Table 1: Dependency Ratio (%), by Consumption Wellbeing in Ethiopia Quintile and Rural/Small Town (poorest)   In 2011/2012, Ethiopiaʼs Central Statistical Agency   Q1     Q2   Q3   Q4   Q5   (CSA), in collaboration with the World Bank, conducted the first wave of the Ethiopia Rural Socioeconomic Rural   121   114   105   102   89   Survey (ERSS), which collects detailed data on household welfare and income-generating activity. The Small   ERSSʼ sample includes 4,000 households that are town   108   100   74   81   65   representative of small towns and rural areas; wave 2 will be expanded to include urban areas and will run Total   121   114   105   102   88   from 2013 to 2014. This note looks at the relationship between wellbeing as measured by consumption and Education & Wellbeing Rankings outcomes related to education, health, and Literacy rates, although notably low throughout the demography. country, increase steadily as we move up the wellbeing ladder. This positive correlation holds true for males and Consumption Quintiles & Household females, as well as rural and small town subgroups. Composition However, note that the male literacy rate is 13 to 20 Consumption quintiles are used to categorize percentage points higher than the female literacy rate in households according to their welfare; the poorest any given quintile (see Figure 1). st households are grouped into the 1 quintile and the th richest households fall in the 5 quintile. We calculated Figure 1: Literacy, by Consumption Quintile & Gender the annualized consumption aggregate using the index of selected food and non-food items that reflect the bulk 70   of the consumption basket in Ethiopia. Final 60   consumption quintiles were constructed based on the Literacy  (%)   1 distribution of consumption per adult equivalent . 50   40   The dependency ratio, defined as the ratio of non- working individuals (<15 and >64 years old) to working 30   Male   individuals (15-64 years), can provide insight into the 20   Female   2 financial burden faced by specific populations . Overall, 10   we find that the dependency ratio for the poorest Total   0   quintile is almost 40 percent larger than the ratio for the top quintile. The poorest have a dependency ratio of 1st   2nd   3rd   4th   5th   (poorest)   121 percent, meaning for every productive-aged adult in ConsumpEon  QuinEles   the household, there are 1.21 non-working age individuals. This figure drops to 88 percent for the top The positive correlation between consumption and quintile (see Table 1). This pattern is also observed literacy is mirrored, and likely driven, by a similar within rural and small town households, though we note relationship between consumption and school a lower burden of dependents in small towns. attendance. As seen in Table 2, the likelihood of having                                                                                                                 ever attended school, regardless of gender, increases 1 Consumption per adult equivalent was calculated based on the as the individual moves up the wellbeing ranking. equivalent scales from Household Consumption and Expenditure (HCS) Survey 2010/11: Analytical Report, Central Statistical Agency Consistent with both males and females, the 2012 percentage of those ever having attended school 2 Households with no working individuals were excluded from the increases more than 50 percent when moving from the dependency ratio calculations.   poorest to the top quintile. indicators of malnutrition are: stunting, or low height-for- age; underweight, low weight-for-age; and wasting, low Table 2: Ever attended school (%), by Consumption weight-for-height. Rates of stunting and underweight Quintile & Gender are inversely correlated with consumption; the stunting   Q1   Q2   Q3   Q4   Q5   and underweight prevalence estimates are 60 percent and 36 percent, respectively, in the poorest quintile, vs. Male   43   53   59   62   67   36 percent and 20 percent in the top quintile. However, wasting prevalence estimates are not significantly Female   33   40   47   47   54   different at each consumption quintile, and we note no pattern as we move up the welfare ranking (see Figure Total   38   47   53   55   61   2). A similar phenomenon emerges for current school Figure 2: Malnutrition estimates, by Consumption enrollment which, when contrasted with having ever Quintile attended school, can shine light on the dynamic 70   changes in the Ethiopian education system. Only 50 Stunted   percent of school-aged children in the poorest 60   Underweight   consumption quintile are currently attending school, 50   3 compared to 70 percent of those in the richest quintile . Wasted   40   However, we note that these figures are 10 percentage %   points higher than their counterparts in Table 2, 30   suggesting improvement in recent years. 20   10   Health & Wellbeing Rankings 0   Although we observe a significant correlation between wellbeing and education outcomes, we find no evidence 1st    (poorest)   2nd   3rd   4th   5th   ConsumpEon  QuinEles   that the poorest are worse off with respect to health outcomes. Self-reports of an incidence of illness, chronic disease, or disability are not significantly Conclusion different for individuals at each of the five consumption A substantial body of literature speaks to the positive quintiles. There is also no statistically significant correlation between household wealth and indicators of difference in type of illness endured (for those reporting wellbeing. Analysis of the ERSS data supports this an incidence of illness) as we move up the wellbeing notion in the dimensions of household composition and ladder. However, malaria prevalence in the top quintile education. However, we find that the relationships is 32 percent, compared to only 23 percent in the between consumption and health are only significant for poorest quintile (see Table 3). stunting and underweight prevalence. Therefore, while household consumption may provide insight into many Table 3: Self-Reported Illness/Disability, by outcomes of wellbeing in Ethiopia, the ERSS data Consumption Quintile suggest that health outcomes are also tied to other (poorest)   factors.   Q1   Q2   Q3   Q4   Q5   4 This brief is based on data collected by the Central Illness   18   16   17   16   20   Statistical Agency as part of the Living Standards Malaria     23   32   34   26   32   Measurement Study – Integrated Surveys on Agriculture (LSMS-ISA) project. The full dataset is available for Diarrhea   14   8   8   12   12   download at CSA via http://www.csa.gov.et. 5 Other   63   60   58   61   56   Disability   10   8   8   8   8   The findings outlined in this brief are drawn from… A Profile of Well-being in Ethiopia: Evidence from the The ERSS also measures height and weight of children Ethiopia Rural Socioeconomic Survey, by Alemayehu ages 6-59 months to monitor malnutrition. The primary Ambel & Biratu Yigezu, The World Bank, June 2013                                                                                                                   3 School aged children are those ages 7 to 18. 4 Illness defined as incidence occurring in two months prior to survey. 5 “Other” illnesses include injury, dental, skin diseases, ear/nose/throat problems, and TB.