WPS6329 Policy Research Working Paper 6329 The Impact of the Global Food Crisis on Self-Assessed Food Security Derek D. Headey The World Bank Development Economics Vice Presidency Partnerships, Capacity Building Unit January 2013 Policy Research Working Paper 6329 Abstract The paper provides the first large-scale survey-based food inflation in some of the most populous countries, evidence on the impact of the global food crisis of particularly India. However, these favorable global trends 2007–08 using an indicator of self-assessed food security mask divergent trends at the national and regional from the Gallup World Poll. For the sampled countries levels, with a number of countries reporting substantial as a whole, this subjective indicator of food security deterioration in food security. The impacts of the global remained the same or even improved, seemingly owing crisis therefore appear to be highly context specific. to a combination of strong economic growth and limited This paper is a product of the Partnerships, Capacity Building Unit, Development Economics Vice Presidency. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at D.Headey@cgiar.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team The Impact of the Global Food Crisis on Self-Assessed Food Security By DEREK D. HEADEY 1 JEL Codes: I32; O11. Keywords: Food crisis, food security, poverty, subjective indicators. Sector Board: Poverty Reduction (POV) 1 Derek Headey, Research Fellow, International Food Policy Research Institute, PO Box 5689, Addis Ababa, Ethiopia. D.Headey@cgiar.org. The author particularly wishes to thank Angus Deaton for the introduction to the GWP data as well as very detailed comments on an early draft. Thanks also to Gallup staff for answering a number of questions and to Shahla Shapouri of the USDA for providing comments and answering questions regarding the USDA model. John Hoddinott, Olivier Ecker, Paul Dorosh, Bart Minten, Maggie McMillan, Maximo Torero, and Shenggen Fan contributed useful comments and suggestions. Participants at various seminars at the FAO and IFPRI provided insightful comments. The author also thanks USAID for financial support and Yetnayet Begashaw, Teferi Mequaninte, and Sangeetha Malaiyandi for excellent research assistance. Any errors are the author’s own. The global food crisis of 2007–08 involved approximately a doubling of international wheat and maize prices in the space of two years and a tripling of international rice prices in the space of just a few months. Understandably, such rapid increases in the international prices of staple foods have raised concerns about the impact on the world’s poor. Household surveys suggest that most poor people earn significant shares of their incomes from agriculture but are nevertheless often net food consumers (World Bank 2008b). Consistent with this stylized fact, several multicountry World Bank simulation studies find that poverty typically increases when food prices increase (holding all else equal), with much of the increase in poverty taking place in poorer rural areas (Ivanic and Martin 2008; de Hoyos and Medvedev 2009; Ivanic, Martin and Zaman 2011). Likewise, the U.S. Department of Agriculture’s (USDA 2009) simulation found that approximately 75–80 million people went hungry during the 2008 food crisis, a number that the Food and Agriculture Organization (FOA) of the UN (FAO 2009) applied to its precrisis baseline numbers in the absence of an FAO model that could simulate a food price shock. 2 Subsequent USDA simulations were used by the FAO to estimate that over one billion people went hungry in 2009, up from 873 million in 2005–06. 3 2 Some basic problems with the FAO model are reviewed in Headey (2011a) and FAO (2002). In the 2008 crisis, the FAO had an underlying model that only incorporated quantities, not prices, so the FAO’s capacity to simulate the effects of food price increases was very limited. Therefore, the FAO relied on a USDA trade model (USDA 2009). A major shortcoming of the USDA model was that it did not include middle-income countries, including large ones such as China, Mexico, and Brazil. Headey (2011a) also shows that the USDA (2009) estimates are contradicted by the USDA’s own historical production and import estimates for 2007–08 (USDA 2011). 3 In addition to the two basic approaches described above (the World Bank poverty simulations and the FAO/USDA hunger simulations), several authors have taken mixed approaches to estimate calorie availability trends, including Anriquez et al. (2010) and Tiwari and Zaman (2010). Dessus et al. (2008) adopt the net benefit ratio approach, but only for urban areas. There are also many country-specific simulation exercises; a particularly good one is Arndt et al. (2008). See Headey (2011a) for a more extensive overview and critique. 2 These studies have led some observers to conclude that global poverty or hunger increased during the 2008 food crisis. Fundamentally, however, most of the simulation studies cited above aim to predict and understand the impacts of higher relative food prices, holding all else equal. The use of this kind of partial simulation approach is justifiable on several grounds. First, partial simulations have an advantage in being able to produce very timely ex ante estimates of what might happen if food prices increase. Second, more sophisticated approaches (Ivanic and Martin 2008; de Hoyos and Medvedev 2009; Ivanic, Martin, and Zaman 2011) are useful for identifying the mechanisms by which higher food prices could influence poverty and the distributional consequences of food price changes. In that sense, they are certainly policy relevant. Third, these approaches provide the scope to explore the sensitivity of results to alternative assumptions. However, the use of partial approaches to infer actual changes in global poverty is inappropriate because there are many ways their predictions might not eventuate. For example, several simulation studies assumed rates of international price transmission to domestic markets rather than using observed price changes (e.g., Ivanic and Martin 2008). There is also the poorly informed question of whether wages (rural and urban) might adjust to higher food prices, with some evidence suggesting that agricultural wages might adjust even in the short run (Lasco et al. 2008). More generally, strong income or wage growth (even without “adjustment�) may have buffered any negative impacts of higher prices in the 2000s, as Mason et al. (2011) observed in urban Kenya and Zambia. More ambiguously, households could mitigate the worst forms of hunger or poverty through any number of coping mechanisms, such as reducing dietary quality, selling assets, working longer hours, or reducing nonfood expenditures. 4 4 Inevitably, measurement and estimation issues constrain these studies. Headey and Fan (2010) and Headey (2011a) provide an overview of some measurement and estimation issues (see also footnote 2). Of course, measurement issues also apply to the data used in this study (see section 2). 3 Because of these complexities, this article takes a different route by providing the first ex post analysis of survey data collected before, during and shortly after the 2008 food crisis across a large number of countries. Specifically, we examine the results from an indicator of self-assessed problems affording sufficient amounts of food, which was collected as part of the Gallup World Poll (GWP). Although subjective data certainly have shortcomings (an issue we discuss in detail below), their advantage in this context is that they are substantially cheaper to collect relative to the more objective monetary or anthropometric indicators found in standard household welfare surveys. Hence, the country and time coverage of the GWP surveys is their primary advantage. Specifically, the GWP surveys allow us to examine self-assessed food insecurity trends in 69 low- and middle-income countries, of which China is the most prominent exclusion. This substantial cross-country coverage also allows us to test whether changes in this indicator are explained by variations in food inflation and economic growth. The basic conclusion from the Gallup data is that at the peak of the crisis (2008), global food insecurity was either not higher or even substantially lower than it was before the crisis. The raw results for the 69 countries for which we have precrisis (2005–06) and mid-crisis (2008) data suggest that 132 million people became more food secure. If 2007 is used as the “precrisis� benchmark, the picture is more neutral because self-assessed food insecurity was essentially unchanged between 2007 and 2008. However, these surprisingly optimistic global trends mask large regional variations. Global trends are clearly driven by declining food insecurity in India and several other large developing countries. However, on average, self-assessed food insecurity increased in many African countries and most Latin American countries. It decreased somewhat in Eastern Europe and Central Asia, but it probably rose in the Middle East (for which the GWP 4 sample is very small). In the average Asian country, there was basically no change, although we again observe variations around the mean. Because this article introduces a new method for gauging trends in global food security, it is especially important to investigate the reliability of the Gallup indicator and to understand the factors that might explain these somewhat surprising results. In the analysis below, we note some of the general shortcomings of subjective indicators, which are now widely used in the contexts of general well-being (e.g., Headey et al. 2010; Kahneman and Deaton 2010; Deaton 2010; 2011), poverty (Ravallion 2012), and food security (Deitchler et al. 2011), as well as some specific problems with the Gallup indicator. We also conduct econometric tests to determine whether the observed trends in self-assessed food security are plausibly explained by changes in per capita GDP and various food price indices. As expected, we find that real economic growth improves self-assessed food security. Real GDP growth already controls for aggregate price changes. We also find some additional effects of aggregate inflation, but we find no significant additional effect of relative food price changes (i.e., changes in the food terms of trade). We also show that in many of the largest developing economies (i.e., those with the largest poor populations), nominal economic growth generally outpaced food inflation, even in 2008. Hence, it appears that strong real income growth has largely offset the adverse impacts of food inflation in many developing countries, including those with the largest poor populations. <>II. AN OVERVIEW OF THE GALLUP WORLD POLL FOOD INSECURITY INDICATOR In this section, we provide an overview of the GWP and the specific food security indicator used in this study. Our goal is limited to answering three questions. First, what is the general quality of the GWP surveys? Second, what limitations might the GWP indicator of self-assessed 5 food insecurity have? Third, do basic cross-country patterns in this indicator align with expectations? Because the GWP is conducted by a private organization and its collaborators, much of the description of the formal survey characteristics relies on Gallup materials. We explore correlations between the GWP indicator and non-GWP welfare indicators by conducting a correlation analysis of a cross-section of countries and, in the next section, a multivariate analysis of the full panel dataset. 5 <>General characteristics of the Gallup World Poll Since 2005–06, the GWP has interviewed households in approximately 150 countries, although not always annually. Most questions are constructed to have yes or no answers to minimize translation errors. In developing countries, all but one of the GWP surveys are conducted face to face (China 2009 is the exception), and most take approximately one hour to complete. The surveys follow a complex design and employ probability-based samples intended to be nationally representative of the entire resident civilian noninstitutionalized population aged 15 years and older. In the first stage of sampling, primary sampling units consisting of clusters of households are stratified by population size, geography, or both, with clustering achieved through one or more stages of sampling. When population information is available, sample selection is based on probabilities proportional to population size; otherwise, simple random sampling is used. Gallup typically surveys 1,000 individuals in each country, except in larger countries such as India (roughly 6,000), China (4,000), and Russia (3,000). In the second stage, random route procedures are used to select sampled households within a primary sampling unit, and Kish grids are used to select respondents within households. Finally, the data are internally assessed for consistency and validity and then centrally aggregated and cleaned. Data weighting 5 Much of what follows is drawn directly from the Gallup Worldwide Research Methodology (Gallup 2010a). The present author purchased country-level data directly from Gallup and corresponded with senior Gallup staff about specific questions. 6 is used to ensure a nationally representative sample for each country, with oversampling corrected accordingly. 6 This approach generates margins of error that are generally in the 3–4 percent range at the 95 percent confidence level, with a mean error margin of 3.3 percent. 7 Note, however, that because these surveys have a clustered sample design, the margin of error varies by question. It is therefore possible that the margin of error is greater for certain questions. We also note that the margins of error in China and India tend to be lower than the average (by 1.6 to 2.6 percentage points). However, in China in 2005–06, the food insecurity question followed some fairly detailed questions on income and welfare, which may have primed respondents to be more likely to answer “yes� to the food insecurity question. Although we were aware of this problem in China, there may be similar problems in other countries. It is certainly possible that the first wave of the GWP (2005–06) contains greater measurement error than subsequent waves because Gallup faced a steep learning curve in conducting such an ambitious global survey (we address this issue below in a sensitivity analysis). <>The Gallup World Poll question on food security Although these general characteristics of the GWP surveys are pertinent, we now turn to the specific question of interest, which is phrased as follows: 6 In a handful of cases, certain sections of the population are oversampled (see appendix S3). For example, urban areas were oversampled in Pakistan, Russia, and Ukraine in at least one year, and in the August–September 2009 survey in China, the provinces of Beijing, Shanghai, and Guangzhou were oversampled, possibly because of the unusual switch to telephone surveying. In other contexts, it appears that Gallup oversampled more educated groups (Senegal, Zambia), and in some developing countries, certain parts of the country were not sampled at all because of ongoing political instability or other accessibility problems. 7 Thus, if the survey were conducted 100 times using the same procedures, the “true value� around an assessed percentage of 50 would fall within the range of 46.7 percent to 53.3 percent in 95 out of 100 cases. 7 “Have there been times in the past 12 months when you did not have enough money to buy the food that you or your family needed?� A simple yes or no answer is recorded. For simplification, we refer to this as the “food insecurity� indicator rather than a more cumbersome term such as “unaffordability of food.� 8 What are some of the strengths and weaknesses of this question? The strengths include a focus on access rather than availability, a recall period (12 months) capable of capturing seasonality and other short-run food price movements, and large cross-country and multiyear coverage. This last strength is a significant advantage in the absence of more regular economic or nutritional surveys, but there are also limitations with subjective data. Unlike simulation approaches, for example, subjective data do not provide much information about the mechanisms or magnitudes of welfare impacts. However, there are some indications that the simple yes/no indicator used here may not lead to much loss of information in practice. The GWP has data for Africa in which a similar question is asked that allows for five different answers based on the frequency of deprivation. Those data show a similar trend to the dichotomous indicator (see fig. S.1 in the supplemental online appendix, available at http://wber.oxfordjournals.org/). A more significant problem is that the definition of food needs is not universal. For a well-off or well-educated family accustomed to a high-quality diet, “food� may mean a food bundle of 8 We note that there are other welfare indicators measured by Gallup, including a question pertaining to hunger rather than food affordability as well as a general life satisfaction question (scaled from one to ten). In earlier versions of this paper, we considered the hunger variable, but the sample size for that indicator was much smaller, and trends in that variable could not be significantly explained by economic growth or food inflation. The life satisfaction question was not explored because it is not obvious that changes in this indicator over 2006–08 would be substantially related to food inflation. Even so, that indicator generally suggests sizeable improvements in well-being in developing countries, with only a handful of exceptions (Pakistan, Sierra Leone, Egypt, and Afghanistan). Hence, we concentrate on the more relevant food insecurity question. 8 sufficiently high quality (e.g., meat, eggs, dairy). For a very poor family, however, “food� may just mean enough cereals or other staple foods. Hence, it is possible that the food insecurity measure is biased upward by education or income or downward by overly low standards of food intake. There is some indication of such biases in the data, although formal tests of the presence of biases proved to be inconclusive (Headey 2011a). For example, there is surprisingly high self- assessed food insecurity in developing countries with relatively high levels of education/literacy, such as the former Soviet Bloc countries and Sri Lanka (see the online supplemental appendix S2 for individual country-year observations). At the other extreme, food insecurity appears too low in several countries where we know that undernutrition is quite prevalent. In Ethiopia, for example, where diets are very monotonous and undernutrition is very high, self-assessed food insecurity was just 14 percent in 2006 (although it subsequently rose rapidly). However, in cross- country regressions, we did not find an impact of education on food insecurity after controlling for income (see Headey 2011a). There are no indications that large numbers of poor countries systematically underreport food insecurity. To illustrate this issue, table 1 reports regional means (the full Gallup data are presented in appendix S2). At the bottom of table 1, we observe that the mean “global� prevalence of households reporting problems with affording food is almost 32 percent. As expected, however, there are large variations around the world, with some countries reporting almost no food insecurity and others reporting that 80 percent of households had problems affording food. For the most part, the pattern across continents is plausible. Food insecurity is highest in sub-Saharan Africa, which is by far the poorest region in the world in monetary terms. Food insecurity in South Asia is higher than in East Asia, as expected, but only when two large outliers, Nepal and 9 Cambodia, are excluded. 9 In Latin America, food insecurity is surprisingly high (34 percent). This may relate to the greater prevalence of urban poverty and of relatively poor net food consumers, although this is only a speculation. Table 1. Regional Unweighted Means for the Two GWP Measures, Circa 2005, for Developing Countries Only (Percent) Food insecurity Mean No. of obs. sub-Saharan Africa 58.3 27 South Asia* 31.2 5 East Asia* 24.0 6 Middle East & North Africa 26.5 2 Central America & Caribbean 34.7 9 South America 36.0 10 a Transition countries 29.1 23 OECDb 8.3 22 c Low income 48.6 49 Middle incomec 29.6 28 c Upper income 11.0 34 Mean, total sample 31.7 433 Source: Data are from the GWP (Gallup 2011). Note: *Note that two outliers are excluded. Nepal is excluded from the South Asia results, and Cambodia is excluded from the East Asia results. In the case of Nepal, its food insecurity score is much lower than that of the other South Asian countries, whereas Cambodia’s is much higher. With the inclusion of these two outliers, the food insecurity scores for South Asia and East Asia are roughly equal at 31 percent. a Transition refers to former Communist countries. b Members of Organization for Economic Co-operation and Development. c. Low income is defined as a 2005 GDP per capita of less than USD 5,000 purchasing power parity; middle income, as USD 5,000–13,000; and upper income, as greater than USD 13,000. The data also suggest a strong income gradient for food insecurity. Low-income countries have food insecurity rates that are 17 percentage points higher than middle-income countries, and the same difference is observed between middle- and upper-income countries. In terms of correlations with other welfare indicators (table 2), there is some support that cross-country patterns impart meaningful information. Of course, extremely high correlations are not necessarily expected given the well-known problems associated with measuring hunger and 9 Self-assessed food insecurity in Cambodia is unusually high (67 percent), but in Nepal, it is extremely low (9 percent). Including these two countries leaves the South and East Asian means roughly equal, at 31 percent. 10 poverty 10 and the fact that anthropometric indicators are heavily influenced by nonfood factors, such as health, education, family planning, and cultural norms. Bearing this in mind, we find that GDP per capita, mean household income, poverty rates, hunger rates, and anthropometric indicators are significantly correlated with the two GWP indicators, almost invariably at the one- percent level (table 2). The correlations are particularly strong for the (logarithmic) income and poverty indicators. In a very small sample—which excludes six important outliers—the correlation between the GWP indicators and the body mass index (BMI) of adult women is also very high (0.68). Table S1.1 in the appendix presents the full correlation matrix among the variables. It shows that the correlations between the GWP measure and the various benchmarks are at least as strong as the benchmark correlations for the FAO hunger measure and the World Bank poverty measure, if not stronger. 10 Indeed, in the context of critiquing standard poverty measures, Deaton (2010) suggested that the Gallup indicators used in this study might be more reliable than the World Bank poverty estimates. As a rough demonstration of their suitability, Deaton showed that the food security variable is highly correlated with GDP per capita. 11 Table 2. Correlations between the Self-Reported Food Security Indicator and Other Indicators of Income, Poverty, Hunger, and Malnutrition, Circa 2005 Alternative poverty/hunger indicator Self-reported (source) hunger GDP per capita, purchasing power parity, log Correlation −0.71*** (World Bank) No. of obs. 44 Household income per capita, USD, log Correlation −0.68*** (World Bank Povcal) No. of obs. 59 Prevalence of hunger Correlation 0.58*** (FAO) No. of obs. 62 Prevalence of poverty, USD 1/day Correlation 0.77*** (World Bank Povcal) No. of obs. 58 Prevalence of poverty, USD 2/day Correlation 0.67*** (World Bank Povcal) No. of obs. 49 Prevalence of low-BMI women, excluding outliers Correlation 0.73*** (DHS & WHO) No. of obs. 17 Prevalence of underweight preschoolers, log Correlation 0.55*** (DHS & WHO) No. of obs. 45 Prevalence of stunted preschoolers, log Correlation 0.48*** (DHS & WHO) No. of obs. 45 Source: Dependent variable is from the GWP (Gallup 2011) . The sources of the independent variables are as follows: World Bank, World Bank (2010b) WDI; World Bank Povcal, World Bank (2010a); FAO; Food and Agriculture Organization (2011); DHS; Demographic Health Surveys (2010); WHO, World Health Organization (2010). Note: *, **, and *** indicate significance at the 10 percent, 5 percent, and 1 percent levels, respectively. All variables are measured in 2005 or the nearest available year. Log indicates that variable is expressed in logarithms to account for a nonlinear relationship. Excluding outliers refers to the exclusion of six countries with the highest prevalence of low-BMI women in the sample, all above 20 percent: India, Bangladesh, Ethiopia, Cambodia, Nepal, and Madagascar. Without this exclusion, the correlation is statistically insignificant. Samples vary in size because of the paucity of some of the poverty and malnutrition indicators. In table 3, we also show that the GWP food insecurity indicator is significantly explained by “relative food prices,� which is measured as the ratio of the purchasing power parity for food items to the exchange rate (both measured in 2005). This index can be interpreted as the extent to which a country’s food basket is expensive or cheap relative to the costs of importing food 12 (values of more than 100 imply that food is relatively expensive, whereas values of less than 100 imply that food is relatively cheap). However, because of Balassa-Samuelson effects, this indicator is likely to be higher in richer countries than in poorer countries. Hence, we use multivariate regressions to control for GDP per capita. However, even after controlling for GDP per capita, there are still substantial variations in food prices across developing countries (as the continent dummies in regression 1 suggest), which could be explained by transport costs, variations in agricultural productivity, the limited tradability of food (partly due to tastes), or even exchange rate distortions. Indeed, regression 2 suggests that variation in “relative food prices� across countries significantly explains variations in self-assessed food security after controlling for GDP per capita. However, the relationship is nonlinear: at low levels of food prices, the marginal effects of higher prices are quite large, but at the highest observed levels of relative food prices, the marginal effects are insignificantly different from zero. A caveat is that the result of regression 2 in table 3 is not very robust to the inclusion of continental dummies (introduced in regression 3), particularly the dummy for sub-Saharan Africa. This lack of robustness appears to be because relative food prices and self-assessed food insecurity are both very high in Africa. 11 Specifically, the inclusion of continent dummies results in the food price coefficients no longer being significant at the 10 percent level, although this insignificance also applies to the continent dummy coefficients, suggesting that multicollinearity is an issue. 11 An issue here is that food prices may be higher in Africa because of the way in which the 2005 round of the International Comparison Program was conducted on a continental basis. Specifically, it is possible that food prices in Africa are biased upward by methodological issues, although it is difficult to substantiate such a claim. A more general problem with purchasing power parities is the challenge of finding common items to compare across countries. Exchange rate distortions may be problematic for this index, although data on black market premia on exchange rates suggest that most exchange rate distortions have declined markedly over time. 13 Table 3. Whether Self-Assessed is Food Security Explained by Relative Food Prices Regression No. 1 2 3 Food price Food Food Dependent variable level insecurity insecurity No. of observations 99 91 91 Constant 61.74*** 17.0** 31.1** GDP per capita ($1,000s) 2.80*** −3.1*** −2.3*** GDP per capita, squared 0.04*** 0.03*** Food price ratio 63.8*** 48.7 Food price ratio, squared −19.4*** −9.2 Africa dummy 30.4 18.6 Latin America dummy −12.3 10.5 Asia dummy 5.0 4.6 Europe-plus dummy −12.5 5.9 R-squared 0.65 0.73 0.76 Adjusted R-squared 0.63 0.72 0.75 Source: “Food insecurity� is from the GWP (Gallup 2011) and is described in the text. GDP per capita is from the World Bank (2010) and is measured in constant purchasing power parity dollars. “Relative food prices� are measured as the purchasing power parity of food and nonalcoholic beverages relative to the nominal exchange rate for the year 2005. Information is from the World Bank (2008b). Note: *, **, and *** indicate significant at the 10 percent, 5 percent, and 1 percent levels, respectively. “Europe-plus� includes Eastern European countries plus North America and Australasia. Note that self-reported food insecurity data are measured in 2005 or 2006, whereas the food price ratio is measured in 2005. Overall, the results reported above present a mixed picture of the validity of cross-country patterns in the Gallup data. On the one hand, there are certainly some worrying outliers in the GWP indicator (particularly in the 2005–06 round). On the other hand, the data as a whole are plausibly patterned across countries and strongly correlated with other welfare indicators and relative food prices. However, we acknowledge that many social scientists are wary of subjective 14 indicators of welfare, even if this skepticism has been moderated in recent decades. There is, of course, an immense body of economic literature that uses indicators of self-assessed well-being and health (e.g., Headey et al. 2010), including indicators collected by Gallup (Kahneman and Deaton 2010; Deaton 2010; 2011). On the positive front, comparisons of self-assessed poverty and objectively measured indicators of poverty have uncovered close relationships between the two (Ravallion 2012). A recent assessment of food insecurity questions in six developing countries also found that questions pertaining to more severe forms of deprivation were highly comparable across countries, although concepts related to anxiety and dietary quality were not (Deitchler et al. 2010). In addition, there are longstanding concerns that such measures are sensitive to framing, question ordering, and other response biases. In terms of the third item, there is an extensive body of literature that examines biases in self-reported indicators (see, e.g., Benitez-Silva et al. 2004; Krueger and Schkade 2008; Ravallion 2012). A specific concern in the context of food security is that respondents may believe that more negative answers increase their chances of accessing food or cash transfers. Many such biases may only exist at certain levels but disappear when trends in the data are observed. However, any changes in question ordering could bias results, as a recent paper by Deaton (2011) shows. Substantial measurement errors could also mean that subjective indicators perform adequately in the cross-section but poorly in first differences (Bertrand and Mullainathan 2001). Clearly, there are important reasons to explore the validity of trends in the Gallup indicator, not just at certain levels. <>III. EXPLAINING CHANGES IN SELF-ASSESSED FOOD INSECURITY In this section, we explore the validity of changes in self-assessed food insecurity at the national level by gauging whether trends in the GWP indicators are explained by changes in 15 disposable income. The underlying model for these regressions is that the prevalence of food insecurity (F) at time t in country i is a function of disposable income per capita, or nominal income per capita (Y), deflated by a relevant set of prices (P): (1) 𝐹𝑖,𝑡 = 𝑓�𝑌𝑖,𝑡 �𝑃𝑖,𝑡 �. Although intuitive in principle, in practice, disposable income at the national level is measured with considerable error for several reasons. First, income inequality means that GDP per capita may be a flawed indicator of the purchasing power of a poor or vulnerable household in a country (the same is true of GDP growth as an indicator of changes in welfare). Second, the price index (P) used to deflate GDP per capita (the GDP deflator) may not represent the consumption patterns of the food-insecure population because the budget share they allocate to food expenditures will typically be higher than the share employed in calculating the (consumer price index) CPI. Because of these complications, it is not obvious that changes in real GDP per capita adequately capture trends in the purchasing power of the poor. Hence, in the regressions below, we estimate several different specifications. First, we vary the choice of price index used to deflate growth in per capita GDP (the GDP deflator, the total CPI, and the food CPI). Second, we test whether changes in the total CPI or changes in relative food prices (i.e., the food CPI over the nonfood CPI) provide some additional explanatory power. Third, we test whether these relationships vary over income levels, in accordance with Engel effects and the fact that welfare programs may play a larger role in determining food security in wealthier countries than economic growth. Finally, we add fixed effects to the specifications to partially control for unobservable factors, such as income inequality and social safety net. 12 12 Although adding fixed effects would seem desirable in principle, the valid addition of fixed effects rests on the 16 In addition to these issues of specification, there are some measurement considerations. First, in our preferred regression models, we specify the dependent variables as the change in the prevalence of food insecurity across two successive periods. This approach is in contrast to most of the analogous growth-poverty literature, in which it is common to measure the dependent variable as a percentage change. However, taking the percentage changes of a prevalence rate can cause scaling problems and create outliers (Deaton 2006; Headey 2011c[[There is no “Heady 2011c� listed in the references section. Do you mean 2011a or 2011b?]]). 13 The only significant advantage of using a percentage change is that it allows for the derivation of elasticities that can be directly compared to the literature that examines the impact of economic growth on poverty. Therefore, in some of our results, we also report these elasticities, although our preferred estimates focus on first differences. A second issue pertains to measurement error in the Gallup data. Some apparent outliers are indicative of this measurement error. In figure 1, we consider potential outliers more systematically with scatter plots between changes in food security and various indicators of assumption that both right-hand side variables are strictly exogenous at all leads and lags, which is unlikely. Hence, we do not solely rely on the fixed effects estimator. 13 The problem with taking percent changes in prevalence rates can be illustrated with an example of a country with high food insecurity and a country with low food insecurity. In the food-insecure country, suppose that food insecurity decreases from 42 percent at time t − 1 to 40 percent at time t. This yields a first difference of two percentage points and a percent change of approximately −4.7 percent (that is, 2/40 × 100). However, an equally large reduction in malnutrition prevalence in the food- secure country from 4 to 2 percent yields a percent change of 50 percent. Not only is a 50 percent change likely to be an outlier, but it is also 10 times the value of the equally large reduction in malnutrition in the high-malnutrition country. Of course, one could argue that this may not matter if percent differences are applied to the right-hand-side variables. In the case of per capita income, however, this is not true because the denominator (initial income) is invariably large enough to produce more meaningful estimates of percent change. Moreover, percent changes in income make sense if there is a diminishing marginal impact of income on food insecurity. 17 economic growth and price changes. 14 In all of the scatter plots, there are some potentially influential outliers, including Azerbaijan, Angola, and Venezuela, which are three oil producers, several Eastern European countries (Armenia, Latvia, Estonia, and Ukraine) and several African countries (Tanzania, Mali, and Malawi). Note that these outliers are sometimes driven by large changes in the dependent variable as well as by unusual economic growth or inflation rates. Measurement error is therefore a problem in both the left- and right-hand side variables. To gauge the influence of outlying observations, we calculated dfbetas (an indicator of the influence of outliers) and earmarked observations with dfbetas greater than 0.2. 15 One option is to run regressions that exclude outliers, which we do in the case of fixed effects regressions. Another option is to use a robust regressor that downweights outlying observations without completely discounting them. Hence, we use both robust regressors and fixed effects estimates that exclude these outlying observations. Furthermore, we report ordinary least squares regressions in appendix S1, in which all outliers are included. 14 Note that in all our regressions, we exclude observations for Zimbabwe because of its hyperinflationary episode, which leaves the country as an enormous outlier on the food inflation-food insecurity relationship. 15 This cut-off is fairly conservative. The usual cut-off for this sample size, 2/sqrt(N), is equal to 0.12. We calculate these dfbetas for various models and exclude a common set of outlying observations: Algeria, 2009; Angola, 2008; Armenia, 2007 and 2009; Azerbaijan, 2007; Botswana, 2008; China, 2008; Denmark 2007 and 2008, and 2009; Djibouti, 2009; Iraq, 2008 and 2009; Kenya, 2007; Kuwait, 2009 and 2010; Romania, 2007; Rwanda, 2009; Tanzania, 2008; Trinidad and Tobago, 2008; Vietnam, 2009; and Zimbabwe, all observations. A good explanation of dfbetas can be found in Stata Web Books: Regressions with Stata, Chapter 2 – Regression Diagnostics: http://128.97.141.26/stat/stata/webbooks/reg/chapter2/statareg2.htm 18 Figure 1. Scatter plots of self-reported food insecurity, economic growth, and various inflation indicators. BWA BWA TZA TZA TURSEN 20 SEN TUR 20 AZE AZE AGO AGO ARM BFA UGA ETH ARM BFA UGA ETH ECU SDN MOZ ECU MOZ SDN MLI KEN DOM 10 DOM MLI KEN 10 Change in food insecurity Change in food insecurity BFAECU ECU BFA LVA EST MYS CMR LKA MWI EST CMR MYS LVA MWI LKA ITA GHA VNM ZAF THA SER PHL MDG ITAPHL THA ZAF GHAMDG SER VNM HRV USA CMR PAK UGA ROM NPLSLV HND ZMB LKA LAO ARM ALBTTO ZMB USASLV ALB LAO ARM CMR NPL HRV ROM UGA HNDLKAPAK TTO LTU VEN ESP SEN BGD KWT IRQ IND PAN SEN IRQ KWT PAN ESP BGD LTU IND VEN UKR GEO TJK TGO SAU MEX CHLSLE KOR EGY UKR UKR SLEMEX EGY TGO CHL UKR SAU MDA NZLGTM COL BLR PHL PAK NGA NPL MDA LAO URY COL MDA COL NZL BLR COL NPLMDA LAO PAK GTM PHL URY NGA NER HND MRT BOL ZAF JOR BGD GBR KAZ CYPAUTAZE ARG GEO LBN BOLMRT AUT GBR HND BGD ARG GEO CYP NER AZE ZAF KAZ JOR IRL DNK GTM FRA TCD KHM CMR ZMB TUN NIC AUSISR ZMB CAN MYS JPN IDN MNG FIN LVABLR UZB JPN FRAAUS CMR IRL IDN CAN FIN DNK GTM TUN KHM MYS ZMB TCD LVAISR BLR NIC ZMB MNG JPN MRT EST CAN GBR ISR PAK IDN YEM NPL JOR IRL BEL SLV CHL LBN ARG PER KGZ CRI CHN AUS VEN BRA CHNISR JOR BRA JPN YEM CAN CHL BEL MRT IRL LBN ARG AUS GBR CRI SLV NPL EST PER IDN PAK VEN 0 0 MEX AREJPN DEU TCD SYR SWE NZL MRT KAZ THA NLD MRT LBN GRC ECU VNM SGP NGA LTU GHA DOM BRA AFG AZE BLR SWE SGP DEU DEUAZE JPNDEU SGP SGP SWE ECU NLD SWENGA VNMBLR KAZ SYR GRC MEX MRT THA MRT NZL BRA GHA DOM LTUTCD THA ZAF SGP ITA CRI USA ARE ESP KOR BRA BHR SGP AUSNOR DEU UZB HND BGD URY IND BEL ZAFBHR ITA SGP AUS BEL NOR ESP USA BGD BRAHND SGP THA IND DEU URYCRI SWE KGZ BOL CAN TZA CHE NZL KAZ FRA LKABENJPN LBN ISR ESP FRA LBN JPNCAN NZLISR LKA CHE ESP BEN BOL SWE KAZ TZA GBR KWT KEN JOR ARG LBR CRI NGA COL NGA NPL MYS IDN HTI GHA KOR FRA ITA DNK IDN PAN NLD LTU ARG ITA NGA NLD MYS FRADNK IDN GHA JOR NGA LTU GBR IDN COL NPL CRI KWT KEN PAN HTI NIC LVA SLV DOM SEN GEO BGD PRY IND CZE SLV CZE DOM SEN BGD IND GEO PRYLVA NIC PRY DNK CZE COM GTM TJK KHM PERHUN MDA PER DNK GTM CZE HUN KHM MDA PRY BOL KHM PHL BOL KHM PHL CHL USA LKA KEN BDI DZA UKR NER KGZ PAK MAR NER MAR USA CHLDZA PAK LKA UKR URY MAR EGY EST MEX JOR MAR EST MEX URY JOR EGY -10 -10 CMR IND KHM CMR KHM IND TUR UGA VNM BFA DZA TJK BFA DZA TUR UGA VNM TUN POL AFG POL TUN IRQ IRQ UGA TZA TZAUGA KEN ROM ROM KEN MOZ VEN MOZ VEN TCD RWA TCD RWA -20 -20 DJI CHN CHN ARM ARM MWI MWI -30 -30 MLI MLI -20 0 20 40 -20 0 20 40 60 Economic growth Food inflation BWA BWA TZA TZA TUR SEN 20 SEN TUR 20 AZE AZE AGO AGO BFA ARM UGA ETH ARM UGA ETH BFA ECU MOZ ECU MOZ DOM MLI KEN 10 MLI DOM KEN 10 Change in food insecurity Change in food insecurity BFA ECU ECU BFA CMR MYS EST LVA LKA MWI EST MWI LVA CMR MYS LKA VNM THAITA PHL ZAF MDGSER GHA GHA PHL ITA THAMDG ZAF SER VNM USA LAO CMR HNDUGA SLV TTO NPL ARM HRV ALB ROM PAK LKA ZMB ZMB ALB SLV LKAROMUSA PAK ARM HRVCMR NPL UGA HND LAO TTO IRQ SEN KWTESP PAN BGD LTU IND VEN SEN BGD KWT LTUPAN ESP IND IRQ VEN TGO MEX CHL SLE EGY SAU UKR UKR UKR SLE UKR MEX EGY CHL SAU TGO NZL LAO COL URY COLNPL MDA PHL GTM BLR NGA PAK MDA MDA MDA PAK BLR NPL COL NZLCOL GTMLAO NGA PHL URY NER MRT AUT GBR CYPBGD JOR HND GEO KAZ AZE ARG ZAF BOL BOL GBR MRT ARG KAZ AUT HND JOR BGD GEO ZAFCYPAZE NER JPN CMRIDN FRATUN IRL ISRMYS GTM CAN DNK AUS LVA FIN ZMB NIC KHM TCD MNG BLR ZMB TCD ZMB AUS KHM FIN JPNCAN DNK FRAGTM IDN IRL TUN CMR LVA MYSZMB BLR ISR NIC MNG JPN CHNCAN SLV MRT LBN JOR NPL YEM ISR CHL BEL AUS PER IDNGBR PAK EST BRA IRL ARG CRI VEN BRACRI IRLISRARG JOR YEM BEL CHL CHNMRT GBR JPNLBN AUS CANESTPAK SLV IDN NPL PER VEN 0 0 SGP JPN DEU SWE MRT NLDSWE MRTNZL DEU ECU TCD THA MEX BRA AZEVNMSYR NGA LTU GRC KAZ BLR DOM GHA GHA AZE DEU NGABLR DEU JPN KAZ NLD SGP SGP SWESWE VNM ECU SYR GRCDOM THA MRT MEX LTUNZLBRA TCD MRT THA BHR ITASGP ESP NOR USA ZAF DEU BEL SGP BRA AUS URY HND IND BGD CRI ZAF SGP AUS BEL USABGD BHR BRA HND ITA NOR ESP IND SGP DEU URY CRI THA ISR SWE JPNLBN CHE LKA NZL FRA CANTZA BENESP BOL KAZ CAN FRALBN NZL LKA BOL JPN ESP CHE BEN ISR SWE KAZ TZA MYS JOR GBRIDNNLD LTU ITA IDN FRA COL DNK PAN CRI KENNGA NPLNGA ARG GHA HTI KWT NGAARG GHA NLD ITA KWT FRADNK IDN MYS NGA LTU NPL COL GBR HTI CRIKEN IDN JOR PAN DOM SEN PRY NIC SLV CZE BGD IND GEO LVA CZE SEN BGD SLV GEO LVA IND DOM NIC PRY PER DNK GTM CZE KHM HUN PRY MDA PER GTM DNK HUN CZE MDA PRY KHM PHL BOL KHM BOL KHM PHL MARUSA NER CHL DZA PAKLKA UKR NER USA PAK DZA LKA CHL MAR UKR EST MEX URY JOR EGY MAR MAR URYMEX EST EGY JOR -10 -10 CMR KHM IND CMR KHM IND BFA DZA VNM UGA TUR BFA DZATUR UGA VNM POL TUN POL TUN IRQ IRQ UGATZA TZA UGA KEN ROM ROM KEN VEN MOZ MOZ VEN TCD RWA TCD RWA -20 -20 CHN CHN ARM ARM MWI MWI -30 -30 MLI MLI -10 0 10 20 30 40 -20 0 20 40 Non-food inflation Relative food inflation Sources: The Y-axis variable is from the GWP (Gallup 2011) . Economic growth data are from the IMF (2010), and food inflation data are from the ILO (ILO, 2011). Turning to some results, we begin with descriptive statistics and correlations for our dependent and independent variables (tables 4 and 5). Over the entire period, the mean change in the first difference of the food insecurity measure was close to zero (0.2), although the standard deviation and range of this variable is quite large. The statistics for the percentage change in food insecurity show a similar pattern and indicate the presence of some of the previously mentioned problems with the use of percentages of a prevalence variable. There is a tendency to inflate 19 small changes at lower levels of food insecurity due to the small base. Next, the three economic growth indicators show similar variation around the mean, but the relatively rapid rate of food inflation over this period means that the GDP growth deflated by the food CPI has a mean of only 0.4, whereas deflating by the GDP or CPI deflators results in means of 2.7 percent and 2.9 percent, respectively. Thus, food inflation typically exceeded nonfood inflation. Turning to table 5, it is noteworthy that the correlations among different price indices are quite large, as high as 0.82 in the case of the relationship between food inflation and total inflation. Table 5 also presents some bivariate evidence that changes in food security are significantly related to both economic growth and overall inflation but not to our estimates of relative food inflation. Table 4. Descriptive Statistics for Dependent and Independent Variables Count Mean Std. De. Min. Max. Change in food insecurity 296 0.2 7.2 −31.0 24.0 Percent change in food insecurity 290 4.9 31.6 −83.0 200.0 Economic growth (GDP deflator) 291 2.9 5.7 −17.6 32.1 Economic growth (CPI deflator) 276 2.7 8.4 −27.1 41.1 Economic growth (food CPI deflator) 271 0.4 8.4 −34.2 34.0 Total CPI inflation 276 8.6 7.5 −8.9 51.8 Food CPI inflation 276 10.9 9.6 −11.5 67.6 Nonfood CPI inflation 276 6.4 6.2 −9.5 34.4 Relative food inflation 276 4.4 7.7 −20.8 41.8 Source: Food insecurity is from the GWP (Gallup 2011). Economic growth data are from the IMF (2010), and all inflation data are from the ILO (ILO 2011). Note: All data are in percent or percentage points. Economic growth is reported with three different means of deflation: the GDP deflator, the CPI deflator, and the food CPI deflator. Relative food inflation is the change in the ratio of the food CPI to the nonfood CPI. 20 Table 5. Correlations between Changes in Food Insecurity and Various Explanatory Variables Change in Economic Total Food inflation Nonfood food insecurity growtha inflation inflation Economic growtha −0.10** Total inflation 0.18*** 0.20*** Food inflation 0.15*** 0.19*** 0.82*** Nonfood inflation 0.19*** 0.19*** 0.71*** 0.51*** Relative food inflationb 0.04 0.06 0.41*** 0.76*** −0.17*** Note: *, **, and *** indicate significance at the 10 percent, 5 percent, and 1 percent levels, respectively. a Growth in GDP per capita deflated by the GDP deflator. b Changes in the ratio of the food CPI to the nonfood CPI. Table 6 reports the results for the full sample of countries with first differences in food insecurity as the dependent variable and various indicators of economic growth as the sole explanatory variable. Regressions 1 through 3 report results from the analysis using the robust regressor, and regressions 4 through 6 report results from using the fixed effects estimator. The main finding from table 6 is that the economic growth coefficient is always highly negative, significant, and quite large in magnitude. In terms of the size of the coefficients, the point estimates suggest that doubling the GDP per capita would reduce the rate of food insecurity by 12 to 24 percentage points, depending on the estimator and the indicator of economic growth. In general, the fixed effects estimators produce larger estimates. When fixed effects are used and outliers are removed, the choice of deflator makes virtually no difference. In table 6, we report elasticities in addition to the first difference coefficients. The elasticities are quite large, varying from 0.47 to 1.25, and are commensurate in size to growth-poverty elasticities (for example, those reported in Christiaensen et al. 2011). 21 Table 6. Regressions of Changes in Self-Reported Food Insecurity against Economic Growth Regression No. 1 2 3 4 5 6 Means of deflating GDP Total Food GDP Total Food economic growth deflator CPI CPI deflator CPI CPI Outliers removed? No No No Yes Yes Yes Robust Robust Robust Fixed Fixed Fixed Regressor regressor regressor regressor effects effects effects Economic growth Coefficients −0.24*** −0.14*** −0.12*** −0.21*** −0.22*** −0.23*** (0.06) (0.04) (0.04) (0.08) (0.06) (0.07) Elasticities −0.56** −0.55*** −0.47** −1.25** −0.93*** −0.82*** (0.27) (0.20) (0.19) (0.48) (0.29) (0.30) No. of observations 291 275 271 271 256 252 No. of countries 120 112 111 113 106 105 R-squared 0.05 0.04 0.03 0.06 0.05 0.05 Source: Dependent variables are from the GWP (Gallup 2011) . Economic growth data are from the IMF (2010), and food and total CPI data are from the ILO (ILO 2011). Note: *, **, and *** indicate significance at the 10 percent, 5 percent, and 1 percent levels, respectively. Standard errors are reported in parentheses. The robust regressions are estimated using the rreg command in stata, with default settings. For fixed effects regressions, standard errors are adjusted for country clusters. Outliers are identified based on dfbetas greater than 0.20. Economic growth is the percent change in GDP per capita between the two years in which the GWP surveys were conducted. Note that the robust regressor does calculate a pseudo R-squared, but it is generally regarded as inappropriate to report this value. Hence, the R-squared reported in this table is derived from an ordinary least squares regression that excludes outlying values. In table 7, we run the same regressions with the addition of separate price change indicators to determine whether certain types of inflation have additional explanatory power over real economic growth rates. Specifically, we add inflation in the total CPI and food CPI relative to the nonfood CPI. The first represents an aggregate price effect, and the second represents a relative food price effect. Table 7 shows that overall inflation has a significant positive effect on 22 the prevalence of food insecurity. Again, the coefficient point estimates are larger in the fixed effects regressions (0.22 versus 0.11), but these marginal effects are relatively large for both estimators. Doubling the CPI, for example, is expected to increase the prevalence of food insecurity by 11 to 22 percentage points, holding real economic growth constant. Somewhat surprisingly, the relative food inflation coefficients in table 7 are insignificant at the 10 percent level, but they are still positive (in one regression, the relative food inflation coefficient is significant at the 13 percent level). One explanation may be greater measurement error in relative food inflation because we were required to estimate nonfood inflation rates for approximately half of our sample. 16 Nevertheless, the fact that food inflation was the main driver of overall inflation over the period in question (food inflation explained almost 80 percent of variation in total inflation from 2006 to 2008 in developing countries) indirectly points to the generally adverse role of higher food prices on self-assessed food insecurity. Moreover, a significant additional effect of overall price inflation on food insecurity could be consistent with microeconomic theories of labor markets. Specifically, most poor people engaged in wage labor (i.e., those who are not self-employed, such as farmers) tend to work in labor markets that are characterized by substantial slack (unemployment or underemployment). If various food and nonfood prices increase, then the nominal wages of workers in such markets would not be expected to increase commensurately, leading to a fall in real incomes (Headey et al. 2012). 16 The reason for the larger error in the relative food inflation measure is that the ILO only reports the total CPI and the food CPI. Because relative food inflation is measured as changes in the ratio of the food CPI to the nonfood CPI, we had to derive the nonfood CPI from the total CPI, the food CPI, and the share of food in the total CPI. However, only approximately 50 percent of countries reported the food weight to the ILO, so we were required to estimate food CPI weights for the remaining countries using regressions against GDP per capita (i.e., Engel effects). This interpolation is the best we could do, but it may mean that relative food inflation is measured with sizeable error. That said, alternative indicators of relative food prices, such as the change in the food CPI minus the change in the total CPI, essentially yield the same insignificant results. 23 Hence, it is possible for nominal price increases to induce real wage declines, and there is significant evidence pointing to the adverse impact of inflation on poverty reduction (see Ferreira, Prennushi, and Ravallion 2000). 17 17 Ferreira et al. (2000) write, “While changes in the relative short-term returns to holding bonds versus stocks may redistribute income only among the non-poor, there is one major asset-type impact which affects the poor: inflation. The rate of inflation is a tax on money holdings. Because there are barriers to entry in most markets for non-money financial assets, the poor are constrained in their ability to adjust their portfolio to rises in inflation. Typically, they will hold a greater proportion of their wealth in cash during inflationary episodes than do the non-poor. The non-poor are generally better able to protect their living standards from inflationary shocks than the poor.� They go on to cite evidence from India, Brazil, the Philippines, and a larger cross-country review. 24 Table 7. Augmenting the Regressions with Measures of Inflation Regression No. 1 2 3 4 Means of deflating Total Total Total Total economic growth CPI CPI CPI CPI Robust Robust Fixed Fixed Regressor regressor regressor effects effects Economic growth (CPI) −0.14*** −0.13*** −0.22*** −0.19*** (0.04) (0.04) (0.06) (0.07) Relative food inflation 0.04 0.12 (0.05) (0.08) Total inflation 0.11*** 0.22** (0.05) (0.11) Number of countries 105 105 105 105 Number of observations 252 252 252 252 R-squared: overall 0.05 0.04 0.05 0.06 Source: Dependent variables are from the GWP (Gallup 2011) . Economic growth data are from the IMF (2010), and food and total CPI data are from the ILO (ILO 2011). Note: *, **, and *** indicate significance at the 10 percent, 5 percent, and 1 percent levels, respectively. Standard errors are reported in parentheses. Note that outliers are removed for all regressions. Outliers are identified based on dfbetas greater than 0.20. The robust regressions are estimated using the rreg command in stata with default settings. For fixed effects regressions, standard errors are adjusted for country clusters. Economic growth is the percent change in GDP per capita between the two years in which the GWP surveys were conducted deflated by the total CPI. Total inflation is the percent change in the food CPI between the month of the GWP survey and the month of the previous GWP survey, where the food CPI in any given month is actually the average food CPI in the previous 12 months. Relative food inflation is the percentage change in the ratio of the food CPI to the nonfood CPI, where the both CPIs in any given month are actually the average CPIs in the previous 12 months. Note that the robust regressor does calculate a pseudo R-squared, but it is generally regarded as inappropriate to report this value. Hence, the R-squared reported in this table is derived from an ordinary least squares regression that excludes outlying values. 25 Finally, we ran a number of additional specification tests related to income-level effects and alternative inflation effects. Specifically, we ran interaction terms with GDP per capita (in linear and log form) and with income dummy variables (low, middle, upper). Although we strongly expected that changes in food insecurity would be more sensitive to changes in disposable income at lower levels of income, there were no significant interaction terms (results available upon request). We suspect that this result may be driven by the fact that growth rates, inflation rates, and changes in food insecurity were all much lower in upper-income countries, which would have the effect of making the relationships approximately linear. From the perspective of providing validation that changes in self-assessed food insecurity impart useful information, the results in tables 6 and 7 are encouraging. It is particularly encouraging that changes in real GDP per capita significantly explain changes in self-assessed food insecurity, suggesting that the latter is sensitive to changes in disposable income. Despite significant and robust marginal effects, there are some caveats to these results. First, there is the influence of outliers. In the online appendix (table S1.2), we report the results of reestimating the regressions in table 6 and including outliers. Although all of the economic growth coefficients are still significant at the 10 percent level or higher, the standard errors are significantly larger, and the point estimates are sometimes larger and sometimes smaller in magnitude than those in table 6. Our treatment of outliers therefore does not lead to qualitatively different results. Nevertheless, the presence of outliers and the low explanatory power of the regressions reemphasize our concerns about measurement error. These concerns must be tempered, however, because the analogous literature on the impact of economic growth on poverty reduction reports regression models with similarly low explanatory power (see Christiaensen et al. 2011, for 26 example), suggesting that these types of short-run poverty/food insecurity episodes suffer from the measurement errors and misspecification problems noted above. Although the presence of large marginal effects of economic growth and inflation rates on self-assessed food insecurity are encouraging, we must interpret trends in the latter quite cautiously. <>IV. MEASURING AND INTERPRETING KEY TRENDS IN THE GALLUP DATA In the introduction to this paper, we noted our basic result at the global level: 132 million fewer people were food insecure in 2008 relative to 2005–06. In this section, we examine Gallup trends in more detail by observing regional variations within this global trend, considering important exclusions from the sample, engaging in an important sensitivity analysis, and exploring the factors that might explain the surprisingly positive global trend. In table 8, we report simple averages of the GWP food insecurity indicator by various regions of the developing world for 2005–06, 2008, and 2009. These years quite neatly correspond to a precrisis survey round, a food crisis round, and an early financial crisis round. Starting at the top of table 8, we observe what superficially explains the very positive global trend: in the eight most populous developing countries (excluding China), food insecurity decreased by 4.7 percentage points between 2005–06 and 2008. However, in many other regions of the world, food insecurity increased, including coastal West Africa (but not the Sahel), Eastern and Southern Africa, and Latin America. In other developing regions, there was either no change or some improvement. We also note that the deterioration of food insecurity in much of Africa and Latin America is consistent with a number of simulation studies (see Headey and Fan 2010 for a review). 27 Table 8. Regional Trends in Self-Reported Food Insecurity (Percent Prevalence) No. of 2005–06 surveys 2008 surveys 2009 surveys Developing region obs. (precrisis) (food crisis) (financial crisis) Eight most populous developing 8 32.7 28.0 30.6 countries* sub-Saharan Africa 14 55.8 54.6 57.2 West Africa, coastal 4 48.5 51.3 58.0 West Africa, Sahel 5 59.6 49.2 55.2 Eastern & Southern Africa 5 57.8 62.8 58.6 Latin America & Caribbean 15 33.2 36.4 35.7 Central America, Caribbean 7 38.4 41.4 40.3 South America 8 28.6 32.0 31.6 Middle East (including Turkey) 3 19.7 26.0 21.3 Transition countries 13 31.9 30.2 34.6 Eastern Europe 6 21.8 19.7 25.8 Central Asia 7 40.6 39.1 42.1 Asia 12 28.8 29.0 30.8 East Asia 7 30.1 30.6 32.7 South Asia 5 26.8 26.8 28.6 Source: Author’s calculations from GWP (Gallup 2011) self-reported food insecurity prevalence rates. Note: * “Large and fast growing� includes India, Indonesia, Brazil, Pakistan, Bangladesh, Nigeria, Mexico, and Vietnam but excludes China. Although the results in table 8 cover the majority of the developing world’s population, there are still sizeable omissions. Although the GWP surveys cover China, we excluded the 2005–06 rounds due to specific concerns about biases in the responses to the food insecurity question. However, a number of other countries are lacking the requisite data for 2005–06 or 2008. China, of course, has a population of over a billion people, but 16 other omitted developing countries 28 represent close to 600 million people. Hence, one way to explore the sensitivity of our “global� estimate to the omission of these countries is to posit some plausible trends for these omitted countries and then recalculate the global figures. With regard to China, the assessed GWP observations for 2006 and 2008 suggest an unrealistically large drop in food insecurity over that time (20 percentage points), which is probably related to the aforementioned problems with the ordering of questions in the 2006 round. It is therefore pertinent to consider a more plausible scenario for China and what this scenario would suggest about global trends in food insecurity. Given China’s phenomenal economic growth and rather limited level of food inflation (nominal mean incomes increased by 65 percent over 2006–08, whereas the food CPI increased by approximately 30 percent), it is plausible that food insecurity fell several percentage points in China. We thus consider a 3- percentage-point reduction from 2006 to 2008 to be relatively conservative. However, the countries omitted from one of more rounds of the GWP include many that could be suspected to have experienced rapid food inflation, including the Philippines (the largest rice importer in the world), a number of Middle Eastern and North African countries (some of the largest wheat importers in the world), and Ethiopia (the second largest country in Africa, one of the poorest countries in the world, and a country that experienced one of the fastest inflation rates in the world over 2007–08). In table S1.3 in the appendix, we make rather pessimistic assumptions about trends in food insecurity in these 16 countries (based largely on observed food inflation data) and adjust the raw GWP estimates by adding the assumed changes in food insecurity from the omitted countries. The results of this exercise are assessed in table 9. The inclusion of assumed changes for these 16 countries adds 62 million people falling into food insecurity rather than coming out it, but the assumed trend in China would result in close to 40 million people 29 coming out of poverty. In short, the core results reported in the introduction are not highly sensitive to the omission of these admittedly important countries. Table 9. Alternative Estimates of Global Self-Reported Food Insecurity Trends after Allowing for Omitted Countries (Millions of People) Estimated change in global food insecurity, Estimation scenarios 2005–06 to 2007–08 Raw results, 69 countries (excluding China), covering 57% −132 of developing world population As above, plus pessimistic assumptions for 16 omissions, covering 67% of developing −60 world population As above, plus a 3-percentage-point reduction in China, −100 covering 87% of world population Source: Author’s calculations from GWP data (Gallup 2011), FAO Global Information and Early Warning System data (2010), and ILO food inflation data (2011). Note: See text in this section for more details regarding the assumptions and data as well as table A3. Another objection may be that the 2005–06 GWP results are less reliable than subsequent rounds because the first round of the GWP may be regarded as a trial run for Gallup. We have the option of using the second round of the GWP (2007) as a base year instead of the 2005–06 round, but the 2007 round contains fewer countries and does not include China. Nevertheless, the 2007 GWP round includes India and other large countries and therefore covers approximately 43 percent of the population in the developing world. A second potential problem with using the 2007 round as a base year is that maize and wheat prices were already increasing in 2007, so it is difficult to regard 2007 as a pure precrisis period. Thus, we might underestimate the food insecurity impacts of the crisis if the 2005–06 round is shown to be unreliable. However, we note that there is no analogous problem with the 2008 data. The vast majority of the GWP surveys in 30 2008 were conducted in the last three quarters of the year after international food prices peaked. Therefore, they cover the period of peak international prices. Some lag in domestic food inflation may still be a problem, although we have already assessed results for surveys conducted in 2009, which may capture the twin effects of slower growth (due to the financial crisis) and higher food prices. Bearing these caveats in mind, table 10 reports the results of calculating the population- weighted averages of food insecurity prevalence and population numbers for 2007 and 2008. The results thus suggest that there was basically no change in the “global� prevalence of food insecurity between 2007 and 2008. However, table 8 also shows that this result is heavily driven by trends in India, where food insecurity fell 4 percentage points from 2007 to 2008. The bottom half of table 8 calculates trends excluding India (which, admittedly, represents approximately one-quarter of the developing world’s population) and finds that population-weighted food insecurity in the rest of the sample went up by 2.53 percentage points, representing approximately 43 million people. Therefore, using the 2007 round as a base suggests that many developing countries were somewhat worse off in the peak food crisis year relative to the previous year. The largest increases in self-assessed food insecurity occur in Tanzania (23 points), Turkey (21 points), Burkina Faso (14 points), Uganda (14 points), Mozambique (12 points), Kenya (11 points), Ecuador (10 points), Cameroon (9 points), Sri Lanka (9 points), Armenia (7 points), and Honduras (7 points). Although we cannot ignore measurement error and the role of other factors in explaining these trends (the result in Turkey stands out as somewhat implausible), it is notable that many of the countries listed above did experience quite rapid food inflation. Indeed, the average rate of food inflation in these countries was approximately 4 points higher than the rest of the sample. 31 Table 10. Changes in Self-Reported Food Insecurity from 2007 to 2008 Prevalence of food insecurity (%) Population of food insecure (millions) 48 developing countries (43.3% of developing world population) 2007 29.33% 821.4 2008 29.28% 820.1 −0.05 percentage points −1.3 million 47 developing countries excluding India (23.3% of developing world population) 2007 31.51% 532.8 2008 34.04% 575.9 2.53 percentage points 43.1 million Source: Author’s calculations from GWP data (Gallup 2011) . Although we have explored validity issues in previous sections, another relevant question is whether the GWP results are supported by any other survey evidence. One other reasonably large survey of developing countries that was conducted before and during the crisis is the Afrobarometer survey. A recent working paper by Verpoorten, Arora and Swinnen (2011) explores trends in an Afrobarometer indicator that pertains to a very similar question to the one asked in the GWP and finds a 3-percentage-point increase in food insecurity in urban Africa from 2005 to 2008 and a 2-percentage-point increase in rural Africa. Thus, the overall picture of some deterioration in food insecurity in Africa is common across both the GWP and Afrobarometer surveys. Second, and perhaps most important, the most recent World Bank estimates of poverty trends also suggest that global poverty fell between 2005 and 2008 on every continent (World Bank 2012). Third, the FAO (2012) has revised its estimates of large increases in global hunger in 2009. The most recent estimates show a relatively steady decline in global undernourishment, consistent with both the Gallup and World Bank poverty estimates. 32 Finally, it is worth exploring why the GWP results (and the new World Bank and FAO numbers) tell a positive story at the global level. One clear pattern is that events in the largest developing countries heavily influence any appraisal of global trends, not only because of the obvious influence of their sheer size on global trends, but also because many large countries are characterized by limited food inflation, rapid economic growth, or both. The first of these is not surprising. Large countries are very reluctant to rely heavily on significant food imports and are more likely to impose export restrictions and set aside significant food reserves. For example, China, India, Indonesia, and Vietnam all imposed some restrictions on grain exports in 2007 or 2008, and Nigeria abolished a 100 percent tariff on rice imports (Headey and Fan 2010). Of course, the effect of these attempts to insulate domestic markets on global poverty is ambiguous given that effective trade restrictions by large countries may protect their own poor but may hurt the rest of the world’s poor by spurring further international food inflation. Another domestic policy factor that may explain the apparent reduction of food insecurity in some of the larger developing countries is the spread of major social safety net programs in these countries, particularly India’s National Rural Employment Guarantee Scheme. However, in addition to these factors, strong economic growth in most of the world’s largest developing countries clearly provides a plausible explanation for the largely favorable trends in self-assessed data in these countries. To examine disposable income issues more explicitly, we deflate nominal economic growth in recent years by changes in the food CPI rather than by an overall price index (as we did in some of the regressions in tables 6 and 7). This indicator is clearly an imperfect indicator of food security because poor people also spend money on nonfood items (but not much on fuel, which is the major source of nonfood inflation) and because mean GDP growth is often not 33 representative of the income growth of lower income groups. Nevertheless, this crude indicator of “food-disposable mean income� at least indicates whether mean nominal income growth outpaced food inflation. Figure 2. Nominal average per capita GDP deflated by the food CPI, 2005 to 2008. 2500 8000 2005 2006 2007 2008 7000 2000 6000 1500 5000 1000 4000 3000 500 2000 0 1000 0 Brazil Mexico Source: The indicator above is nominal GDP per capita between 2005–06 and 2007–08 from the IMF (2011) deflated by food CPI data from the ILO (2011). 34 Figure 2 plots this indicator for the nine largest developing countries from 2005 to 2008, with Brazilian and Mexican incomes measured on a separate axis because of scaling issues. The results are quite striking. “Mean food-disposable income� rose by over USD 700 per capita in China, over USD 1,800 in Brazil, and over USD 400 in Indonesia. In India, the increase was surprisingly modest (USD 80), but the increase was notably large in Nigeria (USD 240). In Mexico, however, the results are completely reversed, with food-disposable incomes declining sharply in 2007 and especially in 2008, during the so-called “tortilla crisis.� In the other countries, the data show much more modest trends and some general declines in 2008 as food prices rose substantially in Bangladesh, Vietnam, and Pakistan. Although there are variations among these nine countries, it is clear that nominal income growth generally outpaced food inflation in most country-year observations by a large margin in the four most populous countries (China, India, Indonesia, and Brazil). These results apply to the largest countries, but we can also experiment with predicting changes in self-assessed food security for the entire sample of developing countries based on our regression results. Specifically, we use the growth and inflation coefficients derived in regressions 2 and 4 in table 7 in conjunction with actual growth and inflation rates to predict changes in self-assessed food insecurity over 2006–08. From these regressions, it is clear that economic growth reduces self-assessed food insecurity, whereas the total inflation rate increases it. For the most part, the results in table 7 suggest that these two effects cancel each other out over 2006–08. In one estimate, global food insecurity decreases by 25 million people, and in the other, it increases by 6 million. In other words, the econometric predictions lead to a conclusion that is qualitatively similar to the raw descriptive statistics: a decrease in global self-assessed food insecurity or little or no change overall. 35 Table 11. Econometrically Predicted Changes in Self-Assessed Food Insecurity over 2006–2008 (Millions of People) Predicted change in food insecurity Predicted change in food insecurity using regression 2 in table 7 using regression 4 in table 7 Total predicted change in self-reported food insecurity −104.5 −152.8 Change due to economic growth 79.4 158.8 Change due to total inflation −25.1 6.0 Source: Author’s calculations based on the GWP (Gallup 2011) Note: Changes in self-reported food insecurity are estimated by multiplying the coefficients from table 7 by changes in GDP per capita and changes in the total CPI from 2006 to 2008. <>V. CONCLUSIONS The innovation of this study is its use of survey-based evidence, rather than simulations, to assess the impact of the food crisis on global and regional food insecurity. We find no evidence that global food insecurity was higher in 2008 than it was in previous years, although it affected many regions, particularly a number of countries in Africa and Latin America. Though qualified by measurement issues, these results are broadly corroborated by recent World Bank estimates of a declining global poverty trend over 2005–08. Our results also cast doubt on the usefulness of simulation approaches in predicting global poverty trends, although the more sophisticated of these approaches are still useful for exploring the mechanisms and distributional impacts of food price impacts in an experimental setting. Finally, our results raise the question of whether self-assessed indicators might be a useful addition to existing food security metrics. Further research is needed in this regard. As we noted in section 2, self-assessed indicators are susceptible to a number of biases. Nevertheless, these 36 weaknesses must be traded off against the fact that such indicators are easily, quickly, and cheaply measured relative to household expenditure or consumption data. There are also some potentially significant ways to improve self-reported data. For example, a more disaggregated ordering of self-assessed food security (such as using scales of 0 to 5) might reduce measurement errors. King et al. (2004) also argue for anchoring vignettes to make measurements more comparable across different socioeconomic groups. Another approach might be to ask households to report the frequency of consumption across different groups rather than asking about more subjective feelings of deprivation. Such dietary diversity or food consumption scores have been shown to be strong predictors of household calorie consumption and individual anthropometric outcomes (Wiesmann et al. 2009; Arimond and Ruel 2006) and might capture the fact that reducing dietary diversity is a common means of coping with higher staple food prices (Block et al. 2004). Others have argued for the use of sentinel sites to collect higher frequency measurements of food security and nutrition outcomes (Barrett 2010). Regardless of the path that is pursued, there are strong grounds for making a large push to improve the measurement of food security. The global food crisis of 2007–08 revealed some significant deficiencies in our capacity to monitor coping strategies and welfare impacts in an acceptable timeframe. Moreover, if strong economic growth had not been prevalent in substantial parts of the developing world prior to and during the food crisis, the impacts of higher food prices might have been far more disastrous. Indeed, predictions of higher food prices and continued price volatility in the next decade or beyond (Headey and Fan 2010) would seem to justify greater investment and experimentation in food security measurement in the near future. 37 References Anríquez, Gustavo, Silvio Daidone, and Erdgin Mane. 2010. “A Cross-country Inquiry, Gustavo Anríquez, and Silvio Daidone. Rising Food Prices and Undernourishment A Cross- Country Inquiry.� ESA Working Paper No. 10-01. Food and Agriculture Organization, Rome. Arimond, Mary, and Ruel, Marie. 2006. "Dietary Diversity Is Associated with Child Nutritional Status: Evidence from 11 Demographic and Health Surveys." The Journal of Nutrition 134: 2579-85. Arndt, Channing, Rui Benfica, Nelson Maximiano, Antonio M.D. Nucifora, and James T. Thurlow. 2008. "Higher Fuel and Food Prices: Impacts and Responses for Mozambique." Agricultural Economics 39(2008 supplement):497–511. Barrett, Christopher S. 2010. “Measuring Food Insecurity.� Science 327: 825-8. Benitez-Silva, Hugo, Moshe Buchinsky, Hiu Man Chan, Sofia Cheidvasser, and John Rust. 2004. "How Large Is the Bias in Self-Assessed Disability?" Journal of Applied Econometrics 19: 649-70. Bertrand, Marianne, and Sendhil Mullainathan. 2001. "Do People Mean What They Say? Implications for Subjective Survey Data." The American Economic Review 91(2): 67-72. Block, Steven A, Lynnda Kiess, Patrick Webb, Soewarta Kosen, Regina Moench-Pfanner, Martin W. Bloem, and C. Peter Timmer. 2004. “Macro shocks and micro outcomes: child nutrition during Indonesiaʼs crisis.� Economics and Human Biology 2(1): 21-44. Christiaensen, Luc, Lionel Demery, and Jesper Kühl. 2011. "The Role of Agriculture in Poverty Reduction: An Empirical Perspective." Journal of Development Economics 96(2), 239- 254. 38 de Hoyos, Rafael, and Denis Medvedev. 2009. "Poverty Effects of Higher Food Prices: A Global Perspective." Policy Research Working Paper No. 4887. The World Bank, Washington DC. Deaton, Angus. 2006. "Global Patterns of Income and Health: Facts, Interpretations, and Policies." NBER Working Paper No.12735. National Bureau of Economic Research, Cambridge, MA, 2006. Deaton, Angus. 2010. "Price Indexes, Inequality, and the Measurement of World Poverty." Presidential Address to the American Economic Association. Atlanta, GA. Deaton, Angus. 2011. “The Financial Crisis and the Well-Being of America.� NBER Working Paper No. 17128. National Bureau of Economic Research, Cambridge, MA. Deitchler, Megan, Terri Ballard, Anne Swindale, and Jennifer Coates. 2010. “Validation of a Measure of Household Hunger for Cross-Cultural Use.� Food and Nutrition Technical Assistance II Project (FANTA-2) Technical Report. USAID, Washington, DC. Dessus, Sébastien, Santiago Herrera, and Rafael de Hoyos. 2008. "The Impact of Food Inflation on Urban Poverty and Its Monetary Cost: Some Back-of-the-Envelope Calculations." Policy Research Working Paper 4666. The World Bank, Washington DC. FAO. 2002. "Summary of Proceedings." In Measurement and Assessment of Food Depreviation and Undernutrition, xx-xx[[Please include page numbers of book section here without]]. Food and Agriculture Organisation (FAO), Rome. FAO. 2009. "The State of Food Insecurity in the World 2008." Food and Agriculture Organisation (FAO), Rome. FAO. 2012. "The State of Food Insecurity in the World 2012." Food and Agriculture Organisation (FAO), Rome. 39 Ferreira, F., G. Prennushi, and M. Ravallion. 2000. “Protecting the Poor from Macroeconomic Shocks: an Agenda for Action in a Crisis and Beyond.� World Bank, Washington, DC. Gallup. 2010a. "World Wide Research Methodology." Gallup, Inc., Washington, DC. Gallup. 2010b. "Worldwide Research – Country Data Set Details." Gallup, Inc., Washington, DC. Haddad, Lawrence, John Hoddinott, and Harold Alderman. 1996. Intrahousehold Resource Allocation in Developing Countries: Models, Methods and Policy. Baltimore: Johns Hopkins University Press. Headey, Bruce, Ruud Muffels, and Gert G. Wagner. 2010. "Long-Running German Panel Survey Shows That Personal and Economic Choices, Not Just Genes, Matter for Happiness." Proceedings of the National Academy of Sciences Pre-print Online Version, October 4. Headey, Derek, and Shenggen Fan. 2010. “Reflections on the Global Food Crisis: How Did It Happen? How Has It Hurt? And How Can We Prevent the Next One?� International Food Policy Research Institute (IFPRI), Washington, DC. Headey, Derek. 2011a. “Was the Global Food Crisis Really a Crisis? Simulations versus Self- Reporting.� IFPRI Discussion Paper 01087. International Food Policy Research Institute, Washington, DC. Headey, Derek. 2011b. "Turning Economic Growth into Nutrition-Sensitive Growth." Background paper for the “2020 Conference: Leveraging Agriculture for Improving Nutrition and Health,� New Delhi, India. Headey, D., F. Nisrane, I. Worku, M. Dessa, and A. Seyoum Taffesse. 2012. “Urban Wage Behavior during Food Price Hikes: The Case of Ethiopia.� Paper presented at "CSAE 40 Conference 2012: Economic Development in Africa," March 18-21st, 2012, Centre for the Study of African Economies, University of Oxford. IMF. 2011.World Economic Outlook - April 2011. Washington, DC: International Monetary Fund (IMF). Ivanic, Maros, and Will Martin. 2008. "Implications of Higher Global Food Prices for Poverty in Low-Income Countries." Agricultural Economics 39(s1): 405-16. Ivanic, Maros, Will Martin, and Hassan Zaman. 2011. Estimating the Short-Run Poverty Impacts of the 2010–11 Surge in Food Prices. Policy Research Working Paper 5633. The World Bank, Washington, DC. Kahneman, Daniel, and Deaton, Angus. 2010. "High Income Improves Evaluation of Life but Not Emotional Well-Being.� Proceedings of the National Academy of Sciences 107(38): 16489-93. King, Gary, Christopher J. L. Murray, Joshua A. Salomon, and Ajay Tandon. 2004. Enhancing the Validity and Cross-Cultural Comparability of Measurement in Survey Research. American Political Science Review 98(1): 191-207. Krueger, Alan B., and David A. Schkade. 2008. "The Reliability of Subjective Well-Being Measures." Journal of Public Economics 92(8-9): 1833-45. Lasco, Christine D., Robert J. Myers, and Richard H. Bernsten. 2008. Dynamics of Rice Prices and Agricultural Wages in the Philippines. Agricultural Economics 38(3): 339-48. Mason, Nicole M., T.S. Jayne, Antony Chapoto, and Cynthia Donovan. 2011. “Putting the 2007/2008 global food crisis in longer-term perspective: Trends in staple food affordability in urban Zambia and Kenya.� Food Policy 36(3): 350-67. 41 Ravallion, Martin. 2012. “Poor, or just feeling poor? On using subjective data in measuring poverty.� Policy Research Working Paper Series with number 5698. World Bank, Washington, DC, 2010. Swinnen, Johan, and Pasquamaria Squicciarini. 2012. Mixed Messages on Prices and Food Security. Science 335(January 27): 405-6. Tiwari, Sailesh, and Hassan Zaman. 2010. "The Impact of Economic Shocks on Global Undernourishment." Policy Research Working Paper Series with number 5215. World Bank, Washington, DC. USDA. 2008. "Food Security Assessment, 2007." United States Department of Agriculture (USDA) Economic Research Service (ERS), Washington, DC. USDA. 2009. "Food Security Assessment, 2008-09.� United States Department of Agriculture (USDA) Economic Research Service (ERS), Washington, DC. USDA. 2011. "Ps&D Online Database," United States Department of Agriculture (USDA) Economic Research Service (ERS), Washington, DC. Verpoorten, Marijke, Abhimanyu Arora, Johan Swinnen. 2011. Self-Reported Food Security in Africa During the Food Crisis. LICOS Working Paper. LICOS Centre for Institutions and Economic Performance & Department of Economics, University of Leuven (KUL). Wiesmann, Doris, Lucy Bassett, Todd Benson, and John Hoddinott. 2006. Validation of the World Food Programme’s Food Consumption Score and Alternative Indicators of Household Food Security. IFPRI Discussion Paper 00870. International Food Policy Research Institute (IFPRI), Washington, DC. World Bank. 2010. World Development Indicators Online. Washington, DC: The World Bank. 42 World Bank. 2008a. International Comparison Program Tables of Final Results. Washington, DC: World Bank. World Bank. 2008b. World Development Report 2008: Agriculture for Development. Washington, DC: World Bank. World Bank. 2012. An update to the World Bank’s estimates of consumption poverty in the developing world. Washington, DC: The World Bank. 43 SUPPLEMENTAL APPENDICES Contents SUPPLEMENTAL APPENDIX S1. ADDITIONAL STATISTICS AND ECONOMETRIC RESULTS 2 SUPPLEMENTAL APPENDIX S2: RAW GALLUP DATA 6 SUPPLEMENTAL APPENDIX S3: GALLUP DETAILS 52 SUPPLEMENTAL APPENDIX S1. ADDITIONAL STATISTICS AND ECONOMETRIC RESULTS Table S1.1—The full correlation matrix between various indicators of food insecurity, poverty, and hunger GWP GDP per Income Under- GWP FAO US$1/day US$2/day Low BMI, Stunted food capita per capita weight hunger hunger poverty poverty women children insecurity (log) (log) children GWP hunger 1.00*** GWP food insecurity 0.90*** 1.00*** GDP per capita (log) -0.79*** -0.82*** 1.00*** Income per capita (log) -0.67*** -0.61*** 0.93*** 1.00*** FAO hunger 0.58*** 0.49*** -0.59*** -0.61*** 1.00*** US$1/day poverty 0.77*** 0.64*** -0.90*** -0.90*** 0.60*** 1.00*** US$2/day poverty 0.68*** 0.63*** -0.93*** -0.95*** 0.69*** 0.92*** 1.00*** Low BMI, women -0.14 -0.18** -0.57*** -0.65*** 0.37** 0.56*** 0.78*** 1.00*** Underweight children 0.55*** 0.38*** -0.76*** -0.79*** 0.46*** 0.71*** 0.76*** 0.80*** 1.00*** Stunted children 0.48*** 0.33*** -0.73*** -0.76*** 0.45*** 0.68*** 0.72*** 0.63*** 0.90*** 1.00*** Sources: Dependent variables (indicated by GWP) are from the Gallup World Poll (Gallup 2011). Independent variables are sourced as follows: GDP per capita is from World Bank (2010c) World Development Indicators. Poverty and income per capita are from household surveys collated in the World Bank Povcal data bank (2010b). FAO hunger is from Food and Agriculture Organization (FAO 2011b). Low BMI, women, is from the Demographic Health Surveys (Measure DHS 2010), and underweight and stunted children are from the Demographic Health Surveys (Measure DHS 2010) and the World Health Organization (WHO 2010). Notes: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. All variables are measured in 2005 or the nearest available year. Log indicates that variable is expressed in logarithms to account for a nonlinear relationship. Samples vary in size because of the paucity of some of the poverty and malnutrition indicators. Table S1.2—Estimates of changes in self-reported food insecurity with outliers included Regression No. 1 2 3 4 5 6 GDP Total Food GDP Total Food Type of deflator deflator CPI CPI deflator CPI CPI Outliers removed? Yes Yes Yes Yes Yes Yes Fixed Fixed Fixed Regressor OLS OLS OLS effects effects effects Economic growth -0.16** -0.09* -0.09* -0.321*** -0.174* -0.146 Number of observations 291 275 271 291 275 271 Number of countries 120 112 111 120 112 111 R-square 0.02 0.01 0.01 0.047 0.036 0.025 Sources: Dependent variables are from the Gallup World Poll (Gallup 2011). Economic growth data are from the IMF (2010), and food and total CPI data are from the International Labour Organization ILO (2011). Notes: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and # indicates marginal insignificance at the 10% level. The robust regressions are estimated using the rreg command in stata, with default settings. For fixed effects regressions standard errors are adjusted for country clusters. Economic growth is the percent change in GDP per capita between the two years in which the GWP surveys were conducted. 3 Table S1.3—Countries excluded from the “global� estimates and likely impacts of the 2007/08 food crisis on their food insecurity a Assumed Country Self-reported food insecurity data Clues as to impact of global food crisis b impact 2005/06 2006/07 2007/08 2008/09 2009/10 Seven Middle Eastern and North African countries; total population = 230 million Afghanistan 49 38 38 11 points Algeria 22 15 13 All countries are dependent on wheat imports, and GIEWS data often 7 points Iraq 25 12 18 show rising domestic wheat prices, while overall inflation was often high 13 points Egypt 31 23 28 (exceptionally high in Yemen). In many instances self-reported food 8 points Morocco 36 29 insecurity fell from 2008 to 2009, suggesting 2008 might have been a 5 points Tunisia 22 11 9 year of unusually high food insecurity. 11 points Yemen 47 48 10 points Three large African countries; total population = 190 million Ethiopia 24 38 In DRC and Sudan, GIEWS data suggest that many food items 20 points DRC 61 increased in price by 50–100%. In Ethiopia overall inflation peaked at 10 points Sudan 27 38 50 60% in July 2008 but was already high before the global food crisis. 10 points Three medium-sized African countries; total population = 30 million Malawi 76 51 60 GIEWS data suggest rapid increases in maize, bean, and rice prices in 5 points Rwanda 61 43 Rwanda and Malawi, although many poor people produce maize and 5 points beans. Sierra Leone is a large importer of rice; inflation rose to 17% by Sierra Leone 58 63 mid-2008. 10 points Two medium-sized Latin American countries; total population = 33 million Paraguay 40 36 31 In Paraguay there is no strong evidence on food inflation. In Peru, 5 points maize, potato, and wheat prices rose by 50%, but many poor people Peru 50 45 46 produce maize and potatoes. 5 points One large East Asian country; total population = 86 million Philippines 56 64 68 62 Rice prices rose by 50%, and food insecurity trend is upward. 14 points 62.4 million Total estimated change in self-reported food insecurity in all 16 countries people Source: Self-reported food insecurity data are from the Gallup World Poll (Gallup 2011). Notes: a. These clues include an assessment of FAO Global Information and Early Warning System (GIEWS) data (FAO 2010), IMF inflation data (IMF 2011), and trends in the self-reported food insecurity reported in columns 2 through 6. b. This is the assumed change in self-reported food insecurity between 2005/06 and 2007/08. 4 Figure S1.1—Population-weighted estimates of changes in the frequency of self-reported food insecurity problems in sub-Saharan Africa: 2005/06 and 2007/08 Always, 1.5% Always, 2.2% 100% Frequency of having problems affording food Many times, 8.3% Many times, 6.4% 90% 80% Several times, Several times, 23.8% 70% 27.9% in last 12 months 60% Once or twice, 50% Once or twice, 23.8% 24.5% 40% 30% 20% Never, 41.4% Never, 36.7% 10% 0% 2005/06 2007/08 Source: Self-reported food insecurity data are from the Gallup World Poll (Gallup 2011). Notes: This figure reports population weighted changes in self-reported food insecurity for 19 sub-Saharan African countries, excluding South Africa. The question is similar to the main question posed in the text, but asks how often there have been difficulties in affording enough food for you or your family to eat over the last 12 months. 5 SUPPLEMENTAL APPENDIX S2: RAW GALLUP DATA Table S2.1—Self-reported food insecurity and hunger data from the Gallup World Poll World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity AFG Afghanistan December 2008 49 Not available low AFG Afghanistan October 2009 38 22 low AFG Afghanistan April 2010 38 33 low ALB Albania January 2006 23 Not available middle ALB Albania September 2008 30 Not available middle DZA Algeria June 2008 22 Not available middle DZA Algeria September 2009 15 Not available middle DZA Algeria March 2010 13 Not available middle AGO Angola May 2006 63 Not available low AGO Angola September 2008 79 57 low ARG Argentina May 2006 23 11 middle ARG Argentina August 2007 26 11 middle ARG Argentina August 2008 27 Not available middle ARG Argentina August 2009 24 Not available middle ARM Armenia July 2006 47 12 low ARM Armenia July 2007 26 4 low ARM Armenia August 2008 33 8 low ARM Armenia July 2009 47 Not available low AUS Australia December 2005 8 Not available upper AUS Australia April 2007 9 3 upper AUS Australia June 2008 11 4 upper AUS Australia March 2010 10 3 upper AUT Austria April 2006 3 Not available upper AUT Austria April 2008 6 Not available upper AZE Azerbaijan September 2006 37 11 low AZE Azerbaijan December 2007 57 16 low AZE Azerbaijan November 2008 60 15 low AZE Azerbaijan August 2009 60 Not available low BHR Bahrain September 2009 22 Not available upper BHR Bahrain April 2010 21 Not available upper BGD Bangladesh May 2006 25 Not available low BGD Bangladesh May 2007 24 18 low BGD Bangladesh June 2008 27 22 low BGD Bangladesh May 2009 23 17 low BGD Bangladesh April 2010 29 20 low BLR Belarus June 2006 22 4 middle BLR Belarus July 2007 22 4 middle BLR Belarus December 2008 24 4 middle BLR Belarus July 2009 28 Not available middle BEL Belgium July 2005 7 1 upper BEL Belgium May 2007 6 1 upper BEL Belgium June 2008 7 1 upper BLZ Belize October 2007 Not available 22 middle BEN Benin July 2006 66 63 low 6 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity BEN Benin August 2008 64 63 low BOL Bolivia June 2006 41 28 low BOL Bolivia July 2007 39 24 low BOL Bolivia September 2008 42 Not available low BOL Bolivia August 2009 36 Not available low BIH Bosnia-Herzegovina January 2006 Not available 6 middle BIH Bosnia-Herzegovina September 2008 15 6 middle BIH Bosnia-Herzegovina September 2009 Not available 6 middle BWA Botswana May 2006 35 28 middle BWA Botswana July 2008 59 28 middle BRA Brazil November 2005 20 4 middle BRA Brazil August 2007 21 4 middle BRA Brazil October 2008 21 4 middle BRA Brazil September 2009 20 4 middle BGR Bulgaria January 2007 35 10 middle BGR Bulgaria March 2010 Not available 10 middle BFA Burkina Faso June 2006 52 52 low BFA Burkina Faso July 2007 42 40 low BFA Burkina Faso April 2008 56 Not available low BFA Burkina Faso May 2010 66 Not available low BDI Burundi July 2008 74 Not available low BDI Burundi August 2009 67 Not available low KHM Cambodia August 2006 67 20 low KHM Cambodia August 2007 58 34 low KHM Cambodia July 2008 53 35 low KHM Cambodia June 2009 55 12 low KHM Cambodia May 2010 49 15 low CMR Cameroon June 2006 66 65 low CMR Cameroon June 2007 57 59 low CMR Cameroon May 2008 66 Not available low CMR Cameroon April 2009 73 Not available low CMR Cameroon March 2010 75 Not available low CAN Canada December 2005 7 2 upper CAN Canada September 2007 9 2 upper CAN Canada September 2008 7 2 upper CAN Canada August 2009 8 2 upper CAF Central African Rep. November 2007 75 79 low TCD Chad November 2006 72 76 low TCD Chad November 2007 54 59 low TCD Chad November 2008 54 Not available low TCD Chad December 2009 56 Not available low CHL Chile May 2006 27 17 middle CHL Chile August 2007 28 14 middle CHL Chile September 2008 33 Not available middle CHL Chile September 2009 26 Not available middle CHN China October 2006 36 3 low 7 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity CHN China November 2008 16 4 low CHN China September 2009 17 4 low COL Colombia June 2006 32 16 middle COL Colombia July 2007 36 13 middle COL Colombia August 2008 33 Not available middle COL Colombia August 2009 37 Not available middle COM Comoros March 2009 70 Not available low COM Comoros March 2010 65 Not available low COG Congo, D. Rep. September 2008 69 Not available low CRI Costa Rica July 2006 26 7 middle CRI Costa Rica September 2007 27 10 middle CRI Costa Rica September 2008 24 Not available middle CRI Costa Rica August 2009 23 Not available middle CIV Côte d’Ivoire April 2009 53 Not available low HRV Croatia January 2007 10 3 upper HRV Croatia September 2009 17 3 upper CYP Cyprus September 2006 7 4 upper CYP Cyprus May 2009 10 4 upper CZE Czech Republic July 2005 17 2 upper CZE Czech Republic June 2007 13 2 upper CZE Czech Republic January 2009 8 2 upper DNK Denmark July 2005 9 2 upper DNK Denmark May 2007 6 2 upper DNK Denmark April 2008 1 2 upper DNK Denmark December 2009 3 2 upper DJI Djibouti September 2008 44 Not available low DJI Djibouti August 2009 24 Not available low DOM Dominican Republic July 2006 48 36 middle DOM Dominican Republic September 2007 59 37 middle DOM Dominican Republic November 2008 59 Not available middle DOM Dominican Republic September 2009 55 Not available middle ECU Ecuador June 2006 36 26 middle ECU Ecuador July 2007 36 25 middle ECU Ecuador September 2008 46 Not available middle ECU Ecuador September 2009 58 Not available middle EGY Egypt, Arab Rep. September 2005 Not available 23 low EGY Egypt, Arab Rep. July 2007 Not available 23 low EGY Egypt, Arab Rep. May 2008 31 23 low EGY Egypt, Arab Rep. August 2009 23 23 low EGY Egypt, Arab Rep. March 2010 28 23 low SLV El Salvador June 2006 40 25 middle SLV El Salvador September 2007 47 22 middle SLV El Salvador September 2008 48 Not available middle SLV El Salvador July 2009 44 Not available middle EST Estonia July 2006 20 6 upper EST Estonia August 2007 12 3 upper 8 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity EST Estonia July 2008 13 6 upper EST Estonia July 2009 22 Not available upper ETH Ethiopia May 2006 24 22 low ETH Ethiopia July 2007 38 27 low FIN Finland April 2006 5 1 upper FIN Finland April 2008 7 1 upper FRA France July 2005 12 2 upper FRA France December 2006 10 2 upper FRA France June 2008 7 2 upper FRA France May 2009 9 2 upper GEO Georgia February 2006 52 18 low GEO Georgia May 2007 55 18 low GEO Georgia June 2008 51 17 low GEO Georgia May 2009 56 Not available low DEU Germany July 2005 7 2 upper DEU Germany January 2007 7 2 upper DEU Germany October 2008 6 2 upper DEU Germany October 2009 6 2 upper GHA Ghana March 2006 44 39 low GHA Ghana February 2007 41 33 low GHA Ghana April 2008 41 Not available low GHA Ghana July 2009 49 Not available low GRC Greece July 2005 Not available 4 upper GRC Greece May 2007 9 4 upper GRC Greece October 2009 9 4 upper GTM Guatemala June 2006 26 21 low GTM Guatemala September 2007 21 11 low GTM Guatemala September 2008 25 Not available low GTM Guatemala July 2009 27 Not available low GUY Guyana October 2007 Not available 19 low HTI Haiti October 2006 63 73 low HTI Haiti December 2008 60 73 low HND Honduras June 2006 42 29 low HND Honduras September 2007 41 30 low HND Honduras September 2008 48 Not available low HND Honduras July 2009 51 Not available low HUN Hungary July 2005 20 4 upper HUN Hungary May 2007 15 4 upper HUN Hungary January 2009 Not available 4 upper IND India February 2006 35 Not available low IND India May 2007 26 26 low IND India July 2008 22 15 low IND India November 2009 28 18 low IND India June 2010 27 19 low IDN Indonesia July 2006 28 Not available low IDN Indonesia April 2007 25 15 low 9 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity IDN Indonesia March 2008 22 7 low IDN Indonesia May 2009 23 7 low IDN Indonesia April 2010 25 11 low IRQ Iraq June 2008 25 Not available middle IRQ Iraq August 2009 12 Not available middle IRQ Iraq February 2010 18 Not available middle IRL Ireland May 2006 4 1 upper IRL Ireland April 2008 5 1 upper IRL Ireland April 2009 7 1 upper ISR Israel July 2006 14 5 upper ISR Israel August 2007 12 5 upper ISR Israel October 2008 14 5 upper ISR Israel November 2009 15 5 upper ITA Italy July 2005 11 3 upper ITA Italy May 2007 8 3 upper ITA Italy June 2008 16 3 upper ITA Italy May 2009 15 3 upper JAM Jamaica November 2006 Not available 23 middle JPN Japan November 2005 8 Not available upper JPN Japan August 2007 6 2 upper JPN Japan March 2008 6 2 upper JPN Japan August 2009 7 Not available upper JPN Japan June 2010 9 1 upper JOR Jordan September 2005 17 7 low JOR Jordan October 2007 9 7 low JOR Jordan August 2008 12 7 low JOR Jordan October 2009 9 7 low JOR Jordan April 2010 10 7 low KAZ Kazakhstan September 2006 25 8 middle KAZ Kazakhstan December 2007 28 7 middle KAZ Kazakhstan November 2008 26 4 middle KAZ Kazakhstan August 2009 26 Not available middle KEN Kenya April 2006 71 56 low KEN Kenya June 2007 56 52 low KEN Kenya August 2008 67 Not available low KEN Kenya April 2009 64 Not available low KEN Kenya February 2010 57 Not available low KOR Korea, Rep. March 2006 15 6 upper KOR Korea, Rep. May 2007 12 1 upper KOR Korea, Rep. September 2008 17 Not available upper KOR Korea, Rep. September 2009 16 Not available upper KWT Kuwait August 2006 6 7 upper KWT Kuwait August 2009 3 7 upper KWT Kuwait April 2010 9 7 upper KGZ Kyrgyz Republic March 2006 40 12 low KGZ Kyrgyz Republic May 2007 33 10 low 10 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity KGZ Kyrgyz Republic July 2008 34 8 low KGZ Kyrgyz Republic July 2009 32 Not available low LAO Lao PDR July 2006 14 11 low LAO Lao PDR July 2007 21 15 low LAO Lao PDR August 2008 25 13 low LVA Latvia July 2006 16 6 upper LVA Latvia July 2007 18 4 upper LVA Latvia August 2008 14 4 upper LVA Latvia August 2009 23 Not available upper LBN Lebanon September 2005 16 8 middle LBN Lebanon October 2006 16 8 middle LBN Lebanon May 2008 19 8 middle LBN Lebanon March 2009 20 8 middle LBN Lebanon March 2010 18 8 middle LBR Liberia February 2007 81 80 low LBR Liberia May 2008 78 80 low LBY Libya October 2009 14 Not available middle LTU Lithuania July 2006 13 2 upper LTU Lithuania August 2007 10 4 upper LTU Lithuania June 2008 10 3 upper LTU Lithuania August 2009 16 Not available upper MKD Macedonia, FYR September 2008 Not available 7 middle MKD Macedonia, FYR September 2009 Not available 7 middle MDG Madagascar July 2006 58 46 low MDG Madagascar August 2008 66 46 low MWI Malawi October 2006 76 76 low MWI Malawi June 2007 51 45 low MWI Malawi September 2009 60 Not available low MYS Malaysia June 2007 9 3 middle MYS Malaysia September 2008 11 6 middle MYS Malaysia July 2009 20 6 middle MYS Malaysia June 2010 17 3 middle MLI Mali June 2006 60 55 low MLI Mali June 2008 29 55 low MLI Mali October 2009 40 55 low MRT Mauritania September 2006 39 34 low MRT Mauritania August 2007 39 26 low MRT Mauritania July 2008 39 Not available low MRT Mauritania March 2009 40 Not available low MRT Mauritania March 2010 43 Not available low MEX Mexico November 2005 36 19 middle MEX Mexico July 2007 28 19 middle MEX Mexico August 2008 33 19 middle MEX Mexico August 2009 33 19 middle MDA Moldova April 2006 31 10 low MDA Moldova June 2007 35 6 low 11 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity MDA Moldova October 2008 30 5 low MDA Moldova July 2009 34 Not available low MNG Mongolia September 2007 34 12 low MNG Mongolia October 2008 36 13 low MON Montenegro January 2007 21 7 middle MON Montenegro September 2009 22 7 middle MAR Morocco August 2005 36 24 low MAR Morocco December 2007 29 24 low MAR Morocco August 2009 Not available 24 low MAR Morocco March 2010 Not available 24 low MOZ Mozambique May 2006 62 60 low MOZ Mozambique July 2007 46 43 low MOZ Mozambique June 2008 58 Not available low NAM Namibia September 2007 Not available 35 low NPL Nepal June 2006 9 8 low NPL Nepal July 2007 13 13 low NPL Nepal October 2008 10 6 low NPL Nepal July 2009 17 9 low NPL Nepal May 2010 18 10 low NLD Netherlands July 2005 7 1 upper NLD Netherlands May 2007 4 1 upper NLD Netherlands June 2008 4 1 upper NZL New Zealand March 2006 11 4 upper NZL New Zealand February 2007 9 3 upper NZL New Zealand June 2008 13 3 upper NZL New Zealand March 2010 13 6 upper NIC Nicaragua June 2006 Not available 38 low NIC Nicaragua September 2007 51 35 low NIC Nicaragua September 2008 53 Not available low NIC Nicaragua July 2009 49 Not available low NER Niger June 2006 75 74 low NER Niger June 2008 68 Not available low NER Niger June 2009 71 Not available low NGA Nigeria May 2006 58 54 low NGA Nigeria May 2007 55 58 low NGA Nigeria April 2008 55 Not available low NGA Nigeria August 2009 59 Not available low NGA Nigeria April 2010 56 Not available low NOR Norway May 2006 6 3 upper NOR Norway June 2008 5 3 upper PAK Pakistan September 2005 33 Not available low PAK Pakistan June 2007 26 20 low PAK Pakistan June 2008 27 23 low PAK Pakistan May 2009 34 22 low PAK Pakistan May 2010 38 22 low PAN Panama July 2006 30 14 middle 12 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity PAN Panama September 2007 36 13 middle PAN Panama August 2009 33 Not available middle PRY Paraguay May 2006 40 20 low PRY Paraguay July 2007 36 12 low PRY Paraguay August 2009 31 Not available low PER Peru June 2006 50 34 middle PER Peru July 2007 45 30 middle PER Peru August 2009 46 Not available middle PHL Philippines March 2006 56 28 low PHL Philippines August 2007 64 33 low PHL Philippines June 2009 68 35 low PHL Philippines April 2010 62 33 low POL Poland July 2005 29 6 upper POL Poland May 2007 18 6 upper PRT Portugal September 2006 10 2 upper PRT Portugal January 2010 Not available 2 upper PRI Puerto Rico June 2006 Not available 6 upper QAT Qatar March 2009 8 Not available upper ROM Romania July 2005 48 8 middle ROM Romania May 2007 33 8 middle ROM Romania April 2009 40 8 middle RWA Rwanda October 2006 61 61 low RWA Rwanda August 2009 43 61 low SAU Saudi Arabia September 2005 13 9 upper SAU Saudi Arabia March 2009 18 9 upper SEN Senegal May 2006 26 22 low SEN Senegal February 2007 22 21 low SEN Senegal June 2009 43 Not available low SEN Senegal April 2010 49 Not available low SER Serbia January 2007 17 5 middle SER Serbia September 2009 25 5 middle SLE Sierra Leone July 2006 58 67 low SLE Sierra Leone June 2007 63 67 low SGP Singapore March 2006 4 7 upper SGP Singapore May 2007 4 3 upper SGP Singapore February 2008 3 1 upper SGP Singapore June 2009 2 Not available upper SGP Singapore June 2010 2 1 upper SVN Slovenia May 2009 11 1 upper ZAF South Africa March 2006 45 39 middle ZAF South Africa September 2007 48 46 middle ZAF South Africa September 2008 56 Not available middle ZAF South Africa April 2009 55 Not available middle ESP Spain July 2005 11 1 upper ESP Spain April 2007 9 1 upper ESP Spain April 2008 8 1 upper 13 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity ESP Spain April 2009 14 1 upper LKA Sri Lanka March 2006 32 17 low LKA Sri Lanka May 2007 39 12 low LKA Sri Lanka May 2008 48 11 low LKA Sri Lanka June 2009 41 11 low LKA Sri Lanka May 2010 39 15 low SDN Sudan January 2008 27 24 low SDN Sudan March 2009 38 24 low SDN Sudan March 2010 50 24 low SWE Sweden July 2005 7 1 upper SWE Sweden April 2007 7 1 upper SWE Sweden April 2008 7 1 upper SWE Sweden December 2009 5 1 upper CHE Switzerland May 2006 6 1 upper CHE Switzerland December 2009 4 1 upper SYR Syrian Arab Republic August 2008 16 Not available low SYR Syrian Arab Republic March 2009 16 Not available low TJK Tajikistan June 2006 46 16 low TJK Tajikistan November 2007 41 9 low TJK Tajikistan November 2008 31 5 low TJK Tajikistan August 2009 36 Not available low TZA Tanzania March 2006 53 41 low TZA Tanzania June 2007 39 35 low TZA Tanzania July 2008 62 Not available low TZA Tanzania November 2009 60 Not available low THA Thailand July 2006 10 9 middle THA Thailand August 2007 18 14 middle THA Thailand September 2008 18 9 middle THA Thailand November 2009 17 Not available middle TGO Togo August 2006 62 54 low TGO Togo August 2008 67 54 low TTO Trinidad and Tobago November 2006 26 11 upper TTO Trinidad and Tobago October 2008 33 11 upper TUN Tunisia June 2008 22 Not available middle TUN Tunisia August 2009 11 Not available middle TUN Tunisia April 2010 9 Not available middle TUR Turkey August 2005 Not available 11 middle TUR Turkey May 2007 26 11 middle TUR Turkey July 2008 47 11 middle TUR Turkey November 2009 37 11 middle UGA Uganda March 2006 62 56 low UGA Uganda June 2007 48 42 low UGA Uganda July 2008 62 Not available low UGA Uganda June 2009 52 Not available low UGA Uganda March 2010 59 Not available low UKR Ukraine June 2006 29 7 middle 14 Table S2.1—Continued World Self-reported Date of survey Self-reported Bank Country name food Income level completion hunger code insecurity UKR Ukraine July 2007 34 5 middle UKR Ukraine May 2008 27 5 middle UKR Ukraine May 2009 32 Not available middle ARE United Arab Emirates August 2006 6 4 upper ARE United Arab Emirates September 2009 6 4 upper ARE United Arab Emirates April 2010 4 4 upper GBR United Kingdom June 2005 8 3 upper GBR United Kingdom January 2007 11 3 upper GBR United Kingdom June 2008 12 3 upper GBR United Kingdom May 2009 9 3 upper USA United States July 2006 17 3 upper USA United States August 2007 10 3 upper USA United States August 2008 9 3 upper USA United States July 2009 16 3 upper URY Uruguay June 2006 25 10 middle URY Uruguay July 2007 24 10 middle URY Uruguay September 2008 28 Not available middle URY Uruguay August 2009 20 Not available middle UZB Uzbekistan June 2006 37 11 low UZB Uzbekistan July 2008 39 8 low UZB Uzbekistan June 2009 38 Not available low VEN Venezuela November 2005 41 13 middle VEN Venezuela December 2006 25 13 middle VEN Venezuela September 2008 26 13 middle VEN Venezuela August 2009 32 13 middle VNM Vietnam March 2006 27 17 low VNM Vietnam April 2008 17 6 low VNM Vietnam May 2009 25 6 low VNM Vietnam May 2010 25 7 low YEM Yemen, Rep. September 2009 47 Not available low YEM Yemen, Rep. February 2010 48 Not available low ZMB Zambia April 2006 58 53 low ZMB Zambia July 2007 65 67 low ZMB Zambia June 2008 67 Not available low ZMB Zambia November 2009 69 Not available low ZWE Zimbabwe April 2006 72 65 low ZWE Zimbabwe July 2007 71 50 low ZWE Zimbabwe March 2008 79 Not available low ZWE Zimbabwe July 2009 73 Not available low ZWE Zimbabwe March 2010 53 Not available low Source: Gallup World Poll (Gallup 2011). 15 SUPPLEMENTAL APPENDIX S3: GALLUP DETAILS Table S3.1—Gallup World Poll survey details including design effects and margins of error Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Jun 4–Jun 16, Afghanistan 1,000 1.66 4 Face-to-face Dari, Pashto 2009 Sep 20–Oct 12, Afghanistan 1,000 1.68 4 Face-to-face Dari, Pashto 2009 Sep 7–Oct 2, Albania 1,000 1.45 3.7 Face-to-face Albanian 2009 Feb 21–Mar 22, Deep south excluded (25% of the Algeria 1,000 1.27 3.5 Face-to-face Arabic 2009 population). Aug 1–Sep 12, Deep south excluded (25% of the Algeria 1,000 1.24 3.5 Face-to-face Arabic 2009 population). Jul 4–Aug 12, Argentina 1,000 1.36 3.6 Face-to-face Spanish 2009 Jun 10–Jul 7, Armenia 1,000 1.3 3.5 Face-to-face Armenian, Russian 2009 Dec 4–Jan 28, Austria 1,000 1.47 3.8 Telephone German 2010 Nagorno-Karabakh and Jul 29–Aug 16, Azerbaijan 1,000 1.32 3.6 Face-to-face Azeri, territories excluded (10% of the 2009 population). Feb 23–Mar 19, Non-Arabs excluded (25% of the Bahrain 1,051 1.28 3.4 Face-to-face Arabic 2009 population). Aug 17–Sep 15, Non-Arabs excluded (25% of the Bahrain 1,077 1.27 3.3 Face-to-face Arabic 2009 population). Apr 29–May 14, Bangladesh 1,000 1.22 3.4 Face-to-face Bengali 2009 Jun 3–Jul 10, Belarus 1,077 1.29 3.4 Face-to-face Russian 2009 Jul 29–Aug 31, Bolivia 1,000 1.47 3.8 Face-to-face Spanish 2009 Bosnia and Sep 8–Sep 30, Bosnian, Croatian, 1,023 1.81 4.2 Face-to-face Herzegovina 2009 Serbian Aug 11–Sep 1, Brazil 1,031 1.19 3.3 Face-to-face Portuguese 2009 52 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Jan 25–Mar 2, Bulgaria 1,000 1.24 3.4 Face-to-face Bulgarian 2010 Jul 24–Aug 1, Burundi 1,000 1.31 3.5 Face-to-face French, Kirundi 2009 Jun 4–Jun 27, Cambodia 1,000 1.44 3.7 Face-to-face Khmer 2009 Mar 24–Apr 7, Cameroon 1,000 1.71 4.04 Face-to-face French, English, 2009 Aug 7–Aug 25, Yukon, Northwest Territories, Canada 1,011 1.64 4 Face-to-face English, French 2009 and Nunavut excluded. Eastern part of country excluded Nov 20–2–Dec– Chadian Arabic, (20% of the population). Chad 1,000 1.92 4.3 Face-to-face 09 French, Ngambaya Oversampled educated population. Jul 3–Sep 8, Chile 1,009 1.36 3.6 Face-to-face Spanish 2009 Aug 14–Sep 28, Face-to-face and Beijing, Shanghai, Guangzhou China 4,201 1.95 2.1 Chinese 2009 telephone oversampled. Jul 14–Aug 1, Colombia 1,000 1.35 3.6 Face-to-face Spanish 2009 Feb 23–Mar 5, Comoros 1,000 1.44 3.7 Face-to-face French, Comorian 2009 Jul 15–Oct 10, Comoros 1,000 1.5 3.8 Face-to-face French, Comorian 2009 North and South Kivu, Ituri, and Nov 1–Nov 24, French, Lingala, Congo (DRC) 1,000 1.62 3.9 Face-to-face Haut-Uele excluded (20% of the 2009 Kiswahili population). Jul 6–Aug 8, Costa Rica 1,000 1.26 3.5 Face-to-face Spanish 2009 Sep 4–Sep 28, Croatia 1,009 1.07 3.2 Face-to-face Croatian 2009 Apr 23–May 19, Cyprus 502 1.46 5.3 Telephone Greek 2009 Dec 18–Jan 24, Czech Republic 1,077 1.19 3.3 Face-to-face Czech 2009 53 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Dec 7–Dec 22, Denmark 1,000 1.48 3.8 Telephone Danish 2009 Mar 2–Mar 12, Djibouti 1,000 1.89 3.4 Face-to-face French, Afar, Somali 2009 Jul 25–Aug 2, Djibouti 1,000 1.25 3.5 Face-to-face French, Afar, Somali 2009 Jul 21–Sep 2, Dominican Rep. 1,000 1.37 3.6 Face-to-face Spanish 2009 Jul 12–Sep 1, Ecuador 1,000 1.31 3.6 Face-to-face Spanish 2009 Mar 7–Mar 22, Egypt 1,080 1.29 3.4 Face-to-face Arabic 2009 Aug 11–Aug 19, Egypt 1,032 1.28 3.5 Face-to-face Arabic 2009 Jul 4–Jul 17, El Salvador 1,006 1.14 3.3 Face-to-face Spanish 2009 Jun 13–Jul 7, Estonia 607 1.19 4.3 Face-to-face Estonian, Russian 2009 Apr 16–May 18, France 1,000 1.57 3.9 Telephone French 2009 May 2–May 13, Georgian, Russian, South Ossetia and Abkhazia Georgia 1,000 1.26 3.5 Face-to-face 2009 Armenian excluded (7% of the population). Sep 28–Oct 18, Germany 1,000 1.27 3.5 Telephone German 2009 Jul 9–Jul 31, English, Hausa, Ewe, Ghana 1,000 1.52 3.8 Face-to-face 2009 Twi, Dagbani Oct 1–Oct 15, Greece 1,000 1.44 3.7 Face-to-face Greek 2009 Jul 8–Jul 21, Guatemala 1,015 1.18 3.3 Face-to-face Spanish 2009 Jul 11– Jul 25, Honduras 1,002 1.17 3.3 Face-to-face Spanish 2009 Nov 23–Dec 16, Hong Kong 755 1.48 4.3 Telephone Chinese 2009 May 1 – Jun 17, Northeast states and remote islands India 6,000 1.72 1.66 Face-to-face 11 national languages 2010 excluded (<10% of the population). 54 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Oct 1–Nov 30, Northeast states and remote islands India 3,010 2.07 2.6 Face-to-face 11 national languages 2009 excluded (<10% of the population). Apr 18–May 5, Indonesia 1,080 1.41 3.5 Face-to-face Bahasa Indonesia 2009 Apr 4–Apr 24, Indonesia 1,080 1.36 3.5 Face-to-face Bahasa Indonesia 2010 Feb 20–Mar 12, Iraq 1,000 1.43 3.7 Face-to-face Arabic 2009 Aug 10–Aug 20, Iraq 1,000 1.41 3.6 Face-to-face Arabic, Kurdish 2009 Feb 17–Feb 27, Iraq 1,000 1.33 3.6 Face-to-face Arabic, Kurdish 2010 Apr 17–Apr 27, Ireland 500 1.55 5.5 Telephone English 2009 Oct 11–Nov 5, Israel 1,000 1.27 3.5 Face-to-face Arabic, Hebrew 2009 Apr 21–May 6, Italy 1,005 1.71 4 Telephone Italian 2009 Apr 4–Apr 15, Ivory Coast 1,000 1.26 3.5 Face-to-face Dioula, French 2009 Jul 31–Aug 31, Japan 1,000 1.7 4 Telephone Japanese 2009 June 5 – Jun Japan 1,000 1.37 3.6 Telephone Japanese 24, 2010 Mar 18–Apr 2, Jordan 1,015 1.19 3.4 Face-to-face Arabic 2009 Sep 23–Oct 10, Jordan 1,001 1.23 3.4 Face-to-face Arabic 2009 Mar 20–Apr 9, Jordan 1,000 1.29 3.5 Face-to-face Arabic 2010 Jul 2–Aug 6, Kazakhstan 1,000 1.3 3.5 Face-to-face Kazakh, Russian 2009 Feb 5–Feb 17, Kenya 1,000 1.51 3.8 Face-to-face English, Kishwahili 2010 55 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Mar 30–Apr 10, Kenya 1,000 1.42 3.7 Face-to-face English, Kiswahili 2009 Sep 8–Sep 24, Albanian, Serbian, Kosovo 1,000 1.82 4.2 Face-to-face 2009 Montenegrin Feb 23–Mar 18, Non-Arabs excluded (20% of the Kuwait 1,000 1.23 3.4 Face-to-face Arabic 2009 population). Aug 10–Aug 30, Non-Arabs excluded (20% of the Kuwait 1,000 1.15 3.3 Face-to-face Arabic 2009 population). Apr 8–Apr 17, Non-Arabs excluded (20% of the Kuwait 1,000 1.25 3.5 Face-to-face Arabic 2010 population). Jun 13–Jul 10, Kyrgyz, Russian, Kyrgyzstan 1,000 1.55 3.9 Face-to-face 2009 Uzbek Aug 15–Aug 24, Latvia 515 1.19 4.7 Face-to-face Latvian, Russian 2009 Feb 18–Mar 20, Lebanon 1,002 1.23 3.4 Face-to-face Arabic 2009 Aug 2–Aug 30, Lebanon 1,008 1.28 3.5 Face-to-face Arabic 2009 Feb 3–Mar 25, Lebanon 1,008 1.61 3.9 Face-to-face Arabic 2010 Sample includes only Tripoli, Aug 17–Oct 19, Benghazi, and Al Kufra (50% of Libya 1,000 1.59 3.9 Face-to-face Arabic, English 2009 population). Sample skews male and employed. Sample includes only Tripoli, Feb 20–Mar 18, Benghazi, and Al Kufra (50% of Libya 1,000 1.18 3.4 Face-to-face Arabic 2010 population). Sample skews male and employed. Jul 24–Aug 10, Lithuania 500 1.46 5.3 Face-to-face Lithuanian 2009 Sep 10–Sep 22, Albanian, Bosnian, Macedonia 1,008 1.34 3.6 Face-to-face 2009 Macedonian 56 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Sep 5–Sep 17, Chichewa, English, Malawi 1,000 1.47 3.8 Face-to-face 2009 Tumbuka Jun 12–Jul 26, Bahasa Malay, Malaysia 1,011 2.04 4.4 Face-to-face 2009 Chinese, English May 15 – Bahasa Malay, Malaysia 1000 1.34 3.6 Face-to-face Jun17, 2010 Chinese, English Oct 15–Oct 30, Mali 1,000 1.31 3.6 Face-to-face Bambara, French 2009 Arabic, French, Feb 20–Mar 1, Mauritania 1,000 1.43 3.7 Face-to-face Poulaar, Wolof, 2009 Soninke Arabic, French, Jul 25–Sep 26, Mauritania 984 1.75 4.1 Face-to-face Poulaar, Wolof, 2009 Soninke Arabic, French, Feb 28–Mar 11, Tiris and Adrar excluded (5% of Mauritania 1,000 1.52 3.8 Face-to-face Poulaar, Wolof, 2010 the population). Soninke Jul 21–Aug 5, Mexico 1,000 1.35 3.6 Face-to-face Spanish 2009 Transnistria (Prednestrovie) Jun 12–Jul 4, Romanian/ Moldovan, Moldova 1,000 1.34 3.3 Face-to-face excluded (13% of the 2009 Russian population). Sep 6–Sep 21, Albanian, Bosnian, Montenegro 1,003 2.1 4.5 Face-to-face 2009 Montenegrin, Serbian Feb 26–Mar 18, Morocco 1,000 1.21 3.4 Face-to-face Arabic, French 2009 Aug 7–Aug 24, Morocco 1,031 1.41 3.6 Face-to-face Arabic, French 2009 Feb 18–Mar 23, Morocco 1,002 1.26 3.5 Face-to-face Arabic and French 2010 Apr 4–May 4, Nepal 1,000 1.65 4 Face-to-face Nepali 2010 57 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Jun 19–Jul 25, Nepal 1,002 1.37 3.6 Face-to-face Nepali 2009 Feb 11–Mar 10, New Zealand 750 1.38 4.2 Telephone English 2010 Jul 4–Jul 23, Nicaragua 1,012 1.16 3.3 Face-to-face Spanish 2009 Jun 19–Jun 28, French, Zarma, Agadez region excluded (5% of Niger 1,000 1.29 3.5 Face-to-face 2009 Haussa the population). Jul 15–Aug 6, English, Yoruba, Nigeria 1,000 1.35 3.6 Face-to-face 2009 Hausa, Igbo Mar 19–Apr 4, (Pidgin) English, Nigeria 1,000 1.32 3.5 Face-to-face 2010 Hausa, Igbo, Yoruba May 5 – May FATA/FANA excluded (5% of the Pakistan 1,030 1.51 3.7 Face-to-face Urdu 25, 2010 population). May 1–May 17, FATA/FANA excluded (5% of the Pakistan 842 1.41 4 Face-to-face Urdu 2009 population). Urban oversampled. May 1–Jun 30, FATA/FANA excluded (5% of the Pakistan 1,133 1.57 3.7 Face-to-face Urdu 2009 population). Nov 14–Dec 7, FATA/FANA excluded (5% of the Pakistan 1,147 1.56 3.6 Face-to-face Urdu 2009 population). Feb 13–Feb 23, Palestine 1,014 1.44 3.7 Face-to-face Arabic 2009 Aug 3–Aug 17, Palestine 1,000 1.42 3.7 Face-to-face Arabic 2009 Feb 4–Feb 20, Palestine 1,000 1.5 3.8 Face-to-face Arabic 2010 Jul 9–Aug 3, Panama 1,018 1.19 3.4 Face-to-face Spanish 2009 Jul 6–Aug 26, Paraguay 1,000 1.33 3.6 Face-to-face Spanish 2009 Jul 25–Aug 17, Peru 1,000 1.59 3.9 Face-to-face Spanish 2009 Apr 9–Apr 15, Philippines 1,000 1.41 3.7 Face-to-face 7 national languages 2010 58 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Jun 4–Jun 10, Philippines 1,000 1.6 3.9 Face-to-face 7 national languages 2009 Dec 12, 2009– Poland 1,000 1.3 3.5 Face-to-face Polish Jan 16, 2010 Dec 5, 2009– Portugal 1,000 1.39 3.7 Telephone Portuguese Jan 5, 2010 Mar 11–Mar 25, Non-Arabs excluded (50% of the Qatar 1,016 1.44 3.69 Face-to-face Arabic 2009 population) Mar 3–Apr 5, Romania 1,000 1.46 3.75 Face-to-face Romanian 2009 Apr 2–Jun 14, Russia 2,042 1.65 2.8 Face-to-face Russian Urban oversampled. 2009 April 29 – Jun Russia 2,000 1.62 2.8 Face-to-face Russian 16, 2010 Aug 10–Aug 18, French, Rwanda 1,000 1.55 3.9 Face-to-face 2009 Kinyarwandan Feb 17–Mar 20, Non-Arabs excluded (20% of the Saudi Arabia 1,031 1.23 3.39 Face-to-face Arabic 2009 population). Aug 1–Aug 21, Non-Arabs excluded (20% of the Saudi Arabia 1,021 1.41 3.6 Face-to-face Arabic 2009 population). Apr 5–Apr 15, Senegal 1,000 1.66 4 Face-to-face French, Wolof 2010 May 23–Jun 1, Senegal 1,000 2.42 4.8 Face-to-face French, Wolof 2009 Sep 4–Sep 17, Serbia 1,008 1.24 3.4 Face-to-face Montenegrin, Serbian 2009 May 15 – Jun 9, Singapore 1,001 1.42 3.7 Face-to-face Chinese, English 2010 May 30–Jun 18, Chinese, English, Singapore 1,005 1.41 3.7 Face-to-face 2009 Bahasa Malay Apr 16–May 5, Slovenia 500 1.67 5.7 Telephone Slovene 2009 Mar 6–Mar 17, Somaliland 1,000 1.21 3.4 Face-to-face Arabic, Somali, Afar 2009 59 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Aug 1–Aug 11, Somaliland 1,000 1.24 3.4 Face-to-face Arabic, Somali, Afar 2009 Feb 27–Mar 11, Somaliland 1,000 1.24 3.4 Face-to-face Somali 2010 Mar 21–Apr 7, Afrikaans, English, South Africa 1,000 1.68 4 Face-to-face 2009 Sotho, Zulu, Xhosa Sep 2–Sep 27, South Korea 1,000 1.29 3.5 Landline Korean 2009 Apr 14–Apr 24, Spain 1,005 1.64 4 Telephone Spanish 2009 April 24 – May Sri Lanka 1030 1.68 4 Face-to-face Sinhalese, Tamil 21, 2010 Northern and Eastern parts of Sri May 16–Jun 8, Sri Lanka 1,000 1.73 4.1 Face-to-face Sinhalese, Tamil Lanka excluded (10% of the 2009 population). Southern and southwestern Mar 2–Mar 12, Sudan 1,000 1.89 4.2 Face-to-face Arabic, English parts, including Darfur excluded 2009 (25% of the population). Southern and southwestern Jul 29–Aug 9, Sudan 1,000 1.74 4.1 Face-to-face Arabic, English parts, including Darfur excluded 2009 (25% of the population). Feb 19–Mar 4, Darfur excluded (15% of the Sudan 1,000 1.74 4.1 Face-to-face Arabic, English 2010 population). Dec 3–Dec 20, Sweden 1,002 1.41 3.7 Telephone Swedish 2009 Dec 2–Dec 18, French, German, Switzerland 1,003 1.29 3.5 Telephone 2009 Italian Feb 20–Mar 16, Syria 1,082 1.29 3.4 Face-to-face Arabic 2009 Aug 10–Sep 30, Syria 1,018 1.29 3.4 Face-to-face Arabic 2009 Mar 3–Apr 30, Syria 1,029 1.27 3.4 Face-to-face Arabic 2010 60 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing Jul 27–Aug 14, Tajikistan 1,000 1.44 3.7 Face-to-face Russian, Tajik 2009 Nov 2–Nov 14, Tanzania 1,000 1.83 4.2 Face-to-face English, Kishwahili 2009 Oct 1–Nov 1, Thailand 1,019 1.5 3.8 Face-to-face Thai 2009 Feb 20–Mar 25, Tunisia 1,008 1.11 3.3 Face-to-face Arabic 2009 Aug 2–Aug 22, Tunisia 1,006 1.15 3.3 Face-to-face Arabic 2009 Feb 3–Apr 27, Tunisia 1,059 1.35 3.5 Face-to-face Arabic 2010 Oct 24–Nov 17, Turkey 999 1.47 3.8 Face-to-face Turkish 2009 Jul 1–Aug 9, Turkmenistan 1,000 1.2 3.4 Face-to-face Turkmen, Russian 2009 Northern region excluded (10% Mar 19–Mar 30, Ateso, English, Uganda 1,000 1.45 3.7 Face-to-face of the population). Educated 2010 Luganda, Runyankole population oversampled. Northern region excluded (10% May 23–Jun 3, English, Luganda, Uganda 1,000 1.58 3.9 Face-to-face of the population). Educated 2009 Ateso, Runyankole population oversampled. May 11–May Ukraine 1,081 1.73 3.9 Telephone Russian, Ukrainian Urban oversampled. 25, 2009 Mar 1–Mar 31, Non-Arabs excluded (50% of the UAE 1,013 1.35 3.5 Face-to-face Arabic 2009 population). Aug 8–Sep 18, Non-Arabs excluded (50% of the UAE 1,041 1.34 3.5 Face-to-face Arabic 2009 population). Feb 21–Apr 20, Non-Arabs excluded (50% of the UAE 1,037 1.35 3.5 Face-to-face Arabic 2010 population). Apr 17–May 6, UK 1,002 1.45 3.7 Telephone English 2009 61 Table S3.1—Continued Collection # of Design Margin of Mode of Country a b Languages Exclusions or oversampling? Dates Interviews Effect Error Interviewing May 5–Jul 8, United States 1,003 1.48 3.8 Telephone English 2009 Aug 1–Aug 30, Uruguay 1,000 1.29 3.5 Face-to-face Spanish 2009 May 20–Jun 8, Uzbekistan 1,000 1.34 3.6 Face-to-face Russian, Uzbek 2009 Jul 22–Aug 12, Venezuela 1,000 1.69 4 Face-to-face Spanish 2009 Apr 11–May 26, Vietnam 1,009 1.6 3.9 Face-to-face Vietnamese 2009 Apr 6–May 11, Vietnam 1,000 1.35 3.6 Face-to-face Vietnamese 2010 Gender-matched sampling used Feb 24–Mar 19, Yemen 1,000 1.51 3.8 Face-to-face Arabic during the final stage of 2009 selection. Gender-matched sampling used Aug 4–Sept 2, Yemen 1,000 1.43 3.7 Face-to-face Arabic during the final stage of 2009 selection. Feb 12–Feb 27, Yemen 1,000 1.57 3.9 Face-to-face Arabic 2010 Nov 8–Nov 19, Bemba, English, Lozi, Educated population Zambia 1,000 1.75 4.1 Face-to-face 2009 Nyanja, Tonga oversampled. Mar 12–Mar 25, English, Ndebele, Zimbabwe 1,000 1.19 3.38 Face-to-face 2010 Shona Source: Gallup (2010a). Notes: a The design effect calculation reflects the weights and does not incorporate the intraclass correlation coefficients. Design effect calculation: n*(sum of squared weights)/[(sum of weights)*(sum of weights)]. b. Margin of error is calculated around a proportion at the 95% confidence level. The maximum margin of error was calculated assuming a reported percentage of 50% and takes into account the design effect. Margin of error calculation: (0.25/N)^0.5*1.96*(DE)^0.5. 62