Education, Health and Poverty in Sudan November 2018 Poverty and Equity Global Practice, Africa Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. . Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750- 4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. This paper was written by Alvin Etang Ndip (Senior Economist, GPV01). Overall guidance was provided by Pierella Paci (Practice Manager, GPV01). The author would like to thank Guy Morel Amouzou Agbe (Consultant, GPV01) for assisting with some of the data work. Many thanks to Nobuo Yoshida (Lead Economist, GPV01) and Simon Lange (Junior Professional Officer, GPV01) for providing valuable inputs at various stages of writing this paper, as well as Victor Sulla (Senior Economist, GPV07) and Omer Nasir Elseed (Senior Education Specialist, GED01) who acted as peer reviewers. The paper also benefited from comments from Alaa Mahmoud Hamed (Senior Operations Officer, GHN07), Dmitry Chugunov (Consultant, GED01), and Fareed Hassan (Consultant, GWA08), and discussions with Eiman Adil Mohamed Osman, (Consultant, GPV01), Thanh Thi Mai (Senior Education Specialist, GED13), and Tanya June Savrimootoo (Economist, GED01). Vice President Hafez Ghanem Country Director Carolyn Turk Senior Director Carolina Sanchez-Paramo Practice Manager Pierella Paci Task Team Leaders Alvin Etang Ndip and Simon Lange Table of Contents EXECUTIVE SUMMARY .................................................................................................................................. i I. INTRODUCTION .................................................................................................................................... 1 II. EDUCATION AND POVERTY .................................................................................................................. 5 Overview of Sudan’s Education System ...........................................................................................5 Primary School Attendance ............................................................................................................7 Secondary School Attendance....................................................................................................... 10 Determinants of School Enrollment .............................................................................................. 14 Dropouts and Late Entry into the Education System in Sudan ........................................................ 16 Reasons for Not Attending School and Dropouts ........................................................................... 17 Public Expenditure on Education................................................................................................... 19 III. HEALTH AND POVERTY................................................................................................................... 21 Levels and Trends in Infant, Under-five, and Child Mortality .......................................................... 22 Child Health ................................................................................................................................. 27 Determinants of Stunting for Children Ages 0–23 months .............................................................. 33 Public Expenditure on Health........................................................................................................ 34 IV. SUMMARY AND POLICY OPTIONS ................................................................................................. 36 Education Outcomes and Poverty ................................................................................................. 36 Health Outcomes and Poverty ...................................................................................................... 38 REFERENCES ................................................................................................................................................ 40 ACRONYMS CBS Central Bureau of Statistics EA Enumeration Area GNI Gross National Income IPRSP Interim Poverty Reduction Strategy Paper MDG Millennium Development Goal MICS Multiple Indicator Cluster Survey MENA Middle East and North Africa PPP Purchasing Power Parity SDG Sustainable Development Goal SHHS Sudan Household Health Survey NHBPS National Household Budget and Poverty Survey UNESCO United Nations Educational, Scientific, and Cultural Organization UNICEF United Nations Children’s Fund WDI World Development Indicators EXECUTIVE SUMMARY Improving education and health outcomes is a priority for Sudan’s Poverty Reduction Strategy. The objective of this note is to aid better understanding of the current nature of education and health and their relationship with poverty in Sudan. This in turn will inform the dialogue on policies to achieve reductions in poverty in the future through improvements in the education and health sectors. The note seeks to answer two main questions: First, what are the levels and trends in different education and health outcomes between 2010 and 2014, and how are these related to poverty and other household characteristics? Second, has Sudan’s performance on education and health outcomes been at the expected level based on experience in its income- peer countries? Analyzing Sudan’s performance on education and health outcomes for its level of income in comparison to peers helps ascertain whether Sudan is performing below, within, or above expectations compared to its peers. The analysis is mainly based on data from the 2010 and 2014 Multiple Indicator Cluster Survey (MICS) and the main messages are summarized as follows. Education Outcomes and Poverty Overall, Sudan made progress on education outcomes between 2010 and 2014. The literacy rate of young women (15–24 years) increased significantly during this period, particularly among those in poor households and in urban areas. Primary net attendance ratio increased by 5 percentage points standing at 69 percent in 2014, and this improvement was driven by an increase in attendance among children in the bottom 40 percent of the consumption distribution. Yet, primary net attendance rate remains far from being universal. This is likely because while government policy provides free basic education, household out-of-pocket spending is substantial and presents a barrier to education, especially for poor households. It is also observed that primary completion rate has increased between 2010 and 2014. Similarly, attendance in secondary schools increased, although it remains very low. Although primary completion rate has increased, it is not high enough to boost secondary attendance. However, substantial gaps remain in education outcomes across poverty levels, gender, and location. Although literacy rates increased a lot more for young women in poor households, there remains a significant gap of about 49 percentage points between young Sudanese women in the richest quintile and poorest quintile, and about 24 percentage points between those in urban and rural areas. For primary school attendance, the gap between urban and rural areas continues to widen, increasing to about 24 percentage points in 2014, with more urban children attending primary school than children in rural areas. With regard to primary school completion, children from poor households are still lagging. The gap between the richest quintile and the bottom 40 percent remains quite large, at over 70 percentage points. The analysis also reveals a huge gap in primary completion rates between children in urban and rural areas. Despite observed progress, Sudan’s performance is below expectations for all education outcomes, compared to its income peers. Even though progress was made during the last i decade, primary and secondary school enrollment remained low, falling short of achieving the then Millennium Development Goal (MDG) of universal access to basic education by 2015. Compared to many of its income peers, Sudan’s performance is below expectations for all education outcomes analyzed in this note. Female youth literacy, primary attendance, gross primary completion, and secondary attendance rates are all lower than expected levels. On the other hand, Sudan’s primary dropout rate is slightly higher than the expected level and remains a serious concern. Several reasons have been cited for this , including lack of funds to cover high household out-of-pocket spending on education, girls being married off early and being forced to work and support their families. In sum, Sudan is lagging its peers and needs to make further effort to consolidate and improve progress on education outcomes. Given the observed low levels of school enrollment, this note also analyzes determinants of school enrollment to identify the factors causing this issue. Poverty is a major determinant of school attendance for children in primary and secondary schools. Being a boy increases the probability of primary enrollment but decreases the probability of secondary enrollment, and the reason for this that boys have lower transition rates from primary to secondary schools. Also, parental education (particularly mother’s education) is another key factor affecting school enrollment. Children with more educated parents are more likely to enroll in school. It could be that more educated parents are likely to invest more money and time in their children ’s education. These findings have policy implications and the Government of Sudan could take several steps to address the issues highlighted above. Since free tuition alone is not enough for school enrollment and attendance, efforts to improve secondary school enrollments must start with programs that would boost primary school completion rates. For example, appropriate social protection programs could enable households to cope with negative shocks, allowing their children to stay in school when a negative shock hits. Also, social safety net programs that support people with disabilities or health insurance schemes may minimize the rate of dropouts for children who drop out of school to support family members. Conditional cash transfers may help boost school attendance. Similarly, food for education programs should be promoted, particularly targeting children in poor households. Providing on-site meals for school children and take-home rations (targeted food transfer) may encourage children to stay in school as it cuts down out-of-pocket expenses. To reduce school dropouts caused by early marriage of girls, various interventions can be considered to delay marriage, and support girls who marry early. Programs to educate girls and their parents about the negative impacts of this would be very useful. In addition, investing in female education today is important for future education outcomes, given the finding that school attendance increases with the mother’s level of education. For this reason, for adults and parents (especially mothers) who may not have opportunity to go through formal education, literacy programs could be offered. To reduce the burden of out-of-pocket expenses for households, designing and implementing programs that encourage free access to ii textbooks in classrooms for students in poor and disadvantaged communities may not only boost education outcomes (increase school attendance or reduce dropouts) but would also have a direct impact on learning outcomes (quality of education) and poverty (parents saving money they would normally spend on books). This would have a long-term impact on social progress, poverty, and inequality. The highlighted policy options require an increase in public expenditure on education, accompanied by measures to increase efficiency of the system. Health Outcomes and Poverty Overall, Sudan made considerable progress in reducing childhood mortality rates during the last decade, although the numbers remain high. Infant, under-five, and child mortality all declined considerably between 2010 and 2014, and this may be partly the result of recorded improvements on child immunization during this period. The under-five mortality rate, which measures the probability of a child dying between birth and his/her fifth birthday, declined by nearly 28 percent from 83 to 68 deaths per 1,000 live births. The improvements in under-five mortality rates were driven primarily by reductions in infant mortality rates. Infant mortality, which measures the probability of infants dying before their first birthday per 1,000 live birth, dropped from 60 in 2010 to 52 in 2014. Progress in reducing child mortality was slightly slower— falling from 24 to 17 infant deaths per 1,000 live births. This implies that one in every 58 children, who survived the first birthday, does not live to the fifth birthday. On the contrary, indicators of child malnutrition deteriorated between 2010 and 2014, and continues to be widespread. Stunting, defined as low height-for-age and an indicator of chronic malnutrition, increased by 6 percentage points to 38 percent in 2014, with 18 percent being severe cases. The high level of stunting is associated with many long-term factors including deficiencies in nutrition and inappropriate feeding practices over a sustained period. In Sudan, the minimum acceptable diet for children worsened during this period. Wasting, defined as low weight-for- height, remained unchanged during this period, standing at 16 percent. Underweight, defined as low weight-for-age, increased to 34 percent in 2014 by 6 percentage points since 2010. There are substantial variations in health outcomes rates across poverty levels, gender, and location. While childhood mortality rates are on a declining trend, child survival varies widely between different population groups. For example, infant mortality declined in all quintiles, except for children in the poorest quintile among whom it worsened in 2014. For child and under- five mortality, the gap between the poorest and the richest quintiles remains huge. Gender disparities exist in childhood health care in favor of girls. Also, childhood mortality remains a rural phenomenon, and there are substantial variations in mortality rates across states. Similarly, child malnutrition is related to household wealth. Stunting, wasting, and underweight are all more prevalent among male children in poor rural households. For its level of income, Sudan is underperforming on all health and nutrition outcomes. Infant and under-five mortality rates are higher than expected levels compared to peers. Also, stunting, wasting, and underweight rates are much higher in Sudan compared to other countries with iii similar incomes. This suggests that much effort is needed to improve these outcomes, particularly those related to child malnutrition. Because childhood stunting has long-term effects that are often irreversible, this note also analyzes determinants of stunting to identify its causes. Results show that stunting for children between 0 and 23 months of age (the period during which most damage in terms of stunting is done), is weakly correlated with poverty, but strongly correlated with gender, location, and mother’s education. Stunting or ‘chronic malnutrition’ is about more than money or food. Relatively high stunting rates exist even in upper quintiles, with only the richest quintiles being associated with significant decrease in stunting rates. Boys are more likely to be stunted than girls. Finally, mother’s education is strongly correlated with stunting in children. These findings have policy implications and the Government of Sudan could take several steps to address the issues highlighted above. First, the stunting results suggest that improving nutrition alone may not have much impact if no progress is made on education and health care access and quality. There is a need for a multisectoral approach toward reducing stunting, which will depend on having a more holistic view of the inequities and gaps in access to adequate levels of the underlying determinants of nutrition addressed by nutrition-sensitive interventions. For example, the finding that the probability of stunting diminishes among better-educated mothers implies that programs that promote women education would be useful for minimizing stunting levels. Second, there is need to enhance the level and efficiency of the government’s spending on health. The health system in Sudan, like in many developing countries, relies heavily on direct spending by health care users at the place and moment of service delivery, resulting in high out- of-pocket health spending and exposure of individuals to the risk of poverty, and perhaps death. Third, the government should provide better service delivery in the health sector, such as expanding preventive health care. This would increase the access to and quality of health care (demand and supply). Finally, expand immunizations or vaccinations campaign for children. Implement programs that educate families about the importance for children to receive all necessary immunizations or vaccinations in a timely manner. The highlighted policy options require an increase in public expenditure on health. iv I. INTRODUCTION Sudan’s medium-term national development policy framework is embodied in the Interim Poverty Reduction Strategy Paper (IPRSP). The paper was formulated in 2012 in the context of immense political upheaval due to the separation of the North (now Sudan) and South Sudan in 2011, which resulted in substantial loss (about 75 percent) in oil revenue and Sudan’s total revenue. To this end, Sudan launched a Five-Year Development Plan (2012–2016) to serve as a growth-oriented strategy with a primary focus on sustainable development and poverty reduction in the medium term. The IPRSP aims to reduce poverty through rapid, sustainable, and shared economic growth. Developing human resources is one of the four broad pillars of the IPRSP, which recognizes the role of investment in human development to build and enhance the population capabilities through education and better health. The Government of Sudan is now preparing the full PRSP that outlines a medium- to long-term plan for poverty reduction. This aligns with the Sustainable Development Goals (SDGs) and the World Bank Group’s twin goals to eliminate extreme poverty (with US$1.90 per day as the poverty line) and boost shared prosperity by 2030. The most recent estimates of poverty in Sudan are based on the 2014/15 National Household Budget and Poverty Survey (NHBPS), conducted by the Central Bureau of Statistics (CBS). According to the results of the survey, released in November 2017, in 2014/15 an estimated 36.1 percent of Sudan’s population had levels of per capita expenditure below the national poverty line. When measured against the World Bank’s international poverty line of US$1.90 per day (2011 purchasing power parity [PPP]), 13.5 percent of the population was deemed poor. When poverty was measured against the international poverty line for lower- to upper-middle-income countries (US$3.2 per person per day), 46.1 percent Sudanese were deemed poor. While the national poverty rate is high, it also varies significantly across Sudan’s 18 states (Figure 1). For example, while the Northern state had only 12 percent poor people, the Central Darfur state in western Sudan recorded the highest rate of poverty, with 67 percent of its residents deemed poor. All the Kordofan and Darfur states as well as the Red Sea and White Nile states had higher levels of poverty than the national average. The other states had lower poverty rates. 1 Figure 1: Poverty Rate by State 80% 60% 40% 20% 0% Central Darfur White Nile Al-Gadarif River Nile West Kordufan Blue Nile Al-Gezira Khartoum Northern North Kordufan Red Sea Kassala South Kordufan West Darfur North Darfur East Darfur South Darfur Sudan Sinnar Sources: CBS 2017; 2014/15 NHBPS. Based on the US$1.90 per day (2011 PPP) poverty line, the poverty rate in Sudan in 2014/15 (13.5 percent) barely changed from the corresponding number in 2009 (14.9 percent).1 During this period, poverty decreased only in rural areas and increased significantly in urban areas. Investment in human capital (education and health) is central to achieving the goals of eliminating extreme poverty and boosting shared prosperity in Sudan. Education and health have been shown to be strongly correlated with poverty across the world (Appleton and Balihuta 1996; Hughes and Irfan 2007; Jolliffe 2002). Tilak (2007) found that literacy and primary education are positively correlated with poverty. Similarly, Anyanwu (2005, 2013) found that rural households in India and Nigeria, respectively, whose main earning member does not have formal education or has attended only up to primary school are more likely to be poor than households whose earning members have attended secondary or higher education. High levels of education are often associated with better economic opportunities, including better access to jobs and higher lifetime wages. Education is also correlated with healthier life choices and increased voice and agency, the ability to make decisions and act on them. At the country level, economic benefits include increased rates of economic growth through gains in productivity and a greater capacity to adopt new technologies. But education is not only instrumental in promoting development; it is also by itself an end of development. Regarding health, being sick can reduce labor productivity which in turn can reduce household income level and consumption. A measure of child survival, infant mortality rate is one of the main indicators of a country’s well-being (that is its social progress and economic development), since it reflects social, economic, and environmental conditions in which children (and others in society) live, including their health care. Infant mortality reveals a society’s overall ability and willingness to care for its most vulnerable members (Eberstadt 1995; Sen 1998; Waldmann 1992; WHO 1999). Like many other countries, improving the well-being of infants, children, and 1Poverty rates using the national poverty line cannot be compared for 2009 and 2014/15 due to methodological differences between the two surveys. 2 mothers is an important public health goal for Sudan. Their well-being may determine the health of the next generation. Healthy birth outcomes, early identification and treatment of health conditions among infants can prevent death or disability and enable children to reach their full potential. Analyses at country level have found that mortality is correlated with per capita income (Preston 1975; Pritchett and Summers 1996). Child nutrition is an important issue for Sudan as malnutrition contributes to poor health, aggravates disease, and reduces productivity while compounding poverty and its aftereffects. As part of the SDGs’ agenda, Sudan formulated a National Health Policy (2017–2030) whose priorities include (a) promoting the health of mothers, neonates, infants, children, and adolescents by scaling up implementation of integrated high- impact interventions and (b) promoting good nutrition through a multisectoral approach to combat malnutrition in addition to providing access to nutrition services. It would be interesting to see how well the policy contributes to reducing malnutrition. From the forgoing discussion, it is probably not surprising that education and health are among key priorities of Sudan’s interim poverty reduction strategy. The objective of this paper is to aid better understanding of the current nature of education and health and their relationship with poverty in Sudan. This will inform the dialogue on policies to achieve reductions in poverty in the future through improvements in education and health. Two main questions this study seeks to answer are: 1. What are the levels and trends in different education and health outcomes between 2010 and 2014, and how are these related to poverty and other household characteristics? Throughout the analysis, we explore differences in the outcome variables across gender, poverty groups, and locations (rural or urban, states). This breakdown is important for monitoring progress made toward closing the disparities in access to opportunities for self-improvement for women and the very poor and reducing the inequality in human development outcomes between states and regions (as articulated in the IPRSP). We also analyze the determinants of school enrollment and stunting (low height-for-age) in Sudan. While the quality of service delivery in education and health sectors is important, the analyses in this paper focus on access to education and health services. 2. Has Sudan’s performance on education and health been at the expected level based on the experience in its income-peer countries? The paper will analyze Sudan’s performance on education and health outcomes for its level of income compared to peers. This helps ascertain whether Sudan is performing below, within, or above expectations compared to its peers. Cross-country comparisons (based on constant-elasticity regressions) are used to assess whether Sudan is overperforming or underperforming on any given indicator relative to its gross national income (GNI) per capita (Gable et al. 2014). Deviations from predicted indicator values may be viewed as an indication of how well Sudan does relative to its capacity to achieve outcomes and provide inputs. The main sources of data for this study are the 2010 and 2014 Multiple Indicator Cluster Survey (MICS). The MICSs were conducted by the CBS of the Republic of the Sudan, in collaboration with 3 the United Nations Children’s Fund (UNICEF) and the Ministry of Health.2 The sample size was 14,778 households in 2010 and 16,801 households in 2014. The survey employed a two-stage stratified cluster design in which the enumeration areas (EAs) were selected in the first stage based on the 2008 population census, and individual households were randomly selected in the second stage following household listing exercise in the selected EAs. In each EA, 25 households were selected . The sample was designed to be representative both nationally and at the state level, covering all Sudan’s states (15 states in 2010 and 18 states in 2014). The MICSs have a unique advantage over the household budget survey in that MICS is designed to collect comprehensive information that allow generation of child health and nutrition (anthropometric) as well as maternal health indicators. The analysis also makes use of other data sources, particularly the World Development Indicators (WDI) compiled by the World Bank. The paper proceeds as follows. Section II presents the results of selected education outcomes, linking them with poverty. Section III focuses on the link between health outcomes and poverty in Sudan. Section IV provides a summary of the main findings and policy options. 2MICS 2010 was labelled as Sudan Household Health Survey (SHHS) in its second round. In this paper, MICS 2010 refers to SHHS 2010. 4 II. EDUCATION AND POVERTY There is a clear relationship between education and poverty outcomes. From an individual perspective, lack of education is one of the main determinants of poverty. This strong link at the individual level is best illustrated by observing the vast differences in poverty of groups of individuals with different educational levels. For example, the Sudan 2014/15 National Household Budget and Poverty Survey (NHBPS) demonstrates that the poverty rate differs by the household head’s level of education (Figure 2). As the education level increases, the likelihood of being poor diminishes significantly. Two in three poor Sudanese belong to households whose head is illiterate. Those in households whose head has university-level education experience relatively much lower prevalence of poverty. Figure 2: Poverty Incidence by Level of Household Head’s Education High Secondary Intermediate Primary Never attend Khalwa* Sources: CBS 2017; 2014/15 NHBPS. Note: * Indicates people who had attended only koranic school. Enhancing the population capabilities through education is key for poverty reduction in Sudan as articulated in the IPRSP. A growing body of evidence on an international scale shows the paramount role that education plays as a fundamental input into a person’s functioning and capacity and as a great social equalizer as reflected by its impact on lifelong earnings and civic engagement. This section first provides an overview of the current state of the education system in Sudan. It then takes an in-depth look at the extent of inequity in access to education and educational achievement among Sudanese children and identifies the main factors driving the existing disparities. Attention is given to the disparities in resources for and access to education by location, gender, and wealth groups to highlight the several challenges that Sudan faces in developing an education system that adequately supports a sustainable poverty reduction strategy. Overview of Sudan’s Education System Sudan’s formal education system has four levels. The first level, kindergarten and day care (common in urban areas, such as Khartoum), begins at ages 3–4 and consists of up to two grades, depending on parental choice. The next level, primary school, begins at ages 6 –7 and consists of eight grades. Primary education is compulsory and tuition free. The primary-level curriculum is standardized into five subjects: religion, language, mathematics, man and the universe, and 5 applied arts. Primary education consists of eight years of schooling, followed by three years of secondary education. After Grade 8, students take a state examination to receive a certificate for entering high school, the third level of the education system. There are three grades at high school where student age ranges from 14–15 to 17–18 years. There are nine subjects at the secondary level: Arabic, Islamic studies, English, mathematics, physics, chemistry, biology, geography, and history, as well as additional elective courses. According to the international classification of education levels, the fourth level is tertiary education which comprises both higher provided in universities and Technical and Vocational Education and Training (TVET) provided in colleges and vocational schools. TVET subsector is a very important one.. In addition to formal education, there are schools called khalwa; the curriculum in these centers is mainly religious with the objective of teaching children to read, write, and memorize the Koran. Though they can provide any level of education, most are put in the preschool category.3 The primary language of instruction at all levels is Arabic. The analysis examines both stock variables such as literacy rates and flow variables such as school enrollment. Education outcomes analyzed include literacy rates; primary and secondary school attendance ratios, completion, and dropout rates. These are all related to SDG 4 and are priority indicators in the Education Sector Strategic Plan (2018–2022). We also analyze the determinants of enrollment in primary and secondary schools, the reasons for not attending school and dropouts. Youth Literacy Rate Female youth literacy has increased considerably, particularly among the poor. “A high literacy rate among the 15–24-year-olds suggests a high level of participation and retention in primary education, and its effectiveness in imparting the basic skills of reading and writing. Because persons belonging to this age group are entering adult life, monitoring their literacy levels is important with respect to national human resources policies, as well as for tracking and forecasting progress in adult literacy” (UNESCO).4 In addition to the household questionnaire, the MICS also administered a separate questionnaire for individual women only. Literacy is assessed on their ability to read a short simple statement shown to them or based on school attendance. As men did not complete this interview, it is not possible to compare the results by gender. Overall, the literacy rate of young women (15–24 years) increased from 52 percent in 2010 to 73 percent in 2014.5 As Figure 3 shows, the increase is particularly pronounced among individuals in the poor quintiles of the consumption distribution and urban youths. However, there remains a significant gap of about 49 percentage points between young Sudanese women in the richest quintile and poorest quintile (97 percent and 48 percent, respectively, in 2014), and about 24 percentage points between those in urban and rural areas (90 percent and 66 percent, 3 Since it is not possible to distinguish the level of khalwa, they are assumed to be preschools for this analysis. 4 http://uis.unesco.org/en/glossary-term/youthadult-literacy-rate 5 Youth literacy is calculated as the percentage of young women ages 15–24 who are able to read a short simple statement (shown to them) about everyday life or who attended secondary or higher education. 6 respectively, in 2014).6 Sudan seems to have slightly low female youth literacy rates compared to many of its income peers, implying that Sudan is performing below expectations. Figure 3: Trends in Literacy Rates among Young Women (a) Literacy rate (%) (b) Literacy rate versus GNI per capita 100 90 80 70 60 50 40 30 20 10 0 2010 2014 Source: Based on MICS 2010 and 2014. Source: Based on WDI, World Bank. Primary School Attendance Primary net attendance ratio increased slightly over time but remains far from being universal. Net attendance ratio increased overall by 5 percentage points from 64 percent in 2010 to 69 percent in 2014 (Figure 4). The increase is consistent with expectations, but the attendance ratio remains low given that primary education is intended to be free and compulsory in Sudan. The increase was driven by a rise in attendance among children in the bottom 40 percent of the consumption distribution. Net attendance among the top 20 percent increased only slightly (by 6In the absence of monetary measures of household wealth such as expenditure, a wealth or asset index was constructed using observations on household asset ownership (such as consumer durables) and housing characteristics (such as availability of piped water or a flush toilet). Wealth quintiles were then constructed. 7 3 percentage points) but is still about 40 percentage points higher than that of the bottom 40 percent. An interesting finding is that the gap between urban and rural areas continues to widen, increasing to about 24 percentage points in 2014, with more urban children attending primary school than children in rural areas. Gross enrollment rates in primary school remained stable between 2008 and 2017, but increased by 4 percentage points for secondary school during that period (Figure 4c). That primary net attendance ratio remains far from being universal is likely because though government policy provides free basic education, household out-of-pocket spending is substantial and presents a barrier to education, especially for poor households. Due to limited public expenditure on education, households pay for the running of schools in addition to other costs for uniforms, textbooks, and meals for their children (IPRSP Status Report 2016). For the poorest households, out-of-pocket basic education spending poses a barrier to education, and this has implications for school attendance and dropouts. Although net primary attendance ratio is on the rise, it is lower than expected, given the GNI level (and lower than the Sub-Saharan Africa average of 77 percent). Further efforts to consolidate and improve progress requires that education performance disparities across gender and states are fully addressed. 8 Figure 4: Trends in Primary Attendance Ratio (a) Net primary attendance ratio (%) (b) Net primary attendance ratio versus GNI per capita 100 90 80 70 60 50 40 30 20 10 0 2010 2014 Source: Based on MICS 2010 and 2014. Source: Based on WDI, World Bank. (c) Gross Enrolment Rates in General Education 100% 80% 72% 72% 73% 60% 39% 38% 40% 34% 20% 0% 2008/09 2015/16 2016/17 Basic Secondary Source: Education Sector Analysis (2017) The increase in primary attendance was observed for both boys and girls, but the magnitude of increase was slightly higher for girls. The gender gap in primary attendance ratio is shrinking 9 with girls catching up with boys (Table 1). There are significant variations in primary attendance across states, with highest attendance in the Northern, Khartoum, and River Nile states and lowest in the West Kordofan, Blue Nile, and Central Darfur states (Figure 5). According to cross- country regression analysis, Sudan’s net primary attendance ratio is far below the expected level when compared to other countries with similar incomes. Table 1: Trends in Net Primary Attendance Ratio by Gender (%) 2010 2014 Boys 65.8 69.9 Girls 62.0 68.3 All 63.9 69.1 Source: Based on MICS 2010 and 2014. Figure 5: Net Primary Attendance Ratio by State (%) Source: Based on MICS 2014. Secondary School Attendance Attendance in secondary schools remains very low. As shown in Figure 6, overall net attendance in secondary school increased by 7 percentage points between 2010 (20 percent) and 2014 (27 percent). This increase is mainly driven by a considerable rise in attendance among the top 40 percent. The gap in secondary attendance between the poor and rich is huge and continues to 10 widen. For the top 40 percent net attendance increased by 26 percentage points, which is twice the increase for the bottom 40 percent. The analysis also shows that the gap between urban and rural areas is expanding, with about twice as many children attending secondary school in urban areas than in rural areas. The proportion of girls in secondary schools remains slightly more than that of boys, although the gap is gradually closing. There is huge inequality in secondary attendance across states, and the variations are a lot more pronounced here than for primary attendance. The ranking is similar to primary attendance, with Khartoum, River Nile, and Northern states still at the top and Blue Nile, Central Darfur, Kassala, and West Kordofan states still ranked at the bottom (Figure 7). Figure 6: Trends in Secondary Attendance Ratio (a) Net secondary attendance ratio (%) (b) Net secondary attendance ratio by wealth quintile (%) 45 70 40 60 35 50 30 25 40 20 30 15 20 10 10 5 0 0 Urban Rural Male Female All Poorest Second Middle Fourth Richest 2010 2014 2010 2014 Source: Based on MICS 2010 and 2014. Source: Based on MICS 2010 and 2014. 11 Figure 7: Net Secondary Attendance Ratio by State (%) Source: Based on MICS 2014. Sudan is underperforming in secondary attendance compared to the Sub-Saharan Africa average of about 33 percent. Figure 8 plots secondary school attendance ratios for Sudan and other countries. The results show clearly that Sudan is lagging many countries in Africa and falls below the average attendance for Sub-Saharan Africa. Figure 8: Cross-country Comparison of Secondary School Net Attendance Ratio in 2014 50 40 30 20 10 0 Malawi Eritrea Ethiopia Cameroon Liberia Sudan Mozambique Sub-Saharan Africa Source: Based on MICS 2014 and World Bank WDI. School attendance in Sudan, as with many developing countries, follows an inverted-U shape. There is very low enrollment in during early school years, followed by high enrollment rates in compulsory school, and a precipitous decline in enrollment at ages corresponding to upper 12 secondary and higher education (Figure 9a). This pattern is consistent across 2010 and 2014. Only two out of three pupils who start Basic 1 make it to Basic 8 (Figure 9b) Figure 9a: School Attendance Rate by Age 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 2014 2010 Source: Based on MICS 2010 and 2014. Figure 10b: School Profile, 2008-2017 Source: Education Sector Analysis (2017) 13 Determinants of School Enrollment Poverty is a major determinant of school attendance among children between 6 and 18 years of age. Information from the MICS 2014 provides insights on the role of household wealth, parental education, and location on children’s enrollment in school. The disparity between rich and poor children plays out at primary and secondary levels of education and one can expect to see the same thing, at least, in higher education. The probability of school enrollment is about 3 percentage points higher among boys and about 16 percentage points higher among children from families in the top 20 percent of the wealth distribution (Table 2). But it is 7 percentage points lower in rural areas. While the probability of enrollment for boys is higher in primary school, it is higher for girls in secondary school. This is an interesting finding. Maybe this result can be explained by high drop out of male pupils while they get older. This is associated with the high opportunity cost of attending school, which includes the cost of not working in the household. Another reason for this is that boys have lower transition rates from primary to secondary schools. The transition rates rate for boys was 73 percent in 2010 and 80 percent in 2014, which were both lower than corresponding numbers for girls (80 percent and 84 percent, respectively). Parental education is another key factor affecting school enrollment. There are several ways through which parental education may influence school enrollment and attendance. On one hand, more-educated parents are likely to invest more money and time in their children. On the other hand, less-educated parents are more likely to lack both the know-how and the resources to invest similarly in their children´s education. While Sudan considers higher enrollment rate as an important government priority both as part of both the national development strategy and the SDGs, enrollment rates are low at the beginning of school and in the later parts of the education system. Table 2: Determinants of Primary School Enrollment (Probit Regressions, 2014) Primary school (6 to 13 years) Secondary school (14 to 16 years) Coefficient Marginal effect Coefficient Marginal effect Household Head Age 0.06 (0.01)*** 0.02 0.00 (0.02) −0.00 Household Head Age-squared −0.00 (0.00)*** −0.00 0.00 (0.00) 0.00 Household Head Sex (Male=1) 0.08 (0.14) 0.02 0.06 (0.30) * 0.22 Mother’s education_primary 0.44 (0.05)*** 0.13 0.00 (0.24)** 0.11 Mother’s education_secondary 0.48 (0.06)*** 0.14 0.61 (0.12)*** 0.22 Mother’s education_higher 0.34 (0.10)*** 0.10 0.76 (0.20)*** 0.27 Father’s education_primary 0.32 (0.05)*** 0.09 0–0.06 (0.11) −0.02 Father’s education_secondary 0.39 (0.05)*** 0.11 0.16 (0.12) 0.06 Father’s education_higher 0.50 (0.08)*** 0.14 0.49 (0.15)*** 0.19 Location (Rural=1) −0.24 (0.04)*** −0.07 0.05 (0.08) 0.02 Child sex (Male=1) 0.11 (0.03)*** 0.03 −0.16 (0.06)*** −0.06 Second quintile 0.08 (0.05) 0.03 0.27 (0.15) * 0.07 Middle quintile 0.22 (0.06)*** 0.07 0.45 (0.16)*** 0.12 Fourth quintile 0.48 (0.07)*** 0.14 0.99 (0.17)*** 0.31 Richest quintile 0.60 (0.08)*** 0.16 1.30 (0.18)*** 0.44 14 Primary school (6 to 13 years) Secondary school (14 to 16 years) Coefficient Marginal effect Coefficient Marginal effect Constant −1.16 (0.31)*** −2.48 (0.72)*** N 9,578 2,034 R-squared 0.16 0.20 Source: Based on MICS 2014. Note: *, **, and *** denote statistical significance at the 10 percent, 5 percent, and 1 percent level. Standard errors are reported in parentheses. Apart from age, the others are binary variables, and marginal effects are calculated as a discrete change in these variables from zero to one. Reference categories are no education; poorest quintile; urban area. In the full regression model, we controlled for many other factors including the number of children in household between the ages of 6 and 17 and state-fixed effects. Primary completion rate has increased but not high enough to boost secondary attendance. Not enough children complete primary school and move to secondary school. Gross primary completion rates in Sudan have increased between 2010 and 2014, from 70 percent to 83 percent, respectively (Figure 11).7 This positive change is mainly driven by the remarkable increase in completion rates among the bottom 40 percent of wealth distribution. Nevertheless, children from poor households are still lagging. The gap between the richest quintile and the bottom 40 percent remains quite large, at over 70 percentage points. All girls from urban richest households have a chance of completing primary education compared to only half of the boys from rural poorest households. Similarly, all boys from urban richest households are likely to complete primary education compared to only 28 percent of girls from rural poorest households (Education Sector Analysis, 2017). The analysis also reveals a huge gap in primary completion rates between children in urban and rural areas. In 2014, the proportion of urban children who completed primary school was almost double that for rural children (117 percent against 69 percent). It is also found that more boys completed primary school than girls in 2014 (88 percent against 78 percent). Despite the rising trend, Sudan is underperforming with regard to gross primary completion rates relative to its income peers. The results suggest that Sudan is making progress in enrolling children in primary school but unsuccessful in bringing them to full and timely primary completion (due to dropout and repetition). The low completion rates are probably because parents, who have managed to enroll their children in school, are unable continue investment (when a negative shock hits the household), or parents do not consider the investment worthwhile, either due to perceived low returns or the child’s poor performance (Etang and Tsimpo 2017). 7Gross primary completion rate higher than 100 indicates that some of the graduates are older than the official age of 13 years (which remains unchanged as the denominator), while the numerator is all children in Grade 8 regardless of age. 15 Figure 11: Gross Primary Completion Rates (a) Primary completion rate Gross (Grade 8) (b) Primary completion rate versus GNI per capita 120 100 80 60 40 20 0 Urban Rural All Male Female 2010 2014 Source: Based on MICS 2010 and 2014. Source: Based on WDI, World Bank. Dropouts and Late Entry into the Education System in Sudan Cross-country regression analysis shows that Sudan’s primary dropout rate is slightly higher than the expected level when compared to other countries with similar incomes. Figure 12 illustrates this clearly. This implies that Sudan is underperforming on this front, as a large number of Sudanese students exit the education system without completing the full primary education. According to government statistics, the average child attends seven years of schooling, below the eight years required to complete the primary school cycle. Officially the age at which students enter primary school is 6 years, but the average age of a Grade 1 student is 7.5 years. Based on MICS 2014, the proportion of children of primary school entry age (six) entering Grade 1 was only 37 percent nationwide, and much higher in urban areas (57 percent) than rural areas (30 percent). Table 3 gives a picture of late entry by showing the number of out-of-school children, which stands at 27 percent and higher among children aged 6-13 compared to those aged 14-16. Household wealth plays a crucial role in whether or not a child enters Grade 1 at the age of six, with over three-quarters (78 percent) of children in the richest households (top 20 percent) entering school on time compared to only 15 percent of their peers from the poorest households (bottom 20 percent) who are six years old in Grade 1. There is no statistically significant difference between girls and boys in the percentage of students that started primary school on time in the 2014/15 school year. Similarly, there are no gender differences in the average age of first year primary school students. This result suggests that there is a huge delay in entering school in 16 Sudan. The issue of late entry would explain why the gap between the net and gross primary enrollment rates is large in Sudan. This is likely to be largely attributable to delayed entry into primary school. Late school entry varies substantially across states. The state with the lowest proportion of children entering Grade 1 at six years is West Kordofan (13 percent); and the Northern state, which is the richest state, has the highest of children entering Grade 1 on time (74 percent). Figure 12: Primary School Dropout Rate Source: Based on WDI, World Bank. Table 3: Out-of-school children by age category and reason for being out of school, 2014 Ages Population Enrolled Out of school Never Total attended Left school 6-13 (‘000) 7,887 5691 2,197 1,244 953 14-16 (‘000) 2,518 1855 662 241 421 Total 10,405 7,546 2,859 1,485 1,374 6-13 100% 72% 28% 57% 43% 14-16 100% 74% 26% 36% 64% Total 100% 73% 27% 52% 48% Source: Authors’ calculations based on 2014 Sudan MICS 17 Reasons for Not Attending School and Dropouts Affordability constraints are one of the main reasons cited for not attending school in Sudan. From MICS 2014, self-reported reasons for which children and adolescents (ages 4–24 ) do not attend school are mainly high costs and long distance to schools. Figure 13 plots reasons for not attending school. Lack of funding to meet the financial burden of school expenses is the main reason the majority have never attended school (21 percent). Of the respondents, 14 percent stated that they do not attend school because it is too far away, with 4 percent citing lack of schools for their non-attendance. Around 3 percent did not attend school due to a disability or disease and another 3 percent cited the need for the child to support the family. These results clearly suggest that affordability (that is, cost) is the main hindering factor for not attending school rather than availability of a school. Similarly, financial constraints are cited as main reasons for dropout among both boys and girls (19 percent cases). Taking up work to support the family and early marriage of girls are other main reasons for dropping out of school (cited by 9 percent and 6 percent, respectively). Some dropouts reported that disabilities became impediments to attending school, while others mentioned long distances to available schools as a hindrance to continuing school. These main results are consistent with findings of the 2014/15 NHBPS survey. Early marriage of girls is one of the main reasons for school dropouts, and this can have negative effect on education attainment. This is in line with findings that child marriage is a main reason why girls drop out of school prematurely (Wodon et al. 2017). According to MICS 2014 survey, 21 percent of girls (ages 15–19) were married, and this was mainly a rural phenomenon (26 percent) compared to urban areas (11 percent). The issue of child marriage is strongly correlated with poverty and low mother education, mainly among the bottom 40 percent. Early marriages will likely lead to early pregnancies, which will prevent a girl from going to school. In this regard it is probably not surprising that fertility rates are much higher among adolescent girls (ages 15–19) in rural areas than in urban areas (103 per 1000 women and 53 per 1000 women, respectively (based on MICS 2014). 18 Figure 13: Reasons for Not Attending School and Dropouts (a) Reasons for not attending school (b) Reasons for dropping out from school Fees Fees Work to support family School far Early marraige No School Disability/Disease Disability/Disease School far No water/Toilet Work to support family No School Co-education Co-education 0% 5% 10% 15% 20% 25% 0% 5% 10% 15% 20% Source: Based on MICS 2014. Source: Based on MICS 2014. Public Expenditure on Education Public education expenditure has consistently low, especially by regional standards, and given the need to expand education opportunities increased expenditure allocation is required. Public education expenditure as a share of total public expenditure fluctuated between 7.3 and 12 percent during the last two decades, while it fluctuated between 1.3 and 2.7 percent of GDP (Figure 13). However, the budgetary allocation to education in the national budget for 2018 seems to be the lowest since 2007. Given the need to expand and improve education access, increased public education expenditure allocation is required. Sudan spends less on education than do comparable countries. Comparable countries with similar dependency ratio (i.e. school age population as a share of the total population) spend more—including Ethiopia and Kenya, with a level of spending exceeding 20 percent of total public spending. Other lower-middle- income such as Morocco and Tunisia each spent more by this measure (Table 4). Eyeballing Table 4, the numbers show that primary school enrollment were much higher for all these countries than Sudan, indicating a strong correlation between public expenditure and educational outcomes. 19 Figure 13: Public Education Expenditure, 2002-2018 15 11.2 12 8.1 9.2 9 10 7.3 5 1.8 1.9 2.7 2.7 1.3 1.3 0 2000 2002 2005 2007 2009 2018 Education spending as a % of total public spending Education spending as a % of GDP Source: 2018 data from Ministry of Finance administrative data and 2000-2009 data from World Bank (2012). Table 4: Public Education Expenditures: Regional Comparison (2010–2014) Public expenditure Primary school Primary school on education enrollment enrollment % of total public (% gross) (% net) expenditure Sudan 10.6 54 70 Neighboring countries Ethiopia 22.0 n.a. 98 Kenya 20.6 84 114 Uganda 14.0 91 107 Other SSA lower-middle-income countries Cape Verde 15.0 98 114 Benin 22.3 95 124 Senegal 20.7 84 73 MENA lower-middle-income countries Egypt 12.6 115 98 Morocco 18.5 118 98 Tunisia 21.2 110 99 Source: WDI, World Bank. Data refer to latest year available for 2010–2014. Note: SSA = Sub-Saharan Africa. 20 III. HEALTH AND POVERTY In Sudan, life expectancy has remained barely the same between 2011 and 2016 (at 63 and 64, respectively).8 Sudan’s life expectancy is higher than the average for Sub-Saharan Africa, but lags those elsewhere, including lower-middle-income countries (Figure 14). While measures of life expectancy provide a basis for comparing Sudan’s performance with that of other countries, the national averages do not capture many important aspects such as the distribution of health outcomes among the population. Life expectancy may vary substantially across different populations of the country, with some populations living longer than others for various reasons. Because of prolonged conflict and violence in parts of Sudan, historical differences in economic opportunities, and differences in the degree of exposure to health risks across states and population groups, the health status of the population varies substantially. Given that Sudan has large and diverse population groups, it is important to explore variations in health outcomes. Figure 14: Life Expectancy by Region (1990–2016) 90 80 Life Expectancy (Years) 70 Sub-Saharan Africa Sudan 60 World Middle East & North Africa 50 High income 40 1990 2000 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: WDI. This analysis presents a diagnostic assessment of the state of health and its links to poverty in Sudan. Health indicators will be examined to monitor child and maternal health, including infant mortality rate, under-five mortality rate, child mortality rate, and maternal mortality rate (consistent with SDG 3). Nutrition is a very important issue for Sudan as malnutrition contributes to poor health, aggravates disease, and reduces productivity while compounding poverty and its 8Life expectancy is the average number of years a newborn can expect to live if exposed to the age-specific mortality rates prevailing in the population. 21 long-term effects. The paper will also analyze the main indicators that are used to monitor malnutrition in children ages 6–59 months (that is, under-five malnutrition): stunting or low height-for-age, underweight or low weight-for-age, and wasting or low weight-for-height. To be more specific, these measures reflect children whose height-for-age, weight-for-age, and weight- for-height fall more than two standard deviations below the median of internationally accepted growth standards.9 Under-five malnutrition is a complex phenomenon and its eradication depends on more than simply increasing caloric intake. Each indicator conveys unique information about a child or population and monitoring trends over time provide invaluable insight for shaping policy and developing targeted interventions. The analysis will also be conducted based on data from MICS 2010 and 2014. We also look at the determinants of stunting, as well as Sudan’s health financing. Throughout, the links between poverty, gender, location, and health outcomes are emphasized. Levels and Trends in Infant, Under-five, and Child Mortality Levels and trends in infant, under-five, and child mortality are some of the key measures of a country’s population health status and its progress in ensuring effective, safe, and good quality health care. Here, we present estimates of these indicators using data from MICS 2010 and 2014. Figure 15 presents the estimates of the trends in infant, under-five, and child mortality for Sudan. Infant, under-five, and child mortality in Sudan all declined considerably between 2010 and 2014 but remain high. Under-five mortality, which measures the probability of a child dying between birth and the fifth birthday, in Sudan declined by nearly 28 percent from 83 to 68 deaths per 1,000 live births. These improvements in under-five mortality were driven primarily by reductions in infant mortality rates. Infant mortality, which measures the probability of infants dying before their first birthday per 1,000 live births, dropped from 60 in 2010 to 52 in 2014. Progress in reducing child mortality was slightly slower—falling from 24 to 17 infant deaths per 1,000 live births. This implies that one in every 58 children, who survived the first birthday, does not live to the fifth birthday. The gains recorded in children health outcomes between 2010 and 2014 may partly be related to improvements made on their health care during this period. The proportion of children (12– 23 months) who received BCG immunization increased from 76 percent in 2010 to 83 percent in 2014. The magnitude of the increase was more pronounced among girls, by 10 percentage points which was double that for boys. The results also show a rise in the share of children in poorest households receiving BCG immunization, although they are still lagging way behind their peers in richest households (66 percent versus 93 percent in 2014). In the same vein, there has been an increase in the proportion of children (12–23 months) who received immunization against measles, one of the causes of child death. The extent of the increase was 10 percentage points 9WHO Child Growth Standards is the internationally accepted growth standards, launched in 2006. The standards are based on a Multi-Growth Reference Study from sample populations in Brazil, Ghana, India, Norway, Oman, and the United States. 22 between 2010 and 2014, and this was driven by a substantial increase in infants in poorest households who received immunization. Figure 15: Trends in Infant, Child, and Under-five Mortality Rates 90 80 Deaths per 1,000 live births 70 60 50 40 30 20 10 0 2010 2014 Infant mortality rate Child mortality rate Under-5 mortality rate Source: World Bank staff calculations based on MICS 2010 and 2014. Note: (a) Infant mortality: the probability of dying before the first birthday. (b) Child mortality: the probability of dying between the first and the fifth birthday. (c) Under-five mortality: the probability of dying between birth and the fifth birthday. All rates are expressed per 1,000 live births except for child mortality, which is expressed per 1,000 children surviving to 12 months of age. Child survival varies widely between different population groups. It is encouraging to observe that all these indicators are on a declining trend since 2010. However, the national averages mask variations that exist between different groups of the population. For example, infant mortality declined for everyone except for children in the poorest quintile among whom infant mortality worsened in 2014 (Figure 16). For child and under-five mortality, the gap between the poorest and the richest quintiles remains huge. Gender disparities also exist in childhood health care. For all three indicators, boys did worse than girls, and the pace of mortality reduction is slower for boys. The results also show that childhood mortality remains a rural phenomenon. The decline in infant mortality is more pronounced in urban areas, where this indicator was cut by about 16 percentage points compared to a 5percentage point decrease in rural areas. Gender and location differences are more conspicuous for under-five mortality. Urban areas did see reduction twice as much as did rural areas (by 23 percentage points against 11 percentage points respectively). And under-five mortality reduced by 19 percentage points for girls and 10 percentage points for boys. 23 Figure 16: Trends in Childhood Mortality by Poverty, Gender, and Location (a) Infant mortality by consumption quintile (%) (b) Child mortality by consumption quintile(%) 80 35 70 30 60 25 50 20 40 15 30 10 20 10 5 0 0 Poorest Second Middle Fourth Richest All Poorest Second Middle Fourth Richest All 2010 2014 2010 2014 (ci) Under-five mortality by consumption quintile (d) Infant mortality by location and gender 120 70 100 60 50 80 40 60 30 40 20 20 10 0 0 Poorest Second Middle Fourth Richest All Urban Rural Male Female All 2010 2014 2010 2014 24 (e) Child mortality by location and gender (f) Under-five mortality by location and gender 30 100 90 25 80 70 20 60 15 50 40 10 30 20 5 10 0 0 Urban Rural Male Female All Urban Rural Male Female All 2010 2014 2010 2014 Source: Based on MICS 2010 and 2014. There are substantial variations in mortality rates across states. Figure 17 displays state-level trends in infant, child, and under-five mortality in 2014 for Sudan. The Northern and River Nile states had the best outcomes for children, while the Darfur and Kordofan states generally had the worst outcomes. In particular, East Darfur had the highest infant and under-five mortality rates in the country, at nearly 89 and 118 respectively, in 2014. Followed by South Kordofan, with 70 and 95 deaths per 100,000 live births, respectively for the two indicators. Mortality trends in conflict-affected states (such as Darfur) may be driven by the selection effects of emigration and survivorship bias rather than by real changes in survival conditions. 25 Figure 17: Mortality Rate Trends by State in 2014 National East Darfor Central Darfor South Darfor West Darfor North Darfor West Kordofan South Kordofan North Kordofan Blue Nile Sinnar White Nile Gezira Khartoum Gadarif Kassala Red Sea River Nile Northern 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Infant mortality rate Child mortality rate Under-5 mortality rate Source: Based on MICS 2014. Sudan’s infant and under-five mortality rates seem to be higher than the expected level for its income level. A comparison of infant and under-five mortality rates among countries with similar GNI per capita shows that Sudan is underperforming relative to its peers (Figure 18). Maternal mortality rate declined from 534 (per 100,000 women of reproductive age during the same period) in 2010 to 521 in 2013 (IPRSP 2016). This is probably because the proportion of women who received antenatal care from any skilled provider increased over time, mostly women in the richest urban households. However, as with children health outcomes, maternal mortality rate in Sudan seems to be higher than the expected level (Figure 19). 26 Figure 18: Infant and Under-five Mortality Rates in Sudan and International Comparison (a) Infant mortality rate versus. GNI per capita (b) Under-five mortality rate versus GNI per capita Source: Based on WDI, World Bank. Figure 19: Maternal Mortality Rate versus GNI per capita Source: Based on WDI, World Bank. Child Health Childhood malnutrition has significant long-term effects, including diminished adult intellectual ability, work capacity, and productivity, which ultimately lead to economic hardships for individuals and their families. For children of school age, poor nutrition limits cognitive ability and school readiness. From a health standpoint, under-nutrition increases vulnerability to disease and the likelihood of death from disease. Worldwide, 19 percent of 27 under-five deaths are directly attributable to being underweight. Even children who are mild to moderately underweight are at increased risk of death. In Sub-Saharan Africa, an estimated 1.3 million deaths are attributable to undernourished children (weight-for-age less than 1 standard deviation below the reference mean). In addition, there is growing evidence that early malnutrition is a risk factor for developing chronic diseases later in life such as diabetes, hypertension, renal disease, and cardiovascular disease, which are costly to manage. Infants and young children falter in their growth due to inadequate diets and recurrent illnesses, which reduce appetite, increase metabolic requirements, and increase nutrient loss. Anthropometric indicators for young children have deteriorated between 2010 and 2014, with malnutrition continuing to be widespread. The analysis of child malnutrition is not only essential in its own right to monitor the SDGs , but malnutrition also has direct impact on health and educational outcomes, including the ability to learn. Trends in the nutritional status of children under five years for the period between 2010 and 2014 are shown in Figure 20. The outlook seems negative, particularly for stunting and underweight. The results show an upward trend in the percentage of stunted and underweight children between 2000 and 2014, but the percentage of children who are wasted has remained stable. Stunting, defined as low height-for-age and an indicator of chronic malnutrition, increased by 6 percentage points to 38 percent in 2014, with 18 percent being severe cases. The high level of stunting is associated with many long-term factors including deficiencies in nutrition (chronically inadequate levels of protein and energy and or micronutrient deficiencies), as well as inappropriate feeding practices over a sustained period (Seff et al. 2014). In Sudan, the minimum acceptable diet for children worsened during this period. The percentage of breastfed children (ages 6–23 months), who had at least the minimum dietary diversity and the minimum meal frequency during the day preceding the survey, was only 25 percent in 2014. Similarly, the percentage of non-breastfed children (ages 6–23 months), who received at least two milk feeds and had at least the minimum dietary diversity not including milk feeds and the minimum meal frequency during the day prior to the survey, was 37 percent in 2014, having declined considerably from about 56 percent in 2010. Childhood stunting has long-term effects that are often irreversible. It can cause delayed motor function and diminished cognitive ability; and children with low height-for-age in their early years may exhibit poor academic performance later in life (UNICEF 2007). This can adversely affect their economic outcomes, and the country’s productivity and growth. 28 Figure 20: Trends in Nutritional Outcomes of Children under Five Years (%) 40 35 30 25 20 15 10 5 0 2010 2014 Stunting Wasting Underweight Source: Based on MICS 2010 and 2014. Wasting remained unchanged in Sudan between 2010 and 2014, standing at 16 percent, with nearly 5 percent being severe cases. Wasting (low weight-for-height) is a measurement of acute malnutrition characterized by considerable weight loss or failure to gain weight, resulting in a child having a weight substantially below what would be expected of a healthy child of the same height (Seff et al. 2014). Wasting is an indicator of current malnutrition and can change quickly over time; even showing seasonal patterns associated with changes in food availability and disease prevalence. Wasting poses a particularly imminent danger for children under five and can be a strong predictor of child mortality. The incidence of underweight in Sudan stands at 34 percent in 2014, increasing by 6 percentage points since 2010. Underweight (low weight-for-age) is a composite measurement of stunting and wasting as it is influenced by both height and weight. Underweight reflects both chronic and acute malnutrition and is a good indicator for assessing changes in malnutrition over time. Child malnutrition is related to household wealth. Stunting is more prevalent among male children in poor households in rural areas. Figure 21 plots descriptive results of stunting levels by poverty status of the household, location and gender of the child. This consistent across both survey years. The same story holds for wasting and underweight (Figures Figure 22 and Figure 23). This means that efforts to curb child malnutrition need to address the issue of poverty, particularly in rural areas. 29 Figure 21: Stunting by Poverty, Gender, and Location (children under five years) All Richest Poorest 2014 Rural Urban Female Male All Richest Poorest 2010 Rural Urban Female Male 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% Yes, severely stunted Yes, moderately stunted Source: Based on MICS 2010 and 2014. Figure 22: Wasting by Poverty, Gender, and Location (children under five years) All Richest Poorest 2014 Rural Urban Female Male All Richest Poorest 2010 Rural Urban Female Male 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% Yes, severely wasted Yes, moderately wasted Source: Based on MICS 2010 and 2014. 30 Figure 23: Underweight by Poverty, Gender and Location (children under five years) All Richest Poorest 2014 Rural Urban Female Male All Richest Poorest 2010 Rural Urban Female Male 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% Yes, severely underweight Yes, moderately underweight Source: Based on MICS 2010 and 2014. Nutrition outcomes are worse than what would be expected based on Sudan’s GNI per capita. Comparing stunting, wasting, and underweight rates among countries with similar GNI per capita, Figure 24 shows that Sudan is performing below expectations relative to its peers. This indicates much effort is needed to reverse the rising trends for these outcomes. 31 Figure 24: Nutritional Outcomes in Sudan and International Comparison (a) Stunting versus GNI per capita (b) Wasting versus GNI per capita (c) Underweight versus GNI per capita Source: Based on WDI, World Bank. 32 Determinants of Stunting for Children Ages 0–23 months Stunting is weakly correlated with poverty, but strongly correlated with gender, location, and mother’s education. Regression results of the determinants of stunting for children between 0 and 23 months of age (the period during which most damage in terms of stunting is done), are presented in Table . Five results stand out. First, household wealth is weakly correlated with stunting. Stunting or ‘chronic malnutrition’ is about more than money or food alone. There are relatively high stunting rates even in upper quintiles, with only the richest quintiles being associated with significant decreases in stunting. Regarding food, adequate nutrition is important, but other factors may be more important depending on context (access to clean water and sanitation, health care, and early childhood stimulation (World Bank 2017). Second, the gender of the child is strongly correlated when other variables are controlled for. Boys are more likely to be stunted than girls. This seems to be a common finding elsewhere (Galasso and Umapathi 2009). The probability of being stunted is about 8 percentage points higher among boys. Third, descriptive results showed that stunting is more prevalent in rural settings. However, when other factors are held constant, living in rural areas is associated with decrease in the likelihood of stunting, by about 8 percentage points. Table 5: Determinants of Stunting for Children 0–23 months (Probit Regressions, 2014) Coefficient Marginal effect Household Head Age 0.01 (0.02) 0.00 Household Head Age-squared −0.00 (0.00) −0.00 Household Head Sex(Male=1) −0.11 (0.22) −0.04 Mother’s education_primary −0.17 (0.07)** −0.06 Mother’s education_secondary −0.18 (0.10)* −0.07 Mother’s education_higher −0.50 (0.17)*** −0.17 Father’s education_primary −0.07 (0.07) −0.03 Father’s education_secondary 0.04 (0.09) 0.02 Father’s education_higher −0.05 (0.15) −0.02 Location (Rural=1) −0.21 (0.08)*** −0.08 Child sex (Male=1) 0.21 (0.06)*** 0.08 Second quintile 0.04 (0.09) 0.01 Middle quintile 0.03 (0.11) 0.01 Fourth quintile −0.21 (0.14) −0.08 Richest quintile −0.33 (0.17)** −0.12 Household size 0.02 (0.01)* 0.01 Child received immunization −0.01 (0.06) −0.01 Child took Vitamin A −0.21 (0.07)*** −0.08 Has access to improved drinking water −0.08 (0.08) −0.03 Make water safe for drinking −0.10 (0.07) −0.04 Constant −0.41 (0.55) N 2,298 R-squared 0.06 Source: Based on MICS 2014. Note: *, **, and *** denote statistical significance at the 10 percent, 5 percent, and 1 percent level. Standard errors are reported in parentheses. Apart from age, the others are binary variables and marginal effects are calculated as a discrete change in these variables from zero to one. Reference categories are no education; poorest quintile; urban 33 area. In the full regression model, we controlled for many other factors including the number of children in household between the ages of 6 and 17, and state-fixed effects. Fourth and perhaps the most important determinant of stunting is parental education, specifically mother’s education (father’s education does not matter). Mother’s education, which seems to be vital for children school enrollment, is also a key determinant of stunting in children. The probability of stunting decreases as a mother’s level of education increases. The magnitude of the coefficient and marginal effect is far higher for mothers with higher education, associated with 17 percentage points decline in stunting. Fifth and finally, the probability of stunting is significantly lower for children who take vitamin A. Being immunized on the other hand is not statistically significantly associated with a decrease in the probability of being stunted in Sudan. Put together, the stunting results suggest that improving nutrition alone may not have much impact if no progress is made on education and health care access and quality. This calls for a multisectoral approach toward reducing stunting, which will depend on having a more holistic view of the inequities and gaps in access to adequate levels of the underlying determinants of nutrition addressed by nutrition-sensitive interventions (World Bank 2017). Public Expenditure on Health Public expenditure on heath declined between 2012 and 2018 even though health—MDGs have not yet been met. Public expenditure allocation to the health sector, as a share of total public spending fluctuated between 7 and 8 percent during the last decade (Figure 25). Sudan’s level of public spending on health is lower than Ethiopia and Uganda but Sudan spends more than Kenya, Egypt and Morocco (Table 6). This is really encouraging. However, finding that Sudan is performing worse than its income peers suggests that while appropriate budget allocations are necessary, the quality of service delivery is also critical for having improved outcomes. The foregoing analyses suggest a need for substantial improvement on health outcomes. Figure 25: Public Health Expenditure, 2012-2018 8.0 7.4 7.5 8.0 7.3 7.0 7.0 5.9 6.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 2012 2013 2014 2015 2016 2017 2018 Health expenditure as a % of total expenditure Health expenditure as a % of GDP Source: Ministry of Finance and Economic Planning and I-PRSP Status Report, 2016. 34 Table 6: Public Expenditure on Health: Regional Comparison (2014) Public expenditure Under-five Maternal on health (% of total mortality mortality public expenditure) rate (per 1,000) rate (per 100,000 live births) Sudan 8.0 73 512 Neighboring countries Ethiopia 16.4 64 420 Kenya 5.9 71 400 Uganda 24.3 66 360 Other SSA lower middle-income countries Cape Verde 10.0 26 53 Benin 10.7 85 340 Senegal 7.6 55 320 MENA lower-middle-income countries Egypt 5.5 22 45 Morocco 6.0 30 120 Tunisia 13.3 15 46 Source: WHO, Global Health Observatory Data (2013). 35 IV. SUMMARY AND POLICY OPTIONS Improving education and health outcomes is a key priority for Sudan’s Poverty Reduction Strategy. The objective of this paper is to aid better understanding of the current nature of education and health and their relationship with poverty in Sudan. This will inform the dialogue on policies to achieve reductions in poverty in the future through improvements in education and health. The paper attempts to answer two main questions: First, what are the levels and trends in different education and health outcomes between 2010 and 2014? Throughout the analysis, which is based on MICS 2010 and 2014 data, we explore differences in the outcome variables across gender, poverty groups, and locations in both rural and urban areas and at the state level). The analysis highlights the links between education and health and poverty and the role they can play in Sudan’s poverty reduction efforts. Second, has Sudan’s performance on education and health been at the expected level based on experience in its income peer countries? Cross-country constant-elasticity regressions are used to assess whether Sudan is overperforming or underperforming on any given indicator relative to its GNI per capita. The main findings and policy implications are summarized in the following paragraphs. Education Outcomes and Poverty Despite progress made during the last decade, primary and secondary school enrollment remained low, falling short of achieving the then Millennium Development Goal (MDG) of universal access to basic education by 2015. Only three out of every four children of basic school age are receiving formal education. The secondary school net enrollment rate is about 27 percent, which is less than half the basic level. There is inequality in school attendance across gender and states . Gender disparities in educational achievements are substantial in Sudan. Historically, gender discrimination and social norms have limited women’s access to basic education. Further effort to consolidate and improve progress requires that education performance disparities across gender and states are fully addressed. In other words, there is a need for leveling the playing field in children’s access to quality education, irrespective of their location or gender. Sudan is underperforming in all education outcomes, given its income level when compared with peers. School attendance, literary rate, completion, and dropout rates do not meet expected levels when compared to countries with similar incomes. Despite the increase in public spending on education during the last decade, Sudan spends less as a share of total public spending than do comparable peer countries. Basic education in Sudan is intended to be free and compulsory. However, due to limited public expenditure on education, households pay for the running of schools in addition to other costs for uniforms, textbooks, and meals for their children. For the poorest households, out-of-pocket basic education spending poses a barrier to education. Having a spending level commensurate with its level of income and the needs of its education system is crucial for the country’s effort to enhance economic growth and poverty reduction. 36 School dropout is a serious issue in the education sector. Several reasons have been cited for this, including lack of funds to cover high household out-of-pocket spending on education, early marriage of girls, and children dropping out to work and support the family. Late entry into the education system remains a problem and varies substantially across states. Only about one in three children of primary school entry age (six) enters Grade 1 on time. The late entry issue explains the big gap between net and gross primary enrollment rates in Sudan. This is largely attributable to delayed entry into primary school. It would be important to have a full understanding of the reasons for late entry to find ways to resolve this concern. There is a strong relationship between education and poverty outcomes. Household wealth and parents’ (particularly mother’s) education are major determinants of school attendance among children between 6 and 18 years of age. Policy Options The Government of Sudan could take several steps to address the issues highlighted above. First, the findings suggest that free tuition alone is not enough for enrollment and attendance. Efforts to improve secondary school enrollments must start with programs that would boost primary school completion rates. In addition, social protection programs that can enable households to cope with negative shocks would enable their children to stay in school when a negative shock hits. Second, the government needs to design and implement programs to address the issue of school dropouts. For example, having social safety net programs that support people with disabilities or health insurance schemes may minimize the dropout rate of children who leave school to support family members. Conditional cash transfers may help ensure that such children attend school. Third, while the cost of schooling is a major constraint for school attendance of both boys and girls, it disproportionately affects girls. In the face of limited resources parents prioritize boys for schooling over girls (World Bank 2017). Investing in female education is very important for poverty reduction and development, especially as we also find that school attendance increases with the mother’s level of education. Fourth, food for education programs should be promoted, particularly targeting children in poor households. Providing on-site meals for school children and take-home rations (targeted food transfer) may encourage children to stay in school as it cuts down their family’s out-of-pocket expenses. Fifth, in the spirit of reducing the burden of out-of-pocket expenses for households, free access to textbooks may also improve education outcomes for children from poor households with long- term implications for poverty reduction. It is encouraging to learn that textbook provision will be conducted within the new Basic Education Development Project (BEDP). Such programs that encourage free access to textbooks in classrooms for students in disadvantaged 37 communities may not only boost education outcomes (increase school attendance or reduce dropouts) but would also have direct impact on learning outcomes (quality of education) and poverty (parents saving money they would normally spend on books). This would have long-term impact on social progress, poverty, and inequality. Sixth, various interventions can be considered to delay marriage, and support girls who marry early.10 Programs to educate girls and their parents about the negative impacts of this would be very useful. Curbing early marriage and pregnancy will also help reduce the fertility rate, and subsequently the dependency ratio, with positive impact on poverty. Seventh, expand literacy programs for those who may not have opportunity to go through formal education.. Finally, increase public spending and efficiency in education. An increase in public expenditure on education should be accompanied by measures to increase efficiency of the system. Health Outcomes and Poverty Sudan made progress during the last decade on health outcomes, but more improvements are needed. Infant, under-five, and child mortality rates in Sudan all declined considerably between 2010 and 2014, although they remain high. Despite showing progress, Sudan might have missed meeting MDG-related goals on health by 2015. Child survival varies widely between different population groups. In Sudan, there is substantial inequity in health outcomes between females and males, rich and poor, urban and rural areas. For child and under-five mortality the gap between the poorest and the richest quintiles remains huge. Gender disparities exist in childhood health care in favor of girls. Also, childhood mortality remains a rural phenomenon, and there are substantial variations in mortality rates across states. Malnutrition remains a serious issue in Sudan and continues to be widespread. Stunting and underweight both deteriorated between 2010 and 2014, while wasting remained stable. For its level of income, Sudan is underperforming on all health and nutrition outcomes. Infant, under-five mortality, stunting, wasting, and underweight are all higher than expected values compared to countries with similar incomes. Stunting is weakly correlated with poverty, but strongly correlated with gender, location, and mother’s education. 10Such interventions will include: (a) Empowering girls with information, skills, and support networks; (b) Educating and mobilizing parents and community members; (c) Enhancing the accessibility and quality of formal schooling for girls; (d) Offering economic support and incentives for girls and their families; and (e) Fostering an enabling legal and policy framework. See Wodon et al. (2016) for more details. 38 Policy Options The evidence of health outcomes presented in this paper points to areas where the country is performing less than expected and which require special attention. First, stunting results suggest that improving nutrition alone may not have much impact if no progress is made on education and health care access and quality. There is a need for a multisectoral approach toward reducing stunting, which will depend on having a more holistic view of the inequities and gaps in access to adequate levels of the underlying determinants of nutrition addressed by nutrition-sensitive interventions. For example, the finding that the probability of stunting diminishes among better-educated mothers implies that programs that promote women’s education would be useful for minimizing stunting levels. Second, enhance the level and efficiency of the government spending on health expenditures. The health system in Sudan, like in many developing countries, relies heavily on direct spending by health care users at the place and moment of service delivery, resulting in high out-of-pocket health spending and exposing individuals to the risk of poverty, and perhaps death. Third, provide better service delivery in the health sector such as expanding preventive health care. This would result in increased access to and quality of health care (demand and supply). Finally, expand immunizations or vaccination campaigns for children. Implement programs that educate families about the importance for children to receive all necessary immunizations or vaccinations in a timely manner. 39 REFERENCES Anyanwu, J. C. 2005. “Rural Poverty in Nigeria: Profile, Determinants and Exit Paths .” African Development Review 17 (3): 435–460. Anyanwu, J. C. 2013. “The Correlates of Poverty in Nigeria and Policy Implications.” African Journal of Economic and Sustainable Development 2 (1): 23–52. Appleton, S., A. Balihuta. 1996. “Education and Agricultural Productivity: Evidence from Uganda.” Journal of International Development 8 (3): 415–444. Caulfield, L. E., S. A. Richard, et al. 2006. “Stunting, Wasting, and Micronutrient Deficiency Disorders.” Disease Control Priorities in Developing Countries 2nd ed., edited by D. T. Jamison, J. G. Breman, A. R. Measham, et al. New York: The World Bank and Oxford University Press. Education Sector Analysis. (2017). Federal Ministry of Education, Sudan. Eberstadt, N. 1995. The Tyranny of Numbers: Mismeasurement and Misrule. Washington, DC: The AEI Press. Etang, Alvin and Clarence Tsimpo. 2017. “Beyond Income Poverty: Nonmonetary Dimensions of Poverty in Uganda.” Policy Research Working Paper 8111, World Bank, Washington, DC. Gable, Susanna, Hans Lofgren, and Israel Osorio-Rodarte. 2014. Country Development Diagnostics Post-2015: Uganda. Washington, DC: World Bank. Galasso, Emanuela, and Nithin Umapathi. 2009. “Improving Nutritional Status Through Behavioural Change: Lessons from Madagascar.” Journal of Development Effectiveness 1 (1): 60–85. Hughes, B. and M. T. Irfan. 2007. “Assessing Strategies for Reducing Poverty.” International Studies Review 9 (4): 690–710. Jolliffe, D. 2002. “Whose Education Matters in the Determination of Household Income? Evidence from a Developing Country.” Economic Development and Cultural Change 50 (2): 287–312. Preston, S. H. 1975. “The Changing Relation Between Mortality and Level of Economic Development.” Population Studies 29 (2): 231–248. Pritchett, L., and L. Summers. 1996. “Wealthier is Healthier.” Journal of Human Resources 31 (4): 844–868. 40 Seff, I., D. Jolliffe, K. Kafle, and J. Kastelic. 2014. “Dynamics of Wellbeing in Tanzania 2008-2013: Analysis Based on Three Waves of the National Panel Survey.” Background paper prepared for Tanzania Poverty Assessment Report (May 19). Sen, A. 1998. “Mortality as an Indicator of Economic Success and Failure.” The Economic Journal 108 (446): 1–25. Tilak, J. B. G. 2007, “Post-elementary Education, Poverty and Development in India”, International Journal of Educational Development. 27 (4): 435–445. UNICEF. 2007. Progress for Children: A World Fit for Children. Statistical Review. New York: UNICEF. Waldmann, R. J. 1992. “Income Distribution and Infant Mortality.” The Quarterly Journal of Economics 107 (4): 1283–1302. WHO. 1999. Highlights on Health in Uzbekistan. Copenhagen: WHO.Wodon, Q., C. Male, K. A. Nayihouba, A. O. Onagoruwa, A. Savadogo, A. Yedan, J. Edmeades, A. Kes, N. John, L. Murithi, M. Steinhaus, and S. Petroni. 2017. Economic Impacts of Child Marriage: Global Synthesis Report. Washington, DC: World Bank. World Bank. 2017. All Hands on Deck: Reducing Stunting through a Multisectoral Approach in Sub-Saharan Africa. World Bank Report. 41