A prospective birth cohort study initiated at the time of political transition

Rising resistance to Apartheid, the policy of separate and unequal treatment of Black people in South Africa, came to the boil in the 1980s, particularly in the densely populated economic hub on the Witwatersrand, including Johannesburg and Soweto. The Abolition of the Influx Control Act of 1986, which reversed only one of many pieces of inter-related legislation that controlled the movement, living and working conditions of Black people, symbolised the tangible beginning of the breakdown of the Apartheid state.

The free movement of people led to rapid urbanisation from a relatively low level, particularly among Black people. It was anticipated to compound existing problems in urban areas, including lack of employment, inadequate housing, poor services, rising crime, long commuting distances and disrupted family life.8 Among other initiatives to better understand and prepare for changing health patterns and service needs with increasing urbanisation, scientists at the South African Medical Research Council and the University of the Witwatersrand established a birth cohort to study the impact of the anticipated political, economic, nutritional and health transitions.

The Birth to Twenty+ (B20+) cohort study, dubbed ‘Mandela’s Children’ and ‘children of the 90s’, included all children born in the 6 weeks following 23 April 1990 within the municipal boundaries of Soweto and Johannesburg. Of the 5449 singleton births in the recruitment window, 3273 mothers and their children remained resident in the area 6 months after delivery and constitute the cohort. The aim was to study physical, psychological, social and educational aspects of the children’s development in the context of their socio-economic and family conditions in an intensive prospective longitudinal design. Various forms of attrition, including mortality, mobility and circular migration affect sample sizes during data collection rounds9

The sample breakdown by Apartheid race classifications reflects the population as a whole: most children were black African children (79%) who were historically disadvantaged by the Apartheid regime, with a relatively small proportion of Whites (6%) who were historically privileged.10 Now aged 29 years, participants have been assessed on human development indicators 22 times, generally on an annual cycle, with a larger gap between 22 and 28 years.11 12

The study provides a baseline for urban children’s biological, environmental, socio-economic and psychosocial experiences on the eve of South Africa’s democratic transition. It represents a unique source from which to examine whether the extreme disadvantages experienced by adults living under Apartheid continued to the first generation growing up under a democratic dispensation. The period was marked by significant political, economic and social changes; by all expectations, socio-economic gaps (across areas, social classes and race groups that were separated by Apartheid laws) should have narrowed. This paper summarises selected evidence that assessed whether societal shifts also reflected in improved socio-economic and health indicators across generations—those born at the dawn of democracy, and their parents who were born during Apartheid. The discussion focuses on the correspondence between secular socio-economic improvements and the (slow) change experienced by a young generation of Africans to reach a position beyond that of their disadvantaged parents.

Have socio-economic and health gaps narrowed?

South Africa’s first democratically elected government followed a multi-pronged agenda to tackle the country’s glaring socio-economic disparities. The approach ranged from investment in infrastructure to direct welfare initiatives. In 1994, the Reconstruction and Development Programme was adopted with the aim of creating a developmental state.13 South African government spending prioritises human development. Adjusting for inflation, per person public expenditure on health, schooling and social development has doubled in the two decades after Apartheid, and is the largest component of the national budget.14 Primary healthcare and primary schooling are provided without charge to the most vulnerable groups, and feeding programmes are widespread in poor schools. Most notably, more than 17 million poor South Africans receive unconditional cash transfers (approximately 30% of the population). These spending patterns rank South Africa as one of the most redistributive emerging economies. Despite the scale of social policy in monetary terms, inequality and poverty remain high relative to comparative countries (including Brazil).15 In addition, access to housing, water, sanitation and electricity has expanded dramatically—in many cases with subsidies towards the poor. A ‘headline’ view suggests that, at least in part, the programmes have been successful.

The prevalence of poverty—measured as households’ command over minimum levels of expenditure—declined in the early 2000s. The trend is attributed mainly to the rapid expansion of unconditional government cash transfers (for children, pensioners and people living with disabilities).16 The programme is one of the largest of its kind in a middle-income country, and has brought about marginal improvements along a range of socio-economic dimensions, including the linear growth of children and their progress through education.17 Among other benefits, the programme contributed to narrowing large spatial differences in self-reported hunger across former Apartheid homelands (areas to which the majority Black population were confined) and the rest of the country.18 However, these positive trends were not sustained. By 2015 official poverty rates started to rise again. About half of the population can now be described as ‘chronically poor’, never having escaped poverty at any point in time.19

In addition to social programmes that protect basic levels of consumption, public provision of housing, electricity, piped water and sanitation have expanded at large scale. Rural areas, with very low levels of provision, experienced greater improvement.20 For instance, only 26.5% of rural households had access to electricity for lighting in 1993, increasing to 65% in the first decade of democracy. Expansion of access to piped water occurred almost exclusively in rural areas. However, despite reductions in these disparities, historic separate development policies are still strongly evident in current socio-economic indicators.21 The changes brought about progress but not convergence.

Labour market conditions remain a key constraint to bridging socio-economic gaps. Non-discriminatory access to well-paying jobs is a precondition for alleviating inequalities and creating equality of opportunity. Yet, the economic environment has not changed in a manner to support these outcomes. Many South Africans—especially the younger generation—continue to be isolated from jobs that generate incomes necessary to climb the economic ladder.22 Even for those who find work, the playing field remains imbalanced. Despite legislated affirmative action, there is little evidence to suggest that racial discrimination has decreased in the work place, except among the very rich.23 Access to secure jobs—both in terms of permanence and adequate pay—is an essential precursor to creating stability for South Africa’s middle-class.24 However, many South Africans find it difficult to hold onto well-paid work, leaving them economically vulnerable. Upward economic mobility is therefore often temporary. Consequently, socio-economic disparities continue, with limited intergenerational income mobility.2 Despite significant social and economic changes, the World Bank describes South Africa’s experience as an ‘incomplete transition’.25

Other life outcomes closely mirror the unequal socio-economic conditions that exist between people, groups and areas. While under-five mortality dropped gradually between 1990 and 2015 and life expectancy has risen thanks to the provision of antiretroviral treatment for HIV patients, childhood stunting has remained high over the long run. Where improvements have occurred, they have been concentrated among the poorest, who typically have the worst nutrition and are under-served in the health system.26–28 However, access to healthcare continues to be strongly correlated with socio-economic status, despite the democratic constitution’s commitment to universal care.29 Wealthier individuals mainly access high quality private care paid by medical insurance through their employment, while poorer individuals rely on public services, often over-burdened and with inadequate resources. Racial imbalances in employment entail that White individuals are more likely than Black individuals to access private insurance and high quality care; consequently, the burden of disease is also more concentrated among the latter group.30

Because access to services is limited by socio-economic constraints, it is unsurprising that outcomes associated with healthcare follow the same pattern. To understand these gaps, we turn to evidence derived from a number of studies that use data from Bt20+. The study participants come from an urban setting, and therefore the findings cannot detect urban-rural disparities. Nevertheless, significant variation in socio-economic status within this group sheds light how it affects human development, including health.

In 1990, birth weight did not differ significantly by socio-economic category, although conditions of delivery did. Better off women were more likely to deliver their babies in a private facility and to have a Caesarean section delivery.31 In the early years, the physical growth of socio-economic groups diverged, creating substantial early life inequalities.32 At age 5, children from higher socio-economic status households were at par with international height benchmarks and heavier than weight benchmarks; in contrast, poorer children continued to fall below these thresholds. Even though all socio-economic groups grew heavier for their age by age 5, growth since birth was more substantial among the best-off children. Children who already lived in better conditions grew faster, even if social and health policies were aimed to benefit the poorest.

This socio-economic divergence also reflects in different experiences across Apartheid race classifications. White and Black African children had similar birth weights and heights up to 1 year of age. But White children grew better across childhood and adolescence, following unequal trajectories expected under Apartheid but not during democracy.10 Nonetheless, comparisons with data from an earlier study,33 show that African children from the Bt20+ cohort were taller on average, at the same ages, than those born and raised before democracy.10 No similar generational shifts occurred within the privileged White population, and this contributed to narrowing the gap. Alternative data sources confirm that there was a (slight) narrowing of the large and persistent racial gap in adult heights, even before democracy.34 Overall, as with other socio-economic indicators, there has been some progress across generations—this is partially attributed to a shift towards high energy diets among urban Africans.10 However, the political transition has not yet eliminated anthropometric inequalities across socio-economic and race groups.

Are children’s outcomes tied to the disadvantages faced by their parents?

There are some indications of biological and socio-economic improvements in the lives of younger generations from poorer backgrounds. However, with younger generations still having the highest unemployment rates over and above usual life cycle disadvantages,22 it stands to reason they are still—to varying degrees—tied to the disadvantages experienced by older generations who grew up during the Apartheid era, though these links may be weakening.

To better understand this, we explore evidence from two separate phases of human development. First, we consider whether there are direct associations between parent’s (especially mother’s) conditions and those of their children at birth. This sets the ‘baseline’ for likely transfers of disadvantages across generations. Second, we follow children to understand how strongly early life experiences continue to be correlated with outcomes across childhood and adolescence. This describes the potential leeway for children to break free of disadvantages that they inherit from their parents.