and to es tab lish ing a frame work for the sus tained use of the ag ricultural pro duc tion ca paci ties which ex ist in low in come na tions;3. to de ter mine and pub li cize those ac tions which could be under taken, and those poli cies which could be adopted by govern ments, re gional and in ter na tional agen cies, to ef fect a contin ued in crease in the quan tity of and qual ity of food sup plies avail able to all peo ple-through en hanced food pro duc tion, wider trade op por tu ni ties, and im proved ef fi ciency and equity in food dis tri bu tion (Hatha way 1975, 5-6).These ob jec tives served as a sort of man date for IFPRI un til 1979. To achieve them, ac cord ing to the pro spec tus, the In sti tute would perform two closely re lated func tions: (1) re search, to be per formed by mul ti dis ci plin ary teams, where pos si ble in col labo ra tion with other na tional and in ter na tional or gani za tions in clud ing CGIAR cen ters; and (2) cur rent analy sis and in for ma tion, draw ing on the work of the re search staff and other or gani za tions to pres ent in for ma tion on selected key is sues re lated to cur rent and pro spec tive con di tions and their pol icy im pli ca tions.The pro spec tus iden ti fied a number of po ten tial ar eas of con centra tion from which se lec tions would be made for the re search, analysis, and in for ma tion pro grams. Sev eral, but not all, of these ar eas did be come ma jor ar eas of IFPRI re search. Sev eral ar eas that later be came im por tant for IFPRI were not men tioned. The sug gested ar eas of concen tra tion fell into three main groups.1. The in ter ac tion be tween new tech nol ogy and poli cies• The ef fects of al ter na tive eco nomic poli cies on the adoption of new tech nolo gies and the im pact of new tech nologies on in come dis tri bu tion and em ploy ment. This first topic has been an im por tant fo cus of IFPRI re search.• In vest ments re quired to sup port new tech nolo gies and how to fi nance those in vest ments. This topic was the sub ject of an early IFPRI re search re port con sid er ing global needs but has been lit tle ad dressed since then.• Iden ti fi ca tion of ap pro pri ate tech nolo gies to be ad dressed by re search in sti tu tions in the light of cur rent or pro spec tive fac tor prices. This sug ges tion has not been promi nent in IFPRI re search.• The ef fect of new tech nolo gies on the com para tive ad van tage of dif fer ent ar eas, and the likely re sult ing trade and ad just ment pat terns. At the in ter na tional level, this is sue was sub sumed in work on analy sis of food trends, in which the im pact of new tech nolo gies was taken as one fac tor af fect ing com para tive ad van tage. Within coun tries, it can be seen as an ele ment of research on pro duc tion, eq uity, and sus tain abil ity.2. Ma jor op tions and op por tu ni ties for ex pand ing ag ri cul tural pro duc tion in de vel op ing coun tries• Costs of al ter na tive meth ods of ex pand ing pro duc tion, such as the choice be tween in vest ments in rain fed and ir ri gated sys tems, and al ter na tive crop ping sys tems, in clud ing the as sess ment of the po ten tial of ma jor re gions. This is as close as the list comes to pro pos ing re search on the manage ment of natu ral re sources. The topic as stated has been ad dressed through com para tive analy sis of vari ous crops and crop ping sys tems, with a par ticu lar fo cus in re cent years on the rela tive re turns to in vest ment in high-ver sus low-potential ar eas.• Com pe ti tion and com ple men tar ity be tween live stock pro duction and food pro duc tion. This topic has played an im por tant role in IFPRI re search, par ticu larly in re la tion to the com pe tition be tween feed and food uses of ce real pro duc tion.• In vest ment, re search, and pric ing poli cies needed to achieve po ten tial in creases in pro duc tion. This is a ques tion re peatedly ad dressed by IFPRI, most re cently in the 2020 Vi sion ini tia tive.• The ef fect of in fla tion on dif fer ent strate gies for ex pand ing food pol icy. This is sue seemed im por tant in the face of sharply ris ing oil prices in the mid-1970s but has not sub sequently at tracted the In sti tute's at ten tion.3. World food pro duc tion, trade, and con sump tion• Cur rent and pro spec tive food pro duc tion and dis tri bu tion, in clud ing key in puts, and the pol icy im pli ca tions of these analy ses. This was the first area of ma jor re search ac tiv ity for IFPRI and has re mained part of the re search pro gram. A sec ond is sue con cern ing the ade quacy of world food stocks was ef fec tively merged with the first.• The ef fects of food aid on con sump tion and pro duc tion, and poli cies to miti gate ad verse ef fects. This topic has been a long-term part of the IFPRI re search pro gram, al though the em pha sis was more on posi tive as pects of food aid than on miti ga tion of po ten tial nega tive ef fects.• The ef fects of ag ri cul tural and re lated poli cies of de vel oped coun tries on the world food situa tion and es pe cially on devel op ing coun tries. This has been an ac tive area for IFPRI re search and a sub ject of con tro versy con cern ing IFPRI's ap pro pri ate role.• Trade bar ri ers and prac tices that ad versely af fect pro duction and con sump tion in de vel op ing coun tries. This point partly over laps with the pre vi ous one but also fore shad ows the work of the Trade Pro gram on the bias against ag ri culture, a ma jor IFPRI ac tiv ity.• Po ten tial for ex pand ing developing-country ex ports. IFPRI has done a mod est amount of re search on this topic.• Im prove ment of sta tis tics re lated to the world food situation. This has re ceived lit tle at ten tion from IFPRI over the years.In dis cuss ing the types of skills that the In sti tute would need to address these is sues, the pro spec tus said that the first two groups would re quire in ter ac tion with sci en tists knowl edge able about ag ri cul tural pro duc tion. It raised the pos si bil ity that pro duc tion ex per tise might be needed on the staff.Al though much pol icy re search was rec og nized as be ing lo ca tion spe cific, the pro spec tus did not pro pose over seas post ings of In sti tute staff. In stead, IFPRI would hire re search ers from de vel op ing coun tries to par tici pate in IFPRI re search for up to three years and then re turn to their home in sti tu tions. This, to gether with col labo ra tion with na tional in sti tu tions, would pro vide needed lo cal in ter ac tion and help en gage de ci sion mak ers in the out come of re search. Train ing would also be con ducted by par tici pa tion of developing-country schol ars in mul tidis ci plin ary pol icy re search.The In sti tute was ex pected to have a rela tively small staff. There was pro vi sion for 5 or 6 long-term sen ior staff (with con tracts of three years or more), 12 short-term pro fes sion als of vary ing sen ior ity, a support staff of 10, and a budget of $5 mil lion over five years.Of the three pos si ble themes for a food pol icy re search in sti tute de scribed ear lier (tech nol ogy, ru ral de vel op ment, and in ter na tional), the first and the third are fully re flected in the three ob jec tives of the pro spec tus. The ru ral de vel op ment theme, rep re sented most strongly in later IFPRI re search by work on link ages be tween ag ri cul tural and eco nomic de vel op ment and by re search on food con sump tion and nutri tion, was not in the pro spec tus at all. The in ter na tional theme is clearly pres ent, as it was in the ear lier re port of the TAC sub com mittee. The tech nol ogy theme, on the other hand, is much more evi dent in the planned re search pro gram than it was in the sub com mit tee's re port. For in stance, the grant docu ment re ceived from IDRC, the prin ci pal do nor, said in part that IFPRI would work \"to iden tify ma jor op por tuni ties for ex pand ing world food pro duc tion with par ticu lar em pha sis on the de vel op ment ac tions and poli cies best suited to re move pres ent con straints to pro duc tion and to es tab lish the frame work for the sustained use of the po ten tial ag ri cul tural ca paci ties ex ist ing in lowincome na tions\" (quoted in TAC 1985, 5). With strong en cour age ment from IDRC, IFPRI was thus al ready mov ing sig nifi cantly away from the ap proach of the 1974 TAC pro posal.I n Sep tem ber 1978, three years af ter IFPRI opened its doors, the spon sor ing foun da tions re quested that the CGIAR con sider support ing the In sti tute. The spon sors ar gued that IFPRI's per form ance to date dem on strated the value of con tinu ing, but IFPRI's re source demands had grown be yond the ca pac ity of the three to meet them. It seems un likely that the three had ever aban doned the goal of se cur ing CGIAR sup port for their crea tion. It is a stan dard foun da tion prac tice to start en ter prises in the ex pec ta tion that they will be come selfsupporting or get help else where. The re ac tion of the CGIAR was still mixed, but the tone of the pre limi nary dis cus sion was gen er ally fa vorable. TAC was asked for its rec om men da tion and sent a team headed by Pro fes sor Carl Thom sen of Den mark, a TAC mem ber, to make a review of IFPRI.TAC it self had changed con sid era bly since the ini tial pro posal for a food pol icy re search in sti tute in 1974, and all of the prin ci pal TAC ac tors on the IFPRI is sue were new. Ralph Cum mings, pre vi ously the Ford Foun da tion rep re sen ta tive in In dia and first di rec tor gen eral of ICRISAT, had re placed Sir John Craw ford as TAC chair man. Crawford was still chair man of the IFPRI Board. He was re placed in this role by Sa mar S. Sen of In dia in mid-1979. Hop per and Rut tan were no longer TAC mem bers. Pe ter Oram had moved from FAO, where he had been ex ecu tive sec re tary of TAC, to IFPRI in 1976 and was now dep uty di rec tor.The Thom sen mis sion was more broadly rep re sen ta tive of CGIAR mem ber ship than the TAC sub com mit tee that made the 1974 pro posal. That group had mem bers from Aus tra lia, Can ada, the United King dom, and the United States. Headed by a Dane, the TAC mis sion of 1979 included mem bers from Chile, In dia, and the United States. Philippe Mahler, ex ecu tive sec re tary of TAC, a French na tional, played an im por tant role as sec re tary to the mis sion and prin ci pal drafter of the re port.When it vis ited Wash ing ton in Janu ary 1979, the TAC mis sion found an ac tive in sti tute, headed by John Mel lor, who had been in place for a lit tle more than a year. Hatha way, hav ing got ten IFPRI well started on the track laid out in the pro spec tus of 1975, left early in 1977 to be come as sis tant sec re tary for in ter na tional re la tions and com modity pro grams in the U.S. De part ment of Ag ri cul ture. Mel lor, a Cor nell Uni ver sity pro fes sor who spe cial ized in the role of ag ri cul ture in economic growth, par ticu larly with ref er ence to South Asia, had re cently been serv ing as chief econo mist at USAID. He took over as IFPRI's direc tor in Sep tem ber 1977. Mel lor was to lead IFPRI for 13 years and is widely cred ited with es tab lish ing the In sti tute's strong re search program and ex cel lent repu ta tion.In 1979 IFPRI had four ac tive pro grams: Trends Analy sis, Produc tion and In vest ments, Food Con sump tion and Dis tri bu tion, and Trade and Food Se cu rity. IFPRI had pub lished a large vol ume of research in a rela tively short pe riod, as sem bled a com pe tent and widely rep re sen ta tive staff, and, per haps most im por tant, had not got ten into any trou ble on sen si tive is sues.By the time Thom sen and his team ar rived, IFPRI had un der taken, and in some cases com pleted, sub stan tial work on • trends in food pro duc tion and con sump tion and their im pli ca tions;• meas ures to en sure food se cu rity at the in ter na tional level;• es ti mates of in vest ments needed to close the gap be tween food pro duc tion and con sump tion in de vel op ing coun tries;• cri te ria for al lo ca tion of re sources to re search, gen er ally and in Ni ge ria, and an ap praisal for the CGIAR of train ing needs of na tional sys tems of re search and ex ten sion;• ex plo ra tion of the in ter ac tions be tween ag ri cul tural growth and the rest of the econ omy, with a par ticu lar fo cus on in come and em ploy ment ef fects among the poor; and• stud ies of food price and sub sidy is sues as they af fected the poor in Bang la desh, Bra zil, and In dia.The Thom sen re port and TAC's rec om men da tions based on it contained some im por tant con di tions but were fa vor able to IFPRI. In Octo ber 1979, more than a year af ter the re quest from Ford, Rocke fel ler, and IDRC, the CGIAR for mally agreed to ac cept re spon si bil ity for IFPRI be gin ning in 1980. The pro cess was man aged in such a way that ac cep tance in prin ci ple came rela tively early, and fi nal de ci sions on some con ten tious is sues were made later, in stages. Some is sues ended up be ing re ferred for con tin ued con sid era tion in TAC's an nual re view of pro gram and budget sub mis sions. Three is sues, IFPRI's lo ca tion, the role of trends and trade re search, and the pro vi sion of serv ices to in ter na tional or gani za tions de serve spe cial at ten tion:T he ques tion of IFPRI's lo ca tion was the most con ten tious is sue and the one that pre oc cu pied in ter nal dis cus sion at the In sti tute. In con trast to the first TAC re port, which sug gested Rome, the Thomsen mis sion rec om mended that the board give se ri ous con sid era tion to a move from Wash ing ton to a de vel op ing coun try, to en able the staff to have a di rect and con tinu ous ap pre cia tion of the prob lems they were study ing and to avoid un due in flu ence by do nors lo cated in Wash ington. TAC strength ened the point in its own rec om men da tion to the CGIAR by say ing that the im por tance of such a trans fer \"jus ti fied a rec om men da tion that the Con sul ta tive Group when grant ing its fi nancial sup port to IFPRI should have a suf fi cient as sur ance from the Board that it would be ac tu ally ef fected and this as soon as pos si ble\" (quoted in TAC 1985, An nex V, 3). The CGIAR asked IFPRI to study this ques tion but did not make mov ing a con di tion of sup port. In due course, af ter CGIAR fi nan cial sup port had be gun, the In sti tute pro duced a re port that ar gued con vinc ingly that no sin gle lo ca tion provided an ade quate con text for global food is sues, that Wash ing ton was more con ven ient and ef fi cient than other lo ca tions, and that it was not sen si ble to go to the trou ble and ex pense of mov ing. The CGIAR as a whole went along, al though there was enough dis sent, both ex pressed and un ex pressed, to keep the is sue open for fu ture re con sid era tion.I n rec om mend ing IFPRI to the CGIAR in 1979, TAC took a very dif - fer ent line about re search pri ori ties from the one it had fol lowed in 1974. In the ear lier pro posal, the in ter na tional theme domi nated, but in 1979 TAC pushed the tech no logi cal model and mini mized the in ter national one. TAC minced no words about where the pri or ity should lie:TAC rec og nized that the man date of IFPRI in its pres ent for mula tion was very broad and could be read and in ter preted in many differ ent ways. The way this man date was trans lated in ac tual pro grams was of cru cial im por tance in de ter min ing the de gree of con cur rence of objectives be tween the CGIAR and IFPRI. TAC recom mended that, from the point of view of CGIAR sup port, the mandate of the In sti tute should give its prin ci pal em pha sis to the prob lems of de vel op ing coun tries and that the cen tral tasks in its pro gram should be con cerned with the link ages and interrelationships be tween the micro-level prob lems of the adop tion of new tech nolo gies and the wider eco nomic and socio-economic aspects of ag ri cul tural de vel op ment. Thus the work on trends analy sis and in ter na tional food trade should be con sid ered only as sup porting ac tivi ties to the main re search pro gram. The Com mit tee also con sid ered that more em pha sis should be given to the col labo ra tion with na tional in sti tutes in de vel op ing coun tries and to the pos si bilities of use ful in ter ac tion with the In ter na tional Serv ice for Na tional Ag ri cul tural Re search (ISNAR) which was be ing cre ated by the CGIAR at the time the Thom sen re port was be ing writ ten, and opened its doors in No vem ber 1979. The Com mit tee there fore recom mended that IFPRI re-examine its man date in the light of the above con sid era tions (quoted in TAC 1985, An nex V, 1-2).Avail able docu ments do not fully ex plain the sharp change in the TAC view of IFPRI be tween 1974 and 1979, par ticu larly with re gard to trade. The ar gu ment for con sid er ing trends re search in a sup port ing rather than a pri mary re search role was quite ex plicit in the Thom sen re port. Anxie ties about the world food situa tion were still abun dant, but 1978 was a year of strong im prove ment in pro duc tion and stocks (FAO 1979, vi). In ter na tional food data re port ing and analy sis had been sig nifi cantly im proved fol low ing the 1974 World Food Con ference, re duc ing the level of in ter na tional con cern on that front. The mis sion saw this re search as pro vid ing the con text within which research ers at IFPRI could ad dress na tional and in ter na tional food policy. It would also be of value to in ter na tional or gani za tions with more di rect re spon si bil ity in this field and pro vide a ba sis for iden ti fy ing prob lems re quir ing at ten tion. The rec om men da tion was not for ter mina tion of trends re search, sim ply for re gard ing it as out side of IFPRI's front line ef fort. The need to main tain a sen si ble dis tinc tion be tween the work done by FAO and that of IFPRI was very strongly put.The mis sion re port made no such pro posal re gard ing trade, however, but came close to re peat ing the ra tion ale for that re search given by TAC in 1974 for an IFPRI role in this field:Most prob lems of trade and food se cu rity are be yond the con trol of sim ple [sic, an ob vi ous mis print for \"sin gle\"] de vel op ing countries, be ing in ter na tional in na ture. Fur ther more, these coun tries have lit tle re search ca pac ity to ana lyze the prob lem and ad vance sound pol icy op tions for them selves. IFPRI has thus a clear-cut valu able role to play in this situa tion: its in de pend ent re search, geared to pro vide al ter na tive pol icy choices, di rected par ticu larly at the lower in come coun tries and peo ples, fills in a se ri ous gap in research ca pa bil ity in the in ter na tional scene (TAC 1979, 18).The only criti cisms in the re port con cern ing trade re search noted some lack of fo cus, which was iden ti fied with the need to avoid du pli cating re search by oth ers. The mis sion sug gested con fin ing IFPRI's trade research to the area where in ter na tional trade prob lems af fect do mes tic pro duc tion and con sump tion poli cies. The In sti tute should avoid spending too much time re spond ing to re quests from other agen cies but should rather col labo rate more closely with other CGIAR cen ters on these is sues.The sug ges tion to treat trade as sup port ing rather than cen tral research seems to have come from TAC it self. It found some ech oes among do nors in the CGIAR. Ac cord ing to a tran script of CGIAR dis cus sions on May 4, 1979, Su zanne Ver valcke, rep re sent ing Bel gium, said:The is sue of trade in this sec tor does, of course, have in ter na tional, or trans-national, im pli ca tions. In this re spect we do have a number of res er va tions, or a number of ques tions. In par ticu lar, what would be the ex act role of IFPRI in this prob lem? TAC it self, I be lieve, pointed out that there was a chance of per haps a greater de gree of over lap ping with the in ter na tional or gani za tions re spon si ble for trade mat ters.Simi lar sen ti ments were ex pressed in the 1979 dis cus sion by several of the do nors that had op posed sup port to IFPRI in 1974.De spite the strong po si tion taken by TAC and sev eral do nors, IFPRI suc cess fully fended off the rec om men da tion to nar row the mandate on trends and trade. As part of the pro cess of ob tain ing CGIAR sup port, IFPRI was re quired to sub mit a new state ment of its man date to re flect the nar rowed fo cus iden ti fied by TAC and deal with cer tain other is sues. When the pro posed man date state ment was dis cussed by the CGIAR on No vem ber 1, 1979, TAC Chair man Ralph Cum mings chided IFPRI for not con form ing with TAC guid ance and for fail ing to state clearly that its prin ci pal em pha sis would be on prob lems of devel op ing coun tries. IFPRI's cen tral re search task should be to study the link ages and in ter re la tion ships be tween the mi cro level prob lems of the adop tion of new tech nolo gies and the wider eco nomic and socio eco nomic as pects of ag ri cul tural de vel op ment. Trends and trade, he said, should be viewed only as sup port ing ac tivi ties. IFPRI was accord ingly asked to make fur ther re vi sions in its man date.The IFPRI re sponse, pro vided in a let ter to CGIAR Chair man Baum on March 28, 1980, de clined to ac cept TAC's views. Board Chair man Sen said, in part:The Board rec og nizes the need to re flect in the man date the im por tant re la tion ship be tween IFPRI and the other mem bers of the CGIAR system, and has amended the text . . . ac cord ingly. Cer tain other changes have been made to high light the fact that the In sti tute's main con cern is with poli cies lead ing to the al le via tion of hun ger and mal nu tri tion in the poorer coun tries, and for pur poses of clari fi ca tion. . . . At the same time the Board feels that be cause IFPRI is a pol icy re search in sti tute it has to deal with wide-ranging is sues. To re strict its man date too nar rowly to micro-level re search might so cir cumscribe IFPRI's work as to up set the needed bal ance be tween macro and mi cro ap proaches and to af fect ad versely the rele vance and qual ity of the micro-level in ves ti ga tions them selves. There fore, while IFPRI will re main se lec tive in its ac tivi ties, it must, in the Board's view re main in volved in work on trends, trade, food aid, and food se cu rity. It is evi dent from the re sponses to IFPRI's research re ports that there is con sid er able in ter est in the pol icy im plica tions of such re search, which is com ple men tary to, but not du pli ca tive of micro-level re search.The com ments about mi cro level re search, which seem to over state TAC's po si tion, were prompted, in the mem ory of one board mem ber, by con cern that there was a ten dency to wish to re strict IFPRI to the microlevel prob lems of the adop tion of new tech nolo gies (David son 1996). Over all, how ever, Chair man Sen's let ter was a straight for ward re jec tion of TAC's de mand on the nar row ing of pri ori ties. The dis cus sion of this issue con tin ued for the fol low ing sev eral years in TAC's an nual re view of the IFPRI pro gram sub mis sion, with only slightly dif fer ent re sults and no sig nifi cant im pact on the con tent of IFPRI's work.A third point in the Thom sen mis sion re port to which TAC drew par ticu lar at ten tion was the dan ger of IFPRI's spend ing too much of its time re spond ing to re quests for analy sis from in ter na tional organi za tions and im plic itly from do nors more gen er ally. TAC feared that the In sti tute might be come a serv ice agency for do nors, rather than per form ing re search for de vel op ing coun tries. Un der the ru bric of \"link ages to pol icy,\" IFPRI had called at ten tion to its work for in ter national de vel op ment or gani za tions as a means of show ing the rele vance of and de mand for its prod uct. It was too early to find ex am ples of im pact in the be hav ior of de vel op ing coun tries. To prove that it had found an audi ence, the In sti tute cited the work it had done with a va riety of in ter na tional or gani za tions in clud ing the Asian De vel op ment Bank, the World Bank, the In ter na tional Wheat Coun cil, the UN Protein-Calorie Ad vi sory Group, FAO, and the Brandt Com mis sion, as well as the U.S. De part ment of Ag ri cul ture. TAC, it self named as a bene fi ci ary of IFPRI in put, found \"a po ten tial con flict be tween the role of IFPRI as a re search or gani za tion and as a serv ice or gani za tion. Many or gani za tions, the CGIAR in par ticu lar, are likely to ex pect IFPRI to re spond to their spe cial needs and de mands.\" TAC urged IFPRI to de fine its re la tion ships with in ter na tional or gani za tions in spe cific un der stand ings and pur sue those re la tion ships through welldefined con tracts. At the same meet ing at which it rec om mended IFPRI to the CGIAR for sup port, TAC con sid ered a draft pri ori ties paper that ac knowl edged IFPRI con tri bu tions to the analy sis of re search pri ori ties.In ad di tion to mak ing these three main points, TAC in vited IFPRI to pur sue more ac tively its re la tion ships with the CGIAR and with other in ter na tional ag ri cul tural re search cen ters (IARCs). \"IFPRI could cer tainly be of great help to the CGIAR, TAC, and the IARCs in tack ling some com plex prob lems such as those of eq uity in dis tri bution of re search bene fits and pro vid ing broader per spec tive analy sis which could have an im por tant bear ing on the over all pri ori ties for and ap proaches to in ter na tional re search\" (quoted in TAC 1985, An nex V, 2). This ad vice, and other points raised by TAC, such as con tin ued study of link ages be tween ag ri cul ture and over all de vel op ment, closer work ing re la tion ships with developing-country re search or gani zations, and en hanced re search on con sump tion is sues, were read ily accepted by the IFPRI Board. took place in Food Pro duc tion Pol icy and De vel op ment Strat egy and in Food Con sump tion and Nu tri tion Strat egy. The two other pro grams, Food Trends Analy sis and In ter na tional Food Trade and Food Se cu rity re mained more or less at the same level as be fore. On the ad min is trative front, IFPRI was able, as a CGIAR-sponsored ac tiv ity, to ob tain for mal rec og ni tion of its in ter na tional status by the U.S. Gov ern ment in April 1982. The main prac ti cal bene fit of this rec og ni tion was to free IFPRI staff who were not U.S. citi zens or per ma nent resi dents from U.S. in come tax and im mi gra tion re stric tions.As its pro gram grew, IFPRI came un der in creas ing pres sure to make its re search strat egy more co her ent. In June 1982, in re sponse to urg ing from TAC and do nors, the In sti tute pub lished a docu ment called Look ing Ahead: The De vel op ment Plan for the In ter na tional Food Pol icy In sti tute (IFPRI 1982). This state ment framed the fu ture in terms of six ques tions \"ex pected to domi nate food pol icy for at least the next dec ade.\" The six are as fol lows, with brief com ments:1. What food pol icy ad just ments are needed in re sponse to rapid growth in food im port de mand by de vel op ing coun tries? This is sue was posed pri mar ily to the Food Trends Analy sis Program. Most de vel op ing coun tries were ex pected to in crease net im ports of food sta ples, be cause of the linked im pact of agri cul tural growth, over all eco nomic growth, and rapid increases in food con sump tion. In creas ing de mand for feed by a grow ing live stock sec tor was fore seen as a ma jor fac tor.The Pro duc tion Pol icy and De vel op ment Strat egy Pro gram would be re spon si ble for this work, with an al most dou bled effort ex pected. Key top ics were ag ri cul tural re search pol icy (with ISNAR) and the criti cal in puts of fer til izer and wa ter.3. What com bi na tion of farm pro ducer in cen tives can achieve growth and eq uity si mul ta ne ously? All four pro grams were involved, with a rela tively sta ble ef fort over time.4. What rela tive weight should be given to al ter na tive ag ri cultural com modi ties in fu ture pro duc tion pat terns? Again all four pro grams were in volved. The Trends Pro gram was charged with iden ti fy ing the shifts tak ing place, but the weight of re sources was planned to shift from Pro duc tion to Trade in re sponse to rap idly chang ing trade pat terns.5. What poli cies are needed for tech no logi cal change in ag ri culture to stimu late the growth in in come and em ploy ment nec essary to al le vi ate ru ral pov erty? All pro grams were in volved; most of the re sources would be man aged by the Pro duc tion Pol icy and De vel op ment Strat egy Pro gram. The re search would cover the com pe ti tion for in vest ment re sources between ag ri cul ture and other sec tors, as well as the posi tive link ages be tween ag ri cul tural growth and eco nomic de vel op ment. 6. How can food se cu rity be pro vided to the world's poor est people in the face of une qual dis tri bu tion of in come, fluc tu at ing pro duc tion, and high costs of stor age? Again all pro grams were in volved. The Trade Pro gram would main tain its fo cus on the in ter na tional as pects of food se cu rity at the same level as be fore. A pre pon der ant share of fu ture re sources, how ever, would go to Con sump tion, where they would be de voted to poli cies de signed to pro tect the poor against fluc tua tions in income and food avail abil ity and against secu lar in creases in food prices, and to en sure posi tive ef fects from new ag ri cultural tech nolo gies. IFPRI planned to study the im pact of a broad number of poli cies and pro grams on the nu tri tional status of the poor in se lected coun tries, in clud ing a gen der and house hold fo cus, and at tempt to gen er al ize the find ings to provide pol icy ad vice in ter na tion ally. (One can de tect here early evi dence of the con cept be hind the in te grated mul ti coun try research pro grams of the 1990s.)Of the six ques tions, num bers 2, 5, and 6, rep re sent ing tech no logical and de vel op men tal is sues, were sin gled out for two-thirds of the planned re sources.E ach cen ter in the CGIAR sys tem is the sub ject of pe ri odic ex ter nal re views cov er ing pro gram and man age ment, usu ally con ducted at five-year in ter vals. These re views are the prin ci pal means the sys tem uses to ap praise the ac com plish ments and con di tion of the cen ters. The first ex ter nal re views of IFPRI's pro gram and man age ment took place in 1984. The ex ter nal pro gram re view was pre pared by a panel chaired by Lloyd T. Evans of the Com mon wealth Sci en tific and In dus trial Re search Or gani za tion in Aus tra lia, a plant sci en tist, who had sev eral econo mists and an ecolo gist as col leagues on his panel (TAC 1985). The man age ment re view was chaired by an other bio logi cal sci en tist with man age ment expe ri ence, Mi chael Ar nold, who was as sisted by a man age ment ex pert. While an im por tant docu ment in its own right, the man age ment re view is of less im me di ate con cern to the as pects of IFPRI's his tory ad dressed in the pres ent study (CGIAR 1984).T he TAC chair man was now Guy Ca mus, former di rec tor gen eral of the French Of fice of Sci en tific and Tech ni cal Re search Overseas (ORSTOM), who had been a TAC mem ber at the time of the Thom sen mis sion. In trans mit ting the pro gram re view to the CGIAR, Ca mus noted that the re view was more de tailed than nor mal be cause TAC mem bers agreed \"that the ex er cise should bring to light all the ele ments nec es sary to dis pel the am bi gui ties which have sur rounded IFPRI since its en try into the CGIAR Sys tem.\" Ca mus ex pressed TAC's con fi dence that this goal had been at tained (TAC 1985, vii).Is sues of re search pri ori ties needed to be ad dressed, as well as the lo ca tion of IFPRI's head quar ters. The breadth of IFPRI's man date, and par ticu larly the role of trends and trade re search in that man date, were is sues on which TAC had taken dif fer ent po si tions in 1974 and 1979. The Evans panel adopted yet a third po si tion and per suaded TAC and the CGIAR to agree. The panel noted a shift in both the mandate and the work of IFPRI dur ing the 10 years since it was founded \"in re sponse to chang ing per cep tions of where the great est need lay. Partly this shift had re flected im proved un der stand ing of the prob lems, partly a change in em pha sis from world sur veil lance to bet ter nu tri tion for the poor in de vel op ing coun tries, and partly changes in the com plemen tary work of other in sti tu tions\" (TAC 1985, 9). The panel ex plicitly agreed with the 1980 IFPRI man date docu ment in es tab lish ing the \"pre cise ob jec tive of con trib ut ing to the re duc tion of hun ger and malnu tri tion\" and that this would re quire \"analy sis of un der ly ing processes and ex tend ing be yond a nar rowly de fined food sec tor.\"Re search on food con sump tion and nu tri tion was found to be the pro gram most clearly meet ing the \"pre cise ob jec tive\" of the man date. All of the pro grams were, how ever, on tar get, and trade re search was \"an es sen tial com po nent of the In sti tute's over all re search in its own right.\" In stat ing its con cur rence with this view, TAC ex plic itly reversed its own 1979 po si tion that trade re search should be a sup port ing ac tiv ity only. Trends re search should con tinue to serve the other programs and turn its ma jor at ten tion to im proved sys tems for data col lection in Af rica. This was a fresh ap proach, re flect ing changed times (TAC 1985, 21-43, 75-78).Through the in stru ment of this ex ter nal re view and ac tions taken in re sponse to it, the role of trade re search as a cen tral part of IFPRI's program was de cided and has not been re opened to this writ ing. The status of trends re search was put in a new frame work but not fully re solved. The more gen eral ques tion of whether IFPRI was be ing se lec tive and fo cused enough within its man date was rec og nized to be a mat ter of cir cum stance and judg ment, hence open to be re vis ited con tinu ously in the fu ture. The re view dis sected care fully and criti cally the six ques tions IFPRI propounded in 1982 as a frame work for its fu ture pro gram and sug gested that the In sti tute dis cuss them fur ther with a view to re for mu la tion. In par ticular the panel ob jected to the omis sion of in ter na tional co or di na tion of trade and aid as a sepa rate ma jor is sue.The 1980 ver sion of IFPRI's man date, pre vi ously left in a sort of limbo, re ceived en dorse ment. Through what must have been a cleri cal error, the ver sion sub mit ted to the CGIAR and TAC in 1979, rather than the re vised 1980 ver sion of the man date, was pro vided to the pro gram re view panel, was cited and blessed in their re port, and was used by IFPRI un til 1994. There were sev eral dif fer ences be tween the 1979 and 1980 mandate drafts, in clud ing a whole new para graph, but the changes were of em pha sis and pres en ta tion, not of sub stance. The fact that this slip was not no ticed and cor rected for more than 10 years may say some thing about the cost-benefit ra tio of fine-tuning man date docu ments. IFPRI's head quar ters lo ca tion was an other is sue on which TAC had taken dif fer ent po si tions in the past, sug gest ing first Rome and then a de vel op ing coun try. The 1984 re view panel made a third choice in this case as well, rec om mend ing and se cur ing TAC and CGIAR approval for IFPRI to re main in Wash ing ton \"while rec og niz ing that the is sue is a com plex one which mer its on-going con sid era tion by the Board as the na ture of IFPRI's work con tin ues to evolve\" (TAC 1985, 73). Un like the man date ques tion, this one did not dis ap pear completely but was set tled for all prac ti cal pur poses for the me dium term.A fur ther is sue to which TAC had called par ticu lar at ten tion in 1979 had been IFPRI's serv ices to in ter na tional de vel op ment and fi nan cial organi za tions. By 1984 this prob lem had greatly di min ished, as the prominence of such ac tiv ity dropped sharply within a larger and more bal anced IFPRI pro gram. The re view sub sumed this ques tion in a de tailed dis cussion of IFPRI's cli en tele. While ac cept ing the In sti tute's view that main cli ents should be poli cy mak ers in de vel op ing coun tries, the panel added pol icy ana lysts and re search ers in those coun tries as al most equally impor tant cli ents. The pol icy ana lysts and re search ers were a scat tered group, of ten lack ing both train ing and ex pe ri ence, who were con sid ered the ma jor tar get for both IFPRI's re search (as op posed to pol icy con clusions) and its institution-building ac tivi ties per formed largely as part of re search col labo ra tion. In ter na tional fi nan cial and na tional do nor agen cies were seen as in ter me di ate cli ents and as le giti mate tar gets of IFPRI re search and ap praisal. The re port said:De vel op ing coun tries them selves see a very im por tant role for IFPRI in en larg ing the scope for pol icy dia logue be tween them and the ma jor fi nan cial in sti tu tions such as the World Bank, and in in depend ently ana lyz ing the poli cies and con di tion ali ties of these agencies. IFPRI should not act as ad vo cate or apolo gist for de vel op ing coun tries, but should, through in de pend ent analy sis, ex am ine the com plex, and of ten counter-intuitive, ef fects of aid poli cies and fash ions. Such work could, at times, make IFPRI vul ner able in a way that other CGIAR cen ters are not, and may re quire con sid erable un der stand ing from do nors (TAC 1985, 75).T he re view also made many other sug ges tions and rec om men da tions.One sig nifi cant pro posal was to cre ate a sepa rate de vel op ment strate gies pro gram, ab sorb ing the work on link ages be tween the ag ri cultural sec tor and eco nomic growth, which was part of the Pro duc tion Program, but also in clud ing in ter sec to ral link ages more gen er ally, struc tural and in fra struc tural con straints, and the ef fects of mac roeconomic policies. The In sti tute re sponded by cre at ing a sepa rate, but more lim ited, Agri cul tural Growth Link ages Pro gram, drawn from the stead ily ex pand ing ac tivi ties of the Pro duc tion Pro gram.The panel saw its pro posal for more stra te gic re search as a means of en gag ing all of the In sti tute's pro grams in ad dress ing these broader prob lems. Along the same lines, it sug gested some con cen tra tion of effort \"in one or two lo ca tions where pro duc tion, con sump tion, nu tri tion and trade as pects and in ter link ages can be ana lyzed more com pre hensively\" (TAC 1985, 32). In the pan el's view the need for strat egy research and some coun try con cen tra tion was linked with its criti cism of a per ceived lack of breadth among IFPRI re search staff. It found that not only were most of IFPRI's staff econo mists, but they were economists be long ing to one part of their dis ci pline, that is, the part em phasiz ing in cen tives and in puts. The panel sug gested that the range of ex per tise needed to be broad ened by add ing econo mists in ter ested in po liti cal and eco nomic power, in ter est groups, struc tures, and in sti tutions, as well as sen ior tal ent in po liti cal sci ence and so cial sci ence. In re sponse, the Board com mented that this and a number of other sugges tions in volved ad di tional re sources at a time when IFPRI was strug gling to keep the ex ist ing pro gram funded.An other pro posal, con sis tent with the broad view the panel took of IFPRI's role, was that the di rec tor of IFPRI should give a re port to the CGIAR every two years on the food and ag ri cul ture situa tion worldwide. This thought, which harked back to the origi nal 1974 pro posal for an an nual IFPRI re port on world food and ag ri cul ture, caused a ripple of con cern among those par ticu larly anx ious to avoid com pe ti tion be tween IFPRI and FAO. It was nev er the less im ple mented start ing in 1984 and has con tin ued ever since, chal leng ing suc ces sive IFPRI leaders to find new and ex cit ing ma te rial on the global level to pres ent to the CGIAR at two-year in ter vals.The re view panel also urged IFPRI to in crease its ef forts to build up the ca pac ity of developing-country in sti tu tions, though con tinu ing to pursue this goal in for mally through re search col labo ra tion rather than through sepa rate pro grams. It en dorsed the on go ing shift in em pha sis from Asia and Latin Amer ica to ward Af rica, where the most dif fi cult food pro duc tion prob lems were per ceived to ex ist and re search would be harder and more ex pen sive. The panel and TAC en cour aged IFPRI to con tinue to col labo rate with other CGIAR cen ters, al though there was rec og ni tion that the In sti tute did not have re sources suf fi cient for ac tive in volve ment with more than a few of its sis ter cen ters at any one time.The pro gram re view panel ef fec tively closed the door on what had been an am bigu ous role for IFPRI in help ing TAC to set sys tem pri ori ties. Rec og niz ing that much of IFPRI's work would be rele vant to the con sidera tion of pri ori ties by TAC, the panel pointed out that these pri ori ties were fixed only in part on the ba sis of eco nomic analy sis. For IFPRI to become in volved di rectly in the pro cess could be a slip pery slope that would lead to di ver sion of re sources away from re search of in ter est to de vel oping coun tries. Moreo ver, a per cep tion that IFPRI had any re spon si bil ity for re source al lo ca tions within the CGIAR would dam age IFPRI's abil ity to col labo rate ef fec tively with other cen ters. This rec om men da tion was simi lar to one found in the Thom sen mis sion re port of 1974. Nev er theless, there were oc ca sions af ter 1985 when IFPRI did par tici pate in de fining pri ori ties and long-term ob jec tives for the CGIAR sys tem.T he net re sult of the ex ter nal re view and its con sid era tion by the CGIAR was to do what Guy Ca mus sought. The re view di minished the am bi gu ity and un cer tainty about IFPRI as a CGIAR cen ter, not to the van ish ing point but to a much more bear able level. Ten years af ter its crea tion, IFPRI was strongly en gaged in the is sue of what policies would en cour age the ap pli ca tion of new tech nolo gies to food produc tion in the de vel op ing world. It was also strongly en gaged in analy sis of the world food situa tion in the me dium and longer term from the point of view of the in ter ests of de vel op ing coun tries. And it was study ing se lected is sues of broader de vel op ment in ter est, for exam ple, in its growth link ages work and in food con sump tion and nu trition, but al ways in re la tion to food. IFPRI was, there fore, in some sense re spond ing to all three of the themes-tech no logi cal, de vel opmen tal, and in ter na tional-used to jus tify its crea tion. At the same time, the re view chal lenged IFPRI to re think some of its po si tions, a chal lenge that was both ap pro pri ate and wel come.Many in di vidu als made im por tant con tri bu tions to bring ing IFPRI to the well-established po si tion it held in 1985, af ter 10 years of ex istence. Some of the names are men tioned here, but it is not pos si ble to cite all of them. This pa per should not close, how ever, with out a brief ap pre cia tion of the role-more ac cu rately, roles-played by Sir John Craw ford in cre at ing IFPRI. He was a force ful and early ad vo cate, based on his per sonal ex pe ri ence in Aus tra lia, In dia, and else where, of the im por tance of food pol icy re search, and he had the pro fes sional stand ing to make him self heard. He ex er cised great in flu ence among the de vel op ment com mu ni ty's de ci sion mak ers, bi lat eral, mul ti lat eral, and pri vate, and was seen by developing-country lead ers as one who un der stood their needs. Sir John knew how to make pro cesses work and how to lead, even on oc ca sion to drive, col leagues to ward ef fective ac tion. He dem on strated this skill par ticu larly in his re la tion ship as chair man of the IFPRI Board of Trus tees with the first two di rec tors of the In sti tute and with the other mem bers of the Board. He left a unique stamp on IFPRI, and it seems quite pos si ble that with out him, some of the bat tles that led to IFPRI's crea tion and sur vival might not have been won. Sir John's close as so cia tion with IFPRI ended in 1979, just as it was join ing the CGIAR sys tem, and he died in Oc to ber 1984. bib li og ra phy with ci ta tions of docu ments that are not gen er ally avail able to in ter ested read ers. I have listed in the bib li og ra phy some docu ments from IFPRI or CGIAR files that are the source of ex ten sive quo ta tions or where I thought that a spe cific ci ta tion might pro mote un der stand ing. Many of these docu ments will shortly be come avail able through the CGIAR web site at www.cgiar.org in a col lec tion of core docu ments being cre ated by the CGIAR Sec re tar iat.Pub lished sources, which are rela tively few in this chap ter, are cited in the nor mal man ner, and listed in the bib li og ra phy. Among these, I would par ticu larly rec om mend three. Those in ter ested in learn ing more about that ex traor di nary per son, Sir John Craw ford, and his con tri bu tions to in ter na tional ag ri cul tural re search and other im por tant causes should con sult Pol icy and Prac tice: Es says in Hon our of Sir John Craw ford (Evans and Miller 1987). For the story of the CGIAR, the stan dard source, writ ten with the un derstand ing of an in sider, is Part ners against Hun ger (Baum 1986). And for a more de tailed, con tem po rary ac count of IFPRI's first 10 years, see Re port of the Ex ter nal Pro gram Re view of the In ter national Food Pol icy Re search In sti tute (TAC 1985), one of the bestwritten bu reau cratic docu ments I have read. This re port also includes ex cerpts from some of the more im por tant his tori cal documents on which this pa per is based.","tokenCount":"17509","images":[],"tables":["-407380121_1_1.json","-407380121_2_1.json","-407380121_3_1.json","-407380121_4_1.json","-407380121_5_1.json","-407380121_6_1.json","-407380121_7_1.json","-407380121_8_1.json","-407380121_9_1.json","-407380121_10_1.json","-407380121_11_1.json","-407380121_12_1.json","-407380121_13_1.json","-407380121_14_1.json","-407380121_15_1.json","-407380121_16_1.json","-407380121_17_1.json","-407380121_18_1.json","-407380121_19_1.json","-407380121_20_1.json","-407380121_21_1.json","-407380121_22_1.json","-407380121_23_1.json","-407380121_24_1.json","-407380121_25_1.json","-407380121_26_1.json","-407380121_27_1.json","-407380121_28_1.json","-407380121_29_1.json","-407380121_30_1.json","-407380121_31_1.json","-407380121_32_1.json","-407380121_33_1.json","-407380121_34_1.json","-407380121_35_1.json","-407380121_36_1.json","-407380121_37_1.json","-407380121_38_1.json","-407380121_39_1.json","-407380121_40_1.json","-407380121_41_1.json","-407380121_42_1.json","-407380121_43_1.json","-407380121_44_1.json","-407380121_45_1.json","-407380121_46_1.json","-407380121_47_1.json"]}

data/part_2/0427343198.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"60ec7c5baaf5719fd4a2f8cc0ca7784e","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/edc72005-9eb9-4cf4-a497-c18d92a599f1/retrieve","description":"Recent attention has focused on \"repurposing\" and redirecting agricultural support programs towards achieving environmental, climate and nutritional outcomes. Under these proposals, typically equivalent levels of subsidies and other forms of government support would be focused on the reducing GHG emissions, environmental externalities and other broader public policy objectives such as improving nutrition. But questions arise as to whether new support programs would necessarily be consistent with WTO disciplines. This paper examines various measures aimed at reducing GHG emissions including imposition of carbon standards and taxes, border measures to reduce slippage, and so-called \"Climate Smart\" domestic support measures and considers how such measures comport with WTO trade rules.","id":"981117006"},"keywords":["WTO","climate smart agriculture","carbon border adjustment measures","product standards"],"sieverID":"62fb57b1-b89c-46a5-bd49-7d89de7af8d5","pagecount":"23","content":"in 1975, provides research-based policy solutions to sustainably reduce poverty and end hunger and malnutrition. IFPRI's strategic research aims to foster a climate-resilient and sustainable food supply; promote healthy diets and nutrition for all; build inclusive and efficient markets, trade systems, and food industries; transform agricultural and rural economies; and strengthen institutions and governance. Gender is integrated in all the Institute's work. Partnerships, communications, capacity strengthening, and data and knowledge management are essential components to translate IFPRI's research from action to impact. The Institute's regional and country programs play a critical role in responding to demand for food policy research and in delivering holistic support for countryled development. IFPRI collaborates with partners around the world. Notices 1 IFPRI Discussion Papers contain preliminary material and research results and are circulated in order to stimulate discussion and critical comment. They have not been subject to a formal external review via IFPRI's Publications Review Committee. Any opinions, findings, and conclusions or recommendations stated herein are those of the author(s) and are not necessarily representative of or endorsed by IFPRI, Norway's International Climate and Forest Initiative (NICFI), or the World Resources Institute (WRI).2 The boundaries and names shown and the designations used on the map(s) herein do not imply official endorsement or acceptance by the International Food Policy Research Institute (IFPRI) or its partners and contributors. 3 Copyright remains with the authors. The authors are free to proceed, without further IFPRI permission, to publish this paper, or any revised version of it, in outlets such as journals, books, and other publications.Since the World Trade Organization (WTO) was formed more than 25 years ago, trade has become increasingly important to meeting global food needs. Agricultural exports have more than tripled in value and more than doubled in volume since 1995 (Figure 1). Import penetration rates (imports as a percent of consumption) have increased for most major cereals and oilseeds and countries have become increasingly reliant on imports as increases in consumption caused by income and population growth outstrip productivity gains within those countries. Developing countries have become increasingly important suppliers and consumers in world markets, and now account for about 40 percent of world food trade (Glauber 2022). South-South trade (not shown) now accounts for over 20 percent of world food trade (UNCTAD 2022).Trade helps buffer domestic prices. Countries can make up for production shortfalls by importing or by exporting surpluses when there are bumper crops. These stabilizing effects are expected to become increasingly important with climate change as production shifts due to increases in global temperatures and weather variability increases (Gouel and Laborde 2018).Arguably, the growth in agricultural trade over the past 25 years is in no small part due to the trading rules established under the WTO, particularly the Uruguay Round Agreement on Agriculture (AoA). The AoA was implemented in 1995 and brought substantial discipline to the areas of market access, domestic support, and export competition. Yet while trade has grown, the appetite for trade liberalization has stalled. The Doha Development Agenda was launched in 2001 with an ambitious work program aimed at reducing agricultural domestic support, lowering agricultural tariffs, and eliminating export subsidies. However, after years of negotiation, , members failed to reach an agreement in July 2008 (Glauber 2019). Since then, some progress has been made, particularly in the 2015 Ministerial decision to phase out export subsidies, but progress in many other areas such as domestic support, market access and public stockholding has been lacking.In the meantime, global agriculture has come under increasing scrutiny, in particular by those who criticize some agricultural subsidies and their potentially harmful impacts on health and the environment. Economists have long criticized agricultural support for distorting farmers' production decisions and agricultural markets (Johnson 1972). Such subsidies create large negative impacts for the broader community, contributing to poorer water quality, increased greenhouse gas (GHG) emissions, soil erosion in fragile lands, loss of wildlife habitat, deforestation, and other environmental concerns (OECD 2019). As shown in Figure 2, direct GHG emissions from agriculture exceeded 7.2 gigatonnes (7.2 million kilotonnes) in 2017 and accounted for 13.2 percent of total global GHG emissions (FAOSTAT 2022). Moreover, emissions from indirect land use change were over 3.5 gigatonnes in 2017 (6.5 percent of global emissions)1 . Total agricultural emissions, including pre-and postproduction emissions, accounted for about 30 percent of total global emissions (FAOSTAT 2022). Some subsidies have also been criticized by nutritionists as encouraging \"unhealthy\" diets and contributing to obesity and major health care challenges such as heart disease and diabetes (Alston, Sumner and Vosti 2008;Abay, Ibrahim, and Breisinger 2022).Recent attention has focused on \"repurposing\" and redirecting agricultural support programs towards achieving environmental, climate and nutritional outcomes (FAO, IFAD, UNICEF, WFP and WHO 2022; Gautam et al. 2022;FAO, UNDP and UNEP 2021). Under these proposals, typically equivalent levels of subsidies and other forms of government support would be focused on the reducing GHG emissions, environmental externalities and other broader public policy objectives such as improving nutrition. But questions arise as to whether new support programs would necessarily be consistent with WTO disciplines (Glauber 2018). Redirecting subsidies to \"more healthy\" foods, for example, could increase their production and export which could run afoul of WTO domestic support disciplines or be challenged under the more general Agreement on Subsidies and Countervailing Measures.Domestic policies aimed at improving climate, environmental, or nutrition goals could also have implications for border measures such as tariffs or non-tariff measures such as performance standards. For example, agricultural policy reform efforts such as the EU's Farm-to-Fork strategy would impose sustainability criteria on producers that could potentially raise the costs of production (Guyomard, Bureau et al. 2020). Unless border measures are implemented, importers could import cheaper substitutes from foreign producers who do not face such standards, thus undermining the purpose of the standards (\"leakage\"). For that reason, the European Green Deal includes the possibility of a carbon border adjustment mechanism (CBAMs) for selected sectors (European Council 2022). In the United States, Congress is considering a bill (Clean Competition Act) that would introduce CBAMs to incentivize deeper decarbonization among foreign producers while protecting U.S. firms (Reinsch and Duncan 2022).Such measures could prompt retaliation by foreign suppliers and challenges within the WTO and other bilateral and plurilateral trade agreements. Some critics have charged that measures like CBAMs and product standards introduce new forms of protectionism which ultimately hurt the most vulnerable developing countries (de Melo and Solleder 2020). Do current trade rules hamstring countries from implementing reforms that would improve climate, environmental and nutrition outcomes or is there sufficient flexibility in the AoA and other agreements that would allow such changes?In economics, an externality is the cost or benefit that affects a party, who did not choose to incur that cost or benefit. Unregulated markets with significant negative externalities are considered inefficient in that, market prices do not reflect the true (societal) cost of that good or service. Because the cost of the negative externality is not included, producers produce more (and consumers consume more), than they would had, had the cost of the externality been reflected in the price. Taxing the producer for the cost of the externality would, in theory, \"internalize\" the externality, so that costs and benefits will affect mainly parties, who choose to incur them. Carbon taxes have been proposed to address the societal externalities caused by GHG emissions through global warming.A carbon tax applied to agricultural emissions, would have a number of consequences. The immediate effect would be to raise prices of agricultural products. The map in Figure3 shows the effects of a carbon price of USD 50/MT CO2eq on effective carbon taxes for beef for selected countries. The data on GHG emissions calculated using data from FAOSTAT that calculates direct GHG emissions per kg of product produced for various countries,. The taxes were then compared to average annual price for beef 2021(World Bank Group, 2022) to demonstrate their relative magnitudes.While the map in Figure 3 is useful for illustrative purposes to show the large disparity in implied tax rates, in practice, a carbon tax would be much more difficult to implement, since GHG emissions depend on numerous factors, including practice, type of agricultural product and region (Eve et al., 2014). Any estimates derived for a carbon tax would be likely be inexact, since emissions are often from non-point sources, and this is the reason why they are difficult to measure. While direct taxes on agricultural emissions have been proposed in New Zealand (Withers 2022), a more likely approach would be for countries to impose a more targeted tax on a specific sector, such as fossil fuel production that would then have indirect impacts in energy-intensive sectors such as agriculture. Alternatively, cap and trade schemes could be introduced that limit the total amount of emissions and allow trading of emission permits. Cap and trade schemes would penalize higher emitting products and services, while providing incentives for adaption of more efficient technologies. But cap and trade schemes would still be challenging for agriculture, due to the high costs of monitoring, necessary to ensure the integrity of the trading scheme. Domestic taxes are typically not covered under international agreements since they are likely to depress, rather than enhance output. Nonetheless, as Blandford (2013) points out, exempting sectors like agriculture from emissions caps, could raise the issue of implicit subsidization by allowing exempt sectors to sell emissions credits. Like a carbon tax, the value of carbon offsets would be subject to similar uncertainty in measurement, as well as monitoring and enforcement issues. Offsets could also present food security concerns, for example, if cropland were taken out of production for carbon sequestration purposes (see Climate Smart policies below).A major concern of domestic policies such as carbon taxes is the issue of leakage--that domestic regulations could be undermined by imports that do not face the same regulations. In a analysis of an EU carbon policy, Fellmann et al. (2018) show that without sufficient border protection measures, carbon policies could lead to an increase in EU agricultural imports that would have two potential consequences. First, increased imports could lead to price declines which would harm domestic producers. Second, there could be a significant leakage effect, with extra pollution and GHG emissions in non-EU countries (assuming those imports are from countries with relatively high emitting production However, as Blandford (2018) points out, not all carbon leakage leads to a net increase in global GHG emissions. Carbon leakage may still be a net positive for global emissions if relatively highemitting domestic production is replaced by lower-emitting imports (carbon reallocation).Addressing leakage through trade measures such as carbon border adjustment measures (CBAMs) is likely to confront provisions in agreements under the WTO, particularly key provisions under the revised General Agreement on Tariffs and Trade (GATT)2 . Two core provisions are GATT Article I (most favored nation treatment or MFN) and GATT Article III (national treatment). GATT Article I holds that any advantage accorded to an imported product has to be accorded to a \"like\" product from any other WTO member. The principal of national treatment in GATT Article III holds that an imported product should be treated no less favorably than a like domestic product.WTO rules allow members to adjust product prices at the border. Governments can impose product taxes on imports or rebate taxes on exports, or both. For example, if a member imposes a consumption tax directly on a product, such as a sales tax, or environmental tax, it is permitted to impose like taxes on imports of that product3 . Much has been written on whether CBAMs would be permissible under WTO rules (Hufbauer, Charnovitz and Kim 2009;Pauwelyn 2012;Howse 2015, Trachtman 2017;Hufbauer, Kim and Schott 2021). The consensus is that if border measures are consistent with Article I and Article III, they would likely be judged consistent with WTO provision; however, implementation of \"equivalent measures\" may be problematic if border measures are based on compliance costs (as opposed to an explicit carbon tax for example).As an example, consider a cap-and-trade system where emission caps are imposed on a particular sector (for example, energy) but indirect costs are borne by other sectors (for example, agriculture). Establishing border measures for agricultural imports based on indirect impacts on production costs would likely be more problematic and likely challenged under WTO provisions as potential Article III violations. Other issues would arise when foreign supplies originate in countries that also impose carbon taxes or cap-and-trade schemes. Taxing them at the same rate as imports from countries with no such schemes may result in double taxation (Blandford 2018); exempting those countries while taxing others would likely raise Article I issues (non-discriminatory treatment between most favored nations).Article XX of the GATT does allow exceptions for specific GATT provisions if the measure is intended to protect \"human, animal, or plant life or health\" (Art XXb) or the \"conservation of exhaustible natural resources if such measures are made in conjunction with restrictions on domestic production or consumption\" (Art XXg). However, any Article XX exemption is \"[s]ubject to the requirement that that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restrictions on international trade.\" Bound tariffs are specific commitments made by individual WTO member governments. The bound tariff is the maximum MFN tariff level for a given commodity line. Members have the flexibility increase or decrease their actual or \"applied\" tariffs (on a non-discriminatory basis) so long as they didn't raise them above their bound levels.WTO members can always raise tariffs if their applied tariff bindings are less than their bound rates (Figure 4). Applied tariff rates for agricultural products are often far below bound rates, particularly for developing countries, but any tariff adjustment must be done in a non-discriminatory (that is, MFN) manner. As discussed above, this would preclude setting duties based on the emission intensity of a member. Thus, non-discriminatory tariffs could disproportionately affect lower carbon emitters with relatively high costs of production (including those who have internalized the cost of emissions in production) and favor higher carbon emitters with relatively low costs of production. The technical and legal constraints on the effective application of border measures for food and agricultural products to prevent leakage in a way that reduces global emissions but avoids trade distortions and protectionism have led some to advocate an alternative approach through the use of carbon standards and labelling. Blandford (2018) points out that an added advantage of labelling is that it can help to promote changes in consumption that will be needed to reduce the carbon footprint of food and agriculture.In contrast to a carbon tax, carbon intensity standards could be devised for agriculture in a way that could be imposed equally on both imports and domestic production. For example, a country could set a standard based on greenhouse gas (GHG) emissions intensities. If the GHG emissions from producing the product exceeded the standard, then the product could not be sold4 .Product standards are covered under the WTO Agreement on Technical Barriers to Trade (TBT), as well as other agreement such as the Agreement on Sanitary and Phytosanitary Measures (SPS). If a carbon performance standard were analyzed under the TBT agreement, a key question would be whether it was based on an international standard. If a domestic carbon standard were not based on an international standard, then under Art 2.2 of the TBT agreement, the standard would be subject to the requirement that it \"shall not be more traderestrictive than necessary to fulfill a legitimate objective\" such as protection of the environment. If the standard were decided not to be a TBT measure then it would have to comply with Art III.4 of the GATT to treat imported products no less favorably than domestically produced products.Whether standards could be imposed on developing countries is unclear as the TBT agreement states that developing countries should not be expected to use international standards that are \"not appropriate to their development, financial and trade needs.\" (TBT Article 12.4) Further, a proliferation of different sets of standards among major trading partners risks creating a \"spaghetti bowl\" of competing trade rules that increase transaction costs for businesses through variable tariffs, complicated rules of origin, and various business requirements (Baghwati 1995). For example, recent regulations put forward by the European Commission would prohibit imports of some agricultural products, such as soy, beef, palm oil, cocoa, and coffee, if they are shown to contribute to deforestation (European Commission 2021). Under its Renewable Fuel Standards (RFS), the United States determines eligibility for biofuels based on Lifecycle Analysis which includes indirect impacts on land use change. The Renewable Energy Directive of the European Union also establishes standards for eligible biofuel feedstocks that are similar to the RFS but with key differences on how maize and soybeans are treated (Panichelli and Gnansounou 2017). Smaller developing countries may be forced to choose among major trading partners or risk being left further behind, as they do not have the capacity to meet multiple sets of standards.The use of a standards approach to regulate carbon could be facilitated through the adoption of international standards for carbon measurement and labelling. For example, the Codex Alimentarius Commission, established by the Food and Agriculture Organization (FAO) and World Health Organization (WHO) in 1963, develops harmonized international food standards, guidelines, and codes of practice to protect the health of consumers and ensure fair practices in food trade. The Codex Alimentarius is well recognized and integrated into the WTO Sanitary and Phytosanitary Agreement. Likewise the International Organization for Standardization (ISO) is a non-governmental organization with a membership of 167 national standards bodies has developed a series of standards for environmental labelling5 . Establishing a similar set of standards for product labeling based on scientific research on GHG emissions and agricultural production practices could ensure that standards contribute to GHG reductions in a nondiscriminatory fashion.Climate Smart Agriculture (CSA) refers to an approach developed by the Food and Agriculture Organization of the United Nations (FAO) to develop the technical, policy and investment conditions to achieve sustainable agricultural development for food security under climate change (FAO, 2013). The CSA approach has three main pillars: 1) sustainably increasing agricultural productivity and incomes; 2) adapting and building resilience to climate change; and 3) reducing and/or removing greenhouse gas emissions, where possible. CSA aims to improve food security, help communities adapt to climate change and contribute to climate change mitigation by adopting appropriate practices, developing enabling policies and institutions, and mobilizing needed finances (FAO, 2013).In the analysis which follows we consider CSA policies relative to existing domestic support disciplines established under the WTO Agreement on Agriculture.Domestic support disciplines under the WTO Agreement on Agriculture (AoA) distinguish between programs that are viewed as minimally trade distorting (green box subsidies) and those that are not (amber box subsidies). Green box subsidies are judged to have no, or at most minimal, production-distorting effects and are exempt from reductions under the AoA and thus could be provided without budgetary limits. Amber box subsidies are judged to have more than minimal trade-distorting effects and are capped. Amber box support includes payments to producers that are tied to current production levels, market price support programs, and other policies that make payments based on current output and current market prices such as countercyclical income support programs. Amber Box subsidies are further classified into two groups-product-specific and nonproduct-specific supportand both categories are subject to de minimis tests that exempt support below a specific share of the current value of production from the reported AMS. For developed countries, if the estimated level of support is less than 5 percent of the value of current production, support is considered de minimis and excluded from calculations of the total current AMS. The de minimis threshold for developing countries is 10 percent of the value of current production6 . WTO members who had larger subsidies than the de minimis levels at the beginning of the post-Uruguay Round reform period were committed to reduce these subsidies in aggregate (by 20 percent for developed countries and 10 percent for developing countries). Countries without domestic support reduction commitments are effectively bound by de minimis levels.A third category of trade-distorting support, called Blue Box support, is addressed in Article 6.5 of the AoA. Any subsidies and other forms of income transfers that would normally be included in the Amber Box are placed in the Blue Box if the program under which those income transfers occur also requires farmers to limit production and base payments on fixed area and yields or made on 85 percent or less of a base level of production, or, in the case of livestock, such payments are based on a fixed number of head. Under the AoA, Blue Box expenditures are not capped and, therefore, not subject to any limitation or reduction commitments. There are also exemptions for developing countries (sometimes called a \"Special and Differential (S&D) Box\"), covered under provisions in Article 6.2 of the AoA. These exemptions include government measures of assistance to encourage agriculture and rural development, generally available investment subsidies, and agricultural input subsidies generally available to low-income or resource-poor farmers (Glauber 2018).To be exempt under the green box, policies must meet specific criteria laid out in Annex 2 of the AoA. To qualify, green box subsidies must not distort production or trade, or at most cause minimal distortions (Annex 2, paragraph 1). Subsidies must be government funded (not through indirect support such as higher consumer prices) and must not involve price supports. Paragraphs 2-13 of Annex 2 detail specific categories for government service programs and lay out specific qualifying criteria (Table 1). Programs for environmental programs 13Payments to producers in disadvantaged regions under regional investment programsSource: World Trade Organization, Agreement on Agriculture, Annex 2. https://www.wto.org/english/docs_e/legal_e/14-ag_02_e.htm#annIIGenerally, increasing agricultural productivity is one of the most effective ways to address climate change. For example, a 2009 study showed that if production technologies were frozen at 2005 levels, because of land-use change, carbon emissions (assuming no mitigation strategies) would be over 70 billion tons higher over the 21st century because greater amounts of land would have been necessary to produce the same amount of food. Some of these productivity gains may be partially offset by increased production at least at the national level because of improved profitability, which is sometimes called the \"rebound effect.\" At the global level, rebound impacts are likely smaller since consumption only moderately responds to lower prices caused by productivity gains (Jones and Sands 2013).Livestock and dairy production are both large sources for GHG emissions (principally methane). Improving productivity among those countries with the highest GHG intensities could have significant impacts on global agricultural emissions. Figure 5 shows country-level data collected by the Food and Agriculture Organization (FAO) of the United Nations for 2017. The data show how GHG emissions per unit of milk produced decline rapidly as productivity per cow increases until milk production reaches roughly 2 metric tons per cow, at which point, GHG emissions per metric ton of output remain roughly the same. In a recent report, Gautam et al. (2022) found that repurposing a portion of government spending on agriculture each year to develop and disseminate more emission-efficient technologies for crops and livestock could reduce overall emissions from agriculture by more than 40 percent. As Figure 5 illustrates, improving productivity in many developing countries could have large improvements in GHG efficiency.Public expenditures for CSA research and development would most likely be consistent with paragraph 2 of Annex 2 of the AoA. Other examples of general services policies would be technical assistance programs aimed at helping crop and livestock producers develop and implement nutrient management plans and research activities that promote soil health and reduce GHG emissions from cropland and livestock. Those policies do not involve direct payments to producers or processors, nor are the activities funded by transfers from consumers. Such policies would therefore qualify as green box measures under WTO rules. International funding would likely be necessary to assist poorer countries in making large development assistance of this kind.Under the AoA, the Green Box accommodates several adaptation measures that can help producers adjust to increased weather volatility that will likely accompany climate change. Agricultural insurance is available in over 100 countries (Mahul and Stutley, 2010). Annex 2 of the AoA also accommodates public reserve programs that can help buffer the impacts of production shortfalls to meet food security needs. For producers, whose livelihoods are adversely affected by climate change, Annex 2 allows structural adjustment programs, such as producer retirement programs to transition producers, to more economically viable livelihoods.Agricultural insurance. The strict criteria in paragraph 7 (income insurance) or paragraph 8 (natural disaster and crop insurance) of Annex 2 of the AoA may make it difficult to report insurance programs in the Green Box (Glauber 2017). For example, both Paragraphs 7 and 8, limit coverage to 70 percent of expected income or yield and establish guarantees based on 5 years or less of historical data. Most area-based yield programs or weatherbased derivative products tend to offer higher coverage levels and coverage levels tend to be based on expected yield or income outcomes, which may differ from averages of past outcomes. In part because insurance programs do not meet the Annex 2 criteria, most countries that notify insurance programs to the WTO notify them as Amber Box programs8 .Increased yield variability due to climate change will increase the costs of insuring these risks, as premium costs will rise. This may reduce the attractiveness of agricultural insurance as an adaptation option, unless governments continue to subsidize a large share of the premium costs. And insurance companies may be less willing to underwrite risks without large public support in the form of reinsurance and/or delivery subsidies (Glauber, 2004).Structural Adjustment payments. Climate change could have severe adverse implications for producers in affected regions. In cases where climate change has dramatically affected crops or livestock production, structural adjustment programs could help countries mitigate those impacts by providing aid to build regional infrastructure to support alternative crops, or even by taking land out of crop or livestock production, or in more extreme cases by assisting producers who leave agriculture altogether. Structural measures that are designed to assist permanently disadvantaged producers or regions, are covered in Annex 2 of the AoA under Paragraph 9 (producer retirement programs), Paragraph 10 (resource retirement programs), Paragraph 11 (regional investment aids) and Paragraph 13 (regional adjustment programs). Due to the costs of such programs, the measures -covered under Paragraphs 9, 10, 11 and 13 -have been largely used by developed members, such as the European Union and the United States of America. Like with general services, international funding would likely be necessary to assist poorer countries in making large structural adjustments of this kind.Public stockholding programmes. Like trade, stocks serve to mitigate the impact of production shortfalls (Williams and Wright, 1991). However, some public stockholding (PSH) programs are considered to be trade distorting when they involve purchases from farmers at prices fixed by the governments, known as \"supported\" or \"administered\" prices. Under the provisions of Annex 3 of the Agreement on Agriculture, market price support is calculated as the gap between a fixed external reference price and the administered price, multiplied by the quantity of eligible production. For most members the fixed external reference price is based on a 3-year average price between the years 1986-19889 .After almost a decade of relatively flat price levels since the implementation of the Agreement on Agriculture, commodity prices rose significantly between 2005 and 2011, yet the fixed external reference prices have not been updated (Figure 6). Some members such as the G-33 developing country coalition have argued that the current rules reduce policy space for PSH programs because of the effective price gap implied by the \"outdated\" reference period used in the calculation of support (Glauber and Sinha, 2021).The WTO's interim 2013 Bali decision to exempt PSH programs from legal challenge under certain conditions appears to be the most promising avenue for resolution. More technical fixes, such as updating the methodology for establishing a fixed external reference price, would have broader implications for calculating support under more general price support programs. Moreover, changing the way in which the fixed external price is calculated could raise questions on whether domestic support bindings would need to be adjusted to reflect the new methodology 10 . One potential avenue would be to expand the Bali decision to include a broader group of eligible foodstuffs than traditional staple crops, but only if the reporting requirements were maintained. Members could consider whether to agree to refrain from challenges under the dispute settlement mechanism support provided by LDCs.CSA measures to encourage practices that reduce GHG emissions or sequester carbon are likely to be consistent with Annex 2 criteria and thus not subject to reduction commitments under the AoA (and are thus unlimited) if measures are not tied to input use and production and meet the specific criteria of paragraphs 6 through 13.Payments Tied to Output or Input Use. Direct output or input subsidies to encourage adoption of new GHG technologies would generally be considered amber box programs since they are tied to production or the input itself. For example, paying farmers to adopt no-tillage practices would presumably be based on planted area, which would arguably tie the payment directly to production. The AoA does allow certain exceptions for resource-poor, low-income farmers in developing countries who could claim those measures under Article 6.2 of the AoA 11 .10 And global prices could reverse recent trends which could result in new reference price based on an updated base period higher than current price levels and thus distortionary. 11 Article 6.2 of the AoA exempts from reduction commitment government measures of assistance to encourage agriculture and rural development, generally available investment subsidies, and agricultural input subsidies generally available to low-income or resource-poor farmers in developing countries.Cost-Share Subsidies. Cost-share programs for establishing conservation practices on agricultural and have supported implementation of farming practices and structures that reduce loss of fertility through soil erosion; facilitate improved drainage, water storage, and more efficient irrigation; and provide manure storage and assistance with meeting nutrient management regulations (Claassen, Horowitz, Duquette and Udea, 2014).Infrastructural services such as water supply facilities, dams and drainage schemes, and infrastructural works associated with environmental programs are generally exempt from reduction commitments under paragraph 2(f) of Annex 2 of the AoA. To qualify as an environmental program consistent with paragraph 12 of Annex 2 of the AoA, the amount of the payment must be limited to the \"extra costs or loss of income involved with complying with the government programme\" (Paragraph 12(b)). To the extent that such measures also provide an incentive component to encourage adaptation, this may make them ineligible for Green Box protection. The reduction in land use intensity, provided by long term easements and set asides, can provide multiple environmental benefits, including substantial GHG mitigation, that occurs as carbon is sequestered in soils or vegetation. To be consistent with Annex 2 criteria for resource retirement programmes (paragraph 10), eligibility for such payments shall be determined by reference to clearly defined criteria in programmes designed to remove land or other resources, including livestock, from marketable agricultural production. Payments must be conditional upon the retirement of land from marketable agricultural production for a minimum of three years, and in the case of livestock on its slaughter or definitive permanent disposal. Payments shall not require or specify any alternative use for such land or other resources which involves the production of marketable agricultural products. Payments shall not be related to either the type or quantity of production or to the prices, domestic or international, applying to production undertaken using the land or other resources remaining in production.Biofuel Policies. Biofuel mandates have been touted as climate smart policies, though they remain controversial, because of their potential impact on GHG emissions, when agricultural production practices and direct and indirect land use effects are considered (USDA, 2016). WTO domestic support issues are potentially raised, when subsidies are used to encourage biofuel production or consumption, which, in turn, has potential impacts on feedstock production (for example, corn for ethanol production, soybeans for soybean oil-based biodiesel production). As such, biofuel policies potentially distort feedstock production and exports. Policies which affect imports of feedstocks and biofuels are covered above under the discussion of border measures and standards.Addressing the challenges of climate change through mitigation measures aimed at lower GHG emissions could have important implications for global trade and how such measures comport with existing trade provisions.Ideally, the externalities caused by GHG emissions would be fully incorporated in market prices. In practice, this will likely never be the case. Particularly challenging will be designing border measures that reduce the potentially undermining effects of leakage but are implemented in a way that provides like treatment for domestic and foreign suppliers.Product standards and labeling may provide means by which countries can prevent leakage in a way that reduces global emissions but avoids trade distortions and protectionism, but without international coordination such efforts may encourage a \"spaghetti bowl\" of standards that would likely result in legal challenges at the WTO.The creation of a standards body, modeled on the OIE or Codex Alimentarius, and building on the work of ISO 14064, International Standard for GHG Emissions Inventories and Verification, could help facilitate development and potentially avoid future disputes.In general, CSA policies aimed at reducing GHG emissions, or sequestering CO2, or mitigating the effects of climate change itself, should not be backdoor ways of promoting protectionism or distorting domestic production. Such policies, while attractive from a narrow domestic perspective, will likely have more consequential adverse impacts in international markets, particularly for developing countries who will be less likely to afford such support measures. In this sense, CSA policies should be compatible with the overarching conditions of the Green Box of providing minimal trade-or production-distorting support.That is not to say that reforms to the provisions in Annex 2 should not be considered. Annex 2 provision governing insurance are not well adapted to current insurance programs, particularly index insurance and weather derivatives, which are increasingly being used in developing countries (Miranda and Farrin, 2012). That said, heavily subsidized insurance programs, particularly those that provide price and revenue protection, can distort production since they are directly coupled to actual production (Glauber et al. 2021).Increasing productivity, particularly in developing countries where yield gaps are often greatest, should be a priority of any CSA policy. More public R&D should be directed towards climate resilient varieties and towards an aim of increasing GHG efficiency. While Article 6.2 of the AoA allows for subsidized input use to increase productivity in resource poor developing countries, input subsidies for developing countries that promote less GHG-efficient production technologies should be viewed with more caution.Resource retirement programs may offer significant opportunities for carbon sequestration through afforestation. However, the impact of sequestered carbon on GHG emissions must be analyzed in a global context in terms of indirect land use as well as food security concerns. Likewise, biofuel mandates should be limited to non-food and non-feed feedstocks, though land use effects of converting potential cropland to biofuel use remain an issue of concern.","tokenCount":"5858","images":["981117006_1_1.png","981117006_6_1.png","981117006_7_1.png","981117006_8_1.png","981117006_11_1.png","981117006_15_1.png","981117006_17_1.png"],"tables":["981117006_1_1.json","981117006_2_1.json","981117006_3_1.json","981117006_4_1.json","981117006_5_1.json","981117006_6_1.json","981117006_7_1.json","981117006_8_1.json","981117006_9_1.json","981117006_10_1.json","981117006_11_1.json","981117006_12_1.json","981117006_13_1.json","981117006_14_1.json","981117006_15_1.json","981117006_16_1.json","981117006_17_1.json","981117006_18_1.json","981117006_19_1.json","981117006_20_1.json","981117006_21_1.json","981117006_22_1.json","981117006_23_1.json"]}

data/part_2/0438778219.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"376e0d2d901975f557152528d32f3ac4","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/c4e7ea88-90ab-49c5-8d5f-b2c136cafec1/retrieve","description":"David Dawe IFPRI-FAO conference, \"Accelerating the End of Hunger and Malnutrition: November 28-30, 2018 Bangkok, Thailand","id":"1796657522"},"keywords":[],"sieverID":"9128e56e-2623-43ed-a0ba-c30bd9f3ecf2","pagecount":"19","content":"▪ Urban residents are more likely to be of prime working age in general and more likely to be young in particular (15 -24 years of age).▪ Urban residents are more likely to be female in E & SE Asia and the Pacific, but more likely to be male in South Asia ▪ Refrigeration is more common in urban areas (longer storage of perishables).▪ Urban areas have a greater diversity of markets, including supermarkets and wet markets, and demand for convenience is more prominent.▪ Purchasing food, especially highly processed food, is more common in urban areas.▪ Less land and lower water quality for urban homestead food production ▪ More exposure to advertising and/or more susceptible to advertising in urban areas ▪ More sedentary behavior in urban areas.▪ Eating out is more common in urban areas.Differences in urban and rural food systems/environments mean that different interventions will be required ▪ Urban residents are more likely to be of prime working age in general and more likely to be young in particular (15 -24 years of age).▪ Urban residents are more likely to be female in E & SE Asia and the Pacific, but more likely to be male in South Asia ▪ Refrigeration is more common in urban areas (longer storage of perishables).▪ Urban areas have a greater diversity of markets, including supermarkets and wet markets, and demand for convenience is more prominent.▪ Purchasing food, especially highly processed food, is more common in urban areas.▪ Less land and lower water quality for urban homestead food production ▪ More exposure to advertising and/or more susceptible to advertising in urban areas ▪ More sedentary behavior in urban areas.Urban areas in South Asia have relatively more males, but elsewhere in the region females are more commonSex ratio is defined as number of females to number of males▪ Urban residents are more likely to be of prime working age in general and more likely to be young in particular (15 -24 years of age).▪ Urban residents are more likely to be female in E & SE Asia and the Pacific, but more likely to be male in South Asia ▪ Refrigeration is more common in urban areas (longer storage of perishables).▪ Urban areas have a greater diversity of markets, including supermarkets and wet markets, and demand for convenience is more prominent.▪ Purchasing food, especially highly processed food, is more common in urban areas.▪ Less land and lower water quality for urban homestead food production ▪ More exposure to advertising and/or more susceptible to advertising in urban areas ▪ More sedentary behavior in urban areas.▪ Eating out is more common in urban areas.Differences in urban and rural food systems/environments mean that different interventions will be required ▪ Urban residents are more likely to be of prime working age in general and more likely to be young in particular (15 -24 years of age).▪ Urban residents are more likely to be female in E & SE Asia and the Pacific, but more likely to be male in South Asia ▪ Refrigeration is more common in urban areas (longer storage of perishables).▪ Urban areas have a greater diversity of markets, including supermarkets and wet markets, and demand for convenience is more prominent.▪ Purchasing food, especially highly processed food, is more common in urban areas.▪ Less land and lower water quality for urban homestead food production ▪ More exposure to advertising and/or more susceptible to advertising in urban areas ▪ More sedentary behavior in urban areas.Urban residents purchase more highly processed foods and grow less of their own food Differences in urban and rural food systems/environments mean that different interventions will be required ▪ Urban residents are more likely to be of prime working age in general and more likely to be young in particular (15 -24 years of age).▪ Urban residents are more likely to be female in E & SE Asia and the Pacific, but more likely to be male in South Asia ▪ Refrigeration is more common in urban areas (longer storage of perishables).▪ Urban areas have a greater diversity of markets, including supermarkets and wet markets, and demand for convenience is more prominent.▪ Purchasing food, especially highly processed food, is more common in urban areas.▪ Less land and lower water quality for urban homestead food production ▪ More exposure to advertising and/or more susceptible to advertising in urban areas ▪ More sedentary behavior in urban areas.▪ Eating out is more common in urban areas.","tokenCount":"739","images":["1796657522_1_1.png","1796657522_1_2.png","1796657522_1_3.png","1796657522_1_4.png","1796657522_2_1.png","1796657522_2_2.png","1796657522_2_3.png","1796657522_3_1.png","1796657522_3_2.png","1796657522_4_1.png","1796657522_4_2.png","1796657522_4_3.png","1796657522_4_4.png","1796657522_5_1.png","1796657522_5_2.png","1796657522_6_1.png","1796657522_6_2.png","1796657522_6_3.png","1796657522_6_4.png","1796657522_7_1.png","1796657522_7_2.png","1796657522_8_1.png","1796657522_8_2.png","1796657522_8_3.png","1796657522_8_4.png","1796657522_9_1.png","1796657522_9_2.png","1796657522_10_1.png","1796657522_10_2.png","1796657522_10_3.png","1796657522_10_4.png","1796657522_11_1.png","1796657522_11_2.png","1796657522_12_1.png","1796657522_12_2.png","1796657522_12_3.png","1796657522_12_4.png","1796657522_13_1.png","1796657522_13_2.png","1796657522_13_3.png","1796657522_14_1.png","1796657522_14_2.png","1796657522_14_3.png","1796657522_15_1.png","1796657522_15_2.png","1796657522_15_3.png","1796657522_16_1.png","1796657522_16_2.png","1796657522_16_3.png","1796657522_17_1.png","1796657522_17_2.png","1796657522_17_3.png","1796657522_18_1.png","1796657522_18_2.png","1796657522_18_3.png","1796657522_18_4.png","1796657522_19_1.png","1796657522_19_2.png","1796657522_19_3.png"],"tables":["1796657522_1_1.json","1796657522_2_1.json","1796657522_3_1.json","1796657522_4_1.json","1796657522_5_1.json","1796657522_6_1.json","1796657522_7_1.json","1796657522_8_1.json","1796657522_9_1.json","1796657522_10_1.json","1796657522_11_1.json","1796657522_12_1.json","1796657522_13_1.json","1796657522_14_1.json","1796657522_15_1.json","1796657522_16_1.json","1796657522_17_1.json","1796657522_18_1.json","1796657522_19_1.json"]}

data/part_2/0438984007.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"686eac8828f026844cc0177b3913f9c6","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/3f7ac3dd-bbb8-4fc3-898e-fbd30f5d58ac/retrieve","description":"This study aims to explore and quantify systematic similarities and differences between four major global cropping systems products: the dataset of monthly irrigated and rainfed crop areas around the year 2000 (MIRCA2000), the spatial production allocation model (SPAM), the global agroecological zone (GAEZ) dataset, and the M3 dataset developed by Monfreda, Ramankutty, and Foley. The analysis explores not only the final cropping systems maps but also the interdependencies of each product, methodological differences, and modeling assumptions, which will provide users with information vital for discerning between datasets in selecting a product appropriate for each intended application.","id":"1688536799"},"keywords":["global cropland","yield","harvested area","cropping systems","mapping","downscaling v"],"sieverID":"9fee14cd-42c7-4b28-a349-789720a0f626","pagecount":"44","content":"established in 1975, provides evidence-based policy solutions to sustainably end hunger and malnutrition and reduce poverty. The Institute conducts research, communicates results, optimizes partnerships, and builds capacity to ensure sustainable food production, promote healthy food systems, improve markets and trade, transform agriculture, build resilience, and strengthen institutions and governance. Gender is considered in all of the Institute's work. IFPRI collaborates with partners around the world, including development implementers, public institutions, the private sector, and farmers' organizations, to ensure that local, national, regional, and global food policies are based on evidence. IFPRI is a member of the CGIAR Consortium.Cropland accounts for approximately 15 to 18 million square kilometers of Earth's land coveramounting to 12 percent of the planet's ice-free land surface-yet information on the distribution and performance of specific crops is often available only through national or subnational statistics (Ramankutty et al. 2008;Foley et al. 2005). Cataloging the increasing extent and yield of cropland has implications for food security analyses, studies of land degradation, and resource management. Whereas subnational statistics provide information on raw quantities, they provide only limited information useful for spatially explicit applications. Detailed mapping of human-induced land-use change is vital to understanding water usage (Rosegrant, Cai, and Cline 2002), nutrient cycling (Bondeau et al. 2007;Liu et al. 2010), soil erosion (Yang et al. 2003), loss of biodiversity (Foley et al. 2011), and impact on regional climate (Ramankutty, Delire, and Snyder 2006;Voldoire et al. 2007).Remote-sensing products offer spatially disaggregated information, but those currently available on a global scale are ill suited for many applications due to the limited separation of crop types within the area classified as cropland. Recently, however, several initiatives have begun to incorporate the detailed information available from statistical surveys with supplemental spatial information to produce a spatially explicit global dataset specific to individual crops for the year 2000. These studies have generated increasingly sophisticated results portraying the downscaling of crop production statistics across the global extent at a moderately high spatial resolution. Global studies have been reported by Leff, Ramankutty, and Foley (2004), Monfreda, Ramankutty, and Foley (2008), Portmann, Siebert, and Döll (2010), and most recently You et al. (2013) and Fischer et al. (2010). Although such datasets provide analysts and decisionmakers with improved information on global cropping systems, the final global cropping maps differ from one another substantially. This study aims to explore and quantify systematic similarities and differences between four major global cropping systems products and compare and contrast the general conceptual, methodological, and data underpinnings of the four products: the global dataset of monthly irrigated and rainfed crop areas around the year 2000 known as MIRCA2000 (Portmann, Siebert, and Döll 2010), the spatial production allocation model, or SPAM (You et al. 2013), the global agroecological zone (GAEZ) dataset (Fischer et al. 2013), and the M3 dataset developed by Monfreda, Ramankutty, and Foley (2008).We begin with an overview of the methods and outputs of each dataset in Section 2 before comparing and contrasting the downscaling methodology each product uses in Section 3. Section 4 compares the input datasets each model uses and evaluates how interdependencies between models may propagate through the downscaling methodology to affect the final product. In Section 5, we perform a quantitative comparison across the four products, focusing on the world's three major crops: rice, wheat, and maize. We conclude with a summary and some recommendations for users of these products.Research on cropping system models has been reported at the global scale by Leff, Ramankutty, and Foley (2004), You et al. (2013), Monfreda, Ramankutty, and Foley (2008), and Portmann, Siebert, and Döll (2010), while regional applications have been reported for Latin America and the Caribbean (You and Wood 2006) and Africa south of the Sahara (You, Wood, and Wood-Sichra 2009). This paper focuses on four global cropping system models: MIRCA, M3, GAEZ, and SPAM.Of the four cropping system models considered, the M3 approach attempts spatial downscaling of the most complete coverage of crops (175, including tree and forage crops and managed grasslands) for both harvested area and yield. M3 is described in the companion paper to Ramankutty et al. (2008), which uses remote-sensing products to construct a new dataset for croplands and pasture circa 2000 at a 5-arc-minute resolution. The M3 dataset applies minimal modeling to distribute subnational statistics of yield and harvested area, opting for ease of interpretation and a limit to requisite assumptions over complexity (see Section 3 of Monfreda, Ramankutty, and Foley [2008] for further detail).MIRCA downscales 26 crops and two aggregate categories of \"other annual\" and \"other perennial\" crops, all of which are divided into rainfed and irrigated production areas. But, unique among the four approaches, MIRCA also performs a temporal downscaling so as to provide rainfed and irrigated area estimates disaggregated by month. MIRCA uses M3 downscaled crop data results as its starting point for allocating the total harvested area for each crop into rainfed and irrigated areas, and apportions the M3 crop area allocations into 402 spatial \"calendar units\" globally for which the MIRCA team has been able to compile unique sets of ancillary information on irrigation, crop-specific irrigated water use, crop calendars, and cropping intensities.SPAM covers the fewest crops, just 20, but downscales the area and yield for each crop into three different production systems: high-input irrigated, high-input rainfed, and low-input rainfed. The lowinput rainfed category is itself further subdivided into low input and subsistence as described in You et al. (2013). SPAM relies on a separate collection of subnational statistical data from that of MIRCA, focusing on increased coverage in developing countries. The SPAM approach relies on using ancillary information-including crop prices, population density (CIESIN, IFPRI, and WRI 2000, and cropspecific biophysical suitability (Fischer et al. 2010)-to distribute subnational statistics based on a method known as cross entropy (Golan, Judge, and Miller 1996;Lencer and Miller 1998;Zhang and Fan 2001). See Section 2 of You et al. (2013) for detail on the cross-entropy approach.The GAEZ dataset (Fischer et al. 2010) downscales 23 crops, including forages and other cereals in either irrigated or rainfed production systems for both harvested area and yield. GAEZ develops a cropland extent independent from SPAM, MIRCA, and M3-which all use Ramankutty et al. (2008)-and relies on an extensive analysis of crop-specific agroclimatic and edaphic suitability. Using a methodology similar but not identical to SPAM, GAEZ incorporates ancillary information such as population density, biophysical suitability, and market access into a cross-entropy framework as a means of distributing subnational statistics.Given their different approaches, the four datasets offer four different sets of products at the same 5-arcminute resolution (about 9 kilometers at the equator) globally for the year 2000. As Table 2.1 shows, the final products of the four models are as follows:• M3-350 global grids. One harvested area grid and one average yield grid for each of 175 crops.• MIRCA-624 global grids. One harvested area grid for rainfed and one harvested area grid for irrigated production of 26 crops for each month. • SPAM-240 global grids. One physical area grid, one harvest area grid, and one average yield grid for each of 20 crops for irrigated, high-input rainfed, and low-input rainfed (divided into low input and subsistence) production. • GAEZ-138 global grids. One harvested area grid, one average yield, and one production value grid for each of 23 crops for irrigated and rainfed production systems. As a first step toward delineating crop-specific harvested area and yield, each cropping system model defined a spatially explicit layer of cropland extent, representing the proportion of cropland in each 5 arcminute pixel globally. Because each subsequent step in the modeling process relies on the definition of cropland extent, the degree to which each pair of cropland extent products agree represents an upper bound of intermodel agreement on the spatial distribution of crop physical areas. M3, MIRCA, and SPAM all rely on the same base dataset for cropland extent- Ramankutty et al. (2008), which is an extension of Leff, Ramankutty, and Foley (2004). Leff, Ramankutty, and Foley synthesize satellite-derived land cover data and agricultural census data worldwide to assess the distribution of major crops across a global 5-arc-minute grid in terms of the proportion of the total harvested area of each of the crops in each administrative unit. Following and improving on this work, Ramankutty et al. (2008) developed a new global land cover dataset for croplands and pasture circa 2000 (at the same 5-arc-minute resolution of the original dataset) by combining Boston University's MODISderived land cover data (Friedl et al. 2002) and the SPOT VEGETATION-based GLC2000 (Bartholome and Belward 2005). Ramankutty et al. (2008) apply a multiple linear regression model to relate the combined satellite-derived datasets to the agricultural statistics using a least-squares-error framework. The optimization is applied separately to six different regions of the world.The cropland extent developed by Ramankutty et al. (2008) is used directly by M3 and with modifications by MIRCA and SPAM. By combining Ramankutty et al. and the Global Map of Irrigation Areas (GMIA), MIRCA produced a global dataset of monthly growing areas of 26 irrigated crops on the same 5-arc-minute grid. SPAM similarly reconciles the GMIA and the Ramankutty cropland extent by setting the cropland extent to be at least equal to the irrigated area in a preprocessing step. For more information on each of these methodologies, see Appendix A.GAEZ uses GLC2000 data and GMIA, but it also considers a global land cover categorization (IFPRI 2002) that is based on a reinterpretation of the Global Land Cover Characteristics Database version 2.0 (EROS Data Center 2000), a layer of forestland from the Forest Resources Assessment of FAO (FAO 2001), the IUCN-WCMC protected areas inventory (WDPA 2009), and an estimate of land required for housing and infrastructure for the year 2000 derived from FAO-SDRN, based on LANDSCAN 2003 and calibrated to United Nations 2000 population figures (Fischer et al. 2008;Bhaduri et al. 2002;Dobson et al. 2003). GAEZ runs a cross-sectional regression on the land cover distributions to derive weights that are then applied in an iterative adjustment procedure to match estimated reference values such that the geographic and statistical data are consistent.GAEZ and SPAM further constrain potential crop distribution using biophysical and socioeconomic suitability prior to allocating the harvested area and yield of each crop. M3 and MIRCA do not consider suitability criteria. SPAM directly uses the suitable area product from GAEZ, meaning that despite using different cropland extent products to constrain the distribution of crops, the two models use identical constraints on biophysically suitable land. The GAEZ suitability product integrates an extensive set of edaphic and climatic factors into its biophysical suitability analysis to produce a suitability index by production system and crop. Further information on the suitability index analysis developed as part of the GAEZ model may be found in Fischer et al. (2013).In addition to biophysical suitability criteria, both SPAM and GAEZ model the socioeconomic factors that often constrain or encourage crop production. As a means of differentiating between low, medium, and high input or management conditions, GAEZ divides the land into land-use types. Land-use types are derived using information on road infrastructure, livestock density, population density, and distance to market. For example, whereas low input relies on available human or livestock labor, high input is market oriented. Similar to GAEZ, SPAM explicitly models different production systems, which include high-input irrigated, high-input rainfed, low-input rainfed, and subsistence (always low-input rainfed). SPAM also includes data on crop prices and market access to construct a realistic market scenario that includes not only biophysical barriers to producing crops but also social economic forces (see Appendix A for an explicit mathematical formulation).Perhaps the largest methodological differences between M3, MIRCA, SPAM, and GAEZ lie in the approaches used to downscale statistical data reported at the administrative-unit level into grid-cellspecific values. M3 uses the most straightforward method, allocating each crop into each grid cell as the same proportion of grid cell cropland area as the crop occupies in the total harvested area of each statistical reporting unit. Crop yield in each grid cell is assigned as being the same as the yield reported for the statistical unit as a whole. This approach implicitly assumes that both environmental conditions and management/production systems are uniform across the cropland extents of each statistical reporting unit, or that there is insufficient information to characterize the spatial variations of crop production within a statistical unit. As a result, the distinct tolerances of individual crops to those spatial patterns are not incorporated in the downscaling procedure (for example, wetter, higher elevations within a province for which we hold statistical data may provide the production zones for cassava, banana, and coffee, while the lower elevations, which are drier and hotter, contain all the millet and sorghum areas). This approach does not, furthermore, acknowledge the very significant differences between the yield levels of irrigated and rainfed production systems, or of commercial and smallholder producers within these sometimes large and highly diverse statistical reporting units.MIRCA primarily focuses on reconciling the differences among information derived from subnational crop production statistics, M3 crop distributions, and the Siebert et al. (2005) irrigated areas database. MIRCA deals only with harvested area and essentially uses the relative share of rainfed and irrigated cropland within each grid cell to break out M3 total crop areas into grid-cell-specific rainfed and irrigated areas. The model also includes use of numerous checks and adjustments to reconcile differences between the cropland area of Ramankutty et al. (2008) and the irrigated area estimates of Siebert et al. (2005) within each grid cell, given that total cropland area should at all times be greater than or equal to the irrigated cropland area. Similar to M3, MIRCA does not consider any form of suitability in its downscaling procedure.The downscaling approaches of GAEZ and SPAM are predicated on the importance, in terms of subsequent utility of the downscaled estimates, of attempting to take explicit account of available evidence of the spatial variation of production conditions within the cropland extent and of the significantly different yield levels of different types of production systems even in the same production environment. Both GAEZ and SPAM use an approach that produces a result mathematically equivalent to that of a cross-entropy formulation, but GAEZ uses an iterative rebalancing procedure to adjust weighting factors until all constraints in the model are met, while SPAM uses a cross-entropy formulation. Although the two models incorporate similar information (see Table 4.1), the manner in which the information is used to constrain the model differs (see Appendix A for details on the mathematical formulation of each model). Additionally, GAEZ differs from SPAM in that it uses a location factor to incorporate spatially explicit information including geo-referenced household survey data. Although the prior in SPAM is used to capture spatially explicit information as well, the model does not include household survey data but instead leverages the field presence of the CGIAR network to incorporate an extensive dataset of expert elicitations.The major determinants of the potential reliability of downscaling efforts are (1) the quality of the cropland extent dataset indicating the physical extent and area intensity of cropland (for example, share of cropland area in each 5-arc-minute grid cell), and (2) the resolution and reliability of the subnational crop statistics. Each model builds on a common set of available data as well as previous work in cropping systems modeling. Table 4.1 illustrates both the broad linkages and increasing sets of input data and assumptions that each of the M3, MIRCA, GAEZ, and SPAM datasets relies upon. All four datasets draw on FAOSTAT national data to provide control totals for cropland area, the harvested area, and yields of specific crops, while also spending considerable efforts to collect subnational crop statistics to allow as detailed as possible disaggregation of national totals within subnational administrative boundaries. Since MIRCA relies on M3 to provide its input data on the spatial allocation of the total area and average yield (that is, not yet disaggregated between rainfed and irrigated production), it relies initially on the same sources of subnational crop statistics. SPAM relies on a separate collection of subnational statistical data sources, focusing on increased coverage in developing countries.M3 reports a total of 22,106 statistical reporting units globally, of which 56 were national, 2,299 were first-level subnational disaggregation (for example, US state level), and 19,751 were second-level (for example, US county level) reporting units. SPAM reports 24,507 statistical units of which 251 were national, 2,758 were first level, and 21,498 were second level. SPAM focused its data collection efforts particularly in developing countries. For example, in Africa the M3 and SPAM datasets were developed using around 300 and 4,150 second-level statistical reporting units, respectively. The GAEZ model uses data from FAOSTAT to constrain the model at a national level and, similar to MIRCA, uses the M3 subnational statistics for select crops in countries that had subnational statistics covering more than 50 percent of the country.Those models that distinguish between rainfed and irrigated cultivation-MIRCA, SPAM, and GAEZall use GMIA version 4.0 released in 2007 (Siebert et al. 2005) to identify the location and area intensity of irrigated production. However, the MIRCA and SPAM teams compiled information in the national and subnational shares of different production systems and cropping intensities independently. MIRCA and SPAM both draw on FAO's AQUASTAT and national databases for gaining greater insights into national and crop-specific irrigation extents and practices, but MIRCA relies on a richer collection of national data, including a more complete collection of national/subnational crop calendars and cropping intensities (in part because the goal of MIRCA is to produce monthly and not annual crop distribution maps). In contrast to MIRCA and SPAM, GAEZ relies on International Institute for Applied Systems Analysis AEZ data for information about cultivation intensity of irrigated crops.SPAM and GAEZ incorporate datasets beyond those used by M3 and MIRCA as a means of differentiating between production levels within cropping systems. The SPAM approach requires additional sets of data because it attempts further disaggregation of its rainfed production statistics among commercial and subsistence categories, and bases its approach to distribution of individual crops within the cropland extent on agronomic, economic, and demographic principles and assumptions. These include crop area and production shares among irrigated production and large-scale/commercial and smallholder rainfed production, the spatial differences in the biophysical suitability of individual crops for irrigated and rainfed (commercial and subsistence) production, and estimates of the spatial patterns of population density as well as crop prices. GAEZ similarly divides the land into land-use types to reflect variable management and input conditions. Data used to differentiate among land-use types reflect the specific requirements of each and include road infrastructure, livestock density, population density, and distance to market. Table 4.1 reflects the overlapping and separate ancillary datasets used in SPAM and GAEZ.In addition to available ancillary datasets, SPAM leverages the international network and field presence of CGIAR to undergo a systemic validation process. The feedback from this validation is used to inform future model simulations. This process is unique to the SPAM model.Each model studied produced spatially explicit cropping system maps on the same 5-arc-minute grid; however, comparing those grids directly (pixel-wise comparison) may produce artificially inflated disagreement between products. Using a pixel-specific approach fails to account for the spatial dimension of the data and implicitly assumes each pixel to be a result independent from any neighboring pixels.Each product was therefore assessed using methods that incorporated the spatial dimension of the data in biophysically and mathematically meaningful ways. As a means of accounting for the biophysical evolution of crops and cropland by growing region while still allowing for methodological differences in crop distribution, the sum of crops or cropland for each product was compared by latitude. This analysis of each product provides a biophysically meaningful broad-brush-stroke supplement to the subsequent evaluation of the distribution of crops and cropland in both spatial dimensions. To compare the twodimensional distribution of each product to one another, a Gaussian filter with a kernel density of three standard deviations was applied to the results of each product prior to a pixel-wise comparison. Preprocessing the data using a filter expands the analysis to incorporate neighboring pixels. The kernel density for the Gaussian filters-that is, the number of neighboring pixels to consider -was chosen following a sensitivity analysis using a kernel density of one, two, three, and four standard deviations. The results of the sensitivity analysis and full documentation of the implementation of the Gaussian filter are documented in Appendix B.The pixel-wise, by-latitude, and Gaussian filter analyses were each applied to assess the cropland extent, the harvested area, and the yield for each product. SPAM, M3, and MIRCA all rely on the Ramankutty cropland, while GAEZ has developed its own cropland extent. The cropland extent delineates the domain to which each product's downscaling approaches are applied and therefore represents an upper limit to the agreement between products. Each model produced maps of harvested area for wheat, rice, and maize. But only M3, GAEZ, and SPAM produced maps of yield. The results of each comparison are described in following sections.Although both Ramankutty et al. (2008) and GAEZ use the GLC2000 land cover dataset as one input to the definition of cropland extent, Ramankutty et al. blend GLC2000 with remote-sensing observations from the Boston University MODIS dataset while GAEZ supplements the GLC2000 data with independent information on the extent of protected areas, forests, and agricultural extent (Table 4.1) (Ramankutty et al. 2008;FAO 2001;WDPA 2009;Friedl et al. 2002). The methodological differences in delineating cropland extent result in distributions of cropland that broadly resemble one another but that differ significantly over select regions. By latitude, the two cropland extents largely agree over the majority of potential cropland-north of 15° N-but the distributions contain significant discrepancies further to the south. These discrepancies are particularly pronounced in the ranges of 5° to 15° N, 10° to 20° S, and 35° to 40° S (Figure 5.1).Source: Author's calculations.The pixel-wise and Gaussian filter analyses provide greater detail on the source of the discrepancies identified in Figure 5.1. The GAEZ and Ramankutty cropland extent maps largely agree in Europe, southern Africa, East Africa, and through much of China. The products significantly disagree in the Great Plains of North America, West Africa, Southeast Australia, India, and Southeast South America (see Figure 5.2). These differences, which occur in the first step of the crop distribution process, propagate through each model to underpin the difference between products of those models using the Ramankutty et al. (2008) versus the GAEZ cropland, as discussed in the following sections. The differences in the harvested area of wheat between GAEZ and the products that use the Ramankutty cropland extent (Figure 5.3) largely mirror the differences in cropland extent depicted in Figure 5.2. The differences in the harvested area of wheat for Australia, the United States, and Russia are nearly identical to those same differences in cropland extent. The lack of disagreement between products in Africa and Brazil despite large discrepancies in cropland extent reflects how little wheat is grown in those areas, and is not an indication of model agreement (Figure 5.4). Differences between the M3 and MIRCA products are minor, which is to be expected given that MIRCA uses M3 output directly as the model's input (Table 4.1 and Figure 5.4, panel e). The difference between SPAM and M3 (and MIRCA) is much larger, in particular in Europe, North India, and coastal China. The large difference is a reflection of their different downscaling methodology and the different subnational data collections.Source: Author's calculations. Note: Histograms in each panel display the normalized percent of pixels as a function of harvested area, y-axis limits (0, 50%),x-axis limits (-5000, 5000) hectares.The estimated yields from M3, GAEZ, and SPAM (MIRCA does not produce estimates of yields) matched less well than did the harvested areas from each product. GAEZ predicted larger yields at higher latitudes, while M3 predicted a far larger wheat yield in the tropics than either SPAM or GAEZ (Figure 5.3). Judging from the spatial distribution, the anomalous mid-latitude wheat yield in M3 is predominantly in Africa, where M3 is alone in predicting significant yields over much of the continent (Figure 5.5). Similarly, in South America, M3 predicts more wheat across much of the continent than either of the other two models. GAEZ predicts wheat at much higher latitudes than SPAM or M3, stretching up into Canada and northern Europe. GAEZ also predicts concentrated yields in central/eastern China that neither of the other products predict.Source: Author's calculations. Note: Histograms in each panel display the normalized percent of pixels as a function of yield, y-axis limits (0, 50%), x-axis limits (-20, 20) metric tons/hectare.The harvested areas of rice appear to be less influenced by discrepancies in the cropland extent products and instead differ as a function of downscaling method or input data. The spatial characteristics of the differences between products using Ramankutty versus GAEZ cropland extent do not mirror the differences in cropland extent, which reflects the fact that cropland extent of GAEZ and Ramankutty agree relatively well in the major rice-producing areas of the world when compared to those areas that produce wheat (Figures 5.2 and 5.4).Evaluating the harvested areas of rice by latitude for each product reveals that MIRCA predicts significantly more harvested area for rice north of 30° N than do the other products (Figure 5.6). This discrepancy may stem from the fact that the MIRCA method of crop distribution does not consider biophysical limitations, or it may reflect differences in the input statistical crop yields at a subnational level given that the above-average estimation by MIRCA appears to be concentrated in eastern China (Figure 5.7).With the exception of MIRCA's large harvested area in eastern China, the relation between GAEZ and each of the products that use the Ramankutty cropland extent (M3, MIRCA, and SPAM) is nearly identical (Figure 5.7), which may indicate that all three use similar subnational rice data in China. M3, MIRCA, and SPAM differ from GAEZ in their distribution of rice within India: GAEZ distributes more rice area to the southwest while other products distribute more rice to the northeast.The rice yields match even less well than did the wheat yields, showing significant discrepancies between the products over all latitudes (Figure 5.6). M3 again predicts yields higher than do SPAM and GAEZ over most latitudes. Similar to the wheat yields, M3 consistently predicts greater yields in Africa than either of the other products as well as significantly higher yields in western South America and in north-central Asia (Figure 5.8). The differences between M3 and SPAM are most pronounced in Europe and Asia. SPAM predicts consistently higher yields across all of Europe while GAEZ predicts higher yields in India, China, and Southeast Asia.   As with the harvested area of wheat, the discrepancies in the harvested areas of maize relate closely to differences in the cropland extent products in many regions. This relation is particularly apparent in South America, Africa, and North America (Figure 5.9). This suggests that once again, the underlying cropland extent dataset has a large effect on the final distribution of crops. Source: Author's calculations. Note: Histograms in each panel display the normalized percent of pixels as a function of harvested area, y-axis limits (0, 35%),x-axis limits (-5000, 5000)The maize yields show consistently different patterns but broadly agree on the distribution by latitude. M3 does not display consistently higher yields, but does so in the extra-tropical latitudes (Figure 5.10). Those differences largely stem from maize yields in Chile, South Africa, Canada, and northern Europe/Russia (Figure 5.11). M3 also predicts significantly higher yields than either SPAM or GAEZ in the eastern United States.  This paper explores and quantifies the systematic similarities and differences between four major global cropping systems modeling frameworks and their products. While the models have some similarities (for example, use the same input data) and interdependence (for example, MIRCA builds upon M3 and both GAEZ and SPAM use crop suitability), the modeling methods and their final products vary considerably. We found that the differences between the Ramankutty cropland extent and the GAEZ cropland extent manifest themselves in the final products, as demonstrated by the consistent discrepancies between the GAEZ harvested area products and the Ramankutty harvested area products (those of M3, MIRCA, and SPAM). When considering each harvested area product, regardless of the specific crop, the M3 and MIRCA datasets produce the most similar harvested area maps, which is to be expected because MIRCA uses M3 as a starting point. Furthermore, these models take the least complex modeling approach to distributing the statistical cropping data. Finally, we conclude that the harvested area products matched one another more closely than did the yield products, which differed significantly even in basic spatial patterns.There are many reasons why the differences among SPAM, M3, MIRCA, and GAEZ are generally large. As this paper demonstrates, the most important reason lies in the input data and the methodologies used in the four models. M3 relies heavily on cropland extent while MIRCA relies mainly on the irrigated area and water use statistics of FAO's AQUASTAT and GMIA. The SPAM and GAEZ approaches synthesize the various data sources (satellite-based land cover, ground-based data, and modeling results) in an attempt to best estimate the crop production distribution. Although each model has its own weaknesses and strengths, our past experiences (for example, You and Wood 2006) demonstrate that relying on only one input layer alone (either cropland, crop suitability, or irrigated area) may work under certain circumstances but may not always be sufficient. Conversely, more input data and increasingly complex modeling does not necessarily lead to better or more accurate results.In this paper we evaluate the extent to which the four models agree, and explore the root causes of their discrepancies. As the true crop distribution is unknown, we could not judge which product is more accurate than the other. Rather, such comparison gives the reader some sense of discrepancy among the four products and provides information for users to make a knowledgeable choice.Moving forward, room certainly exists for the four modeling teams to collaborate and develop some community of practice. This has started in a recent workshop convened by IFPRI, but more needs to be done. In the meantime, the four products provide users with a range of alternative approaches to modeling cropping systems. Potential users of these cropping system products need to understand the input data and modeling differences prior to choosing the model best fit for their purposes. Although each model approaches the same problem using many of the same underlying datasets, the methods employed and assumptions made in each product significantly affect the final cropping system map.In each grid cell that had agricultural inventory data, the map of crop area was calculated as follows:where \uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53 \uD835\uDC56\uD835\uDC56 is the harvested area of a specific crop in pixel i, \uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53 \uD835\uDC56\uD835\uDC56 is the fraction of pixel i designated as cropland, \uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53 \uD835\uDC58\uD835\uDC58 is the harvested area of a specific crop in statistical reporting unit k, and \uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53\uD835\uDC53 \uD835\uDC58\uD835\uDC58 is the amount of cropland in statistical reporting unit k. Yield was distributed uniformly across each grid cell as equivalent to the yield reported in the statistical reporting unit as a whole.MIRCA primarily reconciles the differences between the Siebert et al. ( 2005) dataset of areas equipped for irrigation (AEI), the cropland extent (CE) of Ramankutty et al. (2008), and the harvested area (HA) maps of the M3 dataset (Monfreda, Ramankutty, and Foley 2008) to provide a monthly cropping map for irrigated and rainfed crops. Table A.1 outlines the priorities used to reconcile inconsistencies between the datasets. MIRCA first produces a condensed crop calendar of harvested area for each subcrop c and Statistical Reporting Unit (SRU) k (HAcccc,k). A subcrop is used to represent multicropping systems or different subgroups of a crop that grow at different points in the year. For a complete description of how the condensed cropping calendar is produced, see Portmann et al. (2008).Area equipped for irrigation (Siebert et al. 2005) In each month and grid cell, the sum of crop-specific areas is lower than or equal to the area equipped for irrigation.Cropland extent (Ramankutty et al. 2008) In each grid cell and month, the sum of crop-specific irrigated and rainfed areas is lower than or equal to the cropland extent.Harvested crop area (Monfreda, Ramankutty, and Foley 2008) In each grid cell and for each crop class, the annual sum of the irrigated and rainfed harvested crop area is equal to the total harvested area of the specific crop.Source: Author's compilation.The MIRCA process is a production-system-specific cell-wise approach to disaggregating harvested area by month. Area equipped for irrigation and cropland extent both include fallow land in their definition, so these classes need not be used completely so long as the annual harvested area is disaggregated and designated as either rainfed or irrigated. Area equipped for irrigation is prioritized over cropland extent and harvested area in the following process:             ","tokenCount":"5461","images":["1688536799_1_1.png","1688536799_15_1.png","1688536799_15_2.png","1688536799_16_1.png","1688536799_16_2.png","1688536799_17_1.png","1688536799_18_1.png","1688536799_19_1.png","1688536799_19_2.png","1688536799_20_1.png","1688536799_21_1.png","1688536799_22_1.png","1688536799_23_1.png","1688536799_23_2.png","1688536799_23_3.png","1688536799_29_1.png","1688536799_30_1.png","1688536799_31_1.png","1688536799_31_2.png","1688536799_32_1.png","1688536799_32_2.png","1688536799_33_1.png","1688536799_33_2.png","1688536799_34_1.png","1688536799_34_2.png","1688536799_35_1.png","1688536799_35_2.png","1688536799_36_1.png","1688536799_36_2.png","1688536799_37_1.png","1688536799_37_2.png"],"tables":["1688536799_1_1.json","1688536799_2_1.json","1688536799_3_1.json","1688536799_4_1.json","1688536799_5_1.json","1688536799_6_1.json","1688536799_7_1.json","1688536799_8_1.json","1688536799_9_1.json","1688536799_10_1.json","1688536799_11_1.json","1688536799_12_1.json","1688536799_13_1.json","1688536799_14_1.json","1688536799_15_1.json","1688536799_16_1.json","1688536799_17_1.json","1688536799_18_1.json","1688536799_19_1.json","1688536799_20_1.json","1688536799_21_1.json","1688536799_22_1.json","1688536799_23_1.json","1688536799_24_1.json","1688536799_25_1.json","1688536799_26_1.json","1688536799_27_1.json","1688536799_28_1.json","1688536799_29_1.json","1688536799_30_1.json","1688536799_31_1.json","1688536799_32_1.json","1688536799_33_1.json","1688536799_34_1.json","1688536799_35_1.json","1688536799_36_1.json","1688536799_37_1.json","1688536799_38_1.json","1688536799_39_1.json","1688536799_40_1.json","1688536799_41_1.json","1688536799_42_1.json","1688536799_43_1.json","1688536799_44_1.json"]}

data/part_2/0458798944.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"0f7d46b9a7c855ca0969653776fda9ae","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/0c8b84e2-6888-44fc-9e13-d514fa85bf02/retrieve","description":"Demographic transition due to population aging is an emerging trend throughout the developing world, and it is especially acute in China, which has undergone demographic transition more rapidly than have most industrial economies. This paper quantifies the distributional effects in the context of demographic transition using an integrated recursive dynamic computable general equilibrium model with top-down behavioral microsimulation. The results of the poverty and inequality index indicate that population aging has a negative impact on the reduction of poverty while its impact is positive with regard to equality. In addition, elderly rural households are experiencing the most serious poverty, and their inequality problems compared with other household groups and within group inequality worsens with demographic transition. These findings not only advance the previous literature but also deserve particular attention from Chinese policy makers.","id":"-371608326"},"keywords":["demographic transition","poverty","inequality","CGE model JEL codes: J11","D58","D63","D10"],"sieverID":"27e3ba09-8a5b-48f3-8090-283bae72e053","pagecount":"32","content":"established in 1975, provides evidence-based policy solutions to sustainably end hunger and malnutrition, and reduce poverty. The institute conducts research, communicates results, optimizes partnerships, and builds capacity to ensure sustainable food production, promote healthy food systems, improve markets and trade, transform agriculture, build resilience, and strengthen institutions and governance. Gender is considered in all of the institute's work. IFPRI collaborates with partners around the world, including development implementers, public institutions, the private sector, and farmers' organizations, to ensure that local, national, regional, and global food policies are based on evidence. IFPRI is a member of the CGIAR Consortium.As the world's most populous country, China is experiencing an increasing number of young people and declining fertility, which boost economic productivity, and this is taken as a demographic dividend. China's abundance of cheap labor has made it internationally competitive in many low-cost, laborintensive manufacturing sectors (Wang, Mayes, and Wan 2005). Scholars estimate that the demographic dividend accounts for one-fourth of China's economic growth since 1978 (Wang and Mason 2004;Cai 2009). Simultaneously, the standard of living in China has improved significantly, and poverty has been substantially reduced.However, due to the combined influence of the strict implementation of the one-child policy and socioeconomic development, China has completed a demographic transition from the interim pattern (population with low death rate, high birth rate, and high growth rate) to the final pattern (population with low death rate, low birth rate, and low growth rate) within approximately 30 years, a short period of time when compared to the transitions of most developed countries (Cai and Wang 2010). According to the sixth national census in 2010, released by the National Bureau of Statistics, the proportion of the population older than 60 is 13.32 percent, and the population older than 65 accounted for more than 8.92 percent of the total population in China. China is positioned to undergo a period of rapid aging, with the proportion of the old population (older than 65 years old) reaching 16.52 percent by 2030. In addition, the total population is predicted to arrive at its highest point, which is 1.395 billion, in 2026, and the total labor force to decline by 2015 World Population Prospects (United Nations 2013). These projections imply that China's demographic dividend will soon be exhausted (Wang and Mason 2004) and will turn into a demographic deficit with important adverse economic consequences (Peng and Mai 2008). During the same period, the distribution of income in China has become much more unequal between rural and urban areas, coastal regions and inland regions, men and women, and different industry sectors. For example, in 1990, the income per capita for urban households was 1,516 yuan, which was 1.53 times that of rural households. However, this urban and rural inequality increased to 2.75 in 2014. The Gini coefficient was only 0.16 before China's reform and opening up policies were implemented in 1978. According to the National Bureau of Statistics' recent report, the Gini coefficient was 0.49 in 2008 and 0.47 in 2014, both of which crossed the international warning line, which implies that China's inequality is becoming dangerously severe. Is there any relationship between demographic transition and income distribution? What's the impact of population aging on poverty and inequality? Future inequality in the context of demographic transition in the middle-and long-run period will be a salient issue in the near future.Due to the deterioration of income inequality that follows rapid economic growth and population aging, population transition with income distribution research became popular in the 1990s for developed countries. However, the general trend of such research is inconclusive as there is evidence supporting both a positive and a negative relationship. On one hand, some empirical research indicates that population aging worsens income inequality (Ohtake and Saito 1998;Deaton and Paxson 1994;Lan, Wei, and Wu 2014;Lin, Lahiri, and Hsu 2015). Deaton and Paxson (1995) analyze the relationship between population aging and inequality and conclude that population aging leads to greater inequality for both within-cohort inequality and between-cohort inequality. Their results fit the conditions of the Taiwanese economy and also predict increases in inequality in other fast-growing Asian countries. Followed by Deaton and Paxson (1995), Ohtake and Saito (1998) analyze how consumption inequality within a fixed cohort grows with age, using Japanese household microdata. Their results show that half of the rapid increase in economywide consumption inequality during the 1980s was caused by population aging. Miyazawa's (2006) analytical results reveal that the population's aging enlarges the inequality between different generations and inequality within a generation and finally expands total inequality. Lan, Wei, and Wu's (2014) empirical study uses a panel dataset from 76 countries and regions from 1970 and 2011 and finds that the population's aging does increase income inequality significantly. However, on the other hand, some studies find that aging may have a negligible effect on inequality (Jantti 1997;Bishop, Formby, and Smith 1997;Barrett, Crossley, and Worswick 2000;Schultz 1997) or even a positive effect on equality. Chu and Jiang (1997) examine the effects of age structure on family income using the Gini coefficient, and the results demonstrate that changes in Taiwan's demography reduced inequality in family earnings between 1980 and 1990. By applying the Overlapping Families Model, Lee and Mason (2003) find that population aging had little effect on income inequality. Morley (1981) expands Paglin's (1975) method, which decomposes inequality by age structure and finds that a younger age structure would widen income inequality while countries with serious population aging are less unequal. Goldstein and Lee (2016) try to find out the impact of different factors related to population aging on inequality and show that the changing age structure is found to have a small effect on aggregate inequality.As the issue of population aging emerged at the beginning of the 21st century for developing countries, China was among the few developing countries that managed to transform into a society characterized by an aging population. These profound demographic changes, however, raise concern about the sustainability of China's economic growth (Cai 2009;Cai and Wang 2005). For example, scholars and Chinese government officials worry that the looming demographic challenge may undermine China's ability to grow rich before its population grows old (Jackson and Howe 2004;Cai and Lu 2013). There has been a great deal of theoretical and empirical research on the relationship between demographic transition and economic growth in China. Generally speaking, the literature indicates that population aging generates negative economywide effects that will slow economic growth. Peng and Fausten's (2006) simulation results show that the labor force decline caused by population aging will decelerate China's economic growth rate by 2 percentage points annually during the 2020s and by 3 percentage points annually during the 2040s. Cai and Lu (2013) estimate the average annual growth rate of potential output to be 7.2 percent during the 12th five-year plan period and 6.1 percent during the 13th five-year plan period.However, there are only a few existing studies on the relationship between demographic transition and income distribution that focus on the developing world. Research on China's demographic transition and income distribution has begun to emerge in recent years, and the results regarding this relationship are still not clear. Some empirical research indicates that population aging will expand income inequality. Zhong (2011) investigates the relationship between population aging and income inequality in rural China by using five years' panel data from the China Health and Nutrition Survey and argues that a significant portion of the sharp increase of income inequality at the beginning of this decade can be attributed to demographic change. Dong, Wei, and Tang (2012) employ provincial-level panel data between 1996 and 2009 and confirm that population aging positively and significantly affects income inequality. But on the other hand, research finds that demographic transition accounts for only a tiny fraction of inequality. Cai, Chen, and Zhou (2010) analyze urban household survey data between 1992 and 2003 by employing a regression-based inequality decomposition and find that age has a negligible effect on inequality. Qu and Zhao (2008) investigate the relationship between inequality and population aging in rural China using the life cycle model with three years' rural household surveys from the China Household Income Project (CHIP). The results show that population aging plays only a small role in the inequality increase in rural China. Guo, Lu, and Jiang (2014) adopt the advanced decomposition method based on Ohtake and Saito (1998) by using the Urban Household Survey data from 1988 to 2009. Their research finds that the population aging effect explains only 16.33 percent of the total income inequality and this effect keeps decreasing. There are few other studies that have comprehensively analyzed the relationship between income distribution and problems associated with population aging in China.The existing studies use mainly panel data and decompose inequality by age to simulate the contribution of aging on inequality. However, demographic transition is a long process; the economic and social impact of population aging has not yet totally emerged throughout China. The impact of any demographic change can be split into supply effects (consequences for labor and capital) and demand effects (consequences for public and private consumption, international trade, and domestic and foreign investment) (Poot 2008). A review of the literature indicates that demographic transition has economic and social impacts related to changes both in the supply of labor and in household consumption and investment demand. Subsequently, it tends to affect household income and expenditure via two channels:(1) the direct channel, which affects individual employment and wages, and (2) the indirect channel, which affects the sensitivity of commodities' supply and price due to productivity changes. Therefore, a computable general equilibrium (CGE) model, which is a type of economic model that uses economic data to estimate how an economy might react to changes in policy, technology, or other external factors, can be used to capture the macro impacts on factor price and household income in the context of demographic change. However, it is not enough to measure the micro-level impact using only a CGE model. The macro-micro modeling frameworks that integrate CGE models with microeconomic models have proved useful in capturing the effect of macroshocks to microdistribution.This paper aims to examine the distributional effects of the demographic transition that is under way in China. The rest of the paper is organized as follows: Section 2 provides a methodological review and introduces the methodological framework for this study. Section 3 and Section 4 introduce the models and datasets. The macro model, micro model, and model linkages for this study will be set out in these two sections, and a detailed specification will be provided. Section 5 contains the empirical results on poverty and inequality, and finally, we present the conclusions of the paper and propose the policy implications arising from the results in the last section.In this section, we first introduce the recent developments in macro-micro modeling methodology and then describe the framework of this study. Dervis, de Melo, and Robinson (1982) and Gunning (1983) were the first to analyze income distribution by connecting a CGE model with microdata. After that, a series of various approaches were developed with different methods on the linkage between CGE models and micro-household data. These recent developments in macro-micro modeling frameworks have been proved to be helpful in capturing the impact of macro shocks to microdistribution.As concluded by Debowicz (2012), there are mainly two channels that can link CGE models with microdata, including integrated links and layered links. Integrated links integrate the selected information about representative household groups into a macro-CGE model; they can be further divided into representative household integrated (RHI) (such as the studies in Dervis, de Melo, and Robinson 1982;Colatei and Round 2001;Agénor, Izquierdo, and Fofack 2001) and multihousehold integrated (MHI) (such as the studies in Decaluwé, Dumont, and Savard 1999;Cogneau and Robilliard 2000;Boccanfuso, Decaluwé, and Savard 2008). From the empirical research, the RHI link method can integrate the microdata into the macro model, and it works well when there is a small number of groups in the model. However, it does not allow researchers to take into account within-group changes in income distribution because the same groups in the macro model are assumed to be identified. The MHI link method can solve the problem of the RHI link method; however, it accounts for the majority of the micro-household data in the CGE model that makes the CGE model hard to converge when running such a large-scale CGE model (Chen and Ravallion 2004). What's more, it is difficult to balance the household income and expenditure account and to calibrate between micro-household data and macro data (Savard 2005;Boccanfuso, Decaluwé, and Savard 2008;Rutherford, Tarr, and Shepotylo 2005).As such, the layered link method is more reasonable and popular. The layered link method involves connecting a separate micro model with a CGE model, and it can be further split into behaviorlayered link and non-behavior-layered link. For the non-behavior-layering approaches, which assume that all the households in a group are affected in the same way by changes in the macro variables from the macro model, this can eliminate the within-group differences induced by individual heterogeneity. The behavior-layered link method connects a CGE model with a micro model that simulates the microbehavioral model separately. The behavior-layered link method can be further divided into topdown microsimulation and top-down (such as Bourguignon, Robilliard, and Robinson 2003;Debowicz 2012) and bottom-up microsimulation links (such as Savard 2005Savard , 2010) ) due to the reflected difference. The former links the CGE model results to the microsimulation model, while the latter can further deliver the feedback from the microsimulation model back to the CGE model. However, the latter is much more difficult as the convergence cannot be guaranteed and needs to be confirmed at each stage.Households receive income from factor endowments (labor income and capital) and transfers from the government. The CGE model is used to get the results of both the factor market price changes and the household income changes due to China's demographic transition. Therefore, the macro-micro linkage via household income or through the factor markets is helpful. Though the layered behavior link method seems better as described below, additional household behavior models are required for which not all data are suitable for these methods (Wang, Mi, and Liang 2015). Taking into consideration the tradeoff between the advantages and disadvantages of the micro-macro linkage method, data availability, and technical feasibility, we attempt both the behavior-layered link and the non-behavior-layered link approaches to connect the CGE model's macro results to microdata.In our research, we first solve the dynamic CGE model with demographic transition to get the results on a vector of changes for commodity prices, labor price, nonlabor price, and government transfer. Then, (1) we link the household labor income, which is the majority part of the household income and remains at the individual level in this study, via the top-down behavior-layered link; (2) we link the nonlabor income changes, which are difficult to estimate via the micro-household regression model, through a top-down non-behavior-layered link; and (3) we deal with the micro level's household expenditure changes based on the consumption structure in each commodity sector and link the macro commodity price changes from the results of the CGE model via non-behavior-layered connection. Finally, we adopt poverty and inequality indexes to evaluate the distributional impact of the shock to the demographic transition with the results from the microsimulation model (Figure 2.1). In this section, we introduce the specification of the macro CGE model and micro model, respectively, as well as the setting of parameters and datasets.The dynamic CGE model adopted in this research is developed by the International Food Policy Research Institute and is an extension of the International Food Policy Research Institute's static standard model that was developed by Lofgren, Harris, and Robinson (2002). The model is a recursive dynamic model that is solved one period at a time, which indicates that the behavior of the model's institutions is based on adaptive expectations rather than on the forward-looking expectations that underlie intertemporal optimization models (Thurlow 2012). In this paper, we will briefly introduce the demographic-transitionrelated aspects of the model, such as labor factor, production function, household income and expenditure, and data settings.Labor and capital factors are included as value added of factor input in our study. The demographic transition is expected to affect the size and structure of the labor supply. To consider demographic transition within the larger context of real economic development, four types of demographic changes are introduced for the basic scenario: population aging, gender shifts, urbanization, and human capital structure changes. Therefore, besides the population aging issue, labor is divided into regions and gender as well as rural-urban regions, with eight categories in all. Together with the capital factor, which is considered a factor value-added input in the sector production function, there are nine factor segments.In this model, based on the producer's profit maximization principle, the value-added input is estimated by the constant elasticity of substitution (CES) function with different factor input ratio. The factor value-added input in sector a can be defined in the following CES function in equation 1:where, va a α is the efficiency parameter in this CES function and va f a δ denotes the share parameter for factor f input in sector a.f a QF is the factor demand of f in the production activity, and va a ρ is the exponent in the CES function. It is the variant of factor elasticity substitution, which can be denoted as.Therefore, equation 1 can measure the value-added input changes in each sector a due to demographic transition. Consistent with the micro-household survey data and taking into consideration the household's work and consumption sectors, this model classifies production activities into 12 sectors in all, including (1) agriculture, forestry, animal husbandry, and fishery; (2) mining; (3) manufacture of foods, beverage, and tobacco; (4) manufacture of nondurable consumer goods; (5) other manufacture; (6) power, water, gas, and electricity processing industry; (7) construction; (8) transport, storage, post, and information; (9) wholesale, retail trades, and hotel; (10) financial intermediation; (11) real estate, leasing, and business services; and (12) other services. The parameters of factor elasticity substitution for the 12 sectors are based on Zhang, Wang and Chen (2014), who list 17 sectors in all. The efficiency parameter va a α can be taken as total factor productivity, which is enhanced with changes in human capital accumulation. In this paper, we assume that the increase of the ratio for the high-skilled labor force will improve sector total factor productivity. Finally, the total sector production for a sector is calculated using both the value-added input and the aggregate intermediate input, which is expressed by a Leontief function with a fixed share of disaggregated intermediate inputs.Accordingly, the factor price is based on the optimal principle of cost minimization that the factor cost is subjected to equation 1 and calculated with the Lagrange method's first-order condition. Therefore, the price of factor f (eight types of labor factor and one capital factor) in sector a can be measured by equation 2, where the left side is the marginal cost of each factor and the right side can be taken as production marginal revenue. W f is the average factor price, and WFDIST fa is the wage distortion parameter for factor f in sector a. It is set exogenously.( ) ( )The Definition of Household Income and ExpenditureThe household is disaggregated into rural household and urban household in our model. Household income is from factor reward (labor factor and capital factor included) and transfers income from both government and nongovernment institutions. Equation 3 defines the total factor income for a household. It is the sum of all the factor income, which is the average factor price (WF f ) with wage distortion parameter times factor input (QF fa ). The demographic transition would influence the labor factor income, and therefore the total household income would be changed accordingly.The total household income for individual i is the sum of factor incomes (defined in equation 3), transfers from other nongovernment institutions, transfers from the government (indexed to the consumer price index [CPI]), and transfers from the rest of the world (indexed to the exchange rate). Equation 4shows the household total income. ' ' 'The transfers from domestic nongovernment institution i' to household i are paid as fixed shares of total institutional incomes net of direct taxes and savings. Equation 5 defines the transfer from institution i' to household i. MPS is the marginal saving propensity, tins is the direct tax rate, and shii ii' is the share of transfer rate.(5)Finally, the total household disposable income is defined as the income that remains after direct taxes, savings, and transfers to other domestic nongovernment institutions. It is shown in equation 6.( )The household's commodities expenditure is calculated based on the linear expenditure system (LES) function (equation 7). The LES function is calculated by assuming that each household maximized a Stone-Geary utility function subject to a consumption expenditure constraint and captures its first-order conditions. There is a basic survival consumption for each commodity within a household in the LES function; it is measured by the share of this consumption ( m ch γ )times the price of each commodity.Besides the basic consumption, there is an extra consumption for each commodity with the share of m ch β , which varies for different types of households. Therefore, the demographic transition will change the total consumption structure for all the commodities. ' ' 'The household consumption parameters include the commodity demand elasticity in each commodity sector and the basic consumption parameter m ch γ . For the commodity demand elasticity in the 12 sectors in our study, we adopt the calculation by Xie (2008), who employs both the Bayesian rules and generalized maximum entropy methods to estimate the substitution elasticity of 14 sectors based on the panel data throughout China's 31 provinces with household income and expenditure. Due to the difficulty of calculating the basic consumption parameter, previous research mainly uses the Frisch parameter to measure this parameter. Frisch's (1959) research finds that the value of the Frisch parameter for poor households is -10.0, whereas it is -0.7 for rich households. Based on these results and empirical experience, we set the Frisch parameter for rural and urban households at -2.5 and -2.0, respectively. Besides affecting labor supply changes, the amount of population change will affect household consumption and expenditure. Therefore, it will significantly affect commodity sales, saving, and total investment changes. In our study, population growth is exogenously imposed in the model based on the United Nations' World Population Prospects (United Nations 2013) and extends over a time period from 2010 to 2030 to simulate population changes. Despite changes in household population, the model assumes the marginal rate of consumption for commodities to be unchanged, which implies that new consumers have the same preferences as existing consumers.A social accounting matrix (SAM), which is a comprehensive and economywide data framework, serves as the database of the CGE model (Lofgren et al. 2002). SAM is built using the latest input-output (2010), which is an extension table of the input-output table for 2007; different kinds of yearbooks and the microhousehold survey data are employed for SAM. There are 12 activities and commodities sectors, 8 segments of the labor force, and two types of households in SAM. To overcome the difficulty of collecting data for 8 types of labor factor inputs as value added on 12 sectors, we employ the household survey data with an econometrics model to calculate the ratio of the factor input distribution on 12 sectors. The wage difference between the 8 labor segments is estimated by a wage regression with region dummy, skill dummy, gender dummy, and their cross-variable dummy as well as other individual and household characteristics for independent variables. After getting the predicted slopes of these dummy variables, the marginal wage difference of the 8 types of the labor force can be calculated from the slopes. Then the wage differences for different labor forces are induced based on the average wages by sector, which can be collected from the China Stock Market and Accounting Research Database. Besides the parameters in CES and LES function that have been introduced in the previous sector, the elasticity in the constant elasticity transformation function and Armingtion function are mainly from Zhai and Hertel's (2005) results, which provides the elasticity of 53 sectors.As mentioned before, to consider demographic transition within a context of real economic development, four types of demographic changes are introduced for the basic scenario. (1) First is population age structure change, which is the central focus of this paper. (2) Second is population gender structure change. The first two will be simulated based on World Population Prospects by the United Nations. (3) Third is human capital accumulation change. This is represented by the proportional changes for labor force with tertiary education to total labor force aging population. The share of the labor force with tertiary education was 19.52 percent and 2.63 percent, respectively, for rural and urban individuals in 2010, as reported by the China Statistics Yearbook (NBS 2011). On the basis of China's past growth rate for the share of the labor force with tertiary education as well as the current situation for developed countries, the simulation for the tertiary education share change is assumed to double for rural individuals and increase 1.5 times for urban individuals in 2030. (4) The population spatial change with urbanization. China has the largest urban population in the world, with 749 million urban dwellers in 2014, accounting for 54.77 percent of China's total population, and it is supposed to reach 68.7 percent by 2030 and 70.0 percent by 2033 (United Nations 2014). All four of these demographic transitions are linked to the labor factor supply and population changes within the periods of the CGE model's dynamic component.A comparative scenario without population aging is used for comparing the base scenario to determine the real impact of population aging. In this scenario, the population age structure change from 2010 to 2030 follows the structure change from 1990 to 2010 and holds the other three demographic transitions constant. Figure 3.1 shows that the general trend of demographic transition for each of the two scenarios is similar, which shows that the population growth rate is slowing due to China's birth control policy since the 1970s. However, in the basic scenario, population growth becomes negative in 2027, and the average population growth rate is 0.189 percent from 2010 to 2030. While things are different in the comparative scenario without the population-aging problem, the population growth rate slows down but remains positive, and the average growth rate remains at 0.794 percent during the same period. Therefore, the results of the scenario can provide the absolute impact of population aging by comparing the base scenario with that which integrates population aging.  Consistent with the macro model, micro-household income comprises labor income, capital income, government transfer income, and other incomes that cannot be classified into any one category. Both employed and self-employed wage income is classified as labor income that provides the majority of the income for households. The labor income is linked to the CGE results with a layered behavioral methodology that can reveal individual heterogeneity. Other nonlabor income, which cannot be estimated by an individual behavior function and may take only a small part of the income, is linked to the CGE results through a non-behavior-layering approach in which the income is changed with the same ratio within the same group segment according to the changes from the CGE model.The labor income can be represented by function 8.where \uD835\uDC4A\uD835\uDC4A \uD835\uDC56\uD835\uDC56 is the nominal labor income of working individual i and dependent \uD835\uDC4B\uD835\uDC4B \uD835\uDC56\uD835\uDC56 variable denotes the vector of the characteristic of the individual, household, and regions. \uD835\uDC4E\uD835\uDC4E \uD835\uDC60\uD835\uDC60 \uD835\uDC4E\uD835\uDC4E\uD835\uDC4E\uD835\uDC4E\uD835\uDC4E\uD835\uDC4E \uD835\uDC4F\uD835\uDC4F \uD835\uDC60\uD835\uDC60 are the intercepts and slopes in the logarithm of the wage, respectively. \uD835\uDC63\uD835\uDC63 \uD835\uDC56\uD835\uDC56 is the residual term that describes the effects of unobserved earning determinants and possibly measurement errors. The total household income can be defined as equation 9. The nonlabor income and a household specific CPI are described in equation 10 and equation 11, respectively.\uD835\uDC4C\uD835\uDC4C 0ℎ = \uD835\uDC36\uD835\uDC36\uD835\uDC4E\uD835\uDC4E\uD835\uDC36\uD835\uDC36\uD835\uDC36\uD835\uDC36 ℎ + \uD835\uDC47\uD835\uDC47\uD835\uDC47\uD835\uDC47\uD835\uDC4E\uD835\uDC4E\uD835\uDC4E\uD835\uDC4E\uD835\uDC47\uD835\uDC47 ℎ + \uD835\uDC42\uD835\uDC42\uD835\uDC42\uD835\uDC42ℎ\uD835\uDC52\uD835\uDC52\uD835\uDC47\uD835\uDC47 ℎ ； (10)where \uD835\uDC4C\uD835\uDC4C\uD835\uDC4C\uD835\uDC4C ℎ is the total household income for the members within a household. It includes the household labor income and nonlabor income. For the labor income, \uD835\uDC3C\uD835\uDC3C\uD835\uDC4A\uD835\uDC4A \uD835\uDC56\uD835\uDC56ℎ stands for a dummy variable, which denotes the individual work status (1 for work, 0 for not work). W ih is the individual labor income that is calculated in equation 8. Therefore, the sum of the total labor income within a household is ∑ \uD835\uDC4A\uD835\uDC4A \uD835\uDC56\uD835\uDC56ℎ \uD835\uDC3C\uD835\uDC3C\uD835\uDC4A\uD835\uDC4A \uD835\uDC56\uD835\uDC56ℎ \uD835\uDC56\uD835\uDC56∈ℎ . Nonlabor income (\uD835\uDC4C\uD835\uDC4C 0ℎ ), comprises the capital income, land income, transfer income, and other income, which are calculated at the household level and shown in equation 10. To compare the real household income changes as a result of the demographic transition shocks, household income is adjusted by a household-specific CPI (\uD835\uDC43\uD835\uDC43 ℎ ), which is introduced in the household income function (equation 9) following Bourguignon, Robilliard, and Robinson's (2003) method.\uD835\uDC43\uD835\uDC43 ℎ is measured in equation 11, which can eliminate the effects caused by price differences in various households and in different years. Therefore, the adjusted household income is comparable within different years and households. \uD835\uDC46\uD835\uDC46 ℎ\uD835\uDC58\uD835\uDC58 is the observed budget shares of a household's consumption for commodity k, and \uD835\uDC36\uD835\uDC36 \uD835\uDC58\uD835\uDC58 denotes the price of various consumption goods k.The parameters of the labor income function are estimated for transmitting the CGE results to the microsimulation using a behavior method. Labor income is a nonlinear function of the observed characteristics of individuals, households, and regions. Consistent with the CGE model, this labor income function is defined independently across eight labor segments, which are classified by area (rural/urban), skill (with high education level), and gender (male/female). Therefore, eight separate regressions are run to estimate the parameters (\uD835\uDC4E\uD835\uDC4E \uD835\uDC60\uD835\uDC60 \uD835\uDC4E\uD835\uDC4E\uD835\uDC4E\uD835\uDC4E\uD835\uDC4E\uD835\uDC4E \uD835\uDC4F\uD835\uDC4F \uD835\uDC60\uD835\uDC60 ) of the labor income equation for each labor market segment. The subject of these regressions is the aging labor force, which may include the working-age population who do not participate in the workforce. To correct for the possibility of sample selection problems, we use the two-step Heckman procedure, which includes an inverse Mills ratio derived from a preprobit model that estimates the work status of the individual.The household survey data we employ are from CHIP, which is carried out by the Institute of Economics, Chinese Academy of Social Sciences, with assistance from the National Bureau of Statistics. CHIP carried out the survey in 1995, 2002, and 2007. However, the question design for the income and expenditure sector in the 2002 survey is much more consistent with the requirements of our macro-CGE model. The micro-macro linkage requires that the sectors be consistent between the CGE model and microsimulation model. For example, the question design of the job sectors for the micro household in CHIP 2002 has 16 sectors, 25 sectors, and 18 sectors for urban household, migrant workers, and rural household, respectively, which is consistent with our CGE model's sector classification. After matching all the micro households' job sectors with the CGE models', 12 sectors are classified in our research. Only the 1995 and 2002 data were public when we started this study, and thus, the 2002 CHIP data are used for our research. To solve the problem that the income and consumption data in 2002 are too old, we updated the labor income data and household consumption data of 2002 to 2010 using an appropriate method that will be introduced in the following paragraph. The CHIP 2002 data were collected through a series of questionnaire-based interviews conducted in rural and urban areas and covered 22 provinces in China. There are a total of 6,835 urban households, 9,200 rural households, and 2,000 migrant households included in the survey, with a total number of 37,969 individuals. Because the sampling for migrants is based on the place of residence of migrants, migrants living in a dormitory or workplace (such as a construction site) are excluded in the sample (Zhao et al. 2010), and we merge the migrant households (which can be taken as low-skilled urban labor) with the urban households in our study.To connect the CGE results with the base year 2010, we update the labor income data of 2002 to 2010 using a similar method for connecting CGE results with microdata. Different from the method of updating the micro income data according to the corresponding proportion from the macro data (such as is performed by Zhang, Wang, and Chen 2014), the advantage of this method is that it can sustain the individual heterogeneity for the wage changes to overcome the consistency of the within-group variance. Labor income is updated based on the sectors the individual worked for, and the macro data are collected from the China Stock Market and Accounting Research Database, which has the average wages in different sectors for both 2002 and 2010. Labor is classified by sectors, and each regression income model is done to estimate the parameters of different types of labor segments. Other income is updated to 2010 based on the proportion changes between 2002 and 2010 from the data collected in the China Statistics Yearbook. The dependent variables for the labor income regression function are the observed characteristics of the individual, household, and region, which include the individual's age, the individual's education year, the sector in which the individual works, the number of children and labor force in the household, whether the individual is the household head, and finally the region (east, middle, or west of China).As introduced in the previous section, the non-behavior-linkage approach conveniently links the microhousehold income data by directly corresponding with the ratio from the CGE model's results. The following part of this section focuses more on the transmission channels from the CGE model's macrolabor factor price changes to the micro-labor income variation using the layered behavioral methodology in a top-down fashion, which can capture the distributional differences of both the within group and the between group by considering the individual's and household's heterogeneity.After solving the dynamic CGE model for a period of 20 years from 2010 to 2030 in the context of demographic transition, new labor wage changes are generated in each of the periods. From the simulation of the dynamic CGE model, in the basic scenario, the simulation results show that the income difference between rural and urban households is decreasing. For example, the annual per capita income growth rate is 3.18 percent for urban households, while it is 10.08 percent for rural households. From the primary labor factor distribution, the results show that rural labor wages are growing faster than are those for urban labor, while low-skilled labor price growth is faster than that of high-skilled labor (Table 4.1).To determine the micro level's household income and expenditure changes due to demographic transition, we need to link the macro CGE models' results to the micro household for both the basic scenario and the comparative scenario. The following section focuses more on the transmission channels for the communication from the CGE model to the labor income model. Note: BASE = base scenario; NODE = comparative scenario.The labor wage change in a specific simulation year is represented as \uD835\uDC4A\uD835\uDC4A � \uD835\uDC5F\uD835\uDC5F,\uD835\uDC60\uD835\uDC60,\uD835\uDC54\uD835\uDC54,\uD835\uDC5D\uD835\uDC5D ，; here, r, s, g, and p denote the region, skill level, gender, and time period, respectively, and there are eight segments of the labor force in all as mentioned in the previous section. The microsimulation model is applied to generate the changes for the individual's labor income, consistent with the labor wage changes from the CGE model. Let us express the set of the original macro labor wage using equation 12, which is the consistency-adjusted microdata with the sample weights for each of the labor segments. The macro targets in a specific simulation period vector are indicated in equations 13 and 14. In equation 14, \uD835\uDC4A\uD835\uDC4A � \uD835\uDC5F\uD835\uDC5F,\uD835\uDC54\uD835\uDC54,\uD835\uDC60\uD835\uDC60 is taken as the percentage change from the base year of the macro simulation. The parameter changes are assumed to be neutral with respect to the individual characteristic. So only the intercepts of the labor income function are adjusted to generate a proportional change of all the income in each of the labor segments irrespective of individual characteristics. Then, consistent adjusting involves finding a row vector \uD835\uDC65\uD835\uDC65 = \uD835\uDC4E\uD835\uDC4E \uD835\uDC5F\uD835\uDC5F,\uD835\uDC54\uD835\uDC54,\uD835\uDC60\uD835\uDC60 to be consistent with the \uD835\uDC53\uD835\uDC53 * (\uD835\uDC65\uD835\uDC65) macro target vector. Following Bourguignon, Robilliard, and Robinson's (2003) and Debowicz's (2012) research, this problem can be solved using the Newton-Raphson method, which is a root-finding algorithm that uses the first few terms of the Taylor series of a function in the vicinity of a suspected root to find successively better approximations to the root of a real-value function. This requires a Jacobian matrix (equation 15) with all the possible combinations of partial derivatives of the element for the original macro labor wage, \uD835\uDC53\uD835\uDC53(\uD835\uDC65\uD835\uDC65). A detailed discussion of the specification for this methodology can be found in Debowicz (2012).In addition, as the computation of a Jacobian matrix and the solution of a linear system in each iteration makes the Newtonian technique costly with two scenarios and spanning 20 years for the simulation periods, it would not be a good idea to simulate all 40 periods in a microsimulation model. To compare the income distribution associated with the demographic transition in a more practical manner, we choose 2010, 2015, 2020, 2025, and 2030 as the year points with two scenarios to quantify the trend of the distribution impact in our research instead.In addition, the microdata are static cross-section data that can't reflect the demographic transition. Based on the ideas of Bussolo, De Hoyos, and Medvedev's (2008) study, we construct a new weight that can show the demographic characteristics and their tendency. In our study, the population is divided by 20 age groups and further split by regions. We calculate the new weight according to China's sixth nationwide population census as well as the United Nations' population projection.At this point, we are able to obtain the new micro-household income divided by specific household CPI. Both the Foster-Greer-Thorbecke (FGT) index and the Gini index are employed to estimate the changes in poverty and inequality. The scenarios with and without population aging are used to compare the impact of demographic transition so that we can get an idea of the relationship between population aging and income distribution. To study the distributional effect among different household age groups, we decompose the FGT index and Gini index into eight household groups that are classified by area and household head's age.In this section, we briefly describe the FGT index and the Gini index first. Then, the results from the updated income from the microsimulation are introduced to the poverty and inequality index, and a general conclusion on the impact of demographic transition on income distribution is finally summarized.There are quite a lot of poverty measurement indexes, the most popular of which are the FGT index, the Watts index, and the Sen-Shorrocks-Thon poverty index. The FGT index proposed by Foster, Greer, and Thorbecke in 1984 is used in this research. The normalized FGT index is estimated in equation 16.,where \uD835\uDC64\uD835\uDC64 \uD835\uDC56\uD835\uDC56 denotes total population when the sampling weight is taken into consideration, z is the poverty line, and y i represents the household income per capita. The parameter α can be valued at 0, 1, and 2. When α = 0, FGT can measure the poverty incidence or heat-count, which is the ratio of poor people to total population. For α = 1, the poverty gap can be calculated, measuring the gap between income per capita for poverty and the poverty line. For α = 2, the severity of poverty can be calculated, measuring the equilibrium level of the poverty distribution. Both the poverty incidence and the poverty gap are measured in this paper. The choice of poverty line is crucial when measuring poverty. We use both the World Bank's poverty line and China's official poverty standard for comparison. The World Bank's poverty line is US$1.25 per day based on the purchasing-power parity in 2005. China's official poverty line has been adjusted every year during the past decades, and it was set at 2,300 yuan per year in 2011, which is equal to US$1.80 per day based on the purchasing-power parity in 2005. It is worth noting that both urban and rural household incomes in different years are divided by a household-specific CPI (equation 11) so that we can use the same poverty line for both rural and urban households as well as for different simulated years.The form of FGT decomposition can be represented aswhere G is the number of population subgroups. This can estimate the FGT index in each of the subgroups and measure the contribution of subgroups to total poverty as well. The relative contribution of each of the subgroups to total poverty can be expressed asIn this study, to estimate the demographic transition impact, the household groups are classified by household head's age, which can represent the age status for a household. If the household head's age is younger than 30, the household is classified as a young household; if 30 to 45, it is defined as an adult household; if 45 to 60, it is taken to be a senior household; and if older than 60, it is considered an old household. There are eight segments for households: rural-young, rural-adult, rural-senior, rural-old, urban-young, urban-adult, urban-senior, and urban-old.The Gini coefficient, which was developed by Gini (1912), is the most commonly used measure of inequality. This index is usually defined based on the Lorenz curve (Figure 5.1). In Figure 5.1, the straight line ab with 45 degrees represents perfect equality of income distribution curve while the curve ab represents the real income distribution curve, which is just the Lorenz curve. The area between these two curves is marked with \"A,\" and the Lorenz curve with its right area is marked with \"B.\" The Gini coefficient can then be calculated as G = A / (A + B). ( 19) The Gini coefficient ranges from 0 to 1. The smaller the Gini coefficient, the more equal is the society. The inequality 0.4 is internationally recognized as a warning line, and the inequality is dangerously large in a society if the Gini coefficient exceeds 0.4. The Gini index can be decomposed by population subgroups as follows:where ∅ g is the population share of group g to total population, φ g denotes the income share of group g, ∑ ∅ g G g=1φ g I g indicates the between-group inequality, I ̅ is the within-group inequality, and R is the residual implied by group income overlap. Both poverty's and inequality's effects on evolution in the context of demographic transition are presented in this section with FGT and the Gini index as well as their decomposition methods as mentioned before. Moreover, to quantify aging's impact on both poverty and inequality, the scenario of non-population aging across five time points is used for comparison with the basic demographic evolution scenario.From the simulation results, we obtain the following preliminary conclusions.1. Demographic transition will further reduce poverty in China; however, population aging itself has a negative impact on the reduction of poverty.Both the poverty incidence and the poverty gap associated with the two poverty lines are presented in Table 5.1. The poverty incidence and poverty gap are, respectively, 7.25 percent and 3.53 percent using the Chinese poverty line of 2,300 yuan per year in 2010. Generally speaking, in the context of basic demographic evolution and economic growth, poverty has been greatly reduced by 2015. For example, when using the 2,300 yuan poverty line, poverty incidence decreased from 7.25 percent in 2010 to 4.09 percent in 2015 and further dropped to 2.97 percent in 2020. However, poverty reduction after 2020 slowed to only a 0.95 percentage point reduction from 2020 to 2030. This is because poverty is a universally persistent problem. The government has to improve the well-developed social assistance system to address poverty. In a comparative scenario without population aging, poverty is estimated to decrease faster. For example, poverty incidence is estimated to decrease to 3.59 percent in 2015 and to drop further to 1.68 percent in 2030. This is due to the faster macroeconomic development because of the demographic dividend with the relative abundance of the labor force. In other words, the general demographic transition will reduce poverty while population aging itself opposes poverty reduction as it slows economic development. Note: Comparative scenario is the simulation without considering the population-aging issue.2. Inequality is increasing with China's demographic transition; however, population aging itself has a positive impact on equality.However, the story is somewhat different for inequality. Population aging is estimated to improve equality according to our research in China. From the results in Figure 5.2, we can see both that inequality decreases in 2015 and that it increases in the two scenarios. However, the change rate is different from the base scenario. The Gini coefficient is estimated to decrease from 0.475 in 2010 to 0.468 in 2015. This can be attributed to the fact that China's total labor force is predicted to drop from 2015. Compared with the non-population-aging scenario, inequality is much better in the basic demographic transition scenario. For example, the Gini coefficient increases to 0.4799 in 2020 and further increases to 0.5188 in 2030, but it is 0.4847 and 0.6227, respectively, in the scenarios of non-population aging. These indicate that though population aging has negative effects on economic growth and, as a result, is not good for poverty reduction, it may lead to improved equality of distribution. However, in the case of basic demographic transition, which includes four types of demographic change, inequality still increases. This may be due to other demographic transitions. Further research on different scenarios of various types of demographic change can involve studies to estimate the exact causes of inequality. 3. The rural-old group is the poorest group in China, while the young household group is better poised to escape poverty.To further understand the distributional impact on a specific household group, we decompose the FGT poverty index and Gini coefficient into different population groups by areas and age to estimate the poverty and inequality in each of the different household groups and their contribution.From the results of the FGT poverty index decomposition in Table 5.2, we can see the following: First, all household-specific poverty is declining with the development of economic growth and demographic transition. Second, from the cross-section data, the general poverty of rural areas is much more serious than that of urban areas, and the old group is worse off than the young group. Among them, the rural-old household group is the group most seriously threatened by poverty. In 2010, for example, the FGT index for rural-old households was 19.65 percent, while it is only 0.10 percent for urban-adult households. The entire FGT index is greater than 10 percent in the rural household group, while the largest poverty incidence is only 1 percent for the urban group. This can be attributed to migration in rural China where the old population is left in rural areas while the working-aging population migrates to urban areas for work. Third, the relatively young household group can escape poverty much easier than can the old household group. For example, the rural young and adult household groups are estimated to experience a reduction in poverty incidence from 16.38 percent and 13.49 percent in 2010, respectively, to 2.61 percent and 3.76 percent in 2030, while the rural-old household group's poverty incidence remains greater than 11 percent in 2030. It is worth noting that the poverty incidence for the urban-old household group experiences a tiny increase during the whole period of our scenarios. Fourth, after considering the population share, the rural-senior group contributes the most to the total FGT poverty incidence with a relative contribution of 42.45 percent to the total population poverty incidence in 2010. But with the relatively faster reduction of poverty for this group as well as the decrease of the rural population due to urbanization and population aging, this group's contribution to poverty is decreasing, and it is estimated to contribute only 30.89 percent in 2030.  Demographic transitions influenced by population aging have been attracting increasing attention throughout China and are becoming recognized as an important issue in most developing countries. However, only sparse research has studied the relationship between income distribution and demographic transition. In this paper, we investigated the evolution of poverty and inequality in the context of demographic transition. An integrated recursive dynamic CGE model with a layered microsimulation model is used to measure the income changes in light of the shock of demographic changes, such as population aging, gender shifts, urbanization, and human capital structure changes that contribute to real economic development. A comparative scenario with demographic change simulations other than population aging is adopted to capture the real impact of population aging. With the two scenarios in hand, both the FGT index and the Gini coefficient are employed to estimate the poverty and inequality changes due to demographic transition.From previous studies, we cannot find a conclusion about the relationship between population aging and inequality immediately. This paper's results show that a significant decrease in poverty and an increase in inequality are expected in the context of the multi-demographic transition. However, inequality is negative during population aging as there would be a sharp increase in income inequality with the comparative scenario, which excludes the population-aging transition. This is consistent with the macro results that the population-aging tendency would have a positive impact on reducing the ruralurban income gap and when the wage growth rate of the rural labor force and unskilled labor force is higher than that of the urban and skilled labor forces, respectively. This could be explained by China's high saving rate for rural populations. The social welfare system is not comprehensive, especially pension insurance for the rural population. This suggests that rural residents may be unable to get pensions from the government after retiring and that people expect to have to feed themselves as they get older. As a result, rural people have to work hard and start to save money for their retirement when they are young. In addition, most of the rural old people would still work the land even in their 70s. These factors may result in the decreasing rural-urban income gap with the population-aging issue. However, this result is inconsistent with some empirical research such as Zhong (2011) and Dong, Wei, and Tang (2012), which indicate that population aging would expand income inequality. However, such research mainly decomposes inequality by age and simulates the contribution of aging on inequality based on historical cross-section data or panel data, which is quite different with our study. In addition, our study shows that the process of poverty reduction is much slower when considering population. This is because the aging population has been shown to have a negative impact on economic growth from both theoretical and empirical studies in China (Peng and Fausten 2006;Cai and Lu 2013;Huang et al. 2014), which would slow down the poverty reduction process.Furthermore, the reduction of poverty and inequality are important policy objectives for China as well as for other developing countries. This study on China's case indicates that the old population, especially the rural-old population, should be prioritized because both poverty and inequality are more serious among these groups than among other household groups. This is due to urban-biased economic policies (Lu and Chen 2004) on one hand and fast urbanization development on the other hand. Our research further shows that the urbanization process, which can be measured by urban population share in our scenario, may help to reduce poverty but not inequality. This is because China's policy makers focus too much on the quantity of urbanization, not the quality of urbanization. Although China has experienced rapid urbanization, 298 million people still did not have Hukou 1 in the urban areas in 2014 although they lived and worked in these areas. As a result, they cannot enjoy the same social welfare as do urban citizens, which leads to even worse income distribution. This suggests that relevant measures such as the improvement of the social pension insurance system, especially for China's rural areas where China's social insurance coverage is insufficient, as well as the enhancement of the educational system, social security, and health insurance for migrant workers in urban areas may be helpful for reducing the inequality associated with the process of urbanization. Meanwhile, the local governments must put more effort into implementing urban-planning support systems such as public infrastructure, educational infrastructure, and the medical care system.Further research on specific demographic structure changes (such as urbanization, human capital accumulation, and gender ratio shifts) with different scenarios can be studied to find out the specific demographic reason for poverty and inequality. Future studies also should focus on China's rural population and the related economic and social problems. As for methodological considerations, there are quite a lot of scholars attempting to link CGE models with micro models, and it is proving useful in analyzing the distributional impact of exogenous policy shocks. This is an area of great interest as approaches and techniques are still under development. This paper is a layered behavioral methodology in a top-down fashion that links the results from a CGE model to the microdata. However, poverty and inequality changes can in turn induce changes in the macro economy itself, and therefore a trial of a topdown and bottom-up linkage would be much better for connecting the macro model with the micro model. This may provide fertile areas of study for future research.","tokenCount":"8930","images":["-371608326_1_1.png","-371608326_11_1.png","-371608326_11_2.png","-371608326_11_3.png","-371608326_11_4.png","-371608326_11_5.png","-371608326_11_6.png","-371608326_11_7.png","-371608326_11_8.png","-371608326_11_9.png","-371608326_11_10.png","-371608326_11_11.png","-371608326_11_12.png","-371608326_11_13.png","-371608326_11_14.png","-371608326_11_15.png","-371608326_11_16.png","-371608326_11_17.png","-371608326_11_18.png","-371608326_11_19.png","-371608326_11_20.png","-371608326_11_21.png","-371608326_11_22.png","-371608326_11_23.png","-371608326_11_24.png","-371608326_21_1.png"],"tables":["-371608326_1_1.json","-371608326_2_1.json","-371608326_3_1.json","-371608326_4_1.json","-371608326_5_1.json","-371608326_6_1.json","-371608326_7_1.json","-371608326_8_1.json","-371608326_9_1.json","-371608326_10_1.json","-371608326_11_1.json","-371608326_12_1.json","-371608326_13_1.json","-371608326_14_1.json","-371608326_15_1.json","-371608326_16_1.json","-371608326_17_1.json","-371608326_18_1.json","-371608326_19_1.json","-371608326_20_1.json","-371608326_21_1.json","-371608326_22_1.json","-371608326_23_1.json","-371608326_24_1.json","-371608326_25_1.json","-371608326_26_1.json","-371608326_27_1.json","-371608326_28_1.json","-371608326_29_1.json","-371608326_30_1.json","-371608326_31_1.json","-371608326_32_1.json"]}

data/part_2/0471280472.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"d2d2c4ecdbb724388f4d65c7e97f02ca","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/57cb1801-f342-4259-83f1-2e32c153d700/retrieve","description":"In line with the Maputo Declaration that established the Comprehensive Africa Agriculture Development Programme (CAADP) in 2003 and with the 2014 Malabo Declaration, African Union (AU) Member States pledged to conduct a continentwide Biennial Review (BR) to monitor and report on progress in achieving seven thematic commitments outlined in the Declaration. The inaugural 2017 BR Report, the first of its kind in Africa, was launched and endorsed by the AU General Assembly in January 2018. The second BR report was adopted at the AU General Assembly in February 2020.","id":"931980468"},"keywords":[],"sieverID":"3d1efd95-d824-4c5f-a4ed-09400b5d9dad","pagecount":"10","content":"Following the 2014 Malabo Declaration, African Union (AU) Member States pledged to conduct a continentwide Biennial Review (BR) to monitor and report on progress in achieving seven thematic commitments outlined in the Declaration, systematically reviewing progress and capacity to deliver on these commitments. The Inaugural Biennial Review Report (2017), the first of its kind in Africa, was launched and endorsed by the AU General Assembly in January 2018. The second BR report was presented at the AU General Assembly in February 2020.The seven Malabo Commitments were translated into the following seven thematic areas of performance in the BR:1. Re-commit to the principles and values of the CAADP process. 2. Enhance investment finance in agriculture. 3. End hunger by 2025. 4. Halve poverty through agriculture by 2025. 5. Boost intra-African trade in agricultural commodities and services. 6. Enhance resilience to climate variability. 7. Strengthen mutual accountability for actions and results.The AU Commission, with support from various partners including the Regional Strategic Analysis and Knowledge Support System (ReSAKSS), Alliance for a Green Revolution in Africa (AGRA), Bill & Melinda Gates Foundation, Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), World Bank, United States Agency for International Development, Africa Lead, and CAADP Non-State Actors Coalition, developed the Africa Agricultural Transformation Scorecard (AATS) as part of the BR to evaluate member states' progress toward the Malabo goals and recognize the best-performing countries.The AATS provides a gradual, incremental scale for assessing improvement, with \"milestones\" that indicate whether a country or region is on track to achieve the goals by 2025. For each BR year, the AUThis commitment captures member states' ability to meet the food and nutrition needs of their respective populations through successful agricultural production, access to the means of production, reduced food waste, food safety, and social protection programs for the most vulnerable populations. In some sense, Commitment 3 is the most complex, with 21 (of the 47 total) distinct sub-indicators. Progress on this commitment also shows whether the broad investments in the Malabo goals and food systems resilience do in fact result in one of the primary purposes of food and agriculture: reducing hunger and malnutrition. Within the overall AATS, the relative importance of Ending Hunger increases as the 2025 deadline approaches. In future years, the Ending Hunger theme will account for 20-24 percent of the overall AATS score, compared with only 7 percent between 2017 and 2019. Thus, while less important to the 2019 score, attention to the trends of this indicator will be essential if the AU and its member states aim to achieve their 2025 goals.Notably, the 2019 BR also added the new sub-indicator category Food Safety (3.6) to Commitment 3, reflecting the important role that safe food plays in resilient food systems, livelihoods, and trade. To measure progress toward Ending Hunger by 2025, countries reported the sub-indicators listed in Table 1, under six sub-themes: (3.1) Access to agricultural inputs and technologies; (3.2) Agricultural productivity;(3.3)  Reducing postharvest losses;(3.4)  Social protection;(3.5) Food security and nutrition; and (3.6) Food safety.The overall milestone score for Commitment 3 was set at 5.04 in 2019, an increase from 3.71 in 2017. For the 3. 2, 3.3, 3.5, and 3.6 sub-themes, the 2019 milestones were all set at 3.0, or 30 percent of the final 2025 goal. Among the indicators for 3.1, most of the 2019 milestones were set at their final 2025 levels, except for irrigation (3.1ii) at 6.5 and seed and breeding inputs (3.1iii) at 3.0. The social protection sub-theme (3.4) was set at 100 percent of the final 2025 milestone.Data for sub-themes 3. 1, 3.2, 3.4, and 3.6 were derived primarily from available national data sources. However, as the results indicate, only a handful of the 49 reporting countries submitted data for several of the 21 Ending hunger indicators. Additional data sources include FAO (Food and Agriculture Organization of the United Nations), national departments of human services, the World Bank's World Development Indicators, the World Health Organization, UNICEF, and IFPRI (the International Food Policy Research Institute). However, most indicators lacked complete data for several member states. The Technical Guidelines for preparing country Biennial Review Reports details the full methodology used and data sources for the AATS scores.   Performance on the Ending Hunger by 2025 commitment has been the lowest in both BR rounds, with zero countries on track in 2017 and only one, Uganda, surpassing the 2019 benchmark. Table 1 presents all subthemes and indicators for Commitment 3 and identifies the number of member states that reported and are on track for each indicator and the six sub-themes. Except for Uganda, which moved from not being on track in 2017 to being on track in 2019, all other countries remained off track in both rounds. Figure 1 shows the six sub-themes of Commitment 3 graphically depicted, with green indicating countries that are on track, grey indicating no submitted data, and red and yellow indicating the degree to which countries fell short of the 2019 benchmark. As Table 1 and the sub-theme maps show, zero countries were on track for sub-theme 3.1 (access to agricultural inputs); 11 were on track for 3.2 (agricultural productivity); 10 for 3.3 (post-harvest losses); 8 for 3.4 (social protection); 7 for 3.5 (food security and nutrition); and 24 for 3.6 (food safety).While the access to agricultural inputs sub-theme (3.1) had the fewest number of countries on track, it is the one with the most complete data for all indicators. For other sub-themes, some indicators suffered from poor data reporting, notably: increase in agricultural labor productivity (3.2i); increase in proportion of women with minimum dietary diversity (3.5v); proportion of children who meet minimum acceptable diet (3.5vi); and progress on the Food Safety Trade Index (3.6iii).Member states report reported low fertilizer use (3.1i) and very limited access to agricultural advisory education or extension services (3.1iv). Only 3 of 37 countries indicated that farmers had access to secure land rights (3.1vi); in 14 of the 37 reporting countries, less than 10 percent of farm households own or have secure rights to their farmland. Though only 13 member states submitted data on post-harvest losses (3.3), 10 of those reporting were on track for post-harvest loss reduction. Only 8 of 26 reporting countries met the 2019 milestone for allocating enough social protection resources to meet the required needs of vulnerable populations (3.4).Various indicators for the sub-commitment on food security and nutrition (3.5) suffered from either low reporting, poor performance, or both.While indicator 3.6iii on the Food Safety Trade Index indicates low reporting and performance, most countries performed relatively well on this new sub-commitment, with 46 out of 47 reporting countries achieving the milestone of a functional food safety system (3.6i). However, overall improvement on sub-commitment 3.6 will need more attention to meet the requirements for Food Safety Health Index (3.6ii) and Food Safety Trade Index (3.6iii).Overall sub-theme performance was poorest on agricultural productivity (3.2), post-harvest losses (3.3) and food security and nutrition (3.5), primarily due to low data reporting for agricultural productivity and post-harvest losses, and both low scores and low reporting for food security and nutrition.    The rather dismal performance on Commitment 3 in both 2017 and 2019 across the continent calls for urgent interventions. In particular, the following subset of Commitment 3 indicators and sub-themes merit extra attention due to both their low performance in the 2019 BR and their relationship to other indicators: fertilizer consumption (3.1i); proportion of farmers with access to agricultural advisory services (3.1iv); proportion of farmers with secure land rights (3.1vi); and diet quality indicators (3.5iv, 3.5v, and 3.5vi).3 Of the 42 countries that reported on this indicator, only 4 (Eswatini, Mauritius, Senegal, and Tunisia) met the milestone of 50 kg of nutrients per hectare of arable land, with Malawi close behind (49 kg/ha). Thirty countries report an average of 10 kg or less per hectare. Extreme soil degradation across the continent requires organic or inorganic soil amendments to reduce yield gaps (Tully et al. 2015;Symeonakis and Drake 2004;Vågen, Lal, and Singh 2005). Despite significant international investment in fertilizer consumption in Africa, many barriers remain, including poor or inconsistent access to fertilizer markets-a condition exacerbated in 2020 as borders closed in response to the pandemic (Ayanlade and Radeny 2020). In addition to accessibility, additional challenges include the fact that fertilizer may be of poor or unknown quality (Bold 2015); many farmers lack access to soil testing and thus may apply the wrong nutrient to address specific soil nutrient deficiencies, or they may apply fertilizer without proper erosion protection, thus wasting fertilizer while nominally improving outcomes and polluting water sources (Michelson 2017). For this reason, efficient fertilizer consumption must be appropriately paired with advisory services and soil management practices. For example, Davis et al. (2012) showed that reducing nitrogen (N) fertilizer from 80 kg/ha to 16 kg/ha or 11 kg/ha (respectively, for 3-and 4-year rotations) had no negative effect on yield; rather, maize and soybean yields increased in 3-year and 4-year rotation systems that incorporated locally available organic amendments. Inorganic fertilizers benefit one season of productivity, making producer yields highly vulnerable to market volatility, market access, and geopolitical phenomena from year to year. Strong up-front investments in integrated soil management systems can build longterm resilience while improving yields and ecosystem services and maximizing the yield effect of fertilizer inputs.Though CAADP has not prioritized soil management (MaMo 2014), stronger emphasis on nutrient recycling, crop rotation, reduced tillage, agroforestry, and other good conservation management can maximize the impact of Africa's limited fertilizer use, as indicated in Davis et al. (2012), where N application more closely matched African N consumption. Like many indicators, accessible data-reporting systems should be developed to enable all farmers to report seasonal usage of both organic and inorganic fertilizers, as well as rewarding practices that enhance fertilizer utility. When effective demand for chemical inputs is highly constrained, such as during the pandemic, farmers need straightforward guidance on how to report use of organic or traditional fertilizer methodologies, along with management practices (some practices overlap with Commitment 6) so that data reflect all soil-enhancing inputs.Only 7 of 41 reporting countries indicated that 100 percent of farmers have access to agricultural advisory services. An additional 14 countries indicated that between 50 and 95 percent of farmers have access. Yet the majority reported that less than half of farmers have access to essential advisory or extension training. A recent book published by IFPRI on agricultural extension and advisory services globally, with a particular focus on Africa south of the Sahara, shares similar findings and also notes that advisory services are chronically underfunded and poorly coordinated among pluralistic advisory providers, which may include the government, private sector, and civil society (Davis, Babu, and Ragasa 2020). Over-burdened extension agents find themselves pulled in multiple directions (with government advisors spending only 40 percent of their time advising farmers) and often receive minimal training or continuing education. These chronic challenges have led researchers and policy analysts to question how pluralistic services can most synergistically interact with centralized government services. Shared platforms must be developed to reduce redundancy of services and improve communication between providers. Other recommendations include moving away from a top-down approach to more horizontally scalable information-sharing to reduce the burden on underfunded services. Incorporating farmers and farming communities (including women and youth) in participatory research increases farmer participation in decision making and shows strong potential to address food security, climate resilience, and equity goals, and help farmers engage regional markets (IPES-Food 2020). The aforementioned IFPRI book recommends emboldening producers to identify and prioritize their needs so that they become active problem solvers of their own agronomic challenges instead of taking a supply-driven approach of pushing technologies that may not suit farmers' needs. Participatory research in Malawi, in which farmers selected and tested a range of agroecological practices according to their preferences and labor resources, showed that in addition to improved management practices, households experienced greater food security, improved dietary diversity, and gender equity (Bezner- Kerr et al. 2019). Engaging farmers as participatory researchers also facilitates learning and community information-sharing, even when extension agents cannot visit communities for extended periods of time, and encourages further local experimentation. Lastly, logistical and data-collection barriers must be overcome. Due to transportation time required to support remote farmers, service providers should be provided motorcycles or other vehicles, and advisory services should strategically integrate information and communication technologies-from radios to smartphones and other digital technologies-to extend the reach of information exchange and allow producers to report on both formal and farmer-led advisory services.Secure land rights and/or guarantee of land tenure remain a primary factor in determining a farmer's willingness to invest in long-term management practices and infrastructure that improves on-farm yields and environmental sustainability. Nonetheless, land insecurity persists as one of Africa's greatest challenges to agricultural development. Instead of increasing farmer land tenure, millions of hectares of land have been subject to largescale land acquisitions, leading to the eviction of small-scale farmers and loss of access to grazing land. While member states should improve the quality of data for this indicator, they also must establish mechanisms to prevent loss of existing land rights. Africa south of the Sahara has been a hot-spot of farmland acquisitions, with international speculators, state-owned enterprises, members of the diaspora, and domestic political elites viewing land-based investments as increasingly attractive; in many cases, governments provide the regulatory framework and financial, technical, and administrative support conducive to land acquisitions (Anseeuw et al. 2012;Deininger and Byerlee 2011). True progress on this indicator will require challenging negotiations among stakeholders, and must include widespread participation from farmer communities at risk of land acquisition. In some cases, the assumption of large-scale investment as an effective pathway for economic development and poverty alleviation will need to be reconsidered. Identifying land for development requires inclusive legislative and negotiation processes, with checks and balances on customary authorities, state agencies, and others to protect the rights of small farmers. Monitoring functions with full public disclosure of findings must be upheld by governmental and nongovernmental agencies. Because of the difficulties of safeguarding customary rights even in countries currently providing \"best practice\" legal protections, human agency among civil society and political leaders will play a role in shaping outcomes (German, Schoneveld, and Mwangi 2013) outcomes that have severe consequences for food security and farmers' ability to make long-term investments in sustainable management practices and agricultural resilience (see Commitment 6 brief on Climate Resilience for more details).Diet quality (3.5iv, 3.5v and 3.5vi)Poor data on nutritionmost notably, on diet qualityremains a major reporting challenge in most countries. Indicators such as the increase in proportion of women with minimum dietary diversity and proportion of children (6-23mo) who meet minimum acceptable diet were especially underreported. This challenge requires improved data capacity as well as continentwide prioritization of diet quality, not mere caloric sufficiency or economic growth. Nutrition experts note that agricultural development programs alone do not sufficiently improve diets of all household members; strong health and nutrition communication and women's empowerment activities are recommended. Social factors like gender equality significantly influence diet quality. For example, a four-year study in Malawi showed that households in which spouses discussed farming were 2.4 times more likely to have improved food security and dietary diversity compared with those in which spouses did not discuss farming (Bezner- Kerr et al. 2019). Finally, emphasis on diet quality will help policymakers prevent the potential negative dietary changes associated with some agricultural development programs, where rising incomes in low-and middle-income countries have resulted in rapid rises in overweight and obesity (Popkin, Corvalan, and Grummer-Strawn 2020).Achieving Malabo Declaration commitments will pave the road for Africa to achieve the Sustainable Development Goals. The expected benefits for smallholder farmers and millions of ultra-poor and food-insecure households are tremendous. However, progress will require persistent investment in both the Malabo commitments themselves and member states' capacity to measure and report on those commitments. Building resilient food systems across Africa demands extensive investments in land, water, locally relevant knowledge, and vulnerable communities, as well as robust systems to track progress and predict and mitigate the risks of increasing shocks. The COVID-19 outbreak poses significant challenges to already strained food systems in Africa. The pandemic is affecting health, in terms of morbidity and mortality, with negative repercussions on non-health sectors such as agriculture, tourism, transportation, and entertainment. The impacts of COVID-19 further exacerbate a situation of ongoing shocks such as desert locust swarms, fall armyworm spread, droughts, conflict, and insecurity. As pointed out by FAO/AU (2020), the continent has made important progress in terms of prioritizing social protection as a core component of poverty reduction and rural development strategies, including in the context of the Malabo Declaration; the COVID-19 pandemic offers a critical opportunity to scale up these efforts. Previous crises including the HIV/AIDS epidemic, food crises, and Ebola outbreaks have shown that while responding health needs may be the main priority, impacts on income, food security, and livelihoods must also be addressed by employing both immediate and medium-term strategies to build resilience and prevent backsliding on poverty reduction and food security gains.","tokenCount":"2827","images":["931980468_1_1.png","931980468_1_2.png","931980468_4_1.png","931980468_4_2.png","931980468_4_3.png","931980468_4_4.png","931980468_4_5.png","931980468_4_6.png"],"tables":["931980468_1_1.json","931980468_2_1.json","931980468_3_1.json","931980468_4_1.json","931980468_5_1.json","931980468_6_1.json","931980468_7_1.json","931980468_8_1.json","931980468_9_1.json","931980468_10_1.json"]}

data/part_2/0471592009.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"69bd133e769ea4946c5efac8e91e1998","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/7e993528-62d2-4f36-9e63-b4c70bf2a076/retrieve","description":"The paper develops a metric of structural transformation that can account for the production of new varieties of goods embodying advancements in technological know-how and design. Our measure captures the dynamics of an economy's transformation and can be viewed as an extension of Hausmann and Klinger's static measure. We apply our measure to four-digit-level SITC trade data of China, Malaysia, and Ghana over the period 1962-2000. The results show that two important factors characterize the rapid transformation of the Chinese economy: the high proximity of its export basket to three main industrial clusters-capital goods, consumer durable goods, and intermediate inputs-and the increase in the values of the new goods belonging to those three clusters. Malaysia exhibits a similar but more modest pattern. In contrast, the structure of the Ghanaian economy appears unchanged over the entire 1962-2000 period. That economy is dominated by primary goods clusters, and the values of the goods in those clusters have remained relatively low. We also discuss qualitatively the role of policies and institutions in spurring transformation in the three countries.","id":"57611989"},"keywords":["structural transformation","discovery","technological change JEL Codes: F19","O14","O33","O40"],"sieverID":"b54aa2fe-96e3-402e-a2ac-021ac49c65c5","pagecount":"40","content":"The paper develops a metric of structural transformation that can account for the production of new varieties of goods embodying advancements in technological know-how and design. Our measure captures the dynamics of an economy's transformation and can be viewed as an extension of Hausmann and Klinger's static measure. We apply our measure to four-digit-level SITC trade data of China, Malaysia, and Ghana over the period 1962-2000. The results show that two important factors characterize the rapid transformation of the Chinese economy: the high proximity of its export basket to three main industrial clusters-capital goods, consumer durable goods, and intermediate inputs-and the increase in the values of the new goods belonging to those three clusters. Malaysia exhibits a similar but more modest pattern. In contrast, the structure of the Ghanaian economy appears unchanged over the entire 1962-2000 period. That economy is dominated by primary goods clusters, and the values of the goods in those clusters have remained relatively low. We also discuss qualitatively the role of policies and institutions in spurring transformation in the three countries.Wealthy countries are not only recognized for producing more output per worker than poor countries, they also produce a larger number of evolving varieties of more complex products associated with higher unit value. Such products embody varieties of intermediate factors that entail advances in technological know-how and design to meet the changing wants of own households as well as those around the world. This evolutionary process characterizes a structural transformation within and across industrial sectors. Determinants of this transformation process have received increasing attention in recent years. Hausmann, Hwang, and Rodrik (2005), for example, observed that the types of goods a country produces and exports affect its long-run economic performance. Building on that work, Hausmann and Klinger (HK) (2006) focus on the determinants of structural transformation as a process in which developing countries approach those with advanced technologies and economic structures. They find the rate of structural transformation relates to the proximity of new goods to the ones a country is currently producing and the values of the new goods. The slow change in structure experienced by many lower-income countries can be explained by the specificity of their skills and assets to certain types of goods that do not facilitate the transformation toward more complex industrial products.While HK's study provides new insights for understanding differences observed in the evolution of industry and the determinants of such differences, some important issues remain unaddressed. First, the HK study does not provide a measure of the dynamic performance of industrial clusters within individual countries. Without such information, it is difficult to inform the policy process of the experience of successful countries. Second, although HK conduct a country-specific analysis of proximity for selected developing countries, they do so only by applying the data of a single year. A single-year analysis is static and unlikely to address many important dynamic questions. These include: What are the features over time of structural transformation of those developing countries that have increased substantially their proximity to upscale or more complex products? Why have some developing countries successfully moved to produce upscale or more complex products while others continue to produce the same lowvalue goods year after year? Is the industrial transformation in a country an automatic process resulting from the accumulation of fundamentals or an outcome of more pragmatic policy and institutional reforms?This paper extends the HK study and fills the aforementioned gaps by adapting HK's methodology to analyze the dynamics of structural transformation of selected countries in the context of the evolution of world trade. We use the same data set as HK, World Trade Flows from Feenstra et al. (2005), and focus on three of six countries in the HK study-China, Malaysia, and Ghana. To complement the empirical analysis, we also discuss the role of policies and institutions that appear to have spurred the transformation in the three countries.HK measure the distance or proximity of a country's export basket to a particular good. To study the dynamics of structural transformation, we instead calculate the distance between a country's current export basket and a group of goods a country is not producing. We call this new measure the density gravity center. Using this measure, we find that the change in Chinese industrial structure over 1962-2000 is the result of two important factors: the high proximity of the country's export basket to three industrial clusters-capital goods, consumer durable goods, and intermediate inputs-and the high values of the new goods in these three industrial clusters. This suggests that these clusters contain goods that are relatively sophisticated or complex. A similar transformation pattern is found in Malaysia but of a relatively more modest magnitude compared with China.Ghana stands out. Its export structure appears to be unchanged over the 39-year period (i.e., 1962-2000). Not only are this country's exports continuously dominated by non-energy primary goods, a less sophisticated industrial cluster, but also the values of individual products in this cluster have remained low. In the qualitative discussion section we further find that although the evolution of China's institutions is seen to be out of alignment with the Washington Consensus (Rodrik 2006a), its rapid transformation appears to have benefited from openness to multinational enterprises, the relatively large scale of its economy, the abundant endowment and low cost of labor, and the supply of relatively lowcost materials as intermediate inputs. The institutions developed in Malaysia inherit the features of that country's former colonial partner that facilitated its transformation much earlier than China. As a result, Malaysian industry has advanced and continues to evolve. However, initial conditions as well as structural factors have allowed China to outperform Malaysia in the transformation process. Policies and other barriers appear to have prevented an industrial transformation of the Ghanaian economy.The forces of structural transformation are broadly discussed in the literature, including the centrality of research and development (R&D) in driving innovation and the role of policies and institutions in this process. R&D is viewed as facilitating the acquisition of the technological know-how needed to upgrade the quality of industrial products (Stokey 1988;Grossman 1989;Segerstrom, Anant, and Dinopoulos 1990;Grossman and Helpman 1989c, 1990a, 1991;Aghion and Howitt 1992), the development of new industrial products (Romer 1990; Grossman and Helpman 1989a, 1989b, 1989d, 1990a, 1990b), and/or the reduction in the cost of production (Corriveau 1988). Innovation is the result of actions taken by firms in response to market incentives that in turn are dependent upon a well-functioning market environment supported by the government. This implies that the pace of innovation is likely to be higher where policies and institutions are well designed to induce knowledge discovery. Many of these studies also provide insights into mechanisms through which poor countries can achieve a more rapid structural transformation. Openness to trade provides opportunities for domestic firms to exploit the discoveries in advanced countries. Also, a good policy environment creates conditions for domestic firms to become multinational, which allows them to take advantage of abundant and low-cost resources in other parts of the world, extend the scale of their enterprise to larger markets, and further facilitates the transfer and absorption of more advanced foreign technology.The rest of paper is organized as follows. Section 2 revisits and summarizes the findings of the HK study. Section 3 extends the HK method by developing a new measure, the density gravity center, and applies that measure to analyze the dynamics of structural transformation of the same three countries-China, Malaysia, and Ghana-studied by HK so that we may compare the results and insights of our approach with theirs. Section 4 investigates the determinants of the dynamics of structural transformation in those countries. Section 5 discusses the role of policies and institutions in the determination of structural transformation in those countries. Section 6 concludes.The HK paper develops the concept of product space and uses it to investigate the process of structural transformation and its determinants. According to HK, the assets used in the production of currently existing goods in a country can be adapted to produce new goods. The adaptation capacity of that country depends on how close the technologies employed to produce goods in a specific period are to the technologies used to produce new goods and whether the new goods are upscale, or of higher value. The theoretical model underpinning HK's empirical analysis is a two-period overlapping generation model of firms with each producing in each period one unit of either the existing/standard good (good 1) or the new good (good 2). The new good is more attractive since it bears a higher price compared with the standard good, that is, P2 > P1. It is also associated with a fixed cost C. This cost increases with the distance between the standard and new goods by the parameter δ 12 . A positive externality exists for subsequent firms (entrants) because they do not incur the fixed cost C. Such externalities in adapting capabilities are the force that drives innovation in this model.The empirical assessment of the HK model is achieved using the World Trade Flows data from Feenstra et al. (2005) covering the period 1962-2000. HK first construct a matrix of proximity of pairs of goods in each period, where the proximity (inverse of distance) measure is approximated by the conditional probability between two goods. 1 The proximity between two goods (say i and ) j at time t is given byis the probability for a country to produce a good i in period t given that it is already producing a good j in the same period, and  is the probability for a country to produce a good j in period t given that it is already producing a good i in the same period. It is expressed aswhere the numerator of equation  2 is the joint probability of producing both goods i and , j and the denominator is the marginal probability of producing good i or good . jIt should be clear from the preceding description that the subscripts i and j are used to designate two goods in the product space in a specific period and do not refer to a new or an old good. Also, it is important to stress that the proximity between two goods is the same regardless of the countries producing them, and it is bounded below by 0 and above by 1. A value of of 1 implies that goods i and j are homogeneous, while the value ofof 0 implies that the two goods are heterogeneous.Between 0 and 1, goods i and j can be close to being homogeneous or heterogeneous depending on whether the value of is close to 1 or to 0. A path or distance-weighted number of products around a good i is then constructed such asUsing the average proximity over 1998-2000, HK make the first cut into the characteristics of product space. Fifteen products are located in the densest part of the forest, that is, the part of the forest with more sophisticated or complex products. These 15 products include 10 manufactured products, 3 machinery products, and 2 chemical products. 3 On the other hand, 15 products are located in the least dense part of the forest, including 10 unprocessed agricultural and animal goods, 3 low-tech manufactured products, and 2 chemical products.The firms' decisions or abilities to jump to producing new products depend on the locations of the currently produced goods in the economy, the distances to the new goods, and the values of the new goods. HK build upon those three factors to test their model of structural transformation described earlier. They first construct the measure of distance as well as the prices for both standard and new goods. They next extend the measure of proximity between two products to the proximity between the current export basket, taken to represent the existing economic structure of the country, and a particular good. The measure of this proximity is termed density. The intuition behind this measure is that if a country produces goods that surround or are close to a particular other product, then the probability of this country to develop a RCA in that particular product in the future should be high. The density is a scaled sum of paths that lead to the good in which a country has a RCA, where the scale is the number of all paths. The value of the density of good i in country c at time t is between 0 and 1. 4 More formally, it is given bywhere andx , , are defined as before. Unlike the proximity measure, which is the same across countries for two specific goods, the density of a good varies across countries.Then, HK define the prices for the standard goods as well as for the new goods by constructing two additional measures, Expy and Prody. The Prody measure is commodity specific and defined for good i at time .t Prody i,t is calculated as the weighted sum of per capita gross domestic product (GDP) of all countries exporting this good, where the weights are the countries' respective RCAs in good . i Prody is defined globally, that is, it is the same for all countries in the case of good i at time t. Expy, on the other hand, is country specific, but the same for all goods in which this particular country has a RCA. The Expy measure is calculated as the weighted sum of Prody over all goods exported by this country, where the weights are this country's share of each exporting product in the country's total exports. With these constructed measures, HK also test the structural transformation model empirically through a cross-country regression (probit and ordinary least squares) using World Trade Flows data from 1985 to 2000. They test whether the price of a new good (Prody) as well as the proximity of the new product (density) have positive effects on the probability of developing a RCA in the product in the next period, controlling for the price of the standard good (Expy), and whether the country has a RCA in the product in the current period. The regression results confirm the prediction of the model. An increase of one standard deviation in the density increases the probability of exporting a new good in the next period by 1.3%, while an increase of one standard deviation in Prody causes this probability to increase by 0.008%.The analysis of structural transformation at the individual country level is illustrated by plotting the difference between Prody and Expy of a good against its distance 5 (inverse of density) using one-year data (1999) and for the selected three countries. HK's results for the three countries (China, Malaysia, and Ghana) are presented in Figure 1. 6 As shown, patterns of industrial structure as well as potential transformation are quite different among the three countries in 1999. Starting with China, it is obvious that the nearby goods 7 to its current export basket that represents its industrial structure in 1999 are downscale, or low-value primary goods (food and animal, crude materials [agricultural and natural resources], fuels, and beverages). But at a distance of about unit 1, more upscale or high-value products exist. These products, which include chemicals, manufactured, and machinery, represent a more sophisticated industrial structure than the 1999 structure. The 1999 structure indicates a relatively high potential for China to further change its industrial structure in the years following 1999. The same pattern can be seen for Malaysia in which upscale products in the chemicals, manufactured, and machinery clusters exist at a slightly farther distance than in China (starting at unit 1.5). For Ghana, on the other hand, a totally different pattern is observed in the figure. In fact, no product exists at a distance of approximately 1.5 from Ghana's current (1999) export basket. The goods near to its current export basket start at a distance of 1.75, and they are downscale products as exemplified by food and animal and crude materials. The close distance to the upscale products is as far as 1.8, and most of these products belong to primary industrial clusters. The upscale, high-value goods such as those in the chemicals, manufactured, and machinery clusters are far from the country's current export basket with a distance of more than 2.5. Although only one year's data are used, these results show key differences in potential for structural transformation across the three countries. However, without a different measure that can capture the process or dynamics of the transformation process over time, these figures can provide only a conjecture as to the different transformation rates across countries in the future.Figure 1. Visual representation of proximity for selected three countries 5 The measure of the inverse of density is in log. 6 The HK study depicts this relationship for six countries: China, Malaysia, Columbia, Venezuela, El Salvador, and Ghana. 7 A distance (inverse of density) of a good to the existing productive capabilities can be read on the x axis. A nearby good is the one located at a distance relatively close to 0 compared with other goods. The results of Section 2 show how the proximity of the current export basket to new products as well as the price differential between the current and new products determines the patterns of structural transformation. Although the results are consistent with the predictions of the model developed by HK, the analysis is for only one year (1999). A one-year analysis is rather static and may miss key factors unobservable at a specific point in time. Our contribution is to develop the methodology so the analysis can be conducted over time at the country level. Such an analysis will allow for the comparison of patterns of structural transformation across countries and time, thus helping to identify the dynamic differences in the process of transformation.To achieve this goal, we need to develop a new measure of proximity in which the distance from the current export basket to a broader set of products can be captured. HK's measure of proximity (density) captures only the distance between the current export basket and a particular good. It does not, however, provide any idea about the diversity of the basket as a country develops RCAs in various sectors. Specifically, if a country's export basket includes goods that are located in diverse parts of the product space,8 then the probability is high for that country to have its export basket surrounded by many new goods. As a result, such a country may develop RCAs in many other goods, possibly including some high-value goods, in the future, and hence achieve a more rapid structural transformation. Conversely, a country with an export basket composed of products located in a few parts of the product space may develop RCAs in a very few new products in the future. The possibility of new, upscale goods becoming part of this country's export basket is less than in comparison to the former country. The structural transformation in this country will tend to be stagnant.The concept of comparative advantage we use in this study is broader than the one used in the Heckscher-Ohlin theory of trade. According to that theory, a comparative advantage in a specific good is endowment driven and does not admit any role for differences in technology across countries. As an implication, the structural transformation is shaped primarily by factor endowments as countries will have a tendency to move toward the production of goods intensive in factors they are abundantly endowed in. However, recent patterns of comparative advantages across countries have shown that the static aspect of comparative advantage (endowment driven) is not always justified. For instance, the East Asian countries (first and second tiers) are highly endowed in labor and natural resources but have moved lately toward the production of high-tech products, including automobiles, electronics, and other capital goods. This indicates that the comparative advantage has a dynamic aspect that goes beyond just the factor endowment. A country can develop comparative advantages over time by investing intensively in R&D, human capital, and learning by doing, and by implementing policies that allow factor endowments to play a supporting role in this process (see Grossman 1989;Costinot 2009).We develop a measure of proximity we call the density gravity center (DGC). The density gravity center measures the distance between a country's current export basket and all goods in the product space in which the country is not present but are produced by the other countries. The value of DGC is high when a country has a diversified export basket surrounded by many new goods and low for a country with a less diversified export basket. The DGC for country c at time t is defined as a sum of densities of all goods i in which a country has RCAs (i.e., 1 , ,). This sum of densities is further normalized by densities of all goods regardless of whether this country has a RCA in good i or not. Specifically, DGC is given by(Before applying this measure, we first describe how it relates to the other determinants of structural transformation developed by HK. Recall that the price differential between the existing and new goods is the other important factor that drives innovation, in addition to the distance we discussed earlier. Now the question is whether the prices HK developed (Prody and Expy) are still relevant for the new measure developed here (i.e., DGC). Since Expy is country specific at time t, it is also suitable for the measure of the level of sophistication of the country's export basket in a particular year. 9 The Prody, on the other hand, is commodity specific and uniform for all countries because this measure includes all countries' RCA in commodity i. Given the similarity of Prody for a particular product across countries, we need only to find weights for individual good i in order to construct the Prody of a group of goods. We can use a global weight such as trade share of each individual good in world total trade. The difference between Expy for a country and Prody for a group of goods in the world can be used to assess whether an export basket for a country includes several new goods that are upscale or downscale with respect to the country's current export basket. If new goods are predominantly upscale, then they should translate into an increase in the value of Expy, and if they are downscale, they should translate into a decrease in its value.Using the above simplification, we first analyze the relationship between Expy and DGC for a specific country. If a country's export basket is diversified and surrounded by many new goods, then the country is likely to develop RCAs in these goods. Developing RCAs in these goods should translate into an increase in the value of a country's export basket if most new goods are upscale, or a decrease in the value of its export basket if most new goods are downscale.We use the same data set as in the previous section to show the relationship between Expy and DGC. We select three countries to study this relationship-China, Malaysia, and Ghana. Two reasons motivate our choice of these countries. First, the three countries were selected by HK (2006) in their static analysis of structural transformation. Choosing them in our study allows a comparison of our results with theirs. Second, we want to see whether differences in the nature as well as the patterns of comparative advantages explain differences in the structural transformation process across countries. Choosing these three countries helps find this explanation since Ghana has a production/export profile that is endowment based, while the production/export profiles of China and Malaysia are based on both endowments and R&D.We apply the Expy and DGC measures to data of the three countries. Figure 2 depicts this relationship for China, Malaysia, and Ghana. Each point in the figure is a year-specific coordinate of the proximity of the export basket to all other goods a country is not producing/exporting (DGC) and the value of the export basket in the next period (lnExpy). A trend line reveals the general extent of transformation. A clear upward-sloping trend suggests a positive relationship between Expy and DGC, and is observed for China and Malaysia but not for Ghana. This trend indicates that the increased value of the respective country's export basket is associated with the increases in the value of DGC, while in the case of Ghana a flat trend indicates no relationship between those two measures. Also, the slope of the trend in the case of China is steeper than that of Malaysia, suggesting that China's industrial structure and hence export composition have changed more rapidly than that of Malaysia. Moreover, these changes are associated with larger increases in the number of new and more complex goods with higher values in the case of China compared with Malaysia. While Malaysia's export structure changed during the same period, the speed of change seems to be slower than that in China. In contrast, the commodity 9 We will use Expy interchangeably as the price or the value or the level of sophistication or the level of income of a country's export basket. composition of exports in Ghana shows no change. This implies that while Ghana may have added new goods to its export basket, those new goods are dominated by the presence of downscale goods.   of 9.69, 7.69, and 0.51 for China, Malaysia, and Ghana respectively. This result confirms the significantly different magnitude of transformation for each of the three countries. This estimate can also help us assess the density elasticity of Expy. A 1% increase in the value of DGC translates into an increase in the value of Expy of 3.7% for China, 1.53% for Malaysia, and only 0.05% for Ghana. Indeed, these results suggest that the transformation speed has been dramatically different between China and Ghana and modestly different between China and Malaysia. In other words, the speed of structural transformation for China was 2.4 times that of Malaysia and 72.6 times that of Ghana over the period 1962-2000. These results prompt the question of whether Ghana can \"catch up\" to one of the other countries in this structural sense. Notice that two periods can be distinguished from China's graph: the 1960s and 1970s period (pre-reform period) and the 1980s and 1990s period (post-reform period). If we draw a trend line for the first period (below the shown trend line), we see little difference in the process of transformation between China and Malaysia. However, a trend line for the second period is steeper than the one shown, that is, a change in the estimated intercept and the slope. This is an indication that the process of transformation was accelerated during the post-reform era. Thus China's experience might reflect the potential that other countries would aspire to.Also, these results show that China's structural transformation entailed diversification and increasing sophistication of its export basket. While the replication of the one-year HK analysis of the previous section is consistent with these results, the single-year analysis could not tell how each of the three countries has transformed its structure of exports over time, prior and after 1999.To further assess the process of industrial transformation of the three selected countries, we include in the analysis for comparison two countries that have the most sophisticated export profiles in the world (the United States and Japan) and one country that features a very rapid transformation (Korea). In Figure 3 we include DGC (panel one), distance (panel two), and Expy (panel three) for the six countries over time. As the first panel of Figure 3 reveals, the patterns of transformation of China stand out. 10 A linear equation, y = a + bx, is used here. We also tried other types of equations including logarithmic, polynomial, power, and exponential. However, the linear trend seems to fit the data better than the other ones.  , 1962-1971, and a modest reversal period, 1971-1973, has improved its proximity from 1973 on. However, that country's DGC curve stayed below those of the United States, Japan, and China. On the other hand, Ghana remained at the bottom of the figure in the entire period. The graph on the distance (panel 2 of Figure 3) reinforces these conclusions. Starting at a distance of 3.09 in 1962, China reduced the distance between its export basket and the rest of the goods in the product space over time, cutting it down by 22% with respect to the ideal distance (distance = 0) and approaching the group of richer countries by year 2000. Malaysia and Ghana reduced their respective distances too, but they were still far from those of the rich countries by the end of the period. Between 1962 and 2000, Malaysia reduced its distance by 4%. The distance for Ghana remained the largest, at 7.20 from the ideal distance.   The intersecting or the crossing of the DGC lines of some advanced countries does not imply that a country (e.g., China) has reached a similar or more advanced export basket than an advanced country (Japan). Instead, this result implies that the less advanced country is likely to develop RCAs in a number of new goods in the near future that are currently in the advanced countries' export basket. The development of such new goods will accelerate the transformation of the industrial structure only if they contribute to an increase in the level of sophistication of the export basket as captured by the value of Expy. A look at the last panel of Figure 3 indicates that the value of China's Expy increased during the period under study. This suggests that the improvement in its DGC has translated into the development of the RCAs in more upscale goods. However, the sophistication levels of its export basket were still below those of the rich countries (the United States, Japan, and Korea) by the year 2000. Malaysia was further below the advanced countries as well as China, but undercut the latter in 1992. In contrast, Ghana is the less-performing country on the basis of the sophistication of its goods. After a relatively short period of increase in its Expy (1962)(1963)(1964)(1965)(1966)(1967)(1968)(1969)(1970)(1971)(1972)(1973)(1974)(1975)(1976)(1977), the country displayed an episode of erratic movement in the value of its export basket that suggests a stagnation in its process of structural transformation.As emphasized in Section 2 and shown in Section 3, the characteristics of the product space determine the patterns of structural transformation. The close proximity of China's current export basket to new, highvalue goods has helped China transform its structure of production/export more rapidly than Malaysia and Ghana. Now we extend the analysis to assess how the proximity as well as the price differential between current and new goods influenced the speed of structural transformation across countries.We first classify all products that could be included in the product space into six industrial clusters. These six groups of products are (1) capital goods, (2) consumer durable goods, (3) consumer nondurable goods, (4) intermediate inputs, (5) primary energy, and ( 6) nonenergy primary. 11  We then measure the proximity of each cluster to new goods in the same category to assess which clusters have contributed the most to the process of structural transformation. The contribution also depends upon the values of new goods. This dependence requires a value index for each of the six clusters over time. With some modification, we construct an index similar to the Prody measure in HK. Specifically, instead of constructing the value (price) for a particular new good we construct a value index for a group of new goods. This is accomplished by defining the group Prody (GPrody) for a group (or cluster) of goods as the weighted sum of Prodys of goods in that cluster, where the weights are the world export share for each product included in that cluster scaled by the their respective total shares. The calculation is given byThe results of GPrody for the six product groups appear in Figure 4. The capital goods group has the highest values of GPrody over the entire period 1962-2000. The gap between capital GPrody and the other GProdys widened with time. GPrody for consumer durables has the second highest value between 1962 and 1971, after which the intermediate input group prevails in [1972][1973][1974][1975][1976][1977][1978][1979][1980][1981][1982][1983][1984][1985][1986][1987][1988][1989][1990], and the consumer nondurable group since 1990. Also, this figure shows that the two primary product groups always have the lowest values in GPrody with a few exceptions corresponding to world primary resource shocks in the 1970s and early 1980s.The magnitude of the GPrody index for each cluster has a major effect on our measure of structural transformation. The proximity of a country's export basket to the capital goods and consumer durables will affect our estimate of the speed of structural transformation, since these goods not only bear higher values but also often embody recent advancement in technological know-how and designs (which has been well studied in the endogenous innovation literature cited in the introduction). The proximity to the intermediate inputs and to some extent to consumer nondurables is also important in the transformation process. However, the proximity to the two primary product groups appears to not contribute to a country's process of transformation. This result can be explained as follows. First, primary products often bear low values of Expy compared with other categories and do not contribute to the 11 These six classification groups are consistent with the United Nations classifications (see United Nations 2000). Capital goods include industrial and nonindustrial equipment and transportation engines. Consumer durable goods encompass durable and semi-durable goods and passenger vehicles. Consumer nondurable goods include food and nondurable goods used mainly for household consumption. Intermediate inputs include parts and accessories, processed products, and other products mainly used as inputs in the industrial production process. Primary energy goods are composed of energy resources such as hydrocarbons, coal, and so on. Nonenergy primary goods include mineral resources, industrial minerals, and primary agricultural goods. We prefer this classification to Leamer's commodity clusters (1984). Indeed, some of Leamer's clusters include products that are not homogeneous in terms of factor shares and capabilities required to produce them. For instance, cluster 2 (crude materials) includes unprocessed animal and agricultural products, other natural resources, fuels, processed agricultural and animal products, chemicals, and so on. increase in the value of the export basket. Second, primary products are mostly located in the sparse part of the forest and tend not to provide technological links to the production of other more complex products. Finally, little advanced technology is embodied in such products, which constrains them from serving as a driving force of the structural evolution of a country's industry through complementarities or technological spin-offs in designing and inventing new products. Whether the presence of primary groups hinders the transformation process, as some of the primary resource literature might suggest, is worthy of further study. We now turn our attention to the characteristics of the product space that have influenced the speed of transformation differentially across countries by constructing the group DGC (GDGC) for each individual country. The question we address here is whether the group proximity of each industrial cluster to the new goods in this cluster has translated into the development of RCAs in new goods for an individual country. The results are reported in Figure 5 for each of the three countries.We start the discussion for China. As shown in the first chart of Figure 5, the GDGC for two industrial clusters, that is, consumer nondurables and primary nonenergy, has decreased over time, from 0.50 and 0.33 in 1962 to 0.43 and 0.30 in 2000, declining by 14% for consumer nondurables and 15% for primary nonenergy. In contrast, GDGC for the other two clusters, capital goods and consumer durables, increased substantially in the same period. It rose from 0.05 and 0.40 in 1962 to 0.31 and 0.79 in 2000 for capital goods and consumer durables, respectively, a total of 520% and 100% increases for these two groups in this period. China also made modest improvements in the proximity to the intermediate inputs with GDGC rising from 0.30 to 0.34, or an increase of 13% in this period.In comparison, Malaysia's GDGC paths are similar to China's, with relatively low initial values. The GDGC for consumer nondurables and primary nonenergy fell from 0.33 and 0.28 in 1962 to 0.28 and 0.19 in 2000, with similar degrees of decline as in the case for China (15% and 32%, respectively). Also, similar to China, Malaysia's GDGC for both capital goods and consumer durables has increased, rising from 0.03 and 0.09 in 1962 to 0.33 and 0.21 in 2000, or 725% and 133% increases, respectively. The GDGC for intermediate inputs also rose slightly, from 0.161 to 0.164, in this period. 1 9 6 5 1 9 6 8 1 9 7 1 1 9 7 4 1 9 7 7 1 9 8 0 1 9 8 3 1 9 8 6 1 9 8 9 1 9 9 2 1 9 9 5 1 9 9 8  In contrast to China and Malaysia, Ghana's structural transformation is quite different. The only clusters in which substantial improvement is observed are consumer nondurables and nonenergy primaries with their GDGC rising from 0.02 and 0.11 in 1962 to 0.22 and 0.26 in 2000. The GDGC for the intermediate inputs, on other hand, fell to 0.12 in 2000 from 0.16 in 1962. Furthermore, the GDGC for consumer durables shows a stagnant pattern with a similar low value in the beginning and ending of the period (0.071), while the GDGC for capital goods fell to 0.011 in 2000.Figure 5 suggests that differences in the rate of structural transformation across the three countries is associated with the proximity of each country's export basket to the upscale (more complex) goods such as capital goods and consumer durables, and to some extent to intermediate inputs. Increases in the values of GPrody for these three industrial clusters and the large magnitude of such increases are an indication of a relatively rapid rate of transformation. However, the initial conditions also matter. That China started with a proximity to each of the three categories higher than Malaysia in 1962 is such an indication. China also maintained its leading position for the entire period under study. The consistency of such a pattern seems to suggest that China has developed RCAs in more new goods in this transition than Malaysia and Ghana. Indeed, the cumulative number of all new goods exported in 1963-2000 was 924 for China, 529 for Malaysia, and only 245 for Ghana. Of the 924 goods that China exported, 103 were capital goods (11% of total number), 127 were consumer durables (14%), 112 were consumer nondurables (12%), 418 were intermediate inputs (45%), 163 were nonenergy primary goods (18%), and only 1 was primary energy.The structure of the new goods exported in Malaysia is similar to that of China, although the former is half the distance from the latter in level terms. In the case of Malaysia, the shares of clusters in the total number of new goods are 16% for capital goods, 16% for consumer durables, 13% for consumer nondurables, 35% for intermediates, and 20% for non-energy primary goods.The structure of new goods in Ghana is different from those in China and Malaysia. Capital goods have an extremely low cluster share of 5%, followed by consumer durables with a share of 7%. While the shares for consumer nondurables (15%) and intermediates (38%) are comparable with the corresponding shares for China and Malaysia, a much larger share (34%) is observed for nonenergy primaries in Ghana. 12 While Ghana continuously showed strength in the development of new products in the nonenergy primary cluster, products in this category had low value (low GPrody), which indicates that these goods did not contribute to the country's structural transformation. This comparison again raises the question of whether comparative advantage in primary good exports tends to crowd out the production of new, more complex goods.To support this argument, we depict in Figure 6 the contribution of each industrial cluster to each country's export basket represented by Expy. As the figure shows, in China and Malaysia the structure of exports in 2000 was significantly different from the one prevailing in 1962. The initial (in 1962) top three clusters for China were intermediate inputs (51% of Expy), consumer nondurables (28%), and nonenergy primaries (14%). Consumer durables and capital goods, the two more advanced clusters, had the lowest share, 6% and 1%, respectively. In China in 2000, a totally different structure is observed. The consumer durables, capital goods, and intermediate inputs became the most important three clusters with shares of 37%, 33%, and 16%, respectively, while two of the country's initial top three categories turned into the last two with shares of only 12% for consumer nondurables and 3% for nonenergy primaries. 1 9 6 5 1 9 6 8 1 9 7 1 1 9 7 4 1 9 7 7 1 9 8 0 1 9 8 3 1 9 8 6 1 9 8 9 1 9 9 2 1 9 9 5 1 9 9 8 Likewise, Malaysia's structure of exports in 2000 was totally different from its 1962 structure. Its initial structure was dominated by non-energy primaries with a share of 68%. The shares for consumer nondurables, capital goods, and consumer durables were all very small. However, by 2000, the capital goods cluster became the most important category in export structure with a share of 66%, followed by intermediate inputs (12%) and consumer durables (9.5%). In comparison, Ghana's product/export structure in 2000 closely resembled its product/export structure of 1962. Its top two categories-nonenergy primaries and intermediate inputs together-contributed 99% and 84% to the country's total value (Expy) in 1962 and 2000, respectively. The only slight change was the combined contribution of the remaining three categories, increasing from 1% in 1962 to 16% in 2000.Analysis conducted in the previous three sections focuses on the determinants of the structural transformation in the selected three countries based on extending measures developed by HK. The analysis has shown that the proximity of the export baskets (which represents the current economic structure) of China and Malaysia to more complex goods, such as capital goods, consumer durables, and intermediate inputs, and the higher values of the new goods in these industrial clusters have been major determinants of the patterns and rate of transformation in these two economies. Now we focus on the institutional factors associated with the transformation process. Specifically, we focus our discussion on whether the timing of policy and institutional reforms in the three countries matches with the patterns of structural transformation shown in figures 3, 5, and 6.For background purposes, we draw upon a series of papers by Rodrik (2004Rodrik ( , 2006aRodrik ( , 2006b) in which he focuses on the role of policies and institutions in driving innovation. Obviously, economic fundamentals such as initial factor endowments, macroeconomic stability, and well-functioning markets are important factors in understanding what a country will produce, but those factors alone appear insufficient in explaining the rate of structural transformation or, in the case of Ghana, structural stagnation. Rodrik advances the notion of the existence of information and coordination externalities that impede a \"jump\" to innovative activities. These externalities require government intervention through policy and institutional reforms in order to alleviate constraints to the design and investment in new products.An overview of Chinese economic history over the last four decades reveals the most dramatic change in institutions and policies in recent world history. China's market-related institutions as well as policies are generally recognized as being among the most backward before the early 1980s. With a socialist regime prevailing until the early 1980s, government employed a number of direct and indirect instruments to control most economic activities, which also made the economy relatively closed to world markets. Although imports and exports did exist, as shown in the data, they were not only controlled and determined by the government, but also imports tended to be limited necessities with exports serving as a source to provide foreign exchangeThe paths describing changes in China's production and export structure are shown in Figures 3,5,and 6. They show the different patterns in both the structure and change in the structure of the economy between the two sub-periods, pre-reform and after the reform that started in the early 1980s. Moreover, the export structure in the pre-1980 period partially reflected the state-driven industrialization process under the socialist regime, which resulted in a relatively high share of intermediate goods and some capital goods in total exports (Figure 6).The most impressive dynamics in all the figures of this paper are observed in the second subperiod for China, the period starting in the mid-1980s. At that point, China initiated reform of both its policies and institutions. Many other impressive facts of this sub-period, which are widely recognized and not presented in the previous sections, include China's persistent double-digit growth in GDP, an extraordinarily high share of GDP involved in trade,13 a high level of foreign direct investment (FDI) inflows, and the increasing market power in the world economy. While it is difficult to list all the factors the literature has associated with China's economic evolution, we summarize the following factors that are most relevant to this study. According to Lo and Chan (1998), Kraemer and Dedrick (2001), Prasad (2004), Sutton (2004), andRodrik (2006a), China has succeeded by taking advantage of key fundamentals in the reform period, including the relatively large size of markets for many products, abundant and low-cost labor, relatively low material costs, a relatively high level of human capital thanks to the education system developed under the socialist regime, and high saving rates that link to the nature of Chinese culture. These factors together with the process of gradual policy and institutional reforms have provided incentives for enterprises to increase investments in new activities primarily guided by the market signals. And most importantly, these factors have attracted foreign investors to participate in such activities and hence to bring in new technology know-how and designs that China could exploit to foster its growth and transformation process. While the fundamentals played an important role in realizing China's miracle, almost all these factors existed before the reforms, but they alone were not sufficient to create the miracle. Without the \"right\" institutions and policies that opened the economy to foreign markets and multinational enterprises, the fundamentals appear, in hindsight, unable to launch the process of growth and transformation. However, the so-called \"right\" institutions and policies had to be tailored to China's case, and they are often not the copies of those that prevailed in the advanced economies from which China has \"copied\" technology know-how.The important roles of China-specific \"right\" policies and institutions have attracted the attention of many. The literature documents a number of issues, including a gradual reform process that lowered economic risk associated with rapid reform; the creation of special economic zones (SEZs) in the early 1980s aimed at attracting foreign investment; the gradual and eventual increases in the number of SEZs and the adoption of policies and institutions that are more attractive for FDI such as duty-free access to imports of intermediate goods for exports from SEZs, various tax exemptions with a relatively longer period for foreign companies, gradual liberalization in labor mobility, and the determination of wages in SEZs; the eventual enforcement of contracts; the limited direct intervention of government; and the fostering of the spillovers of such new policies and institutions beyond the boundaries of SEZs to almost the entire country.While the market forces created by the reforms are driving forces of the structural transformation, a set of safeguard measures were employed to foster technological transfers to domestic firms. Those safeguards, for example, required multinational firms to achieve certain local product content over an agreed-upon period of time. Further, the safeguards also sought to restrain the volume of multinational firms' sales in domestic markets to a certain proportion of their exports. This promoted exports and allowed domestic firms to have the time and opportunity to \"copy\" or adapt the technology know-how and design associated with the new products. The safeguards also forced multinational firms to enter into joint ventures with local entrepreneurs in the early stage of reform. Thus, as discussed by Rodrik (2006a), the per capita income typically associated with the type of goods that China exports is much higher than China's actual income, indicating that the skill content of China's exports is likely to be much higher than its endowment may imply. By the year 2000 (the endpoint of our data set), a majority of the world's large multinational firms were active in China either through their subsidiaries or in joint ventures, and a majority of such multinational firms' products were high-tech and destined for exports. The industries in which these firms are actively involved include mobile phones (Motorola, Nokia, TCL, Sagem, Samsung, and Siemens), personal computers (Acer, Arima, AST, Compal, Compaq, DEC, Dell, Epson, FIC, GVC, HP, Huashang, IBM, Quanta, and Toshiba), home electronics (Sony, Philips, Toshiba, TCL, Siemens, Samsung, Electrolux, LG, Mitsubishi, Sanyo, Sigma, and Toshiba Carrier), and automobiles (VW Automotive, Citroen, GM, Honda, and Daihatsu). The number of privately owned domestic firms has grown and resulted in the development of high-tech domestic industries. 14 Although China remains mainly an assembler of imported components for high-tech products that are exported with a relatively low value-added component (Koopman et al. 2008), the country has become more integrated into the global production chain that it depends on for the production and exports of other high-value and hightech products.China's patterns of structural transformation analyzed in the previous sections seem consistent with its institutional and policy \"transformation\" discussed in literature and summarized in this section. Obviously, the policy and institutional environment created by the reforms after the mid-1980s provided incentives for both domestic and foreign enterprises to be actively involved in the development process, the consequences of which are observed in the data and analyzed in the previous sections. This match in timing and in patterns between transforming the economy and reforming the institutions provides strong support to the argument that policies and institutions have played the most important role in the extent and rate of transformation.The important role of export-led industrialization in explaining the rapid transformation of many East Asian economies into the category of newly industrialized countries (NICs in both the first and second tiers) has received considerable attention in development literature. Malaysia is one such country. Starting as early as the 1960s, Malaysia adopted a series of policies and institutional arrangements aimed at promoting its most promising export sectors to become the vanguard for the country's industrialization process. Similar to developments in other East Asian countries, these reforms consisted of promoting private entrepreneurship and opening the economy to international competition through the promotion of exports while gradually reducing import restrictions (Carbaugh 2002). The early reforms also included a series of tax incentive policies initiated in the early 1970s and the creation of export zones aimed at attracting foreign investors and technology transfers to domestic firms. In 1972, Malaysia created its first free-processing zones, where multinational firms were exempted from import duty, sales and excise taxes (Rasiah 2004). By 1975, the success of these reforms featured many multinational firms and numerous joint ventures and subsidiaries and helped the country evolve from an agricultural-and-primary-resourcebased economy to a producer of many capital and consumer durable goods. In the early 2000s, it was ranked the fifth most competitive economy in Asia after Singapore, Hong Kong, Japan, and China (MIDA 2005). It was also ranked the world's ninth largest personal computer producer in 1999 (Kraemer and Derrick 2001) and one of the world's top five exporters of semiconductor devices in 2000 (MIDA 2004).In terms of the capital goods sector, Malaysia is a major producer of industrial and nonindustrial machines (heavy and precision engines) and transportation equipment (such as buses, refrigerator trucks, and so on). It has attracted multinational firms in aerospace activities (Boeing, General Electric, Honeywell Aerospace, Parker Hannifin, MTU Maintenance, Hamilton Standard, and Eurocopter) and developed an aerospace industry that assembles, maintains, and repairs light aircraft and manufactures aircraft parts and components. Tax incentives encouraged these multinational firms to extend their activities to shipbuilding (yachts, jet skis, sail-and speedboats, inboard/outboard boats, canoes, barges, trawlers, ferries, and cement carriers) and ship repair. The production of these more complex capital goods has helped transform the structure of Malaysia's industry and the composition of its exports.The timing of institutional and policy reforms in Malaysia seems to coincide with the movements of Malaysian industry toward capital and consumer durable products described in Figure 6. This is an indication that these policies and institutions may have determined the patterns of change of the structure of Malaysian exports. Figure 6 shows that Malaysia started developing its capital goods cluster in the mid-1970s but waited until the early 1980s to foster a faster rate of expansion. By the mid-1980s, this cluster grew at a high rate, resulting in an export share of 66% in 2000. The consumer durables cluster also experienced expansion in its share of total exports. However, that expansion was not sufficient to change its share in foreign trade. Although this cluster moved from the fifth largest cluster in 1975 to the third largest cluster in 2000, its share in exports was only 10%.The slow expansion of the consumer durables cluster can in part be attributed to structural and institutional factors that counteracted the otherwise beneficial effects of multinational firms in the economy. Among those factors are the shortage of labor as well as of skills, 15 low investment in R&D necessary to absorb foreign technology, 16 the high costs of inputs due to local content requirement (Kuchiki 2007), low competitiveness in international markets due to a high level of protection of some industries in this cluster, 17 and the Asian financial crisis of 1998. These factors may explain why, having begun its structural transformation in the 1960s and reached a middle-income standard of living in the 1980s, and hence prior to China's \"takeoff,\" Malaysia has not advanced as rapidly as China in the entire period. 18As the first independent country in Sub-Saharan Africa in the late 1950s, the initial conditions in Ghana were obviously very different from those either in China or in Malaysia. As a major producer and exporter of cocoa and gold during the colonial period, there were little social, economic, and institutional assets left from such history that an independent Ghana could rely on to initiate its modernization. Thus, from this point of view, a lack of fundamentals can explain the lack of transformation in Ghana, at least in the early years of the period we study. However, after many lost years between the 1960s and 1980s, Ghana started its policy and institutional reform in the late 1980s, and in the years following the reform saw sustainable economic growth with a reduction in poverty. While both the total and per capita growth rates cannot compare with those of China and Malaysia, they are high compared with the country's own history and other countries in Africa (Breisinger et al. 2008).Although the country has experienced persistence in economic growth over the last 20 years, its economic structure has not changed appreciably in terms of the share of agricultural and manufacturing in national product or in terms of export structure (Breisinger et al. 2008). It may be unrealistic to expect a country like Ghana that faces human capital and infrastructural constraints to experience rapid structural change within a period of 20 years. However, the factors necessary for transformation but not yet in place are worth discussion. The former discussion suggests that structural transformation has been led by both domestic and foreign enterprises investing in new goods with high value that the country did not previously produce. There are certain necessary conditions to provide enterprises an incentive to be innovative and invest in new products. Such conditions include both physical infrastructure and institutional conditions. Ghana's colonial history and political and social instabilities in the first 30 years of its postcolonial history have left the country with extremely poor infrastructure, such as roads, electrification, and efficient adjudication of commercial disputes and in many other basic conditions for doing business. These together with the lack of human capital and basic skills are the result of little public investment in education and health, which has made relatively costly the supply of effective labor services to private enterprises, both domestic and foreign. Realizing such constraints in development, the government of Ghana has rapidly increased investment in all these aspects over the last 10 years. However, the growth rate in providing these necessary conditions still falls short of demand even under the current economic structure.In terms of policy and institutional conditions, Ghana has pursued an open economic policy in the last 20 years through both policy and institutional reforms. The country has liberalized its economy and trade; privatized almost all state-owned firms; pursued the control of inflation through needed to achieve reverse engineering. For instance, it had an average of only 400 scientists per million population by the end of 1990s, a number far below the standard in industrialized countries, which is between 4,000 and 6,000 scientists and engineers per million. Rasiah (2004) points out that a low supply of human capital prevents the movement of firms toward higher R&D activities. Sadoi (2000), on the other hand, attributes the problem partially to the Malaysian worker attitude toward skill upgrading. He points out that a Malaysian worker pays less attention to precision and is less motivated to learn by doing than a worker of an industrialized country. 16 According to Rasiah (2004), the R&D intensity of electronics products in Malaysia (0.088) was far below that of Taiwan (0.546) and of Korea (0.212) in 2000.17 Kuchiki (2007) documents tariffs on vehicles that range from 40% to 300% in 1998. 18 For example, Malaysia was the world's ninth largest producer of PCs with a share of 2.8% in 1999, lagging four places behind China (fifth largest producer with a share of 5.5%) despite having made a technology jump to PCs earlier than China (Kraemer and Dedrick 2001). macroeconomic stabilization; and imposed a series of policies to encourage foreign investment including tax exemptions, protection of foreign companies' intellectual property rights, guarantee of free transfer of capital, profits, and dividends abroad, and guarantees against expropriation and nationalization.In spite of such reforms and incentive policies, Ghana has not succeeded in attracting foreign investors interested in investing and producing new goods of higher value. The cumulative value of FDI from 1994 to 2000 was only $1.32 billion (U.S. Commercial Service 2004), and most of that investment was in mining. The limited magnitude of foreign investment that is concentrated in producing a few primary products is unlikely to allow for the transfer of technology needed to transform the Ghanaian industry. That new investors are mainly interested in mining is an indicator that the country may be suffering from the natural resource curse (Auty 1993;Sachs and Warner 1995), which reduces the competitiveness of other sectors in the economy. Despite abundant natural resources (gold, timber, industrial diamond, bauxite, manganese, rubber, timber, petroleum, and so on), Ghana's mismanagement and inefficient use of such resources has failed to boost the economy through the improvement of infrastructure and investment in human capital and in R&D. It has also distorted the allocation of resources across sectors, preventing the promotion or creation of the manufacturing subsectors that have a relatively short distance to the more sophisticated and high-value goods. Without additional policies, the current industrial structure and relevant institutional factors will continue to prevent foreign investment flows into more complex activities. Thus, the initial industrial structure of Ghana is a major challenge if the country is to follow the paths of China and Malaysia. As shown in the previous sections, a light manufacturing industry (electronics and car assemblies) existed in China and Malaysia prior to policy and institutional reforms. Such light manufacturing exhibits a relatively short distance to other more sophisticated new goods, and hence firms can more easily \"jump\" between the trees in the relatively dense forest. That made it easier for China and Malaysia to move from their previous industrial structure to a more sophisticated one as compared with Ghana.This paper contributes to the literature first by developing new measures to analyze the dynamics of a country's structural transformation as reflected by the changing structure and value of a country's exports in the context of the evolution of the world economy. The second contribution is to provide insights into the features and determinants of transformation of the Chinese, Malaysian, and Ghanaian economies. The new measures are an extension of those developed by Hausmann and Klinger. We find that China's relatively rapid structural transformation is determined by the high proximity of its export basket to capital goods, consumer durables, and intermediate inputs coupled with high values of new goods in these three clusters. In the 924 new products in which China has developed RCAs and exported during 1962-2000, 648 (70%) belonged to these top three industrial clusters. Many products in the three clusters embody high levels of technological know-how that tends to facilitate further movement toward more complex goods. As a result, not only has China transformed its industrial structure and developed an export profile that is skewed toward goods often associated with advanced economies, but the profile also increases China's potential to sustain this path.Malaysia started its transformation process earlier than China and achieved industrial clusters exhibiting sophisticated export profiles that by the year 2000 also resembled those of advanced economies. In its 529 new products exported over 1962-2000, 352 (67%) belonged to the top three industrial clusters mentioned earlier. However, certain structural factors appear to have impeded a faster expansion of industries in the country's consumer durables cluster. Toward the end of the study period, Malaysia lags behind China in the advanced goods component of its export profile.In contrast to the two former countries, the transformation of the Ghanaian economy appears far behind in new product content and increasing value. Only 245 new products appear in the country's export basket in the 39 years after 1962, of which only 13 are capital goods. Ghana's export profile and hence economic structure are dominated by agricultural and other primary products throughout the period. The relatively less technical nature of these clusters prevents the country from advancing to a product space featuring more complex products, and hence slows the evolution of its industrial structure.This study also discusses the role policies and institutions may have played to improve the proximity of each country's existing production structure to more innovative and sophisticated activities. Drawing from existing literature, the discussion suggests that policy and institutional reforms seem to be key factors in inducing the transfer and absorption of foreign technology, allowing both China and Malaysia to advance more rapidly into the capital and consumer durable goods components of the forest. Ghana offers interesting insights in contrast to those two countries. In its case, unfavorable initial conditions and a relatively short life of transformation have not enabled the policy and institutional reforms initiated in the late 1980s and early 1990s to stimulate anywhere near a rapid transformation of the economy. Although Ghana significantly improved its investment environment to analyze the dynamics of structural transformation in China, Malaysia, and Ghana can be extended to the other developing countries in the Feenstra et al. (2005) data set. Such generalization will help classify these countries based on the nature and patterns of their comparative advantages, and hence on the similarity of their structural transformation processes. With such a classification, it may be possible to recommend to similar countries policies that facilitate a stepwise jump of private enterprises into innovative activities in order to accelerate the structural transformation process. The patterns of this transformation may differ from those exhibited by China and Malaysia, but they will at least lay the foundation for the development of RCAs in goods with high technological intensity instead of just primaries. products A and X in total exports of more than 5%. Once a country satisfies the above criterion, it is dropped in the data set not only in that particular year but in all other years. In HK (2006), it is not obvious how the selection of countries on the basis of the 5% threshold is done. ","tokenCount":"10728","images":["57611989_1_1.png","57611989_1_2.png","57611989_1_3.png","57611989_1_4.png","57611989_1_5.png","57611989_1_6.png","57611989_1_7.png","57611989_1_8.png"],"tables":["57611989_1_1.json","57611989_2_1.json","57611989_3_1.json","57611989_4_1.json","57611989_5_1.json","57611989_6_1.json","57611989_7_1.json","57611989_8_1.json","57611989_9_1.json","57611989_10_1.json","57611989_11_1.json","57611989_12_1.json","57611989_13_1.json","57611989_14_1.json","57611989_15_1.json","57611989_16_1.json","57611989_17_1.json","57611989_18_1.json","57611989_19_1.json","57611989_20_1.json","57611989_21_1.json","57611989_22_1.json","57611989_23_1.json","57611989_24_1.json","57611989_25_1.json","57611989_26_1.json","57611989_27_1.json","57611989_28_1.json","57611989_29_1.json","57611989_30_1.json","57611989_31_1.json","57611989_32_1.json","57611989_33_1.json","57611989_34_1.json","57611989_35_1.json","57611989_36_1.json","57611989_37_1.json","57611989_38_1.json","57611989_39_1.json","57611989_40_1.json"]}

data/part_2/0475395660.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"ba85cb42fff42734bc403746f9c50162","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/9b2810fb-c8aa-43bd-81f7-47dbed6b89f8/retrieve","description":"In this issue of Abstract Digest, we present to you a collection of articles on various aspects of nutrition, including its determinants, tools of measurement, and strategic responses through policy and program. Here are some of the highlights.","id":"-1317551948"},"keywords":[],"sieverID":"59f49343-abb1-4e15-b297-489f60218899","pagecount":"17","content":"Digest, we present to you a collection of articles on various aspects of nutrition, including its determinants, tools of measurement, and strategic responses through policy and program. Here are some of the highlights:• Two studies have examined the influences of maternal depression on children's growth:o Based on evidence from rural India, Nguyen and colleagues (2018) demonstrated that maternal depressive symptom is significantly associated with both child undernutrition and development delay, calling for practical interventions to address maternal depression to simultaneously target multiple outcomes for both women and children. o From a birth cohort in India, Babu and colleagues (2018) found an association between antepartum depression and small for gestational age (SGA) babies, thus highlighting the need to explore implementation of screening, diagnostic services and evidence-based antenatal mental health services by modifying the provisions of ongoing national programs.• Harding and colleagues ( 2018) explored the factors associated with wasting among children aged 0 to 59 months and found that the underlying determinants of wasting and stunting in South Asia are similar, therefore cost-effective interventions to prevent both stunting and wasting, and to treat severe wasting, need to be scaled up urgently without separating these two manifestations of child undernutrition. • Using nationally representative data from the Multiple Indicator Cluster Surveys in 31 lowand middle-income countries (LMICs), Jeong and colleagues (2018) found links between birth registration and children's early growth and development outcomes, emphasizing that efforts to increase birth registration may be promising for promoting early childhood development in LMICs. • From a systematic review of cohort studies and a meta-analysis, Khalil and colleagues (2018)   demonstrated that there is a causal relation between childhood diarrhoea caused by Cryptosporidium infection and childhood growth, calling for interventions designed to prevent and effectively treat infection in children younger than five years.• Richter and colleagues (2018) analysed longitudinal data from four birth cohort studies in low-and middle-income countries and found that early child growth faltering is determined by both biological and social factors, underscoring the need for interventions that address both biological and social determinants over the long and short term.• Using country-level data and an ecological study design, Harding and colleagues (2018)   explored the global associations between women's educational attainment and the micronutrient status of children, women and the general population, drawing policyrelevant connections between formal education, anemia and micronutrient status globally.• Knowles and colleagues (2018) conducted regression analyses of data from stratified, cluster sample, and household iodine surveys in four countries to identify factors associated with household access to adequately iodised salt and concluded that improving household access to refined iodised salt in sealed packaging would improve iodine intake from household salt. • Dror and Allen (2018) conducted a review of current knowledge and summarized how nutrient concentrations change through the initiation and progression of lactation, and howMaternal depressive symptoms are negatively associated with child growth and development: Evidence from rural India Nguyen, P.H., J. Friedman, M. Kak, P. Menon, and H. Alderman. 2018. Maternal & Child Nutrition. doi: https://onlinelibrary.wiley.com/doi/full/10.1111/mcn.12621.Maternal depression has been suggested as a risk factor for both poor child growth and development in many low-and middle-income countries, but the validity of many studies is hindered by small sample sizes, varying cut-offs used in depression diagnostics, and incomplete control of confounding factors. This study examines the association between maternal depressive symptoms (MDSs) and child physical growth and cognitive development in Madhya Pradesh, India, where poverty, malnutrition, and poor mental health coexist. Data were from a baseline household survey (n = 2,934) of a randomized controlled trial assessing an early childhood development programme. Multivariate linear and logistic regression analyses were conducted, adjusting for socio-economic factors to avoid confounding the association of mental health and child outcomes. MDS (measured using the Center for Epidemiologic Studies Short Depression Scale) was categorized as low, medium, and high in 47%, 42%, and 10% of mothers, respectively. The prevalence of child developmental delay ranged from 16% to 27% for various development domains. Compared with children of mothers with low MDS, those of high MDS mothers had lower height-for-age, weight-for-age, and weight-for-height z-scores (0.22, 0.21, and 0.15, respectively), a higher rate of stunting and underweight (~1.5 times), and higher rate of developmental delay (partial adjusted odds ratio ranged from 1.3-1.8 for different development domains and fully adjusted odds ratio = 1.4 for fine motor). Our results-that MDS is significantly associated with both child undernutrition and development delay-add to the call for practical interventions to address maternal depression to simultaneously address multiple outcomes for both women and children.• Maternal depressive symptoms are significantly associated with child undernutrition in India, with possible mechanisms operating though poorer home environment, less engagement with children, lower use of health services, and suboptimal complementary feeding practices.• Depressive symptoms also are associated with development delay among children <4 years, although the strength of the association depends on the domain of development considered. • As poverty, poor mental health, and undernutrition coexist in India, it is crucial to consider maternal depression in endeavours to simultaneously address multiple outcomes for both women and children.Small for gestational age babies and depressive symptoms of mothers during pregnancy: Results from a birth cohort in India Babu, G.R., G.V.S. Murthy, Y. Reddy, R. Deepa, A. Yamuna, S. Prafulla, M. Rathnaiah, and S. Kinra. 2018. Wellcome Open Research 3(76). doi: https://wellcomeopenresearch.org/articles/3-76.Background: More than one million babies are born with Low birthweight (LBW) in India every year, often afflicting disadvantaged families. Earlier studies on LBW in India have mostly focused on addressing poverty, nutritional status, and obstetric factors for LBW babies, comprising of preterm babies (<37 weeks) or small for gestational age (SGA) or both. We aim to find the association between antepartum depression and SGA in a public hospital. Methods: Pregnant women with gestational age between 14 to 32 weeks were recruited. The Edinburgh Postnatal Depression Scale (EPDS) was administered to assess depression. Newborn anthropometry was performed soon after delivery. Birth weight less than 10 percentile were classified as SGA, between 10 to 90th percentile was appropriate for gestational age (AGA), and greater than 90th percentile was large for gestational age (LGA). Results: We found that 16.51% (108) of the antenatal mothers had depressive symptoms (EPDS score >11). The women with depressive symptoms delivered a greater proportion of SGA babies (21.3 v/s 15.8) and LGA (9.3 v/s 3.3) compared to women with no symptoms. The odds of women giving birth to a child with SGA were twice as high for women with EPDS scores >11 (adjusted OR = 2.18; 95% CI = 1.23 -3.87) compared to the women with EPDS scores of ≤11. In terms of Area under curve (AUC), EPDS 11 cut off (AUC: 0.533) narrowly outperformed EPDS 12 cutoff (AUC: 0.4738), which in turn was better than EPDS 13 cut off (AUC: 0.4687) for screening depression in mothers. Conclusions: We have demonstrated the usefulness of the 10-item EPDS screening tool in screening for symptoms of antenatal depression. There is a need to explore implementation of screening, diagnostic services and evidence-based antenatal mental health services by modifying the provisions of ongoing national programs. South Asia continues to carry the greatest share and number of wasted children worldwide.Understanding the determinants of wasting is important as policymakers renew efforts to tackle this persistent public health and development problem. Using data from national surveys in Bangladesh, India, the Maldives, Nepal, Pakistan and Afghanistan, this analysis explores factors associated with wasting among children aged 0 to 59 months (n = 252,797). We conducted multivariate mixed logistic regression and backwards stepwise methods to identify parsimonious models for each country separately (all p values <0.05). Younger children (0 to 5 months), and those whose mothers had a low body mass index (<18.5 kg/m2) had greater odds of being wasted in all countries. Later birth order, being male, maternal illiteracy, short maternal stature, lack of improved water source, and household poverty were also associated with wasting in various countries, but not systematically in all. Seasonality was also not consistently associated with wasting in the final models. These findings suggest that pre-conception (adolescence), pregnancy and early postpartum, represent windows of opportunity for tackling child wasting, not only stunting. Our analysis suggests that the underlying determinants of wasting and stunting in South Asia are similar, but not universal across geographies. Cost-effective interventions to prevent both stunting and wasting, and to treat severe wasting, need to be scaled up urgently. Separating these two manifestations of child undernutrition in conceptual and programmatic terms may unnecessarily impair progress to reach the Sustainable Development Goals targets aimed at addressing both child stunting and wasting.Associations between birth registration and early child growth and development: evidence from 31 low-and middle-income countries Jeong, J., A. Bhatia, and G. Fink. 2018. BMC Public Health 18(673). https://doi.org/10.1186/s12889-018-5598-z Background: Lack of legal identification documents can impose major challenges for children in lowand middle-income countries (LMICs). The aim of this study was to investigate the association between not having a birth certificate and young children's physical growth and developmental outcomes in LMICs. Methods: We combined nationally representative data from the Multiple Indicator Cluster Surveys in 31 LMICs. For our measure of birth registration, primary caregivers reported on whether the child had a birth certificate. Early child outcome measures focused on height-for-age z-scores (HAZ), weight-for-age z-scores (WAZ), weight-for-height z-scores (WHZ), and standardized scores of the Early Childhood Development Index (ECDI) for a subsample of children aged 36-59 months. We used linear regression models with country fixed effects to estimate the relationship between birth registration and child outcomes. In fully adjusted models, we controlled for a variety of child, caregiver, household, and access to child services covariates, including clusterlevel fixed effects. Results: In the total sample, 34.7% of children aged 0-59 months did not possess a birth certificate. After controlling for covariates, not owning a birth certificate was associated with lower HAZ (β = - 0.18; 95% CI: -0.23, - 0.14), WAZ (β = - 0.10, 95% CI: -0.13, - 0.07), and ECDI zscores (β = - 0.10; 95% CI: -0. Background: The protozoan Cryptosporidium is a leading cause of diarrhoea morbidity and mortality in children younger than 5 years. However, the true global burden of Cryptosporidium infection in children younger than 5 years might have been underestimated in previous quantifications because it only took account of the acute effects of diarrhoea. We aimed to demonstrate whether there is a causal relation between Cryptosporidium and childhood growth and, if so, to quantify the associated additional burden. Methods: The Global Burden of Diseases, Injuries, and Risk Factors study (GBD) 2016 was a systematic and scientific effort to quantify the morbidity and mortality associated with more than 300 causes of death and disability, including diarrhoea caused by Cryptosporidium infection. We supplemented estimates on the burden of Cryptosporidium in GBD 2016 with findings from a systematic review of published and unpublished cohort studies and a meta-analysis of the effect of childhood diarrhoea caused by Cryptosporidium infection on physical growth. Background: Early growth faltering accounts for one-third of child deaths, and adversely impacts the health and human capital of surviving children. Social as well as biological factors contribute to growth faltering, but their relative strength and interrelations in different contexts have not been fully described. Objective: The aim of this study was to use structural equation modelling to explore social and biological multidetermination of child height at age 2 y in longitudinal data from 4 birth cohort studies in low-and middle-income countries. Methods: We analyzed data from 13,824 participants in birth cohort studies in Brazil, India, the Philippines, and South Africa. We used exploratory structural equation models, with height-for-age at 24 mo as the outcome to derive factors, and path analysis to estimate relations among a wide set of social and biological variables common to the 4 sites. Results: The prevalence of stunting at 24 mo ranged from 14.0% in Brazil to 67.7% in the Philippines. Maternal height and birthweight were strongly predictive of height-for-age at 24 mo in all 4 sites (all P values <0.001). Three social-environmental factors, which we characterized as \"child circumstances,\" \"family socioeconomic status,\" and \"community facilities,\" were identified in all sites. Each social-environmental factor was also strongly predictive of height-for-age at 24 mo (all P values <0.001), with some relations partly mediated through birthweight. The biological pathways accounted for 59% of the total explained variance and the social-environmental pathways accounted for 41%. The resulting path coefficients were broadly similar across the 4 sites. Conclusions: Early child growth faltering is determined by both biological and social factors. Maternal height, itself a marker of intergenerational deprivation, strongly influences child height at 2 y, including indirect effects through birthweight and social factors. However, concurrent social factors, many of which are modifiable, directly and indirectly contribute to child growth. This study highlights opportunities for interventions that address both biological and social determinants over the long and short term.Harding, K.L., V.M. Aguayo, W.A. Masters, and P. Webb. 2018. BMC Public Health 18: 470. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5894221/ Background: Formal education can be a nutrition-sensitive intervention that supports the scale-up and impact of nutrition-specific actions. Maternal education has long been linked to child survival, growth, and development while adult earnings and nutrition are tied to years in school as a child. However, less is known about the relationship between maternal education and the micronutrient status of children, women and the general population. Methods: Using country-level data and an ecological study design, we explored the global associations between women's educational attainment and: a) anemia and vitamin A deficiency (VAD) in children aged 6-59 months; b) anemia in non-pregnant women; and c) zinc deficiency, urinary iodine excretion (UIE), and the proportion of infants protected against iodine deficiency in the general population Cross-sectional relationships (2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013) were assessed using linear regression models. Results: Percentage of women without schooling was negatively associated with all outcomes. Number of years of schooling among women was positively associated with all outcomes except for UIE and the proportion of infants protected against iodine deficiency. Income level was a significant effect modifier of the effect of years of women's schooling on child anemia as well as of the proportion of women without formal education on zinc deficiency in the population. The relationship was strongest in low-income countries for child anemia, and was not significant in upper middle-income countries. For zinc deficiency, the relationship was not significant in low or lower middle income countries, which may suggest that a minimum threshold of resources needs to be reached before education can influence zinc status. Conclusions: While relationships between maternal schooling and micronutrient outcomes vary around the globe, more schooling is generally linked to lower rates of deficiency. These findings draw policy-relevant connections between formal education and anemia and micronutrient status globally. It is necessary to examine the mechanisms through which this relationship may be working at both household and country level. Regression analyses of data from stratified, cluster sample, household iodine surveys in Bangladesh, India, Ghana and Senegal were conducted to identify factors associated with household access to adequately iodised salt. For all countries, in single variable analyses, household salt iodine was significantly different (p < 0.05) between strata (geographic areas with representative data, defined by survey design), and significantly higher (p < 0.05) among households: with better living standard scores, where the respondent knew about iodised salt and/or looked for iodised salt at purchase, using salt bought in a sealed package, or using refined grain salt. Other country-level associations were also found. Multiple variable analyses showed a significant association between salt iodine and strata (p < 0.001) in India, Ghana and Senegal and that salt grain type was significantly associated with estimated iodine content in all countries (p < 0.001). Salt iodine relative to the reference (coarse salt) ranged from 1.3 (95% CI 1.2, 1.5) times higher for fine salt in Senegal to 3.6 (95% CI 2.6, 4.9) times higher for washed and 6.5 (95% CI 4.9, 8.8) times higher for refined salt in India. Subnational data are required to monitor equity of access to adequately iodised salt. Improving household access to refined iodised salt in sealed packaging, would improve iodine intake from household salt in all four countries in this analysis, particularly in areas where there is significant small-scale salt production. Exclusive breastfeeding is recommended by the WHO for the first 6 mo of life because human milk protects against gastrointestinal infections and supplies balanced and adequate nutrient contents to the infant. However, reliable data on micronutrient concentrations in human milk are sparse, especially because some micronutrients are affected by maternal diet. Microbiological and competitive protein-binding assays, nuclear magnetic resonance or inductively coupled plasma spectroscopy, and chromatographic analyses are among the methods that have been applied to human-milk micronutrient analysis. However, the validation or evaluation of analytical methods in terms of their suitability for the complex human-milk matrix has been commonly ignored in reports, even though the human-milk matrix differs vastly from blood, plasma, or urine matrixes. Thus, information on the validity, accuracy, and sensitivity of the methods is essential for the estimation of infant and maternal intake requirements to support and maintain adequate milk micronutrient concentrations for healthy infant growth and development. In this review, we summarize current knowledge on methods used for analyzing water-and fat-soluble vitamins as well as iron, copper, zinc, iodine, and selenium in human milk and their different forms in milk; the tools available for quality control and assurance; and guidance for preanalytical considerations. Finally, we recommend preferred methodologic approaches for analysis of specific milk micronutrients.Grellety, E., and M.H. Golden. 2018. Emerging Themes in Epidemiology 15(8). https://doi.org/10.1186/s12982-018-0075-9Background: Representative surveys collecting weight, height and MUAC are used to estimate the prevalence of acute malnutrition. The results are then used to assess the scale of malnutrition in a population and type of nutritional intervention required. There have been changes in methodology over recent decades; the objective of this study was to determine if these have resulted in higher quality surveys. Methods: In order to examine the change in reliability of such surveys we have analysed the statistical distributions of the derived anthropometric parameters from 1843 surveys conducted by 19 agencies between 1986 and 2015. Results: With the introduction of standardised guidelines and software by 2003 and their more general application from 2007 the mean standard deviation, kurtosis and skewness of the parameters used to assess nutritional status have each moved to now approximate the distribution of the WHO standards when the exclusion of outliers from analysis is based upon SMART flagging procedure. Where WHO flags, that only exclude data incompatible with life, are used the quality of anthropometric surveys has improved and the results now approach those seen with SMART flags and the WHO standards distribution. Agencies vary in their uptake and adherence to standard guidelines. Those agencies that fully implement the guidelines achieve the most consistently reliable results. Conclusions: Standard methods should be universally used to produce reliable data and tests of data quality and SMART type flagging procedures should be applied and reported to ensure that the data are credible and therefore inform appropriate intervention. Use of SMART guidelines has coincided with reliable anthropometric data since 2007. Objective: In this secondary analysis of data from an intervention trial, we assessed the performance of Mid Upper Arm Circumference (MUAC) as a predictor of mortality in children aged 6-59 months from Delhi, India, one year after their initial MUAC measurements were taken. Additionally, we assessed MUAC as an absolute value and MUAC z-scores as predictors of risk of mortality. Methods:In the trial, children were screened using MUAC prior to referral to the study clinic. These children were revisited a year later to ascertain their vital status. Baseline MUAC and MUAC z-scores were used to categorize children as severely (MUAC <115 mm, MUAC z-score <-3SD) or moderately (MUAC 115 to <125 mm, MUAC z-score <-2SD) malnourished. The proportion of malnutrition, risk of mortality, relative risk estimates, positive predictive value and area under the curve (AUC) by MUAC and MUAC z-scores were calculated. Results: In the resurvey, the first 36159 children of the 48635 in the initial survey were contacted. Of these, vital status of 34060 (94.2%) was available. The proportion of severe malnutrition by MUAC (<115 mm) was 0.5% with an associated mortality of 4.7% over a one year period and an attributable mortality of 13% while the proportion of the severe malnutrition by MUAC z-score (<-3SDwas 0.9% with an associated mortality of 2.2%. Conclusions: MUAC is a significant predictor of subsequent mortality in under-five children. In settings where height measurement is not feasible, MUAC can be used as a screening tool for identifying severely malnourished children for management.Strategies for optimizing maternal nutrition to promote infant development Hambidge, K.M., and N.F. Krebs. 2018. Reproductive Health 15(Suppl 1): 87. https://doi.org/10.1186/s12978-018-0534-3Background: The growing appreciation of the multi-faceted importance of optimal maternal nutrition to the health and development of the infant and young child is tempered by incompletely resolved strategies for combatting challenges. Objective: To review the importance of maternal nutrition and strategies being employed to optimize outcomes. Methods: Selected data from recent literature with special focus on rationale for and currently published results of maternal nutrition supplements, including lipid based nutrition supplements. Results: 1) An impelling rationale for improving the maternal and in utero environment of low resource populations has emerged to achieve improved fetal and post-natal growth and development. 2) Based partly on population increases in adult height over one-two generations, much can be achieved by reducing poverty. 3) Maternal, newborn and infant characteristics associated with low resource environments include evidence of undernutrition, manifested by underweight and impaired linear growth. 4) Apart from broad public health and educational initiatives, to date, most specific efforts to improve fetal growth and development have included maternal nutrition interventions during gestation. 5) The relatively limited but real benefits of both iron/folic acid (IFA) and multiple micronutrient (MMN) maternal supplements during gestation have now been reasonably defined. 6) Recent investigations of a maternal lipid-based primarily micronutrient supplement (LNS) have not demonstrated a consistent benefit beyond MMN alone. 7) However, effects of both MMN and LNS appear to be enhanced by commencing early in gestation. Conclusions: Poor maternal nutritional status is one of a very few specific factors in the human that not only contributes to impaired fetal and early post-natal growth but for which maternal interventions have demonstrated improved in utero development, documented primarily by both improvements in low birth weights and by partial corrections of impaired birth length. A clearer definition of the benefits achievable by interventions specifically focused on correcting maternal nutrition deficits should not be limited to improvements in the quality of maternal nutrition supplements, but on the cumulative quantity and timing of interventions (also recognizing the heterogeneity between populations). Finally, in an ideal world these steps are only a prelude to improvements in the total environment in which optimal nutrition and other health determinants can be achieved. Background: The uptake of maternal healthcare services by young women in rural India is limited. This study aims to assess the effectiveness of community intervention model to improve the maternal healthcare service uptake of young married couples (15-24 years) in rural India. A three year project was carried out to reach young married women through a multi-pronged community intervention involving sensitizing family members, community mobilization, and capacity building of frontline health functionaries. Methods: The study was conducted among the young married couples aged 15-25 years in states of Uttar Pradesh and Rajasthan of India. A quasi-experimental evaluation design was adopted for this study. Two rounds of cross-sectional surveys at baseline and end line were carried out at both intervention and control sites. Net impact of intervention (Difference-in-Difference and multivariate regression) on key outcomes was assessed adjusting for control variables. Composite maternal healthcare uptake score significantly increased in intervention area compared to control area. Results: Women who were able to discuss about delivery care with family, were five times more likely to go for institutional delivery, also the utilization of maternal health care services was higher among these women (β=1.58). Likelihood of uptake for more than three visits for antenatal care (3+ANC) service indicated three times (OR=3.14, p<0.001) increase in intervention area than those in control area. Regression result on composite maternal health care uptake score significantly increased by 2.5 (β=2.23, p<0.001) in intervention area compared to control area. Conclusion: This study demonstrated that the community intervention to foster enabling environment was effective in improving the awareness and uptake of maternal healthcare services.   for 15-49 year old women with complete data for anthropometric measures in 58 low-income and middle-income countries (LMICs). We compared estimates from multilevel variance component models for BMI before and after adjusting for age and socioeconomic factors (place of residence, education, household wealth, and marital status). The hierarchical structure of the sample included three levels with women at level 1, communities at level 2, and countries at level 3. The primary outcome was BMI. We did a sensitivity analysis using the 2002-03 World Health Surveys. Findings: Of 1212 758 women nested within 64764 communities and 58 countries, we found that most unexplained variation for BMI was attributed to between-individual differences (80%) and the remaining was between-population differences (14% for countries and 6% for communities). Socioeconomic factors explained a large proportion of between-population variance in BMI (14•8% for countries and 47•1% for communities), but only about 2% of interindividual variance. In country-specific models, we found substantial variation in the magnitude of between individual differences (variance estimates ranging from 7•6 to 31•4, or 86•0-98•6% of the total variation) and the proportion explained by socioeconomic factors (0•1-6•4%). The disproportionately large unexplained between individual variance in BMI was consistently found in additional analyses including more comprehensive set of predictor variables, both men and women, and populations from low-income and high-income countries. Interpretation: Our findings on variance decomposition in BMI and explanation by socioeconomic factors at population and individual levels indicate that inferential questions that target within and between populations are importantly inter-related and should be considered simultaneously.  • Despite the rise in diet related chronic diseases and associated costs, government policies continue to have conventional perspectives on agricultural production, industry support, food security, economics, and trade • New, evidence informed government nutrition policies are needed to reduce the risk of chronic diseases and reduce dietary and health inequities • The complementary and synergistic nature of different policies supports the need for an integrated, multicomponent government strategy that uses and adapts existing structures and systems • To translate evidence into action, governments must have the appropriate knowledge, capacity, and will to act and the governance and partnership to support action • Specific actions by major stakeholders should promote, facilitate, and complement policy efforts • Strong government policy is essential to help achieve a healthy, profitable, equitable and sustainable food system that benefits all. It is clear that the current rate of decline in the prevalence of anaemia is insufficient to meet the Global Nutrition Target 20255 and the situation prompts us to re-examine the current approaches for control of nutritional anaemia in the country. Under this premise, the ICMR set up a task force on childhood and adolescent anaemia, to brainstorm, evaluate the evidence and prioritise research questions for immediate implementation, under the chairmanship of Prof. M.K. Bhan. The coordination unit, ICMR commissioned an overview of reviews based on Cochrane and nonCochrane systematic reviews, aiming to contextualize the evidence on Anaemia. Additionally, a trend analysis of existing data from the national surveys was conducted and a document based on recent conferences was prepared and shared with the task force committee. These efforts were geared towards out of the box thinking to result in most appropriate research questions, to be taken up in research mode immediately.","tokenCount":"4597","images":["-1317551948_1_1.png","-1317551948_1_2.png","-1317551948_2_1.png","-1317551948_3_1.png","-1317551948_4_1.png","-1317551948_5_1.png","-1317551948_6_1.png","-1317551948_7_1.png","-1317551948_8_1.png","-1317551948_9_1.png","-1317551948_10_1.png","-1317551948_11_1.png","-1317551948_12_1.png","-1317551948_13_1.png","-1317551948_14_1.png","-1317551948_15_1.png","-1317551948_16_1.png","-1317551948_17_1.png"],"tables":["-1317551948_1_1.json","-1317551948_2_1.json","-1317551948_3_1.json","-1317551948_4_1.json","-1317551948_5_1.json","-1317551948_6_1.json","-1317551948_7_1.json","-1317551948_8_1.json","-1317551948_9_1.json","-1317551948_10_1.json","-1317551948_11_1.json","-1317551948_12_1.json","-1317551948_13_1.json","-1317551948_14_1.json","-1317551948_15_1.json","-1317551948_16_1.json","-1317551948_17_1.json"]}

data/part_2/0480445670.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"0c23f49ec07e80b66339331e193b50e9","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/4783fef0-5ca7-41fe-abd2-bc5a60335511/retrieve","description":"This research report highlights findings from a set of studies applied economists have undertaken on the impact that improved banana cultivars and management practices have had on farmers in the Lake Victoria region of Tanzania and Uganda. A particular focus is on the impact of genetically transformed cooking bananas. Genetic transformation to achieve pest and disease resistance is a promising strategy for smallholder farmers in this region: biotic constraints are severe there and not easily addressed through conventional breeding or control methods. Moreover, exports to world markets are currently negligible, so the risks of reduced exports because of policies against genetically modified foods are low. The crop’s status as both an important food source and significant generator of rural income means that improving productivity could have major social and economic benefits.","id":"-930892376"},"keywords":[],"sieverID":"0c7939a4-245d-4349-9d6e-e8d29fc0af3a","pagecount":"2","content":"T his research report highlights findings from a set of stud- ies applied economists have undertaken on the impact that improved banana cultivars and management practices have had on farmers in the Lake Victoria region of Tanzania and Uganda. A particular focus is on the impact of genetically transformed cooking bananas. Genetic transformation to achieve pest and disease resistance is a promising strategy for smallholder farmers in this region: biotic constraints are severe there and not easily addressed through conventional breeding or control methods. Moreover, exports to world markets are currently negligible, so the risks of reduced exports because of policies against genetically modified foods are low. The crop's status as both an important food source and significant generator of rural income means that improving productivity could have major social and economic benefits.The report's findings demonstrate the potentially pro-poor benefits of transgenic cooking varieties and the likely social consequence of the choice of host cultivar for trait insertion. Also, simulations illustrate the extent to which supporting public investments in education, market infrastructure, and extension would augment farmer demand for new planting material. Demand for planting material of potential host varieties for gene insertion varies according to household and farm characteristics, market access, and the attributes of varieties. The evidence also confirms that adoption of recently developed hybrids reduced the vulnerability of Tanzanian households to yield losses from pests and disease. For farmers with few nonfarm sources of income, reducing production risk can smooth both consumption and income from local sales. In addition, social capital evidently plays a significant role in the use of current best practices for managing soil fertility in Ugandan banana production. Most of the villages in the banana-producing areas foster active social organization, and village social characteristics will likely have especially important implications for planting-material systems of the East African highland banana. For this crop, transfers of planting material and related information are heavily based on farmerto-farmer exchange, compared to the more formal, commercialized systems of some cereal crops.The productivity of cooking bananas depends on the economic, climate, and soil characteristics of regions, and policies will need to be tailored to local conditions. Some evidence suggests that the efficiency of banana production can be improved, particularly in the Central Region of Uganda, where labor productivity is higher than elsewhere in the country. Soil pH and the application of manure have positive and significant effects on productivity, especially in the southwest and central areas of the country. Labor is critically important in banana productivity. Given the presence of surplus labor in the southwest, investment in technology development could shift banana productivity there. Wages are higher for casual labor in the central areas, where a more developed market for unskilled labor exists, creating labor deficits. Extension visits enhance productivity in the southwest, and education improves technical efficiency there, but not in the central areas. Banana production appears to be more efficient when managed by men than by women, probably due to underlying differentials in access to resources. Access to credit increases efficiency, while rent and remittances from other income sources reduce efficiency.Simulations of the gross economic benefits that could be generated by a set of technology options indicates that more widespread adoption of current best practices is likely to generate the greatest investment payoffs. Yet the authors argue that the longer-term strategy endorsed by Uganda's National Agricultural Research Organization, which combines conventional and transgenic approaches to mitigate the biotic pressures that cause major economic losses, is essential for sustaining banana production systems.The results of this study have implications for R&D policies in five categories, as outlined below.Findings confirm the vulnerability of the endemic banana genotypes to pests/diseases and the need for a research policy to commit long-term investments to the development of resistant banana genotypes. The choice of host cultivar and use group will have social consequences in terms of which farmers will be most likely to use and benefit from the technology. In general, however, greater social costs are likely to be inflicted through benefits foregone as the result of delays in banana improvement. Time lags in research and adoption have often been shown to be the single most important parameter determining the social payoff of investments.A policy supporting investments in agricultural education, extension, marketing infrastructure, and access to good roads will enhance demand and supply of improved banana varieties, in turn raising banana productivity and efficiency. We predict a demand for pest-and disease-resistant material, given the evidence that farmers value these traits, especially if other supporting investments in education, extension, and market infrastructure are made.Findings support the current policy of the Government of Uganda, which emphasizes farmer association and human capital development as pillars of technology and knowledge dissemination. Farmers are price-responsive, adopting at greater rates when output prices are high relative to input prices. The high rates of dissemination of improved soilfertility practices are promising, despite the labor they demand. Farmer-and socially based mechanisms appear to be a crucial factor in the dissemination of both planting material and technologies. This finding reflects, in part, the clonally propagated nature of the banana plant.A demand-driven strategy for scaling up farmer use of approved banana varieties is needed. Recently introduced hybrids have been more widely adopted in Tanzania than in Uganda, possibly because of greater disease pressures, heavy dissemination efforts, and farmers' past pursuit of planting material free from pests and diseases. Adoption definitely shows an impact on vulnerability to disease losses. Long-term analysis is needed, however, to determine whether diffusion and benefits are sustained and whether income effects can be observed. We recommend a farmer-and socially based network design, with farmer-supplied planting material, possibly scaling up from some of the experiences in Uganda. The strategy of providing materials free-of-charge in large quantities is not sustainable.In the high-elevation areas, developing and promoting best cultural practices and marketing improvements are of primary importance. In the low-elevation areas, developing and promoting endemic cultivars that are resistant to pests and diseases are priorities, alongside cultural practices focusing on reviving productivity. To support the success of these efforts, major investments will need to be made in dissemination.Technical change is a continuous, multidimensional process. Social science research can support research program design and investors' decisionmaking by identifying the impediments to widespread adoption of promising new technologies by farmers. Genetic transformation offers the rare opportunity to maintain the end-use qualities preferred by East African consumers while enhancing agronomic traits. Given the economic role of the banana plant in the region, the social benefits of technical change will be significant. IFPRI gratefully acknowledges unrestricted funding from the following financial contributors and partners: Australia, Canada, China, Denmark, Finland, France, Germany, India, Ireland, Italy, Japan, Netherlands, Norway, Philippines, Sweden, Switzerland, United Kingdom, United States, and World Bank.Printed on Astrolite PC100 ® paper.To order by post, please fill out and send this coupon to Publication Services at IFPRI. Please send me a copy of Research Report 155:An Economic Assessment of Banana Genetic Improvement and Innovation in the Lake Victoria Region of Uganda and Tanzania, edited by Melinda Smale and Wilberforce Kateera Tushemereirwe.Name/Title Organization Address If your order is not received within 2 weeks (USA) or 6 weeks (outside USA), please let us know.","tokenCount":"1194","images":[],"tables":["-930892376_1_1.json","-930892376_2_1.json"]}

data/part_2/0492504783.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"9b9ca0239c8d1bd2d3158bd48c6d8bd4","source":"gardian_index","url":"https://www.academicjournals.org/article/article1380904171_Katungi%20et%20al.pdf","description":"This work analyses on-farm adjustments in land allocation and intensification in a commercial crop following the increases in market demand in a developing economy. Drawing from the survey conducted among common bean producers in Ethiopia in 2008, a two stage econometric method was used to investigate the contribution of market access and other micro-level factors in facilitating crop intensification and productivity. Ethiopia is the leading commercial producer and exporter of common bean in Africa but also one of the countries in Africa with high levels of soil nutrient depletion. Understanding factors that influence input use and productivity is critical for food security and agricultural sustainability in the country. Based on farm survey data, it was shown that most farmers had expanded their area under common bean but the use of fertilizer and improved varieties was still low. Increase in the intensity of fertilizer and seed use produces an increase in yield and so is market access. Market access has intensification as well as specialization effects on common bean yield. Access to credit, extension and household wealth are other factors that facilitate common bean intensification while risk increasing factors constrain it.","id":"-1956455953"},"keywords":null,"sieverID":"b46e9d5d-3c93-4b1d-a73a-d4d5d240b253","pagecount":"0","content":"This work analyses on-farm adjustments in land allocation and intensification in a commercial crop following the increases in market demand in a developing economy. Drawing from the survey conducted among common bean producers in Ethiopia in 2008, a two stage econometric method was used to investigate the contribution of market access and other micro-level factors in facilitating crop intensification and productivity. Ethiopia is the leading commercial producer and exporter of common bean in Africa but also one of the countries in Africa with high levels of soil nutrient depletion. Understanding factors that influence input use and productivity is critical for food security and agricultural sustainability in the country. Based on farm survey data, it was shown that most farmers had expanded their area under common bean but the use of fertilizer and improved varieties was still low. Increase in the intensity of fertilizer and seed use produces an increase in yield and so is market access. Market access has intensification as well as specialization effects on common bean yield. Access to credit, extension and household wealth are other factors that facilitate common bean intensification while risk increasing factors constrain it.","tokenCount":"190","images":null,"tables":null}

data/part_2/0494152542.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"1a4dc3568d3af38e022f4c9585aeca37","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/e9cb837c-b5f9-4c47-9a57-fef57eb5d354/retrieve","description":"This study applies regression analysis as well as a non-parametric method to survey data from Burkina Faso to analyze the role of human capital in explaining technical efficiency in smallholder agricultural production. Exploiting the panel nature of the data and explicitly treating human capital inputs as endogenous, a two-stage estimation method is used for the analysis of determinants of data envelopment analysis (DEA) technical efficiency scores in a double-bootstrap procedure. Findings suggest that the impact of human capital on technical efficiency differs strongly by gender. Strong positive returns exist for education of females, whereas male education is associated with higher inefficiency. Body mass index of adult females also positively relates to technical efficiency. At the community level, presence of a clinic, connection to the electrical grid, presence of a secondary school, and year-round accessibility of the community are found to be vital for human capital formation.","id":"219755814"},"keywords":["public services","smallholders","human capital","West-Africa","non-parametrics"],"sieverID":"ba4ad99b-009e-48b6-8797-60fd40875814","pagecount":"24","content":"was established in 1975. IFPRI is one of 15 agricultural research centers that receive principal funding from governments, private foundations, and international and regional organizations, most of which are members of the Consultative Group on International Agricultural Research (CGIAR).The need to achieve the United Nations Millennium Development Goals (MDGs) has led to a renewed focus on social-sector investments in Africa and developing countries elsewhere. Due to limited resources, African governments face the key challenge of ensuring consistent policies and strategies to promote long-term economic growth, raise smallholder productivity, achieve food security, and reduce poverty, while at the same time providing the social services that meet immediate welfare requirements (Badiane and Ulimwengu 2009). Policymakers thus face a trade-off between addressing short-term concerns related to the symptoms of poverty and raising smallholder productivity and incomes in the longer term, thereby addressing the root causes of poverty. This trade-off-having to choose between growth and poverty interventions-has been described as the African dilemma (Badiane and Ulimwengu 2009). In the face of tight budget constraints, the only feasible option to escape this dilemma is to devise strategies that maximize the contribution of social services to labor productivity in agriculture and the rural economy.Social services have long been analyzed from an entitlement point of view, with a primary objective of meeting people's welfare needs. Once it has been recognized that social services are not growth-neutral, a crucial question becomes how synergies between social services and investments that directly enhance productivity can be maximized in the short and long run-that is, what are the types of social services that have the largest impact on labor productivity in rural areas and to what extent does the composition of such services affect their impact on labor productivity? In order to maximize convergence-meeting both the income growth needs and the social needs of populations under tight budget constraints-the heterogeneity in social services thus needs to be recognized (Badiane and Ulimwengu 2009). Human capital has been associated with increases in productive capacity of the individual over the long run, while its formation can be attributed in part to social investments. From a policy perspective, two important issues remain: First, there is a need to disentangle production impacts of human capital from efficiency impacts, and second, in recognition of the heterogeneity of social services, there is a need to identify those that play an important role in human capital formation.In this study, I use nationally representative, three-year panel data from Burkina Faso, regression analysis, and nonparametric techniques in a comprehensive analysis of the converging impact of social services related to human capital formation on the technical efficiency of smallholders. Taking into account the endogeneity of human capital formation, I first estimate four ordinary least squares (OLS) regressions for human capital indicators-years of education and body mass index (BMI) by gender-and use predicted values to explain output-oriented measures of technical efficiency in a truncated regression. A double bootstrap is applied to make consistent inference possible and to take account of the bias that arises due to serial correlation of the efficiency terms. Findings suggest that the relation between human capital and technical efficiency differs by gender: A negative relation exists between education of males and household technical efficiency, while this relation is positive for females, for whom greater body mass is also associated with higher efficiency. Community characteristics that are found to play a critical role in the formation of human capital are the presence of a clinic, a secondary school, year-round accessibility, and electricity provision through the grid.The paper is structured as follows. In Section 2, I discuss the relationship between human capital and productivity in agriculture. Section 3 presents the analytical model used to estimate the relation between rural services and human capital indicators as well as between these indicators and technical efficiency. The data and study area are described in Section 4. Section 5 presents the estimation results. I conclude in Section 6 by discussing some of the implications of my findings for understanding the influence of rural services on human capital variables and therefore on technical efficiency in agricultural production.Technical efficiency is a measure of a farm's productive performance. In the context of rural Burkina Faso, it can be defined as the ability of an agricultural household to obtain maximal output from a given set of inputs. Technical inefficiency should be considered as a measure of management error rather than a measure of income or gross output; higher inefficiency does not correspond to lower yields or less income. Human capital inputs have been recognized as critical factors in achieving recent sustained growth in productivity in some African countries (Schultz 2003). Education may enhance technical efficiency directly by improving the quality of labor, by increasing the ability of farmers to adjust to disequilibria, and through its effect on input utilization (Moock 1981). Farmers affected by ill health could experience lower technical efficiency due to impaired work capacity in the field and reduced management and supervision abilities (Antle and Pingali 1994). Farm work, particularly hoe agriculture, is physically demanding; it is thus likely that nutrition affects labor productivity through its effect on the person's energy expenditure level (Strauss 1986).Schooling and weight for height are human capital attributes of farm household members associated with their current productivity. Both forms of heterogeneity are to some degree reproducible. Schooling is created by well-described processes, and weight for height is formed by the biological process of human growth, in which the inputs of nutritional intakes, protection from exposure to disease, health care, and activity levels combine to yield a net cumulative effect on the individual's realization of his or her genetic potential. These characteristics of farm household members are viewed here as indicators of human capital because they can be augmented by social or private investments, but they also vary across individuals because of genetic and environmental factors that are not controlled by the individual, family, or community (Schultz 2003).There are several difficulties with discerning causal links between human capital and technical efficiency in a regression framework. First, endogeneity may arise due to simultaneous effects. Though it may be intuitively appealing to believe that better-nourished and healthier individuals are more efficient, the causality of the relationship between nutrition, health, and productivity is difficult to establish. Improved nutritional status and better health could lead to increased productivity, but it is equally plausible that increased productivity leads to higher incomes, thereby improving nutritional and health status (Garcia and Kennedy 1994). Second, endogeneity could manifest itself if there are exogenous unobserved differences across individuals in their original endowments, and if these endowments influence how parents and children invest in human capital, in either a compensatory or a complementary manner (Schultz 2003). Examples of this phenomenon could be -ability‖ affecting the demand for education (Willis and Rosen 1979) and -frailty‖ affecting the demand for health inputs (Rosenzweig and Schultz 1983). These forms of innate, unobserved heterogeneity could cause a correlation between the human capital inputs and the error in determinants of technical efficiency. Third, in the case of both health status and schooling, there may be lags during which the formation of human capital occurs before a farm household member becomes more productive. These potentially long gestation lags add to uncertainties about the precise quantitative payoff of policy interventions (Schultz 1999).To obtain consistent estimates in the presence of endogeneity, one may use instrumental variable methods in which the instruments are sufficiently correlated with the human capital variables but strictly not correlated with the technical efficiency equation error. Using data from Sierra Leone on local relative prices, household demographic characteristics, and farm assets as instruments to predict household energy intake per capita, Strauss (1986) found that household energy consumption was a positive, significant determinant of farm output. Sahn and Alderman (1988) used a similar approach with data from Sri Lanka. Here again, predicted household energy consumption per capita was used as the measure of nutritional status and related to wage earnings. Interestingly, household energy per capita was a significant, positive determinant of men's but not of women's wages. This differential result between men's and women's productivity is a finding that emerges in a number of the studies linking nutrition to productivity. Sahn and Alderman (1988) also found that years of formal education were positively related to the wage rate. Haddad and Bouis (1991) used BMI and height measures to explain agricultural wage rates. They developed equations to instrument individual energy consumption and BMI, and used these in the wage equations. Identifying instruments for the wage equations were household size, nonfarm income, and distance to the nearest market. Haddad and Bouis (1991) found a lack of impact of short-run nutritional status on agricultural wage determination. However, they qualified this finding by noting that variation in wages across individuals is a crude measure of differences in productivity.Exploiting the panel nature of his data for India to eliminate the potential bias in regression coefficient estimates due to time-invariant, individual-specific effects, Deolalikar (1988) concluded that while the human body can adapt to inadequate nutrition in the short run, it cannot adapt as readily to chronic malnutrition that eventually results in loss of weight for height. He also found years of schooling of the household head not to be a significant determinant of agricultural productivity. More recently, in a study of reproducible human capital and wage rentals for Côte d'Ivoire and Ghana, Schultz (2003) resorted to an instrumental variable approach and analyzed demand for four human capital factors: education, migration, height, and body mass index. He emphasized the important role that public spending plays in explaining demand for human capital and included community characteristics: distance to local schools, existence of medical facilities, and community infrastructure in health and sanitation as proxy for public spending on social services. In the current study, I continue along these lines and use the first-stage estimations of human capital demand not only to address problems of endogeneity but also to analyze the role of public spending on social services in human capital formation.The measurement of productive efficiency has important implications for both economic theory and economic policy. Measuring productive efficiency allows for the testing of competing hypotheses regarding sources of efficiency or differentials in productivity. Furthermore, the measurement of productive efficiency makes it possible to quantify the potential increases in output that might be associated with an increase in efficiency (Farrell 1957). Two main methods are generally used to analyze the efficiency of production. The parametric approach, as proposed by Aigner, Lovell, and Schmidt (1977), consists of specifying and estimating a parametric production function frontier, and calculating technical inefficiency. A production frontier reflects the maximum obtainable output given a set of inputs; technical efficiency, in this case, describes the proximity of a farm household's output to this maximum feasible output (Coelli, Rahman, and Thirtle 2002). Although this approach provides a convenient framework for conducting hypothesis testing, the results can be sensitive to the parametric form chosen (Chavas, Petrie, and Roth 2005).This study makes use of the second method and applies nonparametrics, which has the advantage of removing the necessity of making arbitrary assumptions regarding the functional form of the frontier and the distributional form of the error. A second advantage of the nonparametric approach is that it is less data-demanding and thus works better with small samples than does the parametric approach. However, a major drawback is that because the nonparametric method is deterministic and attributes all the variation from the frontier to inefficiency, the frontier it estimates is likely to be sensitive to measurement errors or other noise in the data. In particular, efficiency tends to be overpredicted in finite samples. I employ a bootstrap method, set out below, to address this problem. Data envelopment analysis (DEA) is used to compute technical efficiency scores. DEA involves the use of linear programming methods to construct a nonparametric piecewise frontier over the data in order to calculate efficiencies relative to this surface, along the lines suggested by Farrell (1957). Given that many households are not perfectly competitive, the assumption of constant returns to scale (CRS) is often not appropriate. Banker, Charnes, and Cooper (1984) suggest an extension of the CRS DEA model to account for variable returns to scale (VRS) situations; theirs is the approach used here. To measure production efficiency, both inputand output-oriented efficiency measures have been used. Although the two approaches are equivalent under CRS, they differ under VRS. In the context of missing markets, it is likely that households are using fixed quantities of inputs (land, labor) to produce a maximum amount of output, which means an output-oriented efficiency measure, as also used by Chavas, Petrie, and Roth (2005) and Wouterse (2010) for rural households in, respectively, The Gambia and Burkina Faso is appropriate.To estimate technical efficiency, we can use the Farrell (1957) measure of output technical efficiency, which is the reciprocal of the Shephard (1970) output distance function. Define a measure for some point such that ,where is a vector of inputs and is a vector of outputs for farm household , and is a production set defined by . An overview of agricultural inputs is given in Table 1. Land is the major agricultural input; a missing land market means that the land input can be considered as exogenous. Labor is the second most important input. The survey does not contain data on labor input in days; I therefore use the number of adult males and females that are active on their own farm as a proxy for labor input. The value of equipment accounts for only large agricultural equipment, mainly plows and carts, which can be considered semi-fixed. However, an endogeneity problem may arise for variable inputs. In order to correct for this possible bias, I replaced the input variable with a predicted value obtained by running a regression using a set of instruments. Beyond the exogenous variables included in the production frontier, primary identifying instruments include the average age of household members and household asset holdings represented by the number of houses in the compound. Tests of identifying instrumental variables for regression estimates of the production function have not rejected the assumption of validity of the instrumental variables. The boundary of is sometimes referred to as the technology or production frontier and is given by the intersection of and the closure of its complement. Farm households that are technically efficient operate somewhere along the production frontier defined by the boundary of .The conditioning operates through the following mechanism:where farm household is faced with environmental variables , drawn from ; is a smooth continuous function; is a vector of parameters; and is a continuous, independent and identically distributed (iid), random variable, independent of . in (2) is distributed with right truncation at for each . The estimator of can be written in terms of the linear program .(3)In the second stage, estimated technical efficiency scores are regressed on environmental factors including human capital variables: , (4) where ,(5)where refers to the type of human capital: for years of schooling of males and females and for body mass index of males and females (BMI: weight in kilograms divided by height in meters squared) as an indicator of adult nutritional status and current health (Strauss and Thomas 1998;Schultz 2003). The formation of these two types of human capital of adult household members is modeled as a function of local public services, relevant conditions, and the endowment of the parents of the individual. Following Schultz (2003), instruments for the human capital inputs that are assumed to affect only the demand for human capital and identify the technical efficiency equation include these: (1) community health infrastructure, presence of a secondary school and permanent market in the village, and year-round accessibility of the village; (2) utilities: connection of community to the electricity grid;(3) education level of the father and mother of the individual and whether they are employed in agriculture; and (4) eight community-level food prices.1 An often-used approach in the literature is to analyze the determinants of efficiency using a Tobit regression, considered appropriate since the dependent variable, the calculated efficiency scores from the DEA analysis, is censored at 1. However, recent contributions (Simar and Wilson 2007) have emphasized two possible problems associated with applying a Tobit model in this context. First, the efficiency scores are not independent observations since the calculation of the efficiency score for one farm household necessarily involves all other farm households in the sample. As a consequence, the error term will be serially correlated and standard inference is not valid. The second problem is that the efficiency scores are likely to be biased in finite samples. I therefore apply the double bootstrap procedure as developed by Simar and Wilson (2007), which consists of the following steps: (1) standard DEA efficiency point estimates are calculated; (2) these estimates are integrated in a bootstrap procedure that is similar to the smoothed bootstrap procedure of Simar and Wilson (1998) ); (3) the bootstrap procedure produces biascorrected efficiency estimates; (4) the bias-corrected efficiency estimates are used in a parametric bootstrap on the truncated maximum likelihood; (5) standard errors are thus created for the parameters of the regression; (6) confidence intervals are then constructed for the regression parameters as well as for the efficiency scores. To be in line with these authors, I set the number of replications for the first bootstrap equal to 100 and the number of bootstrap replications for the second bootstrap equal to 2,000.Burkina Faso is a poor, landlocked country situated in the West African semiarid tropics. With a population of around 15.2 million people, Burkina Faso is one of the most densely populated countries of the West African Sahel (World Bank 2009). To achieve development, emphasis has long been put on the agricultural sector because economic growth and improvement in standard of living are thought to be difficult to achieve without the sector due to its importance in employment and export revenue (Asenso-Okyere, Benneh, and Tims 1997). During the period from 2000 to 2006, the country's economy experienced GDP growth of around six percent. The agricultural sector, which forms the main source of subsistence for the majority of the population, has been an important engine of this growth. Agriculture in Burkina Faso is generally extensive and practiced in family exploitations, dominated by small landholdings of between three and six hectares. Rainfed agriculture, which is subject to climatic variation, is prominent. Notwithstanding the important role of agriculture in GDP growth, strong performance of the sector is mainly due to land expansion. Generally, agricultural productivity has not increased; even in the cotton sector, productivity has remained relatively low. In Burkina Faso, GDP per capita grew around 2.5 percent between 1990 and 2006, and stood at around US$517 in 2009 (World Bank 2009).Despite the good levels of growth recorded by the Burkina Faso economy, the results of the three priority surveys conducted in 1994, 1998, and 2003 reveal an increase in poverty. Based on an absolute poverty threshold estimated at 82,672 FCFA in 2003, the proportion of the poor increased from 45.3 percent to 46.4 percent between 1998 and 2003. In fact, if Burkina Faso were to stay on the current agricultural growth path, it would not halve poverty by 2015 and thus would fail to meet MDG 1. Although urban poverty is on the increase, poverty remains a largely a rural phenomenon. Promotion of the basic social sectors (basic education and basic health, including reproductive health, clean drinking water, nutrition, hygiene, and sanitation) has always been the cornerstone of Burkina Faso's development strategy. In fact, around 16 to 19 percent of national resources and official development assistance are devoted to these services (Burkina Faso, Ministry of Economy and Development 2004). However, the country continues to suffer from a low level of human capital development that limits labor productivity, particularly in the agricultural sector, the source of employment and incomes of nearly 80 percent of the labor force.As part of activities aiming to promote rural development, since 2004, the Deuxième Programme National de Gestion des Terroirs (Second National Land Management Program, or PNGT 2) has undertaken socioeconomic studies with the overall objective of evaluating progress in the livelihoods of rural households following a number of interventions. These studies are based on field surveys that cover the national territory. The surveys constitute a three-year panel on living conditions of 2,000 households, with data collected in 2004, 2005, and 2006. Households included in the panel were drawn from 60 villages and all 45 provinces of Burkina. The villages were drawn according to probability proportional to their size. The random drawing of the sample and the coverage of all provinces ensures the representativeness of the data at the national level. Data include household characteristics, education, health, income-generating activities, asset holdings, and expenses, as well as community characteristics.A land market does not exist in rural Burkina Faso. Data from the surveys show that land is generally cultivated on a hereditary basis and that most households do not hold a title for their land. While theory predicts that better property rights on land can increase investment through increased security, enhanced trade opportunities, and increased collateral value of land, the presence and size of these effects depend crucially on whether those rights are properly enforced. In rural Africa land markets often barely function and are generally quite thin (Lanjouw, Quiznon, and Sparrow 2001). In Burkina Faso commercial land market transactions were found to be extremely rare (Ouedraogo et al. 1996). The lack of commercial land market transactions implies that land cannot function as collateral for credit. In terms of land quality, soils tend to be sandy and to a lesser extent clayey. Only about 30 percent of households surveyed had access to animal traction. Inorganic fertilizer was applied by only a third of households, whereas the application of compost was more common.On average the number of adult males and females active on the farm was about equal. Hired labor was not extensively used. A missing market for labor is characteristic of rural areas lacking a large landless class and with homogeneous factor endowments (de Janvry, Fafchamps, and Sadoulet 1991). In Burkina Faso, there appears to be a cultural barrier to offering one's own labor for a wage because it is thought to be a sign of inability to sustain production in one's own fields. Exchange labor in the form of work parties is slightly more common, but it is limited to a few crops with particular patterns of seasonality (Wouterse and Taylor 2008).Table 2 shows that education, in number of years of formal education received by the individual, was lower for females than for males in all age categories. For BMI, on average there was no significant difference between males and females. However, if we consider the distribution across age categories, we find that women had higher BMIs in the lower age categories and lower BMIs as they got older, compared to males. Table 3 shows that households were, in almost all cases, headed by a married male. Generally, the father of individuals working in agriculture was or had been active in agriculture; for the mother this was less so. Fathers of individuals working in agriculture had received about half a year of formal education, mothers almost none. In about 30 percent of the communities to which the individuals belonged, there was a clinic, in almost 60 percent a permanent market. A secondary school was present in only about 5 percent of communities. Connection to the electrical grid was particularly low, with only 1 percent of communities connected.  2004) Clinic (yes = 1) 0.29 0.45 0-1 Permanent market (yes = 1) 0.58 0.49 0-1 Secondary school (yes = 1) 0.05 0.01 0-1 Year-round accessibility (yes = 1) 0.30 0.46 0-1 Electrical grid (yes = 1) 0.01 0.10 0-1 N 3,020Source: Author's calculations.The results of the OLS regressions of determinants of human capital accumulation are given in Table 4. We are primarily interested in how government and household investments influence the formation of reproducible human capital. Age is negatively related to years of formal education received, with younger people being better educated. A concave relationship exists between age of the individual and BMI, with the latter increasing over a certain age but decreasing as individuals get older. For both males and females, BMI is positively related to the marital status of the household head. Although for developed countries a relationship attributed to social problems has been established between single-parent status and childhood obesity (Huffman, Kanikireddy, and Patel 2010), studies from developing countries tend to show that singleparent status is associated with significantly higher prevalence of food insecurity, hunger, and severe hunger (Isanaka et al. 2007). Similar to findings of, for example, Tansel (1997) for Côte d'Ivoire, schooling of both parents is a crucial determinant for the level of education of males as well as females.In terms of community services, the presence of a clinic in the community is positively related to years of education as well as to the body mass index of males and females. Community health infrastructure and access to medical care are expected to influence the prevalence of diseases and affect net nutritional status as proxied by BMI (Schultz 2003). Year-round accessibility of the community is positively associated with years of education. In cases where the secondary school is located outside the community, good year-round accessibility of the community is likely to play an important role in the possibility of attending school. Simultaneously, access to electricity facilitates and improves the delivery of social and business services from a wide range of village-level infrastructure such as schools, financial institutions, and farming machinery (Kirubi et al. 2009). Technical efficiency estimates are given in Table 5. Overall, the results suggest that substantial shortfalls in production efficiency exist. As mentioned, possible sources of inefficiency relate to managerial ability, endowments of human and physical capital, and access to financial capital. To identify sources of inefficiency, the technical efficiency estimates were regressed on the set of explanatory variables given in Table 1 as well as on predicted values of BMI and years of education for men and women. The estimation results of the truncated regression on the technical efficiency estimates are reported in Table 5. The results clearly illustrate the negative role of gender in technical efficiency, with female-headed households being less efficient. For the Ethiopian highlands, Holden, Shiferaw, and Pender (2001) found that female-headed households achieved much lower land productivity than did male-headed households. This finding was attributed to less availability of male labor and equipment in female-headed households. The findings here are similar of those by Bindlish, Evenson, and Gbetibouo (1993) for Burkina Faso, which show that female-headed households are less productive than those headed by men in most crops, with total values of output that are about 15 percent lower. Bindlish, Evenson, and Gbetibouo (1993) emphasize that fewer farm households are headed by women in Burkina Faso and that a woman comes under the authority of another male family member when her husband is away. In terms of the endowment of physical capital, the larger the household, the higher the technical efficiency. In the absence of a labor market, larger households are likely to be better able to meet their labor needs.An interesting contrast is uncovered in the relation between schooling and technical efficiency. There is a positive relation between years of education of females and efficiency, and a negative relation between technical efficiency and the education of males. As shown in Table 2, males tend to have spent more years in school and thus are likely to have spent less time in the field learning traditional farming methods from the household head; they may even have developed negative attitudes toward farm labor (Weir 1999). A study by Appleton and Balihuta (1996), using data for Uganda, found that educationparticularly secondary education-was associated with a reallocation of labor from the farm to nonfarm self-employment and wage employment. Using evidence from a large number of countries, the existence of higher rates of return on investments in the education of women has been established (Psacharopoulos 1985). One interpretation of this finding is that marginal returns on education among relatively low-educated females, typically affected by supply-side innovations, tend to be relatively high, reflecting women's high marginal costs of schooling rather than low ability that limits their return to education (Schultz 1999).BMI of females is associated with higher technical efficiency, whereas no relationship exists between BMI of males and efficiency. Using cross-section data on hoe-cultivating farm households in Sierra Leone, Strauss (1986) found that ‗‗effective family labor,'' which is a function of actual labor and per capita daily calorie intake, was a significant input in production. Croppenstedt and Muller (2000) showed for Ethiopia that nutrition status affected agricultural productivity. Bhargava (1997) used a panel of Rwandan households to analyze determinants of time allocation. Poor nutritional status was found to hamper the capacity of adults to undertake subsistence tasks. While poor health is associated with reduced labor supply, the fact that farm profits are not affected does not imply that the farmer's own productivity was unaffected but instead implies availability of labor that could substitute for the farmer during the time of illness (Strauss and Thomas 1998). All of these results strongly suggest that body mass is a proxy for strength or aerobic capacity, which are productive assets in the labor market of men, and possibly women, when working in more menial jobs. One explanation for this finding is that women are overrepresented in particular labor-intensive stages of the agricultural production process, such as weeding (Foster and Rosenzweig 1996). In Burkina Faso, where male migration is common, it has been demonstrated that in response to the absence of male labor, women disproportionately increase their weeding effort (Wouterse 2010).","tokenCount":"4948","images":["219755814_1_1.png"],"tables":["219755814_1_1.json","219755814_2_1.json","219755814_3_1.json","219755814_4_1.json","219755814_5_1.json","219755814_6_1.json","219755814_7_1.json","219755814_8_1.json","219755814_9_1.json","219755814_10_1.json","219755814_11_1.json","219755814_12_1.json","219755814_13_1.json","219755814_14_1.json","219755814_15_1.json","219755814_16_1.json","219755814_17_1.json","219755814_18_1.json","219755814_19_1.json","219755814_20_1.json","219755814_21_1.json","219755814_22_1.json","219755814_23_1.json","219755814_24_1.json"]}

data/part_2/0516626622.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"b0eeb4460865bec3c1f0b4731f1d69e9","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/b1d669b9-731a-44e7-bed6-be404708a2c9/retrieve","description":"This paper explores the evolution of real agricultural wages, machinery use, and the relationship between farm size and productivity in Vietnam during its dramatic structural transformation over the course of the 1990s and 2000s. Using six rounds of nationally representative household survey data, we find strong evidence that the inverse relationship between rice productivity and planting area attenuated significantly over this period and that the attenuation was most pronounced in areas with higher real wages. This pattern is also associated with sharp increases in machinery use, indicating a scale-biased substitution effect between machinery and labor. The results suggest that rural-factor market failures are receding in importance, making land concentration less of a cause of concern for aggregate food production.","id":"1830988482"},"keywords":["farm size-productivity relationship","structural transformation","Vietnam"],"sieverID":"b4e02e9b-f657-4e6f-a25a-841d1b7b9678","pagecount":"24","content":"The International Food Policy Research Institute (IFPRI), established in 1975, provides evidence-based policy solutions to sustainably end hunger and malnutrition and reduce poverty. The Institute conducts research, communicates results, optimizes partnerships, and builds capacity to ensure sustainable food production, promote healthy food systems, improve markets and trade, transform agriculture, build resilience, and strengthen institutions and governance. Gender is considered in all of the Institute's work. IFPRI collaborates with partners around the world, including development implementers, public institutions, the private sector, and farmers' organizations, to ensure that local, national, regional, and global food policies are based on evidence. IFPRI is a member of the CGIAR Consortium.The structural transformation of low-income agrarian economies is both cause and consequence of the steady amelioration of rural-factor market imperfections that have led to intersectoral and interhousehold heterogeneity in shadow prices of land, labor, and capital (Timmer 1988(Timmer , 2007)). Urbanization, the rise of modern industrial and service sectors, and a declining share of agriculture in gross domestic product and employment go hand in hand with economic growth and diminution of market failures that obstruct factor price equalization among production units. In low-income agrarian economies, household-specific market failures appear commonplace (de Janvry, Fafchamps, and Sadoulet 1989). When multiple markets fail, factor (shadow) prices vary across units, leading to heterogeneous input application rates and partial productivity measures, the most common of which is the oft-observed inverse relationship between farm size and crop yields (Feder 1985). The existence of any relationship between farm productivity and farm size-negative or positive-has attracted much scholarly and policy maker attention, as perhaps indicative of key market failures that motivate agricultural and rural development policy interventions. Examples of such interventions include progressive land reform and smallholder agricultural credit subsidies that might generate both efficiency and equity gains. If economic transformation were to bring improved factor market participation and performance, this might shift the farm size-productivity relationship and the economic rationale for concerted intervention to address rural market failures.In the past twenty-five years, Vietnam has undergone some of the most rapid transformation of any low-income agrarian nation. During that time, the country experienced rapid real gross domestic product growth of 4-8 percent annually, mainly driven by the development of nonfarm sectors and the activation of markets for credit, labor, land, machinery, and other factors of production throughout the country (McCaig and Pavcnik 2013). As the nonfarm sectors provided more job opportunities, people moved out of farming, resulting in increased agricultural real wages and shrinkage in the wage gap between agricultural and nonagricultural sectors, as predicted by dual economy models (Lewis 1954;Gollin 2014).Such changes likely have important implications for agriculture. It has long been observed that, in developing-country agriculture, smaller farms are typically more productive per unit area cultivated than larger ones (Chayanov 1926(Chayanov /1986;;Sen 1962;Berry and Cline 1979;Carter 1984;Barrett 1996;Benjamin and Brandt 2002;Barrett, Bellemare, and Hou 2010;Carletto, Savastano, and Zezza 2013). The evidence of such an inverse farm size-productivity relationship has often justified land policies supporting small landholders and deterring farm size expansion, as well as agricultural credit policies to promote smallholder access to commercial inputs.However, as a low-income agrarian economy undergoes rapid structural transformation, do factor markets for agricultural labor and machinery become more active, driving up real wages and attenuating the inverse relationship? Conceptually, Otsuka (2013) suggested that increasing real wages would reduce demand for agricultural labor, promote the use of machinery as a substitute for labor, and decrease the disadvantage of larger farms-perhaps even flipping the inverse relationship to a direct relationship-due to scale economies in machine use. Similarly, using nationally representative household data from India, Foster and Rosenzweig (2011) estimated an increase in optimal farm size due to the substitution of machinery for labor.This paper explores the evolution of real agricultural wages, machinery use, and the relationship between farm size and productivity in Vietnam during its dramatic structural transformation over the course of the 1990s and 2000s. This inverse relationship has been attributed to imperfections in multiple factor markets, especially for land, labor, and credit (Sen 1966;Feder 1985). Even without significant changes in land policy, if the inverse relation had been partially driven by imperfect credit or labor markets, improved factor market functioning through structural transformation, as manifested in increased real wage rates and more active machinery rental markets, may have lessened or may even have reversed the long-standing inverse farm size-productivity relationship. Because the inverse relationship has long served as a powerful metaphor for pervasive rural-factor market failures that motivate government interventions, any such evolution has powerful implications for policy.This paper uses six rounds of nationally representative household survey data from Vietnam, from 1992 through 2008, to examine whether and the extent to which increasing real wages and increased machinery rentals are associated with change in the inverse relationship between rice productivity and cultivated area. 1 To our knowledge, this study is the first to look at intertemporal change in the inverse farm size-productivity relationship over such a long period. There is strong evidence that the inverse relationship between rice productivity and planting area attenuated significantly over this period. Those effects are most pronounced in areas with higher real wages, which are also associated with sharp increases in machinery use, indicating a scale-biased substitution effect between machinery and labor. The empirical results of this study suggest that, in Vietnam, larger farmers' disadvantage in land productivity, compared to smaller farmers, has lessened sharply, which is probably attributable to improved labor and machine rental markets.The empirical analysis that is the primary contribution of this study is framed with a simple theoretical model. Assume the land market does not function, so farmers have fixed landholdings. In this case, labor and machinery are the only two variable inputs. The production function of farm \uD835\uDC56\uD835\uDC56 iswhere \uD835\uDC44\uD835\uDC44 \uD835\uDC56\uD835\uDC56 is output, ℎ \uD835\uDC56\uD835\uDC56 is landholding, \uD835\uDC3F\uD835\uDC3F \uD835\uDC56\uD835\uDC56 is labor input, and \uD835\uDC40\uD835\uDC40 \uD835\uDC56\uD835\uDC56 is machine use. Assuming constant returns to scale, equation ( 1) is rewritten aswhere \uD835\uDC66\uD835\uDC66 \uD835\uDC56\uD835\uDC56 is household-specific yield (output per hectare) and \uD835\uDC59\uD835\uDC59 \uD835\uDC56\uD835\uDC56 = \uD835\uDC3F\uD835\uDC3F \uD835\uDC56\uD835\uDC56 /ℎ \uD835\uDC56\uD835\uDC56 and \uD835\uDC5A\uD835\uDC5A \uD835\uDC56\uD835\uDC56 = \uD835\uDC40\uD835\uDC40 \uD835\uDC56\uD835\uDC56 /ℎ \uD835\uDC56\uD835\uDC56 represent input application rates per hectare. Assuming that farmer \uD835\uDC56\uD835\uDC56 maximizes profitability and the price of output serves as numeraire, the factor demand functions can be derived based on the first-order conditions.where \uD835\uDF03\uD835\uDF03 \uD835\uDC56\uD835\uDC56 is shadow price of labor and \uD835\uDC58\uD835\uDC58 \uD835\uDC56\uD835\uDC56 is price of machine use. If \uD835\uDF03\uD835\uDF03 \uD835\uDC56\uD835\uDC56 = \uD835\uDF03\uD835\uDF03 and \uD835\uDC58\uD835\uDC58 \uD835\uDC56\uD835\uDC56 = \uD835\uDC58\uD835\uDC58 ∀\uD835\uDC56\uD835\uDC56, then factor prices are exogenous to individual households, and markets will allocate factors so as to equalize productivity across the fixed land endowments. But if a second market (besides land) functions imperfectly, then the (shadow) price of one or both factors will vary across production units, leading to a relationship between farm size and productivity (Feder 1985). Following Feder (1985), Benjamin and Brandt (2002), and Barrett, Sherlund, and Adesina (2008), let \uD835\uDF03\uD835\uDF03 \uD835\uDC56\uD835\uDC56 increase in ℎ \uD835\uDC56\uD835\uDC56 to capture search and or supervision costs; following Foster and Rosenzweig (2011) and Feder (1985), let \uD835\uDC58\uD835\uDC58 \uD835\uDC56\uD835\uDC56 decrease in ℎ \uD835\uDC56\uD835\uDC56 to reflect high-capacity (that is, lower unit price) machinery's greater suitability to larger plots, borrowing costs that decrease in collateralizable landholdings, or nontrivial fixed search or contracting costs of machinery rental. Plugging (3) and (4) into (2) gives \uD835\uDC66\uD835\uDC66 \uD835\uDC56\uD835\uDC56 = \uD835\uDC53\uD835\uDC53(\uD835\uDC59\uD835\uDC59 \uD835\uDC56\uD835\uDC56 , \uD835\uDC5A\uD835\uDC5A \uD835\uDC56\uD835\uDC56 ) = \uD835\uDC53\uD835\uDC53�\uD835\uDC59\uD835\uDC59(\uD835\uDF03\uD835\uDF03 \uD835\uDC56\uD835\uDC56 , \uD835\uDC58\uD835\uDC58 \uD835\uDC56\uD835\uDC56 ), \uD835\uDC5A\uD835\uDC5A(\uD835\uDF03\uD835\uDF03 \uD835\uDC56\uD835\uDC56 , \uD835\uDC58\uD835\uDC58 \uD835\uDC56\uD835\uDC56 )� .(5)Assuming machine and labor are pure substitutes, so that toward zero, then the relationship between crop productivity and farm size will attenuate. The empirical regularity in low-income agriculture is that labor market imperfections dominate, leading to an inverse farm size-productivity relationship. Therefore, as structural transformation proceeds, driving up real wages, there should be attenuation of the commonplace downward-sloping relationship between crop yields and farm size. The next section explores that prediction using data from Vietnam.Based on the model of the preceding section, subindex t represents period and c represents commune. The shadow price of labor is specified as a function of the prevailing local (real) wage rate, w \uD835\uDC50\uD835\uDC50\uD835\uDC50\uD835\uDC50 , adjusted for the household-specific landholding that influences labor search and supervision costs:where \uD835\uDEFC\uD835\uDEFC \uD835\uDC56\uD835\uDC56 is a household fixed effect that also captures time-invariant, location-specific effects; \uD835\uDC67\uD835\uDC67 \uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50\uD835\uDC50\uD835\uDC50 is a vector of household-specific, time-varying characteristics; \uD835\uDC37\uD835\uDC37 \uD835\uDC50\uD835\uDC50 is a time dummy; and \uD835\uDF00\uD835\uDF00 \uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50\uD835\uDC50\uD835\uDC50 is an independent and identically distributed (iid), mean zero, normally distributed random error term. The main coefficient of interest, \uD835\uDEFC\uD835\uDEFC 2 , is expected to be positive. The price of machine use is specified aswhere \uD835\uDEFE\uD835\uDEFE \uD835\uDC56\uD835\uDC56 is a household fixed effect that also captures the time-invariant, location-specific effects; \uD835\uDEFE\uD835\uDEFE 1\uD835\uDC50\uD835\uDC50 captures period-specific fixed effects (including interest rates, which are assumed uniform across communes); \uD835\uDEFE\uD835\uDEFE 2 captures the shadow price effect of landholdings, which is expected to be negative; and \uD835\uDC62\uD835\uDC62 \uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50\uD835\uDC50\uD835\uDC50 is an iid, mean zero, normally distributed random error term. Equations ( 3), ( 4), ( 9), and (10) combine to create the log linear factor demand equations:where \uD835\uDEFD\uD835\uDEFD \uD835\uDC56\uD835\uDC56 and \uD835\uDEFF\uD835\uDEFF \uD835\uDC56\uD835\uDC56 capture household fixed effects and \uD835\uDC37\uD835\uDC37 \uD835\uDC50\uD835\uDC50 is a period fixed effect. In the labor demand equation ( 11), \uD835\uDEFD\uD835\uDEFD 1 and \uD835\uDEFD\uD835\uDEFD 3 should be negative and \uD835\uDEFD\uD835\uDEFD 2 positive. In the machine demand equation ( 12), \uD835\uDEFF\uD835\uDEFF 1 and \uD835\uDEFF\uD835\uDEFF 3 are expected to be positive and \uD835\uDEFF\uD835\uDEFF 2 negative. After plugging the factor demand functions into the crop yield equation:where \uD835\uDF09\uD835\uDF09 \uD835\uDC56\uD835\uDC56 is the household fixed effect, \uD835\uDF09\uD835\uDF09 1 is expected to be negative, \uD835\uDF09\uD835\uDF09 2 is expected to be positive, and \uD835\uDF09\uD835\uDF09 3 can be either positive or negative depending on whether the capital or labor market imperfection dominates. Data permitting, one could estimate the system of equations ( 11)-( 13) to test the model predictions about coefficients of interest. By testing the hypothesis that the parameters \uD835\uDF09\uD835\uDF09 2 and \uD835\uDF09\uD835\uDF09 3 change over the course of structural transformation in the economy, one can explore whether any initial relationship between farm size and productivity is indeed attenuated by increased factor market activity, as theory would predict. The commune survey provides information on local agricultural wage by gender and by task (land preparation, planting, tendering, and harvesting). We generate median wage by gender, combining across all tasks for each commune for each panel round. To calculate the real wage, we deflate local nominal wage rates by the national consumer price index (CPI), which captures intertemporal inflation, and by the regional CPI, which captures spatial price variation.2 Figure 4.1 plots the median female and male real agricultural wages from 1992 to 2008. The real wage increased significantly from 1992 to 1998. It leveled off from 1998 to 2004, probably reflecting the lagged effects of the Asian financial crisis of 1997/1998. From 2004 to 2008, the real wage again picked up rapidly, at a rate even faster than seen during 1992-1998. Table 4.1 reports the median male real wage by regions from 1992 to 2008, using VLSS and VHLSS commune survey data. 3 Although the real wage was consistently lower in the northern regions than in the southern regions, regional wage differences narrowed considerably by 2008, indicating an increasingly spatially integrated national labor market. This pattern is consistent with the stylized facts of structural transformation (Timmer 2007).Source: Authors' computations based on commune surveys in VLSS 1992and 1998and VHLSS 2002, 2004, 2006, and 2008.  VLSS 1992and 1998and VHLSS 2002, 2004, 2006, and 2008. The household survey data provide information on household demographics, rice output by type of rice, rice planting area, landholdings, machinery use, labor hiring, and production cost by category. Figure 4.2 plots the proportion of households that rented machinery, owned tractors, or hired labor from 1992 to 2008. There is not much change in tractor ownership; the rate remains almost zero, which is not surprising given the rather low median landholdings in Vietnam (Figure 4.3). The percentage of cultivating households that rented machines more than tripled, however, from 19 percent in 1992 to 63 percent in 2008. This observation mirrors recent findings from China (Yang et al. 2013). The percentage of households that hired labor also increased sharply, from 32 percent in 1992 to 55 percent in 2008; by 2008, most cultivating households hired both labor and machinery services.  Source: Authors' computations based on household surveys in VLSS 1992and 1998and VHLSS 2002, 2004, 2006, and 2008.Source: Authors' computations based on household surveys in VLSS 1992and 1998and VHLSS 2002, 2004, 2006, and 2008.   1992, 1998, 2002, 2004, 2006, and 2008. 4 The figure shows point estimates with solid lines and the 95 percent confidence intervals with dotted lines of the same color as the corresponding point estimates. Although the curve is downward sloping throughout, pointing to the presence of an inverse farm size-productivity relationship, it flattens considerably in 2006 and 2008, suggesting that productivity advantage for small farmers had decreased. The sharp upward displacement of the yield curves between each successive survey round, especially the first three, demonstrates the rapid growth in rice productivity. Source: Authors' computations based on household surveys in VLSS 1992 and 1998 and VHLSS 2002, 2004, 2006, and 2008.   4 The procedure used is the \"lpoly\" in Stata 13 SE with default optimal bandwidth. Ideally, we would estimate equations ( 11)-(13) as a system for improved efficiency. Due to the lack of total labor input data, however, we cannot estimate the labor demand function, equation ( 11). For machine use, we only know the total amount of money spent on machine rental and fuel but not the days of machine use, as required for estimation of equation ( 12). 5 We therefore estimate only the yield equation ( 13). For comparison, we use the 1992/1998 panel, the 2002/2004 panel, and the 2006/2008 panel separately. We also demean the variables ln w \uD835\uDC5A\uD835\uDC5A\uD835\uDC50\uD835\uDC50 and ln ℎ \uD835\uDC5A\uD835\uDC5A\uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50 around the sample means, so the coefficients \uD835\uDF09\uD835\uDF09 1 and \uD835\uDF09\uD835\uDF09 3 can be interpreted as the mean partial effects (or partial effects evaluated at the same mean). Columns ( 1)-(3) of Table 5.1 report regression results on rice yield aggregated over all rice varieties, for the three panels.6   VLSS 1992and 1998and VHLSS 2002, 2004, 2006, and 2008. Notes: The variables \"log total area of rice\" and \"log male real agricultural wage\" are centered around their sample means.Standard errors in parentheses: * p < 0.10; ** p < 0.05; *** p < 0.01.There are two main findings. First, the coefficient estimate on planting area is statistically significantly negative in all panels, suggesting the existence of an inverse farm size-productivity relationship, which is consistent with most of the published literature. However, with a value of -0.061, the estimated coefficient in the 2006/2008 panel is lower than that in the 1992/1998 panel (-0.150) and in the 2002/04 panel (-0.119). The result is consistent with the model prediction that the inverse farm sizeproductivity relationship attenuates as the labor and machine rental markets become more active and spatially integrated.Second, the coefficient estimate on the interaction term of planting area and real wage is statistically significantly positive in the 2006/2008 panel, though it is insignificant in the 1992/1998 and 2002/2004 panels. This suggests that the inverse relationship attenuates most quickly in areas where the real wage rate is higher, creating greater incentives to substitute machinery for labor. The statistically insignificant estimate from 1992/1998 and 2002/2004 is consistent with the existence of imperfect labor markets in the earlier years in the survey, before wages began to reach a level at which substituting capital for labor began to appear potentially profitable to farmers.To test for the significance of changes in farm size-productivity relationship (that is, to test the differences in key coefficient estimates between the panels), we pool the three panels, interact all the control variables with panel dummies, and run the same regression with the pooled dataset. The results are reported in columns ( 4)-( 6) of Table 5.1. Although the coefficient estimate of the planting area is not significantly different between the 1992/1998 panel and the 2002/2004 panel, indicating that attenuation of the inverse relationship was modest at best, the effect was statistically significantly lower for the 2006/2008 panel than for either of the earlier two panels. Similarly, the coefficient estimate on the interaction term of planting area and real wage for the 2006/2008 panel was statistically significantly higher than for either the 1992/1998 or 2002/2004 panels. Together, these results reinforce the story that by the latter rounds of the survey, increasingly active rural-factor markets significantly reduced the inverse farm size-productivity relationship, and these effects were most pronounced in areas with higher real wages. The point estimates for the 2006/2008 panel (column (3)) suggest that an 80 percent increase in the real wage for male workers offsets the effect of doubling farm size.In Table 5.1, the dependent variable is land productivity of rice aggregated over all rice varieties. These results may be biased if the choice of rice varieties is correlated with farm size and if the productivity differs across rice varieties. We thus run the same regressions for spring ordinary rice and autumn ordinary rice separately as a robustness check. The 2002 VHLSS does not distinguish between these rice varieties; therefore, the 2002/2004 panel is left out of these analyses. The results, reported in Tables 5.2 and 5.3, are similar to those reported in Table 5.1, showing a significantly decreasing inverse relationship for both spring and autumn rice over 1992/1998 and 2006/2008. The interaction term is significantly positive for both spring and autumn rice from the 2006/2008 panel, in line with the result from Table 5.1. For the 1992/1998 panel, the interaction is insignificant for spring rice but significantly positive for autumn rice.  VLSS 1992and 1998and VHLSS 2002, 2004, 2006, and 2008. Notes: The variables \"log total area of spring ordinary rice\" and \"log male real agricultural wage\" are centered around their sample means. Standard errors in parentheses: * p < 0.10; ** p < 0.05; *** p < 0.01.  VLSS 1992and 1998and VHLSS 2002, 2004, 2006, and 2008. Notes: The variables \"log total area of autumn ordinary rice\" and \"log male real agricultural wage\" are centered around their sample means. Standard errors in parentheses: * p < 0.10; ** p < 0.05; *** p < 0.01.This study uses three household panel data sets from the 1990s and the 2000s from the same survey to explore the changing relationship between land productivity and rice planting areas in Vietnam. The findings show that the inverse relationship attenuates considerably over the course of those two decades. This change is associated with rising real wages and increasingly active machine rental and agricultural labor markets in rural Vietnam. In addition, the inverse relation attenuated most in areas with higher agricultural real wages by the mid-2000s, as real wages reached levels sufficiently high enough to induce some substitution of machinery for labor. As a result, the long-standing productivity advantage assumed to exist among smaller farmers appears to have diminished or disappeared altogether by the latter part of the period. Indeed, as real wages keep increasing, the inverse relationship may be reversed, leading to increased land concentration among farmers increasingly likely to employ machinery, without adverse effects on aggregate food production or prices.For the machine use equation, we generate a dummy variable as a dependent variable from the cost of machine rental and machine ownership. The variable takes value 1 if the household owned any tractors or spent on machine rental; otherwise, it takes 0. The estimated equation is:\uD835\uDC51\uD835\uDC51 \uD835\uDC5A\uD835\uDC5A\uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50 = \uD835\uDEFF\uD835\uDEFF 0,\uD835\uDC5A\uD835\uDC5A\uD835\uDC56\uD835\uDC56 + \uD835\uDEFF\uD835\uDEFF 1 \uD835\uDC59\uD835\uDC59\uD835\uDC59\uD835\uDC59 \uD835\uDC64\uD835\uDC64 \uD835\uDC5A\uD835\uDC5A\uD835\uDC50\uD835\uDC50 + \uD835\uDEFF\uD835\uDEFF 2 \uD835\uDC59\uD835\uDC59\uD835\uDC59\uD835\uDC59 \uD835\uDC64\uD835\uDC64 \uD835\uDC5A\uD835\uDC5A\uD835\uDC50\uD835\uDC50 × \uD835\uDC59\uD835\uDC59\uD835\uDC59\uD835\uDC59 ℎ \uD835\uDC5A\uD835\uDC5A\uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50 + \uD835\uDEFF\uD835\uDEFF 3 \uD835\uDC59\uD835\uDC59\uD835\uDC59\uD835\uDC59 ℎ \uD835\uDC5A\uD835\uDC5A\uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50 + \uD835\uDEFF\uD835\uDEFF 4 \uD835\uDC67\uD835\uDC67 \uD835\uDC5A\uD835\uDC5A\uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50 + \uD835\uDEFF\uD835\uDEFF 5 \uD835\uDC37\uD835\uDC37 \uD835\uDC50\uD835\uDC50 + \uD835\uDF16\uD835\uDF16 \uD835\uDC5A\uD835\uDC5A\uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50 , (A1)where \uD835\uDC51\uD835\uDC51 \uD835\uDC5A\uD835\uDC5A\uD835\uDC56\uD835\uDC56\uD835\uDC50\uD835\uDC50 is the dummy variable indicating machine use for household i in commune m and year t. We estimate equation (A1) using the 1992/98, the 2002/04, and the 2006/08 panels separately.The results are presented in Table A.1. For all three panels, larger farmers are more likely to use machines, consistent with theoretical prediction. Interestingly, machine use is not responsive to real agriculture wage in the 1990s panel or in the early 2000s panel. In contrast, it is significantly higher when real wage is higher in the 2006/08 panel, suggesting the efficiency improvement of rural-factor markets. ","tokenCount":"3358","images":["1830988482_1_1.png"],"tables":["1830988482_1_1.json","1830988482_2_1.json","1830988482_3_1.json","1830988482_4_1.json","1830988482_5_1.json","1830988482_6_1.json","1830988482_7_1.json","1830988482_8_1.json","1830988482_9_1.json","1830988482_10_1.json","1830988482_11_1.json","1830988482_12_1.json","1830988482_13_1.json","1830988482_14_1.json","1830988482_15_1.json","1830988482_16_1.json","1830988482_17_1.json","1830988482_18_1.json","1830988482_19_1.json","1830988482_20_1.json","1830988482_21_1.json","1830988482_22_1.json","1830988482_23_1.json","1830988482_24_1.json"]}

data/part_2/0527197965.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"d2ca8505643495b2449e69cb7f9798cf","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/4bd91700-186b-430b-8c69-122ec7face36/retrieve","description":"","id":"564139902"},"keywords":[],"sieverID":"b3457205-ced3-4fda-b9e2-9782e42b3f33","pagecount":"16","content":"• Innovation platforms are essential to enhance communications .More evidence needed on which scaling approaches and whole systems research are the most effective to achieve impact.Stepwise approach to integration of sustainable intensification interventions is a reality in most cases.","tokenCount":"39","images":["564139902_1_1.png","564139902_1_2.png","564139902_2_1.png","564139902_2_2.png","564139902_3_1.png","564139902_4_1.png","564139902_4_2.png","564139902_5_1.png","564139902_5_2.png","564139902_6_1.png","564139902_6_2.png","564139902_7_1.png","564139902_7_2.png","564139902_8_1.png","564139902_8_2.png","564139902_9_1.png","564139902_9_2.png","564139902_9_3.png","564139902_10_1.png","564139902_10_2.png","564139902_11_1.png","564139902_11_2.png","564139902_12_1.png","564139902_12_2.png","564139902_13_1.png","564139902_13_2.png","564139902_14_1.png","564139902_14_2.png","564139902_14_3.png","564139902_15_1.png","564139902_16_1.png","564139902_16_2.png","564139902_16_3.png","564139902_16_4.png","564139902_16_5.png"],"tables":["564139902_1_1.json","564139902_2_1.json","564139902_3_1.json","564139902_4_1.json","564139902_5_1.json","564139902_6_1.json","564139902_7_1.json","564139902_8_1.json","564139902_9_1.json","564139902_10_1.json","564139902_11_1.json","564139902_12_1.json","564139902_13_1.json","564139902_14_1.json","564139902_15_1.json","564139902_16_1.json"]}

data/part_2/0539084041.json

@@ -0,0 +1 @@
+{"metadata":{"gardian_id":"8789b400c2f85a060b4d856cdf631c00","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/d88d5c82-f155-4a43-9630-ff64fe6fd316/retrieve","description":"Land provides the basis for food production and is an indispensable input for economic livelihoods in rural areas. Landownership is strongly associated with social and economic power, not only across communities and households, but also within households. The link between landownership and women’s empowerment has been relatively well documented in general, but not specifically in relation to agriculture. This paper aims to fill this gap by analyzing how ownership of land is associated with agency and achievements in agriculture among female and male farmers in northern Ghana, a region transitioning from customary land tenure without individual ownership rights towards a more individualized and market-based tenure system. We use a recursive bivariate probit model and focus on eight different indicators in four distinct domains: decisions on agricultural cultivation, decisions on farm income, agricultural association membership, and time allocation. Our empirical estimates indicate that landownership is positively correlated with men’s and women’s agency in agriculture, namely in decisions on agricultural cultivation and membership in agricultural association. Yet, we also find that the gender gaps in participation in cultivation decisions, the use of agricultural earnings, and in agricultural workload continue to persist among those who own land. While the results underscore the importance of land as a resource that can enhance women’s agency, they also point out that policies aiming to solely advance land rights may not be sufficient to eradicate or even reduce gender inequality in agriculture.","id":"1751057039"},"keywords":["land tenure","gender","women empowerment in agriculture","agricultural decision making","Ghana"],"sieverID":"e9fbf2fa-9414-48a4-b476-e10c907764ac","pagecount":"51","content":"established in 1975, provides research-based policy solutions to sustainably reduce poverty and end hunger and malnutrition. IFPRI's strategic research aims to foster a climate-resilient and sustainable food supply; promote healthy diets and nutrition for all; build inclusive and efficient markets, trade systems, and food industries; transform agricultural and rural economies; and strengthen institutions and governance. Gender is integrated in all the Institute's work. Partnerships, communications, capacity strengthening, and data and knowledge management are essential components to translate IFPRI's research from action to impact. The Institute's regional and country programs play a critical role in responding to demand for food policy research and in delivering holistic support for country-led development. IFPRI collaborates with partners around the world.Land provides the basis for food production and most income-generating activities and is an indispensable input for economic livelihoods in developing countries (Meinzen-Dick et al., 2019). Equally, if not more important, however, are the strong linkages between landownership and social and economic power (Agarwal, 1997;Kabeer, 1999;Goldstein and Udry, 2008). The direct and indirect pathways of landownership to power and empowerment have been documented at macro-level, across and within societies and communities (Binswanger et al., 1995;Goldstein and Udry, 2008). More recently, the connection between landownership and empowerment also received increasing attention at a micro-level, specifically focusing on gender relations within communities and households (Kabeer, 1999).Landownership can help promote women's empowerment and enhance their well-being (Panda and Agarwal, 2005;Mishra and Sam, 2016). More specifically, research indicates that land access and secured land rights can increase women's economic security and bargaining power at the household level (Agarwal, 1997;Kabeer, 1999;Quisumbing and Briere, 2000;Anderson and Eswaran, 2009;Wigg, 2013). The improvement in women's bargaining positions, in turn, strengthens women's influence in household decisions, enabling women to reallocate household resources towards their preferences and contributing to an improvement in own and household well-being (Panda and Agarwal, 2005;Allendorf, 2007;Mishra and Sam, 2016). Allendorf (2007) and Mishra and Sam (2016), for example, show that landownership has a positive association with women's ability to make decisions concerning their own health care, large household purchases, purchases for daily needs, and visits to family, friends, and relatives in Nepal. With regards to women's empowerment in agriculture, Wiig (2013) finds that joint property rights considerably increase women's participation in household-decision-making on agricultural production and land-related investment in Peru. Santos et al. (2014) find that women whose names are included on land titles in India are more likely to participate in decisions concerning how to use the land, what to grow on it, and whether to sell the produce from that plot. Doss et al. (2014) shows that women who own land in Malawi, Mali, and Tanzania participate into a greater number of agricultural decisions.Given that land is a key asset for economic livelihoods and that agriculture remains the main income source for many rural households in developing countries, given that improving 'women's empowerment in agriculture' has become a target for many projects, and given that land rights are a key entry point for gender advocacy, it is surprising that so few studies look at the association between landownership and women's agricultural decision making or participation in agriculture. The assumption that each landowner makes all decisions regarding his or her plot alone is likely incorrect in many parts of the world (Twyman et al. 2015), and ownership of resources, such as land, does not necessarily translate into agency and achievements in any given domain (Kabeer, 1999). This paper examines the relationships between landownership and consequent agency and achievements in agriculture for farmers in northern Ghana, and how this differs for men and women. We focus on four different domains that are directly related to agriculture: decisions about agricultural production, decision about earnings from agriculture, membership in agricultural associations, and workload in farm and non-farm activities. Ghana provides a relevant case study setting as agriculture is still the main source of income for rural households (World Bank, 2017).Especially in the north of the country, land rights are predominately governed by customary law where kinship and lineage are key determinants of land access (Higgins and Fenrich, 2011).Traditionally, customary law does not grant exclusive control and ownership to an individual (Higgins and Fenrich, 2011;Doss et al., 2019), yet there is an ongoing transition towards increasingly more and stronger individual-held land rights and land markets are developing (Lambrecht and Asare, 2016). As in many other countries, reforming land law is high on Ghana's policy agenda (Bruce, 2014;Narh et al., 2016). Furthermore, Ghana performs poorly in terms of gender equality, and gender discrepancies are more prominent in the north compared to the southern part of the country (Apusigah, 2009;Osei-Assibey, 2014).This study contains three features that distinguish this research from other existing studies.First, this paper extends the empirical evidence on the relationships between landownership and women's empowerment (Allendorf, 2007;Deere and Twyman, 2012;Wiig, 2013;Deere et al., 2013;Santos et al., 2014;Doss et al., 2014;Twyman et al., 2015;Mishra and Sam, 2016). We specifically focus on empowerment in agriculture and use a variety of decision-making and participation measures in agriculture that have received scant attention in the landownership literature. Other studies (Wiig, 2013;Santos et al., 2014;Doss et al., 2014;Twyman et al., 2015) include empowerment measures that capture household decision-making in agriculture but focus predominately on decisions regarding agricultural production, such as types of crops to grow for consumption and sale, and agricultural input purchases. While these are fundamental agricultural decisions and are included in our analysis, other decisions, namely agency over the use of income generated from food and cash crop farming, membership in agricultural groups, and workloads in agricultural activities, are equally critical for agrarian households. These are related yet distinct dimensions in agricultural decision-making, and their association with landownership may differ.Second, our study takes place in a very different setting compared to previous studies on land and women's empowerment in agriculture. Wiig (2013) and Santos et al. (2014), for example, focus specifically on land titling efforts. This analysis, on the other hand, focuses on a setting in which land tenure systems are gradually changing from customary tenure where no one owns the land to a statutory market-based system that allows women and men to acquire ownership rights. Third, we focus not only on women, but we also include men in our analysis. A large number of studies on landownership and women's empowerment, such as Panda and Agarwal, (2005), Doss (2006), Allendorf (2007), Doss et al. (2014), Santos et al. (2014), and Mishra and Sam (2016), focus solely on women. As a result, their studies are unable to recognize the potential gender differences in the connections between landownership and decision-making roles. Critical insights are likely to be gained by understanding how these linkages between landownership and empowerment differ between men and women.The remainder of the paper is organized as follows. The next section presents a brief conceptual framework. Section 3 provides a detailed discussion on how traditional gender norms govern landownership in Ghana. Section 4 discusses the data, and section 5 explains the empirical models. Section 6 reports the results, and section 7 concludes.The conceptual framework for this analysis is based on Kabeer's (1999) definition of empowerment but relies on indicators developed as part of the women's empowerment in agriculture index (WEAI), developed by Alkire and co-authors (2013). In a paper that has shaped much of the current thinking on measuring empowerment, Kabeer (1999) defined empowerment as \"the process by which those who have been denied the ability to make strategic life choices acquire such an ability.\" This process entails three dimensions: resources, agency, and achievements, which are essentially interrelated and to some extent indivisible. Resources are defined as access and rights to material, human, and social resources, which serve to enhance the ability to exercise choice. Agency is the ability to take action in decision-making that would ultimately result in desired achievements or outcomes.Theory and common sense suggest that, in societies where arable land is the most critical form of property for livelihoods, any sustained improvements in women's well-being are tied to improvements in women's land rights (Agarwal, 1997). The literature on women's empowerment therefore interprets land mostly as a resource influencing women's empowerment (e.g. Mason, 1996;Agarwal, 1997;Allendorf, 2007;Wiig, 2013;Mishra and Sam, 2016), whereas many others use land as an indicator for women's empowerment as such (e.g. Roy and Tisdell, 2002).Yet, individual or joint agricultural landownership does not necessarily translate into agency and achievements in agriculture (Kabeer, 1999). Agricultural households in developing countries are often both production and consumption units, where farm and household decisions are essentially interrelated and intertwined (Bardhan and Udry, 1999). The assumption that each household member makes decisions regarding his or her plot alone, or no agricultural decisions in the absence of having individual or joint ownership of a plot, is likely false in many households (Twyman et al. 2015). Rather, household members provide labor on others' plots and may consult one another regarding strategic choices (Doss & Meinzen-Dick, 2015). Whether women's landownership, as opposed to men's landownership, could effectively increase their agency and achievements in agriculture will depend to a large extent on the pervasiveness of customary norms, beliefs and values that characterize the gendered roles of men and women in agriculture (Agarwal, 1997;Kabeer 2016).In line with Kabeer (1999)'s understanding of empowerment, we ask the following questions: (i) is ownership of land associated with the ability of individuals make strategic choices in agriculture?; and, (ii) does this association differ for men and women? Even though the three dimensions of empowerment are essentially intertwined, we consider self-reported ownership of land to reflect mostly upon the resources (pre-conditions) domain. This encompasses both the more conventional economic meaning as well as the institutional aspect enabling ownership rights. Decision-making in agricultural production and earnings, and membership in agricultural groups are considered as signs of agency, whereas agricultural workload and total (farm and non-farm) workload are considered as (negative) achievements.Ghana experienced rapid economic growth over the past 20 years, and by the end of 2010, the country obtained the status of a lower middle-income country (McKay et al., 2016). Yet, the overall reduction in poverty has come at a cost of rising inequality, with the northern part of the country lagging behind (McKay et al., 2016). Agriculture remains the backbone of the Ghanaian economy (African Development Bank, 2019) and subsistence agriculture continues to be the main economic activity for rural households (Lanz et al., 2018;World bank, 2017). Nevertheless, poverty is also higher among those working in agriculture (McKay et al., 2016).The government of Ghana recognizes both statutory and customary laws of access to agricultural land. This has led to an essentially pluralistic setting of overlapping, and often also contradicting laws, with customary as well as state-established laws defining inheritance and property rights (Lanz et al., 2018). Ghana's statutory laws are, in theory, gender-equal. However, they do not align with prevailing customary laws on land rights that are still commonly practiced, and access to, inheritance and ownership of land is rarely gender-equal (Kutsoati and Morck, 2016;Lambrecht, 2016).Roughly 80 percent of the land in Ghana is under customary tenure (Pande and Udry, 2005). Customary land is controlled by either the family head or the traditional head of the lineage or clan (Lambrecht, 2016). Under the customary practices, individuals are allocated a piece of land that they may use either temporarily or permanently, but whether the individual has the right to sell, rent, sharecrop, borrow, or bequeath the land rests on the community and the specific situation (Lambrecht and Asare, 2016). In some, typically more remote communities, land can still only be accessed through nonmarket transactions such as borrowing, allocation of customary land, gifts, or inheritance, and farmers consider themselves as users rather than owners of the land (Bakang and Garforth, 1998;Lambrecht and Asare, 2016;Doss et al., 2019).Similar to other countries in West Africa, men and women within the same household cultivate separate plots (Doss, 2002), and joint ownership or landholding is rare in Ghana (Lambrecht, 2016). Social norms and customary practices have a considerable influence on women's and men's access to land because they define who is and who is not part of the family or community and what is acceptable in the community (Lambrecht, 2016). Customary access to land is mainly organized through family or kinship that follows maternal or paternal bloodlines, known as matrilineal or patrilineal systems, respectively (La Ferrara, 2007). The Akan ethnicity is the matrilineal ethnic group in Ghana, whereas there are several other patrilineal ethnic groups in the country (Ickowitz and Mohanty, 2015). Most women, especially those belonging to the patrilineal groups, do not inherit their fathers' land because their families do not want the land to be transferred to another family upon marriage (Oduro et al., 2011). Yet, as men are traditionally assigned primary responsibility for providing the main necessities for the households, women's land rights are regarded as secondary and are subject to the primary rights of their male relatives in both matrilineal and patrilineal communities (Lambrecht, 2016). As a result, in rural farm households in Ghana, most or all of the farmland and thus agricultural earnings are under the husband's control (Lambrecht, 2016). Nevertheless, Akan women are reported to have greater decision-making power compared to women in patrilineal ethnic groups (Oduro et al., 2012). The gradual change of customary tenure towards more individualization of land rights and the emergence of land markets provides new opportunities and challenges for men and women to access land, possibly affecting gender relations in households and communities (Bruce, 2014;Doss et al., 2019).The study makes use of Ghana's Feed the Future (FtF) Baseline Dataset. The data was collected from June to August 2012 in the Northern Brong Ahafo, Upper West, Upper East, and Northern Regions to monitor impact of FtF-supported activities (USAID, 2013). A two-staged probability sampling methodology was adopted in the selection of the survey sample. Using the probability proportional to size method, a total of 230 enumeration areas (EAs) were selected from all the EAs within the zone of influence in the first stage. In stage two, 20 households were randomly selected from each of the selected EAs. Probability weights were created to take into account differential probabilities of selection and non-responses from the households, allowing the data to be representative of the population in northern Ghana. A total of 4,410 households were surveyed.Aside from relevant modules on household demography and socio-economic characteristics, the survey contains a module that allows for an estimation of the WEAI. Developed as a measure to evaluate empowerment and inclusion of women in the agriculture sector, the WEAI contains individual-level data collected from both primary male and female decision makers in each household 1 (Alkire et al., 2013). In particular, it contains information on primary male and female decision makers' access to productive capital, influence over agricultural production decisions, decision-making roles over the uses of earnings generated from various incomegenerating activities, membership in economic and social groups, and time allocation. This paper restricts its attention to households with married men and women who were interviewed for the WEAI and engage in at least one agricultural activity. Households living in communities in which no one reports owning land are eliminated. 2 These restrictions reduce the number of observations for our main analyses to 3,468 individuals from 1,734 households.The aim of the analysis is to provide insights in the relationship between landownership on the one hand, and agency and achievements in agriculture on the other hand. Evidently, such analysis is plagued by concerns about endogeneity. Self-reported ownership of land is endogenous to different indicators of agency and achievements in agriculture, as they are driven by similar observed and unobserved factors, such as social network, farming experience and skills, intrinsic agency, and prevailing customary practices and gender norms. In other words, land ownership could lead to empowerment, but more empowered individuals are also are more likely to claim 1 In polygamous households, the wife who has the most decision-making authority in absence of the husband was chosen to be interviewed. 2 We dropped these communities because we are interested in the relationship between self-reported landownership and decision-making in agriculture in settings where landownership is feasible. In several communities in Ghana, customary tenure dominates to the extent that it prohibits anyone from claiming landownership. Inclusion of these communities in the analysis prevents us from distinguishing between the influence of living in a traditional setting that complies with customary rules (as a result, no one reports to own land regardless of the strength of potential claims to using the land) and with the influence of not owning land in a setting where landownership is effectively possible.landownership. Thus, we expect a significant upward bias in the coefficient for landownership when an ordinary probit model is used (probit estimates can be found in appendix Table A1).In our main analyses, we model the association between self-reported landownership and agency and achievements in agriculture by using recursive bivariate probit models3 proposed by Maddala (1986). This model is recursive because the first dependent variable, landownership, appears on the right-hand side of the second equation, making this a recursive, simultaneousequation model (Greene, 2011).The model can be mathematically written as follows: district-level fixed effects, that affect landownership along with agency and achievements in agriculture, respectively. Although not explicitly shown in the above equations, several relevant control variables are also interacted with the gender of the respondent. 4 The two equations are interrelated through the error terms, µ \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 and \uD835\uDF00\uD835\uDF00 \uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56\uD835\uDC56 (Greene, 2011) and are estimated with a recursive bivariate probit model via the full-information maximum likelihood estimator in Stata using the cmp command developed by Roodman (2009).Given that the recursive bivariate probit model allows for joint estimation of landownership with the outcome variables of interest with correlated error terms, we expect that the association between landownership and measures of agency and achievements will be smaller in a bivariate probit model compared to a simple probit model. However, the absence of a strong exclusion variable for landownership precludes us from making causal statements.Additionally, one could advocate to employ ordered probit models rather than binary probit models since the original responses are a scale of input into decision-making. Yet, a Brant test (Brant, 1990) shows that the proportional odds assumption was violated for the data used in our analysis and therefore the use of ordered probit models is not appropriate (Fu, 1998). Below, we discuss the key variables in our analysis.Our variable on women's and men's ownership of land is derived from the WEAI module on access to and decision-making power over productive resources. The survey asks the following questions to each male and female respondent individually: \"Does anyone in your household currently have agricultural land?\" And if so, \"who would you say owns most of the land?\".5 Based on these questions, we construct a dichotomous variable capturing each respondent's ownership of land. Individuals who report that they own most of the land, either solely or jointly, are considered as landowners in our analysis. 6This measure of landownership comes with two main caveats. First, the survey asks about agricultural landownership at the household-level rather than at the parcel-level. Consequently, one cannot distinguish between cases of joint ownership of land compared to different individuals of the same household owning different parcels of land. In Ghana, joint landownership is rare (Deere et al. 2013, Lambrecht 2016, and section 3 above). We therefore interpret responses of 'joint' landownership as if the respondent is the individual owner of at least one plot, whereas at the same time, another household member owns at least one other agricultural plot.Second, we use a subjective and self-reported measure of landownership. In places where land access is at least partly governed by customary tenure arrangements and where land markets have not fully developed, the notion of landownership does not necessarily align with legally recognized ownership and is not easily captured in a dichotomous fashion. In northern Ghana, land is rarely rented or sharecropped. Based on the authors' calculations using the data from the 2012 Ghana Living Standard Survey (GLSS6) for parcels in the Upper-West, Upper-East and NorthernRegions, agricultural parcels are seldom bought, rented or sharecropped in, respectively 0.43%, 0.37% and 0.15% of parcels. Rather, parcels are mainly distributed by the family or community (73%) or used free of charge (26%), providing evidence that land in northern Ghana is still mainly accessed through non-market based, customary land tenure systems. 7 All households in our analysis sample cultivate land, and none of them is therefore de facto landless. However, landownership indicates having land rights that go beyond mere use rights, including the rights to household. Second, from a contextual point of view, the distinction may not be meaningful as explained later in this section. 7 These are similar also to the authors' estimations using the baseline survey from the EGC-ISSER Ghana Socio-Economic Panel dataset collected in 2009-2010. Based on those data, we find that only 0.43%, 0.74% and 0.83% of agricultural parcels are respectively purchased, rented or sharecropped in the Northern, Upper-East and Upper-West Regions of Ghana.decide whether to sell or give away the land. Hence, landownership is not only a resource in the narrow economic sense but also refers to a bundle of rights from an institutional point of view.We focus on farmers' agency and achievements in agriculture using eight different indicators in four domains: agricultural production decisions, agricultural income decisions, participation in farmers' associations, and time allocation. The first three indicators capture agricultural production decisions: agricultural production, agricultural input purchases, and types of crops to grow. In this module, respondents are asked: \"When decisions are made regarding each of these indicators, who is it that normally takes the decisions?\" If the respondent indicates that s/he is the sole decision maker, the questionnaire asks \"And to which extent do you feel you can make your own personal decisions if you wanted to?\" For each of these decisions, respondents who normally and individually make the decision or feel that they can make personal decisions to at least a medium extent are considered to be adequately empowered in that decision. This threshold is identical to the thresholds used for the WEAI sub-components (Alkire et al., 2013). 8The fourth and fifth measures focus on decisions regarding the use of food and cash crop income and are derived from the module capturing household decision-making around production and income generation. The questionnaire of this module first asks: \"Did you participate in [ACTIVITY] in the past 12 months?\" Conditional on participating, it then asks: \"How much input did you have in making decisions about income from [ACTIVITY]?\" Agricultural activities included in this module are food crop farming (crops that are grown primary for household food consumption) and cash crop farming (crops that are cultivated largely for sale in the market).Response options are the following: no input, input into very few decisions, input into some 8 Respondents who did not make decisions regarding the specific activity and did not answer the second questions are coded as missing.decisions, input into most decisions, and input into all decisions. For each agricultural activity, individuals who report that they participate in the activity and have input into at least some decisions are considered to have an adequate say in each of these two decision-making outcomes.Again, this threshold is the same as is used for the WEAI (Alkire et al., 2013). 9Similar to the landownership indicator, a key caveat of these five indicators is that these questions are asked at household-level rather than plot-level. Hence, joint decision-making could either mean that decisions on at least one plot are taken jointly, or it could mean that different individuals decide separately on different plots. Another caveat related to decision-making on income from food and cash crop farming is that the survey limits this question to only those who participate in the activity. In practice, some respondents probably still join in decision making on income from an activity that they do not participate in and therefore we cannot automatically code their answer to zero. Hence, our sample is limited to those who participate in the activity.The next indicator captures membership in agriculture-related groups and is drawn from the group membership module. This module reflects the existence of a variety of groups in each respondent's community and whether the respondent is an active member of each group. Given that agriculture/livestock/fisheries producer's groups, water and forest users' groups, and trade and business associations play an important role in agriculture and in expanding agricultural producers' networks, respondents who report that they are active members of at least one of these groups are considered to have membership in agricultural-related groups in their communities.The final two measures reflect workloads in agricultural work and in productive and reproductive work, respectively, and are based on the time allocation module. This module 9 Respondents who did not participate in the activity and did not answer the second question as well as those who did participate in the activity but did not answer the second question are coded as missing.provides information on time spent on primary and secondary activities At individual level, we control for age and education. Age reflects a person's maturity, experience and authority (Allendorf, 2007;Mishra and Sam, 2016). As individuals become older and thus more experienced or mature, their propensity to own agrarian land and influence decisionmaking is expected to increase. We include age squared to account for a diminishing or eventually reversing effect of ageing on landownership or agency and achievements in agriculture.Educational attainment is proxied by a binary variable equal to zero if a respondent did not complete primary education and one if the individual has completed primary education. Education can empower farmers by allowing them to be more informed and have stronger sense of self-worth, 10 Secondary activities are not included in the estimation of workload in agricultural and non-agricultural activities due to the fact that many respondents (96.53%) do not report any time in agricultural activities performed as secondary work activities.thereby enhancing their social status (Kabeer, 1999;Goldstein and Udry, 2008). Thus, education is another essential factor associated with both landownership and decision-making in agriculture.At household-level, we control for religion, ethnicity, household type and composition, household wealth, and distance to the nearest secondary road. Moreover, we include interaction terms for religion, ethnicity and household type with gender of the respondent. Muslim households typically grant greater land rights and household autonomy to men than women (Sait and Lim, 2006), whereas matrilineal ethnic groups, such as the Akan ethnic group, may give women greater access to land and autonomy compared to the patrilineal ethnic groups (Duncan, 2010). Living in polygamous households has also been documented to affect individuals' landownership along with decision-making roles (Ghebru and Lambrecht, 2017;Ichowitz and Mohanty, 2015). Through a number of pathways, the household composition, such as the number of men and women of similar, older or younger age categories, can affect whether one is able to cultivate land and an individual's agency and achievements in agriculture (Wiig, 2013;Quisumbing et al., 2001).Household wealth is proxied by household asset terciles. 11 Wealthy individuals are less credit constrained and experience greater land tenure security (Goldstein and Udry, 2008), and they often have greater access to information and productive resources (Mishra and Sam, 2016).Lastly, distance to the nearest secondary road is likely to affect self-reported landownership. More accessible locations are more likely to be integrated in more market-oriented land tenure systems, as opposed to customary systems with little individual landownership (Ghebru and Lambrecht, 2017), yet there may be more competition for land (Kleemann et al., 2017). Moreover, communities in more remote areas are often observed to have more traditional gender attitudes, 11 The asset terciles are based on the household asset index. The household asset index is constructed using a principal component analysis (Filmer and Pritchett, 1999) and includes information on the material of the floor, wall and roof, type of toilet, source of household cooking fuel, access to electricity and water, and ownership of dwelling, large and small livestock, fishing equipment, non-farm business equipment, vehicles, and large and small electronic devices.potentially reducing women's landownership compared to men (Fenrich and Higgins, 2001).Finally, district-level dummies are included to account for localized differences in gender norms and perceptions.An important shortcoming of our analysis is that we do not control for the landowning status at household-level, which is impractical in the context of a recursive bivariate probit model.Hence, there is a risk of confounding individual ownership of land with household-level ownership of land. We conduct additional analyses to verify the interpretation of our findings from the main analyses by (i) using a household-, rather than individual-, level landownership variable, and (ii)limiting the sample to only landowning households. In the first set of alternative analyses, the landownership variable is not based on individual landownership, but on household-level selfreported landownership. 12 The rationale for this analysis is to verify whether the observed coefficients in our main analysis are truly driven by individual-level reported ownership. In interpreting the results, one should remember that women generally have secondary land rights in Ghana. In practice, when only one person owns land in the household, this is generally a man.Hence, especially for female farmers, we expect that coefficients will be different.The second set of additional analyses maintains the individual-level landownership variable as in the main regressions, but it restricts the sample to landowning households. More specifically, we limit the sample to households where either the husband or the wife (or both) report that they own land. The aim of this sample restriction is to focus particularly on the role of individual landownership within landowning households. We expect individual landownership to be of primary importance as compared to household-level landownership. In this case, these resultsshould not be significantly different from the main results. Yet, minor changes are to be expected given that this concerns a subsample of the main sample. Similar to the former sensitivity analyses, the relatively small share of men who do not report to own land while living in a landowning household provides an important caveat for the estimations and the interpretation of the results.Table 1 reports gender differences in self-reported landownership and indicators for agency and achievements in agriculture. About two thirds of men in our sample (69%) report that they own land, which is more than five times the amount of women who report owning land (12%). In line with the prevailing customary norms, a significantly greater proportion of men report that they have input into decision-making in agriculture compared to women, respectively about 96-98% versus 41%-44%. These discrepancies are less stark when looking at decisions related to income from agriculture, yet there are still significantly more men than women who have influence over the use of these sources of income. In particular, 97% of men indicate that they are able to influence the use of income generated from food crop farming and cash crop farming, compared to 64% and 58% of women, respectively. As hypothesized, men are more likely to be members of agriculturerelated groups than women; 32% of men are members in agricultural groups, compared to 17% of women.In terms of involvement in work activities, men are more likely to have a high agricultural workload than women. In particular, 38% of men are overworked in agricultural activities, compared to 10% of women. Taking into account all types of work activities, however, women are more likely to have high total workload than men, respectively 35% compared to 15%. a. Stars denote significant gender differences in a two-sided t-test using clustered standard errors by the enumeration area at * p < 0.10; ** p < 0.05; *** p < 0.01. b. Bold denotes significant differences between men (resp. women) who do and do not own land in a two-sided t-test using clustered standard errors by the enumeration area at p<0.05. All estimates are calculated using sampling weights.Table 1 also compares agricultural agency and achievements of men and women who do not own land with men and women who own land. Three key observations can be made. First, men and women who own land are significantly more likely to have agency in agriculture compared to those who do not own land. Nevertheless, those who do not own land are still engaged in agriculture to a relatively large degree. Second, we find a sizeable gender gap not only when comparing men and women who do not own land, but also when comparing men and women who own land. Third, non-landowning men still have significantly more agency and achievements in agriculture compared to landowning women except for membership in agricultural groups.Table 2 provides summary statistics on relevant individual and household characteristics by landownership status and gender. Regardless of their landownership status, men are more likely to be older than their female counterparts. Overall, very few respondents completed primary school, but women did so to a significantly lower extent than men, respectively 5% and 14%.There are, however, no differences in educational attainment between women (men) who own land and women (men) who do not own land. While there is no significant difference in the proportion of male respondents who are Muslim among both landowning categories, we do find that significantly fewer female landowners are of Islam religion. Interestingly, the share of women in polygamous marriages is not significantly different across both landowning categories (around 17%), but we find significantly more men in polygamous marriage arrangements among those who own land, respectively 20% compared to 10% among those who do not own land. Women who own land are also more likely to belong to the Akan when compared to male landowners and to women who do not own land, but there is no significant difference when we compare men and women who do not own land. We observe no significant differences in household composition between women and men with and without landownership. One exception is for the number of female and male elderly, who are more numerous in households of landowning compared to households of non-landowning respondents.Table 2 also sheds light on the gender differences in the relationship between landownership and household wealth. A majority of men (72%) who do not own land live in households that belong to the lowest wealth tercile, whereas a significantly lower proportion of landowning men are in the lowest wealth tercile (19%). Consequently, a significantly larger proportion of men who own land belong to the highest wealth tercile. These patterns are similar, but less pronounced among female respondents. Finally, we find no significant difference in the distance to the secondary road for men who own and men who do not own land, but women who own land live significantly closer to the secondary road, suggesting that proximity to a road relates positively to women's land rights. a. Stars denote significant gender differences in a two-sided t-test using clustered standard errors by the enumeration area at * p < 0.10; ** p < 0.05; *** p < 0.01. b. Bold denotes significant differences between men (resp. women) who do and do not own land in a two-sided t-test using clustered standard errors by the enumeration area at p<0.05. All estimates are calculated using sampling weights. does not significantly affect men's landownership, but we do find that women of Akan origin are more likely to own land than women of other ethnicities. However, women of Akan ethnicity are still significantly less likely to own land compared to men of Akan ethnicity. The latter confirms the persisting gender inequalities in matrilineal ethnic groups (Quisumbing et al. 2001;Lambrecht, 2016;Lambrecht et al. 2018).Age is positively associated with landownership. We do not find a significant effect of primary school completion. The lack of variation in the education variable (less than 10% completed primary education) may mask the relationship between education and landownership.The number of male children under the age of 16 along with the number of working-age women are negatively related to landownership. As expected, respondents in households in the lowest wealth tercile are significantly less likely to own land. Distance to a secondary road is positively and significantly correlated with landownership for men. The reverse is found for women; women who live further away from a secondary road are less likely to own land. The latter is in line with observations that land in more remote areas is governed by a more traditional interpretation of the customary law, which grants men greater land rights than women (Fenrich and Higgins, 2001). Table 4 reports the estimates of the recursive biprobit models for our indicators of agency and achievements. As hypothesized, the negative and statistically significant female dummies suggest that women who do not own land are less likely to participate in decisions concerning agricultural production, agricultural input purchases, types of crops to grow, and use of food and cash crop earnings compared to their male counterparts. Women who do not have landownership are, however, less likely to have high agricultural workload than men who do not own land.For men, owning land is positively associated with decision-making in agricultural input purchases, types of crops to grow, and membership in agricultural groups. However, there is no significant association between landownership and participation in agricultural production decisions and decisions on the use of food crop income, agricultural workload, and total workload for men. The negative and statistically significant coefficient for landownership associated with the use of income generated from cash crop farming may not be meaningful since nearly all men make decisions on how cash crop earnings are used, as reported in Table 1.Contrary to the positive association of landownership with outcome indicators in most of the domains for men, the picture is much more nuanced for women. Compared to women who do not own land, women with landownership are more likely to participate in decisions regarding agricultural input purchases and types of crops to grow and are also more likely to be members of agricultural groups. The associations of landownership with participation in farm income decisions and with workloads do not significantly differ between women with and without landownership. Despite the optimistic findings above, the interaction term of being female and owning land is insignificant in all main regressions. As a consequence, the gender gap observed between men and women who do not own land is similar if we compare men and women who own land.Compared to male landowners, female landowners are less likely to be involved in cultivation and farm income decisions. Women are however less likely to have high agricultural workload.Again, we find that religion, household type, and ethnicity have heterogeneous impacts on men and women. Muslim women have less agency than non-Muslim women in terms of production, purchase, and cropping decisions, whereas women in polygamous marriages are less likely to participate in decision on use of food crop income compared to women in monogamous marriages. Akan women are more likely to participate in decisions on use of income from food and cash crop income compared to non-Akan women, but also have a higher agricultural workload.To the contrary, men of Akan origin are less likely to participate in income decisions compared to non-Akan men.Nevertheless, we continue to observe gender gaps in agriculture within each of these different groups. Compared to their male counterparts, Muslim women are considerably less likely to participate in decisions on agricultural production, input purchases, types of crops to grow, and the use of food and cash crop income. Similarly, women living in polygamous households are less likely to have a say in production, input purchase, and farm income decisions than men in polygamous households. Akan women are also less likely to participate in decisions on agricultural production decisions as well as on the use of income generated from farming compared to Akan men. Akan women are more likely to have high total workload than their male counterparts.Regardless of their religion, household type, and ethnicity, men are more likely to have high agricultural workload than women.Age is positively associated with having agency and achievements in agriculture, but eventually this effect becomes negative at older age. Individuals who completed primary education are less likely to participate in decisions on the types of crops to grow. While this negative correlation may be surprising, they could reflect the fact that better-educated individuals are less likely to be involved in agricultural work than lesser-educated persons. This may also account for why better-educated individuals are less likely to have high agricultural workload.While household composition does not have a significant association with agency and achievements in agriculture, the number of working-age men is negatively correlated with the probability of having high agricultural and total workloads. Household wealth is significantly associated with agency and achievement in agriculture. Compared to individuals belonging to the top asset tercile, those in the lowest tercile are more likely to have a say over purchases of agricultural inputs and types of crops to grow. A likely reason is that impoverished individuals are more likely to rely on agricultural activities as their main source of livelihoods; thus, they would have greater input in agricultural production decisions than wealthier persons. On the contrary, those who belong to the poorest wealth tercile are less likely to be members of agricultural associations. Individuals belonging to the middle asset tercile are less likely to have high agricultural and total workloads than those in the highest asset tercile.Tables 5 andA2 presents the main regression results of the recursive bivariate probit analyses in which the landownership indicator is specified at household-level rather than individual-level (sensitivity analysis a). Three key messages emerge. First, we still observe significant gender gaps in agriculture among those living in households who do not own land for all outcomes, except for membership in agricultural groups. Second, for men, the household's landownership is associated with significantly higher participation in input purchase and crop cultivation decisions and agricultural association membership. The relationship between household-level landownership and production decision and decisions on cash crop income is no longer statistically significant. Third, gender gaps persist for decisions on use of food crop income and workload in landowning households and are even wider compared to non-landowning households for crop cultivation decisions and agricultural group membership. This can be observed from the interaction terms, which have become negative and statistically significant for agricultural production decisions and membership in agricultural groups. Overall, women in landowning households do not have significantly more input into decision-making compared to women in nonlandowning households, except in decision on the use of food crop income and participation in agricultural groups.Tables 5 andA3 reports the regression estimates that use self-reported individual landownership as key variable of interest, but are based on the sub-sample of landowning households (sensitivity analysis b). In other words, it excludes households in which neither the husband nor wife report owning land. Again, we point at three key messages emerging from these results. First, we continue to find negative and statistically significant associations of being female with the outcome variables for farmers who do not report to own land, except for input in decisions regarding agricultural input purchases, membership in agricultural groups, and having high agricultural workload. Second, we no longer observe any significant associations between landownership and our outcome variables, except for input in decisions on food crop income and agricultural association membership. This however, may be partly due to large standard errors related to having only few male respondents who do not own land but live in a landowning household. For the same reason, the negative relationships between landownership and income decisions may not be meaningful. Third, however, the gender gap is maintained or even larger when we consider men and women who own land within landowning households. Women who own land are significantly less likely to have agency in agriculture compared to men who own land.Land provides the basis for food production and most income-generating activities and is the most important asset among agricultural households. Using the 2012 Feed the Future Baseline Survey data, we investigate how landownership is associated with agency and achievements in agriculture for male and female farmers. More specifically, this paper studies whether individuals with landownership are more likely to have agency and achievements in agriculture than those who do not own land, and whether landownership is associated with improvements in women's agency (i.e. participation in decisions on crop cultivation, decisions on income from agriculture, and membership in agricultural associations) and achievements (excessive workload) in agriculture relative to their male counterparts.The results from our analyses show a large gender gap in participation in decisions surrounding agricultural cultivation (production, input purchases, and types of crops). This gap is especially large for women in Muslim households. Not surprisingly, self-reported landownership significantly increases participation in input purchase and crop decisions for both men and women.Nonetheless, we continue to observe a similar gender gap among the landowners. Missing value for the interaction term in the cash crop income regression because there are no men that do not have influence on decisions regarding the use of cash crop income and live in non-landowning households. Missing values for the equation predicting the probability of having high total workload in sensitivity analysis b due to lack of convergence of the regression estimates.We also observe a significant gender gap when considering participation in decisionsmaking on income from cash or food crop farming for those who do not own land, and a similar gender gap among those who own land. Given that women are responsible for household food purchases and other small household expenses such as clothes or school fees (Lambrecht, 2016), women can often participate to some extent on the use of income from farming and we find that they are more likely to participate in these decision compared to cultivation decisions. Yet, men continue to be decision makers on the use of income for household food consumption and for household or agricultural investments.We find that men are more likely to be member of agricultural associations compared towomen. Yet, agricultural association membership is the only domain where we do not observe a gender gap in agricultural association membership when controlling for landownership. In other words, we do not find significant gender differences when comparing men and women who do not own land, and when comparing men and women who are landowners. Landownership is positively correlated with being members of agricultural groups for both women and men.Women are less likely to have a high agricultural workload than men, regardless of their landowning status. To the contrary, however, our descriptive statistics show that women more often have a higher total workload compared to men. Yet these differences are not significant in our regression analyses.These findings paint a modest picture of the association of landownership with an improvement of women's agency and achievements. They substantiate the work of Agarwal (1997) and Kabeer (1999) by showing that both women and men are more likely to participate in decisions on agricultural cultivation and in agricultural groups when they own land. However, owning land does not close the observed gender gap. In northern Ghana, as elsewhere, patriarchal social norms and rules govern men's and women's behavior in the household and on the farm and are not easily overridden by changing notions of landownership. The image of the male breadwinners is pervasive in Ghana, and across many other cultures, and continues to dominate norms and perceptions on household organization (Kabeer, 2016;Lambrecht, 2016). In an agrarian economy, this aligns with men taking up leading roles in agriculture, and a stereotypical image of men being more knowledgeable concerning farming practices (Jost et al., 2016). These norms and perceptions seem to be maintained and applied, irrespective of landownership.In terms of policy implications, the observed discrepancies in men's and women's landownership, gaps in participation in agricultural decision making as well as an unequal workload show that continued efforts to improve women's access to land and voice in agriculture are not misplaced. However, a mere focus on women's landownership may not necessarily lead to reduced inequalities in decision-making in agriculture or ensure a more equal workload for men and women. Improving women's participation in, and benefits from, decision-making in agriculture, will require gender-sensitive development strategies that go beyond advancing women's land rights in Ghana.Although our analysis adds valuable insights to the understanding of how ownership of land is associated with agency and achievement in agriculture, it suffers from shortcomings related to the construction and interpretation of some of the key indicators as well as the lack of a causal analysis that can quantify the impact of landownership. Specifically, in a setting where there is a ","tokenCount":"8047","images":["1751057039_1_1.png","1751057039_1_2.png"],"tables":["1751057039_1_1.json","1751057039_2_1.json","1751057039_3_1.json","1751057039_4_1.json","1751057039_5_1.json","1751057039_6_1.json","1751057039_7_1.json","1751057039_8_1.json","1751057039_9_1.json","1751057039_10_1.json","1751057039_11_1.json","1751057039_12_1.json","1751057039_13_1.json","1751057039_14_1.json","1751057039_15_1.json","1751057039_16_1.json","1751057039_17_1.json","1751057039_18_1.json","1751057039_19_1.json","1751057039_20_1.json","1751057039_21_1.json","1751057039_22_1.json","1751057039_23_1.json","1751057039_24_1.json","1751057039_25_1.json","1751057039_26_1.json","1751057039_27_1.json","1751057039_28_1.json","1751057039_29_1.json","1751057039_30_1.json","1751057039_31_1.json","1751057039_32_1.json","1751057039_33_1.json","1751057039_34_1.json","1751057039_35_1.json","1751057039_36_1.json","1751057039_37_1.json","1751057039_38_1.json","1751057039_39_1.json","1751057039_40_1.json","1751057039_41_1.json","1751057039_42_1.json","1751057039_43_1.json","1751057039_44_1.json","1751057039_45_1.json","1751057039_46_1.json","1751057039_47_1.json","1751057039_48_1.json","1751057039_49_1.json","1751057039_50_1.json","1751057039_51_1.json"]}