Stunting Reduction in Sub-Saharan Africa CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 1 Contents Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Executive Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Chapter 1. The Income Elasticity of Stunting Reduction in Sub-Saharan Africa . . . . . 8 Chapter 2. Investing in the Early Years: Nutrition in Africa . . . . . . . . . . . . . . 25 Chapter 3. Benin: An Investment Framework for Nutrition. . . . . . . . . . . . . . . 41 Chapter 4. Cote d’Ivoire: An Investment Framework for Nutrition. . . . . . . . . . . 55 Chapter 5. Ethiopia: An Investment Framework for Nutrition. . . . . . . . . . . . . 69 Chapter 6. Niger: An Investment Framework for Nutrition . . . . . . . . . . . . . . . 83 Chapter 7. Rwanda: An Investment Framework for Nutrition. . . . . . . . . . . . . . 97 Annex 1. Income Elasticity Tables. . . . . . . . . . . . . . . . . . . . . . . . . . 112 Annex 2. Brief Review of Empirical Evidence on the Determinants of Undernutrition. . 120 Annex 3. Levels of Malnutrition and World Bank Investment, by Country . . . . . . . 123 TABLE OF CONTENTS 3 Acknowledgments Task Team Leaders (TTLs) for the analytics are Patrick Eozenou and Meera Shekar. Patrick Eozenou led the work on Chapter 1 and Meera Shekar led Chapters 2 through 7. The team consisted of Jonathan Kweku Akuoku, Julia Dayton Eberwein, Jakub Kakietek, Michelle Mehta, Parendi Mehta, Linda Schultz, and Dylan Walters. For Chapter 1, Patrick Eozenou conceived the study, prepared the first draft of the report, and conducted the econometric analysis; Parendi Mehta constructed the aggregate dataset, prepared descriptive statistics tables, reviewed the empirical literature, and provided inputs to the draft report; and Jakub Kakietek contributed to background section and provided key inputs to the report. For Chapters 2 through 7, Meera Shekar defined the overall parameters for the analyses and the selection of the five high-burden countries, and she provided overall quality-control and guidance. Jakub Kakietek undertook the regional costing and impact analysis for stunting and wasting and wrote parts of that section with Michelle Mehta and Julia Dayton Eberwein. Jonathan Kweku Akuoku prepared the financing and costing analysis for stunting and wasting for the five country studies and drafted most of those sections. Dylan Walters prepared the costing and impact analysis for anemia and breastfeeding for the regional analyses and the five country studies. The background sections for the country studies were drafted by Linda Schultz. Julia Dayton Eber- wein and Michelle Mehta provided technical guidance and overall management of the country studies and the regional brief. The team is grateful to Lisa Saldanha, who provided useful inputs for the Ethiopia country analysis, and Menno Mulder-Sibanda, who provided helpful inputs to the Benin analysis. The report was edited by Hope Steele and designed by Nicole Hamam. Peer reviewers Andrew Dabalen (Practice Manager, GPV07), Lars Sondergaard (Program Leader, EACTF), Emmanuel Skoufias (Lead Economist, GPV04), Omar Arias (Lead Economist, GSPDR), and Erika Lutz (Senior Nutrition Specialist, GHN07) provided invaluable comments that contributed to strengthen the final report. Trina Haque and Lynne Sherburne-Benz provided overall guidance toward this work. The authors are grateful for support from the Bill & Melinda Gates Foundation. 4 in Sub-Saharan Africa Stunting Reduction  Executive Summary Reducing all forms of malnutrition, including stunting, is central to the World Bank Group's twin goals of end- ing extreme poverty and promoting shared prosperity, as well as building resilience and preventing instability. Maternal and child undernutrition is estimated to be responsible for about 45 percent of child mortality and 11 percent of the global disease burden. Conversely, reductions in stunting are estimated to potentially increase overall economic productivity, as measured by GDP per capita, by 4 to 11 percent in Africa and Asia – mak- ing investments in early nutrition one of the most cost-effective development actions to yield permanent and inalienable benefits. Since 2000, progress in stunting reduction has been slower in Africa than in other regions. While both Asia and Latin America and Caribbean have managed to reduce stunting rates by over one third, Africa saw a reduction of only one sixth during the same period. In 2016, over 40 percent of the 159 million stunted children globally were in Africa (UNICEF, WHO, and World Bank, 2016). Accelerating the reduction of stunting in Africa will be key to maximizing the return on investments in early childhood development, in education, and more broadly in policies aimed at fostering and enhancing human capital accumulation and job creation. Investing in the Early Years now in Africa is even more critical to the extent that the region is entering a demographic transition process with an expected increase in the working age population share from 54% in 2010 to a peak of 64% in 2090. Scaling up investments today in effective inter- ventions and policies to reduce stunting will be a necessary condition to harness the potential benefits of the demographic dividend in the region. This report consists of seven chapters. The first chapter focuses on the income elasticity of stunting reduction in Sub-Saharan Africa. Chapters 2 through 7 focus on the potential financing needs and impacts of investing in scaling up stunting reduction interventions in the Africa region as a whole, and in five of the high-burden countries in Africa (Benin, Côte d’Ivoire, Ethiopia, Niger, and Rwanda). While the first chapter offers a broad assessment of the empirical relationship between income and stunting reduction at the aggregate level across countries, the subsequent chapters focus on country specific policy recommendations designed to accelerate progress in stunting reduction. The Income Elasticity of Stunting Reduction in Sub-Saharan Africa The first chapter of the report estimates the income elasticity of stunting reduction and compares elasticities in Sub-Saharan Africa (SSA) to countries in other regions. Using cross-country panel data for 151 countries the income elasticity of stunting reduction is estimated to be, on average, close to −0.44. The statistical dif- ference in income elasticity between SSA countries and non SSA countries is formally tested. Although a 10 percent increase in income is associated with a 5.5 percent reduction in the stunting rate in non-SSA countries, the return of stunting reduction to income is more than 2.5 times lower in SSA countries (−0.2). These results are robust to endogeneity bias, to unobserved country-specific heterogeneity, and to alternative modeling assumptions. Differences in access to water and sanitation account for about half of the gap in the magnitude of income elasticities between non-SSA and SSA countries. Differences in the degree of control of corruption and of government effectiveness between SSA and non-SSA countries account for a sizeable part of the difference in the estimated income elasticity. Projecting these results forward to 2030 implies that the number of stunted children is likely to increase in Sub-Saharan Africa—from about 52 million in 2015 to approximately 60 million in 2030—while the global target endorsed at the 65th World Health Assembly called for a 40 percent reduction in the number of stunted children between 2010 and 2025. These results, together with the existing body of empirical evidence on intervention effectiveness, suggest that scaling up a set of high-impact nutrition-specific interventions is needed to accelerate the pace of reduction and to achieve progress against the United Nations’ 2030 Sustainable Development Goal targets in the region. EXECUTIVE SUMMARY 5 Investing in Nutrition in the Africa Region The report then estimates the expected impacts and financing needs for scaling up a set of high-impact nutri- tion-specific interventions for the Africa region as a whole (Chapter 2) and in five countries (Benin, Côte d’Ivo- ire, Ethiopia, Niger, and Rwanda) in Chapters 3 to 7. In 2012, the World Health Assembly endorsed six global targets for improving nutrition in an effort to boost investments in cost-effective interventions, spearhead better implementation practices and catalyze prog- ress toward decreasing malnutrition. Scaling up a package of high-impact nutrition-specific interventions in Africa to address four of the six global nutrition targets of stunting, anemia, breastfeeding, and wasting would require, on average, an additional $2.7 billion per year over the next 10 years and would provide substantial health and economic benefits: It would prevent nearly 17 million cases of child stunting and 2 million child deaths. The economic benefits generated over the productive lives of beneficiaries would be enormous: the region would gain $67 billion from investments in preventing stunting, $16 billion from preventing anemia, $20 billion from increased breastfeeding, and $13 billion from treating severe wasting (acute malnutrition)1 . Returns on every dollar invested in this set of interventions range from $4 for stunting to $12 for wasting, $13 for anemia, and $18 for investing in exclusive breastfeeding. Without these investments, the numbers of stunted children will continue to rise in Africa, depriving economies of future growth potential. Mobilizing the required resources for nutrition is possible, but it will require the coordinated efforts of African govern- ments, traditional multilateral and bilateral donors, and innovative sources of financing such as the Power of Nutrition. Over the next 10 years, African governments would need to increase their collective average annual expenditure on nutrition by $0.8 billion, an amount equal to about 2.4 percent of the current government expenditure on health. International donors would need to increase average annual allocations to nutrition in Africa by $1.8 billion, an amount equivalent to about 4.3 percent of total official development assistance (ODA), and innovative financing sources would need to leverage these domestic and ODA resources. The Power of Nutrition is already committed to providing such innovative financing in at least two countries in Africa: Tan- zania ($20 million) and Ethiopia ($20 million), with several other African countries in the pipeline. The regional perspective is followed by country studies for five high-burden countries: Benin, Côte d’Ivoire, Ethiopia, Niger, and Rwanda. Four of these countries (the exception is Benin) are in the first wave countries for the Investing in the Early Years (IEY) initiative. They have stunting rates ranging from 40 percent in Niger to 38 percent in Rwanda and Ethiopia and 30 percent in Côte d’Ivoire. Benin carries a high burden as well, with child stunting rates at 34 percent. Each country study estimates the financing needs and the health, nutrition, and economic impacts of scaling up a set of high-impact nutrition-specific interventions to meet the global nutrition targets. The cost-effectiveness of both the package of interventions as a whole and for each interven- tion individually is assessed. Finally, two lower-cost financing packages are estimated for each country. In an environment of constrained resources, one of these alternative financing packages could be a strong first invest- ment, but it would need to be followed by increased investments, along with investments in strengthening the national platforms for service delivery, to contribute to meeting the global nutrition targets. 1 All economic gains reported throughout this report are net gains. 6 in Sub-Saharan Africa Stunting Reduction  CHAPTER 1: The Income Elasticity of Stunting Reduction in Sub-Saharan Africa Introduction Stunting—being short for one’s age—is defined as height-for-age ratio z-score (HAZ) lower than 2 standard devi- ations below the World Health Organization (WHO) Growth Standard median (WHO 2009). Stunting is a mani- festation of chronic malnutrition, resulting from an inadequate quantity and quality of food intake and repeated bouts of infection (WHO 2015). It occurs early in life, with the highest risk during a child’s first two years of life, and after that period is largely irreversible (Black et al. 2013). Stunting is associated with weaker immune responses, and stunted children have elevated morbidity and mortality risk. Meta-analytic studies reported that stunted children (HAZ<2) have a two times higher risk and severely stunted children (HAZ<3) a six times higher risk of dying from common childhood infections—including acute respiratory infections, diarrheal disease, and measles—than children who were not stunted (McDonald et al. 2013). It is estimated that malnutrition is an underlying cause of about 45 percent of all deaths in children under age five globally (Black et al. 2008) and that about 14 percent of all deaths of children under five can be attributed directly to stunting (Danaei et al. 2016). The negative consequences of stunting extend beyond increased mortality and morbidity. Stunted physical growth is strongly associated with slower cognitive development, with delays in school enrollment (Fink et al. 2016), and with lower educational attainment (Adair et al. 2013). Furthermore, stunting has long-lasting eco- nomic impact. A recent systematic review showed a 1 centimeter increase in adult height was associated with a median 4 percent increase in wages in men and about a 6 percent increase wages in women, and that the wages of adults who were stunted in childhood were, on average, 20 percent to 40 percent lower than the wages of their peers who were not stunted as children (McGovern et al. 2017). Cohort studies also suggest that adults who were not stunted as children had a 21 percent higher household consumption and a 10 percent lower risk of living in poverty (Hoddinot et al. 2013). The 2013 Lancet series on maternal and child nutrition presented a conceptual framework describing the causes of malnutrition, including stunting, and strategies to improve it. According to the framework, the direct causes of malnutrition include inadequate food intake (inadequate quantity and quality of food) and repeated bouts of disease that compromise nutrient intake and absorption. The more distal (underlying) causes of malnutrition include food security—that is, food availability and food diversity; access to and use of health services; and a safe and hygienic environment. Finally, the basic causes of malnutrition impacting the underlying and direct causes include macro-level economic, social, political, and environmental factors such as national wealth, pov- erty and wealth distribution, ubiquity of armed conflict, climate events, and so forth (Walker et al. 2011). Global Nutrition Targets In 2012, the World Health Assembly (WHA) established a set of six global nutrition targets to address key dimensions of maternal and child malnutrition including stunting. The stunting target has subsequently been incorporated into the Sustainable Development Goals (SDG) framework under SDG 2: “End hunger, achieve food security and improved nutrition and promote sustainable agriculture; and SDG target 2.2: By 2030, end all forms of malnutrition, including achieving, by 2025, the internationally agreed targets on stunting and wasting in children under 5 years of age, and address the nutritional needs of adolescent girls, pregnant and lactating women and older persons.”1 The WHA stunting target calls for reducing the number of children who are stunted worldwide by 40 percent from the 2010 baseline of 171 million to approximately 100 million in 2025. The target is based on an analysis of global trends in the prevalence of stunting among pre-school-age children from 1990 through 2020 (de Onis et al. 2011). In the context of this report it is important to note that the WHA targets, including the stunting target, are set at the global level and are not accompanied by regional or national targets. 8 in Sub-Saharan Africa Stunting Reduction  Stunting in Sub-Saharan Africa Since 1990, the global prevalence of stunting has declined from 39.5 percent to 22.9 percent and the number of stunted children in the world has declined by over 100 million, from 254 million in 1990 to about 155 million in 2015 (UNICEF, WHO, and World Bank 2017). However, progress in reducing stunting has been slower in Africa than in other regions. Among all World Bank regions, the average annual decline in stunting preva- lence has been the lowest in Africa (see Figure 1.1 and Figure 2.1 on page 27). Figure 1.1: Stunting Trends by Region, 1990–2015 60 50 Stunting prevalence (%) 40 30 20 10 1990 1995 2000 2005 2010 2015 SAR Source: UNICEF−WHO−World Bank Joint Malnutrition SSA Estimates, Global Database on Child Growth EAP and Malnutrition (2017) MENA LAC Figure 1.2: Stunting Prevalence and the Number of Stunted Children in Sub-Saharan Africa, 1990–2015 56.6 60 55.2 60 52.9 50.1 47.3 50 45.2 50 Number of stunted children (million) 49% 46% Stunting prevalence (%) 40 43% 40 40% 38% 35% 30 30 20 20 10 10 0 0 1990 1995 2000 2005 2010 2015 Stunting prevalence Number of stunted children Source: UNICEF−WHO−World Bank Joint Malnutrition Estimates, Global Database on Child Growth and Malnutrition (2017) CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 9 Consequently, although the East Asia and Pacific region managed to reduce stunting prevalence by almost two- thirds, Sub-Saharan Africa (SSA) achieved a reduction of only one-quarter during the same period. Further- more, because of high fertility and population growth, the number of stunted children on the continent within that time frame increased by about 12 million (see Figure 1.2). If population over that period of time did not grow, rather than the 57 million stunted children currently living in Africa, there would be only about 32 million stunted African children. This suggests that the high rate of population growth alone is responsible for about 24 million cases of stunting in Africa today. Although there is some variation, in virtually all African countries more than one in five children is stunted, and in nine countries (the Central African Republic, the Democratic Republic of Congo, Eritrea, Ethiopia, Madagascar, Malawi, Mozambique, Niger, and Zambia) stunting prevalence exceeds 40 percent, or two in five children (see Figure 2.3 on page 27). Empirical Evidence Annex 2 provides a brief overview of the existing empirical evidence by looking at the determinants of undernutrition. This section of the report focuses on the role of growth in income per capita to contribute to reducing stunting. Estimating the income elasticity of stunting reduction has important policy implications. A low income elas- ticity of stunting reduction would provide a strong rationale for increasing investments in targeted nutrition interventions to complement broader growth-promoting policies. A substantive body of evidence exists to suggest that a set of nutrition-specific interventions is highly cost-effective (Bhutta et al. 2013). The potential to improve nutritional outcomes through nutrition-sensitive interventions (i.e., interventions in non-health sectors such as agriculture, social safety nets, early child development, water and sanitation, and schooling) also exists, but the body of evidence to date is not as strong as it is for nutrition-specific interventions (Ruel, Alderman, and The Maternal and Child Nutrition Group 2013). Overall, the review of existing evidence shows that income per capita has a significant effect on the nutritional status of children (Alderman et al. 2001; Headey 2012; Heltberg 2009; Smith and Haddad 2000, 2002, 2015). Using all available data on stunting and on constant gross national income (GNI) per capita suggests that there is a significant negative relationship between these two variables (Figure 1.3). Based on existing cross-country analyses, the estimated income elasticities fall in the range of [−1.26; −0.17] depending on the sample used and the modeling approach adopted (see Table A.1 in Annex 1). Since progress in terms of stunting reduction has been weaker in SSA than it has in other world regions, this analysis looks more closely at the income elasticity of stunting reduction in SSA. One of the key contributions of this section of the report is to test empirically whether the relation between income growth and stunting reduction differs significantly in SSA compared to non-SSA countries. Another contribution is that it assesses the extent to which the estimated income elasticity of stunting reduction differs across SSA subregions and across different levels of stunting. A third contribution of this work is to use the estimated income elasticities and project forward the expected changes in stunting at the 2030 horizon to assess how far SSA countries would lie from the global SDG targets on stunting. A wide range of parametric regression models is adopted to assess the robustness of the results to alternative modeling assumption. In particular, the robustness of the results to potential endogeneity bias using instrumental variables methods is assessed, as well as their robust- ness to unobserved country-specific heterogeneity by modeling country fixed and random effects. The dataset is presented in the next section, and the estimation strategy is described in the following one. The subsequent section details the main results, and is followed by a discussion of their implications. 10 in Sub-Saharan Africa Stunting Reduction  Figure 1.3: Stunting and Income per Capita 100 LOWER UPPER LOW INCOME MIDDLE MIDDLE HIGH INCOME INCOME INCOME 80 Stunting rate (% ) 60 40 20 0 270 1,025 4,035 12,475 100,000 GNI per capita (USD, constant) Source: World Development Indicators. Non−SSA SSA The Data The dataset is constructed at the country-year level, and the primary outcome variable of interest is the preva- lence of stunting. Specifically, the stunting rate is the percentage of children under age five whose height-for- age z-score is less than 2 standard deviations below the median of the global reference population of children. Stunting data originating from the Joint Malnutrition Estimates (JME) were prioritized (UNICEF, WHO, and World Bank 2017). This source of data was complemented with data from the World Bank’s World Develop- ment Indicators (WDI), WHO, or from country Demographic and Health Surveys (DHS) when data points were missing in the JME dataset. All countries with available data are included, unlike most of the previous studies that focus on developing countries. Several sources of data were explored that contain relevant variables in the following categories: basic determinants—macro-fiscal, inclusiveness, governance, and conflicts; and underlying determinants—food security and household environment. The key variables used in this analysis are summarized in Annex 1 in Table A.1 and Table A.2. Table A.1 includes a detailed definition as well as each variable’s source. As specified in the table, most variables come from the World Development Indicators. The government health expenditures, governance, conflict, and food availability indicators originate from other databases that are freely available for public use. Variables capturing some key elements of the macroeconomic environment, such as the number of economic recessions (episodes for which annual economic growth is negative), the degree of diversification of the economy, the share of natural rents on GDP, and the degree of trade openness are included. On the policy side are variables related to the share of public spending allocated to social sectors (health and education). Variables related to the degree of inclusiveness—such as the degree of income inequalities (Gini index) and the share of total consumption of the poorest 40 percent of the population—are also included. The female-to- male life expectancy ratio and the proportion of females having completed lower secondary education are also considered. CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 11 Indicators capturing different dimensions of the quality of governance and exposure to violence and conflicts are also factored in the analysis. Finally, variables related to the aggregate food supply are included, as well as those related to the coverage of key interventions such as vitamin A supplementation and improved water and sanitation. Table A.1 shows the sample mean for each variable of interest, as well as the SSA and non-SSA mean. The mean difference between SSA and non-SSA is formally tested for by using two-sample Student tests. Almost all the variables of interest are shown to be statistically different between the two groups. Overall, Table A.2 confirms that income is lower, closer, more concentrated (especially in natural rents and in agriculture), more volatile, and less inclusive in SSA than in other regions in the world. Indicators related to the quality of governance are lower in the region and so are coverage of key interventions and aggregate food supply. Estimation Strategy The objective of this section is fourfold: 1. Estimate the average income elasticity of stunting for all available country points. 2. Estimate the average income elasticity for SSA countries and test whether this elasticity differs from non- SSA countries. 3. Assess whether the income elasticity differs (i) across SSA subregions and (ii) across countries with differ- ent rates of stunting. 4. Assess the extent to which country characteristics contribute to explaining the heterogeneity in estimated income elasticity parameters. To answer these four questions, the following mixed effects 2-level model specification is used:2 Ln(Stuntirt )=α+β1*Ln(Yirt )+β2*Ln(Yirt )*SSA+β3*SSA +β4*Xirt+μt+ui..+uir.+εirt (1) with ui..~N(ui.., γ²) (2) uir.~N(uir., τ²) (3) εirt~N(0,σ²) (4) where Ln(Stuntirt ) is the log of the stunting rate in country i, region r in year t; Ln(Yirt ) is the log of income per capita expressed in the same constant currency unit across countries; the dummy variable SSA marks countries belonging to the Sub-Saharan Africa region; 12 in Sub-Saharan Africa Stunting Reduction  Xirt is a set of control variables capturing key country characteristics such as macroeconomic conditions, governance, conflicts, and underlying nutritional determinants; α, β1, β2, β3, and β4 are fixed parameters to be estimated; a random parameter is included to capture unobserved time-invariant heterogeneity at coun- try level (ui..); unobserved heterogeneity at regional level where country effects are nested in region effects; γ², τ², and σ² are the estimated variance residuals for (respectively) the country effect, the regional effect, and the error term; γ² measures the dispersion of the country specific effects around their country specific mean ui..; and τ² measures the dispersion of the region effects around their respective means uir.. A mixed effects multilevel model is chosen because it is the most flexible specification to address the key questions with the data structure at hand (i.e., repeated longitudinal observations for countries that belong to specific regional groups). The empirical strategy used is to first impose specific constraints on the model parameters and then to relax each of these constraints in sequence by testing their significance. To answer the first question—what is the average income elasticity across all countries?—the following con- straints are first imposed on the model: β2=β3=β4=0. Since the model is specified in log-log, β1 can be directly interpreted as the income elasticity of stunting—that is, the average expected percentage change in the depen- dent stunting rate caused by one percentage change in income. The second question—what is the average income elasticity for SSA countries, and does this elasticity differ from that of non-SSA countries?—is addressed by relaxing the constraint on β2 and β3 and testing whether the resulting income elasticity for the SSA countries differs significantly from the income elasticity estimated for non-SSA countries. The third question—does the income elasticity differ (i) across SSA subregions and (ii) across countries with different rates of stunting?—is addressed similarly, but instead of creating an interaction variable between the log of income and the whole set of SSA countries, the interactions are constructed with respect to different subregions (south, west, central, and east SSA). To assess the extent to which the estimated income elasticities also vary according to the degree of stunting, the same model is estimated using quantile regressions. The focus is on the 25th percentile, the median, and the 75th percentile, which allows for a test of whether countries with a relatively low incidence of stunting (25th percentile) have different income elasticities than countries with a relatively high incidence of stunting (75th percentile). Finally, the fourth question—to what extent do country characteristics contribute to explaining the heterogene- ity in estimated income elasticity parameters?—is addressed by relaxing the constraint on β4 and by allowing other control variables to enter the model. Although the constrained model allows the overall income elastic- ity to be estimated, irrespective of the specific channels through which income causes stunting to change, the unconstrained specification enables an evaluation of how specific characteristics or policy instruments affect stunting, independently of changes in aggregate income. To examine these four questions, the model is first estimated using pooled ordinary least squares (pooled OLS). This is equivalent to imposing a set of distributional assumptions and parameter constraints on Equation (1). More specifically, the pooled OLS model assumes that μt=ui..=uir.=0, or that the constant α is common across all countries. Under this specification, the error term is supposed to be independently and identically distributed, CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 13 and longitudinal observations within countries are supposed to be uncorrelated. Observations related to coun- tries embedded in a similar region are also assumed to be uncorrelated. Next, these restrictions are relaxed by allowing country-specific and time-invariant effects to be modeled to control for country-specific unobserved heterogeneity. This is done by estimating the main specification using a random effects (RE) model and a fixed effects (FE) model. The main distributional assumption that differs between the random effects and the mixed effects is the correlation between the country effects and the residu- als. While the RE model imposes zero correlation between the country effects and the error term, the FE model relaxes this assumption. Another important difference between the FE and the RE models is that the FE model is estimated based exclusively on the within-country (longitudinal) variation in the data. The distributional assumption between the country effects and the error term can be tested empirically using a Hausman test. Finally, the mixed effects (ME) model allows the estimation to be generalized further by allowing both fixed and random effects to be modeled, and by allowing observation to be correlated across countries within com- mon regional boundaries. Results The main variables of interest for the income elasticity estimation are the stunting rates and a measure of income. To measure income, GDP per capita in constant international dollars at 2011 PPP as well as constant GNI per capita are used.3 The average stunting rate in the sample lies at around 30 percent, and it is signifi- cantly higher in SSA (39 percent) than in non-SSA countries (26 percent) (see Table A.2). Pooled OLS-Instrumental Variable Results Results for the pooled OLS and instrumental variable (IV) results are reported in Table A.4. The overall income elasticity is estimated to lie around −0.5. These parameters are estimated first in column (1) and (2). The only difference between these estimates is the fact that column (2) controls for a time trend. It is worth noting that income alone accounts for over 60 percent of the overall variance in stunting. For both variables, the variation in the sample is driven mostly by variation between countries. Within-country variation represents 25 per- cent or less of total variance in stunting and income. The point estimate suggests that, over the long run, a 10 percent change in GDP per capita is associated with a reduction in stunting of about 5 percent, and the effect is statistically significant. An important issue to consider when estimating the income elasticity of stunting is potential reverse causal- ity. Improvements in stunting rates are indeed expected to contribute to stronger economic growth over the long run (Fogel 1994). Endogeneity would translate into biased estimates of the income elasticity parameter. To address this concern, IV estimates are performed by treating income as an endogenous variable. Money and quasi-money (M2) in percentage of GDP, and the capital share in the economy, are used as instrumental variables.4 The underlying identification restriction here is that these variables affect stunting only in so far as they influence GDP per capita. The results are reported in column (3) of Table A.4. The point estimates for the income elasticity drops slightly but remains significant. Statistical tests are conducted to assess (i) the validity and (ii) the relevance of the instruments. The Sargan-Hansen statistic is reported and the null hypothesis that the instruments are valid cannot be rejected. The under-identification and weak identification tests are rejected, suggesting that the instruments are also relevant and cannot be rejected. Finally, the endogeneity of GDP per capita is tested. Two endogeneity tests are reported in Table A.4, and neither can reject the null hypothesis that income per capita is exogenous. The analysis is therefore pursued by assuming that GDP per capita is exogenous. Next, whether the estimated income elasticity differs between SSA and non-SSA countries is tested. This is done by relaxing the hypothesis that the coefficient associated with the interaction term between (log) GDP per capita and (log) stunting (β2) is equal to zero. The income elasticity for non-SSA countries is given by β1 and 14 in Sub-Saharan Africa Stunting Reduction  the income elasticity for SSA countries by β1+β2. The standard errors for the income elasticity in SSA countries is not directly reported by Stata (StataCorp 2015), but can be derived by constructing std(β1+β2 )=√([Var(β1)+- Var(β2)+2*Cov(β1,β2 )]). The coefficient of the interaction term is statistically significant, suggesting that the income elasticity of SSA countries differs from that of non-SSA countries (see column (4) of Table A.4). The point estimate for the income elasticity in non-SSA countries is −0.57. In SSA countries, the elasticity is still negative, but three times lower (−0.19) in magnitude. This implies that a 10 percent increase in per capita GDP is associated with a 5.7 percent decrease in stunting rates in non-SSA countries, but with only a 1.9 percent decrease in SSA. The same logic is followed in the fifth column of Table A.4, but instead of considering the region as a whole, the parameter heterogeneity is assessed at the subregional level. The elasticity for south SSA, west SSA, central SSA, and east SSA is tested to determine whether they differ from that of non-SSA countries. The interaction terms are all statistically significant, suggesting different magnitudes in the income elasticity of stunting between subregions. The region with the lowest response of stunting to changes income is the south SSA region. For the countries in south SSA,5 a 10 percent increase in GDP per capita is associated on average with a reduction in stunting rates of about 0.9 percent. For countries located in east SSA, however, the point estimate of the income elasticity of stunting lies around −0.24, which is still more than twice as low as it is for non-SSA countries. Figure 1.4 summarizes the estimated elasticity parameters across regions and subregions, with their associated 95 percent confidence intervals. Figure 1.4: Estimated Income Elasticity of Stunting (OLS) 0 Income elasticity estimates −.2 −.4 −.6 −.8 SA A A A A A SS SS SS SS SS tS n− al uth st es ntr No Ea W So Ce Point estimate 95% CI Source: Authors calculations from WDI. Fixed, Random, and Mixed Effects Models Next, some important assumptions imposed on the model are relaxed by taking into account the structure of the dataset and allowing more flexibility in the estimation. More specifically, first the common intercept assumption imposed by the pooled OLS model to introduce country-specific effects is relaxed. These coun- try-specific effects allow the model to control for any source of unobserved and time-invariant country heterogeneity. CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 15 First a random effects (RE) model is estimated; this assumes zero correlation between the country-specific effects and the residuals. The main results for the RE model are reported in column (1) of Table A.5. The main coefficients of interest—β1 and β2—both remain statistically significant. The point estimates for the non-SSA income elasticity (−0.55) and for the SSA income elasticity (−0.20) are also both statistically significant and are similar to those obtained with pooled OLS regression. A fixed effects (FE) model is then estimated. This relaxes the zero correlation assumption between the coun- try effects and the residuals. FE models are estimated by taking the first difference of the variables expressed in level. As a result, the estimated elasticities rely on within-country variation only, and can be interpreted as representing short- to medium-term elasticities in comparison to the long-term elasticity estimated in pre- vious models. The non-SSA income elasticity drops to −0.5, and the SSA income elasticity to −0.18 (column (3)). A Hausman test is conducted to assess the validity of the zero correlation assumption imposed under the RE specification. Under the null hypothesis, both the RE and FE estimates are consistent. In that case, the RE estimates are more efficient. Under the alternative hypothesis, however, the RE model delivers inconsistent estimates. The Sargan-Hansen statistic is reported in columns (1) and (2) with their associated p-values. In both cases, the null hypothesis that both models are consistent cannot be rejected. Finally, the empirical specification is generalized further, and the nested structure of the dataset is factored in by estimating an ME model with nested effects (2-level) in which countries are nested into regions (i.e., non-zero correlation is allowed between countries within the same region). The main results are preserved: the point esti- mate for the non-SSA income elasticity is around −0.55 and around −0.20 for SSA countries (column (5)). For all the three panel models considered here, the price to pay for a more flexible set of imposed assumptions is a loss in the precision of estimates. This is visible if the subregional elasticity parameters are considered; these are now not as precisely estimated as they used to be under the pooled-OLS model—see columns (2), (4), and (6). The key result, however—that the income elasticity for SSA countries is negative but substan- tially lower in magnitude than it is in non-SSA countries—remains robust to changes in the underlying model assumptions. Figure 1.5 shows the estimated elasticities with 95 percent confidence interval from the mixed effects (ME) model. Figure 1.5: Estimated Income Elasticity of Stunting (ME Model) .2 Income elasticity estimates 0 −.2 −.4 −.6 A A A A A A SS SS SS SS SS SS n− t al uth st es ntr No Ea W So Ce Point estimate 95% CI Source: Authors calculations from WDI. 16 in Sub-Saharan Africa Stunting Reduction  Quantile Regression Results The analysis of parameter heterogeneity for the income elasticity of stunting is pursued by allowing the estimated parameter to vary with the distribution of stunting. Quantile regression methods are used to ask whether countries with a relatively low incidence of stunting have a different average elasticity than countries with a relatively large incidence of stunting. Quantile regression models are estimated for the 25th percentile of the stunting distribution, its median, and the 75th percentile. The 25th percentile of the stunting distribution regroups countries for which the incidence is lower or equal to 17.4 percent. The median of the distribution is similar to the mean (30.1 percent). Countries with relatively high incidence (75th percentile) are countries for which the rate of stunting is equal to 42.1 percent or more. The results are reported in Table A.6 and a similar logic is followed: first a distinct parameter is estimated for SSA as a whole, and then the region is broken down into subregions. The income elasticity estimates are larger on the lower side of the stunting distribution. For non-SSA, the income elasticity is around −0.66 at the lower part of the stunting distribution (column (1)), and −0.42 at the higher part of the stunting distribution (column (5)). For SSA countries, the elasticity is around −0.22 at the 25th percentile of the stunting distribution, and around −0.16 at the larger part. The median estimates are similar in magnitude to the OLS results (column (3)). Income elasticity estimates for SSA are three times lower than for non-SSA coun- tries for countries with low stunting incidence, and about 2.6 times lower for countries with relatively higher distribution in stunting. As one might expect, disaggregation across moments of the stunting distribution and across SSA subregions yields imprecise estimates, although the point estimates are ranked similarly to the esti- mated OLS elasticities (columns (2), (4), and (6)). Augmented Regression Results Finally, additional control variables are introduced to the main specification to assess the extent to which the difference between the estimated income elasticity in non-SSA countries and in SSA countries can be accounted for by factors related to the different country characteristics that differ significantly between SSA and non-SSA countries (see Table A.2) with indicators that characterize the broad macroeconomic environment, the degree of inclusion in the economy, the share of public spending allocated to the social sectors, the quality of gover- nance, the exposition to violence and conflicts, and to coverage of key interventions. The results for these augmented mixed effects regressions are reported in Table A.7a and in Table A.7b. In each of these tables, column (1) represents the baseline (no additional control) specification to which subsequent estimates are compared. Economic diversification, measured by a Herfindhal index based on the share of GDP accounted for by the agriculture, industry, and service sectors, is associated with lower stunting rates, above and beyond the effect of economic diversification on GDP per capita (column (2), Table A.7a). The share of public spending allocated to health and education is also associated with lower stunting rates, after controlling for income (column (4), Table A.7a). Improved food availability and access to improved sanitation induce lower incidence of stunting, still con- trolling for income differences across countries (columns (7) and (8), Table A.7a). Although some of the additional control variables are associated with changes in stunting rates, the main result remains stable. The income elasticity of stunting is negative, and substantially lower in SSA countries than it is in non-SSA countries. Controlling for access to improved sanitation has the larger effect on the relative magni- tude of the estimated income elasticities. Differences in access to water and sanitation account for about half of the gap in income elasticities between non-SSA and SSA countries. By comparison, differences in the degree of economic diversification account for only 6 percent of the difference in income elasticity estimates. CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 17 Turning now to governance and exposure to violence and conflicts, Table A.7b shows that a lower degree of corruption (column 2), improved government effectiveness (column 3), and a stronger rule of law (column 6) are associated with lower incidence of stunting, independently of the effect of these variables on GDP per capita. An increase in the state fragility index is also associated with higher rates of stunting. Again, controlling for these additional variables does not alter the main result. Differences in the degree of control of corruption and of government effectiveness between SSA and non-SSA countries account for about 20 percent each of the difference in the estimated income elasticity. Discussion Using the estimated income elasticity parameters, the expected changes in stunting induced by changes in per capita income are projected forward, using the International Monetary Fund’s World Economic Outlook (WEO) projections for GDP growth between 2016 and 2022.6 These projections are extended to 2030 by estimating a simple Solow growth model and assuming steady state growth. The United Nations Department of Economic and Social Affairs (UN-DESA) World Population Prospects, based on a medium fertility assumption, are used to derive population projections to 2030. Based on the dataset, the global number of stunted children is estimated to be about 155 million in 2015, 51.9 million of whom (33 percent) are in SSA. Relying on current income and population projections, and applying the estimated income elasticities for SSA and non-SSA countries, the number of stunted children is estimated to decrease in non-SSA countries from 115.8 million in 2010 to about 64.5 million in 2030 (representing a decrease of about 44 percent). For SSA, however, the number of stunted children is estimated to increase slightly to 60.3 million in 2030 (Figure 1.6). If the income elasticity of stunting in SSA was equal to the global average (−0.44), then the projected number of stunted children would still remain slightly higher than in 2015 at 52.5 million. Even in the case where the income elasticity of SSA countries was the same as in non-SSA countries (−0.55), the number of stunted children in SSA would fall only slightly below its 2015 level, to 49.3 million, quite far from the implicit WHA target of 33.1 million in 2025. Figure 1.6: Projections of the Number of Stunted Children (SSA versus non-SSA) 100 90 Stunting population (million) 80 70 60 50 40 30 20 2015 2020 2025 2030 SSA countries (status quo) Non−SSA countries SSA countries (avg. elasticity) SSA countries (Non−SSA elasticity) SSA countries (WHA target) Non−SSA countries (WHA target) Source: Authors calculations from the modeled income elasticities (mixed effects multilevel model). GDP growth projections are taken from the IMF until 2020 and extrapo- lated using a Solow growth model after. Population projections from UN−DESA World Population Prospects. 18 in Sub-Saharan Africa Stunting Reduction  These projected trends would correspond to a reduction in the stunting rate in SSA countries from 34.7 percent in 2015 to 30.7 percent in 2030 under a status quo scenario (25 percent if the income elasticity was the same in SSA countries and in non-SSA countries). Again, this is far from the implicit WHA target, which would cor- respond to a stunting rate of 18.2 percent in 2025 (Figure 1.7). For non-SSA countries however, the projected stunting rate for 2030 would fall close to the implicit WHA target of about 16.9 percent (for 2025). Figure 1.7: Projections of the Stunting Rate, SSA versus non-SSA countries, 2015–30 36 34 32 30 28 Stunting rate (%) 26 24 22 20 18 16 14 12 10 2015 2020 2025 2030 SSA countries (status quo) Non−SSA countries SSA countries (avg. elasticity ) SSA countries (Non−SSA elasticity ) SSA countries (WHA target) Non−SSA countries (WHA target) Source: Authors calculations from the modeled income elasticities (mixed effects multilevel model). GDP growth projections are taken from the IMF until 2020 and extrapo- lated using a Solow growth model after. Population projections from UN−DESA World Population Prospects. Conclusion Income growth is associated with a reduction in stunting. On average, a 10 percent increase in GDP per capita translates into a 4.4 percent decrease in stunting rates. In SSA, however, not only are stunting rates higher and income per capita growth rates lower, but this report shows that the income elasticity of stunting reduction in this region is substantially lower (more than 2.5 times lower) than in non-SSA countries. Although a 10 percent increase in income would be associated with a reduction of about 5.5 percent in the stunting rate in non-SSA countries, the same increase in income would translate into only a 2 percent reduction in stunting in SSA. Moreover, and consistent with the previous result, the income elasticity of stunting is lower in countries where the stunting rate is the highest. These results have important policy consequences. First, income growth alone will clearly not be sufficient to reach the 2025 target for stunting established in 2012 by the 65th World Health Assembly (WHA) and adopted by the global community for the Sustainable Development Goals (SDGs). Based on estimated elasticity param- eters and on current income and population projections, the number of stunted children is expected to increase in SSA countries from 51.9 million in 2015 to 60.3 million in 2030, while the implicit WHA target would call for a reduction to about 33.1 million by 2025. Second, these results highlight the importance of scaling up invest- ments in targeted nutrition-specific interventions to accelerate the pace of reduction in stunting worldwide, and even more so in SSA where current coverage of key effective interventions is lower than it is in other regions of the world. Third, these results suggest that relying on income growth alone would not only trigger efficiency concerns for stunting reduction in SSA, but it would also contribute to increasing inequalities in CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 19 stunting between countries. Therefore, scaling up targeted nutrition-specific interventions is necessary not only because of an efficiency argument, but also with an objective of reducing inequalities between countries. In addition to bringing nutrition-specific interventions to scale in SSA, a set of nutrition-sensitive interventions might also contribute to the global community target of stunting reduction, especially if one considers the important synergies across sectors to reduce stunting rates (Skoufias et al. 2017). Endnotes Note: All dollar amounts are U.S. dollars unless otherwise indicated. 1 Information about the Sustainable Development Goals can be found at https://sustainabledevelopment. un.org/topics/sustainabledevelopmentgoals . 2 See Rabe-Hesketh and Skrondal (2012) for a detailed exposition on multilevel and longitudinal modeling. 3 The results presented in this section of the report are based on GDP per capita. All results were robust to the inclusion of GNI per capita instead of GDP. GDP per capita is used as the main income variable because the implied sample size is larger than if GNI were used. 4 These are similar to the instruments used for income in Smith and Haddad (2015). 5 The countries belonging to southern Africa include Botswana, Lesotho, Namibia, South Africa, Swaziland, and Zimbabwe. 6 See the World Economic Outlook Database, 2017, available at https://www.imf.org/external/pubs/ft/ weo/2017/01/weodata/index.aspx. References Aguayo, V. M., R. Nair, N. Badgaiyan, and V. Krishna. 2016. “Determinants of Stunting and Poor Linear Adair, L. S., C. H. Fall, C. Osmond, A. D. Stein, R. Martorell, M. Ramirez-Zea, H. S. Sachdev, et al. 2013. “Associations of Linear Growth and Relative Weight Gain during Early Life with Adult Health and Human Capital in Coun- tries of Low and Middle Income: Findings from Five Birth Cohort Studies.” The Lancet 382 (9891): 525–34. Alderman, H., H. Hoogeveen, and M. Rossi. 2006. “Reducing Child Malnutrition in Tanzania: Combined Effects of Income Growth and Program Interventions.” Economics & Human Biology 4(1). Alderman, H., S. Appleton, L. Haddad, L. Song, and Y. Yohannes. 2001. “Reducing Child Malnutrition: How Far Does Income Growth Take Us?” Centre for Research in Economic Development and International Trade, University of Nottingham, Nottingham. Bhutta, Z. A., J. K. Das, A. Rizvi, M. F. Gaffey, N. Walker, S. Horton, P. Webb, et al. 2013. “Evidence-based inter- ventions for improvement of maternal and child nutrition: what can be done and at what cost?” The Lancet 382 (9890): 452–77. Black, R. E., C. G. Victoria, S. P. Walker, Z. A. Bhutta, P. Christian, M. de Onis, M. Ezzati, et al. 2013. “Maternal and Child Undernutrition and Overweight in Low-Income and Middle-Income Countries.” The Lancet 382 (9890): 427–51. 20 in Sub-Saharan Africa Stunting Reduction  Black, R. E., L. H. Allen, Z. A. Bhutta, L. E. Caulfield, M. de Onis, M. Ezzati, C. Mathers, J. Rivera, and The Maternal and Child Undernutrition Study Group. 2008. “Maternal and Child Undernutrition: Global and Regional Exposures and Health Consequences.” The Lancet 371 (9608): 243–60. Brown, C., M. Ravallion, and D. van de Walle. 2017. Are Poor Individuals Mainly Found in Poor Households? Evi- dence Using Nutrition Data for Africa. Washington, D.C.: The World Bank, Policy Research Working Paper 8001. Christiaensen, L., and H. Alderman. 2001. Child Malnutrition in Ethiopia: Can Maternal Knowledge Augment the Role of Income? Washington, DC: The World Bank, Africa Region Working Paper 22. Danaei, G., K. Andrews, C. Sudfeld, G. Fink, D. McCoy, E. Peet, A. Sania, M. S. Smith Fawzi, M. Ezzati, and W. W. Fawzi. 2016. “Risk Factors for Childhood Stunting in 137 Developing Countries: A Comparative Risk Assessment Analysis at Global, Regional, and Country Levels.” PLoS Medicine 13 (11): 1–18. de Onis, M., A. Onyango, E. Borghi, A. Siyam, M. Blössner, C. Lutter, and for the WHO Multicentre Growth Reference Study Gro. 2011. “Worldwide implementation of the WHO Child Growth Standards.” Public Health Nutrition 1–8. Delpeuch, F., P. Traissac, Y. Martin-Prevel, J. P. Massamba, and B. Maire. 2000. “Economic Crisis and Malnu- trition: Socioeconomic Determinants of Anthropometric Status of Preschool Children and their Mothers in an African Urban Area.” Public Health Nutrition: 3 (1): 39–47. Fink, G., E. Peet, G. Danaei, K. Andrews, D. C. McCoy, C. R. Sudfeld, M. C. Smith Fawzi, M. Ezzati, and W. W. Fawzi. 2016. “Schooling and Wage Income Losses Due to Early- Childhood Growth Faltering in Developing Countries: National, Regional, and Global Estimates.” The American Journal of Clinical Nutrition 104 (1): 104–12. Fogel, R. W. 1994. “Economic growth, population theory, and physiology: The bearing of long-term processes on the making of economic policy.” American Economic Review 84 (3): 369-395. Frongillo, Jr., E. A., M. de Onis, and K. M. P. Hanson. 1997. “Socioeconomic and Demographic Factors Are Asso- ciated with Worldwide Patterns of Stunting and Wasting of Children.” The Journal of Nutrition 2302–09. Glewwe, P., S. Koch, and B. L. Nguyen. 2004. “Chapter 10: Child Nutrition, Economic Growth, and the Provi- sion of Health Care Services in Vietnam.” Economic Growth, Poverty, and Household Welfare in Vietnam (The World Bank), 351–89. Haddad, L., H. Alderman, S. Appleton, L. Song, and Y. Yohannes. 2002. Reducing Child Undernutrition: How Far Does Income Growth Take Us? Washington, DC: International Food Policy Research Institute, Food Consumption and Nutrition Division, Discussion Paper 137. Headey, D. D. 2012. “Developmental Drivers of nutritional Change: A Cross-Country Analysis.” World Develop- ment 42: 76–88. Heltberg, R. 2002. Does Economic Growth Reduce Child Malnutrition? Studiestr 6, 1455 København K, Denmark: Institute of Economics University of Copenhagen. Heltberg, R. 2009. “Malnutrition, Poverty and Economic Growth.” Health Economics 18 (Supp): S77–S88. Hoddinot, J., H. Alderman, J. R. Behrman, L. Haddad, and S. Horton. 2013. “The Economic Rationale for Invest- ing in Stunting Reduction.” Maternal and Child Nutrition 9 (2): 69–82. CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 21 Kabubo-Mariara, J., G. K. Ndenge, and D. M. Kirii. 2006. “Determinants of Children’s Nutritional Status in Kenya: Evidence from Demographic and Health Surveys.” Conference on Reducing Poverty and Inequality: How Can Africa Be Included? Oxford: Centre for the Study of African Economies (CSAE). Marini, A., and M. Gragnolati. 2003. Malnutrition and Poverty in Guatemala. Washington, DC: The World Bank, Policy Research Working Paper 2967. McDonald, C., I. Olofin, S. Flaxman, W. W. Fawzi, D. Spiegelman, L. E. Caulfield, R. E. Black, M. Ezzati, and G. Danaei. 2013. “The Effect of Multiple Anthropometric Deficits on Child Mortality: Meta-Analysis of Individual Data in 10 Prospective Studies from Developing Countries.” American Journal of Clinical Nutrition 97 (4): 896–901. McGovern, M. E., A. Krishna, V. M. Aguayo, and S. V. Subramanian. 2017. “A review of the evidence linking child stunting to economic outcomes.” International Journal of Epidemiology 1–21. Milman, A., E. A. Frongillo, M. de Onis, and J. Hwang. 2005. “Differential Improvement among Countries in Child Stunting is Associated with Long-Term Development and Specific Interventions.” The Journal of Nutrition 1415–22. Rabe-Hesketh, S., and A. Skrondal. 2012. Multilevel and Longitudinal Modeling Using Stata, Volumes I and II. USA: Stata Corporation. Ruel, M. T., H. Alderman, and The Maternal and Child Nutrition Group. 2013. “Nutrition-Sensitive Interven- tions and Programmes: How Can They Help to Accelerate Progress in Improving Maternal and Child Nutri- tion.” The Lancet 382 (9890): 536–51. Sahn, D. E. 1994. “The Contribution of Income to Improved Nutrition in Côte d’Ivoire.” Journal of African Econo- mies 3 (1): 29–61. Skoufias, E. 2017. “Laying the Groundwork for More Effective Multi-Sectoral Action for the Reduction of Undernutrition in Africa” (forthcoming report). Smith, L. C. and L. Haddad. 2000. “Explaining Child Malnutrition in Developing Countries: A Cross-Country Analysis.” International Food Policy Research Institute. Smith, L. C. and L. Haddad. 2002. “How Potent Is Economic Growth in Reducing Undernutrition?” Economic Development and Cultural Change 51 (1): 55–76. Smith, L. C. and L. Haddad. 2015. “Reducing Child Undernutrition: Past Drivers and Priorities for the Post- MDG Era.” World Development 68: 180–204. StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP. Subramanyam, M. A., I Kawachi, L. F. Berkman, and S. V. Subramanian. 2011. “Is Economic Growth Associated with Reduction in Child Undernutrition in India?” PLoS Medicine 8 (3): e1000424. Tiwari, S., H. Zaman, and J. Saavedra. 2013. “Chapter B: Economic Growth, Poverty, and Nutrition.” Improving Nutrition Through Multisectoral Approaches (The World Bank), 45–57. UNICEF, WHO, and World Bank. 2017. “Joint Child Malnutrition Estimates. Global Database on Child Growth and Malnutrition.” Vollmer, S., K. Harttgen, M. A. Subramanyam, J. Finlay, S. Klasen, and S. V. Subramanian. 2014. “Association between economic growth and early childhood undernutrition: Evidence from 121 Demographic and Health Surveys from 36 low-income and middle-income countries.” The Lancet Global Health, Vol 2 e225–34. 22 in Sub-Saharan Africa Stunting Reduction  Walker, S. P., T. D. Wachs, S. Grantham-McGregor, M. M. Black, S. A. Nelson, S. L. Huffman, H. Baker-Henning- ham, et al. 2011. “Inequality in Early Childhood: Risk and Protective Factors for Early Child Development.” The Lancet 378 (9799): 1325–38. Webb, P., and S. Block. 2012. “Support for Agriculture during Economic Transformation: Impacts on Poverty and Undernutrition.” PNAS (Proceedings of the National Academy of Sciences) 109 (31): 12309–14. WHO. 2009. WHO Child Growth Standards and the Identification of Severe Acute Malnutrition in Infants and Chil- dren: A Joint Statement by the World Health Organization and the United Nations Children’s Fund. Geneva: WHO. WHO. 2015. Global Targets Tracking Tool. Geneva: WHO. CHAPTER 1 The Income Elasticity of Stunting Reduction in Sub-Saharan Africa 23 CHAPTER 2 INVESTING IN THE EARLY YEARS: Nutrition in Africa Benefits of Investing Key Messages in Nutrition in Africa • About 56.6 million children under age five in Africa are stunted. If no action is taken, even if the prevalence of stunting remains unchanged, the number of stunted children on the continent will 16.9 MILLION increase by 12 million over the next decade because of the region’s high fertility rates. cases of stunting prevented in 2025 • Economic growth alone is insufficient to address the challenge of maternal and child malnutrition in Africa. The impact of economic growth on the prevalence of stunting is weaker in Africa than in any other region in the world (a 1 percent increase in gross national income is associated with 0.2 percent reduction in stunting in Africa 1.9 MILLION vs. 0.6 percent in other regions): even among the richest households, child deaths prevented in 2025 as many as one-fifth of children are stunted. • Scaling up a package of high-impact nutrition-specific interventions in Africa to address the global nutrition targets of stunting, anemia, breastfeeding, and wasting would require on average an additional $2.7 billion per year over the next 10 years and would provide 56.9 MILLION enormous benefits (see panel on right), including preventing nearly 17 case-years of anemia in women million cases of child stunting. prevented in 2025 • The economic benefits generated over the productive lives of beneficiaries would be enormous: the region would gain $67 billion from investments in stunting, $16 billion for anemia, $20 billion for breastfeeding, and $13 billion for the treatment of severe wasting. 20.8 MILLION • Mobilizing the required resources for nutrition is possible but babies exclusively breastfed would require the coordinated efforts of African governments, traditional multilateral and bilateral donors, and innovative sources of financing such as the Power of Nutrition. Over the next 10 years, African governments would need to increase their average annual expenditure on nutrition by $0.8 billion, an amount 11.4 MILLION equal to about 2.4 percent of the current government expenditure cases of severe wasting treated on health. International donors would need to increase average annual allocations to nutrition in Africa by $1.8 billion, an amount equivalent to about 4.3 percent of total official development assistance (ODA), and innovative financing sources would need to leverage these domestic and ODA resources. $4–$18 return for every dollar invested • Returns on every dollar invested in this set of interventions range from $4 for stunting to $12 for wasting, $13 for anemia, and $18 for investing in exclusive breastfeeding. $67 BILLION generated from investments to reduce stunting* *The economic benefits are calculated over the productive lives of the children benefiting from the interventions that prevent stunting. CHAPTER 2 Investing in the Early Years: Nutrition in Africa 25 Investment Case for Nutrition Ensuring optimum nutrition—particularly during the 1,000-day period from pregnancy to a child’s second birthday—can alter an individual’s development trajectory and maximize her or his productive potential. Chronic malnutrition has important lifelong consequences for health and cognitive development. Losses to cognitive development in early childhood resulting from chronic malnutrition are irreversible. Being stunted (low height-for-age) in early childhood is associated with a delayed start at school, reduced schooling attain- ment, and substantially decreased adult incomes at both the individual and country level (Daniels and Adair 2004; Fink et al. 2016; Hoddinott et al. 2008; Martorell et al. 2010). These consequences add up to overall gross domestic product (GDP) losses of 4 to 11 percent in Africa and Asia (Horton and Steckel 2013). Importantly, chronic undernutrition can be transmitted through an inter-generational cycle, where malnourished mothers are more likely to have stunted children (Aguayo et al. 2016; Ozaltin et al. 2010). Investments in nutrition are highly cost-effective and among the best value-for-money development actions (Copenhagen Consensus Center 2015; Hoddinott et al. 2013). An Investment Framework for Nutrition developed by the World Bank in partnership with R4D, 1000 Days, and the Bill & Melinda Gates Foundation estimated high returns on every dollar invested in nutrition: from $4 in returns for treating acute malnutrition (wasting) to $11 for preventing stunting, $12 for the treatment and prevention of anemia, and $35 for increasing the prev- alence of exclusive breastfeeding (Shekar et al. 2017). Not only do investments in nutrition produce substantial economic benefits, but they also lay the groundwork for the success of investments in other sectors. Investments in the early years—including early life nutrition in the first 1000 days, early learning and stimula- tion, and nurturing care and protection from stress—ensure that all children reach their human potential and contribute to the economic growth of their nation. The analysis presented below focuses on high-impact nutri- tion-specific interventions with strong evidence of efficacy in reducing malnutrition, and estimates the costs, impact, and economic benefits of scaling up these interventions in Africa. Nutrition in Africa Since 1995, progress in reducing chronic malnutrition and its principal manifestation, stunting, has been slower in Africa than in other regions. Over the past two decades, among all World Bank regions, the Africa region has seen the lowest average annual decline in stunting prevalence (see Figure 2.1). While both the Europe and Cen- tral Asia region and the East Asia and the Pacific region have managed to reduce stunting prevalence by almost two-thirds, Africa achieved a reduction of only one-quarter during the same period. Furthermore, because of high fertility and population growth, the number of stunted children on the continent within that time frame actually increased by about 12 million (Figure 2.2) and this upward trend will likely continue in the future. The analyses show that if population growth had declined to zero during this period, the number of stunted chil- dren could have declined to 32 million by 2015 instead of increasing to more than 56 million. Although there is some variation, in virtually all African countries more than one in five children is stunted and in nine countries stunting prevalence exceeds 40 percent, and in another 14 countries – 30 percent. (see Figure 2.3). Furthermore, national prevalence figures mask marked differences within countries, with stunting prevalence in high-prevalence regions being significantly higher than the national average (see chapters 3 to 7 for specific examples). Household data consistently show that, although stunting prevalence tends to be higher in lower-income quin- tiles, even among the richest households it is very high, often exceeding 20 percent (Figure 2.4). Recent analyses show that the association between economic growth and reduction in stunting prevalence is weaker in Africa, where a 1 percent increase in per capita gross national income (GNI) is associated with a 0.2 percent reduction in stunting prevalence, than it is in other regions, where the same increase in GNI is associated with a 0.6 percent decrease in the prevalence of stunting (Eozenou et al. 2017). In fact, based on these estimates, economic growth would not be enough to offset the impact of rapid population growth; despite increasing the purchasing power 26 in Sub-Saharan Africa Stunting Reduction  Figure 2.1: Reduction Rates by Region, 1990–2015 Figure 2.2: Trends in Stunting Prevalence and the Number of Stunted Children in the Africa Region, 1990-2015 4 3.8 Average annual reduction in stunting prevalence (%) 3.8% 3.8% South Asi 60 52.9 55.2 56.6 50.1 3 2.8 45.2 47.3 STUNTING PREVALENCE (%) Sub-S h r n 50 2.8% Afric 40 49% E st Asi + 2.2 46% 43% 2.2% 40% th P cific 1.9 30 38% 35% 2 1.9% 45.2 42.4 Middl E st + 39.7 20 37.0 34.5 North Afric 31.9 1.3% 1.3 Europ + 10 C ntr l Asi 1 0 L tin Am ric + th C rribb n 1990 1995 2000 2005 2010 2015 Stuntin Pr v l nc Numb r of stunt d childr n 0 Numb r of stunt d childr n if no poul tion rowth EAP LAC MENA SAR SSA Source: UNICEF, WHO and World Bank. 2015 Figure 2.3: Prevalence of Stunting in Sub-Saharan Africa, 2015 CABO VERDE MAURITANIA MALI NIGER SUDAN ERITREA SENEGAL CHAD THE GAMBIA BURKINA FASO DJIBOUTI GUINEA-BISSAU GUINEA BENIN NIGERIA CÔTE ETHIOPIA SIERRA LEONE D’IVOIRE GHANA CENTRAL AFRICAN SOUTH REPUBLIC SUDAN LIBERIA CAMEROON SOMALIA TOGO EQUATORIAL GUINEA UGANDA SÃO TOMÉ AND PRÍNCIPE REP. OF KENYA GABON CONGO RWANDA DEM. REP. BURUNDI OF CONGO TANZANIA SEYCHELLES COMOROS ANGOLA Prevalence of stunting MALAWI ZAMBIA in Sub-Saharan Africa, 2014 <20 ZIMBABWE MOZAMBIQUE MAURITIUS MADAGASCAR 20 – 29.9 NAMIBIA BOTSWANA 30 – 39.9 ≥40 SWAZILAND SOUTH LESOTHO AFRICA Sources: UNICEF, WHO, World Bank, 2015. IBRD 42988 | JUNE 2017 These maps were produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denominations and any other information shown on these maps do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. CHAPTER 2 Investing in the Early Years: Nutrition in Africa 27 IBRD 42988 | JUNE 2017 This map was produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Figure 2.4: Socioeconomic Disparities in Stunting among Select Figure 2.5: Estimated Impact of Economic Growth on the African Countries Number of Stunted Children in Africa, 2015–2025 Ethiopi Ni ri 60 60 PREVALENCE, % PREVALENCE, % 80 NUMBER OF STUNTED CHIDREN (Millions) 40 40 67.4 70 20 20 60 0 0 61.7 50 h d st st h d st st dl rt n dl rt n co co id u r h id u r h o Fo o 40 Fo M M Po Po Hi Hi S S 30 K n DRC 20 60 PREVALENCE, % PREVALENCE, % 60 10 40 40 0 20 20 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 0 0 h d st st h d st st dl rt n Ch n in numb r of Ch n in numb rs of stunt d dl rt n co id u r h co id u r o h Fo M o Fo Po stunt d childr n if no childr n du to GDP rowth lon M Hi S Po Hi S ction is t k n Data source: UNICEF, WHO and World Bank. 2015 and World Bank 2017. of households, the absolute number of stunted children on the continent would continue to rise (Figure 2.5). This low income elasticity is likely due to the ubiquity of the risk factors—such as food insecurity, lack of access to clean water and improved sanitation, the low socioeconomic status of women, and political instability, among others—which attenuate the impact of higher incomes. At the same time, some countries with relatively low income levels, such as Senegal, have been able to achieve dramatic reductions in stunting prevalence, while relatively rich countries, such as Nigeria, have seen only small declines. In sum, the evidence from within-coun- try and cross-country analyses demonstrates that economic growth alone will not be sufficient to substantially reduce stunting in Africa and that direct action and specific interventions are needed. Other aspects of malnutrition in Africa are equally alarming. Wasting (low weight for height, an indicator of acute malnutrition) prevalence is higher than in any other region in the world with the exception of South Asia and, in 2015, about 13 million (7.8 percent) of children under age five on the continent suffered from wasting. In a number of countries, largely in the Sahel region (Burkina Faso, Chad, Eritrea, Mali, Mauritania, Niger, Soma- lia, South Sudan, and Sudan), the prevalence of wasting permanently exceeds the World Health Organization (WHO) public health emergency threshold of 10 percent. Similarly, the prevalence of maternal anemia remains very high—about 40 percent of all pregnant women in Africa are anemic (UNICEF et al., 2015; data from 2011). No recent data exist on the prevalence of other micronutrient deficiencies in Africa. However, the historical data suggest that the prevalence and the associated burden of disease is high. In 2007, about 57.7 million (33 percent) of school-aged children in Africa suffered from iodine deficiency (de Benoits, 2008). In 2005, about 2 percent of all school-aged children in Africa suffered from night blindness as a result of vitamin A deficiency (WHO, 2009). This meant that Africa also had the highest absolute number of children suffering from night blindness – about 2.5 million, roughly half of all children suffering from night blindness worldwide. Similarly, about 3 million (9.8 percent) of pregnant women in Africa were estimated to suffer from night blindness, about a third of the global burden (WHO, 2009). The prevalence of exclusive breastfeeding tends to be higher in Africa than in other regions. In 18 African countries, more than 50 percent of children are exclusively breastfed (Figure 2.6). However, recent analyses suggest that breastfeeding is inversely associated with income and, as incomes rise, fewer women breastfeed their children (Victora et al., 2016). This in turn implies that, in the absence of interventions aimed at promoting breastfeeding, its prevalence in Africa will decline in the coming years, with a detrimental impact on physical and cognitive development of children on the continent. 28 in Sub-Saharan Africa Stunting Reduction  Figure 2.6: Prevalence of Exclusive Breastfeeding in Africa CABO VERDE MAURITANIA MALI NIGER SUDAN ERITREA SENEGAL CHAD THE GAMBIA BURKINA FASO DJIBOUTI GUINEA-BISSAU GUINEA BENIN NIGERIA CÔTE ETHIOPIA SIERRA LEONE D’IVOIRE GHANA CENTRAL AFRICAN SOUTH REPUBLIC SUDAN LIBERIA CAMEROON SOMALIA TOGO EQUATORIAL GUINEA UGANDA SÃO TOMÉ AND PRÍNCIPE REP. OF KENYA GABON CONGO RWANDA DEM. REP. BURUNDI OF CONGO TANZANIA SEYCHELLES COMOROS ANGOLA Prevalence of exclusive breastfeeding MALAWI ZAMBIA in Sub-Saharan Africa, 2016 ≥50 ZIMBABWE MOZAMBIQUE MAURITIUS MADAGASCAR 20 – 49.9 NAMIBIA BOTSWANA 10 – 19.9 <10 SWAZILAND No data SOUTH LESOTHO AFRICA Sources: WHO 2016 IBRD 42989 | JUNE 2017 This map were produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denominations and any other information shown on these maps do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. IBRD 42989 | JUNE 2017 This map was produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Global Targets for Nutrition Substantial improvements in the nutritional status of women and children can be realized if adequate invest- ment is made in a set of evidence-based nutrition-specific interventions that ensure optimum nutrition during the critical 1,000-day window between the start of a woman’s pregnancy and the child’s second birthday (Black et al. 2008, 2013). For women, these include interventions to prevent anemia before and during pregnancy as well as those aimed at improving protein energy intake during pregnancy. Interventions targeted toward children and their mothers aim to improve breastfeeding and complementary feeding practices, micronutrient status of children, and to treat acute malnutrition in children. In 2012—to rally the international community around improving nutrition—the 176 members of the World Health Assembly endorsed the first-ever global nutrition targets, focusing on six areas: stunting, anemia, low birthweight, childhood overweight, breastfeeding, and wasting (Table 2.1).1 These targets aim to boost invest- ments in cost-effective interventions, spearhead better implementation practices, and catalyze progress toward reducing malnutrition. The targets for stunting and wasting are enshrined within the United Nations’ Sustain- able Development Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. CHAPTER 2 Investing in the Early Years: Nutrition in Africa 29 Table 2.1: Four Global Targets for Nutrition 1 STUNTING*  f stunted children under five by 40% Reduce the number o Stuntin 2 ANEMIA  f women of reproductive age with anemia by 50% Reduce the number o An mi 5 BREASTFEEDING Increase the rate of exclusive breastfeeding in the first six months up to at least 50% Exclusiv br stf din 6 WASTING* Reduce and maintain childhood wasting (acute malnutrition) to less than 5% W stin Source: WHO 2014. *Stunting and wasting are included within the United Nations’ Sustainable Development Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. Despite evidence regarding their effectiveness and cost-effectiveness (see the next section), coverage of key nutrition-specific interventions in Africa remains largely inadequate. Although coverage rates are higher for some childhood interventions, they remain well below the levels necessary to progress in reducing malnutrition. Table 2.2 and Figure 2.7 summarize the current coverage of and delivery platforms available for nutrition-spe- cific interventions in Africa. It should be noted that the analysis presented here does not capture some important, high-impact nutrition interventions, including zinc and oral rehydration solution (ORS) for the treatment of diarrhea, iodization of salt for the prevention of iodine deficiency disorders, calcium supplementation in pregnancy for the prevention of pregnancy-related hypertensive disorders, and others. This is because this report focuses on high-impact inter- ventions with effectiveness ratios for reaching the four specific global nutrition targets adopted by the WHA and included among the SDGs (for stunting and wasting). However, it needs to be noted that addressing malnutri- tion in a comprehensive manner will require expanding the coverage of those high-impact interventions as well. Table 2.2: Delivery Platforms for Nutrition-Specific Interventions in Africa INTERVENTION PLATFORMS Antenatal micronutrient supplementation (iron and folic acid only) Health facility and community Complementary feeding education Health facility, community, and communication campaigns Breastfeeding promotion Health facility, community, and communication campaigns Balanced energy protein supplementation for pregnant women Health facility, community, and social safety net programs Intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions Health facility and community Vitamin A supplementation Health facility, community, and food fortification Public provision of complementary foods Health facility, community, and social safety net programs Treatment of severe acute malnutrition Health facility and community Iron and folic acid supplementation for non-pregnant women of reproductive age School, community, health facility, and marketplace Staple food fortification Marketplace Pro-breastfeeding social policies Government policies National breastfeeding promotion campaigns Media 30 in Sub-Saharan Africa Stunting Reduction  Figure 2.7: Coverage of Key Nutrition-Specific Interventions in Africa, Weighted Average Antenatal micronutrient supplementation - (iron and folic acid only) Pro-breastfeeding Complementary social policies feeding education 100 80 60 Staple food Breastfeeding fortification 40 promotion 20 Iron and folic acid Balanced energy supplementation for non-pregnant protein supplementation women of reproductive age for pregnant women Treatment of severe Intermittent presumptive acute malnutrition treatment of malaria in pregnancy in malaria-endemic regions Public provision of Vitamin A complementary foods supplementation Financing Needs, Impacts, and Cost-Effectiveness of Scaling Up High-Impact Nutrition-Specific Interventions Using the methodology detailed in An Investment Framework for Nutrition (Shekar et al. 2017), this brief pres- ents estimates of the resources needed to scale up a package of 12 high-impact nutrition-specific interven- tions in Africa to contribute toward achieving the global nutrition targets for stunting, anemia, breastfeeding, and wasting, along with their estimated nutrition, health, and economic impacts. To scale up the package of key interventions to reach the global nutrition targets, an investment of about $70 billion is needed over the next 10 years, in addition to what low- and middle-income countries currently spend on nutrition. About 39 percent of this total, or an additional $27.4 billion, needs to be invested in Africa. Interventions to reduce stunting will require the most resources, accounting for nearly 80 percent of the total amount required for scale-up. However, some of the stunting interventions would also affect breast- feeding and anemia targets. Figure 2.8 represents the distribution of total cost across interventions to address the four targets. CHAPTER 2 Investing in the Early Years: Nutrition in Africa 31 Figure 2.8: Ten-Year Financing Needs to Scale Up Interventions ($, billions) 1.4 South Asi 2.8 1.5 Sub-S h r n 2.4 4.9 Afric Stuntin E st Asi + 2.3 th P cific W stin 27.3 Middl E st + 16.8 North Afric An mi Europ + Br stf din C ntr l Asi 23.3 L tin Am ric + 16.8 th C rribb n ALL LOW & MIDDLE INCOME COUNTRIES AFRICA Note: For Africa, some costs for anemia, breastfeeding, and stunting are shared across interventions. Combined with investments over the next decade in water and sanitation envisaged under the Water, Sanita- tion and Hygiene (WASH) SDG goals, and with improvements in other underlying determinants of malnutri- tion (such as food availability and diversity and women’s health, education, and empowerment), investing in the nutrition-specific interventions in Africa would reduce the number of children on the continent who would be stunted in 2025 by 30 percent compared to the 2015 baseline (Figure 2.9). In addition, the scale up of this package of high-impact interventions would prevent over 1.9 million child deaths and 56.9 million case-years of anemia in women of reproductive age. It would also allow the treatment of 11.4 cases of severe acute malnu- trition in children and result in 20.8 million of babies being exclusively breastfed. Figure 2.9: Impact of the 10-Year Scale-Up of Nutrition-Specific Interventions, Africa STUNTED CHILDREN UNDER AGE FIVE, MILLIONS 70 Und rl in d t rmin nts of stuntin 60 WASH 17 million f w r childr n childr n stunt d 50 in 2025 Nutrition-sp cific int rv ntions 40 40% r duction in numb r 30 of stunt d childr n b 2025 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 Note: a. Includes food availability and diversity, women’s education, women’s empowerment and health, and water, sanitation and hygiene (WASH). Among the set of proposed interventions, prophylactic zinc supplementation would be the most effective for stunting reduction, preventing more than 4.8 million cases of stunting over 10 years (Table 2.3). However, it needs to be emphasized that this is an expensive intervention that lacks tested delivery platforms, and WHO has not as yet issued global guidelines for scaling it up, so it will most likely not be possible to deliver it at scale. Vitamin A supplementation and educating mothers about correct complementary feeding practices are the most cost-effective interventions for stunting prevention with a cost per case of stunting prevented of $333 and $429, respectively. Vitamin A supplementation is also the most cost-effective intervention from the point 32 in Sub-Saharan Africa Stunting Reduction  Table 2.3: Estimated 10-Year Financing Needs and Cost-Effectiveness of Scaling Up Nutrition-Specific Interventions, Africa TOTAL 10-YEAR COST PER CASE INTERVENTION COST PER DEATH FINANCING NEEDS OF STUNTING (NUTRITION TARGET) AVERTED (US $) (US $M) AVERTED (US $) For pregnant women and mothers of infants Antenatal micronutrient supplementation (stunting, anemia) 1,051 11,815 7,250 Infant and young child nutrition counseling (complementary feeding 2,410 6,681 778 education and breastfeeding promotion combined) Complementary feeding education (stunting) 1,255 13,419 429 Breastfeeding promotion (stunting, breastfeeding) 1,156 4,324 6,491 Balanced energy protein supplementation for pregnant women (stunting) 4,347 45,896 41,123 Intermittent presumptive treatment of malaria in pregnancy in malaria- endemic regions (stunting, anemia) 466 4,981 1,531 For infants and young children Vitamin A supplementation (stunting) 327 3,557 333 Prophylactic zinc supplementation (stunting) 5,767 18,420 1,201 Public provision of complementary food (stunting) 8,958 62,902 2,158 Treatment of severe acute malnutrition (wasting) 2,316 4,047 n.a. For non-pregnant women and general population Iron and folic acid supplementation for non-pregnant women (anemia) 119 n.a 13,527 Staple food fortification (anemia) 94 n.a Pro-breastfeeding social policies (breastfeeding) 30 n.a n.a National breastfeeding promotion campaigns (breastfeeding) 193 n.a n.a TOTAL: 27,442 19,139 1,616 Note: Financing needs and impacts assume a linear scale-up of interventions from current coverage level to 90 percent over five years, then maintained at 90 percent for an additional five years. Unit costs for each intervention were drawn from available unit costs from neighboring countries, global costs, or estimates available in the literature. The estimated costs include an additional 12 percent (11 percent for pro-breastfeeding social policies and promotion campaigns) to account for monitoring, evaluation, capac- ity and policy development that may be necessary to reach full scale-up of the interventions. The Lives Saved Tool (LiST; see LiST 2015) was used to estimate the impact of interventions that target pregnant women and children. The impact of interventions that target the general population or non-pregnant women was estimated using a Microsoft Excel model. It should be noted that the LiST model does not capture potential synergies between specific interventions (e.g. the fact that the impact of behavior change communication interventions may be higher in populations that have access to affordable and diversified foods or in populations with higher levels of educational attainment). Therefore, it is possible that the impact estimates generated using LiST in fact underestimate the true impact of the interventions in some contexts. n.a. = not applicable of view of mortality prevention, with the cost of $3,557 per death prevented. However, its absolute impact is modest because, among other factors, vitamin A supplementation has already been scaled-up in many coun- tries in Africa—the further scale up of vitamin A supplementation to 90 percent coverage would, over 10 years, prevent about an additional 92,000 deaths (about 8 percent of the total mortality reduction from all interven- tions combined), but it is critical to keep the high coverage rates to prevent back-sliding. Breastfeeding pro- motion through counseling of mothers is projected to increase the number of infants exclusively breastfed by 20.8 million, and prevent nearly 270,00 deaths. For preventing maternal anemia, blanket iron/folic acid supple- mentation for all women of reproductive age would be the most cost-effective strategy, at a cost of $2.2 for each case of anemia prevented. However, this intervention, like prophylactic zinc supplementation, currently lacks a delivery platform that would allow to reach all women and therefore cannot be rapidly scaled up. In contrast, fortification of staple foods like wheat and maize is also relatively cost effective, with the cost of $9.9 per case of anemia prevented, but has well established and easily scalable delivery platforms. CHAPTER 2 Investing in the Early Years: Nutrition in Africa 33 Economic Benefits of Investing in Nutrition There is a strong body of evidence that shows high economic returns to investing in nutrition (Alderman et al. 2016; Copenhagen Consensus Center 2015; Hoddinott et al. 2013). Scaling up these proven nutrition-specific interventions can ensure that mothers are healthy and well nourished and that they can provide optimal nutri- tion to their children, that children realize their full physical and cognitive development potential, and that women’s productivity is not hampered by illness, especially anemia (Figure 2.10). Figure 2.10: How Reaching the Global Nutrition Targets Generates Economic Benefits Co nitive Development STUNTING Childr n’s H lth & Le rnin & Educ tion l Nutrition St tus Att inment WASTING Glob l Nutrition T r ts R ch d Adult Productivit EASTFEE BRE EDING W es ANEMIA Econom Wom n’s H lth & (GDP) Nutrition St tus In Africa, scaling up the package of nutrition-specific interventions would produce substantial economic ben- efits over the productive lifetimes of the affected women and children (Figure 2.11). Additional health system cost-savings would also be likely because many of these investments reduce the burden of childhood illnesses such as diarrhea and pneumonia. Figure 2.11: Investments in Africa to Meet the Global Nutrition Targets Have Enormous Economic Returns Total Economic $1 Invested Benefits (Billions)a Yieldsb STUNTING 66.8 $4 $ 20.0 $18 1 BREASTFEEDING $ $ ANEMIA 16.1 $ $13 WASTING 13.0 $12 a. Total economic benefits over 10 years for women and over the productive lives of children who benefit from these interventions, defined as the period between the age of 18 and a “retirement” age - the life expectancy or the age of 65, whichever is lower. b. Benefit calculation assumes a 3 percent discount rate for both financing needs and benefits and a GDP growth rate of 3 percent. 34 in Sub-Saharan Africa Stunting Reduction  Financing the Scale-Up of Nutrition-Specific Interventions in Africa The global investment framework estimates that about $0.7 billion is currently allocated to nutrition-specific interventions in Africa (see D’Alimonte et al. 2017 for detailed methodology). This includes allocations from official development assistance (ODA) ($0.55 billion; 79 percent of the total) and from national governments ($0.15 billion; 21 percent of the total).2 A very small contribution from household purchasing of nutrition com- modities was assumed, largely by nonpregnant women above the poverty line for purchase of iron and folic acid supplementation through private retailers. Although most of these costs are borne by the public sector and donors, some scale-up costs would theoretically be offset by household spending over time (see Shekar et al. 2017 for more details). In order to cover the estimated cost of the scale-up, the annual allocation for nutrition from all sources would need to be increased sevenfold (to about $4.5 billion) by 2025 (Figure 2.12). Mobilizing the required resources for nutrition is possible but would require a coordinated effort of African governments, traditional multilateral and bilateral donors, and innovative sources of financing such as the Power of Nutri- tion initiative.3 Over the next 10 years, African governments would need to increase their average annual expenditure on nutrition by 0.8 billion, an amount equal to about 2.4 percent of current government expen- diture on health. International donors would need to increase average annual allocations to nutrition by $1.8 billion, an amount equivalent to about 4.3 percent of the current ODA contribution and innovative financing sources would need to catalyze these contributions. The investment framework proposes that, given the current constrained fiscal space for nutrition in Africa, the bulk of the investment would initially come from international donors. In subsequent years, ODA for nutrition would be gradually replaced by domestic resources. According to this financing scenario, in 2025 about 57 per- cent of the financing would come from domestic public sources, about 40 percent from ODA, and the remain- ing 3 percent from the innovative financing sources (e.g., the Power of Nutrition initiative and the Global Financing Facility in Support of Every Woman Every Child)4, and from the households. In order to achieve this level of financing, the average government expenditure for nutrition in Africa would need to reach an amount equivalent to about 7 percent of its current public expenditure on health by 2025. Similarly, but by 2021, the ODA allocated for nutrition would need to reach about 6 percent of the current total ODA and then taper to about 3 percent of ODA. The World Bank Group is making significant efforts to accelerate nutrition investments under the umbrella of the Investing in the Early Years agenda across the Africa region. Annex 3 provides a list of World Bank Group on-going and planned investments in nutrition as of September 1, 2017. Figure 2.12: Financing Required to Scale-Up Nutrition-Specific Interventions in Africa, by Source 4.5 4.5 4.3 4.4 4.2 3.6 3.0 $US BILLIONS 2.4 1.8 1.2 2016 2017 2018 2019 2020 2021 2022 2022 2023 2024 B s lin Addition l donor Addition l dom stic Innov tiv sourc s !" Source: Regional analysis prepared for D’Alimonte et al. 2017. Note: Annual average household contributions are small relative to other contributions and, as such, are not pictured. CHAPTER 2 Investing in the Early Years: Nutrition in Africa 35 Two Alternative Investment Packages In an environment of constrained resources in which it will not be possible to raise $27 billion over the next 10 years, two alternative investment packages are laid out for consideration. The Priority Package: The first—the “priority package”—includes interventions that are the most cost-effec- tive; that is, that have the lowest cost per health outcome (e.g., case of stunting averted), and that have well-es- tablished global policy guidelines and delivery platforms. Based on those two criteria, the priority package includes antenatal micronutrient supplementation, infant and young child nutrition counseling, intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions, vitamin A supplementation, treat- ment of severe acute malnutrition, intermittent weekly iron and folic acid supplementation for girls 15–19 years of age attending school, and fortification of wheat and maize flour with iron and folic acid. These interventions would be scaled up to full program coverage in the first five years and maintained at full coverage levels for the last five years. This priority package would require an estimated $721 million annually, on average (see Table 2.4). Table 2.4: Benefits and Cost-Effectiveness by Investment Package, Africa GLOBAL PRIORITY CATALYZING FULL PACKAGE: BENEFIT All interventions needed TARGET PACKAGE PROGRESS PACKAGE to meet targets 7.2 billion over 13.1 billion over 10 years 27.4 billion over 10 years 10 years STUNTING Cases of stunting prevented 5.7 million 9.0 million 16.9 million Case-years of anemia in women ANEMIA prevented 39.6 million 52.5 million 56.9 million BREASTFEEDING Babies breastfed over 10 years 20.8 million 20.8 million 20.8 million Child deaths averted 1.2 million 1.4 million 1.9 million over 10 years ALL TARGETS Cost per death averted 5,865 9,357 19,139 Cost per case of 1,254 1,456 1,616 stunting averted a. Total impacts of proposed intervention package combined with other health and poverty reduction efforts. During the 10 years of scale-up, this package would prevent more than 5.7 million cases of stunting and avert 1.2 million deaths in children under five years of age. It would also prevent more than 39.6 million case-years of anemia in women in 2025 and would result in 20.8 million children under six months of age being exclu- sively breastfed. The Catalyzing Progress Package: The second package—the “catalyzing progress package”—includes scale-up of all interventions in the priority package, plus a phased approach to scaling up public provision of complementary foods, balanced energy protein supplementation, prophylactic zinc supplementation, and weekly iron-folic acid supplementation for women outside of schools. It is assumed that, for the latter set of 36 in Sub-Saharan Africa Stunting Reduction  interventions, during the first five years emphasis will be placed on establishing global guidelines and on operational research to develop effective delivery platforms, or to develop less expensive products or more cost-effective technologies. Costs are approximated as the cost of scaling up this set of interventions from 0 to 10 percent coverage only in the first five years. In the subsequent five years, it is assumed that the coverage expansion of those interventions will accelerate and reach 60 percent by 2025. This package would require, on average, $1.3 billion per year (Table 2.4), a total of $13.1 billion over 10 years. It would prevent 1.4 million deaths and more than 9 million cases of stunting among children under age five, increase the number of exclu- sively breastfed children under six months of age by 20.8 million, and prevent more than 52.5 million case- years of anemia in women in 2025. In comparing the relative cost-effectiveness of the three investment packages, the two alternative packages are more cost-effective in preventing deaths and stunting. However, neither is as effective as the full package in making progress toward achieving the stunting, wasting, and anemia targets. The priority and catalyzing prog- ress packages would prevent 1.2 million and 1.4 million deaths respectively, compared with 1.9 million deaths prevented with the full package over 10 years. Under the full package scenario, almost 17 million cases of childhood stunting would be prevented, compared with 9 million cases under the catalyzing progress scenario and fewer than 6 million cases under the priority package scenario. Furthermore, there would be nearly 28 mil- lion and 9 million more case-years of anemia prevented in women under the priority package and catalyzing progress package, respectively. A Call to Action As the world stands on the cusp of the new Sustainable Development Goals, there is an unprecedented oppor- tunity to save children’s lives, build future human capital and cognitive development, and drive faster eco- nomic growth. Scaling up key nutrition interventions during the critical 1,000 day window of early childhood will pay lifelong dividends, translating to healthier societies and more robust economies. If this window is missed, it is missed for life. The additional financing required to reach the global nutrition targets will require coordinated efforts by all stakeholders and a supportive policy environment. To achieve these targets, Africa will need to increase the funding allocated to nutrition by $2.7 billion annually, in addition to the region’s current spending. Although this amount may seem high, they are much lower than other types of investments with much lower rates of return. For example, the International Monetary Fund (IMF) estimated that in 2015 alone Africa’s governments have spent over $26 billion on fossil fuel subsidies (Coady et al. 2015; Whitley and van der Burg 2015). Accelerating the reduction of stunting in Africa will be essential for maximizing the return on investments in early childhood development, in education, and more broadly in policies aimed at fostering and enhancing human capital accumulation and job creation. Investing in the early years is even more critical because the Africa region is entering a demographic transition with an expected increase in the working-age population from 54 percent in 2010 to 64 percent in 2090. This creates a golden opportunity for the continent to reap the benefits of the demographic dividend—an acceleration in economic growth resulting from changes in the structure of the population. As a fertility and population growth rates drop and the share of the working-age population rises, the dependency ratio will diminish with a potential to raise per capita output, savings, and investments in human capital. As mentioned earlier, past high fertility rates in Africa mean that over the last decade, stunting prevalence on the continent has decreased from 49 percent to 35 percent, but despite this, the number of stunted children has increased. If population over that period of time did not grow, rather than nearly 57 million stunted children living Africa currently, there would only be about 32 million stunted children. This suggests that the high rate of population growth alone is responsible for about 24 million cases of stunting in Africa today. Some of the factors that precipitate the demographic transition – access to family planning, later marriage, older age at first birth, longer intervals between births, will contribute to the reduction in child malnutrition (these constitute CHAPTER 2 Investing in the Early Years: Nutrition in Africa 37 some of the so-called nutrition-sensitive interventions). At the same time, however, in order for Africa to reap the benefits of the demographic dividend, it will be imperative to dramatically intensify more targeted efforts to reduce child malnutrition and stunting. The scale-up of the key nutrition-specific interventions is estimated to generate considerable returns in economic benefits over the productive lives of beneficiaries, and is a nec- essary condition to build human capital through investments in the early years and to harness the potential benefits of the demographic dividend. Endnotes Note: All dollar amounts are U.S. dollars unless otherwise indicated. 1 Two of the global nutrition targets—those for low birthweight and for child overweight—were not included in the analyses because of insufficient data on the prevalence of low birthweight and a lack of consensus on effective interventions to reach the target for child overweight. 2 Current financing by source is from the Results for Development Institute and can be found at http://www. investinnutrition.org/. 3 Information about the Power of Nutrition initiative is available at https://ciff.org/grant-portfolio/ the-power-of-nutrition/. 4 More information about the Global Financing Facility in support of Every Woman Every Child program can be found at https://www.everywomaneverychild.org/2014/10/04/global-financing-facility/. References Aguayo, V. M., R. Nair, N. Badgaiyan, and V. Krishna. 2016. “Determinants of Stunting and Poor Linear Growth in Children under 2 Years of Age in India: An In-Depth Analysis of Maharashtra’s Comprehensive Nutrition Survey.” Maternal & Child Nutrition 12 (Suppl. 1): 121–40. Alderman , H ., J. R. Behrman , and C. Puett . 2016 . “ Big Numbers about Small Children: Estimating the Eco- nomic Benefits of Addressing Undernutrition .” World Bank Research Observer 31 ( 2 ). Black, R. E., L. H. Allen, Z. A. Bhutta, L. E. Caulfield, M. de Onis, M. Ezzati, C. Mathers, J. Rivera, and the Maternal and Child Undernutrition Study Group. 2008. “Maternal and Child Undernutrition: Global and Regional Exposures and Health Consequences.” The Lancet 371 (9608): 243–60. Black, R. E., C. G .Victora, S. P. Walker, Z. A. Bhutta, P. Christian, M. de Onis, M. Ezzati, S. Grantham-Mcgregor, J. Katz, R. Martorell, R. Uauy, and the Maternal and Child Nutrition Study Group. 2013. “Maternal and Child Undernutrition and Overweight in Low-Income and Middle-Income Countries.” The Lancet 382: 427–51. Coady, D., I. Parry, L. Sears, and B. Shang. 2015. “How Large Are Global Energy Subsidies?” IMF Working Paper WP/15/105. https://www.imf.org/external/pubs/ft/wp/2015/wp15105.pdf Copenhagen Consensus Center. 2015. Smart Development Goals: The Post-2015 Consensus. http://www.copenha- genconsensus.com/sites/default/files/outcomedocument_col.pdf D’Alimonte, M., Rogers, H., de Ferranti, D. 2017. “Financing the Global Nutrition Targets” in Shekar, Kakietek, Dayton Eberwein, Walters eds. An Investment Framework for Nutrition: Reaching the Global Targets for Stunting, Anemia, Breastfeeding, and Wasting. Directions in Development. Washington, DC: World Bank. doi:10.1596/978- 1-4648-1010-7. https://tinyurl.com/InvestmentFrameworkNutrition 38 in Sub-Saharan Africa Stunting Reduction  Daniels, M. C. and L. Adair. 2004. “Growth in Young Filipino Children Predicts Schooling Trajectories through High School.” Journal of Nutrition 134: 1439–46. Eozenou, P. et al. 2017. Reducing Stunting in Africa. Washington, DC: World Bank. Fink, G., E. Peet, G. Danaei, K. Andrews, D. C. McCoy, C. R. Sudfeld, M. C. Smith Fawzi, M. Ezzati, and W. W. Fawzi. 2016. “Schooling and Wage Income Losses Due to Early-Childhood Growth Faltering in Developing Countries: National, Regional, and Global Estimates.” The American Journal of Clinical Nutrition 104 (1): 104–12. Hoddinott, J., H. Alderman, J. R. Behrman, L. Haddad, and S. Horton. 2013. “The Economic Rationale for Investing in Stunting Reduction.” Maternal and Child Nutrition 9 (Suppl. 2): 69–82. Hoddinott, J., J. A. Maluccio, J. R. Behman, R. Flores, and R. Martorell. 2008. “Effect of a Nutrition Intervention during Early Childhood on Economic Productivity in Guatemalan Adults.” Lancet 371 (9610): 411–16. Horton, S. and R. Steckel. 2013. “Malnutrition: Global Economic Losses Attributable to Malnutrition 1900–2000 and Projections to 2050.” In The Economics of Human Challenges, edited by B. Lomborg, 247–72. Cambridge, U.K.: Cambridge University Press. LiST (Lives Saved Tool). 2015. Baltimore, MD: Johns Hopkins Bloomberg School of Public Health. http:/livessa- vedtool.org/ Martorell, R., B. L. Horta, L. S. Adair, A. D. Stein, L. Richter, C. H. D. Fall, S. K. Bhargava, S. K. Dey Biswas, L. Perez, F. C. Barros, C. G. Victora, and Consortium on Health Orientated Research in Transitional Societies Group. 2010. “Weight Gain in the First Two Years of Life Is an Important Predictor of Schooling Outcomes in Pooled Analyses from Five Birth Cohorts from Low- and Middle-Income Countries.” Journal of Nutrition 140: 348–54. Ozaltin, E., K. Hill, and S. V. Subramanian. 2010. “Association of Maternal Stature with Offspring Mortality, Underweight, and Stunting in Low- to Middle-Income Countries.” JAMA 303 (15): 1507–16. Shekar, M., J. Kakietek, J. D. Eberwein, and D. Walters. 2017. An Investment Framework for Nutrition: Reaching the Global Targets for Stunting, Anemia, Breastfeeding, and Wasting. Directions in Development. Washington, DC: World Bank. doi:10.1596/978-1-4648-1010-7. https://tinyurl.com/InvestmentFrameworkNutrition UNICEF, WHO, and World Bank (United Nations Children’s Fund, World Health Organization and World Bank). 2015. Joint Malnutrition Estimates. Global Database on Child Growth and Malnutrition, http://www. who.int/growthdb/estimates2014/en/ Victora , C., R. Bahl, A. Barros, G. V. A. Franca, S. Horton, J. Krasevec, S. Murch, M. J. Sankar, N. Walker, and N. C. Rollins. 2016. “Breastfeeding in the 21st Century: Epidemiology, Mechanisms and Lifelong Effect.” The Lancet 387 ( 10017 ): 475–90. Whitley, S. and L. van der Burg. 2015. “Fossil Fuel Subsidy Report in Sub-Saharan Africa: From Rhetoric to Reality.” The New Climate Economy, ODA Working Paper. http://newclimateeconomy.report/2015/wp-con- tent/uploads/sites/3/2015/11/FFS-Reform-in-Africa_NCE-ODI_final.pdf WHO. 2009. “Global prevalence of vitamin A deficiency in populations at risk 1995–2005”de Benoits B, McLean E, Andersseon M, Rogers L. 2008. Iodine Deficiency in 2007: Global Progress Since 2003. Food and Nutrition Bulletin 29(3):195-202. WHO (World Health Organization). 2014. Comprehensive Implementation Plan on Maternal, Infant and Young Child Nutrition. Geneva: WHO. http://apps.who.int/iris/bitstream/10665/113048/1/WHO_NMH_NHD_14.1_eng. pdf?ua=1 CHAPTER 2 Investing in the Early Years: Nutrition in Africa 39 WHO (World Health Organization). 2016. Global Targets Tracking Tool Version 2.3 (November 2016), http:// www.who.int/nutrition/trackingtool/en/ World Bank. 2017. World Development Indicators (database), World Bank, Washington, DC (accessed May 1, 2017), http://data.worldbank.org/data-catalog/world-development-indicators 40 in Sub-Saharan Africa Stunting Reduction  CHAPTER 3 BENIN: An Investment Framework for Nutrition Benefits of Investing in Nutrition Key Messages • Thirty-four percent of children in Benin are chronically malnourished (stunted). Between 2006 and 2014, those in the wealthiest quintiles experienced the fastest rates of decline in 179,000 stunting prevalence, which remains high across all departments cases of stunting but is highest in Donga, Alibori, and Plateau, where at least half of prevented in 2025 children under five are stunted. • Benin has succeeded in reaching the global target for reducing wasting (acute malnutrition) to below 5 percent, and is currently seeing a significant downward trend in stunting prevalence, 29,000 which dropped from 45 percent in 2006 to 34 percent in 2014. child deaths prevented in 2025 • Scaling up a package of high-impact nutrition-specific interventions in Benin to contribute toward achieving the global nutrition targets would require an additional $30.9 million per year over 10 years and would yield massive benefits (see panel on the right). These investments are over and above those needed for 3.7 MILLION improving water and sanitation and for addressing issues around case-years of anemia in women’s empowerment and food security. women prevented in 2025 • This scale-up would require additional financing equivalent to a 16 percent increase in current government health expenditures and could be financed from a combination of domestic budgets, official development assistance (ODA) and innovative financing 433,000 sources such as the Power of Nutrition.1 babies exclusively breastfed • The economic benefits generated over the productive lives of beneficiaries from this investment would be $2.8 billion for the prevention of stunting, $1.9 billion for breastfeeding, $455 million for the prevention of anemia, and $272 million for the treatment of severe wasting. 121,000 • Returns on every dollar invested in reaching the global nutrition cases of severe wasting treated targets range from $14 for stunting and anemia to $23 for wasting and $58 for exclusive breastfeeding. • To finance the nutrition scale-up, two lower-cost scale up $14–$58 scenarios are estimated to require between $8 and $15 million per year over the next 10 years. In an environment of constrained return for every dollar invested resources, starting with one of these two scenarios would be a strong first investment, but it would need to be followed by increased investment toward full scale-up to achieve the global nutrition targets. $2.8 BILLION generated from investments to reduce stunting* *The economic benefits are calculated over the productive lives of the children benefiting from the interventions that prevent stunting. CHAPTER 3 Benin: An Investment Framework for Nutrition 41 Investment Case for Nutrition Ensuring optimum nutrition—particularly during the 1,000-day period from pregnancy to a child’s second birthday—can alter an individual’s development trajectory and maximize her or his productive potential. Chronic malnutrition has important lifelong consequences for health and cognitive development. Losses to cognitive development in early childhood resulting from chronic malnutrition are irreversible. Being stunted (low height-for-age) in early childhood is associated with a delayed start at school, reduced schooling attain- ment, and substantially decreased adult incomes at both the individual and country level (Daniels and Adair 2004; Fink et al. 2016; Hoddinott et al. 2008; Martorell et al. 2010). These consequences add up to overall gross domestic product (GDP) losses of 4 to 11 percent in Africa and Asia (Horton and Steckel 2013). Importantly, chronic undernutrition can be transmitted through an inter-generational cycle, where malnourished mothers are more likely to have stunted children (Aguayo et al. 2016; Ozaltin et al. 2010). Investments in nutrition are highly cost-effective and among the best value-for-money development actions (Copenhagen Consensus Center 2015; Hoddinott et al. 2013). An Investment Framework for Nutrition developed by the World Bank in partnership with R4D, 1000 Days, and the Bill & Melinda Gates Foundation estimated high returns on every dollar invested in nutrition: from $4 in returns for treating acute malnutrition (wasting) to $11 for preventing stunting, $12 for the treatment and prevention of anemia, and $35 for increasing the prev- alence of exclusive breastfeeding (Shekar et al. 2017). Not only do investments in nutrition produce substantial economic benefits, but they also lay the groundwork for the success of investments in other sectors. Investments in the early years—including early life nutrition, early learning and stimulation, and the provision of nurturing care and protection from stress—ensure that all children reach their human potential and contrib- ute to the economic growth of their nation. The analysis presented below focuses on high-impact nutrition-spe- cific interventions with strong evidence of efficacy in reducing malnutrition, and estimates the financing needs, impacts, and economic benefits of scaling up these interventions in Benin. Country Context The Republic of Benin is a small coastal country in Sub-Saharan Africa, with a population of almost 10.9 million and a population growth rate of 2.7 percent. Benin has a young population, with approximately 15 percent of the country under age five (UN DESA 2015). Low human capital remains one of the key challenges in reducing poverty and achieving greater socioeconomic equity. More than half of the population resides in rural areas (World Bank 2016); agriculture accounts for the majority of employment in Benin, with cotton being the primary export commodity. Over 98 percent of young people are employed, largely in the agriculture sector, with less than 8 percent of those between ages 15 and 34 working in non-agriculture salaried jobs (World Bank 2014). Poverty rates are moderately higher in rural areas (39.7 percent) than in urban centers (31.4 percent), and the data show income disparities between districts as well as between rural and urban households. In 2011, 64.4 percent of the rural population subsisted on less than $1.25 per day, nearly double the rate observed in urban areas. Similarly, the poverty headcount in 2011 was estimated at over 75 percent of the population in Atacora compared to 7 percent in Littoral (World Bank 2014). The Human Development Index (HDI) showed Benin ranked 167 out of 188 countries in 2016 (UNDP 2016). High income inequality exacerbates vulnerability and prevents poor households from meeting basic needs. Child malnutrition, an underlying cause of up to 45 percent of deaths of children under age five (Black et al. 2013), has emerged as one of the key markers of poverty and vulnerability as well as one of the key challenges in ensuring optimal accumulation of human capital in the country. 42 in Sub-Saharan Africa Stunting Reduction  To address these issues, the Benin government developed the Third Poverty Reduction Strategy (SCRP), covering the period 2011–15. This strategy aimed to improve the quality of life in Benin through health and education and to place the country on the road to emerging-market status by 2025 by strengthening human capital (as its third pillar) and making gains toward sustainable regional development (as its fifth pillar) (IDA and IMF 2011). Progress has been made in reducing total fertility rates (TFR) in Benin, which through the 1990s was up above 6.0 and is now declining, with a current TFR of 4.8. Declining TFR can be a positive contributor to improved health and nutrition outcomes (World Bank 2016).3 Nutritional Status in Benin Persistently high rates of undernutrition remain a serious human development challenge in Benin. More than a third of children (34 percent) under five years of age are stunted and 4.5 percent are wasted (low weight-for-height) (UNICEF 2015). Between 1996 and 2006, data from Benin Demographic and Health Surveys (BDHS) showed an increasing trend in stunting prevalence, from 39.1 percent to 44.7 percent. However, the latest data from the 2014 Benin Multiple Indicator Cluster Survey (MICS) showed promising declines in both chronic and acute malnutrition (Figure 3.1), albeit the different surveys—BDHS and MICS—may not be comparable because of their varying meth- odologies. Nevertheless, Benin ranks 103rd of 130 countries ranked from lowest to highest stunting prevalence (IFPRI 2016). National estimates mask socioeconomic and geographic and disparities in stunting prevalence. Significant differences in stunting prevalence remain among children in poorer and wealthier households, and similarly among children in rural and urban households. Between 2006 and 2014, stunting declined at a faster rate among children living in households in the wealthiest quintile when compared with those in the poorest quintile (Figure 3.2). In the wealthiest households, nearly one in five children is stunted, as compared to the poorest quintile where almost half of children are stunted. Not only was the decline less dramatic among children in the poorest quintile but rates of stunting in this group remain much higher in 2014 as compared with 1996 (46 percent versus 36 percent). The highest stunting prevalence rates are concentrated in the north where two departments have stunting prev- alence of over 40 percent (Figure 3.3). This is in stark contrast to the stunting prevalence in the southernmost department, Littoral, at 17.8 percent. Given the regional variation in stunting prevalence, it would be important to understand the key drivers of undernutrition in these highest prevalence regions, and to design targeted interven- tions to address them. Figure 3.1: Trends in Undernutrition in Benin, 1996–2014 Figure 3.2: Disparities in Stunting by Wealth Quintile, 1996–2014 70 64.3 61.3 70 64.3 60 61.3 60 AMONG CHILDREN UNDER FIVE 60 AMONG CHILDREN UNDER FIVE 60 50 AMONG CHILDREN UNDER FIVE STUNTING PREVALENCE (%) AMONG CHILDREN UNDER FIVE 50 44.7 41.4 50 STUNTING PREVALENCE (%) PREVALENCE (%) 39.1 50 39.1 44.7 41.4 40 PREVALENCE (%) 40 39.1 39.1 34 40 34 40 30 30 30 30 17.1 20 17.1 20 20 20 9.3 8.4 9.3 8.4 4.5 10 10 4.5 10 10 0 0 0 0 1996 2001 1996 2006 2001 2011-12 2006 2014 2011-12 2014 1996 1996 2001 2001 2006 2006 20142014 Stunting Stunting Wasting Wasting Anemia Anemia Poorest quintile Poorest quintile Richest quintile Richest quintile National National Sources for Figures 3.1 and 3.2: BDHS for 1996, 2001, 2006, and 2011–2012; Benin MICS 2014 CHAPTER 3 Benin: An Investment Framework for Nutrition 43 30 COTE D’LVOIRE Swaziland STUNTING PREV 25 Kenya Mauritania 20 Congo Ghana 15 Sao Tome and Principe 10 5 Figure 0 3.3: Number of Stunted Children and Stunting Prevalence in Benin by Region, 2014–2015 500 1500 2500 3500 4500 5500 6500 7500 8500 9500 GDP PER CAPITA (INTERNATIONAL $ PPP) ALIBORI ALIBORI ATAKORA ATAKORA BORGOU BORGOU Stunting Prevalence Total Children By Region (%) DONGA NIGER Stunted By Region DONGA 50 ≥ 40%: Very high 86,672 65,000 30–39%: High 45,000 20–29%: Medium DRC Mozambique 35,000 20,000 Chad Ethiopia STUNTING PREVALENCE (%) <20%: Low 19,250 40 COLLINES Rwanda COLLINES Department Boundaries Department Boundaries Malawi Sierra Leone Tanzania ZOU Benin ZOU PLATEAU PLATEAU 30 COUFFO Uganda COUFFO Guinea OUÉMÉ OUÉMÉ MONO Zimbabwe MONO LITTORAL Senegal LITTORAL 20 ATLANTIQUE ATLANTIQUE Data source: Benin MICS 2014. IBRD 42940 | MAY 2017 These maps were produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denom- inations and any other information shown on these maps do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any IBRD 42940 | JUNE 2017 endorsement acceptance of such boundaries. or10 500 1000 1500 2000 2500 3000 The prevalence of stunting in Benin is comparable to the average in the region (35.2 percent) (World Bank 2017). Nevertheless, some African countriesGDP PER with CAPITA lower (INTERNATIONAL per $ PPP) as Guinea, Uganda, and Zim- capita income—such babwe—exhibit similar or lower prevalence of child stunting, which belies the fact that economic development needs to be coupled with other investments to attain improvements in childhood nutrition (Figure 3.4). Figure 3.4: Prevalence of Stunting and GDP per Capita: Benin and Selected Low-Income Countries 50 DRC Mozambique Chad STUNTING PREVALENCE (%) Niger Ethiopia 40 Malawi Rwanda Tanzania Sierra Leone 30 Uganda BENIN Guinea Zimbabwe 20 Senegal 10 500 1000 1500 2000 2500 3000 GDP PER CAPITA (INTERNATIONAL $ PPP) Sources: Benin MICS 2014; World Bank 2017. Wasting, also known as acute malnutrition, is typically classified as either severe or moderate. Wasting can result from food insecurity in resource-poor settings with insufficient dietary quality, quantity and diversity, suboptimal breastfeeding, and recurrent episodes of illness such as diarrhea. Wasting prevalence across Sub-Sa- haran Africa is second highest in the world, after South Asia, with 13 million children (7.8 percent) suffering 44 in Sub-Saharan Africa Stunting Reduction  from acute malnutrition. However, this varies at country level. In 2014, Benin exceeded their goal to reduce wasting to under 5 percent with a downward trending wasting prevalence of 4.5 percent (Benin MICS 2014). Micronutrient deficiencies (a form of malnutrition that relates to a deficiency in essential vitamins and minerals needed for body functions and is sometimes referred to as hidden hunger) are highly pervasive in Benin. Anemia, a condition caused by inadequate dietary intake of iron, helminth infections, and malaria, among other causes, has cross-generational impacts. Approximately 41 percent of women of reproductive age in Benin are anemic (BDHS 2011–2012), which affects not only women’s own health, but also contributes to the intergenerational cycle of undernutrition. Over half (58 percent) of children aged 6 to 59 months are anemic, with one department having an anemia prevalence of over 85 percent (BDHS 2011–2012). Although overall anemia has declined significantly from more than 80 percent in 2001 (BDHS 2001), largely as a result of malaria prevention measures, maintaining and promoting dietary diversity with iron-rich foods to prevent micronutrient deficiencies in women of reproduc- tive age is essential to set the stage for their children to achieve optimal nutrition and development. Benin has made some gains in child survival and maternal health, including reduced maternal and child mortality ratios and improvements in the essential package of services for maternal and newborn health, such as deliveries in a health facility, but significant challenges remain in improving nutrition outcomes and addressing the multiple determinants of undernutrition. At the underlying level, stunting in Benin is associated with insufficient access to a nutritionally rich and diverse diet, inadequate hygiene and sanitation practices, and suboptimal care and feeding practices, among other causes. Demand- and supply-side barriers influence optimum feeding practices such as breastfeeding, appropriate food consumption, and dietary diversity. An analysis conducted in 2014 looked at household surveys and found that 11 percent of households were food insecure, with the Couffo, Mono, and Atacora departments experiencing the highest prevalence of food inse- curity (exceeding 25 percent of households) (WFP 2014). Similarly, the rate of food insecurity is estimated to be twice as high in rural areas (approximately 15 percent) as in urban settings (8 percent). Households with poor food consumption were found to have a limited diversity of diet, which consisted mainly of cereals and starchy roots. More than 85 percent of families are dependent on markets, making them vulnerable to climatic and economic shocks as well as seasonal and price fluctuations. The increase in market rate for staple crops in times of crisis translates to a reduction in purchasing power for the most vulnerable families (WFP 2014). Recogniz- ing that factors other than poverty and food insecurity put children at risk of chronic malnutrition, effective multisectoral strategies are needed to address undernutrition across the country. Political Commitment to Reduce Malnutrition The policy environment around nutrition is gaining momentum in Benin. For several decades the institutional leadership responsible for implementing and overseeing nutrition policies moved between various ministries. In 2007, Benin, with the support of the World Bank, undertook reforms related to food and nutrition. Since 2009, Benin has adopted several nutrition-specific policies and plans, beginning with the Strategic Plan for Food and Nutrition Development (PSDAN) and the National Policy for the Protection, Encouragement, and Promotion of Breastfeeding (UN SCN Secretariat 2013). Following on these achievements, Benin established a multisector national Council on Food and Nutrition (CAN) platform in 2011 to bring together all sectors involved in nutrition to improve food, health, and nutrition outcomes. CAN is attached to the Presidency of the Republic and fosters synergies between nutrition research and policy recommendations (SUN 2016). Also in 2011, Benin joined the SUN movement, solidifying the coun- try’s commitment to ending malnutrition. Food and nutrition is increasingly seen as a development priority at the commune level as well, evidenced by the integration of nutrition within the National Association of Communes of Benin (ANCB) support funds. Benin is implementing a common results framework to align actions to address chronic malnutrition at the commune level (SUN 2016). CHAPTER 3 Benin: An Investment Framework for Nutrition 45 Benin has also integrated nutrition within agriculture policy by establishing the Law on Agriculture and Food and Nutrition Security (SUN 2016). The growing association of nutrition and agriculture is further grounded in the Actions for Environment and Development (ACED) strategic plan, which includes an arm designed to improve nutrition security of vulnerable communities by adapting agriculture production to nutrition needs (AECD-Benin, No date). Current Financing for Nutrition In 2015 in Benin, the government and foreign donors spent a total of $1.64 million on interventions that will contribute to reaching the global targets for nutrition.2 Of that amount, $383,000 came from the government and $1.26 million came from ODA. This contribution from ODA included $876,000 for stunting, $106,000 for anemia, $183,000 for breastfeeding, and $256,000 for wasting.3 These estimates reflect the current spending on nutrition, and the following sections detail additional financing needed in order for Benin to contribute to reaching the global targets on nutrition. Global Targets for Nutrition Substantial improvements to the nutritional status of women and children can be realized if adequate invest- ment is made in a set of evidence-based nutrition-specific interventions that ensure optimum nutrition during the critical 1,000-day window between the start of a woman’s pregnancy and the child’s second birthday (Black et al. 2008, 2013). For women, these include interventions to prevent anemia before and during pregnancy as well as those aimed at improving protein energy intake during pregnancy. Interventions targeted toward children and their mothers aim to improve breastfeeding and complementary feeding practices, enhance the micronutrient status of children, and treat acute malnutrition in children. In 2012—to rally the international community around improving nutrition—the 176 members of the World Health Assembly endorsed the first-ever global nutrition targets, focusing on six areas: stunting, anemia, low birthweight, childhood overweight, breastfeeding, and wasting. These targets aim to boost investments in cost-effective inter- ventions, spearhead better implementation practices, and catalyze progress toward reducing malnutrition. The tar- gets for stunting and wasting are enshrined within the United Nations’ Sustainable Development Goal 2 (SDG 2), Table 3.1: Four World Health Assembly Targets for Nutrition and Benin’s Contribution toward Meeting Them RANK PREVALENCE PROGRESS Reduce the number of stunted children 1 STUNTING* 103/132 34 under five by 40% Stuntin Reduce the number of women of 2 ANEMIA 177/185 41.4 reproductive age with anemia by 50% An mi Increase the rate of exclusive breastfeeding  5 BREASTFEEDING 55/141 41.4 in the first six months up to at least 50% Exclusiv br stf din Reduce and maintain childhood wasting 6 WASTING* 64/130 4.5  alnutrition) to less than 5% (acute m W stin LEGEND: Off course, no progress Off course, some progress On course, good progress *Stunting and wasting are included within the United Nations’ Sustainable Development Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. Sources: Nutrition targets from WHO 2014; Rank and progress from IFPRI 2016; Prevalence data from BDHS 2011-2012. 46 in Sub-Saharan Africa Stunting Reduction  which commits to ending malnutrition in all its forms by the year 2030. The 2016 Global Nutrition Report ranked each country’s progress in contributing toward achieving the global targets (Table 3.1) (IFPRI 2016).4 Coverage of key nutrition-specific interventions in Benin is low and remains well below the levels necessary to advance progress in reducing malnutrition among Beninese children. Table 3.2 and Figure 3.5 summarize the current coverage of and delivery platforms for nutrition-specific interventions in Benin. Table 3.2: Delivery Platforms of Nutrition-Specific Interventions in Benin INTERVENTION PLATFORM Antenatal micronutrient supplementation (iron and folic acid only) Health facility and community Complementary feeding education Health facility, community, and communication campaigns Breastfeeding promotion Health facility, community, and communication campaigns Balanced energy protein supplementation for pregnant women Health facility, community, and social safety net programs Intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions Health facility and community Vitamin A supplementation Health facility, community, and food fortification Public provision of complementary foods Health facility, community, and social safety net programs Treatment of severe acute malnutrition Health facility and community Iron and folic acid supplementation for non-pregnant women of reproductive age School, community, facility, and marketplace Staple food fortification Marketplace Pro-breastfeeding social policies Government policies National breastfeeding promotion campaigns Media Figure 3.5: Coverage of Key Nutrition-Specific Interventions: Benin and Sub-Saharan Africa Antenatal micronutrient supplementation - (iron and folic acid only) Pro-breastfeeding Complementary social policies feeding education 100 80 60 Staple food Breastfeeding fortification 40 promotion 20 Iron and folic acid Balanced energy supplementation for non-pregnant protein supplementation women of reproductive age for pregnant women Treatment of severe Intermittent presumptive acute malnutrition treatment of malaria in pregnancy in malaria-endemic regions Public provision of Vitamin A Note: Red shading represents Benin and light blue shading complementary foods supplementation represents average Sub-Saharan Africa coverage CHAPTER 3 Benin: An Investment Framework for Nutrition 47 Economic Benefits of Investing in Nutrition There is a strong body of evidence that shows high economic returns to investing in nutrition (Alderman et al. 2016; Copenhagen Consensus Center 2015; Hoddinott et al. 2013). Scaling up these proven nutrition-specific interventions can ensure that mothers are healthy and well nourished, that they can provide optimal nutrition to their children, that children realize their full physical and cognitive development potential, and that wom- en’s productivity is not hampered by illness, especially anemia (Figure 3.6). Figure 3.6: How Reaching the Global Nutrition Targets Generates Economic Benefits Co nitive Development STUNTING Childr n’s H lth & Le rnin & Educ tion l Nutrition St tus Att inment WASTING Glob l Nutrition T r ts R ch d Adult Productivit EASTFEE BRE EDING W es ANEMIA Econom Wom n’s H lth & (GDP) Nutrition St tus In Benin, scaling up the package of nutrition-specific interventions would produce substantial economic ben- efits over the productive lifetimes of the affected women and children (Figure 3.7). Additional health system cost-savings would also be likely because many of these investments reduce the burden of childhood illnesses such as diarrhea and pneumonia. Figure 3.7: Investments in Benin to Meet the Global Nutrition Targets Have Enormous Economic Returns Total Economic $1 Invested Benefits (Millions)a Yieldsb STUNTING $2,821 $14 $ $1,933 $58 1 BREASTFEEDING $ $ ANEMIA $455 $ $14 WASTING $272 $23 a. Total economic benefits over 10 years for women and over the productive lives of children who benefit from these interventions, defined as the period between the age of 18 and a “retirement” age - the life expectancy or the age of 65, whichever is lower. b. Benefit calculation assumes a 3 percent discount rate for both financing needs and benefits and a GDP growth rate of 3 percent. 48 in Sub-Saharan Africa Stunting Reduction  Financing Needs, Impacts, and Cost-Effectiveness of Scaling-Up Nutrition-Specific Interventions Using the methodology detailed in An Investment Framework for Nutrition (Shekar et al. 2017), this brief presents estimates of the resources needed to scale up a package of high-impact nutrition-specific interventions in Benin to meet the global nutrition targets for stunting, anemia, breastfeeding, and wasting, along with their estimated nutrition, health, and economic impacts. An additional $30.9 million per year over 10 years is needed to scale up the package of key interventions (Table 3.3). The health and nutrition impacts of this investment are shown in Table 3.4. Table 3.3: Estimated 10-Year Financing Needs and Cost-Effectiveness of Scaling Up Nutrition-Specific Interventions, Benin TOTAL 10-YEAR COST PER CASE COST PER DEATH INTERVENTION (NUTRITION TARGET) FINANCING NEEDS OF STUNTING AVERTED (US $) (US $M) AVERTED (US $) For pregnant women and mothers of infants Antenatal micronutrient supplementation (stunting, anemia) 9.1 68,675 49,673 Infant and young child nutrition counseling (complementary feeding 21.0 2,255 501 education and breastfeeding promotion combined) Complementary feeding education (stunting) 11.0 4,799 284 Breastfeeding promotion (stunting, breastfeeding) 9.9 1,419 3,282 Balanced energy protein supplementation for pregnant women (stunting) 41.8 24,602 25,721 Intermittent presumptive treatment of malaria in pregnancy in malaria- 6.8 3,330 1,144 endemic regions (stunting, anemia) For infants and young children Vitamin A supplementation (stunting) 2.7 1,317 184 Prophylactic zinc supplementation (stunting) 66.2 13,255 1,254 Public provision of complementary food (stunting) 104.7 33,321 1,930 Treatment of severe acute malnutrition (wasting) 14.6 3,338 n.a For non-pregnant women and general population Iron and folic acid supplementation for non-pregnant women (anemia) 13.8 n.a 41,926 Staple food fortification (anemia) 3.7 n.a Pro-breastfeeding social policies (breastfeeding) 5.0 n.a n.a National breastfeeding promotion campaigns (breastfeeding) 20.0 n.a n.a TOTAL: 309.3 10,465 1,455 Note: Financing needs and impacts assume a linear scale-up of interventions from current coverage level to 90 percent over five years, then maintained at 90 percent for an additional five years. Unit costs for each intervention were drawn from available unit costs from neighboring countries, global costs, or estimates available in the literature. The estimated costs include an additional 12 percent (11 percent for pro-breastfeeding social policies and promotion campaigns) to account for monitoring, evaluation, capacity and policy development that may be necessary to reach full scale-up of the interventions. The Lives Saved Tool (LiST; see LiST 2015) was used to estimate the impact of interventions that target pregnant women and children. The impact of interventions that target the general population or non-pregnant women was estimated using a Microsoft Excel model. It should be noted that the LiST model does not capture potential synergies between specific interventions (e.g. the fact that the impact of behavior change communication interventions may be higher in populations that have access to affordable and diversified foods or in populations with higher levels of educa- tional attainment). Therefore, it is possible that the impact estimates generated using LiST in fact underestimate the true impact of the interventions in some contexts. n.a. = not applicable CHAPTER 3 Benin: An Investment Framework for Nutrition 49 ANEMIA Total 10%: Among the set of proposed interventions, vitamin A supplementation is the most cost-effective for preventing stunting, averting nearly 15,000 cases of stunting and 22,000 child deaths over 10 years. Complementary feed- ing education prevents nearly 40,000 cases of stunting at a cost of $284 per case averted, and total cost of $11 million over 10 years. Breastfeeding promotion through counseling mothers WASTING is projected to increase the number Total: of infants exclusively breastfed by more than 430,000, prevent 77,000 deaths, 22% and cost just $9.9 million over 10 years. Among the package of nutrition-specific interventions, breastfeeding promotion is the most cost-effec- tive intervention for preventing child mortality, and costs $56 per child exclusively breastfed. For preventing NIGER STUNTING maternal anemia, staple food fortification proves to be most cost-effective, at a cost of $44 for each case-year of anemia Total 71%: prevented BREASTFEEDING in women. Over 10 years, staple food fortification will prevent nearly 940,000 cases of anemia in women at a cost of $3.7 million. Among pregnant women, antenatal Total: 8% micronutrient supplementation will prevent nearly 570,000 case-years of anemia at a cost of $16 per case averted, or a total of $9.1 million over 10 years. TOTAL Interventions to reduce stunting would require the most resources, TEN-YEAR accounting FINANCING for about NEEDS: 82 percent $771 million of the total amount required for scale-up. However, some of the stunting interventions would also affect the breastfeeding and anemia targets. Figure 3.8 represents the distribution of total cost across interventions to address the four targets. Figure 3.8: Ten-Year Financing Needs for Scaling Up a Package of Nutrition-Specific Interventionsin Benin, by Percent per Intervention WASTING ANEMIA Total: 5% Total 11%: BREASTFEEDING Total: 11% Benin STUNTING TOTAL TEN-YEAR Total 82%: FINANCING NEEDS: $309 million Note: Some costs for anemia, breastfeeding, and stunting are shared across interventions. Costs for breastfeeding promotion ($9.99 million) have been included in both the- total cost for the breastfeeding target and the total cost for the stunting target; the costs of intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions ($6.8 million) and antenatal micronutrient supplementation ($9.1 million) have been included in both the total cost for the anemia target and the total cost for the stunting target. Two Alternative Investment Packages Relative to current expenditures on health, the investment required to scale up the set of effective nutrition-spe- cific interventions may present significant challenges for Benin. In an environment of constrained resources in which Benin may not be able to raise $309 million over the next 10 years, two alternative investment packages are laid out for consideration. 50 in Sub-Saharan Africa Stunting Reduction  Table 3.4: Benefits and Cost-Effectiveness by Investment Package, Benin GLOBAL PRIORITY CATALYZING FULL PACKAGE: BENEFIT All interventions needed TARGET PACKAGE PROGRESS PACKAGE to meet targets $8.1 million/year in $14.6 million/year in $30.9 million/year in financing need financing need financing need Cases of stunting reduced by STUNTING 66,000 94,000 179,000 2025 (vs 2015)a Cases of anemia in women ANEMIA 2 million 2.4 million 3.7 million prevented by 2025 Additional babies BREASTFEEDING 433,000 433,000 433,000 breastfed over 10 years Child deaths averted 20,000 23,000 29,000 over 10 years ALL TARGETS Cost per death averted 4,072 6,361 10,465 Cost per case of 601 1,074 1,455 stunting averted a. Total impacts of proposed intervention package combined with other health and poverty reduction efforts. The Priority Package: The first—the “priority package”—includes interventions that are the most cost-effec- tive; that is, have the lowest cost per health outcome (e.g., case of stunting averted), and that have well-es- tablished global policy guidelines and delivery platforms. Based on those two criteria, the priority package includes antenatal micronutrient supplementation, infant and young child nutrition counseling, intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions, vitamin A supplementation, the treatment of severe acute malnutrition, weekly iron and folic acid supplementation for girls 15–19 years of age attending school, and fortification of wheat and maize flour with iron and folic acid. These interventions would be scaled up to full program coverage in the first five years and maintained at full coverage levels for the last five years. This priority package would require an estimated $81 million over 10 years, or $8.1 million annually (see Table 3.4). During the 10 years of scale up, this package would prevent more than 66,000 cases of stunting, prevent nearly 2 million case-years of anemia in women and would result in 433,000 children under six months of age being exclusively breastfed. It would also avert over 20,000 deaths in children under five years of age. The Catalyzing Progress Package: The second alternative—the “catalyzing progress package”—includes scale-up of all interventions in the priority package, plus a phased approach to scaling up public provision of complementary foods, balanced energy protein supplementation, prophylactic zinc supplementation, and weekly iron and folic acid supplementation for women outside of schools. It is assumed that, for the latter set of interventions, during the first five years emphasis will be placed on establishing global guidelines and on operational research to develop effective delivery platforms, or to develop less expensive products or more cost-effective technologies. Costs are approximated as the cost of scaling up this set of interventions from 0 to 10 percent coverage only in the first five years. In the subsequent five years, it is assumed that the coverage CHAPTER 3 Benin: An Investment Framework for Nutrition 51 expansion of those interventions will accelerate and reach 60 percent by 2025. This package would require $14.6 million per year, or a total of $146 million over 10 years (Table 3.4). It would prevent more than 94,000 cases of stunting among children under age five and nearly 2.4 million case-years of anemia in women. It would also increase the number of exclusively breastfed children under six months of age by 433,000 and avert nearly 23,000 deaths. In comparing the relative cost-effectiveness of the three investment packages, the two alternative packages are more cost-effective in preventing deaths and stunting. However, neither is as effective as the full package in making progress toward achieving the stunting, wasting, and anemia targets. The priority and catalyzing prog- ress packages would prevent 20 ,000 and 23,000 deaths respectively, compared with 29,000 deaths prevented with the full package over 10 years. Under the full package scenario, 179,000 cases of childhood stunting would be prevented, compared with 94,000 cases under the catalyzing progress scenario and 66,000 cases under the priority package scenario. Furthermore, there would be nearly 1.7 million and 1.3 million more case-years of anemia prevented in women under the priority package and catalyzing progress package, respectively. A Call to Action As the world stands on the cusp of the new Sustainable Development Goals, there is an unprecedented oppor- tunity to save children’s lives, build future human capital and cognitive development, and drive faster eco- nomic growth. Scaling up key nutrition interventions during the critical 1,000 day window of early childhood would pay lifelong dividends, translating to healthier societies and more robust economies. If this window is missed, it is missed for life. The additional financing needed to reach the global nutrition targets will require coordinated efforts by all stakeholders and a supportive policy environment. To achieve these targets, Benin will need to increase its general government health expenditure by approximately 16 percent.5 Although this level of domestic financ- ing is ambitious, Benin has demonstrated its commitment to improving the nutritional status of children and mothers. In the long term, nutrition interventions have significant potential to reduce poverty and boost shared prosperity. Accelerating the reduction of stunting in Benin will be essential for maximizing the return on investments in early childhood development, in education, and more broadly in policies aimed at fostering and enhancing human capital accumulation and job creation. Investing in the early years is even more critical because the Africa region is entering a demographic transition with an expected increase in the working-age population from 54 percent in 2010 to 64 percent in 2090. The scale-up of the key nutrition-specific interventions to reduce stunting is estimated to generate considerable returns in economic benefits over the productive lives of ben- eficiaries, and is a necessary condition to build human capital through investments in the early years and to harness the potential benefits of the demographic dividend. Endnotes Note: All dollar amounts are U.S. dollars unless otherwise indicated. 1 Information about the Power of Nutrition initiative is available at https://ciff.org/grant-portfolio/ the-power-of-nutrition/. 2 Current financing is sourced from Results for Development Institute and can be found at http://www.inves- tinnutrition.org/. 3 Note that because some funded interventions contribute to more than one target, the sum of funding across the four targets is less than the total funding for each target added together. 52 in Sub-Saharan Africa Stunting Reduction  4 Two of the global nutrition targets—those for low birthweight and for child overweight—were not included in the analyses because of insufficient data on the prevalence of low birthweight and a lack of consensus on effective interventions to reach the target for child overweight. 5 WHO National Health Accounts database indicates general government health expenditure in Benin was US $197m in 2014. At that level, this will need to be increased by 16 percent to accommodate the $30.9 million per year required for scale-up of the interventions. References AECD-Benin. No date. Actions pour l’environnement et le developpement durable : Strategic plan 2014–2018. Abom- ey-Calvi, Benin: ACED. http://www.aced-benin.org/sites/default/files/strategic-plan-aced_summary.pdf Aguayo, V. M., R. Nair, N. Badgaiyan, and V. Krishna. 2016. “Determinants of Stunting and Poor Linear Growth in Children under 2 Years of Age in India: An In-Depth Analysis of Maharashtra’s Comprehensive Nutrition Survey.” Maternal & Child Nutrition 12 (Suppl. 1): 121–40. Alderman , H ., J. R. Behrman , and C. Puett . 2016 . “ Big Numbers about Small Children: Estimating the Eco- nomic Benefits of Addressing Undernutrition .” World Bank Research Observer 31 ( 2 ). BDHS (Benin Demographic Health Survey). 1996. Benin Demographic Health Survey, 1996. http://dhsprogram. com/what-we-do/survey/survey-display-87.cfm ———. 2001. Benin Demographic Health Survey, 2001. http://dhsprogram.com/pubs/pdf/SR93/SR93.pdf ———. 2011–2012. Benin Demographic Health Survey: Final Report 2011-2012. Calverton, Maryland, USA: INSAE and ICF International (published in 2013). http://dhsprogram.com/publications/publication-FR270-DHS-Fi- nal-Reports.cfm Benin MICS (Benin Multiple Cluster Survey). 2014. Enquête par grappes à indicateurs multiples 2014, Rapport final. Cotonou, Bénin: Institut national de la statistique et de l’analyse économique (INSAE) published in 2015). Black, R. E., L. H. Allen, Z. A. Bhutta, L. E. Caulfield, M. de Onis, M. Ezzati, C. Mathers, J. Rivera, and the Maternal and Child Undernutrition Study Group. 2008. “Maternal and Child Undernutrition: Global and Regional Exposures and Health Consequences.” The Lancet 371 (9608): 243–60. Black, R. E., C. G .Victora, S. P. Walker, Z. A. Bhutta, P. Christian, M. de Onis, M. Ezzati, S. Grantham-Mcgregor, J. Katz, R. Martorell, R. Uauy, and the Maternal and Child Nutrition Study Group. 2013. “Maternal and Child Undernutrition and Overweight in Low-Income and Middle-Income Countries.” The Lancet 382: 427–51. Copenhagen Consensus Center. 2015. Smart Development Goals: The Post-2015 Consensus. http://www.copenha- genconsensus.com/sites/default/files/outcomedocument_col.pdf Daniels, M. C. and L. Adair. 2004. “Growth in Young Filipino Children Predicts Schooling Trajectories through High School.” Journal of Nutrition 134: 1439–46. Fink, G., E. Peet, G. Danaei, K. Andrews, D. C. McCoy, C. R. Sudfeld, M. C. Smith Fawzi, M. Ezzati, and W. W. Fawzi. 2016. “Schooling and Wage Income Losses Due to Early-Childhood Growth Faltering in Developing Countries: National, Regional, and Global Estimates.” The American Journal of Clinical Nutrition 104 (1): 104–12. Hoddinott, J., H. Alderman, J. R. Behrman, L. Haddad, and S. Horton. 2013. “The Economic Rationale for Investing in Stunting Reduction.” Maternal and Child Nutrition 9 (Suppl. 2): 69–82. CHAPTER 3 Benin: An Investment Framework for Nutrition 53 Hoddinott, J., J. A. Maluccio, J. R. Behman, R. Flores, and R. Martorell. 2008. “Effect of a Nutrition Intervention during Early Childhood on Economic Productivity in Guatemalan Adults.” Lancet 371 (9610): 411–16. Horton, S. and R. Steckel. 2013. “Malnutrition: Global Economic Losses Attributable to Malnutrition 1900–2000 and Projections to 2050.” In The Economics of Human Challenges, edited by B. Lomborg, 247–72. Cambridge, U.K.: Cambridge University Press. IFPRI (International Food Policy Research Institute). 2016. Global Nutrition Report 2016: From Promise to Impact – Ending Malnutrition by 2030. Washington, DC: IFPRI. LiST (Lives Saved Tool). 2015. Baltimore, MD: Johns Hopkins Bloomberg School of Public Health. http:/livessa- vedtool.org/ Martorell, R., B. L. Horta, L. S. Adair, A. D. Stein, L. Richter, C. H. D. Fall, S. K. Bhargava, S. K. Dey Biswas, L. Perez, F. C. Barros, C. G. Victora, and Consortium on Health Orientated Research in Transitional Societies Group. 2010. “Weight Gain in the First Two Years of Life Is an Important Predictor of Schooling Outcomes in Pooled Analyses from Five Birth Cohorts from Low- and Middle-Income Countries.” Journal of Nutrition 140: 348–54. Ozaltin, E., K. Hill, and S. V. Subramanian. 2010. “Association of Maternal Stature with Offspring Mortality, Underweight, and Stunting in Low- to Middle-Income Countries.” JAMA 303 (15): 1507–16. SUN (Scaling Up Nutrition). 2016. Institutional Transformations in 2015-16. http://docs.scalingupnutrition. org/wp-content/uploads/2016/11/Benin-SUN-Movement-Annual-Progress-Report-2016.pdf Shekar, M., J. Kakietek, J. D. Eberwein, and D. Walters. 2017. An Investment Framework for Nutrition: Reaching the Global Targets for Stunting, Anemia, Breastfeeding, and Wasting. Directions in Development. Washington, DC: World Bank. doi:10.1596/978-1-4648-1010-7. https://tinyurl.com/InvestmentFrameworkNutrition UN DESA (United Nations Department of Economic and Social Affairs). 2015. World Population Prospects, the 2015 Revision. https://esa.un.org/unpd/wpp/Download/Standard/Population/ UNDP (United Nations Development Programme). 2016. Human Development Report 2016. New York, NY: UNDP. http://hdr.undp.org/sites/default/files/2016_human_development_report.pdf UN SCN Secretariat (United Nations Standing Committee on Nutrition Secretariat). 2013. Draft UN Agencies Country Level Actions in Nutrition: Mapping of agencies’ nutrition actions in 21 countries – FAO, UNICEF, WFP, WHO and IFAD. http://unscn.org/files/Activities/SUN/NewYork2013/Final_report_Mapping_of_UN_nutri- tion_actions_in_21_SUN_countries_Sept2013.pdf WFP (World Food Programme). 2014. Republic of Benin: Global Analysis of Vulnerability and Food Security. https://www.wfp.org/content/ republique-benin-analyse-globale-vulnerabilite-securite-alimentaire-janvier-2014 World Bank. 2014. Benin Economic Update: Fall 2014. Macroeconomics and Fiscal Management, The World Bank. http://documents.worldbank.org/curated/en/388521468228911802/pdf/916550WP0Benin- 00Box385342B00PUBLIC0.pdf ———. 2016. World Bank Data Bank (database). http://data.worldbank.org/indicator/SP.RUR.TOTL. ZS?page=1 ———. 2017. World Development Indicators (database), World Bank, Washington, DC (accessed May 1, 2017), http://data.worldbank.org/data-catalog/world-development-indicators 54 in Sub-Saharan Africa Stunting Reduction  CHAPTER 4 CÔTE D’IVOIRE: An Investment Framework for Nutrition Benefits of Investing in Nutrition Key Messages • Thirty percent of Ivorian children under five suffer from chronic malnutrition (stunting); this rate has remained virtually 529,000 unchanged over the last two decades. Children in the northern regions of the country and those in poorest households are the cases of stunting prevented most affected. in 2025 • Scaling up a package of high-impact nutrition-specific interventions in Côte d’Ivoire to address global nutrition targets would require an additional $47.7 million per year over 10 years and would make an enormous health and nutrition impact (see 45,000 panel on the right). These investments are over and above those child deaths prevented in 2025 needed for improving water and sanitation and for addressing issues around women’s empowerment and food security. • This scale-up would require additional financing equivalent to an 8.3 percent increase in current government health expenditures and could be financed from a combination of domestic budgets, 6.5 MILLION official development assistance (ODA), and innovative financing case-years of anemia sources such as the Power of Nutrition.1 in women prevented in 2025 • The economic benefits generated over the productive lives of beneficiaries would be massive: $8.1 billion for the prevention of stunting, $2.5 billion for breastfeeding, $1.5 billion for the prevention of anemia, and $287 million for the treatment of severe 695,000 wasting. babies exclusively breastfed • Returns on every dollar invested in reaching the global nutrition targets range from $9 for wasting to $24 for anemia, $27 for investing in reducing stunting, and $54 for exclusive breastfeeding. 217,000 • To finance the nutrition scale-up, two lower-cost scale-up scenarios are estimated to require between $13.5 million and cases of severe wasting treated $21.3 million per year over the next 10 years. In an environment of constrained resources, starting with one of these two scenarios would be a strong first investment, but it would need to be followed by increased investments to contribute toward meeting $9–$54 the global nutrition targets. return for every dollar invested $8.1 BILLION generated from investments to reduce stunting* *The economic benefits are calculated over the productive lives of the children benefiting from the interventions that prevent stunting. CHAPTER 4 Côte d’ivoire: An Investment Framework for Nutrition 55 Investment Case for Nutrition Ensuring optimum nutrition—particularly during the 1,000-day period from pregnancy to a child’s second birthday—can alter an individual’s development trajectory and maximize her or his productive potential. Chronic malnutrition has important lifelong consequences for health and cognitive development. Losses to cognitive development in early childhood resulting from chronic malnutrition are irreversible. Being stunted (low height-for-age) in early childhood is associated with a delayed start at school, reduced schooling attain- ment and substantially decreased adult incomes at both the individual and country level (Daniels and Adair 2004; Fink et al. 2016; Hoddinott et al. 2008; Martorell et al. 2010). These consequences add up to overall gross domestic product (GDP) losses of 4 to 11 percent in Africa and Asia (Horton and Steckel 2013). Importantly, chronic undernutrition can be transmitted through an inter-generational cycle, where malnourished mothers are more likely to have stunted children (Aguayo et al. 2016; Ozaltin et al. 2010). Investments in nutrition are highly cost-effective and among the best value-for-money development actions (Copenhagen Consensus Center 2015; Hoddinott et al. 2013). An Investment Framework for Nutrition, developed by the World Bank in partnership with R4D, 1000 Days, and the Bill & Melinda Gates Foundation, estimated high returns on every dollar invested in nutrition: from $4 in returns for treating acute malnutrition (wasting) to $11 for preventing stunting, $12 for the treatment and prevention of anemia, and $35 for increasing the prev- alence of exclusive breastfeeding (Shekar et al. 2017). Not only do investments in nutrition produce substantial economic benefits, but they also lay the groundwork for the success of investments in other sectors. Investments in the early years—including early life nutrition, early learning and stimulation, and the provision of nurturing care and protection from stress—ensure that all children reach their human potential and contrib- ute to the economic growth of their nation. The analysis presented below focuses on high-impact nutrition-spe- cific interventions with strong evidence of efficacy in reducing malnutrition, and estimates the financing needs, impacts, and economic benefits of scaling up these interventions in Côte d’Ivoire. Country Context Côte d’Ivoire has a population of 22.7 million and a population growth rate of 2.4 percent (UN DESA 2015). Côte d’Ivoire has a young population, with approximately 16 percent of the country under age five (UN DESA 2015). Over 3.7 million children—or 30 percent of those under five—are affected by the largely irreversible cognitive and developmental impacts associated with stunting. Côte d’Ivoire’s economic development has been built on agriculture, but it is also emerging as an oil-rich coun- try. Agriculture accounts for 27 percent of GDP and over three-quarters of non-oil exports, and provides income for two-thirds of all households. The labor market in Côte d‘Ivoire is undergoing a transformation: the country has seen a 12 percent drop in agricultural employment between 2012 and 2015 and an increase in self and wage employment (World Bank Group 2016). After almost two decades of strong economic growth, Côte d’Ivoire experienced a series of economic and polit- ical crises (2002–07, 2010–11) which culminated in a short war following the 2010 elections. The crises resulted in widespread deterioration of living standards. Since then, Côte d’Ivoire has experienced some of the strongest economic growth on the African continent (9 percent in 2015). Despite this growth, there has been only a mar- ginal 0.3 percent decline in poverty for every 1 percent of economic growth since 2012. Nearly half of Ivorians continue to live in poverty—a rate that far exceeds that of 1985—as a result of successive economic shocks and political instability (World Bank Group 2016). A large proportion of the population lives in a state of high vulnerability without any social protection. Accord- ing to the 2016 UNDP Human Development Report, the incidence of poverty declined only marginally between 2008 and 2011, but the depth and incidence increased in a number of regions, including the North, Center North, 56 in Sub-Saharan Africa Stunting Reduction  and North East regions. Low human capital stock remains one of the key challenges to reducing poverty and achieving greater socioeconomic equity. The Human Development Index (HDI) showed Côte d’Ivoire ranked 171st out of 187 countries, with a value of 0.474 (UNDP 2016). Nutritional Status in Côte d’Ivoire Persistently high rates of undernutrition remain a serious human capital challenge in Côte d’Ivoire. One in three children (29.8 percent) under five years of age are stunted (Côte d’Ivoire DHS 2011–2012). Over the past two decades, stunting prevalence has stagnated at around 30 percent and is of medium public health significance according to WHO standards. Côte d’Ivoire ranks 87th out of 130 countries assessed for highest stunting rates for children under five (IFPRI 2016). Prevalence of wasting (low weight-for-height) was 7.5 percent among children under 5 (Côte d’Ivoire DHS 2011–2012; WHO 2010) (Figure 4.1). National estimates mask geographic and socioeconomic disparities in stunting prevalence. There continue to be significant differences in stunting prevalence among children in poorer and wealthier households, and similarly among children in rural and urban households. Stunting prevalence has remained consistently lower among chil- dren living in households in the wealthiest quintile (15.6 percent) when compared with those in poorest quintile (38.1 percent) (Figure 4.2). However, it is critical to address the key underlying drivers of malnutrition, such as high fertility rates (total fertility rate for Côte d’Ivoire is 5.0), suboptimal infant and young child nutrition prac- tices, and hygiene and sanitation at all levels. Figure 4.1: Trends in Undernutrition in Côte d’Ivoire, Figure 4.2: Disparities in Stunting by Wealth Quintile, 1994–2012 1994–2012 35 30.7 31 29.8 30 35 AMONG CHILDREN UNDER FIVE 31 AMONG CHILDREN UNDER FIVE 30.7 29.8 STUNTING PREVALENCE (%) 40 25 30 AMONG CHILDREN UNDER FIVE AMONG CHILDREN UNDER FIVE STUNTING PREVALENCE (%) 40 PREVALENCE (%) 20 25 30 PREVALENCE (%) 20 30 15 20 10.9 10 15 7.5 20 10.9 6.7 10 5 10 6.7 7.5 10 5 0 0 1994 1998-99 2011-12 1994 1998-99 2011-12 0 0 1994 1998-99 2011-12 1994 1998-99 2011-12 Poorest quintile Richest quintile National Stunting Wasting Stunting Wasting Poorest quintile Richest quintile National Source for Figures 4.1 and 4.2: Côte d’Ivoire DHS 2011–2012. This variation is pronounced at the regional level as well. While stunting prevalence is relatively low in Abi- djan (18 percent), five regions have stunting rates over 30 percent, with prevalence estimates rising to nearly 40 percent in the North and North East regions, and the highest absolute number of stunted children is con- centrated in the north and southwest regions (Figure 4.3). Given the data on regional variations in stunting prevalence, it would be important to understand the key drivers of undernutrition in these highest prevalence regions and to design targeted interventions to address them. CHAPTER 4 Côte d’ivoire: An Investment Framework for Nutrition 57 Figure 4.3: Number of Stunted Children and Stunting Prevalence by Region, Côte d’Ivoire 2011–12 DENGUELE SAVANES DENGUELE SAVANES ZANZAN ZANZAN WOROBA VALLEE DU WOROBA VALLEE DU BANDAMA BANDAMA MONTAGNES SASSANDRA- LACS MONTAGNES SASSANDRA- LACS MARAHOUE YAMOUSSOUKRO MARAHOUE YAMOUSSOUKRO GÔH- COMOE GÔH- COMOE DJIBOUA LAGUNES DJIBOUA LAGUNES BAS SASSANDRA BAS SASSANDRA Burundi 60 ABIDJAN ABIDJAN Madagascar RWANDA STUNTING PREVALENCE (%) 142,220 50 Niger Mozambique 100,000 Stunting Prevalence ≥ 40%: Very high 20–29%: Medium Total Children Chad 75,000 By Region (%) 30–39%: High <20%: Low Ethiopia Stunted By Region 50,000 Region Boundaries 40 DRC Tanzania Malawi Region Boundaries Sierra Leone 17,500 14,897 Uganda 30 Guinea Data source: Côte d’Ivoire DHS 2011–12. IBRD 42941 | MAY 2017 These maps were produced by the Cartography Unit of the World Bank Group. The boundaries, colors, IBRD 42941 | MAY 2017 Zimbabwe Senegal 20 and any other information shown on these maps do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any denominations endorsement or acceptance of such boundaries. 10 The prevalence 0 of stunting in Côte d’Ivoire is lower than the average for the Africa region (35.2 percent) Bank 2017). Nevertheless, (World500 1000 1500 higher than would it is considerably 2500 on Côte d’Ivoire’s 2000 be expected based 3000 per capita income level, which belies the fact that economic development needs to be coupled with other GDP PER CAPITA investments to attain improvements in childhood (see Figure $ (INTERNATIONAL nutrition PPP) 4.4). Figure 4.4: Prevalence of Stunting and GDP per capita: Côte d’Ivoire and Selected Low-Middle-Income Countries 50 45 Zambia STUNTING PREVALENCE (%) 40 Sudan Nigera 35 Cameroon Lesotho 30 CÔTE D’IVOIRE Swaziland 25 Kenya Mauritania 20 Congo Ghana 15 Sao Tome and Principe 10 5 0 500 1500 2500 3500 4500 5500 6500 7500 8500 9500 GDP PER CAPITA (INTERNATIONAL $ PPP) Source: UNICEF, WHO, and World Bank 2015; World Bank 2017. Wasting, also known as acute malnutrition, is typically classified as either severe or moderate. Wasting can result from food insecurity in resource-poor settings with insufficient dietary quality, quantity and diversity, subop- timal breastfeeding, and recurrent episodes of illness such as diarrhea. Wasting prevalence across Sub-Saharan NIGER Africa is second 50 highest in the world, after South Asia, with 13 million children (7.8 percent) suffering from acute malnutrition. Wasting in DRCCôte d’Ivoire is just slightly below the regional prevalence at 7.5 percent. Mozambique Chad Ethiopia ING PREVALENCE (%) 40 Rwanda Malawi Sierra Leone Tanzania Benin 58 30 Uganda in Sub-Saharan Africa Stunting Reduction  Guinea Zimbabwe Micronutrient deficiencies (a form of malnutrition that relates to a deficiency in essential vitamins and minerals needed for body functions and is sometimes referred to as hidden hunger) are highly pervasive in Côte d’Ivoire. Anemia, a condition caused by inadequate dietary intake of iron, helminth infections, and malaria, among other factors, has cross-generational impacts. More than one in two women of reproductive age in Côte d’Ivoire are anemic (Côte d’Ivoire DHS 2011–2012), which affects not only women’s own health, but also contributes to the intergenerational cycle of undernutrition. Three-quarters of children aged 6 to 59 months are anemic, with three regions experiencing an anemia prevalence of over 80 percent (Côte d’Ivoire DHS 2011–2012). Ensuring that women of reproductive age are well nourished sets the stage for their children to achieve optimal nutrition and development. Recognizing that other factors beyond poverty and food insecurity put children at risk of chronic malnutrition, there is a need for effective multisectoral strategies to address undernutrition across the country. The high prevalence of stunting in Côte d’Ivoire is associated with insufficient access to health services, poor water and sanitation, and suboptimal care and feeding practices. Demand- and supply-side barriers influence food consumption and diversity of diet. A survey conducted in 2015 found that households in rural areas spent 56 percent of their income on food, compared with 39 percent in urban areas (National Institute of Statistics 2015), thereby persistently limiting access to the range of nutrients required for growth, health, and devel- opment during the early years and beyond. Food insecurity affects 12.8 percent of the population (Ministry of Planning and Development 2015) and an estimated 20–40 percent of the population are not meeting the minimum recommended caloric intake (IMF 2009). In post-harvest and lean seasons, 72 percent of households reported food-related coping strategies including reductions in number of daily meals and consumption of non-diversified diets (Ministry of Agriculture 2016). Political Commitment to Reduce Malnutrition Political commitment to nutrition is very high in Côte d’Ivoire, with the Prime Minister and the Finance Minister both highly committed to the agenda. In 2013, Côte d’Ivoire joined the SUN movement, solidifying this commitment to nutrition. In 2016, a council of ministers adopted the national Multisectoral Nutrition Plan (2016–2020). This national plan identifies chronic malnutrition as a key priority for improving human and economic development in Côte d’Ivoire. The plan lays the groundwork for cross-sectoral collaboration across seven strategic areas and prioritizes the scale-up of nutrition activities in support of the Sustainable Devel- opment Goals (Ministry of Health and Public Hygiene 2016). Côte d’Ivoire has developed a common policy framework for nutrition, incorporating a common budget and results framework for both the National Nutri- tion Policy and the Multisectoral Nutrition Plan. The Prime Minister launched the plan in September 2016 and is mobilizing financing for its implementation. The commitment to nutrition is also reflected in the 2016–2020 National Development Plan, which lays out tar- gets for reducing chronic malnutrition to 20 percent and wasting to 5 percent by 2020 (Ministry of Health and Public Hygiene 2016). This commitment builds upon the 2012–2015 National Development Plan, which pro- moted action on food security (Côte d’Ivoire, Republic of 2012). A multisectoral Nutrition Council was estab- lished in 2014 and is chaired by the Prime Minister. The Council engages almost a dozen ministries, reflecting the multisectoral nature of food and nutrition policies and programs. Current Financing for Nutrition In 2015 in Côte d’Ivoire, the government and overseas donors spent a total of $4.41 million on interventions that will contribute to reaching the global targets for nutrition. Of that amount, $1.60 million came from the government and $2.82 million came from ODA. This contribution from ODA included $799,000 for stunting, $97,000 for anemia, $190,000 for breastfeeding, and $1.89 million for wasting.2 These estimates reflect the current spending on nutrition, and the following sections detail additional financing needed in order for Côte d’Ivoire to contribute to reaching the global targets on nutrition.3 CHAPTER 4 Côte d’ivoire: An Investment Framework for Nutrition 59 Global Targets for Nutrition Substantial improvements to the nutritional status of women and children can be realized if there is adequate investment in a set of evidence-based nutrition-specific interventions that ensure optimum nutrition during the critical 1,000 day window between the start of a woman’s pregnancy and the child’s second birthday (Black et al. 2008, 2013). For women, these include interventions to prevent anemia before and during pregnancy as well as those aimed at improving protein energy intake and adequate antenatal care. Interventions targeted at children and their mothers aim to improve breastfeeding and complementary feeding practices, enhance the micronutrient status of children, and treat acute malnutrition in children. In 2012—to rally the international community around improving nutrition—the 176 members of the World Health Assembly endorsed the first-ever global nutrition targets, focusing on six areas: stunting, anemia, low birthweight, childhood overweight, breastfeeding, and wasting. These targets aim to boost investments in cost-effective interventions, spearhead better implementation practices, and catalyze progress toward reducing malnutrition. The targets for stunting and wasting are enshrined within the United Nations’ Sustainable Devel- opment Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. The 2016 Global Nutrition Report ranked each country’s progress in contributing toward achieving the global targets (Table 4.1) (IFPRI 2016).4 Table 4.1: Four Global Targets for Nutrition and Côte d’Ivoire’s Contribution toward Meeting Them CÔTE D’IVOIRE PREVALENCE PROGRESS RANK Reduce the number of stunted children 1 STUNTING* 87/132 29.8% under five by 40% Stuntin Reduce the number of women of 2 ANEMIA 173/185 53.7% reproductive age with anemia by 50% An mi Increase the rate of exclusive breastfeeding  5 BREASTFEEDING 126/141 12.1% in the first six months up to at least 50% Exclusiv br stf din Reduce and maintain childhood wasting 6 WASTING* 92/130 7.5%  alnutrition) to less than 5% (acute m W stin LEGEND: Off course, no progress Off course, some progress On course, good progress Sources: Nutrition targets from WHO 2014; Rank and progress from IFPRI 2016; Prevalence data from Cote d’Ivoire DHS 2011-12. *Stunting and wasting are included within the United Nations’ Sustainable Development Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. Coverage of key nutrition-specific interventions in Côte d’Ivoire is largely inadequate, and is well below the levels necessary to advance progress in reducing malnutrition among Ivorian children. Table 4.2 and Figure 4.5 summarize the current coverage of and delivery platforms for nutrition-specific interventions in Côte d’Ivoire. 60 in Sub-Saharan Africa Stunting Reduction  Table 4.2: Delivery Platforms of Nutrition-Specific Interventions in Côte d’Ivoire INTERVENTION PLATFORM Antenatal micronutrient supplementation - (iron and folic acid only) Health facility and community Complementary feeding education Health facility, community, and communication campaigns Breastfeeding promotion Health facility, community, and community campaigns Balanced energy protein supplementation for pregnant women Health facility, community, and social safety net programs Intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions Health facility and community Vitamin A supplementation Health facility, community, and food fortification Public provision of complementary foods Health facility, community, and social safety net programs Treatment of severe acute malnutrition Health facility and community Iron and folic acid supplementation for non-pregnant women of reproductive age School, community, health facility, and marketplace Staple food fortification Marketplace Pro-breastfeeding social policies Government policies National breastfeeding promotion campaigns Media Figure 4.5: Coverage of Key Nutrition-Specific Interventions: Côte d’Ivoire and Sub-Saharan Africa Antenatal micronutrient supplementation - (iron and folic acid only) Pro-breastfeeding Complementary social policies feeding education 100 80 60 Staple food Breastfeeding fortification 40 promotion 20 Iron and folic acid Balanced energy supplementation for non-pregnant protein supplementation women of reproductive age for pregnant women Treatment of severe Intermittent presumptive acute malnutrition treatment of malaria in pregnancy in malaria-endemic regions Public provision of Vitamin A complementary foods supplementation Note: Red shading represents Côte d’Ivoire and light blue shading represents average Sub-Saharan Africa coverage CHAPTER 4 Côte d’ivoire: An Investment Framework for Nutrition 61 Economic Benefits of Investing in Nutrition There is a strong body of evidence that shows high economic returns to investing in nutrition (Alderman et al. 2016; Copenhagen Consensus Center 2015; Hoddinott et al. 2013). Scaling up these proven nutrition-specific interventions can ensure that mothers are healthy and well nourished, that they can provide optimal nutrition to their children, that children realize their full physical and cognitive development potential, and that wom- en’s productivity is not hampered by illness, especially anemia (Figure 4.6). Figure 4.6: How Reaching the Global Nutrition Targets Generates Economic Benefits Co nitive Development STUNTING Childr n’s H lth & Le rnin & Educ tion l Nutrition St tus Att inment WASTING Glob l Nutrition T r ts R ch d Adult Productivit EASTFEE BRE EDING W es ANEMIA Econom Wom n’s H lth & (GDP) Nutrition St tus In Côte d’Ivoire, scaling-up the package nutrition-specific interventions would produce substantial economic benefits over the productive lifetime of the affected women and children (Figure 4.7). Additional health system cost-savings would also be likely because many of these investments reduce the burden of childhood illnesses such as diarrhea and pneumonia. Figure 4.7: Investments in Côte d’Ivoire to Meet the Global Nutrition Targets Have Enormous Economic Returns Total Economic $1 Invested Benefits (Millions)a Yieldsb STUNTING $8,112 $27 $ $2,529 $54 1 BREASTFEEDING $ $ ANEMIA $1,547 $ $24 WASTING $286.6 $9 a. Total economic benefits over 10 years for women and over the productive lives of children who benefit from these interventions, defined as the period between the age of 18 and a “retirement” age - the life expectancy or the age of 65, whichever is lower. b. Benefit calculation assumes a 3 percent discount rate for both financing needs and benefits and a GDP growth rate of 3 percent. 62 in Sub-Saharan Africa Stunting Reduction  Financing Needs, Impacts, and Cost-Effectiveness of Scaling-Up Nutrition-Specific Interventions Using the methodology detailed in An Investment Framework for Nutrition (Shekar et al. 2017), this brief presents estimates of the resources needed to scale up a package of 12 high-impact nutrition-specific interventions in Côte d’Ivoire to meet the global nutrition targets for stunting, anemia, breastfeeding, and wasting, along with their estimated nutrition, health, and economic impacts. An additional $47.7 million per year over 10 years is needed to scale up the package of key interventions (Table 4.3). The health and nutrition impacts of this invest- ment are shown in Table 4.4. Table 4.3: Estimated 10-Year Financing Needs and Cost-Effectiveness of Scaling Up Nutrition-Specific Interventions, Côte d’Ivoire TOTAL 10-YEAR COST PER CASE INTERVENTION COST PER DEATH FINANCING NEEDS OF STUNTING (NUTRITION TARGET) AVERTED (US $) (US $M) AVERTED (US $) For pregnant women and mothers of infants Antenatal micronutrient supplementation (stunting, anemia) 24.7 8,395 4,173 Infant and young child nutrition counseling (complementary feeding 22.7 2,017 125 education and breastfeeding promotion combined) (stunting, breastfeeding) Complementary feeding education (stunting) 11.3 2,269 63 Breastfeeding promotion (stunting, breastfeeding) 11.5 1,820 3,302 Balanced energy protein supplementation for pregnant women (stunting) 56.7 20,551 10,647 Intermittent presumptive treatment of malaria in pregnancy (stunting, 14.6 2,294 1,087 anemia) For infants and young children Vitamin A supplementation (stunting) 5.3 1,630 513 Prophylactic zinc supplementation (stunting) 76.4 6,766 489 Public provision of complementary food (stunting) 181.0 66,705 1,763 Treatment of severe acute malnutrition (wasting) 39.2 10,184 n.a For non-pregnant women and general population Iron and folic acid supplementation for non-pregnant women (anemia) 22.7 n.a 27,855 Staple food fortification (anemia) 8.2 n.a Pro-breastfeeding social policies (breastfeeding) 5.0 n.a n.a National breastfeeding promotion campaigns (breastfeeding) 20.0 n.a n.a TOTAL: 476.6 10,462 801 Note: Financing needs and impacts assume a linear scale-up of interventions from current coverage level to 90 percent over five years, then maintained at 90 percent for an additional five years. Unit costs for each intervention were drawn from available unit costs from neighboring countries, global costs, or estimates available in the literature. The estimated financing needs include an additional 12 percent (11 percent for pro-breastfeeding social policies and promotion campaigns) to account for monitoring, evalua- tion, capacity, and policy development that may be necessary to reach full scale-up of the interventions. The Lives Saved Tool (LiST; see LiST 2015) was used to estimate the impact of interventions that target pregnant women and children. The impacts of interventions that target the general population or non-pregnant women were estimated using a Microsoft Excel model. It should be noted that the LiST model does not capture potential synergies between specific interventions (e.g. the fact that the impact of behavior change communication interventions may be higher in populations that have access to affordable and diversified foods or in populations with higher levels of educa- tional attainment). Therefore, it is possible that the impact estimates generated using LiST in fact underestimate the true impact of the interventions in some contexts. n.a. = not applicable. CHAPTER 4 Côte d’ivoire: An Investment Framework for Nutrition 63 ANEMIA Total: 5% Total 11%: BREASTFEEDING Total: 11% Among the set of proposed interventions, educating mothers about complementary feeding is the most effec- tive for preventing stunting, averting more than half a million cases of stunting and 40,000 child deaths over 10 years. Breastfeeding promotion through counseling mothers is projected to increase TOTAL the number of infants TEN-YEAR Benin STUNTING exclusively breastfed million over 10NEEDS: by 69,000, prevent over 6,000 deaths, and cost $11.5 FINANCING years. For preventing maternal anemia,Total 82%: staple food fortification proves to be the most cost-effective, at a cost of $5.50 for each case- year of anemia prevented in women. Over 10 years, staple food fortification $309 million will prevent nearly 1.5 million case-years of anemia in non-pregnant women at a cost of $8.2 million. Among pregnant women, antenatal micronutrient supplementation will prevent 1.5 million case-years of anemia, at a cost of $16.5 per case averted, or a total of $24.7 million over 10 years. Interventions to reduce stunting will require the most resources, accounting for 80 percent of the total amount required for scale-up. However, some of the stunting interventions will also have impacts for the breastfeeding and anemia targets. Figure 4.8 breaks down the distribution of total financing needs across the four targets. Figure 4.8: Ten-Year Financing Needs for Scaling Up a Package of Nutrition-Specific Interventions in Côte d’Ivoire, by Percent per Intervention WASTING ANEMIA Total: 8% Total 11%: BREASTFEEDING Total: 8% Cote STUNTING TOTAL TEN-YEAR Total 80%: FINANCING NEEDS: $477 million Note: Some costs for anemia, breastfeeding, and stunting are shared across interventions. Costs for breastfeeding promotion ($11.5 million) have been included in both the total cost for the breastfeeding target and the total cost for the stunting target; the costs of intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions ($14.6 million) and antenatal micronutrient supplementation ($24.7 million) have been included in both the total cost for the anemia target and the total cost for the stunting target The analysis above includes only interventions related to the four global targets. Two additional high impact interventions aimed at improving micronutrient status, namely zinc and oral rehydration solute (ORS) for the treatment of diarrhea and deworming, will be scaled up as part of the Cote d’Ivoire’s National Multisectoral Nutrition Plan. The World Bank had previously conducted analysis of the cost and impact of two of these inter- ventions in the context of the Plan. Over a five year period, the scale-up of therapeutic zinc coverage would cost an additional $39.4 million and save 7,700 additional lives, while the scale-up of deworming would cost $1.7 million and result in an additional 6.9 million children being dewormed. Two Alternative Investment Packages Relative to current expenditures on health, the investment required to scale up the set of effective nutrition-spe- cific interventions may present significant challenges for Côte d’Ivoire. In an environment of constrained resources in which Côte d’Ivoire may not be able to raise $477 million over the next 10 years, two alternative investment packages are laid out for consideration. 64 in Sub-Saharan Africa Stunting Reduction  Table 4.4: Benefits and Cost-Effectiveness by Investment Package, Côte d’Ivoire GLOBAL PRIORITY CATALYZING FULL PACKAGE: BENEFIT All interventions needed TARGET PACKAGE PROGRESS PACKAGE to meet targets $13.5 million/year in $21.3 million/year $47.7 million/year in financing need financing need financing need Cases of stunting reduced by STUNTING 253,000 330,000 529,000 2025 (vs 2015)a Cases of anemia in women ANEMIA 2 million 2.6 million 6.5 million prevented by 2025 Additional babies BREASTFEEDING 695,000 695,000 695,000 breastfed over 10 years Child deaths averted 34,000 39,000 45,000 over 10 years ALL TARGETS Cost per death averted 4,019 5,458 10,463 Cost per case of 266 489 801 stunting averted a. Total impacts of proposed intervention package combined with other health and poverty reduction efforts. The Priority Package: The first—the “priority package”—includes interventions that are the most cost-effec- tive; that is, have the lowest cost per health outcome (e.g., case of stunting averted), and that have well-es- tablished global policy guidelines and delivery platforms. Based on those two criteria, the priority package includes antenatal micronutrient supplementation, infant and young child nutrition counseling, intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions, vitamin A supplementation, the treatment of severe acute malnutrition, weekly iron and folic acid supplementation for girls 15–19 years of age attending school, and fortification of wheat and maize flour with iron and folic acid. These interventions would be scaled up to full program coverage in the first five years and maintained at full coverage levels for the last five years. This priority package would require an estimated $135 million over 10 years, or $13.5 million annu- ally (see Table 4.4). During the 10 years of scale up, this package would prevent more than 253,000 cases of stunting and avert 34,000 deaths in children under five years of age. It would also prevent nearly 2 million case-years of anemia in women and would result in 695,000 additional children under six months of age being exclusively breastfed. The Catalyzing Progress Package: The second alternative—the “catalyzing progress package”—includes scale-up of all interventions in the priority package, plus a phased approach to scaling up public provision of complementary foods, balanced energy protein supplementation, prophylactic zinc supplementation, and weekly iron and folic acid supplementation for women outside of schools. It is assumed that, for the latter set of interventions, during the first five years, emphasis will be placed on establishing global guidelines and on operational research to develop effective delivery platforms, or to develop less expensive products or more cost-effective technologies. Costs are approximated as the cost of scaling up this set of interventions from 0 to 10 percent coverage only in the first five years. In the subsequent five years, it is assumed that the coverage expansion of those interventions will accelerate and reach 60 percent by 2025. This package would require $21.3 million per year, or a total of $213 million over 10 years (Table 4.4). It would prevent 39,000 deaths and more CHAPTER 4 Côte d’ivoire: An Investment Framework for Nutrition 65 than 330,000 cases of stunting among children under age five, increase the number of exclusively breastfed chil- dren under six months of age by 695,000, and prevent more than 2.6 million case-years of anemia in women. In comparing the relative cost-effectiveness of the three intervention packages, the two alternative packages are more cost-effective in preventing deaths and stunting. However, neither is as effective as the full package in making progress toward achieving the stunting, wasting, and anemia targets. The priority and catalyzing prog- ress packages would prevent 34,000 and 39,000 deaths respectively, compared with 45,000 deaths prevented with the full package over 10 years. Under the full package scenario, 529,000 cases of childhood stunting would be prevented, compared with 330,000 cases under the catalyzing progress scenario and 253,000 cases under the priority package scenario. Furthermore, there would be nearly 4.5 million and 3.9 million more case-years of anemia prevented in women under the priority package and catalyzing progress package, respectively. A Call to Action As the world stands on the cusp of the new Sustainable Development Goals, there is an unprecedented oppor- tunity to save children’s lives, build future human capital and cognitive development, and drive faster eco- nomic growth. Scaling up key nutrition interventions during the critical 1,000 day window of early childhood will pay lifelong dividends, translating to healthier societies and more robust economies. If this window is missed, it is missed for life. The additional financing required to reach the global nutrition targets will require coordinated efforts by all stakeholders and a supportive policy environment. To achieve these targets, Côte d’Ivoire will need to increase the funding allocated to nutrition by $47.7 million annually, roughly equivalent to an 8.3 percent increase in current general government expenditure on health.5 These investments are over and above those needed for improving water and sanitation and issues around women’s empowerment, and food security. Although this level of domestic financing is ambitious, Côte d’Ivoire is already moving in this direction. In the long term, nutrition interventions have significant potential to reduce poverty and boost shared prosperity. Accelerating the reduction of stunting in Côte d’Ivoire will be essential for maximizing the return on invest- ments in early childhood development, in education, and more broadly in policies aimed at fostering and enhancing human capital accumulation and job creation. Investing in the early years is even more critical because the Africa region is entering a demographic transition with an expected increase in the working-age population from 54 percent in 2010 to 64 percent by 2090. The scale-up of the key nutrition-specific interven- tions to reduce stunting is estimated to generate considerable returns in economic benefits over the productive lives of beneficiaries, and is a necessary condition to build human capital through investments in the early years and to harness the potential benefits of the demographic dividend. Endnotes Note: All dollar amounts are U.S. dollars unless otherwise indicated. 1 Information about the Power of Nutrition initiative is available at https://ciff.org/grant-portfolio/ the-power-of-nutrition/. 2 Note that because some funded interventions contribute to more than one target, total funding across the four targets is less than the total funding for each target added together. 3 Current financing by source is from Results for Development Institute and can be found at http://www. investinnutrition.org/. 66 in Sub-Saharan Africa Stunting Reduction  4 Two of the global nutrition targets—those for low birthweight and for child overweight—were not included in the analyses because of insufficient data on the prevalence of low birthweight and a lack of consensus on effective interventions to reach the target for child overweight. 5 WHO National Health accounts database indicates general government health expenditure in Cote d’Ivoire was US $575m in 2014. At that level, this will need to be increased by 8.3% to accommodate the US $47.7m per year required for scale-up of the 12 nutrition-specific interventions. References Aguayo, V. M., R. Nair, N. Badgaiyan, and V. Krishna. 2016. “Determinants of Stunting and Poor Linear Growth in Children under 2 Years of Age in India: An In-Depth Analysis of Maharashtra’s Comprehensive Nutrition Survey.” Maternal & Child Nutrition 12 (Suppl. 1): 121–40. Alderman , H ., J. R. Behrman , and C. Puett . 2016 . “ Big Numbers about Small Children: Estimating the Eco- nomic Benefits of Addressing Undernutrition .” World Bank Research Observer 31 ( 2 ). Black, R. E., L. H. Allen, Z. A. Bhutta, L. E. Caulfield, M. de Onis, M. Ezzati, C. Mathers, J. Rivera, and the Maternal and Child Undernutrition Study Group. 2008. “Maternal and Child Undernutrition: Global and Regional Exposures and Health Consequences.” The Lancet 371 (9608): 243–60. Black, R. E., C. G .Victora, S. P. Walker, Z. A. Bhutta, P. Christian, M. de Onis, M. Ezzati, S. Grantham-Mcgregor, J. Katz, R. Martorell, R. Uauy, and the Maternal and Child Nutrition Study Group. 2013. “Maternal and Child Undernutrition and Overweight in Low-Income and Middle-Income Countries.” The Lancet 382: 427–51. Copenhagen Consensus Center. 2015. Smart Development Goals: The Post-2015 Consensus. http://www.copenha- genconsensus.com/sites/default/files/outcomedocument_col.pdf Côte d’Ivoire, Republic of. 2012. National Development Plan 2012–2015. Côte d’Ivoire DHS (Côte d’Ivoire Demographic and Health Survey). 2011–2012. Côte d’Ivoire Demographic and Health Survey: Final Report 2011–2012. Calverton, Maryland, USA: INS/Côte d’Ivoire and ICF International (in French; published in 2013). http://dhsprogram.com/publications/publication-FR272-DHS-Final-Reports.cfm Daniels, M. C. and L. Adair. 2004. “Growth in Young Filipino Children Predicts Schooling Trajectories through High School.” Journal of Nutrition 134: 1439–46. Fink, G., E. Peet, G. Danaei, K. Andrews, D. C. McCoy, C. R. Sudfeld, M. C. Smith Fawzi, M. Ezzati, and W. W. Fawzi. 2016. “Schooling and Wage Income Losses Due to Early-Childhood Growth Faltering in Developing Countries: National, Regional, and Global Estimates.” The American Journal of Clinical Nutrition 104 (1): 104–12. Hoddinott, J., H. Alderman, J. R. Behrman, L. Haddad, and S. Horton. 2013. “The Economic Rationale for Investing in Stunting Reduction.” Maternal and Child Nutrition 9 (Suppl. 2): 69–82. Hoddinott, J., J. A. Maluccio, J. R. Behman, R. Flores, and R. Martorell. 2008. “Effect of a Nutrition Intervention during Early Childhood on Economic Productivity in Guatemalan Adults.” Lancet 371 (9610): 411–16. Horton, S. and R. Steckel. 2013. “Malnutrition: Global Economic Losses Attributable to Malnutrition 1900–2000 and Projections to 2050.” In The Economics of Human Challenges, edited by B. Lomborg, 247–72. Cambridge, U.K.: Cambridge University Press. CHAPTER 4 Côte d’ivoire: An Investment Framework for Nutrition 67 IFPRI (International Food Policy Research Institute). 2016. Global Nutrition Report 2016: From Promise to Impact – Ending Malnutrition by 2030. Washington, DC: IFPRI. IMF (International Monetary Fund). 2009. “Côte d’Ivoire: Poverty Reduction Strategy Paper.” IMF Country Report No. 09/156. Washington, DC: IMF. https://www.imf.org/external/pubs/ft/scr/2009/cr09156.pdf LiST (Lives Saved Tool). 2015. Baltimore, MD: Johns Hopkins Bloomberg School of Public Health. http:/livessa- vedtool.org/ Martorell, R., B. L. Horta, L. S. Adair, A. D. Stein, L. Richter, C. H. D. Fall, S. K. Bhargava, S. K. Dey Biswas, L. Perez, F. C. Barros, C. G. Victora, and Consortium on Health Orientated Research in Transitional Societies Group. 2010. “Weight Gain in the First Two Years of Life Is an Important Predictor of Schooling Outcomes in Pooled Analyses from Five Birth Cohorts from Low- and Middle-Income Countries.” Journal of Nutrition 140: 348–54. Ministry of Agriculture, Côte d’Ivoire. 2016. Suivi de la saison agricole et de la vulnérabilité alimentaire. Côte d’Ivoire. Ministry of Health and Public Hygiene, Côte d’Ivoire. 2016. National Multisectoral Nutrition Plan 2016–2020 (in French). http://www.nutrition.gouv.ci/fichier/doc/PNMN_2016_2020_15_08_16.pdf Ministry of Planning and Development, Côte d’Ivoire. 2015. National Household Living Conditions in Côte d’Ivoire. Côte d’Ivoire. National Institute of Statistics. 2015. Enquête sur le niveau de vie des ménages en Côte d’Ivoire. Ozaltin, E., K. Hill, and S. V. Subramanian. 2010. “Association of Maternal Stature with Offspring Mortality, Underweight, and Stunting in Low- to Middle-Income Countries.” JAMA 303 (15): 1507–16. Shekar, M., J. Kakietek, J. D. Eberwein, and D. Walters. 2017. An Investment Framework for Nutrition: Reaching the Global Targets for Stunting, Anemia, Breastfeeding, and Wasting. Directions in Development. Washington, DC: World Bank. doi:10.1596/978-1-4648-1010-7. https://tinyurl.com/InvestmentFrameworkNutrition UN DESA (United Nations Department of Economic and Social Affairs). 2015. World Population Prospects, the 2015 Revision. https://esa.un.org/unpd/wpp/Download/Standard/Population/ UNDP (United Nations Development Programme). 2016. Human Development Report 2016. New York, NY: UNDP. http://hdr.undp.org/sites/default/files/2016_human_development_report.pdf UNICEF, WHO, and World Bank (United Nations Children’s Fund, World Health Organization and World Bank). 2015. Joint Malnutrition Estimates. Global Database on Child Growth and Malnutrition, http://www. who.int/growthdb/estimates2014/en/ WHO (World Health Organization). 2010. Nutrition Landscape Information System – Country Profile Indica- tors: Interpretation Guide. http://www.who.int/nutrition/nlis_interpretationguide_isbn9789241599955/en/ World Bank, World Development Indicators. 2015. Africa region average for 2015 World Bank. 2017. World Development Indicators (database), World Bank, Washington, DC (accessed May 1, 2017), http://data.worldbank.org/data-catalog/world-development-indicators World Bank Group. 2016. Côte d’Ivoire Economic Update. http://documents.worldbank.org/curated/ en/809921468000024657/pdf/102021-v1-FRENCH-WP-PUBLIC-Box394824B.pdf 68 in Sub-Saharan Africa Stunting Reduction  CHAPTER 5 ETHIOPIA: An Investment Framework for Nutrition Benefits of Investing in Nutrition Key Messages • Thirty-eight percent of children in Ethiopia are stunted. Those most affected live in northern regions and in the poorest households. Over the past 15 years, those in the 1.6 MILLION highest wealth quintiles experienced the fastest rates of cases of stunting prevented declines in stunting. in 2025 • Scaling up a package of high-impact nutrition-specific interventions in Ethiopia to address global nutrition targets would require an additional $220 million per year for 10 years and would provide enormous benefits (see 130,000 panel on right). These investments are over and above child deaths prevented in 2025 those needed for improving water and sanitation, and for addressing issues around women’s empowerment and food security. • This investment would require additional financing equivalent to a 15 percent increase in current government 14.4 MILLION health expenditures and could be financed from a case-years of anemia combination of domestic budgets, official development in women prevented in 2025 assistance (ODA), and innovative financing sources such as the Power of Nutrition.1 • The economic benefits generated over the productive lives of beneficiaries would be enormous: $28 billion 1.7 MILLION for stunting, $4 billion for anemia, $12 billion for additional babies exclusively breastfeeding, and $3 billion for the treatment of severe wasting. breastfed • Returns on every dollar invested in this set of interventions range from $10 for wasting to $18 for anemia, $21 for stunting, and $66 for investing in exclusive breastfeeding. 2.7 MILLION • To finance the nutrition scale-up, two lower-cost cases of severe wasting treated scale-up scenarios are estimated to require between $100 and $140 million per year over the next 10 years. In an environment of constrained resources, starting with one of these two scenarios would be a strong first $10-$66 investment, but it would need to be followed by increased investments to contribute to meeting the global nutrition return for every dollar invested targets and the Government of Ethiopia’s pledge to end child undernutrition by 2030 as laid out in the Segota Declaration. $28 BILLION generated from investments to reduce stunting* *The economic benefits are calculated over the productive lives of the children benefiting from the interventions that prevent stunting. CHAPTER 5 Ethiopia: An Investment Framework for Nutrition 69 Investment Case for Nutrition Ensuring optimum nutrition—particularly during the 1,000-day period from pregnancy to a child’s second birthday—can alter an individual’s development trajectory and maximize her or his productive potential. Chronic malnutrition has important lifelong consequences for health and cognitive development. Losses to cognitive development in early childhood resulting from chronic malnutrition are irreversible. Being stunted (low height-for-age) in early childhood is associated with a delayed start at school, reduced schooling attain- ment and substantially decreased adult incomes at both the individual and country level (Daniels and Adair 2004; Fink et al. 2016; Hoddinott et al. 2008; Martorell et al. 2010). These consequences add up to overall GDP losses of 4 to 11 percent in Africa and Asia (Horton and Steckel 2013). Importantly, chronic undernutrition can be transmitted through an inter-generational cycle, where malnourished mothers are more likely to have stunted children (Aguayo et al. 2016; Ozaltin et al. 2010). Investments in nutrition are highly cost-effective and among the best value-for-money development actions (Copenhagen Consensus Center 2015; Hoddinott et al. 2013). An Investment Framework for Nutrition developed by the World Bank in partnership with R4D, 1000 Days, and the Bill and Melinda Gates Foundation estimated high returns on every dollar invested in nutrition: from $4 in returns for treating acute malnutrition (wasting) to $11 for preventing stunting, $12 for the treatment and prevention of anemia, and $35 for increasing the prev- alence of exclusive breastfeeding (Shekar et al. 2017). Not only do investments in nutrition produce substantial economic benefits, but they also lay the groundwork for the success of investments in other sectors. Investments in the early years including early life nutrition, early learning and stimulation, and nurturing care and protection from stress ensure that all children reach their human potential and contribute to the economic growth of their nation. The analysis presented below focuses on high-impact nutrition-specific interventions with strong evidence of efficacy in reducing malnutrition, and estimates the costs, impact, and economic bene- fits of scaling up these interventions in Ethiopia. Country Context Ethiopia is the second-most populous country in Sub-Saharan Africa, with a population of almost 100 million and a population growth rate of 2.5 percent. Ethiopia has a young population, with approximately 15 percent of the country under age five (UN DESA 2015). More than 80 percent of the population resides in rural areas, and agriculture accounts for 77 percent of employ- ment in Ethiopia. Ethiopia’s economy is among the fastest growing in the world, despite slowed growth in 2015–2016 due to the recent drought. The Government of Ethiopia is implementing its National Second Growth and Transformation Plan (GTP II) for the period 2015/16 – 2019/20, outlining the path for Ethiopia to become a lower-middle-income country by 2025, with a focus on sustained economic growth, job creation, promotion of women and youth empowerment, and human development (World Bank 2016). Ethiopia has achieved substantial progress in economic, social, and human development over the past decade. Extreme poverty incidence has fallen dramatically within less than two decades (US$1.90 PPP poverty line), from 60.5 percent in 1996 to 33 percent in 2011, driven by improvements in economic growth and provision of safety nets and basic services (World Bank 2016). Additionally total fertility rate has fallen dramatically from 7.0 in the mid-1990’s to the current 4.3, which is below the average for Sub-Saharan Africa (4.9). With agriculture underlying most livelihoods, the population is highly vulnerable to weather shocks and food insecurity, and malnutrition remains high (MOFED 2011). 70 in Sub-Saharan Africa Stunting Reduction  Nutritional Status in Ethiopia Persistently high rates of undernutrition remain a serious human development challenge in Ethiopia. More than one-in-three children (38.4 percent) under five years of age are stunted and 10 percent are wasted (low weight-for-height) (EDHS 2016). Since 2000, stunting prevalence has declined but remains of high public health significance according to WHO standards. Furthermore, Ethiopia is among the 10 countries globally with the largest numbers of children under five who suffer from acute malnutrition (wasting) (Sanchez-Montero et al. 2010). Prevalence of wasting has slowly declined, with a slight increase documented between 2014 and 2016, likely due to the recent drought (Figure 5.1). Figure 5.1: Trends in Undernutrition in Ethiopia, 2000–2016 Figure 5.2: Disparities in Stunting by Wealth Quintile, 2000–2016 70 70 70 57.4 AMONG CHILDREN UNDER FIVE AMONG CHILDREN UNDER FIVE 6057.4 70 (%) (%) UNDER FIVE UNDER FIVE 60 50.7 60 STUNTING PREVALENCE (%) STUNTING PREVALENCE (%) 50.7 60 PREVALENCE PREVALENCE 50 44.2 44.2 40.4 50 50 40.4 38.4 50 40 38.4 40 40 40 CHILDREN CHILDREN 30 26.6 30 26.6 23 30 STUNTING STUNTING 30 23 16.6 20 20 16.6 20 20 10 AMONG AMONG 10 10 10 12.3 0 10.1 8.7 9.9 0 0 12.3 10.1 9.9 0 8.7 2000 2005 2011 2014 2016 2000 2005 2011 2014 2016 2000 2005 2011 2014 2016 2000 2005 2011 2014 2016 Stuntin W stin An mi Poor st Quintil Rich st quintil N tion l Stuntin W stin An mi Poor st Quintil Rich st quintil N tion l Source for Figures 5.1 and 5.2: EDHS 2016. National estimates mask geographic and socioeconomic disparities in stunting prevalence. There continue to be significant differences in stunting prevalence among children in poorer and wealthier households. Stunt- ing declined at a faster rate among children living in households in the top wealth quintile when compared to those in the bottom quintile (Figure 5.2). Nevertheless, even in the richest households, stunting still remains high (27 percent), underscoring the fact that rising income alone are insufficient to eliminate malnutrition. The highest stunting prevalence rates are concentrated in the northern regions of the country, however much of the country carries a high burden in terms of absolute number of children stunted (Figure 5.3). Four regions have stunting prevalence of over 40 percent, and as many as half of children under five years of age were stunted in the Amhara region. Given the regional variations in stunting prevalence, it would be important to understand the key drivers of undernutrition in these highest prevalence regions, and to design targeted inter- ventions to address them. CHAPTER 5 Ethiopia: An Investment Framework for Nutrition 71 Figure 5.3: Number of Stunted Children and Stunting Prevalence by Region, TIGRAY TIGRAY r the AFAR AFAR ries. BENSHANGUL AMHARA BENSHANGUL AMHARA DIRE DAWA DIRE DAWA ADDIS HARARI ADDIS HARARI ABABA ABABA OROMIYA OROMIYA GAMBELA SOMALI GAMBELA SOMALI SNNP SNNP 1,951,121 350,000 Stunting Prevalence ≥ 40%: Very high 20–29%: Medium Total Children 115,000 Region Boundaries By Region (%) 30–39%: High <20%: Low Stunted By Region 75,000 25,000 Region Boundaries 11,612 Data source: EDHS 2016. IBRD 42913 and IBRD 42912 | MAY 2017 These maps were produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denominations and any other information shown on these maps do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. The prevalence of stunting in Ethiopia is higher than the average for the region (35.2 percent) (World Bank 2017). It is significantly higher than in African countries with similar income levels, such as Zimbabwe and Uganda, indicating that it is possible to achieve better nutrition outcomes at this income level (Figure 5.4). Figure 5.4: Prevalence of Stunting and GDP per capita: Ethiopia and Selected Low-Income Countries 70 Burundi 60 50 STUNTING PREVALENCE (%) Niger Mozambique ETHOPIA Chad 40 DRC Sierra Leone Malawi Uganda Benin 30 Guinea-Bissau Zimbabwe Togo The Gambia Senegal 20 10 0 500 700 900 1100 1300 1500 1700 1900 2100 2300 2500 GDP PER CAPITA (INTERNATIONAL $ PPP) EDHS 2016 and World Bank 2017. Wasting, also known as acute malnutrition, is typically classified as either severe or moderate. Wasting can result from food insecurity in resource-poor settings with insufficient dietary quality, quantity and diversity, suboptimal breastfeeding, and recurrent episodes of illness such as diarrhea. Wasting prevalence across Sub-Sa- haran Africa is second highest in the world, after South Asia, with 13 million children (7.8 percent) suffering from acute malnutrition. Ethiopia’s wasting prevalence is higher than the regional average at nearly 10 percent in 2016, and one of the highest in the world. Thus, treatment of acute malnutrition, as well as efforts to better understand and address the drivers of wasting, are evermore critical to reversing this trend. 72 in Sub-Saharan Africa Stunting Reduction  Micronutrient deficiencies, a form of malnutrition that relates to a deficiency in essential vitamins and minerals needed for body functions, also known as hidden hunger, are highly pervasive in Ethiopia. Anemia, a condition caused by inadequate dietary intake of iron and parasitic infections, among other causes, has cross-generational impacts on health and the economy. Approximately one in four women of reproductive age in Ethiopia are ane- mic (EDHS 2016), which affects not only women’s own health, but also contributes to the intergenerational cycle of undernutrition and lower productivity. More than half of children aged 6 to 59 months are anemic, with five regions having anemia prevalence of over 60 percent (EDHS 2016). Ensuring that women of reproductive age are well nourished sets the stage for their children to achieve optimal nutrition and development. Ethiopia has made significant investments in health services over the past two decades, and has seen impres- sive progress in improving several health outcomes, including child mortality that has declined from 77 to 20 deaths per 1,000 live births between 2000 and 2016 (EDHS 2016). Despite these improvements, significant challenges remain. The high prevalence of stunting in Ethiopia is associated with insufficient access to health services, poor water and sanitation, and suboptimal care and feeding practices. Demand- and supply-side barriers influence food consumption and diversity of diet. A recent analysis found that the cost of a nutrition- ally balanced diet in Ethiopia exceeds households’ incomes, limiting access to the range of nutrients required for growth, health and development during the early years and beyond (De Pee et al. 2010). Recognizing that factors other than poverty and food insecurity put children at risk of chronic malnutrition, there is a need for effective multi-sectoral strategies to address undernutrition across the country. A recent report found that 28 percent of all child mortality and 16 percent of primary school repetitions in Ethio- pia are associated with undernutrition (African Union Commission et al. 2014). Total annual costs associated with undernutrition are estimated at 16.5 percent of the Ethiopian gross domestic product (GDP) (55.5 billion Ethiopian birr) and are driven largely by lost working hours due to mortality associated with undernutrition and lower productivity of adults engaged in manual labor, such as agricultural work (Government of Ethiopia 2013). Political Commitment to Reduce Malnutrition In 2015, the Government of Ethiopia, a member of the SUN movement since 2011, adopted the Seqota Declara- tion to end child undernutrition by 2030 (Denys 2015). This declaration identifies nutrition targets and imple- mentation guidelines, and mobilizes ministries across sectors to address the underlying causes of malnutrition, with a focus on improved agriculture, food quality and micronutrient fortification. The commitment to nutri- tion is also reflected in the inclusion of an indicator to measure stunting in the five-year National Growth and Transformation Plan (GTP II 2016-2020) (Government of Ethiopia 2015). The Government of Ethiopia enacted the National Nutrition Plan (NNP I 2008-2015) to reorient the nutrition focus in the country from humanitarian emergencies to a more systematic and strategic preventive/promotive approach. Building on the success of this initiative, in 2016 the government launched the NNP II (2016-2020) with a focus on multi-sectoral actions to improve nutrition outcomes, particularly within the first 1,000 days, including through agriculture, water, education, social protection, and health sector as well as a nutrition coor- dination structure that extends from sub-woreda level to regional levels. Current Financing for Nutrition In 2015 in Ethiopia, the government and foreign donors spent a total of $71.8 million on interventions that con- tribute to reaching the global targets for nutrition. Of that amount, $2.55 million came from the government and $69.2 million came from ODA. This contribution from ODA included $23 million for stunting, $4.76 million for anemia, $7.26 million for breastfeeding, and $38.2 million for wasting.2 These estimates reflect the current spend- ing on nutrition; the following sections detail additional financing needed in order for Ethiopia to contribute to reaching the global targets on nutrition.3 CHAPTER 5 Ethiopia: An Investment Framework for Nutrition 73 Global Targets for Nutrition Substantial improvements to the nutritional status of women and children can be realized if there is adequate investment in a set of evidence-based nutrition-specific interventions that ensure optimum nutrition during the critical 1,000 day window between the start of a woman’s pregnancy and the child’s second birthday (Black et al. 2008, 2013). For women, these include interventions to prevent anemia before and during pregnancy, as well as those aimed at improving protein energy intake during pregnancy. Interventions targeted at children and their mothers aim to improve breastfeeding and complementary feeding practices, micronutrient status of children, and to treat acute malnutrition in children. In 2012—to rally the international community around improving nutrition—the 176 members of the World Health Assembly endorsed the first-ever global nutrition targets, focusing on six areas: stunting, anemia, low birthweight, childhood overweight, breastfeeding, and wasting. These targets aim to boost investments in cost-effective interventions, spearhead better implementation practices, and catalyze progress toward reducing malnutrition. The targets for stunting and wasting are enshrined within the United Nations’ Sustainable Devel- opment Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. The 2016 Global Nutrition Report ranked each country’s progress in contributing toward achieving the global targets (Table 5.1) (IFPRI 2016)4. Table 5.1: Four Global Targets for Nutrition and Ethiopia’s Contribution Toward Meeting Them ETHIOPIA’S RANK PREVALENCE PROGRESS Reduce the number of stunted children 1 STUNTING* 117/132 38% under five by 40% Stuntin Reduce the number of women of 2 ANEMIA 37/185 23% reproductive age with anemia by 50% An mi Increase the rate of exclusive breastfeeding  5 BREASTFEEDING 38/141 58% in the first six months up to at least 50% Exclusiv br stf din Reduce and maintain childhood wasting 6 WASTING* 98/130 10%  alnutrition) to less than 5% (acute m W stin LEGEND: Off course, no progress Off course, some progress On course, good progress *Stunting and wasting are included within the United Nations’ Sustainable Development Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. Sources: Nutrition targets from WHO 2014; Rank and progress from IFPRI 2016; Prevalence data from EDHS 2016. Coverage of key nutrition-specific interventions in Ethiopia is largely inadequate. Although coverage rates are higher for some childhood interventions, they remain well below the levels necessary to advance progress in reducing malnutrition among Ethiopian children. Table 5.2 and Figure 5.5 summarize the current coverage of and delivery platforms available for nutrition-specific interventions in Ethiopia. 74 in Sub-Saharan Africa Stunting Reduction  Table 5.2: Delivery Platforms For Nutrition-Specific Interventions In Ethiopia INTERVENTION DELIVERY PLATFORM Antenatal micronutrient supplementation (iron and folic acid only) Health facility and community Complementary feeding education Health facility, community, and communication campaigns Breastfeeding promotion Health facility, community, and communication campaigns Balanced energy protein supplementation for pregnant women Health facility, community, and social safety net programs Intermittent presumptive treatment of malaria in pregnancy Health facility, community, and food fortification Vitamin A supplementation Health facility, community, and food fortification Public provision of complementary foods Health facility, community, and social safety net programs Treatment of severe acute malnutrition Health facility and community Iron and folic acid supplementation for non-pregnant women of reproductive age School, community, health facility, and marketplace Staple food fortification Marketplace Pro-breastfeeding social policies Government policies National breastfeeding promotion campaigns Media Figure 5.5: Coverage of Key Nutrition-Specific Interventions in Ethiopia and Sub-Saharan Africa Antenatal micronutrient supplementation - (iron and folic acid only) Pro-breastfeeding Complementary social policies feeding education 100 80 60 Staple food Breastfeeding fortification 40 promotion 20 Iron and folic acid Balanced energy supplementation for non-pregnant protein supplementation women of reproductive age for pregnant women Treatment of severe Intermittent presumptive acute malnutrition treatment of malaria in pregnancy Public provision of Vitamin A complementary foods supplementation Note: Red shading represents Ethiopia and light blue shading represents average Sub-Saharan Africa coverage CHAPTER 5 Ethiopia: An Investment Framework for Nutrition 75 Economic Benefits of Investing in Nutrition There is a strong body of evidence that shows high economic returns to investing in nutrition (Alderman et al. 2016; Copenhagen Consensus Center 2015; Hoddinott et al. 2013). Scaling up these proven nutrition-specific interventions can ensure that mothers are healthy and well-nourished and that they can provide optimal nutrition to their children, that children realize their full physical and cognitive development potential, and that women’s productivity is not hampered by illness, especially anemia (Figure 5.6). Figure 5.6: How Reaching The Global Nutrition Targets Generates Economic Benefits Co nitive Development STUNTING Childr n’s H lth & Le rnin & Educ tion l Nutrition St tus Att inment WASTING Glob l Nutrition T r ts R ch d Adult Productivit EASTFEE BRE EDING W es ANEMIA Econom Wom n’s H lth & (GDP) Nutrition St tus In Ethiopia, scaling-up the package nutrition-specific interventions will produce substantial economic ben- efits over the productive lifetime of the affecte d women and children (Figure 5.7). Additional health system cost-savings are also likely because many of these investments reduce the burden of childhood illnesses such as diarrhea and pneumonia. Figure 5.7: Investments in Ethiopia to Meet the Global Nutrition Targets Have Enormous Economic Returns Total Economic $1 Invested Benefits (Billions)a Yieldsb STUNTING $27.9 $21 $ $10.7 $66 1 BREASTFEEDING $ $ ANEMIA $3.5 $ $18 WASTING $2.9 $10 a. Total economic benefits over 10 years for women and over the productive lives of children who benefit from these interventions , defined as the period between the age of 18 and a “retirement” age - the life expectancy or the age of 65, whichever is lower. b. Benefit calculation assumes a 3 percent discount rate for both costs and benefits, and GDP growth rate of 3 percent. 76 in Sub-Saharan Africa Stunting Reduction  Financing Needs, Impact, and Cost-Effectiveness of Scaling Up Nutrition-Specific Interventions Using the methodology detailed in An Investment Framework for Nutrition this brief presents estimates of the resources needed to scale up a package of 12 high-impact nutrition-specific interventions in Ethiopia to meet the global nutrition targets for stunting, anemia, breastfeeding and wasting, and the estimated nutrition, health and economic impact. An additional $220 million per year over 10 years is needed to scale up the package of key inter- ventions (Table 5.3). The health and nutrition impact of this investment is shown in Table 5.4. Among the set of proposed interventions, educating mothers about complementary feeding is the most effective for stunting, averting more than half million cases of stunting and 12,000 child deaths over 10 years. Breastfeeding promotion through counseling of mothers is projected to increase the number of infants exclusively breastfed by 1.65 million, prevent nearly 30,000 deaths, and cost $93 million over 10 years. Among these interventions, this is the most cost-effective intervention for preventing child mortality, and costs $56 per child exclusively breastfed. For preventing maternal anemia, staple food fortification proves to be most cost-effective, at a cost of $3.70 for each case of anemia prevented in women. Over 10 years, staple food fortification will prevent 4.19 million cases of anemia in women at a cost of $15.4 million. Interventions to reduce stunting will require the most resources, accounting for about 76 percent of the total amount required for scale-up. However, some of the stunting interventions will also have impacts for achieving the targets for breastfeeding and anemia. Figure 5.8 represents the distribution of total cost across interventions to address the four targets. Figure 5.8: Ten Year Financing Needs For Scaling Up a Package of Nutrition-Specific Interventions in Ethiopia ANEMIA Total 13%: WASTING Total: 17% STUNTING Total 76%: BREASTFEEDING Total: 13% TOTAL TEN-YEAR FINANCING NEEDS: $2.2 billion Note: Some costs for anemia, breastfeeding, and stunting are shared across interventions. Costs for breastfeeding promotion (US $93 million) has been included in both the total cost for the breastfeeding target and the total cost for the stunting target; cost of intermittent presumptive treatment of malaria in pregnancy (US $62.2 million) and antenatal micronutrient supplementation (US $110.7 million) have been included in both the total cost for the anemia target and the total cost for the stunting target. CHAPTER 5 Ethiopia: An Investment Framework for Nutrition 77 Table 5.3: Estimated 10-Year Financing Needs and Cost-Effectiveness of Scaling Up Nutrition-Specific Interventions, Ethiopia TOTAL 10-YEAR COST PER CASE COST PER DEATH INTERVENTION (NUTRITION TARGET) FINANCING NEEDS OF STUNTING AVERTED (US $) (US $M) AVERTED (US $) For pregnant women and mothers of infants Antenatal micronutrient supplementation (stunting, anemia) 110.7 10,491 8,170 Infant and young child nutrition counseling (complementary feeding 235.1 5,865 411 education and breastfeeding promotion combined) (stunting, breastfeeding) Complementary feeding education 142.1 11,517 254 Breastfeeding promotion 93.0 3,352 7,671 Balanced energy protein supplementation for pregnant women (stunting) 210.0 28,256 45,537 Intermittent presumptive treatment of malaria in pregnancy (anemia) 62.2 4,230 1,965 For infants and young children Vitamin A supplementation (stunting) 48.5 7,188 384 Prophylactic zinc supplementation (stunting) 541.4 33,852 1,062 Public provision of complementary food (stunting) 453.6 57,743 1,317 Treatment of severe acute malnutrition (wasting) 366.7 11,744 n/a For non-pregnant women and general population Iron and folic acid supplementation for non-pregnant women (anemia) 90.5 n/a 15,252 Staple food fortification (anemia) 15.4 n/a Pro-breastfeeding social policies (breastfeeding) 6.2 n/a n/a National breastfeeding promotion campaigns (breastfeeding) 33.3 n/a n/a TOTAL: 2,173.7 15,078 1,037 Note: Financing needs and impacts assume a linear scale-up of interventions from current coverage level to 90 percent over five years, then maintained at 90 percent for an additional five years. Unit costs for each intervention were drawn from available unit costs from neighboring countries, global costs, or estimates available in the literature. The estimated financing needs include an additional 12 percent (11 percent for pro-breastfeeding social policies and promotion campaigns) to account for monitoring, evaluation, capacity and policy development that may be necessary to reach full scale-up of the interventions. The Lives Saved Tool (LiST) was used to estimate the impact of interventions that target pregnant women and children. The impact of interventions that target the general population or non-pregnant women were estimated using a Microsoft Excel model. It should be noted that the LiST model does not capture potential synergies between specific interventions (e.g. the fact that the impact of behavior change communication interventions may be higher in populations that have access to affordable and diversified foods or in populations with higher levels of educational attainment). Therefore, it is possible that the impact estimates generated using LiST in fact underestimate the true impact of the interventions in some contexts. n.a. = not applicable. Two Alternative Investment Packages In an environment of constrained resources in which Ethiopia may not be able to raise $220 over the next 10 years, two alternative investment packages are laid out for consideration. The Priority Package: The first—the “priority package”—includes interventions that are the most cost- effec- tive, that is, have the lowest cost per health outcome (e.g., case of stunting averted), and that have well-es- tablished global policy guidelines and delivery platforms. Based on those two criteria, the priority package includes:antenatal micronutrient supplementation, infant and young child nutrition counseling, intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions;vitamin A supplementation; treatment of severe acute malnutrition intermittent weekly iron and folic acid supplementation for girls 15–19 years of age attending school; and) fortification of wheat and maize flour with iron and folic acid. These 78 in Sub-Saharan Africa Stunting Reduction  Table 5.4: Benefits and Cost-Effectiveness By Investment Package, Ethiopia GLOBAL PRIORITY CATALYZING FULL PACKAGE: BENEFIT All interventions needed TARGET PACKAGE PROGRESS PACKAGE to meet targets $99 million/year in $138 million/year $220 million/year in financing need financing need financing need Cases of stunting reduced by STUNTING 0.7 million 1.0 million 1.6 million 2025 (vs 2015)a Cases of anemia in women ANEMIA 2.7 million 10.0 million 14.4 million prevented by 2025 Additional babies BREASTFEEDING 1.7 million 1.7 million 1.7 million breastfed over 10 years Child deaths averted 97,000 107,000 130,000 over 10 years ALL TARGETS Cost per death averted 10,247 12,897 16.692 Cost per case of 630 840 1,038 stunting averted a. Total impact of proposed intervention package combined with other health and poverty reduction efforts. interventions would be scaled up to full program coverage in the first five years and maintained at full cover- age levels for the last five years. This priority package would require an estimated $994 million over 10 years, or $99.4 million annually (see Table 5.4). During the 10 years of scale-up, this package would prevent 700,000 cases of stunting and avert 97,000 deaths in children under five years of age. It would also prevent 2.7 million case-years of anemia in women and would result in 1.7 million additional children under 6 months of age being exclusively breastfed. The Catalyzing Progress Package: The second alternative—“catalyzing progress”—includes scale-up of all interventions in the priority package, plus a phased approach to scaling up public provision of complementary foods, balanced energy protein supplementation, prophylactic zinc supplementation, and weekly iron-fo- lic acid supplementation for women outside of schools. It is assumed that, for the latter set of interventions, during the first five years, emphasis will be placed on establishing global guidelines and on operational research to develop effective delivery platforms, or to develop less expensive products or more cost-effective technologies. Costs are approximated as the cost of scaling up this set of interventions from 0 to 10 percent coverage only in the first five years. In the subsequent five years, it is assumed that the coverage expansion of those interventions will accelerate and reach 60 percent by 2025. This package would require $138 million per year, a total of $1.38 billion over 10 years (Table 5.4). It would prevent 107,000 deaths and 1 million cases of stunting among children under five, increase the number of exclusively breastfed children under six months of age by 1.7 million, and prevent 10 million case-years of anemia in women. In comparing the relative cost-effectiveness of the three investment packages, the two alternative packages are more cost-effective in preventing deaths and stunting. However, neither is as effective as the full package in making progress toward achieving the stunting, wasting, and anemia targets. The priority and catalyzing CHAPTER 5 Ethiopia: An Investment Framework for Nutrition 79 progress packages will prevent 97,000 and 107,000 thousand deaths respectively, compared with 130,000 deaths prevented with the full package over 10 years. Under the full package scenario, 1.6 million cases of childhood stunting will be prevented, compared with 1 million cases under the catalyzing progress scenario and 700,000 cases under the priority package scenario. Furthermore, there would be nearly 11.7 million and 4.4 million more cases of anemia in women under the priority package and catalyzing progress package, respectively. A Call to Action As the world stands at the cusp of the new Sustainable Development Goals, there is an unprecedented oppor- tunity to save children’s lives, build future human capital and cognitive development, and drive faster eco- nomic growth. Scaling-up key nutrition interventions during the critical 1,000 day window of early childhood will pay lifelong dividends, translating to healthier societies and more robust economies. If this window is missed, it is missed for life. The additional financing required to reach the global nutrition targets will require coordinated efforts by all stakeholders and a supportive policy environment. To achieve these targets would require an increase in the funding allocated to nutrition by $220 million annually, roughly an equivalent to a 15 percent increase in current general government expenditure on health.5 These investments are over and above those needed for improving water and sanitation, and issues around women’s empowerment, and food security. Although this level of domestic financing is ambitious, Ethiopia is already moving in this direction. In the long term, nutrition interventions have significant potential to reduce poverty and boost shared prosperity. Accelerating the reduction of stunting in Ethiopia will be essential for maximizing the return on investments in early childhood development, in education, and more broadly in policies aimed at fostering and enhancing human capital accumulation and job creation. Investing in the early years is even more critical because the Africa region is entering a demographic transition with an expected increase in the working age population from 54 percent in 2010 to 64 percent in 2090. The scale-up of the key nutrition-specific interventions to reduce stunting is estimated to generate considerable returns in economic benefits over the productive lives of ben- eficiaries, and is a necessary condition to build human capital through investments in the early years and to harness the potential benefits of the demographic dividend. Endnotes 1 Information about the Power of Nutrition initiative is available at https://ciff.org/grant-portfolio/ the-power-of-nutrition/. 2 Note that because some funded interventions contribute to more than one target, total funding across the four targets is less than the total funding for each target added together. 3 Current financing by source is from Results for Development Institute and can be found at http:/www.inves- tinnutrition.org/. 4 Two of the global nutrition targets—those for low birthweight and for child overweight—were not included in the analyses because of insufficient data on the prevalence of low birthweight and a lack of consensus on effective interventions to reach the target for child overweight. 5 WHO National Health accounts database indicates general government health expenditure in Ethiopia was US$1,517 million in 2014. At that level, this will need to be increased by 15 percent to accommodate the US$220 million per year required for scale-up of the 12 nutrition-specific interventions. 80 in Sub-Saharan Africa Stunting Reduction  References African Union Commission, NEPAD Planning and Coordinating Agency, UN Economic Commission for Africa, and UN World Food Programme. 2014. The Cost of Hunger in Africa: Social and Economic Impact of Child Undernutrition in Egypt, Ethiopia, Swaziland, and Uganda. Addis Ababa: UNECA. Aguayo, V. M., R. Nair, N. Badgaiyan, and V. Krishna. 2016. “Determinants of Stunting and Poor Linear Growth in Children under 2 Years of Age in India: An In-Depth Analysis of Maharashtra’s Comprehensive Nutrition Survey.” Maternal & Child Nutrition 12 (Suppl. 1): 121–40. Alderman , H ., J. R. Behrman , and C. Puett . 2016 . “ Big Numbers about Small Children: Estimating the Eco- nomic Benefits of Addressing Undernutrition .” World Bank Research Observer 31 ( 2 ). Black, R. E., L. H. Allen, Z. A. Bhutta, L. E. Caulfield, M. de Onis, M. Ezzati, C. Mathers, J. Rivera, and the Maternal and Child Undernutrition Study Group. 2008. “Maternal and Child Undernutrition: Global and Regional Exposures and Health Consequences.” The Lancet 371 (9608): 243–60. Black, R. E., C. G .Victora, S. P. Walker, Z. A. Bhutta, P. Christian, M. de Onis, M. Ezzati, S. Grantham-Mcgregor, J. Katz, R. Martorell, R. Uauy, and the Maternal and Child Nutrition Study Group. 2013. “Maternal and Child Undernutrition and Overweight in Low-Income and Middle-Income Countries.” The Lancet 382: 427–51. Copenhagen Consensus Center. 2015. Smart Development Goals: The Post-2015 Consensus. http://www.copenha- genconsensus.com/sites/default/files/outcomedocument_col.pdf Daniels, M. C. and L. Adair. 2004. “Growth in Young Filipino Children Predicts Schooling Trajectories through High School.” Journal of Nutrition 134: 1439–46. Denys, C. 2015. “Financing for Development: Mobilizing Leadership and Investment in Nutrition.” Site and Life 29 (2): 90–94. http://www.sightandlife.org/fileadmin/data/Magazine/2015/29_2_2015/20_financing_for_ development_mobilizing_leadership_and_investment_in_nutrition.pdf De Pee, S., T. van den Briel, J. van Hees, and M. Bloem. 2010. “Introducing New and Improved Food Products for Better Nutrition.” In Revolution: From Food Aid to Food Assistance. Rome: World Food Programme. EDHS (Ethiopia Demographic and Health Survey). 2016. Ethiopia Demographic and Health Survey: Key Indicators Report. Addis Ababa, Ethiopia, and Rockville, Maryland, USA: Central Statistical Agency and the DHS Pro- gram, ICF. https://dhsprogram.com/pubs/pdf/PR81/PR81.pdf Fink, G., E. Peet, G. Danaei, K. Andrews, D. C. McCoy, C. R. Sudfeld, M. C. Smith Fawzi, M. Ezzati, and W. W. Fawzi. 2016. “Schooling and Wage Income Losses Due to Early-Childhood Growth Faltering in Developing Countries: National, Regional, and Global Estimates.” The American Journal of Clinical Nutrition 104 (1): 104–12. Government of Ethiopia. 2013. The Cost of Hunger in Ethiopia: Implications for the Growth and Transformation of Ethiopia. Summary Report. Addis Ababa: Government of Ethiopia. https://www.unicef.org/ethiopia/FINAL_ Ethiopia_-_COHA_Summary_Report_June_20_24pg_72dpi.pdf Government of Ethiopia. 2015. The Second Growth and Transformation Plan (GTP II) 2015/16-2019/20. Addis Ababa: National Planning Commission. https://www.africaintelligence.com/c/dc/LOI/1415/GTP-II.pdf Hoddinott, J., H. Alderman, J. R. Behrman, L. Haddad, and S. Horton. 2013. “The Economic Rationale for Investing in Stunting Reduction.” Maternal and Child Nutrition 9 (Suppl. 2): 69–82. Hoddinott, J., J. A. Maluccio, J. R. Behman, R. Flores, and R. Martorell. 2008. “Effect of a Nutrition Intervention during Early Childhood on Economic Productivity in Guatemalan Adults.” Lancet 371 (9610): 411–16. CHAPTER 5 Ethiopia: An Investment Framework for Nutrition 81 Horton, S. and R. Steckel. 2013. “Malnutrition: Global Economic Losses Attributable to Malnutrition 1900–2000 and Projections to 2050.” In The Economics of Human Challenges, edited by B. Lomborg, 247–72. Cambridge, U.K.: Cambridge University Press. IFPRI (International Food Policy Research Institute). 2016 . Global Nutrition Report 2016: From Promise to Impact – Ending Malnutrition by 2030. Washington, DC: IFPRI. Martorell, R., B. L. Horta, L. S. Adair, A. D. Stein, L. Richter, C. H. D. Fall, S. K. Bhargava, S. K. Dey Biswas, L. Perez, F. C. Barros, C. G. Victora, and Consortium on Health Orientated Research in Transitional Societies Group. 2010. “Weight Gain in the First Two Years of Life Is an Important Predictor of Schooling Outcomes in Pooled Analyses from Five Birth Cohorts from Low- and Middle-Income Countries.” Journal of Nutrition 140: 348–54. MOFED (Ministry of Finance and Economic Development). 2011. Growth and Transformation Plan for Ethiopia (GTP): 2010/11–2014/15. Draft. Addis Ababa, Ethiopia: Ministry of Finance and Economic Development. http:// www.ethiopians.com/Ethiopia_GTP_2015.pdf Ozaltin, E., K. Hill, and S. V. Subramanian. 2010. “Association of Maternal Stature with Offspring Mortality, Underweight, and Stunting in Low- to Middle-Income Countries.” JAMA 303 (15): 1507–16. Sanchez-Montero, M., N. Salse Ubach, and M. Sullivan 2010. Undernutrition: What Works? A Review of Policy and Practice. Madrid: ACF International Network and Tripode. Shekar, M., J. Kakietek, J. D. Eberwein, and D. Walters. 2017. An Investment Framework for Nutrition: Reaching the Global Targets for Stunting, Anemia, Breastfeeding, and Wasting. Directions in Development. Washington, DC: World Bank. doi:10.1596/978-1-4648-1010-7. https://tinyurl.com/ InvestmentFrameworkNutrition UN DESA (United Nations Department of Economic and Social Affairs). 2015. World Population Prospects, the 2015 Revision. https://esa.un.org/unpd/wpp/Download/Standard/Population/ World Bank. 2016. 5th Ethiopia Economic Update: Why so Idle? Wages and Employment in a Crowded Labor Mar- ket. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/463121480932724605/ pdf/110730-WP-P155867-PUBLIC-WBEEUdraftforlaunchdecemberfull.pdf ———. 2017. World Development Indicators (database), World Bank , Washington, DC (accessed May 1, 2017), http://data.worldbank.org/data-catalog/world-development-indicators. WHO (World Health Organziation). 2014. Comprehensive Implementation Plan on Maternal, Infant and Young Child Nutrition. Geneva: WHO. http://apps.who.int/iris/bitstream/10665/113048/1/WHO_NMH_NHD_14.1_eng. pdf?ua=1 82 in Sub-Saharan Africa Stunting Reduction  CHAPTER 6 NIGER: An Investment Framework for Nutrition Benefits of Investing in Nutrition Key Messages • More than 40 percent of children in Niger are chronically malnourished (stunted), with the highest prevalences in Diffa, 436,000 Zinder, and Maradi regions. In addition, nearly half (46 percent) of women of reproductive age in Niger are anemic. cases of stunting prevented in 2025 • To address persistent malnutrition, the Nigerien government launched an initiative called the 3N Program, Nigeriens Nourish Nigeriens, to promote sustainable national food security and agricultural development. 86,000 • Scaling up a package of high-impact nutrition-specific child deaths prevented interventions in Niger to contribute to reaching the global in 2025 nutrition targets would require an additional $77.1 million per year over 10 years and would provide enormous benefits (see panel on right). These investments are over and above those needed for improving water and sanitation and for addressing 7.5 MILLION issues around women’s empowerment and food security. case-years of anemia • This investment would require additional financing equivalent to in women prevented in 2025 40 percent of the general government expenditure on health and could be financed from a combination of domestic budgets, official development assistance (ODA), and innovative financing sources such as the Power of Nutrition.1 1.2 MILLION • The economic benefits generated over the productive lives of babies exclusively breastfed beneficiaries would be enormous: $4.1 billion for stunting, $512 million for anemia, $3 billion for breastfeeding, and $1.5 billion for the treatment of severe wasting. • Returns on every dollar invested in reaching the global nutrition 1.4 MILLION targets range from $7 in returns for anemia to $10 for stunting, $11 cases of severe wasting treated for wasting, and $48 for investing in exclusive breastfeeding. • To finance the nutrition scale-up, two lower-cost scale-up scenarios are estimated to require between $29.7 and $46.6 million per year over the next 10 years. In an environment of constrained $7–$48 resources in Niger, starting with one of these two scenarios would return for every dollar invested be a strong first investment, but it would need to be followed by increased investments, along with investments in strengthening the national platforms for service delivery, to contribute to meeting the global nutrition targets. $4.1 BILLION generated from investments to reduce stunting* *The economic benefits are calculated over the productive lives of the children benefiting from the interventions that prevent stunting. CHAPTER 6 Niger: An Investment Framework for Nutrition 83 Investment Case for Nutrition Ensuring optimum nutrition—particularly during the 1,000-day period from pregnancy to a child’s second birthday—can alter an individual’s development trajectory and maximize her or his productive potential. Chronic malnutrition has important lifelong consequences for health and cognitive development. Losses to cognitive development in early childhood resulting from chronic malnutrition are irreversible. Being stunted (low height-for-age) in early childhood is associated with a delayed start at school, reduced schooling attain- ment, and substantially decreased adult incomes at both the individual and country level (Daniels and Adair 2004; Fink et al. 2016; Hoddinott et al. 2008; Martorell et al. 2010). These consequences add up to overall GDP losses of 4 to 11 percent in Africa and Asia (Horton and Steckel 2013). Importantly, chronic undernutrition can be transmitted through an inter-generational cycle, where malnourished mothers are more likely to have stunted children (Aguayo et al. 2016; Ozaltin et al. 2010). Investments in nutrition are highly cost-effective and among the best value-for-money development actions (Copenhagen Consensus Center 2015; Hoddinott et al. 2013). An Investment Framework for Nutrition developed by the World Bank in partnership with R4D, 1000 Days, and the Bill & Melinda Gates Foundation estimated high returns on every dollar invested in nutrition: from $4 in returns for treating acute malnutrition (wasting) to $11 for preventing stunting, $12 for the treatment and prevention of anemia, and $35 for increasing the prev- alence of exclusive breastfeeding (Shekar et al. 2017). Not only do investments in nutrition produce substantial economic benefits, but they also lay the groundwork for the success of investments in other sectors. Investments in the early years—including early life nutrition, early learning and stimulation, and nurturing care and protection from stress—ensure that all children reach their human potential and contribute to the economic growth of their nation. The analysis presented below focuses on high-impact nutrition-specific inter- ventions with strong evidence of efficacy in reducing stunting, and it estimates the costs, impact, and economic benefits of scaling up these interventions in Niger. Country Context The Republic of Niger is a landlocked country in Sub-Saharan Africa with a population of nearly 20 million people. Niger has one of the lowest GDP rates per capita on the continent, coupled with one of the highest pop- ulation growth rates globally (4 percent) (UN DESA 2015). The absolute number of individuals living in poverty in Niger continues to increase because of the country’s high population growth rate, which is highest among the poorest households (World Bank 2014). Niger’s population is young—approximately 20 percent of the country are under age five (UN DESA 2015). Low human capital remains one of the key challenges in reducing poverty and achieving greater socioeconomic equity. Over 80 percent of the population reside in rural areas (World Bank 2016). Only 4 percent of the country is arable, yet the agriculture sector represents the country’s primary source of economic activity, contributing an average of 25 percent of the national GDP (Geesing and Djibo 2001; World Bank 2013). Between 2005 and 2011, income disparities between rural and urban households grew as poverty rates fell more rapidly in urban centers than in rural areas, as measured by household consumption and living conditions (World Bank 2014). During this same time period, the poverty headcount for the poorest populations decreased by four percentage points, reflecting the fact that farming households are highly sensitive to fluctuations in agricultural outputs (World Bank 2014). The Human Development Index (HDI) showed Niger ranked 187 out of 188 countries in 2016, with a value of 0.353 (UNDP 2016), exacerbating vulnerability and compounding a poor household’s ability to meet basic needs. Child malnutrition, an underlying cause of up to 45 percent of deaths of children under age five (Black et al. 2013), has emerged as one of the key markers of poverty and vulnerability as well as one of the key challenges to ensuring optimal accumulation of human capital in the country. 84 in Sub-Saharan Africa Stunting Reduction  Nutritional Status in Niger Persistently high rates of undernutrition remain a serious human development challenge in Niger. More than 40 percent of children under age five are stunted and 18 percent are wasted (low weight-for-height) (NDHS 2012). Between 1992 and 2006, stunting prevalence increased, but this was followed by a decline from 54.8 percent in 2006 to 43.9 percent in 2012 (NDHS 2006). Despite this recent downward trend, Niger still ranks 122nd out of 132 countries assessed for highest stunting prevalence in children under age five (IFPRI 2016). At the same time, wasting prevalence increased sharply, from 12.4 percent in 2006 to 18 percent in 2012 (Figure 6.1) (NDHS 2006, 2012) suggesting that acute forms of malnutrition are on the rise. Although levels of chronic malnutrition show some socioeconomic variation, stunting prevalence is perva- sive across all wealth quintiles (Figure 6.2). In the wealthiest households, stunting is a significant concern (34.5 percent), compared with 46.7 percent in the poorest quintile, underscoring the fact that much of Niger faces extreme poverty and that wealth status may be relative and insufficient to eliminate malnutrition by itself. Among children from rural households, 46 percent are stunted compared with 30 percent of their urban counterparts. Figure 6.1: Trends in Undernutrition in Niger, 1992–2012 Figure 6.2: Disparities in Stunting by Wealth Quintile, 1992–2012 60 60 54.8 54.8 80 80 AMONG CHILDREN UNDER FIVE AMONG CHILDREN UNDER FIVE STUNTING PREVALENCE (%) AMONG CHILDREN UNDER FIVE STUNTING PREVALENCE (%) 46.5 46.5 AMONG CHILDREN UNDER FIVE 50 45.2 45.2 50 43.9 43.9 60 60 PREVALENCE (%) 40 PREVALENCE (%) 40 40 40 30 30 25.5 25.5 18.5 18.5 18 18 20 20 20 12.4 20 12.4 10 10 0 0 1992 1992 1998 1998 2006 2006 2012 2012 0 0 1992 1992 1998 1998 2006 2006 2012 2012 Poor Poor st quintil Rich st quintil Rich st st quintil quintil N tionNl tion l Stuntin Stuntin W stinW stin Source for Figures 6.1 and 6.2: NDHS, 1992, 1998, 2006 and 2012. The highest stunting prevalence is concentrated in the Diffa, Maradi, and Zinder regions, where stunting preva- lence there is estimated to be over 50 percent (Figure 6.3). This is in stark contrast to the capital, Niamey, which has the lowest prevalence of stunting, at 20 percent (NDHS 2012). Given the regional variation in stunting prevalence and burden, targeted interventions are needed to address the key drivers of undernutrition in these highest prevalence and burden regions. Furthermore, Maradi and Zinder regions carry the greatest absolute burden of children who are stunted. While Diffa region’s prevalence is among the highest, the absolute burden is lower, particularly as compared to the Tahoua region. Given the data on regional variations in stunting prev- alence in Niger, it would be important to understand the key drivers of undernutrition in these highest preva- lence and burden regions and to design targeted interventions to address them. CHAPTER 6 Niger: An Investment Framework for Nutrition 85 Burundi 60 Madagascar RWANDA STUNTING PREVALENCE (%) 50 Niger Mozambique Ethiopia Chad 40 DRC Tanzania Malawi Sierra Leone Figure 6.3: Number of Stunted Children and Stunting Prevalence by Region,Uganda Niger 2012 30 Guinea Zimbabwe Senegal 20 10 0 AGADEZ AGADEZ 500 1000 1500 2000 2500 3000 GDP PER CAPITA (INTERNATIONAL $ PPP) TAHOUA DIFFA TAHOUA DIFFA ZINDER ZINDER TILLABERI TILLABERI MARADI MARADI NIAMEY NIAMEY DOSSO DOSSO 50 453,927 400,000 45 Stunting Prevalence ≥ 40%: Very high 20–29%: Medium Total Children By Region (%) 30–39%: High <20%: Low Zambia Stunted By Region 250,000 80,000 Region Boundaries STUNTING PREVALENCE (%) 40 Region Boundaries Sudan Nigera 50,000 46,437 35 Cameroon Data source: NDHS 2012. IBRD 42942 | MAY Lesotho 2017 These maps were produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denominations 30 | MAY 2017 IBRD 42942 COTE D’LVOIRE Swaziland and any other information shown on these maps do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or 25 of such boundaries. acceptance Kenya Mauritania 20 Congo Niger’s stunting prevalence is significantly higher than the average for the region, which is 35.2 percent (World Ghana 15 Bank 2017), and the country Sao Tomeof is one andthe poorest in the region (Figure 6.4). Approximately one in five people Principe 10 in Niger face extreme food insecurity (Save the Children 2009). During the 2009–10 food crisis, more than 30 5 of the population required food assistance (World Bank 2013). It is estimated that, in 2011, the poorest percent 30 percent 0 of the population shared one-seventh of total national consumption, while the highest income group more than1500 shared 500 2500 51 percent 3500 of total national 4500 consumption 5500 (World 6500 Bank 7500 2014). Although 8500 Niger 9500 has made some gains in child survival and maternal health, significant challenges remain GDP PER CAPITA (INTERNATIONAL $ PPP) in improving nutrition outcomes as well as in curbing the high fertility rates, which can seriously impede any progress made in nutrition. Figure 6.4: Prevalence of Stunting and GDP per Capita: Niger and Selected Low-Income Countries NIGER 50 DRC Mozambique Chad Ethiopia STUNTING PREVALENCE (%) 40 Rwanda Malawi Sierra Leone Tanzania Benin 30 Uganda Guinea Zimbabwe Senegal 20 10 500 1000 1500 2000 2500 3000 GDP PER CAPITA (INTERNATIONAL $ PPP) Source: NDHS 2012; World Bank 2017. Wasting, also known as acute malnutrition, is typically classified as either severe or moderate. Wasting can result from food insecurity in resource-poor settings with insufficient dietary quality, quantity and diver- 50 sity, suboptimal breastfeeding, and recurrent episodes of illness such as diarrhea. Wasting prevalence across DRC Mozambique Chad PREVALENCE (%) Niger Ethiopia 40 Malawi Rwanda 86 Tanzania  Stunting Reduction in Sub-Saharan Africa Sierra Leone 30 Uganda BENIN Guinea Sub-Saharan Africa is second highest in the world, after South Asia, with 13 million children (7.8 percent) suffering from acute malnutrition. Niger ranks fifth in the top five countries with the highest wasting preva- lence, with the prevalence at 18 percent in 2012. Particularly with sharp rise in wasting prevalence since 2012, treatment of acute malnutrition, as well as efforts to better understand and address the drivers of wasting, are evermore critical to reversing this trend. Micronutrient deficiencies, a form of malnutrition that relates to a deficiency in essential vitamins and minerals needed for body functions, are highly pervasive in Niger. Anemia, a condition caused by inadequate dietary intake of iron and by parasitic infections, among other causes, has cross-generational impacts. Nearly half (46 percent) of women of reproductive age in Niger are anemic (NDHS 2012), which affects not only women’s own health but also contributes to the intergenerational cycle of undernutrition. Nearly three-quarters of children (73 percent) 6 to 59 months of age are anemic, with anemia prevalence highest in the Diffa region (88 percent) (NDHS 2012). Ensuring that women of reproductive age are well nourished through diverse diets rich in micronutrients and are provided with necessary micronutrient supplements sets the stage for their children to achieve optimal nutrition, growth, and development and for the country to build human capital. Demand- and supply-side barriers influence feeding behaviors, including breastfeeding practices, food con- sumption, and dietary diversity. Market dependence varies by livelihood and wealth quintile. Sheep and cattle farmers largely rely on markets, while wealthier cultivators tend to grow enough to support themselves as well as to purchase additional food to improve dietary diversity (Save the Children 2009). The cost of food can account for 60 to 75 percent of expenditures for the poorest families, and because the percentage of budget allocated for food purchases sharply increases during times of crisis, food consumption can drastically decrease at these times. A crop or grazing failure, climatic event, or an increase in market rate for staple crops translates into a reduction in purchasing power and an impaired ability to meet basic food needs for the most vulnerable families (Save the Children 2009). These underlying causes of malnutrition from food shortages, coupled with difficult access to health centers, inadequate hygiene, lack of proper sanitation, and behavioral factors, further highlight the need for effective multisectoral strategies to address undernutrition across the country. Political Commitment to Reduce Malnutrition The policy environment around nutrition is gaining momentum in Niger. Nutrition, along with reducing fertility and maximizing the demographic dividend—two agendas that share complementarity with nutrition—is one of priorities of the Prime Minister’s ambitious General Development Plan, demonstrating a high-level recog- nition that nutrition outcomes influence economic development (SUN 2014). This commitment was further solidified when Niger joined the fight to end malnutrition as a member of the SUN movement in 2011. Starting in 2012, Niger began implementing a multisector, overarching nutrition strategy known as Nigeriens Nourish- ing Nigeriens (3N). 3N aims to strengthen the agriculture sector’s production capabilities to improve nutrition outcomes and resilience to cyclical food crises (SUN, no date). The Health Ministry implements the majority of the nutrition interventions, the Minister of Public Health chairs the 3N committee, and the underlying guide- lines of the strategy were derived by the Niger Renaissance Program (SUN 2014). Within the 3N initiative, nearly three-quarters of the budget allocation is for nutrition-specific interventions to improve nutrition practices, reduce acute malnutrition, and improve micronutrient intake; about one-quarter is allocated for governance to implement nutrition-specific and nutrition-sensitive interventions; and the remaining smaller portion is allo- cated for nutrition-sensitive interventions such as food security (SUN, no date). Although the 3N program lays the foundation necessary to address malnutrition, enable national food production, and increase income, greater investments are needed from domestic resources, ODA, and other innovative financing mechanisms for Niger to substantially reduce malnutrition. CHAPTER 6 Niger: An Investment Framework for Nutrition 87 Current Financing for Nutrition In 2015 in Niger, the government and overseas donors spent a total of $28.9 million on interventions that will contribute to reaching the global targets for nutrition. Of that amount, $10.7 million came from the government and $18.3 million came from ODA. The contribution from ODA included $5.7 million for stunting, $1.2 million for anemia, $1.6 million for breastfeeding, and $10.7 million for treating wasting.2 The single largest nutrition investment in Niger is for the treatment of wasting. These estimates reflect the current spending on nutrition; the following sections detail additional financing needed in order for Niger to contribute to reaching the global targets on nutrition.3 Global Targets for Nutrition Substantial improvements to the nutritional status of women and children can be realized if adequate invest- ment is made in a set of evidence-based nutrition-specific interventions that ensure optimum nutrition during the critical 1,000-day window between the start of a woman’s pregnancy and the child’s second birthday (Black et al. 2008, 2013). For women, these include interventions to prevent anemia before and during pregnancy as well as those aimed at improving protein energy intake during pregnancy. Interventions targeted toward children and their mothers aim to improve breastfeeding and complementary feeding practices, enhance the micronutrient status of children, and treat acute malnutrition in children. In 2012—to rally the international community around improving nutrition—the 176 members of the World Health Assembly endorsed the first-ever global nutrition targets, focusing on six areas: stunting, anemia, low birthweight, childhood overweight, breastfeeding, and wasting. These targets aim to boost investments in cost-effective interventions, spearhead better implementation practices, and catalyze progress toward reducing malnutrition. The targets for stunting and wasting are enshrined within the United Nations’ Sustainable Devel- opment Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. The 2016 Global Nutrition Report ranked each country’s progress in contributing to the achievement of the global targets (Table 6.1) (IFPRI 2016).4 Table 6.1: Four Global Targets for Nutrition and Niger’s Contribution Toward Meeting Them NIGER’S RANK PREVALENCE PROGRESS Reduce the number of stunted children 1 STUNTING* 122/132 43.9% under five by 40% Stuntin Reduce the number of women of 2 ANEMIA 169/185 46.7% reproductive age with anemia by 50% An mi Increase the rate of exclusive breastfeeding  5 BREASTFEEDING 103/141 23.3% in the first six months up to at least 50% Exclusiv br stf din Reduce and maintain childhood wasting 6 WASTING* 126/130 18.0%  alnutrition) to less than 5% (acute m W stin LEGEND: Off course, no progress Off course, some progress On course, good progress *Stunting and wasting are included within the United Nations’ Sustainable Development Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. Source: Global Nutrition Report 2016. Nutrition targets are from WHO 2014; rank and progress from IFPRI 2016; prevalence data from NDHS 2012. 88 in Sub-Saharan Africa Stunting Reduction  Coverage of key nutrition-specific interventions in Niger is largely inadequate and remains well below the lev- els necessary to advance in reducing malnutrition among Nigerien children. Table 6.2 and Figure 6.5 summa- rize the current coverage and delivery platforms of nutrition-specific interventions in Niger. Table 6.2: Delivery Platforms for Nutrition-Specific Interventions in Niger INTERVENTION DELIVERY PLATFORM Antenatal micronutrient supplementation (iron and folic acid only) Health facility and community Complementary feeding education Health facility, community, and communication campaigns Breastfeeding promotion Health facility, community, and communication campaigns Balanced energy protein supplementation for pregnant women Health facility, community, and social safety net programs Intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions Health facility and community Vitamin A supplementation Health facility, community, and food fortification Public provision of complementary foods Health facility, community, and social safety net programs Treatment of severe acute malnutrition Health facility and community Iron and folic acid supplementation for non-pregnant women of reproductive age School, community, health facility, and marketplace Staple food fortification Marketplace Pro-breastfeeding social policies Government policies National breastfeeding promotion campaigns Media Figure 6.5: Coverage of Key Nutrition-Specific Interventions in Niger and Sub-Saharan Africa Antenatal micronutrient supplementation - (iron and folic acid only) Pro-breastfeeding Complementary social policies feeding education 100 80 60 Staple food Breastfeeding fortification 40 promotion 20 Iron and folic acid Balanced energy supplementation for non-pregnant protein supplementation women of reproductive age for pregnant women Treatment of severe Intermittent presumptive acute malnutrition treatment of malaria in pregnancy in malaria-endemic regions Public provision of Vitamin A complementary foods supplementation Note: Red shading represents Niger and light blue shading represents average Sub-Saharan Africa regional coverage CHAPTER 6 Niger: An Investment Framework for Nutrition 89 Economic Benefits of Investing in Nutrition There is a strong body of evidence that shows high economic returns to investing in nutrition (Alderman et al. 2016; Copenhagen Consensus Center 2015; Hoddinott et al. 2013). Scaling up these proven nutrition-specific interventions can ensure that mothers are healthy and well nourished and that they can provide optimal nutri- tion to their children, that children realize their full physical and cognitive development potential, and that women’s productivity is not hampered by illness, especially anemia (Figure 6.6). Figure 6.6: How Reaching the Global Nutrition Targets Generates Economic Benefits Co nitive Development STUNTING Childr n’s H lth & Le rnin & Educ tion l Nutrition St tus Att inment WASTING Glob l Nutrition T r ts R ch d Adult Productivit EASTFEE BRE EDING W es ANEMIA Econom Wom n’s H lth & (GDP) Nutrition St tus In Niger, scaling-up the package of nutrition-specific interventions would produce substantial economic ben- efits over the productive lifetime of the affected women and children (Figure 6.7). Additional health system cost-savings would also be likely because many of these investments reduce the burden of childhood illnesses such as diarrhea and pneumonia. Figure 6.7: Investments in Niger to Meet the Global Nutrition Targets Have Enormous Economic Returns Total Economic $1 Invested Benefits (Billions)a Yieldsb STUNTING $4,136 $10 $ $3,001 $48 1 BREASTFEEDING $ $ ANEMIA $512 $ $7 WASTING $1,492 $11 a. Total economic benefits over 10 years for women and over the productive lives of children who benefit from these interventions, , defined as the period between the age of 18 and a “retirement” age - the life expectancy or the age of 65, whichever is lower. b. Benefit calculation assumes a 3 percent discount rate for both costs and benefits, and GDP growth rate of 3 percent. 90 in Sub-Saharan Africa Stunting Reduction  TOTAL TEN-YEAR FINANCING NEEDS: $2.2 billion Financing Needs, Impacts, and Cost-Effectiveness of Scaling Up Nutrition-Specific Interventions WASTING ANEMIA Using the methodology detailed Total: 1% in An Investment Framework for Nutrition (Shekar et al. 2017), this brief presents Total 12%: 12 high-impact nutrition-specific interventions in estimates of the resources needed to scale up a package ofBREASTFEEDING Niger to meet the global nutrition targets for stunting, anemia, 13% and wasting, along with their breastfeeding, Total: estimated nutrition, health, and economic impacts. An additional $77.1 million per year over 10 years is needed to scale up the package of key interventions (Table 6.3). The health and nutrition impact of this investment is shown in Table 6.4. The scale up of complementary feeding education would be the most cost-effective intervention for the pre- TOTAL TEN-YEAR anda of stunting, averting more than 71,000 cases and costing $17.6 ventionSTUNTING million over 10 years. Prophylactic zinc supplementation Total 82%: would be the most effective intervention for FINANCING stunting, and wouldNEEDS: prevent 194,000 cases over $273 million through counseling 10 years—but this intervention also has the highest total cost. Breastfeeding promotion mothers would be projected to increase the number of infants exclusively breastfed by 1.18 million, prevent 14,000 deaths, and cost $38 million over 10 years. Among the set of nutrition-specific interventions, vitamin A supplementation would be the most cost-effective intervention for preventing child mortality, but would avert only 5,000 deaths over 10 years. Given that acute malnutrition is of serious public health significance in Niger, treatment of wasting would have the highest impact for child mortality, resulting in more than 30,000 deaths averted. For preventing maternal anemia, staple food fortification would be the most cost-effective, at a cost of $4.50 for each case-year of anemia prevented in women. Over 10 years, staple food fortification would prevent nearly 1.7 million case-years of anemia in women at a cost of $7.4 million. Figure 6.8: Ten-Year Financing Needs for Scaling Up a Package of Nutrition-Specific Interventions in Niger by Percent per Intervention ANEMIA Total 10%: WASTING Total: 22% ER STUNTING Total 71%: BREASTFEEDING Total: 8% TOTAL TEN-YEAR FINANCING NEEDS: $771 million Note: Some costs for anemia, breastfeeding, and stunting are shared across interventions. Costs for breastfeeding promotion ($38 million) has been included in both the total cost for the breastfeeding target and the total cost for the stunting target; cost of intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions ($15.3 million) has been included in both the total cost for the anemia target and the total cost for the stunting target. Interventions to reduce stunting would require the most resources,4 accounting for about 71 percent of the total amount required for scale-up. However, some of the stunting interventions would also affect breastfeeding and WASTING anemia targets. Figure 6.8 represents the distribution of total cost across interventions to address the four targets. ANEMIA Total: 5% Total 11%: BREASTFEEDING Total: 11% CHAPTER 6 Niger: An Investment Framework for Nutrition 91 TOTAL TEN-YEAR Table 6.3: Estimated 10-Year Financing Needs and Cost-Effectiveness of Scaling Up Nutrition-Specific Interventions, Niger TOTAL 10-YEAR COST PER CASE COST PER DEATH INTERVENTION (NUTRITION TARGET) FINANCING NEEDS OF STUNTING AVERTED (US $) (US $M) AVERTED (US $) For pregnant women and mothers of infants Antenatal micronutrient supplementation (stunting, anemia) 26.4 5,980 5,143 Infant and young child nutrition counseling (complementary feeding 55.6 3,272 712 education and breastfeeding promotion combined) Complementary feeding education (stunting) 17.6 5,909 245 Breastfeeding promotion (stunting, breastfeeding) 38.0 2,711 5,999 Balanced energy protein supplementation for pregnant women (stunting) 112.7 24,090 19,322 Intermittent presumptive treatment of malaria in pregnancy in malaria- 15.3 4,117 744 endemic regions (stunting, anemia) For infants and young children Vitamin A supplementation (stunting) 5.9 1,194 110 Prophylactic zinc supplementation (stunting) 181.7 11,255 937 Public provision of complementary food (stunting) 148.0 40,458 1,880 Treatment of severe acute malnutrition (wasting) 167.6 5,479 n.a For non-pregnant women and general population Iron and folic acid supplementation for non-pregnant women (anemia) 25.1 n.a 29,719 Staple food fortification (anemia) 7.4 n.a Pro-breastfeeding social policies (breastfeeding) 5.0 n.a n.a National breastfeeding promotion campaigns (breastfeeding) 20.2 n.a n.a TOTAL: 771.1 8,940 1,524 Note: Financing needs and impacts assume a linear scale-up of interventions from current coverage level to 90 percent over five years, then maintained at 90 percent for an additional five years. Unit financing needs for each intervention were drawn from available unit costs from neighboring countries, global costs, or estimates available in the literature. The estimated costs include an additional 12 percent (11 percent for pro-breastfeeding social policies and promotion campaigns) to account for monitoring, evaluation, capacity, and policy development that may be necessary to reach full scale-up of the interventions. The Lives Saved Tool (LiST) was used to estimate the impact of interventions that target pregnant women and children. The impact of interventions that target the general population or non-pregnant women were estimated using a Microsoft Excel model. n.a. = not applicable. It should be noted that the LiST model does not capture potential synergies between specific interventions (e.g. the fact that the impact of behavior change communication interventions may be higher in populations that have access to affordable and diversified foods or in populations with higher levels of educational attainment). Therefore, it is possible that the impact estimates generated using LiST in fact underestimate the true impact of the interventions in some contexts. n.a. = not applicable. Two Alternative Investment Packages Relative to current expenditures on health, the investment required to scale-up the set of effective nutrition-spe- cific interventions may present significant challenges for Niger. In an environment of constrained resources in which Niger may not be able to raise $771.1 million over the next 10 years, two alternative investment packages are laid out for consideration. The Priority Package: The first—the “priority package”—includes interventions that are the most cost-ef- fective; that is, that have the lowest cost per health outcome (e.g., case of stunting averted), and that have 92 in Sub-Saharan Africa Stunting Reduction  Table 6.4: Benefits and Cost-Effectiveness by Investment Package, Niger GLOBAL PRIORITY CATALYZING FULL PACKAGE: BENEFIT All interventions needed TARGET PACKAGE PROGRESS PACKAGE to meet targets $29.7 million/year $46.6 million/year $77.1 million/year in financing need in financing need in financing need Cases of stunting reduced by STUNTING 140,000 188,000 425,000 2025 (vs 2015)a Cases of anemia in women ANEMIA 2.2 million 3 million 7.5 million prevented by 2025 Additional babies BREASTFEEDING 1.2 million 1.2 million 1.2 million breastfed over 10 years Child deaths averted 62,000 68,000 86,000 over 10 years ALL TARGETS Cost per death averted 4,787 6,811 8,940 Cost per case of 738 1,054 1,524 stunting averted a. Total impact of proposed intervention package combined with other health and poverty reduction efforts. well-established global policy guidelines and delivery platforms. Based on those two criteria, the priority package includes antenatal micronutrient supplementation, infant and young child nutrition counseling, intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions, vitamin A supplemen- tation, treatment of severe acute malnutrition, weekly iron and folic acid supplementation for girls 15–19 years of age attending school, and fortification of wheat and maize flour with iron and folic acid. These interventions would be scaled up to full program coverage in the first five years and maintained at full coverage levels for the last five years. This priority package would require an estimated $297 million over 10 years, or $29.7 million annually (see Table 6.4). During the 10 years of scale-up, this package would prevent more than 140,000 cases of stunting and avert 62,000 deaths in children under five years of age. It would also prevent more than 2.2 million case-years of ane- mia in women and result in 1.2 million children under six months of age being exclusively breastfed. The Catalyzing Progress Package: The second alternative—the “catalyzing progress package”—includes scale-up of all interventions in the priority package, plus a phased approach to scaling up public provision of complementary foods, balanced energy protein supplementation, prophylactic zinc supplementation, and weekly iron and folic acid supplementation for women outside of schools. It is assumed that, for the latter set of interventions, during the first five years emphasis will be placed on establishing global guidelines and on operational research to develop effective delivery platforms, or to develop less expensive products or more cost-effective technologies. Financing needs are approximated as the cost of scaling up this set of interventions from 0 to 10 percent coverage only in the first five years. In the subsequent five years it is assumed that the coverage expansion of those interventions will accelerate and reach 60 percent by 2025. This package would require $46.6 million per year, for a total of $466 million over 10 years (Table 6.4). It would prevent 68,000 CHAPTER 6 Niger: An Investment Framework for Nutrition 93 deaths and more than 188,000 cases of stunting among children under age five, increase the number of exclu- sively breastfed children under six months of age by 1.2 million, and prevent more than 3 million case-years of anemia in women. In comparing the relative cost-effectiveness of the three investment packages, the two alternative packages are more cost-effective in preventing deaths and stunting. However, neither is as effective as the full package in making progress toward achieving the stunting, wasting, and anemia targets. The priority and catalyzing prog- ress packages would prevent 62,000 and 68,000 deaths respectively, compared with 86,000 deaths prevented with the full package over 10 years. Under the full package scenario, 425,000 cases of childhood stunting would be prevented, compared with 188,000 cases under the catalyzing progress scenario and 140,000 cases under the priority package scenario. Furthermore, there would be nearly 5.3 million and 4.5 million more case-years of anemia in women under the priority package and catalyzing progress package, respectively. A Call to Action As the world stands on the cusp of the new Sustainable Development Goals, there is an unprecedented oppor- tunity to save children’s lives, build future human capital and cognitive development, and drive faster eco- nomic growth. Scaling up key nutrition interventions during the critical 1,000 day window of early childhood would pay lifelong dividends, translating to healthier societies and more robust economies. If this window is missed, it is missed for life. The additional financing needed to reach the global nutrition targets will require coordinated efforts by all stakeholders and a supportive policy environment. To achieve these targets, Niger will need to increase its general government health expenditure by approximately 40 percent.5 These investments are over and above those needed for improving water and sanitation and for addressing issues around women’s empowerment and food security. Although this level of domestic financing is ambitious, investing in nutrition interventions have significant potential to reduce poverty and boost shared prosperity. Accelerating the reduction of stunting along with family-planning programs to help realize potential of the demographic dividend in Niger will be essential for maximizing the return on investments in early childhood development, in education, and more broadly in policies aimed at fostering and enhancing human capital accu- mulation and job creation. Investing in the early years is even more critical because the Africa region is entering a demographic transition with an expected increase in the working-age population from 54 percent in 2010 to 64 percent in 2090. The scale-up of the key nutrition-specific interventions to reduce stunting is estimated to generate considerable returns in economic benefits over the productive lives of beneficiaries, and is a necessary condition to build human capital through investments in the early years and to harness the potential benefits of the demographic dividend. Endnotes Note: All dollar amounts are U.S. dollars unless otherwise indicated. 1 Information about the Power of Nutrition initiative is available at https://ciff.org/grant-portfolio/ the-power-of-nutrition/. 2 Note that because some funded interventions contribute to more than one target, the sum of funding across the four targets is less than the total funding for each target added together. 3 Current financing by source is from Results for Development Institute and can be found at http://www.inves- tinnutrition.org/. 94 in Sub-Saharan Africa Stunting Reduction  4 Two of the global nutrition targets—those for low birthweight and for child overweight—were not included in the analyses because of insufficient data on the prevalence of low birthweight and a lack of consensus on effective interventions to reach the target for child overweight. 5 WHO National Health accounts database indicates that general government health expenditure in Niger was $191 million in 2014. At that level, government spending on health will need to be increased by 40 percent to accommodate the $77.1 million per year required to scale up the 12 nutrition-specific interventions. References Aguayo, V. M., R. Nair, N. Badgaiyan, and V. Krishna. 2016. “Determinants of Stunting and Poor Linear Growth in Children under 2 Years of Age in India: An In-Depth Analysis of Maharashtra’s Comprehensive Nutrition Survey.” Maternal & Child Nutrition 12 (Suppl. 1): 121–40. Alderman , H ., J. R. Behrman , and C. Puett . 2016 . “ Big Numbers about Small Children: Estimating the Eco- nomic Benefits of Addressing Undernutrition .” World Bank Research Observer 31 ( 2 ). Black, R. E., L. H. Allen, Z. A. Bhutta, L. E. Caulfield, M. de Onis, M. Ezzati, C. Mathers, J. Rivera, and the Maternal and Child Undernutrition Study Group. 2008. “Maternal and Child Undernutrition: Global and Regional Exposures and Health Consequences.” The Lancet 371 (9608): 243–60. Black, R. E., C. G .Victora, S. P. Walker, Z. A. Bhutta, P. Christian, M. de Onis, M. Ezzati, S. Grantham-Mcgregor, J. Katz, R. Martorell, R. Uauy, and the Maternal and Child Nutrition Study Group. 2013. “Maternal and Child Undernutrition and Overweight in Low-Income and Middle-Income Countries.” The Lancet 382: 427–51. Copenhagen Consensus Center. 2015. Smart Development Goals: The Post-2015 Consensus. http://www.copenha- genconsensus.com/sites/default/files/outcomedocument_col.pdf Daniels, M. C. and L. Adair. 2004. “Growth in Young Filipino Children Predicts Schooling Trajectories through High School.” Journal of Nutrition 134: 1439–46. Fink, G., E. Peet, G. Danaei, K. Andrews, D. C. McCoy, C. R. Sudfeld, M. C. Smith Fawzi, M. Ezzati, and W. W. Fawzi. 2016. “Schooling and Wage Income Losses Due to Early-Childhood Growth Faltering in Developing Countries: National, Regional, and Global Estimates.” The American Journal of Clinical Nutrition 104 (1): 104–12. Geesing D. and H. Djibo H. 2001. Niger: Country Pasture/Forage Resources Profile. http://www.fao.org/ag/ agp/agpc/doc/counprof/niger/niger.htm Hoddinott, J., H. Alderman, J. R. Behrman, L. Haddad, and S. Horton. 2013. “The Economic Rationale for Investing in Stunting Reduction.” Maternal and Child Nutrition 9 (Suppl. 2): 69–82. Hoddinott, J., J. A. Maluccio, J. R. Behman, R. Flores, and R. Martorell. 2008. “Effect of a Nutrition Intervention during Early Childhood on Economic Productivity in Guatemalan Adults.” Lancet 371 (9610): 411–16. Horton, S. and R. Steckel. 2013. “Malnutrition: Global Economic Losses Attributable to Malnutrition 1900–2000 and Projections to 2050.” In The Economics of Human Challenges, edited by B. Lomborg, 247–72. Cambridge, U.K.: Cambridge University Press. IFPRI (International Food Policy Research Institute). 2016 . Global Nutrition Report 2016: From Promise to Impact – Ending Malnutrition by 2030. Washington, DC: IFPRI. CHAPTER 6 Niger: An Investment Framework for Nutrition 95 Martorell, R., B. L. Horta, L. S. Adair, A. D. Stein, L. Richter, C. H. D. Fall, S. K. Bhargava, S. K. Dey Biswas, L. Perez, F. C. Barros, C. G. Victora, and Consortium on Health Orientated Research in Transitional Societies Group. 2010. “Weight Gain in the First Two Years of Life Is an Important Predictor of Schooling Outcomes in Pooled Analyses from Five Birth Cohorts from Low- and Middle-Income Countries.” Journal of Nutrition 140: 348–54. NDHS (Niger Demographic and Health Survey). 2006. Niger Demographic Health Survey 2006. Calverton, Maryland, USA: INS/Niger and Macro International (published in 2007). http://dhsprogram.com/publica- tions/publication-FR193-DHS-Final-Reports.cfm Ozaltin, E., K. Hill, and S. V. Subramanian. 2010. “Association of Maternal Stature with Offspring Mortality, Underweight, and Stunting in Low- to Middle-Income Countries.” JAMA 303 (15): 1507–16. Save the Children. 2009. Understanding Household Economy in Rural Niger. London: Save the Children. https:// www.savethechildren.org.uk/sites/default/files/docs/Understanding_HE_in_Rural_Niger_low_res_comp_1. pdf Shekar, M., J. Kakietek, J. D. Eberwein, and D. Walters. 2017. An Investment Framework for Nutrition: Reaching the Global Targets for Stunting, Anemia, Breastfeeding, and Wasting. Directions in Development. Washington, DC: World Bank. doi:10.1596/978-1-4648-1010-7. https://tinyurl.com/InvestmentFrameworkNutrition SUN (Scaling Up Nutrition). 2014. “Niger.” In SUN Movement Compendium. http://docs.scalingupnutrition. org/wp-content/uploads/2014/11/SUN_Compendium_ENG_20141026_06Niger.pdf ———. No date. Niger: Call for Commitments for Nutrition.” Costed Plan Summary. http://docs.scalingupnu- trition.org/wp-content/uploads/2015/06/Niger-Costed-Plan-Summary.pdf UN DESA (United Nations Department of Economic and Social Affairs). 2015. World Population Prospects, the 2015 Revision. https://esa.un.org/unpd/wpp/Download/Standard/Population/ UNDP (United Nations Development Programme). 2016. Human Development Report 2016. New York, NY: UNDP. World Bank. 2013. “Republic of Niger: Joint IDA-IMF Staff Advisory Note on the Economic and Social Development Plan 2012-2015.” Report No. 76241-NE, April 5. http://documents.worldbank.org/curated/ en/106641468324034369/pdf/762410PRSP0P120OFFICIAL0USE0ONLY090.pdf ———. 2014. Republic of Niger: Trends in Poverty, Inequality and Growth 2005-2011. Report No. 89837-NE. Pov- erty Reduction and Economic Management 4, Africa Region. http://documents.worldbank.org/curated/ en/226071468189571689/pdf/89837-WP-P146536-PUBLIC-Box393186B.pdf ———. 2016. World Bank Data Bank (database). World Bank, Washington, DC (accessed May 1, 2017), . http:// data.worldbank.org/indicator/SP.RUR.TOTL.ZS?page=1 ———. 2017. World Development Indicators (database). World Bank, Washington, DC (accessed May 1, 2017), http://data.worldbank.org/data-catalog/world-development-indicators WHO (World Health Organziation). 2014. Comprehensive Implementation Plan on Maternal, Infant and Young Child Nutrition. Geneva: WHO. http://apps.who.int/iris/bitstream/10665/113048/1/WHO_NMH_NHD_14.1_eng. pdf?ua=1 96 in Sub-Saharan Africa Stunting Reduction  CHAPTER 7 RWANDA: An Investment Framework for Nutrition Benefits of Investing in Nutrition Key Messages • Thirty-eight percent of children in Rwanda are chronically malnourished (stunted); this represents a steady but slow decline since 2005, when half of all children were stunted. Acute 183,000 malnutrition (wasting) has declined rapidly between 2000 and cases of stunting prevented 2015 (from 8.3 to 2.2 percent), and Rwanda is among the few in 2025 countries in the world that have passed the 2030 Sustainable Development Goal (SDG) target for wasting. • Although stunting is concentrated in Rwanda’s western districts and among the poor, one-fifth of children in the richest quintile 6,200 are stunted (as compared with nearly half in the poorest quintile), suggesting that factors other than income affect malnutrition. child deaths prevented in 2025 Despite major successes in exclusive breastfeeding rates (almost 90 percent nationwide), suboptimal complementary feeding practices, poor hygiene and sanitation, and high fertility rates all contribute to persistently high rates of stunting. 1.5 MILLION • Scaling up a package of high-impact nutrition-specific case-years of anemia interventions in Rwanda to contribute toward achieving the global nutrition targets would require an additional $27.3 million per in women prevented in 2025 year over 10 years and would yield enormous benefits (see panel on the right). These investments are over and above those needed for improving water and sanitation and for addressing issues around women’s empowerment and food security. 28,000 • This scale-up would require additional financing equivalent to a babies exclusively breastfed 12 percent increase in current government expenditure on health and could be financed from a combination of domestic sources, official development assistance (ODA), and innovative financing sources such as the Power of Nutrition.1 • To finance the nutrition scale-up, two lower-cost scenarios are 45,000 estimated to require between $5.7 and $12.2 million per year over cases of severe wasting treated the next 10 years. In an environment of constrained resources, starting with one of these two scenarios would be a strong first investment, but it would need to be followed by increased investments toward full scale-up to achieve the global nutrition targets. $4–$24 return for every dollar invested • Making these investments in reducing malnutrition would produce around $3.4 billion in economic benefits over the productive lives of women and children and help the Government of Rwanda meet its ambitious Vision 2020 strategy. $3.4 BILLION generated from investments to reduce stunting* *The economic benefits are calculated over the productive lives of the children benefiting from the interventions that prevent stunting. CHAPTER 7 Rwanda: An Investment Framework for Nutrition 97 Investment Case for Nutrition Ensuring optimum nutrition—particularly during the 1,000-day period from pregnancy to a child’s second birthday—can alter an individual’s development trajectory and maximize her or his productive potential. Chronic malnutrition has important lifelong consequences for health and cognitive development. Losses to cognitive development in early childhood resulting from chronic malnutrition are irreversible. Being stunted (low height-for-age) in early childhood is associated with a delayed start at school, reduced schooling attain- ment, and substantially decreased adult incomes at both the individual and country level (Daniels and Adair 2004; Fink et al. 2016; Hoddinott et al. 2008; Martorell et al. 2010). These consequences add up to overall gross domestic product (GDP) losses of 4 to 11 percent in Africa and Asia (Horton and Steckel 2013). Importantly, chronic undernutrition can be transmitted through an inter-generational cycle, where malnourished mothers are more likely to have stunted children (Aguayo et al. 2016; Ozaltin et al. 2010). Investments in nutrition are highly cost-effective and among the best value-for-money development actions (Copenhagen Consensus Center 2015; Hoddinott et al. 2013). An Investment Framework for Nutrition devel- oped by the World Bank in partnership with R4D, 1000 Days, and the Bill & Melinda Gates Foundation esti- mated high returns on every dollar invested in nutrition: from $4 in returns for treating acute malnutrition (wasting) to $11 for preventing stunting, $12 for the treatment and prevention of anemia, and $35 for increasing the prevalence of exclusive breastfeeding (Shekar et al. 2017). Not only do investments in nutrition produce substantial economic benefits, but they also lay the groundwork for the success of investments in other sectors. The Government of Rwanda, with support from ODA and implementation partners, has an expanding portfo- lio of investments in nutrition that is expected to total more than $100 million between 2014 and 2021.2 Investments in the early years—including early life nutrition, early learning and stimulation, and the provision of nurturing care and protection from stress—ensure that all children reach their human potential and con- tribute to the economic growth of their nation. The analysis presented below focuses on high-impact nutri- tion-specific interventions with strong evidence of efficacy in reducing malnutrition, and estimates the financ- ing needs, impacts, and economic benefits of scaling up these interventions in Rwanda. Country Context Rwanda is a small landlocked country in East Africa. The Rwandan population is estimated to be 11.6 million with a growth rate of 2.4 percent, a total fertility rate (TFR) of 4.2, and one of the highest densities on the conti- nent (470.6 people per square kilometer). Although the country’s TFR has declined from a high of 5.8 in 2000 and is currently below the average for Sub-Saharan Africa (4.9), it still remains a significant impediment to achieving progress on malnutrition reduction. Rwanda has a young population: 43 percent are under age 15 and about 4 percent under age five (UN DESA 2015). More than 84 percent of the population resides in rural areas, and agriculture accounts for 70 percent of employ- ment in Rwanda (RDHS 2015). Rwanda’s economy has maintained a steady growth rate over the past five years and is projected to increase from 6.8 percent in 2016 to 7.1 percent in 2017, exceeding average global growth rate projections (World Bank 2016). The Government of Rwanda is prioritizing achieving the ambitious goals laid out in Vision 2020, a guiding national strategy that aims to transform the country into a knowledge-based, service-oriented middle-income country by 2020 (Republic of Rwanda 2012). Rwanda has made substantial progress in economic, social, and human development. The poverty rate ($1.90 2011 PPP) has declined from 60.4 percent in 2010 to 54.3 percent in 2014, driven by a growing working-age labor force and a rapid inter-sectoral shift from work in the agriculture sector to work in other sectors of the economy (World Bank 2016). With agriculture still underlying most liveli- hoods, the population is highly vulnerable to weather shocks (V20, 2017). 98 in Sub-Saharan Africa Stunting Reduction  Nutritional Status in Rwanda Persistently high rates of undernutrition remain a serious human development challenge in Rwanda. Thirty-eight percent—or 640,000—children under age five are stunted and 2 percent are wasted (low weight-for-height), which suggests that wasting, or acute malnutrition, is less of a problem in Rwanda than chronic malnutrition (RDHS 2015). Overall, stunting prevalence has declined since 2000 but remains of high public health significance according to WHO standards. The prevalence of wasting has declined significantly since 2000, and by 2005 Rwanda had reduced wasting to less than 5 percent, overshooting the 2030 global target (Figure 7.1). National estimates mask socioeconomic and geographic disparities in stunting prevalence. Significant differences in stunting remain among children in poorer and wealthier households. In the past 15 years, stunting declined more quickly among children living in households in the top wealth quintile (with a 3.2 percent average annual rate of reduction) when compared with those in the bottom quintile (a 1 percent average annual rate of reduction) (Figure 7.2). Nevertheless, stunting prevalence remains high (21 percent) in the highest wealth quintile, underscor- ing the fact that rising incomes alone are insufficient to eliminate child malnutrition and build sustainable futures. Figure 7.1: Trends in Undernutrition in Rwanda, 2000–2015 Figure 7.2: Disparities in Stunting by Wealth Quintile, 2000–2015 60 60 51.1 51.1 70 70 AMONG CHILDREN UNDER FIVE 48.3 48.3 AMONG CHILDREN UNDER FIVE AMONG CHILDREN UNDER FIVE AMONG CHILDREN UNDER FIVE STUNTING PREVALENCE (%) STUNTING PREVALENCE (%) 50 50 44.2 44.2 60 60 37.9 37.9 PREVALENCE (%) 50 PREVALENCE (%) 40 40 50 25.6 25.6 40 40 30 30 19.2 19.2 30 30 17.3 17.3 20 20 20 20 8.3 8.3 10 10 4.7 4.7 2.8 10 10 2.8 2.2 2.2 0 0 0 0 2000 2000 2005 2005 2010 2010 2014-15 2014-15 2000 2000 2005 2005 2010 2010 2014-15 2014-15 Stuntin Stuntin W stin W stin An mi An mi Poor Poor st st quintil quintil Rich Rich st st quintil quintil N ltion l N tion Source for Figures 7.1 and 7.2: RDHS 2015. The highest rates of stunting are concentrated in the western districts of the country, but stunting levels remain high across almost all districts (Figure 7.3). Out of 30 districts in the country, 25 had a stunting prevalence of high (more than 30 percent) or very high (more than 40 percent) public health significance. Furthermore, seven districts carry the highest burden in terms of absolute number of stunted children. CHAPTER 7 Rwanda: An Investment Framework for Nutrition 99 Figure 7.3: Number of Stunted Children and Stunting Prevalence by District, Rwanda 2014–15 MUSANZE MUSANZE NYAGATARE NYAGATARE NYABIHU BURERA NYABIHU BURERA GICUMBI GATSIBO GICUMBI GATSIBO RUBAVU GAKENKE RUBAVU GAKENKE RULINDO GASABO RULINDO GASABO KAYONZA KAYONZA NGORORERO NGORORERO RUTSIRO RWAMAGANA RUTSIRO RWAMAGANA MUHANGA MUHANGA KAMONYI KICUKIRO KAMONYI KICUKIRO KARONGI NYARUGENGE KARONGI NYARUGENGE RUHANGO NGOMA RUHANGO NGOMA BUGESERA KIREHE BUGESERA KIREHE NYAMASHEKE NYANZA NYAMASHEKE NYANZA NYAMAGABE NYAMAGABE HUYE HUYE RUSIZI RUSIZI GISAGARA GISAGARA NYARUGURU NYARUGURU 29,208 23,250 Stunting Prevalence ≥ 40%: Very high 20–29%: Medium Total Children 21,500 District Boundaries By District(%) 30–39%: High <20%: Low Stunted By District 18,500 12,500 District Boundaries 8,215 Data source: RDHS 2015.|IBRD IBRD 42943 42943 | MAY 2017 These maps were produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denominations MAY 2017 and any other information shown on these maps do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. The prevalence of stunting in Rwanda is higher than the average for the region (35.2 percent) (Figure 7.4), and higher than that of other Sub-Saharan African countries with similar or lower per capita GDP, such as Guinea, Malawi, Uganda, and Zimbabwe (World Bank 2017). Although the nutritional status of children under five can improve with greater economic growth, more active interventions are necessary to capitalize on such gains. Figure 7.4: Prevalence of Stunting and GDP per Capita: Rwanda and Selected Low-Income Countries Burundi 60 Madagascar RWANDA STUNTING PREVALENCE (%) 50 Niger Mozambique Ethiopia Chad 40 DRC Tanzania Malawi Sierra Leone Uganda 30 Guinea Zimbabwe Senegal 20 10 0 500 1000 1500 2000 2500 3000 GDP PER CAPITA (INTERNATIONAL $ PPP) Source: RDHS 2015; World Bank 2017. Wasting, also known as acute malnutrition, is typically classified as either severe or moderate. Wasting can from food insecurity in resource-poor settings with insufficient dietary quality, quantity and diversity, result50 suboptimal 45 breastfeeding, and recurrent episodes of illness such as diarrhea. Wasting prevalence across Sub-Sa- Zambia haran40Africa is second highest in the world, after South Asia, with 13 million children (7.8 percent) suffering STUNTING PREVALENCE (%) Sudan Nigera from acute malnutrition. However, this varies at the country level. Since 2005, Rwanda has shown a consis- 35 Cameroon Lesotho prevalence and has exceeded its goal to reduce wasting to under 5 percent. tently downward trend in wasting 30 In 2014-15, the wasting prevalence in Rwanda was COTE D’LVOIRE 2.2 percent. Swaziland 25 Kenya Mauritania 20 Congo Ghana 15 Sao Tome and Principe 10 5 100 in Sub-Saharan Africa Stunting Reduction  0 500 1500 2500 3500 4500 5500 6500 7500 8500 9500 Micronutrient deficiencies (a form of malnutrition that relates to a deficiency in essential vitamins and minerals needed for body functions and sometimes referred to as hidden hunger) are pervasive in Rwanda. Anemia—a condition caused by inadequate dietary intake of iron, helminth infections, and malaria, among other causes— has cross-generational impacts. Ensuring that women of reproductive age are well nourished sets the stage for their children to achieve optimal nutrition and development. Approximately one in five women of reproduc- tive age in Rwanda is anemic, which not only affects women’s own health, but also contributes to the inter-gen- erational cycle of undernutrition (RDHS 2015). Although more than one in three children (37 percent) 6 to 59 months of age are anemic, there has been a 30 percent reduction since 2005 (RDHS 2015). Rwanda has made significant investments in health services and has seen impressive progress in improv- ing several health outcomes in recent years. However, a multitude of factors have restricted similar levels of improvement in childhood malnutrition. High fertility rates, along with suboptimal complementary feeding practices, poor hygiene and sanitation practices, insufficient access to health services, and inadequate maternal and child care practices remain obstacles to further reducing chronic child malnutrition. Although Rwanda has one of the highest rates of exclusive breastfeeding for children under six months old (87 percent), the transition to complementary foods after this period presents a significant challenge, with only 18 percent of children 6 to 24 months old having an adequate quantity and diversity of complementary foods. Demand- and supply-side barriers influence food consumption and diversity of diet. A recent analysis found that the cost of the most nutritious diet in Burera District is 25 percent greater than the total income of very poor households, thereby persistently limiting access to the range of nutrients required for growth, health, and development during the early years and beyond (Save the Children 2011a). The high cost of food is a significant barrier: an analysis conducted by Save the Children in the same district estimated that 50 percent of very poor households rely on markets to purchase half of their food, largely as a result of limited land to grow a sufficient amount and variety of food. This reliance on purchasing food makes families vulnerable to seasonality and price fluctuations (Save the Children 2011b). Recognizing that other factors beyond poverty and food insecurity put children at risk of chronic malnutrition, effective multisectoral strategies are needed to address undernutri- tion across the country. A recent report found that 21.9 percent of all child mortality and 12.7 percent of grade repetitions in Rwanda were associated with undernutrition. Total annual financing needs associated with undernutrition in 2012 were estimated at 11.5 percent of the Rwandan GDP (503.6 billion Rwandan francs, equivalent to US$826.2 million in 2012) and are driven largely by lost working hours that result from mortality associated with undernutrition and the lower productivity of adults engaged in manual labor, such as agricultural work (WFP 2013). Political Commitment to Reduce Malnutrition The Government of Rwanda, a SUN movement member since 2011, has demonstrated high-level political commitment, multisectoral coordination, and resource allocation to address chronic malnutrition. The govern- ment highlighted food security, nutrition, and early childhood development as foundational issues to address within its Economic and Poverty Reduction Strategy for 2013–2018, which guides midterm actions to achieve the ambitious Vision 2020 goals. Under this strategy, community-based nutrition programs across the coun- try will identify and address poor maternal, infant, and child feeding practices (Republic of Rwanda 2013). Building on this momentum, the Ministry of Health developed the National Food and Nutrition Strategic Plan for 2013–2018 to provide multisectoral and sector-specific strategies to address chronic malnutrition and food security (Ministry of Health 2014). To operationalize this plan, in 2016 the Government of Rwanda approved the Joint Action Plan to Eliminate Malnutrition (JAPEM) and created the National Food and Nutrition Coordi- nation Secretariat (NFNCS) to oversee its implementation (Ministry of Local Government 2016). The commit- ment to reducing malnutrition is also reflected in the inclusion of an indicator to measure chronic malnutrition and wasting in the Rwanda Vision 2020 strategy (Republic of Rwanda 2012). This reflects official recognition that malnutrition is a barrier to achieving the human resource development necessary to match the country’s evolving market labor needs. CHAPTER 7 Rwanda: An Investment Framework for Nutrition 101 Current Financing for Nutrition In 2015 in Rwanda, overseas donors spent a total of $10.1 million on interventions that will contribute to reach- ing the global targets for nutrition. No data on domestic resources were available at the time of publishing.3 This contribution from donors included $6.61 million to reduce stunting, $0.914 million to prevent anemia, $1.85 million to promote breastfeeding, and $1.93 million to treat wasting.4 Global Targets for Nutrition Substantial improvements to the nutritional status of women and children can be realized if adequate invest- ment is made in a set of evidence-based nutrition-specific interventions that ensure optimum nutrition during the critical 1,000 day window between the start of a woman’s pregnancy and the child’s second birthday (Black et al. 2008, 2013). For women, these include interventions to prevent anemia before and during pregnancy as well as those aimed at improving protein energy intake during pregnancy. Interventions targeted toward children and their mothers aim to improve breastfeeding and complementary feeding practices, enhance the micronutrient status of children, and treat acute malnutrition in children. In 2012—to rally the international community around improving nutrition—the 176 members of the World Health Assembly endorsed the first-ever global nutrition targets, focusing on six areas: stunting, anemia, low birthweight, childhood overweight, breastfeeding, and wasting. These targets aim to boost investments in cost-effective inter- ventions, spearhead better implementation practices, and catalyze progress toward reducing malnutrition. The targets for stunting and wasting are enshrined within the United Nations’ Sustainable Development Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030. The 2016 Global Nutrition Report ranked each country’s progress in contributing toward achieving the global targets (Table 7.1) (IFPRI 2016 ).5 Table 7.1: Four World Health Assembly Targets for Nutrition and Rwanda’s Contribution toward Meeting Them RWANDA’S RANK PREVALENCE PROGRESS Reduce the number of stunted children 1 STUNTING* 110/132 37.9 under five by 40% Stuntin Reduce the number of women of 2 ANEMIA 16/185 19.2 reproductive age with anemia by 50% An mi Increase the rate of exclusive breastfeeding  5 BREASTFEEDING 1/141 87.0 in the first six months up to at least 50% Exclusiv br stf din Reduce and maintain childhood wasting 6 WASTING* 26/130 2.2  alnutrition) to less than 5% (acute m W stin LEGEND: Off course, no progress Off course, some progress On course, good progress *Stunting and wasting are included within the United Nations’ Sustainable Development Goal 2 (SDG 2), which commits to ending malnutrition in all its forms by the year 2030 Sources: Nutrition targets from WHO 2014; Rank and progress from IFPRI 2016; Prevalence data from RDHS 2015 Coverage of key nutrition interventions in Rwanda is largely inadequate. Coverage of interventions that ensure optimal maternal nutrition and micronutrient status, which reduce the burden of low birthweight in children, is very low. Although coverage rates are higher for some childhood interventions, they remain well below the lev- els necessary to reduce malnutrition among Rwandan children. Table 7.2 and Figure 7.5 summarize the current coverage and delivery platforms of key nutrition interventions in Rwanda. 102 in Sub-Saharan Africa Stunting Reduction  Table 7.2: Delivery Platforms of Nutrition-Specific Interventions in Rwanda INTERVENTION DELIVERY PLATFORM Antenatal micronutrient supplementation (iron and folic acid only) Health facility and community Complementary feeding education Health facility, community, and communication campaigns Breastfeeding promotion Health facility, community, and communication campaigns Balanced energy protein supplementation for pregnant women Health facility, community, and social safety net programs Intermittent presumptive treatment of malaria in pregnancy in malaria-endemic areas Health facility, community, and food fortification Vitamin A supplementation Health facility, community, and food fortification Public provision of complementary foods Health facility, community, and food fortification Treatment of severe acute malnutrition Health facility and community Iron and folic acid supplementation for non-pregnant women of reproductive age School, community, health facility, and marketplace Staple food fortification Marketplace Pro-breastfeeding social policies Government policies National breastfeeding promotion campaigns Media Figure 7.5: Coverage of Key Nutrition-Specific Interventions: Rwanda and the Sub-Saharan Africa Region Antenatal micronutrient supplementation - (iron and folic acid only) Pro-breastfeeding Complementary social policies feeding education 100 80 60 Staple food Breastfeeding fortification 40 promotion 20 Iron and folic acid Balanced energy supplementation for non-pregnant protein supplementation women of reproductive age for pregnant women Treatment of severe Intermittent presumptive acute malnutrition treatment of malaria in pregnancy in malaria-endemic regions Public provision of Vitamin A complementary foods supplementation Note: Red shading represents Rwanda and light blue shading represents average Sub-Saharan Africa coverage CHAPTER 7 Rwanda: An Investment Framework for Nutrition 103 Economic Benefits of Investing in Nutrition There is a strong body of evidence that shows high economic returns to investing in nutrition (Alderman et al. 2016; Copenhagen Consensus Center 2015; Hoddinott et al. 2013). Scaling up these proven nutrition-specific interventions can ensure that mothers are healthy and well nourished and that they can provide optimal nutri- tion to their children, that children realize their full physical and cognitive development potential, and that women’s productivity is not hampered by illness, especially anemia (Figure 7.6). Figure 7.6: How Reaching the Global Nutrition Targets Generates Economic Benefits Co nitive Development STUNTING Childr n’s H lth & Le rnin & Educ tion l Nutrition St tus Att inment WASTING Glob l Nutrition T r ts R ch d Adult Productivit EASTFEE BRE EDING W es ANEMIA Econom Wom n’s H lth & (GDP) Nutrition St tus In Rwanda, scaling up the package of nutrition-specific interventions would produce substantial economic ben- efits over the productive lifetimes of the affected women and children (Figure 7.7). Additional health system cost-savings would also be likely because many of these investments reduce the burden of childhood illnesses such as diarrhea and pneumonia. Figure 7.7: Investments in Rwanda to Meet the Global Nutrition Targets Have Enormous Economic Total Economic $1 Invested Benefits (Millions)a Yieldsb STUNTING $3,390 $18 $ $88.6 $4 1 BREASTFEEDING $ $ ANEMIA $235 $ $6 WASTING $72.9 $24 a. Total economic benefits over 10 years for women and over the productive lives of children who benefit from these interventions, defined as the period between the age of 18 and a “retirement” age - the life expectancy or the age of 65, whichever is lower. b. Benefit calculation assumes a 3 percent discount rate for both financing needs and benefits and a GDP growth rate of 3 percent. 104 in Sub-Saharan Africa Stunting Reduction  Financing Needs, Impacts, and Cost-Effectiveness of Scaling-Up Nutrition-Specific Interventions Using the methodology detailed in An Investment Framework for Nutrition (Shekar et al. 2017), this brief pres- ents estimates of the resources needed to scale up a package of high-impact nutrition-specific interventions in Rwanda to meet the global nutrition targets for stunting, anemia, breastfeeding, and wasting, along with their estimated nutrition, health, and economic impacts. An additional $27.3 million per year over 10 years is needed to scale up the package of key interventions (Table 7.3). The health and nutrition impacts of this investment are shown in Table 7.4. Table 7.3: Estimated 10-Year Financing Needs and Cost-Effectiveness of Scaling Up Nutrition-Specific Interventions, Rwanda TOTAL 10-YEAR COST PER CASE COST PER DEATH INTERVENTION (NUTRITION TARGET) FINANCING NEEDS OF STUNTING AVERTED (US $) (US $M) AVERTED (US $) For pregnant women and mothers of infants Antenatal micronutrient supplementation (stunting, anemia) 9.9 18,452 6,788 Infant and young child nutrition counseling (complementary feeding 14.7 22,770 327 education and breastfeeding promotion combined) Complementary feeding education (stunting) 12.2 25,978 277 Breastfeeding promotion (stunting, breastfeeding) 2.5 14,154 2,918 Balanced energy protein supplementation for pregnant women (stunting) 47.5 70,024 52,172 Intermittent presumptive treatment of malaria in pregnancy in malaria- 1.9 12,399 260 endemic regions (stunting, anemia) For infants and young children Prophylactic zinc supplementation (stunting) 54.4 28,336 833 Public provision of complementary food (stunting) 96.0 107,055 1,332 Treatment of severe acute malnutrition (wasting) 3.8 5,370 n.a For non-pregnant women and general population Iron and folic acid supplementation for non-pregnant women (anemia) 16.3 n.a 30,331 Staple food fortification (anemia) 3.9 n.a Pro-breastfeeding social policies (breastfeeding) 5 n.a n.a National breastfeeding promotion campaigns (breastfeeding) 20 n.a n.a TOTAL: 273.3 44,059 1,168 Note: Financing needs and impacts assume a linear scale-up of interventions from the current coverage level to 90 percent over five years, then maintained at 90 percent for an additional five years. Unit costs for each intervention were drawn from available unit costs from neighboring countries, global unit costs, or estimates available in the literature. The estimated financing needs include an additional 12 percent (11 percent for pro-breastfeeding social policies and promotion campaigns) to account for monitoring, evaluation, capacity, and policy development that may be necessary to reach full scale-up of the interventions. The Lives Saved Tool (LiST; see LiST 2015) was used to estimate the impacts of interventions that target pregnant women and children. The impacts of interventions that target the general population or non-pregnant women were estimated using a Microsoft Excel model. It should be noted that the LiST model does not capture potential synergies between specific interventions (e.g. the fact that the impact of behavior change communication interventions may be higher in populations that have access to affordable and diversified foods or in populations with higher levels of educa- tional attainment). Therefore, it is possible that the impact estimates generated using LiST in fact underestimate the true impact of the interventions in some contexts. n.a. = not applicable. CHAPTER 7 Rwanda: An Investment Framework for Nutrition 105 Among the set of proposed interventions, complementary feeding education, prophylactic zinc supplementa- tion, and the public provision of complementary foods would be the most effective in preventing stunting, with ANEMIA each averting 44,000, 65,000, and 72,000 cases of stunting, respectively. Although the intermittent presumptive Total 13%: treatment of malaria in pregnancy would be the most cost-effective for preventing stunting, it would prevent fewer than 8,000 cases. Breastfeeding promotion by counseling mothers would be projected to increase the number of infants exclusively breastfed by 28,000, at a cost $88 per child exclusively breastfed, with a total additional financing need of $2.5 million over 10 years. For preventing anemia in women, staple food fortifi- WASTING cation would be the most cost-effective for non-pregnant women, at a cost Total:of 17%$10 for each case-year of anemia prevented. Over 10 years, staple food fortification and iron and folic acid supplementation for non-pregnant women would prevent about 391,000 and 859,000 case-years of anemia in women, respectively, and require $3.9 million and $16.3 million. Among pregnant women, antenatal micronutrient supplementation would prevent STUNTING Ethiopia 264,000 case-years of anemia, at a cost of $37 per case-year prevented, or $9.9 million over 10 years. BREASTFEEDING Total 76%: Total: 13% Interventions to reduce stunting would require the most resources, accounting for over 80 percent of the total amount required for scale-up. However, some of the stunting interventions would also affect the breastfeed- of total financing ing and anemia targets. Figure 7.8 represents the distributionTOTAL TEN-YEAR needs across interventions FINANCING to NEEDS: $2.2 billion address the four targets. Figure 7.8: Ten-Year Financing Needs for Scaling Up a Package of Nutrition-Specific Interventions in Rwanda, by Percent per Intervention WASTING ANEMIA Total: 1% Total 12%: BREASTFEEDING Total: 13% TOTAL TEN-YEAR Rwanda STUNTING Total 82%: FINANCING NEEDS: $273 million Note: Some costs for anemia, breastfeeding, and stunting are shared across interventions. Costs for breastfeeding promotion ($2.5 million) have been included in both the total cost for the breastfeeding target and the total cost for the stunting target; the costs of intermittent presumptive treatment of malaria in pregnancy in malaria-endemic regions ($1.9 million) and antenatal micronutrient supplementation ($9.9 million) have been included in both the total cost for the anemia target and the total cost for the stunting target. Two Alternative Investment Packages ANEMIA Total 10%: In an environment of constrained resources in which Rwanda may not be able to raise $273 million over the next 10 years, two alternative investment packages are laid out for consideration. WASTINGthat are the most cost-effective; The Priority Package: The first—the “priority package”—includes interventions averted), that is, have the lowest cost per health outcome (e.g., case of stunting Total: 22%and that have well-established global policy guidelines and delivery platforms. Based on those two criteria, the priority package includes antenatal micronutrient supplementation, infant and young child nutrition counseling, intermittent presump- NIGER tive treatment of malaria in pregnancy in malaria-endemic regions, the treatment of severe acute malnutrition, STUNTING weekly iron and 71%: Total BREASTFEEDING folic acid supplementation for girls 15–19 years of age attending school, and fortification of Total: 8% 106 in Sub-Saharan Africa Stunting Reduction  TOTAL TEN-YEAR FINANCING NEEDS: $771 million Table 7.4: Benefits and Cost-Effectiveness by Investment Package, Rwanda GLOBAL PRIORITY CATALYZING FULL PACKAGE: BENEFIT All interventions needed TARGET PACKAGE PROGRESS PACKAGE to meet targets $5.7 million/year in $12.2 million/year in $27.3 million/year in financing need financing need financing need Cases of stunting reduced by STUNTING 39,000 75,000 183,000 2025 (vs 2015)a Cases of anemia in women ANEMIA 754,000 955,000 1.5 million prevented by 2025 Additional babies BREASTFEEDING 28,000 28,000 28,000 breastfed over 10 years Child deaths averted 2,400 3,600 6,200 over 10 years ALL TARGETS Cost per death averted 23,034 33,409 44,059 Cost per case of 686 1,171 1,225 stunting averted a. Total impacts of proposed intervention package combined with other health and poverty reduction efforts. wheat and maize flour. These interventions would be scaled up to full program coverage in the first five years and maintained at full coverage levels for the last five years. This priority package would require an estimated $57 million over 10 years, or $5.7 million annually (see Table 7.4). During the 10 years of scale up, this package would prevent more than 39,000 cases of stunting and avert 2,400 deaths in children under five years of age. It would also prevent more than 754,000 case-years of anemia in women and result in 28,000 children under six months of age being exclusively breastfed. The Catalyzing Progress Package: The second alternative—the “catalyzing progress package”—includes scale-up of all interventions in the priority package, plus a phased approach to scaling up public provision of complementary foods, balanced energy protein supplementation, prophylactic zinc supplementation, and weekly iron and folic acid supplementation for women outside of schools. It is assumed that, for the latter set of interventions, during the first five years emphasis will be placed on establishing global guidelines and on oper- ational research to develop effective delivery platforms, or to develop less expensive products or more cost-ef- fective technologies. Financing needs are approximated as the cost of scaling up this set of interventions from 0 to 10 percent coverage only in the first five years. In the subsequent five years, it is assumed that the coverage expansion of those interventions will accelerate and reach 60 percent by 2025. This package would require $12.2 million per year, a total of $122 million over 10 years (Table 7.4). It would prevent 3,600 deaths and more than 75,000 cases of stunting among children under age five, increase the number of exclusively breastfed children under six months of age by 28,000, and prevent more than 955,000 case-years of anemia in women. In comparing the relative cost-effectiveness of the three investment packages, the two alternative packages are more cost-effective in preventing deaths and stunting. However, neither is as effective as the full package in making progress toward achieving the stunting, wasting, and anemia targets. The priority and catalyzing progress packages would prevent 2,400 and 3,600 deaths respectively, compared with 6,200 deaths prevented CHAPTER 7 Rwanda: An Investment Framework for Nutrition 107 with the full package over 10 years. Under the full package scenario, 183,000 cases of childhood stunting would be prevented, compared with 75,000 cases under the catalyzing progress scenario and 39,000 cases under the priority package scenario. Furthermore, there would be nearly 750,000 and 545,000 more case-years of anemia prevented in women under the priority package and catalyzing progress package, respectively. A Call to Action As the world stands on the cusp of the new Sustainable Development Goals, there is an unprecedented oppor- tunity to save children’s lives, build future human capital and cognitive development, and drive faster eco- nomic growth. Scaling up key nutrition interventions during the critical 1,000 day window of early childhood would pay lifelong dividends, translating to healthier societies and more robust economies. If this window is missed, it is missed for life. The additional financing needed to reach the global nutrition targets will require coordinated efforts by all stakeholders and a supportive policy environment. To achieve these targets, Rwanda would require an increase in the funding allocated to nutrition by $27.3 million annually, roughly equivalent to a 12 percent increase in current general government expenditure on health.6 These investments are over and above those needed for improving water and sanitation and for addressing issues around women’s empowerment and food security. Although this level of domestic financing is ambitious, Rwanda is already moving in this direction. In the long term, nutrition interventions have significant potential to reduce poverty and boost shared prosperity. Accelerating the reduction of stunting in Rwanda will be essential for maximizing the return on investments in early childhood development, in education, and more broadly in policies aimed at fostering and enhancing human capital accumulation and job creation. Investing in the early years is even more critical because the Africa region is entering a demographic transition with an expected increase in the working-age population from 54 percent in 2010 to 64 percent in 2090. The scale-up of the key nutrition-specific interventions to reduce stunting is estimated to generate considerable returns in economic benefits over the productive lives of ben- eficiaries, and is a necessary condition to build human capital through investment in the early years and to harness the potential benefits of the demographic dividend. Endnotes Note: All dollar amounts are U.S. dollars unless otherwise indicated. 1 Information about the Power of Nutrition initiative is available at https://ciff.org/grant-portfolio/ the-power-of-nutrition/. 2 ODA is committed from the Children’s Investment Fund Foundation, USAID, the Embassy of the Kingdom of the Netherlands (EKN), UNICEF, the Government of the Netherlands, and the World Food Programme. Imple- menting partners include Catholic Relief Services, Caritas, UNICEF, and the World Food Programme. 3 Current financing by source is from the Results for Development Institute and can be found at http://www. investinnutrition.org/. 4 Note that because some funded interventions contribute to more than one target, the sum of funding across the four targets is less than the total funding for each target added together. 5 Two of the global nutrition targets—those for low birthweight and for child overweight—were not included in the analyses because of insufficient data on the prevalence of low birthweight and a lack of consensus on effective interventions to reach the target for child overweight. 108 in Sub-Saharan Africa Stunting Reduction  6 The WHO National Health accounts database indicates that general government health expenditure in Rwanda was $227 million in 2014. At that level, this will need to be increased by 12% to accommodate the $27.3 million per year required for scale-up of the 11 nutrition-specific interventions. References Aguayo, V. M., R. Nair, N. Badgaiyan, and V. Krishna. 2016. “Determinants of Stunting and Poor Linear Growth in Children under 2 Years of Age in India: An In-Depth Analysis of Maharashtra’s Comprehensive Nutrition Survey.” Maternal & Child Nutrition 12 (Suppl. 1): 121–40. Alderman , H ., J. R. Behrman , and C. Puett . 2016 . “ Big Numbers about Small Children: Estimating the Eco- nomic Benefits of Addressing Undernutrition .” World Bank Research Observer 31 ( 2 ). Black, R. E., L. H. Allen, Z. A. Bhutta, L. E. Caulfield, M. de Onis, M. Ezzati, C. Mathers, J. Rivera, and the Maternal and Child Undernutrition Study Group. 2008. “Maternal and Child Undernutrition: Global and Regional Exposures and Health Consequences.” The Lancet 371 (9608): 243–60. Black, R. E., C. G .Victora, S. P. Walker, Z. A. Bhutta, P. Christian, M. de Onis, M. Ezzati, S. Grantham-Mcgregor, J. Katz, R. Martorell, R. Uauy, and the Maternal and Child Nutrition Study Group. 2013. “Maternal and Child Undernutrition and Overweight in Low-Income and Middle-Income Countries.” The Lancet 382: 427–51. Copenhagen Consensus Center. 2015. Smart Development Goals: The Post-2015 Consensus. http://www.copenha- genconsensus.com/sites/default/files/outcomedocument_col.pdf Daniels, M. C. and L. Adair. 2004. “Growth in Young Filipino Children Predicts Schooling Trajectories through High School.” Journal of Nutrition 134: 1439–46. Fink, G., E. Peet, G. Danaei, K. Andrews, D. C. McCoy, C. R. Sudfeld, M. C. Smith Fawzi, M. Ezzati, and W. W. Fawzi. 2016. “Schooling and Wage Income Losses Due to Early-Childhood Growth Faltering in Developing Countries: National, Regional, and Global Estimates.” The American Journal of Clinical Nutrition 104 (1): 104–12. Hoddinott, J., H. Alderman, J. R. Behrman, L. Haddad, and S. Horton. 2013. “The Economic Rationale for Investing in Stunting Reduction.” Maternal and Child Nutrition 9 (Suppl. 2): 69–82. Hoddinott, J., J. A. Maluccio, J. R. Behman, R. Flores, and R. Martorell. 2008. “Effect of a Nutrition Intervention during Early Childhood on Economic Productivity in Guatemalan Adults.” Lancet 371 (9610): 411–16. Horton, S. and R. Steckel. 2013. “Malnutrition: Global Economic Losses Attributable to Malnutrition 1900–2000 and Projections to 2050.” In The Economics of Human Challenges, edited by B. Lomborg, 247–72. Cambridge, U.K.: Cambridge University Press. IFPRI (International Food Policy Research Institute). 2016. Global Nutrition Report 2016: From Promise to Impact – Ending Malnutrition by 2030. Washington, DC: IFPRI. Martorell, R., B. L. Horta, L. S. Adair, A. D. Stein, L. Richter, C. H. D. Fall, S. K. Bhargava, S. K. Dey Biswas, L. Perez, F. C. Barros, C. G. Victora, and Consortium on Health Orientated Research in Transitional Societies Group. 2010. “Weight Gain in the First Two Years of Life Is an Important Predictor of Schooling Outcomes in Pooled Analyses from Five Birth Cohorts from Low- and Middle-Income Countries.” Journal of Nutrition 140: 348–54. CHAPTER 7 Rwanda: An Investment Framework for Nutrition 109 Ministry of Health, Republic of Rwanda. 2014. National Food and Nutrition Strategic Plan, 2013–2018. Kigali: Republic of Rwanda. http://www.moh.gov.rw/fileadmin/templates/policies/National_Food_and_Nutrition_ Strategic_Plan_2013-2018-18_scanned.pdf Ministry of Local Government, Republic of Rwanda. 2016. Revised Joint Action Plan to Eliminate Malnutrition (2016-2020). Ozaltin, E., K. Hill, and S. V. Subramanian. 2010. “Association of Maternal Stature with Offspring Mortality, Underweight, and Stunting in Low- to Middle-Income Countries.” JAMA 303 (15): 1507–16. RDHS (Rwanda Demographic Health Survey). 2015. Rwanda: 2014-15 Demographic Health Survey Key Findings. Rockville, Maryland, USA: NISR, MOH, and ICF International. http://dhsprogram.com/pubs/pdf/SR229/ SR229.pdf Republic of Rwanda. 2012. Rwanda Vision 2020, Revised 2012. http://www.minecofin.gov.rw/fileadmin/tem- plates/documents/NDPR/Vision_2020_.pdf Save the Children. 2011a. A Cost of Diet Analysis in a Northern Highland District of Rwanda. Save the Children, UK. https://www.google. com/#q=A+cost+of+diet+analysis+in+a+northern+highland+district+of+Rwanda&spf=1495722395282 ———. 2011b. Household Economy Analysis Rwanda. Save the Children, UK. Shekar, M., J. Kakietek, J. D. Eberwein, and D. Walters. 2017. An Investment Framework for Nutrition: Reaching the Global Targets for Stunting, Anemia, Breastfeeding, and Wasting. Directions in Development. Washington, DC: World Bank. doi:10.1596/978-1-4648-1010-7. https://tinyurl.com/InvestmentFrameworkNutrition UN DESA (United Nations Department of Economic and Social Affairs). 2015. World Population Prospects, the 2015 Revision. https://esa.un.org/unpd/wpp/Download/Standard/Population/ V20. 2017. Climate Vulnerable Forum. http://www.v-20.org. Accessed May, 2017. WFP (World Food Programme). 2013. The Cost of Hunger in Rwanda: Social and Economic Impacts of Child Under- nutrition in Rwanda. Implications on National Development and Vision 2020. http://documents.wfp.org/stellent/ groups/public/documents/newsroom/wfp263106.pdf?_ga=1.118450383.790854755.1490990329 WHO (World Health Organization). 2014. Comprehensive Implementation Plan on Maternal, Infant and Young Child Nutrition. Geneva: WHO. http://apps.who.int/iris/bitstream/10665/113048/1/WHO_NMH_NHD_14.1_eng. pdf?ua=1 World Bank. 2016. Rwanda Economic Update: Rwanda at Work. Edition No. 9. Kigali: World Bank. http://docu- ments.worldbank.org/curated/en/126991468197359041/pdf/103618-WP-P156677-PUBLIC-Rwanda-Econom- ic-Update-Feb-25-2016.pdf ———. 2017. World Development Indicators (database), World Bank, Washington, DC (accessed May 1, 2017), http://data.worldbank.org/data-catalog/world-development-indicators 110 in Sub-Saharan Africa Stunting Reduction  ANNEX 1: Income Elasticity Tables Table A.1: Data Sources Determinant Category Variable Definition details Source Outcome Chronic undernutrition Prevalence of stunting, height for age (% of children Stunting rate (%) JME* under 5) Basic Macro-fiscal Constant GNI per capita (USD) Constant GNI per capita (USD) WDI GNI per cap. growth (%) GNI per capita growth (annual %) WDI Negative GNI per capita growth recorded in a year Economic recession episodes WDI (calculated) Herfindhal index calculated over three sectors--value WDI Economic diversification (1=high added agriculture, industry, and services (each expressed ; 0=low) WDI as % of GDP) General government health expenditure (GGHE) as % Public health exp. (% GDP) WHO GHED of GDP General government health expenditure (GGHE) as % of Public health exp. (% GGE) WHO GHED general government expenditure Public education exp. (% GDP) Government expenditure on education (% of GDP) WDI Inclusiveness GINI index GINI index (World Bank estimate) WDI Share bottom 40% Income share held by bottom 40% of population WDI Governance Control of corruption measured on a scale of -2.5 (weak) Control of corruption WGI to 2.5 (strong) Government effectiveness measured on a scale of -2.5 Government effectiveness WGI (weak) to 2.5 (strong) Political stability and absence of violence/terrorism Political stability WGI measured on a scale of -2.5 (weak) to 2.5 (strong) Regulatory quality measured on a scale of -2.5 (weak) to Regulatory quality WGI 2.5 (strong) Rule of law measured on a scale of -2.5 (weak) to 2.5 Rule of law WGI (strong) Voice and accountability measured on a scale of -2.5 Voice and accountability WGI (weak) to 2.5 (strong) Conflicts State Fragility Index (effectiveness score+legitimacy State fragility index CSP score=25 points max) Total of magnitude scores of major international and civil Violence episodes CSP politically violent episodes Number of conflicts (involving goverments) that occured Number of conflicts (UCDP) UCDP this year with at least 25 deaths recorded Underlying Food security Food availability as measured in kilocalories per capita Food supply availability FAO per day Depth of food deficit Food deficit as measured in kilocalories per capita per day WDI Household environment Improved sanitation facilities (% of population with Improved sanitation WDI access) Improved water source Improved water source (% of population with access) WDI JME=Joint Malnutrition Estimates (UNICEF, WHO, The World Bank); WDI=World Development Indicators; WHO GHED=World Health Organization Government Health Expenditure Database; WGI=Worldwide Governance Indicators; CSP=Center for Systemic Peace; UCDP=Uppsala Conflict Data Program; FAO=United Nations Food and Agriculture Organization * JME was prioritized for the stunting rate variable. However, when stunting data was missing for a year, these sources were used: WDI, HealthStats (Health, Nutrition and Population Statistics), WHO, and DHS (Demographic and Health Surveys). 112 in Sub-Saharan Africa Stunting Reduction  Table A.2: Descriptive Statistics T-TEST: H0: (2) - (3) = 0 MEAN MEAN MEAN NON-SSA SSA (1) t-stat p-value (2) (3) (4) (5) Stunting rate (%) 30.1 25.8 38.7 -13.6 0.0 Constant GNI per capita (USD) 10791.3 13619.9 1860.8 45.1 0.0 Constant GDP per capita (PPP) 15077.7 18727.3 3965.9 39.8 0.0 GNI per cap. growth (%) 2.1 2.3 1.5 3.8 0.0 GDP per cap. growth (%) 2.1 2.4 1.3 5.9 0.0 Economic recession episodes 0.1 0.1 0.2 -13.8 0.0 Economic diversification (1=high ; 0=low) 0.5 0.5 0.4 30.6 0.0 Rents (% GDP) 9.5 8.4 12.3 -10.1 0.0 Agriculture sector (% GDP) 18.9 13.8 30.9 -41.0 0.0 Trade openned (X+M % GDP) 78.4 81.6 69.6 10.2 0.0 Public health exp. (% GDP) 3.7 4.2 2.4 26.6 0.0 Public health exp. (% GGE) 11.2 11.7 9.8 12.4 0.0 Public education exp. (% GDP) 4.4 4.4 4.4 0.7 0.5 GINI index 39.8 39.1 44.7 -7.7 0.0 Share bottom 40% 17.0 17.2 14.9 7.1 0.0 Female to male life expectancy ratio 1.1 1.1 1.1 25.8 0.0 Female having completed lower secondary 64.6 67.8 28.5 11.0 0.0 education (%) Control of corruption 0.0 0.2 -0.6 26.5 0.0 Government effectiveness 0.0 0.2 -0.8 32.8 0.0 Political stability 0.0 0.1 -0.6 17.8 0.0 Regulatory quality 0.0 0.2 -0.7 29.7 0.0 Rule of law 0.0 0.2 -0.7 30.0 0.0 Voice and accountability 0.0 0.2 -0.6 24.1 0.0 State fragility index 9.4 7.1 15.4 -41.1 0.0 Violence episodes 0.7 0.7 0.8 -2.0 0.0 # Conflicts (UCDP) 1.4 1.4 1.2 5.3 0.0 Food supply availability (kcal/cap/d) 2567.2 2694.5 2187.9 54.5 0.0 Depth of food deficit 137.8 111.2 188.6 -17.2 0.0 Coverage Vitamin A (% pop) 60.7 63.5 58.8 2.1 0.0 Open defecation (% pop) 12.9 6.8 31.1 -33.1 0.0 Improved sanitation (% pop) 69.5 81.5 30.8 70.5 0.0 Improved water source (% pop) 84.4 90.6 64.4 45.9 0.0 ANNEXES 113 Table A.3: Summary of Existing Empirical Evidence NUTRITIONAL OUTCOME STUDY INCOME ELASTICITY INCOME VARIABLE SAMPLE ESTIMATION METHOD VARIABLE "1970-1995 Haddad et al. (2002) -0.51 (long-run) GDP per capita Underweight prevalence OLS, country fixed effects 61 countries” "OLS, country fixed % change in GDP per Change in stunting rates “1985-2009 Headey (2013) -0.18 effects (first differences capita (percentage points) 115 countries” approach)” % change in GDP per Change in stunting rate OLS (first differences Heltberg (2009)* -0.2 1969-2000 capita (percentage points) approach) "-1.07 (urban bottom quintile) % change in GNI per “1985-86, 1986-87 Sahn (1994) Height-for-age z-score OLS -0.28 (rural bottom capita" Côte d’Ivoire” quintile)” Log of per capita “1970-1996 63 Smith & Haddad (2000) -1.26 Underweight prevalence OLS, country fixed effects household expenditure countries” “1970-1996 63 Smith & Haddad (2002) -0.63 GDP per capita Underweight prevalence OLS, random effects countries” “Long-run: 2SLS/IV, “-0.63 (long-run) “1970-2012 country fixed effects; Smith & Haddad (2015) GDP per capita Stunting prevalence -0.17 (short-run)” 116 countries” Short-run: OLS, first differences” “1980-2007 Webb & Block (2010) -0.32 GDP per capita Stunting prevalence OLS 29 countries” 114 in Sub-Saharan Africa Stunting Reduction  Table A.4: Regression Results (Pooled OLS and IV-2SLS) DEPENDENT: LN(STUNTING) POOLED OLS (1) POOLED OLS (2) IV-2SLS (3) POOLED OLS (4) POOLED OLS (5) GDP per cap. (Ln) -0.504*** -0.500*** -0.463*** -0.566*** -0.568*** (0.0158) (0.0159) (0.0406) (0.0192) (0.0192) Interaction GDP*SSA 0.377*** (0.0314) Interaction GDP*South SSA 0.476*** (0.0316) Interaction GDP*West SSA 0.419*** (0.0484) Interaction GDP*Central SSA 0.445*** (0.0335) Interaction GDP*East SSA 0.328*** (0.0363) Sub-Saharan Africa -2.560*** (0.215) Southern SSA -3.268*** (0.248) Western SSA -2.935*** (0.320) Central SSA -2.954*** (0.230) Eastern SSA -2.152*** (0.236) Constant 6.982*** 7.034*** 6.786*** 7.549*** 7.566*** (0.109) (0.109) (0.299) (0.141) (0.141) INCOME ELASTICITIES Non-SSA . . . -0.566*** -0.568*** . . . (0.0192) (0.0192) SSA . . . -.189*** . . . . (.025) . South SSA . . . . -.092*** . . . . (.025) West SSA . . . . -.149*** . . . . (.045) Central SSA . . . . -.123*** . . . . (.028) East SSA . . . . -.24*** . . . . (.031) ENDOGENEITY TESTS Sargan-Hansen J-stat. . . .639 .. .. p-value . . .424 KP LM-stat . . 94.639 .. .. p-value . . 0 KP F-stat . . 36.479 .. .. KP F-critical value 10% . . 19.93 Wu-Hausman F-test . . .295 .. .. p-value . . .587 Durbin Wu-Hausman Chi2-test . . .296 .. .. p-value . . .586 Observations 808 808 728 808 808 R-squared 0.607 0.617 0.617 0.662 0.672 Time trend NO YES YES YES YES Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Instrumental variables for income include M2 as % of GDP, and the investment share of the economy ANNEXES 115 Table A.5: Panel Regression Results (Random/Fixed/Mixed Effects) PANEL REGRESSION MODELS DEPENDENT: RANDOM EFFECTS FIXED EFFECTS MIXED EFFECTS LN(STUNTING) (1) (2) (3) (4) (5) (6) -0.552*** -0.549*** -0.498*** -0.489*** -0.546*** -0.544*** GDP per cap. (Ln) (0.0319) (0.0319) (0.0458) (0.0460) (0.0309) (0.0305) 0.348*** 0.318*** 0.341*** Interaction GDP*SSA (0.0550) (0.0694) (0.0529) Interaction GDP*South 0.490*** 0.465*** 0.483*** SSA (0.133) (0.155) (0.128) Interaction GDP*West 0.314*** 0.253** 0.313*** SSA (0.0976) (0.108) (0.0958) Interaction GDP*Central 0.371*** 0.287* 0.373*** SSA (0.110) (0.146) (0.102) 0.323*** 0.296** 0.313*** Interaction GDP*East SSA (0.106) (0.117) (0.104) -2.256*** -2.230*** Sub-Saharan Africa (0.413) (0.464) -3.268*** -3.319*** Southern SSA (1.076) (1.042) -2.115*** -2.213*** Western SSA (0.673) (0.677) -2.286*** -2.410*** Central SSA (0.874) (0.829) -1.988*** -2.037*** Eastern SSA (0.701) (0.706) 7.335*** 7.307*** 6.305*** 6.293*** 7.314*** 7.317*** Constant (0.264) (0.264) (0.294) (0.299) (0.273) (0.268) INCOME ELASTICITIES -0.552*** -0.549*** -0.498*** -0.489*** -0.546*** -0.544*** Non-SSA (0.0319) (0.0319) (0.0458) (0.0460) (0.0309) (0.0305) -.204*** . -.181*** . -.205*** . SSA (.048) . (.061) . (.046) . . -.058 . -.024 . -.062 South SSA . (.131) . (.152) . (.126) . -.235 . -.236** . -.232** West SSA . (.093) . (.1) . (.092) . -.177 . -.201 . -.171* Central SSA . (.106) . (.141) . (.098) . -.226*** . -.193* . -.232** East SSA . (.105) . (.116) . (.102) Observations 808 808 808 808 808 808 Sargan-Hansen statistic 2.961 3.646 . . . . p-value .398 .724 . . . . Time Trend YES YES YES YES YES YES R-squared . . .461 .46 . . Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 116 in Sub-Saharan Africa Stunting Reduction  Table A.6: Quantile Regression Results QUANTILE REGRESSION MODELS DEPENDENT: 25th percentile 50th percentile 75th percentile LN(STUNTING) (1) (2) (3) (4) (5) (6) -0.662*** -0.666*** -0.540*** -0.542*** -0.418*** -0.428*** GNI per cap. (Ln) (0.0303) (0.0290) (0.0185) (0.0180) (0.0165) (0.0158) 0.446*** 0.368*** 0.263*** Interaction GNI*SSA (0.0649) (0.0396) (0.0354) Interaction GNI*South 0.558*** 0.441*** 0.359*** SSA (0.157) (0.0977) (0.0858) Interaction GNI*West 0.459*** 0.401*** 0.330*** SSA (0.144) (0.0896) (0.0787) Interaction GNI*Central 0.472*** 0.377*** 0.357*** SSA (0.120) (0.0743) (0.0653) 0.399** 0.264*** 0.200** Interaction GNI*East SSA (0.164) (0.102) (0.0892) -2.969*** -2.520*** -1.830*** Sub-Saharan Africa (0.460) (0.281) (0.251) -3.694*** -3.075*** -2.643*** Southern SSA (1.220) (0.758) (0.665) -3.177*** -2.839*** -2.374*** Western SSA (0.966) (0.600) (0.526) -3.082*** -2.560*** -2.467*** Central SSA (0.904) (0.561) (0.493) -2.547** -1.793*** -1.405** Eastern SSA (1.042) (0.647) (0.568) 8.013*** 8.055*** 7.404*** 7.423*** 6.756*** 6.815*** Constant (0.244) (0.234) (0.149) (0.145) (0.133) (0.128) INCOME ELASTICITIES -0.662*** -0.666*** -0.540*** -0.542*** -0.418*** -0.428*** Non-SSA (0.0303) (0.0290) (0.0185) (0.0180) (0.0165) (0.0158) -.216*** . -.172*** . -.155*** . SSA (.058) . (.035) . (.031) . . -.108 . -.101 . -.07 South SSA . (.155) . (.096) . (.084) . -.208 . -.141 . -.098 West SSA . (.142) . (.088) . (.077) . -.195* . -.165** . -.072 Central SSA . (.116) . (.072) . (.063) . -.268* . -.278*** . -.228*** East SSA . (.161) . (.1) . (.088) Observations 808 808 808 808 808 808 Time trend YES YES YES YES YES YES Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 ANNEXES 117 Table A.7a: Mixed Effects Regression Results DEPENDENT: MULTILEVEL MIXED-EFFECTS MODEL LN(STUNTING) (1) (2) (3) (4) (5) (6) (7) (8) -0.546*** -0.546*** -0.552*** -0.514*** -0.564*** -0.545*** -0.514*** -0.419*** GDP per cap. (Ln) (0.031) (0.034) (0.032) (0.047) (0.045) (0.031) (0.036) (0.039) 0.341*** 0.323*** 0.345*** 0.265*** 0.334*** 0.334*** 0.369*** 0.266*** Interaction GDP*SSA (0.053) (0.059) (0.055) (0.083) (0.076) (0.053) (0.061) (0.055) -2.230*** -2.179*** -2.264*** -1.674*** -2.180*** -2.204*** -2.444*** -1.822*** Sub-Saharan Africa (0.464) (0.496) (0.477) (0.623) (0.581) (0.455) (0.499) (0.448) Past economic recession (GNI pc 0.029 growth<0) (0.023) Diversification of GDP -1.018*** (1-Herfindhal) (0.278) 0.000 Trade (% of GDP) (0.000) Lagged public health exp. -0.034* (%GDP) (0.019) Lagged public education exp. -0.029* (%GDP) (0.016) 0.002 GINI coefficient (0.003) Female/Male life expectancy -0.883 ratio (0.551) (Ln) Food supply (kcal/capita/ -0.422** day) (FAO) (0.165) People practicing open -0.002 defecation (% of population) (0.002) Improved sanitation facilities (% -0.007*** of population with access) (0.002) Improved water source (% of -0.001 population with access) (0.002) 7.314*** 7.776*** 7.358*** 7.305*** 7.379*** 8.258*** 10.374*** 6.923*** Constant (0.273) (0.292) (0.278) (0.383) (0.383) (0.645) (1.177) (0.306) INCOME ELASTICITIES -0.546*** -0.546*** -0.552*** -0.514*** -0.564*** -0.545*** -0.514*** -0.419*** Non-SSA (0.031) (0.034) (0.032) (0.047) (0.045) (0.031) (0.036) (0.039) -.205*** -.224*** -.207*** -.249*** -.249*** -.249*** -.249*** -.249*** SSA (.046) (.053) (.048) (.071) (.071) (.071) (.071) (.071) RANDOM-EFFECTS PARAMETERS Standard deviation .222*** .22*** .22*** .127*** .183*** .204*** .202*** .178*** (region effect) (.077) (.077) (.077) (.073) (.072) (.074) (.072) (.074) Standard deviation .457*** .457*** .458*** .42*** .394*** .458*** .432*** .434*** (country effect) (.03) (.031) (.03) (.034) (.032) (.03) (.03) (.03) .215*** .211*** .216*** .183*** .212*** .214*** .214*** .19*** Standard deviation (residual) (.006) (.006) (.006) (.01) (.008) (.006) (.006) (.006) Observations 808 738 779 289 476 806 717 690 Time trend YES YES YES YES YES YES YES YES Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 118 in Sub-Saharan Africa Stunting Reduction  Table A.7b: Mixed Effects Regression Results (continued) DEPENDENT: MULTILEVEL MIXED-EFFECTS MODEL LN(STUNTING) (1) (2) (3) (4) (5) (6) (7) (8) (9) -0.546*** -0.448*** -0.432*** -0.493*** -0.467*** -0.444*** -0.477*** -0.564*** -0.483*** GDP per cap. (Ln) (0.031) (0.041) (0.043) (0.040) (0.041) (0.041) (0.038) (0.030) (0.050) 0.341*** 0.272*** 0.262*** 0.285*** 0.280*** 0.276*** 0.338*** 0.385*** 0.327*** Interaction GDP*SSA (0.053) (0.063) (0.063) (0.064) (0.064) (0.063) (0.056) (0.051) (0.084) -2.230*** -1.643*** -1.586*** -1.758*** -1.709*** -1.687*** -2.164*** -2.541*** -2.049*** Sub-Saharan Africa (0.464) (0.529) (0.531) (0.526) (0.527) (0.528) (0.472) (0.453) (0.649) -0.103*** Control of corruption (0.038) -0.127*** Government effectiveness (0.045) Political stability and 0.005 absence of violence (0.024) -0.053 Regulatory quality (0.037) -0.123*** Rule of law (0.041) 0.011** State fragility index (0.006) 0.000 Political violence episodes (0.007) -0.033 Number of conflicts (0.023) 7.314*** 6.613*** 6.479*** 6.986*** 6.767*** 6.556*** 6.704*** 7.480*** 6.924*** Constant (0.273) (0.345) (0.363) (0.334) (0.349) (0.349) (0.349) (0.265) (0.416) INCOME ELASTICITIES -0.546*** -0.448*** -0.432*** -0.493*** -0.467*** -0.444*** -0.477*** -0.564*** -0.483*** Non-SSA (0.031) (0.041) (0.043) (0.040) (0.041) (0.041) (0.038) (0.030) (0.050) -.205*** -.177*** -.17*** -.208*** -.187*** -.168*** -.139*** -.179*** -.156* SSA (.046) (.055) (.056) (.057) (.057) (.056) (.051) (.044) (.081) RANDOM-EFFECTS PARAMETERS Standard deviation .222*** .222*** .222*** .204*** .208*** .227*** .205*** .228*** .267*** (region effect) (.077) (.08) (.081) (.079) (.079) (.081) (.073) (.074) (.114) Standard deviation .457*** .441*** .443*** .458*** .453*** .437*** .41*** .402*** .357*** (country effect) (.03) (.03) (.03) (.031) (.031) (.03) (.03) (.027) (.044) Standard deviation .215*** .187*** .187*** .187*** .187*** .188*** .185*** .209*** .145*** (residual) (.006) (.007) (.007) (.007) (.007) (.007) (.006) (.006) (.01) Observations 808 526 525 525 525 526 593 777 168 Time trend YES YES YES YES YES YES YES YES YES Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 ANNEXES 119 ANNEX 2: Brief Review of Empirical Evidence of the Determinants of Undernutrition To better understand the drivers of child undernutrition, past studies have conducted cross-country regressions and/or run analyses on country-level data. Immediate Causes Because of the lack of child-level data on dietary intake and health status (two immediate determinants), the econometric analyses have focused on the underlying and basic determinants as defined in UNICEF’s concep- tual framework on the factors determining child nutritional status. Underlying Causes The main underlying determinants considered in past studies of child undernutrition include the following: food security, resources for/knowledge of the caregiver, and resources in the household environment. Food Security National food supply is the main variable used in this category of underlying causes since the indicator is readily available. The literature calls for better measurement of food security since national food availability does not encompass food access/distribution or intake. Meanwhile, studies continue to use food availability to identify a “food secure” country without considering differences between regions, communities, households, and within households. Dietary energy availability/food supply (Frongillo, de Onis, and Hanson 1997; Smith and Haddad 2000, 2002, 2015) and food production per capita (Headey 2012) are measures employed by previ- ous studies; these variables are shown to have significant positive effects on nutritional outcomes. Resources for/Knowledge of the Caregiver The most common measure of adequacy/quality of care for a child is female/mother’s education or literacy rate. A couple of micro-level studies use the number of adult women in the household: (Sahn 1994) with data from Côte d’Ivoire and (Marini and Gragnolati 2003) with data from Guatemala. Both cross-country and country-level results point to the importance of female education/literacy to reduce stunting prevalence (Christiaensen and Alderman 2001; Delpeuch et al. 2000; Frongillo, de Onis, and Hanson 1997; Headey 2012; Kabubo-Mariara, Ndenge, and Kirii 2006; Milman et al. 2005; Marini and Gragnolati 2003; Sahn 1994; Smith and Haddad 2000, 2002, 2015). This variable has the most consistent results; more education for the caregiver will always result in better nutritional outcomes for children. Enabling Environment Resources This underlying determinant encompasses improved sanitation, improved water sources, and health care access for households. Safe water has been found to be a significant driver of stunting reduction (Christiaensen and Alderman 2001; Milman, et al. 2005; Marini and Gragnolati 2003; Smith and Haddad 2000, 2002, 2015). However, household infrastructure variables, such as safe water, have not been consistently significant across studies. For example, Headey (2012) found that improved sanitation, improved water, and electricity had no effect on nutritional outcomes (Headey 2012). Likewise, Brown et al. (2017) found no significant effect from improved sanitation and water (Brown, Ravallion, and van de Walle 2017). 120 in Sub-Saharan Africa Stunting Reduction  A key determinant belonging to this category, health care access, is generally not included in cross-country studies because of the lack of quality data. Thus, infrastructure quality variables—easily available and univer- sally recorded in household surveys—are relied upon to evaluate the local environment/community resources. Only Headey (2012) decided to include two variables that proxy for health care access: percentage of births in a medical facility and percentage of children vaccinated (Headey 2012). Country-level studies, however, may have adequate data to assess whether access to health care significantly contributes to child nutrition. Glewwe et al. (2004) use data from Vietnam and show that distance to the nearest medical facility and the availability of oral rehydration salts are significant factors that reduce stunting (Glewwe, Koch, and Nguyen 2004). Basic Causes Following the conceptual framework, the basic determinants fall into five main categories: income, other mac- ro-fiscal, inclusiveness, governance, and conflict. Income Most cross-country studies on the drivers of nutritional outcomes use GDP per capita as the main income variable; only Heltberg uses change in GNI per capita as the main independent variable (Heltberg 2002, 2009). There seems to be a general consensus that increases in income have a positive influence on stunting rates, but also that growth-promoting policies should be complemented with direct health-related interventions. Overall, income per capita has a significant effect on nutritional status of children (Alderman et al. 2001; Headey 2012; Heltberg 2009; Smith and Haddad 2000, 2002, 2015). The brief spotlight on Sub-Saharan Africa in Smith & Haddad (2000), in contrast, concluded that, because of declining national income in the region between 1975 and 1990, national income did not have a positive influence on undernutrition reduction (Smith and Haddad 2000). Heltberg’s 2009 study had a region-by-region analysis that concluded that there indeed was a small but significant effect of income per capita on stunting rates across all regions, including Sub-Saharan Africa, although no formal statistical test was provided to assess whether the income elasticity in Sub-Saharan Africa was different than in other regions of the world (Heltberg 2009). Income composition has also been considered in past cross-country studies with mixed results. Webb and Block (2012) suggest that increasing agricultural income per capita would directly improve nutritional outcomes (Webb and Block 2012). However, Headey (2012) found the opposite effect: non-agricultural growth (measured as a percentage change of GDP per capita) had a significant effect on reducing stunting (Headey 2012). A significant income effect on improving stunting rates has further been found in micro level studies. These country-level analyses use household consumption/expenditure per capita (or per adult equivalent) or an asset index. They support the general findings from cross-country regressions that increasing income per capita at the household-level improves child nutritional outcomes. Table A2.1 identifies these micro-level papers and their countries of study. It is important to note that a few studies found a minimal or no significant association between income and child nutrition. Using Demographic and Health Survey data from 36 countries and income data from the Penn World Tables, Vollmer et al. (2014) found a very small association between income and nutritional outcomes (Vollmer, et al. 2014). Additionally, two country-level studies did not find a significant effect: (1) Subra- manyam et al. (2011) found no association between state-level economic growth and undernutrition in India ANNEXES 121 (Subramanyam, et al. 2011); and (2) Glewwe et al. (2004) found no significant effect of household per capita expenditures on nutrition in Vietnam (Glewwe, Koch, and Nguyen 2004). Economic diversification is also an uncommon variable in these cross-country studies. One study (Milman et al. 2005) concluded that countries with a more economically diverse economy improved stunting rates more than countries with economies concentrated in agriculture. This variable may play an important role when looking at subregions of Sub-Saharan Africa where subsistence agriculture dominates many local economies or in coun- tries where GDP growth is mostly driven by natural resource rents. Lack of economic diversification, especially if GDP is concentrated in primary commodities or in extractive industries, is also often associated with higher macroeconomic volatility, which in turn may have a detrimental incidence on stunting reduction for a given level of economic growth. No studies reviewed included the following variables: economic recession, economic rents, or government expenditure on education. Inclusiveness The GINI index was used in a few studies to control for economic inequality within a country: (Haddad, et al. 2002; Heltberg 2009; and Tiwari, Zaman, and Saavedra 2013). Only Milman et al. (2005) controlled for coun- tries’ income distribution—that is, the share of income held by the richest 20 percent of the population. Once income distribution was controlled for in this study, the significance of GNI per capita on stunting disappeared (Milman et al. 2005). Governance Governance variables in past studies have had different effects, depending on the source and type of variable used. Milman et al. 2005 shows that democracy—a measure of civil liberties and political rights—is not a signif- icant factor affecting child undernutrition, whereas Smith and Haddad (2000, 2002) demonstrate that a democ- racy index positively influences almost all the variables included in their study (Milman, et al. 2005; Smith and Haddad 2000, 2002). Smith and Haddad (2015) note that the regression results are sensitive to the type of gov- ernance indicator used: when they used Worldwide Governance Indicators (WGI), they found no impact from governance on nutritional outcomes; however, when they employed governance indicators published by the Political Risk Services Group (International Country Risk Guide indicators), they found a significant impact on the stunting rate. As a side note, this significant effect did not hold in the short run (Smith and Haddad 2015). Conflict Variables on political violence have not been statistically significant, as shown in one study. Milman et al. (2005) found no significant effect with using conflict variables from Polity IV country reports 2003 and the State Fail- ure Problem Set 1955-2001 (Milman et al. 2005). The analysis undertaken in this section of the report employs three novel variables concerned with conflict: the state fragility index, the frequency of violent conflicts involv- ing governments, and the magnitude score of politically violent episodes (see Table 1). Endnotes Note: All dollar amounts are U.S. dollars unless otherwise indicated. 1 Information about WGI can be found at http://info.worldbank.org/governance/wgi/. 122 in Sub-Saharan Africa Stunting Reduction  ANNEX 3: Levels of Malnutrition and World Bank Investments, by Selected Countries WORLD BANK GROUP INVESTMENTS BY GLOBAL PRACTICE PERCENTAGE OF CHILDREN UNDER 5 YEARS OF AGE (%) 1 ANEMIC – (nutrition and food security themed projects/components) (US $millions)3 PERCENTAGE COUNTRY OF WOMEN HEALTH VITAMIN A 15-49 YEARS NUTRITION AGRI- SOCIAL STUNTED 4 UNDERWEIGHT WASTED ANEMIC EDUCATION TOTAL DEFICIENT OLD2 AND CULTURE PROTECTION POPULATION ANGOLA 37.6 19.0 4.9 64.3 53.3 47.3 0.0 BENIN 34.0 18.0 4.5 - 58.3 51.5 22.0 1.5 23.5 BURKINA FASO 27.3 19.2 7.6 - 86.1 50.5 5.0 5.0 CAMEROON 31.7 14.8 5.2 38.8 60.3 41.7 10.0 10.0 20.0 CENTRAL AFRICAN 40.7 23.5 7.4 68.2 - 46.2 0.0 REPUBLIC CONGO 32.1 16.9 11.1 - 66.7 53.8 0.0 COTE D'IVOIRE 29.6 15.7 7.6 - 74.8 51.8 57.0 2.7 59.7 DEMOCRATIC REPUBLIC OF 42.6 23.4 8.1 61.1 59.8 44.7 23.0 23.0 CONGO ERITREA 50.3 38.8 15.3 - - 36.9 0.0 ETHIOPIA 38.4 23.6 9.9 - 44.2 21.7 179.8 35.0 210.0 422.8 GAMBIA 17.5 6.5 3.4 64.0 72.8 57.2 10.8 10.8 GHANA 18.8 11.0 4.7 - 65.7 48.6 16.6 16.6 KENYA 26.0 11.0 4.0 84.4 36.3 27.5 26.2 26.2 LESOTHO 33.2 10.3 2.8 78.0 50.8 27.2 0.0 MADAGASCAR 49.2 36.8 15.2 42.1 45.3 36.6 55.8 55.8 MALAWI 42.4 16.7 3.8 59.2 62.6 32.3 138.7 138.7 MALI 38.5 27.9 15.3 - 85.4 54.8 7.0 7.00 MAURITANIA 27.9 24.9 14.8 - - 37.2 0.0 MOZAMBIQUE 43.1 15.6 6.1 68.8 68.7 49.9 49.6 16 65.6 NIGER 23.0 5.7 1.5 - 73.4 49.2 20.6 1.8 3.2 6.7 32.4 NIGERIA 32.9 19.4 7.2 29.5 68.4 49.9 720.0 720.0 RWANDA 37.9 9.3 2.2 6.4 36.5 19.4 25.0 5.0 30.0 SENEGAL 20.5 15.5 7.8 - 60.3 53.5 35.8 8.1 60 103.9 SUDAN 14.7 26.3 21.4 - - 29.4 0.0 TANZANIA 26.8 13.3 9.9 24.2 57.7 38.6 30.0 30.0 UGANDA 34.2 12 4.3 27.9 52.8 29.6 3.0 27.6 30.6 ZAMBIA 40.0 14.8 6.3 54.1 - 31.2 13.2 13.2 ZIMBABWE 27.6 11.2 3.3 35.8 36.8 30.1 0.0 Note: - indicates data not available. 1 Stunting, wasting, and underweight prevalence data from the 2017 version of the Joint Malnutrition Estimates; prevalence of vitamin A deficiency data were extracted from country profiles available through the WHO’s Nutrition Landscape Information System (NLiS), with country data being collected between 1993 and 2003; anemia prevalence data were retrieved from statcompiler.com and sourced from the most recent Demographic and Health Surveys (DHS) or Malaria Indicator Surveys (MIS). 2 Data for prevalence of anemia in women of reproductive age were retrieved from the WHO Global Targets Tracking Tool. 3 Data on active and pipeline World Bank Group investments as of September 1, 2017, and include operations with nutrition and food security theme. Project appraisal documents, project information documents, and project papers were reviewed to extract committed or proposed amounts to be dedicated to nutrition and food security theme (detailed on next page). 4 Cutoffs >40% 30-40% 20-30% <20% ANNEXES 123 Notes on current and planned World Bank Group nutrition and food security investments by country: ANGOLA None BENIN 75% of $28M project - Benin Multisectoral Food Health Nutrition Project (P143652) FY2014; 60% of $2.48M project - Nutrition Sensitive Agriculture & Capacity Building of Small & Marginal Farmers Project (P155822) FY 2016; $45M additional financing project (Agricultural Productivity and Diversification Additional Financing (P160029) FY 2017) has clear nutrition component but unclear how much is dedicated to this so not included in total. BURKINA FASO 10% of $50M project - Social Safety Net Project (P124015) FY 2014 CAMEROON 10% of $100M project - Health System Performance Reinforcement Project (P156679) FY 2016; 20% of $50M project - Cameroon Social Safety Nets (P128534) FY 2013 CENTRAL AFRICAN None REPUBLIC CONGO None COTE D'IVOIRE $50M Multisectoral nutrition project FY 2018 (pipeline); 20% of $35M project - Health Systems Strengthening and Ebola Preparedness Project (P147740) FY 2015; $2.7M project – Support to Nutrition-sensitive Agriculture and Capacity Development of Small and Marginal Farmers (P155081) FY 2018 DEMOCRATIC 10% of $230M project - Health System Strengthening for Better Maternal and Child Health. Results Project (PDSS) (P147555) FY REPUBLIC OF CONGO 2015 ERITREA None ETHIOPIA $2.75M Promoting young women’s livelihoods and nutrition project (P157716) FY 2017; 10% of $350M IDA to Second Agricultural Growth Project (P148591) FY 2015; 10% of $600M project - ET Productive Safety Nets Project 4 (PSNP 4) (P146883) FY 2015; $175M of $700M project across 4 sectors – Enhancing Shared Prosperity Through Equitable Services (P161373) FY 2018; 25% of $600M project – Ethiopia Rural Safety Net Project (P163438) FY 2018 GAMBIA 35% of $3.68M on project Maternal and Child Nutrition and Health Results Project (P143650) FY 2014; 40% of $5M additional financing on project GM Maternal and Child Nutrition and Health Results Project (P154007) FY 2015; $7.5M additional financing - AF Maternal and Child Nutrition and Health Results Project (P159693) FY 2017 GHANA 20% of $68M project - Ghana - Maternal, Child Health and Nutrition Project (P145792) FY 2014; $3M Improved Feeding Practices for first 1,000 Days (P159735) FY 2018 KENYA 10% of $150M project - Transforming Health Systems for Universal Care (P152394) FY 2016; 20% of $56.8M project - Health Sector Support Additional Financing (P128663) FY 2012 LESOTHO None MADAGASCAR 15% of $65M - Madagascar Emergency Support to Critical Education, Health and Nutrition Services Project (P131945) FY 2013; 60% of $10M additional financing - Madagascar Emergency Support to Critical Education, Health and Nutrition Services Project (P131945); $40M project - Madagascar: An Integrated Approach to Improving Nutrition Outcomes (P160848) FY 2018 MALAWI 30% of $80M project - Malawi Nutrition and HIV/AIDS Project (P125237) FY 2012; $12.27M of $22.6M additional financing - Additional Financing to Nutrition and HIV/AIDS Project (P156129) FY 2016; $100M project – Investing in the Early Years for Growth and Productivity in Malawi (P164771) FY 2019 MALI 10% of $70M project - Emergency Safety Nets project (Jigiséméjiri) (P127328) FY 2013 MAURITANIA None MOZAMBIQUE 80% of $37M - Mozambique Nutrition Additional Financing (P125477) FY 2013; 40% of $40M project - MZ- AF to Education Sector Support Project (P124729) FY 2012; Mozambique Primary Health Care Strengthening Project (P160967) FY 2018 NIGER 20% of $103M project - Population and Health Support Project (P147638) FY 2015; 70% of $2.5M project - Nutrition-Sensitive Agriculture and Capacity Building of Small and Marginal Farmers (P156863) FY 2016; 15% of $22.5M project - Adaptive Social Safety Nets Project (P155846) FY 2016; 8% of $84.2M project - Niger - GPE - Support to Quality Education Project (P132405) FY 2015 NIGERIA 40% of $50M project - Second Additional Financing to Third National Fadama Development Proj (P158535) FY 2016; $350M project - Accelerating Nutrition Results in Nigeria (P162069) FY 2018; $500M project - Nigeria - Program to Support Saving One Million Lives (P146583) FY 2015 - has significant nutrition component but unclear how much is allocated so not included in total; $350M project - Accelerating Nutrition results in Nigeria (P162069) FY 2018 124 in Sub-Saharan Africa Stunting Reduction  RWANDA $80M project - Strengthening Social Protection - Rwanda(P162646) FY 2018 - This is combined allocation to component with nutrition and ECD activities; $25M project – Rwanda Stunting Prevention and Reduction Project (P164845) FY 2018 SENEGAL 60% of $3M project - Building Resilience to Food and Nutrition Insecurity Shocks (P155475) FY 2017; 20% of $20M project - Senegal Health & Nutrition Financing (P129472) FY 2014; 20% of $40.5M project - Building Resilience to Food and Nutrition Insecurity Shocks (P155475) FY 2014; $30M project – Investing in Maternal and Child Health project (P162042) FY 2018; $60M project - Investing in the early years for human development in Senegal (P161322) FY 2018 SUDAN None TANZANIA 15% of $200M project - Strengthening Primary Health Care for Results (P152736) FY 2015 UGANDA $3M project - An Innovative, Integrated Approach to Enhance Smallholder Family Nutrition (P143324) FY 2013; $27.64M project - Uganda Multisectoral Food Security and Nutrition Project (P149286) FY 2015 ZAMBIA $2.75M project - Zambia Livelihood and Nutrition Project (P147745) FY 2015; 20% of $52M project - Health Services Improvement Project (P145335) FY 2014 ZIMBABWE None ANNEXES 125