INEQUALITY IN SOUTHERN AFRICA: AN ASSESSMENT OF THE SOUTHERN AFRICAN CUSTOMS UNION COUNTRY BRIEF: ESWATINI1 Eswatini has made the slowest progress in reducing inequality among the SACU countries and remains among the top 15 percent of unequal countries. A critical source of inequality is differences in productivity-related characteristics (the primary income distribution), particularly in tertiary or post-secondary educational attainment and skills, which are key to accessing decent jobs. Another important driver of inequality is a widening welfare gap between households with younger and economically active members and those without that demographic dividend. Enhancing the quality of education, promoting skills development, and creating jobs especially for poor and young people are, therefore, key to accelerating the reduction in inequality. A. Progress towards equality The pace of reducing inequality has been slow, and this rate of decline is among the slowest in SACU (panel Eswatini remains among the top 15 percent of the world’s a). Urban areas tend to be more unequal than rural ones; most unequal countries (Figure 1, panel b). The country’s however, inequality in rural areas has increased, notably Gini coefficient2 for consumption per capita fell from 53.2 in in the Shiselweni region, where consumption growth has 2001 to 50.9 in 2016—at only 2.3 Gini points over 15 years, generally been skewed in favor of the rich. Figure 1. The pace of reducing inequality has been slow a. Regional comparison of Gini coefficients b. International inequality comparison 70 68.40 68.77 80 South Africa, 2018 65 66.96 66.52 60 Namibia, 2015 61.60 Botswana, 2015 60.50 Eswatini, 2016 60 Lesotho, 2017 Gini coefficient 59.30 40 Gini coefficient 55 53.20 51.80 54.90 50 20 50.90 45 45.10 0 Countries ordered from lowest to highest Gini 40 Lesotho Eswatini Botswana Namibia South SACU Africa Region* Around 2001 Latest estimate Source: World Bank 2022. Note: Panel a presents the Gini coefficient of consumption for the whole SACU region, based on the earliest and latest rounds of household surveys from member countries. 1 This brief is largely drawn from a SACU regional report on inequality prepared by the World Bank and co-authored by Victor Sulla, Precious Zikhali, and Facundo Cuevas. The report uses the framework highlighted in Box 1: World Bank. 2022. Inequality in Southern Africa: An Assessment of the Southern African Customs Union. Washington, DC: World Bank. 2 The Gini index is a measure of inequality in a distribution. It varies between 0 (perfect equality), where every individual enjoys the same level of consumption per adult equivalent, and 1 (complete inequality), where a single individual accounts for all consumption. 1 B. Key drivers of inequality Differences in tertiary or post-secondary education Human Capital Index (HCI), which is lower than predicted attainment are a key driver of inequality. Disparities for its income level. Eswatini’s HCI score in 2020 suggests in higher education, which is key to human capital that Swazi children born today would be only 37 percent accumulation, account for about 36 percent of overall as productive as adults than they could have been with a inequality. (Figure 2). As in Lesotho, only a small proportion complete education and good health. The main reasons for of Eswatini’s people have a tertiary education. This is this low score are low adult survival rates (e.g., because of consistent with the country’s ranking in the World Bank’s HIV/AIDS) and particularly poor learning outcomes. Figure 2. Decomposition of inequality a. Contribution of selected factors to inequality (%) b. Breakdown of the contribution of selected factors to inequality, percentage points 40 39.9 37.4 Gender 0.1 0.8 Demographics Hh Size 6.8 5.3 Age 15.5 22.8 30 28.9 Contribution to Inequality -1.6 25.8 Secondary -0.9 Education Primary -1.1 2.0 22.3 21.9 Post-Secondary 42.6 20 36.3 -0.3 Industry 0.8 Labor 7.1 12.0 11.8 Participation 6.2 market 19.0 Skills 14.9 10 2.4 Location Region 1.2 Urban 9.6 10.6 0 Demographics Education Labor market Location 0 10 20 30 40 Contribution to Inequality 2001 2016 2001 2016 Source: World Bank 2022. Note: Panel (a) reports the contribution (%) of spatial, demographic, education, and labor market factors to overall inequality. Panel (b) disaggregates the contribution of each of these four factors into its various subfactors. Demographic factors contribute significantly to Disparities in employment outcomes are the primary inequality through the age of household members contributor to inequality. Differences in labor market and household size, with an aggregate contribution to attributes (labor force status, industry of employment, overall inequality of 29 percent. This is largely because the and occupation type) account for 22 percent of overall age profile of the household, and its size, affect the capacity inequality. Among these attributes, differences in of a household to engage in economic activities, and the occupation type (such as senior managers, professionals, resources needed to meet household needs. In that sense, and clerks), which reflect skills differences, contribute most households with more members of working age, and fewer to inequality. The “industry in which people work” does not dependents, have a “demographic dividend”. seem to affect inequality significantly. C. Inequality and the cycle of income generation It is useful to analyze the process of household is affected by access to factor endowments (or assets), income generation to identify the sources of high and such as education, skills, land, and capital, as well as their persistent inequality. The first step focuses on the pre- use and returns from interaction with markets. The third income distribution, which is the inequality of opportunity explores the secondary income distribution, assessing how that arises from differences in circumstances at birth and inequality is influenced by taxes and government transfers. during childhood, such as gender, race, location, parental Finally, the fourth component relates to the tertiary income education, and family wealth; these differences create distribution—the disparities that remain after accounting expected inequalities in income distribution even before for the role of social services (such as education, health, and people interact with factor markets. The second component infrastructure). looks at the primary income distribution—how inequality 2 Box 1. Framework to assess sources of income and consumption inequality The SACU regional inequality report uses an innovative framework built around the process that underlies household income generation to identify sources of high and persistent inequality. The framework is organized into four sequential components, presented in Figure 3. Figure 3. Framework to assess sources of income and consumption inequality Pre-income distribution: Inequality of opportunity Arising from circumstances at birth or family backgroung (including gender, race and parental education). Primary income distribution: Inequality of pre-tax income Influenced by differential access to, use of, and returns to assets (such as education, labor, land and capital). Secondary income distribution: inequality after taxes and transfers Affected by the structure, implementation capacity and incidence of fiscal policy. Tertiary income distribution: Inequality after social services Resulting from the provision of and access to public services (such as health, education and infrastructure). Source: World Bank (2022). High inequality of opportunity Geography has a significant impact on inequality of Inequality of opportunity, arising from inherited opportunity. Relative to the rest of SACU, the contribution circumstances, are a key determinant of overall of the urban-rural divide to inequality (in both consumption inequality.3 An analysis of the inherited circumstances and earnings) is highest in Eswatini. This reflects factors such for which data are available in all SACU countries— as limited access to basic services in rural areas. Such limited gender, age, region of residence (urban-rural, and access also correlates negatively with poverty levels, with regions)—suggests inequality of opportunity explained poor people simultaneously deprived in multiple ways. over a quarter of Eswatini’s overall inequality in per capita High inequality is associated with relatively low consumption in 2017. This was the highest in SACU and intergenerational mobility.4 Although the evidence up from 20.8 percent in 2001. Including other factors for suggests upward educational mobility for young people, which data is available (specifically early education and intergenerational earnings mobility remains limited. The parental education) suggests that as much as 38.5 percent relationship between earnings across two generations of consumption inequality could be attributed to factors is strong; in fact, within SACU, Eswatini is second only to beyond the control of individuals, driven by differences in Namibia in terms of the strength of this relationship. access to early education and parental education (World Bank 2020). 3 Inequality of opportunity is defined as the component of inequality attributable to differences in inherited circumstances beyond the control of the individual. The estimates should be taken as a lower-bound estimate of inequality of opportunity, since they capture the role of only a subset of circumstances—those that were available in the household survey. 4 Intergenerational mobility is defined as the extent to which a person’s life outcomes correlate with those of their parents. 3 Inequities in labor market outcomes benefits poor people in relative terms; however, they receive only a small share of overall health spending, even The lack of new jobs results in poor labor market relative to other SACU countries. Progressivity for hospital outcomes. At 23  percent in 2016, unemployment levels health care is the lowest in the region: poorer people are high, particularly among young people, women, and have limited access to hospital care, given the prohibitive rural residents. New jobs tend to be in low-productivity costs of access. Although growing numbers of Swazi have services sectors and are associated with high economic better access to health facilities, health outcomes remain vulnerability. Agriculture (most of which is subsistence) poor: Eswatini has the world’s highest proportion of adults suffers from low productivity, further widening the labor (ages 15–49) living with HIV, along with poor maternal and market disparities between rural and urban residents. child health, and a rising incidence of noncommunicable Government is the largest formal employer, and the diseases. earnings premium for working in the public sector is large. The informal sector, self-employment, and dependence on remittance flows from South Africa have grown. High vulnerability to climate change risks and economic shocks For those in employment, earnings are starkly polarized. Earnings from employment are the second most polarized Because poor people have limited coping mechanisms, in the region (behind Namibia) and are more polarized than they suffer disproportionately from the adverse income. The sources of wage inequality include differences effects of climate change. Estimates of the incidence in the sector of employment, education, and location. and distribution of drought and floods in the 2015/16 El The contribution of differences in education to earnings Niño event suggest that 82 percent of people in the lowest inequality is highest in Eswatini. Also, as in other SACU quintile were affected by the drought, as against only 75 countries, there is a substantial gender wage gap that is not percent in the richest quintile. Controlling for observable explained by differences in occupation or education. characteristics, households in areas affected by droughts or floods experienced, on average, a 9.2 percent loss in per capita consumption relative to their non-affected Inequities and inefficiencies in social spending peers. This is of particular concern because reductions in Social assistance helps limit inequality, but it is less poor people’s consumption can lead to undernutrition, effective in Eswatini than in the rest of SACU. The with negative health consequences in the short term and reduction in the Gini coefficient because of social assistance potentially serious long-term consequences. These include ranges from 1.9 percent in Eswatini to 10.5 percent in South a high risk of stunting, impaired cognitive development, Africa. The relatively low impact on inequality in Eswatini lower school attendance rates, reduced human capital is due to the low value of the benefits and poor targeting attainment, and higher risks of chronic disease and health of some programs. Although the education grant for problems in adulthood. orphaned and vulnerable children subsidizes secondary The COVID-19 pandemic has increased economic school fees (including for poor children), the 2021 vulnerability and likely widened disparities. The Education Sector Analysis conducted by the World Bank pandemic has exacerbated the triple problem of high found that an estimated 70 percent of children who were unemployment, poverty, and inequality. Poor and eligible for the grant did not receive it (World Bank 2021). vulnerable people are again disproportionately affected Further, there is no social assistance program that provides because they lack the productive capacity to deal with and direct financial support to help poor households keep their recover from the adverse socio-economic impacts of the children in school or to allow them to access higher levels crisis. of education. Social protection programs can potentially offset A significant share of Eswatini’s resources is allocated consumption losses from shocks; however, they to education, but spending could be more efficient. currently cover only a small fraction of climate- About 16 percent of total public spending is on education, vulnerable households. Also, although most of the and public spending per pupil is relatively high. Most of this benefits of social protection programs do reach poor is on recurrent costs, mainly teachers’ wages and salaries, people, they perform less well at reaching the poorest especially at the primary level. Spending per primary people. Substantial numbers of poor and vulnerable Swazi student is a fraction of that per tertiary student. Tertiary people remain unprotected, despite government efforts. education spending is the most regressive in the region Further, benefit levels are low, and the programs do not and disproportionately benefits the wealthy. scale up in the event of a disaster (World Bank 2020). Overall, Health outcomes are not commensurate with spending, enhancing the social protection system’s responsiveness to suggesting that spending is inefficient. Health spending shocks will strengthen resilience. 4 D. Policy considerations to accelerate the reduction in inequality Strengthen human capital to help equalize priorities. A better targeting mechanism, with support from opportunities. In education, this requires enhancing the the social registry, would improve targeting outcomes and quality of education (ensuring that children are learning in make the use of resources more effective. In addition, more school) by improving literacy and numeracy outcomes in complementarity between social protection programs and the early grades. Student retention also is a concern, with education, nutrition, and health investments would help only 89 percent of children enrolled in grade 1 completing maximize their impact, especially among young people. primary school; the completion rate at secondary level is only 47 percent. Retention is even lower among orphans Strengthen the resilience of poor and vulnerable and vulnerable children. In health care, the focus is on households to climate risks and economic shocks. improving poor people’s access to health services. This requires investing in poor and vulnerable households’ capacity to prepare for, cope with, and adapt to shocks Create jobs and ensure equitable access. Growing without falling (deeper) into poverty. It implies improving the private sector to enhance its capacity to generate institutional and legal frameworks for disaster management productive jobs for all is key to reducing poverty and and ensuring disaster management agencies are well- inequality. This requires increasing labor returns in important capacitated and well-prepared to identify and reach poor sectors such as agriculture, which is characterized by low and vulnerable households when disasters and/or shocks investment and productivity. Overall, growing the private strike. Because these households tend to rely heavily on sector and fostering equitable access to employment natural resources, sustainable environmental and natural opportunities requires demand- and supply-side reforms, resource management is an important element of climate such as improving the business environment, deepening resilience. A focus on sustainable land management regional integration, and improving entrepreneurship practices, including the adoption of climate-smart through targeted policies that promote skills development. agriculture, and integrated water resource management A credible commitment to transparent and effective policy is needed to help mitigate the impact of growing water implementation and a level playing field for the private scarcity. Enhancing the responsiveness of the social sector would help create a conducive environment for protection system to shocks is critical; this requires an private investment and job creation. integrated social registry with automated databases, along Improve the equity and efficiency of spending. In with an early warning system that informs, through clear social protection, an integrated social registry, supported and transparent rules, when and where to scale support. by modernization of the social protection system, would It is also important to digitize government-to-person improve efficiency and policy coordination. Modernizing (G2P) payments faster, but without excluding vulnerable social protection systems and improving policy populations. coordination among different ministries are already policy Selected references World Bank. 2020. The Kingdom of Eswatini Toward Equal Opportunity: Accelerating Inclusion and Poverty Reduction. Systematic Country Diagnostic. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/34970 World Bank. 2021. Eswatini Education Sector Analysis 2021. Washington, DC: World Bank. https://openknowledge.worldbank.org/ handle/10986/35787 World Bank. 2022. Inequality in Southern Africa: An Assessment of the Southern African Customs Union. Washington, DC: World Bank. 5