A. Socioeconomic consequences of Illness and Death  

Research Question
This ongoing project investigates the economic impact of the AIDS pandemic on households and individuals. Our completed work to date and our current projects rely on the panel nature of ACDIS to identify the effects of illness and death in households.

Data sources and methods
This project uses the data collected in ACDIS, with an intensive focus on the socioeconomic data collected in the Household Socioeconomic modules (HSEs) and the Illness and Death study (IAD). We rely on the longitudinal nature of ACDIS data to identify the effects of mortality on households and individuals. Observing individuals and households both before and after a death allows us to move some distance in evaluating alternative explanations for a variety of outcomes following the death.

Findings
In “The Impact of Parental Death on School Outcomes: Longitudinal Evidence from South Africa,” (Case and Ardington, Demography 2006), we use the first two waves of the HSE, together with information about parental deaths, to investigate educational outcomes for orphans. We find that maternal orphans are significantly less likely to be enrolled in school, and have completed significantly fewer years of schooling, conditional on age, than children whose mothers are alive. Less money is spent on their educations on average, conditional on enrollment. Moreover, children whose mothers have died appear to be at an educational disadvantage when compared to non-orphaned children with whom they live. While this may simply reflect the fact that children are themselves ill, or are poor students, we find no evidence to support these hypotheses. Children whose mothers were alive when the first data were collected on educational attainment in 2001, but whose mothers subsequently died, were not lagging other children in enrollment or educational attainment in 2001. However, they had fallen behind when observed after their mothers’ deaths. Our evidence is consistent with mothers’ deaths having a causal effect on children’s educational outcomes.

We have also documented the economic impact of death in the households followed in ACDIS, by collecting information on funeral costs and financing for all deaths that occurred between 2003 and 2005 in the DSA. In “Paying the Piper: The High Cost of Funerals in South Africa,” (Case, Garrib, Menendez and Olgiati, 2008), we analyze funeral arrangements following the deaths of 3,751 people who died between January 2003 and December 2005 in the Africa Centre Demographic Surveillance Area. We find that, on average, households spend the equivalent of a year’s income for an adult’s funeral, measured at median per capita African (Black) income. Approximately one-quarter of all individuals (generally the elderly) had some form of insurance, which helped surviving household members defray some fraction of funeral expenses. However, an equal fraction of households borrowed money to pay for the funeral.

Policy implications
One of the lasting effects of the HIV/AIDS crisis will be the impact it has on the education of the generation of children now of school going age. Throughout sub-Saharan Africa, the crisis is reducing educational attainment, a result that can be expected in turn to dampen economic growth and the health and general well-being of Africans. Policy makers are currently grappling with selecting policies to deal with the crisis, and disagreement abounds. Some researchers-cum-policymakers in South Africa argue that it would be unfair to provide special services to orphans when there are many poor children in South Africa whose parents are alive, who are also at risk for poor schooling outcomes. The evidence from ACDIS speaks directly to this question, showing that, in both poor and wealthy households, children who have lost mothers are at risk of poor outcomes, relative to the children with whom they live.

The financing of burials has implications for a household’s ability to maintain a stock of productive assets, to stake migrants in urban areas until they find work, to finance schooling, and more broadly to provide adequate nutrition and a healthy environment within which to raise children. This project has provided quantitative evidence on the extent to which funerals place households in the DSA at risk, taking productive resources and turning them into consumption (coffins, meat, groceries) at the expense of investment (education, capital goods). In a quarter of all funerals for individuals who died between 2003 and 2005 in the DSA, households borrowed money for the funeral, which can be anticipated to drain household resources well into the future. Our estimates suggest that households are expected to spend half of household income on funerals, net of the spending expected given the status of the deceased. Such elaborate funerals are unlikely to be sustainable, if the AIDS pandemic continues to take lives at such a rapid rate. Understanding coordination failures between communities, or among members of extended households, will be important if one wishes to move to smaller, less expensive funerals.

Ongoing research
We are currently building on our previous research by using the first five waves of HSE data, and information on household deaths, to investigate the medium and longer run economic impact of a household death, on children and adults.

B. Intergenerational support and labour supply  

Background
In many parts of the developing world, rural areas exhibit high rates of unemployment and underemployment. Understanding what prevents people from migrating to find better jobs is central to the development process. This issue is especially salient in South Africa, where differences in earnings and employment rates between rural and urban areas are large and persistent. 

Research question
We examine whether, and to what extent, the arrival of a large, stable source of income in the form of the South African old-age pension leads to change in the labour force attachment of the prime-aged adults in households containing pensioners. If households pool income, we might expect prime-aged adults who share resources with pensioners to reduce their work hours, or choose not to participate in the labour market, when pension receipt begins. Alternatively, if social transfers allow households to overcome credit constraints, enabling households to bankroll potential migrants or potential work seekers who need financial support to look for jobs, then social transfers like the pension may promote employment and help households break out of poverty traps. It is an empirical question whether, and to what extent, resources channelled into households in the form of the pension change the labour market behaviour of household.

Data sources and methods
The longitudinal nature of ACIDS data allows us to use the timing of events—pension receipt, migration, labour force participation— to estimate causal pathways. Our ability to compare households and individuals before and after pension receipt, and pension loss, allows us to control for a host of unobservable household and individual characteristics that may determine labour market behaviour.

Findings
In “Labor supply responses to large social transfers: Longitudinal evidence from South Africa,” (Ardington, Case and Hosegood, 2009), we examine whether binding credit constraints and childcare constraints limit the ability of households to send labor migrants, and whether the arrival of a large, stable source of income – the South African old-age pension – helps households to overcome these constraints. Specifically, we quantify the labor supply responses of prime-aged individuals to changes in the presence of pensioners, using data from ACDIS. Our ability to compare households and individuals before and after pension receipt, and pension loss, allows us to control for a host of unobservable household and individual characteristics that may determine labor market behavior. We find that large cash transfers to elderly South Africans lead to increased employment among prime-aged members of their households, a result that is masked in cross-sectional analysis by differences between pension and non-pension households. Pension receipt also influences where this employment takes place. We find large, significant effects on labor migration upon pension arrival. The pension’s impact is attributable both to the increase in household resources it represents, which can be used to stake migrants until they become self-sufficient, and to the presence of pensioners who can care for small children, which allows prime-aged adults to look for work elsewhere.

Policy implications
Previous evidence on labour supply responses to pension receipt relied on the analysis of cross sectional data. An influential paper by Bertrand, Mullainathan, and Miller (2003) found that prime-aged adults living with pensioners had significantly lower rates of labour force participation than those in households without a pensioner. They concluded that “the pension dramatically reduces the labour supply of the prime-age members of the household.” In drawing this conclusion they assumed that the introduction of the pension had no causal effect on household composition.

When we estimate labour supply effects using only cross-sectional data from the household socioeconomic survey, we replicate the findings of the earlier cross-sectional analyses. When we turn to longitudinal analysis, however, we find that the pension has a positive impact on the probability of employment and labour migration. One of the reasons why the cross why the cross-sectional and panel data results differ is because the arrival of a pension attracts new household members who are significantly less educated and less likely to be employed.

In addition to documenting a positive effect of the pension on prime aged adult labour supply, our work highlights the role of credit constraints in prohibiting job search. 

Ongoing work
The potential integration of ART data with ACDIS will create new research opportunities. Conditional on the data being available, we plan to build on our previous research by investigating the socioeconomic determinants of ART adherence. Although identifying causal effects of socioeconomic status on adherence will be difficult, we may be able to use the South African old age pension to quantify the impact of resources on adherence to medical protocols. The presence of a pension may help us understand whether and how the presence of a stable source of income affects ART compliance. We plan to compare compliance patterns of individuals who are living with a pensioner with those of people who are not. The fact that women become eligible for a pension at precisely age 60, and men at age 65, allows us to look at the impact of the pension separately from that of having an older person in the household. Both may matter for compliance: older persons may provide the sort of stability necessary to adhere to treatment, independently of any resources attached to them. We plan to follow compliance in three types of households: those in which there are no older adults; those in which there are older adults just shy of pension age (containing women aged 55 to 59, and/or men aged 55 to 64); and those containing a pensioner. Testing for differences in compliance in these three types of households will identify the importance of the presence of older members and, separately, the impact of household resources. Over time, we should be able to document whether the loss of a stable source of income (for example, when a pensioner dies) has negative effects on health behaviors, including ART compliance.

C. The socioeconomic consequences of early childbearing

Background
Fertility rates have fallen steadily in South Africa in recent decades.  Census data indicate significant declines in the Total Fertility Rate in all regions and all population groups between 1996 and 2001 (Moultrie and Dorrington, 2004).  Perhaps more interesting than the trends in the TFR, however, are the striking patterns observed in the age profile of fertility.  While overall fertility has dropped considerably, fertility rates continue to be relatively high during teenage years, with most of the fertility non-marital.  Garenne et al. (2000) document the high level of teen fertility in the Agincourt demographic surveillance site.  They interpret their results as indicating that reproductive health services do not adequately serve young unmarried women. Similarly, Camlin et al. (2004) and Moultrie et al. (2008) examine fertility trends in the Africa Centre demographic surveillance site and find that while overall fertility declined rapidly, fertility levels among adolescents remain high and have not changed in decades. 

Research question
One obvious policy question coming out of South African fertility patterns is whether high levels of teen fertility have a negative impact on women’s human capital, which in turn could have a negative impact on their employment, their earnings, and the well-being of their children.  We plan to investigate the relationship between teen fertility and a range of socio-economic outcomes.

Data sources and methods
The consequences of teen fertility are difficult to evaluate, as the large and highly contentious literature on teen fertility in the United States has demonstrated (see Ribar (1999), Hotz et al. (2005) and Ashcraft and Lang (2006) for a reviews of this literature). Since schooling and fertility are simultaneously determined, it is difficult to identify the causal relationship of early childbearing on educational outcomes and subsequent socioeconomic status. Researchers have applied a range of techniques in the attempt to identify the causal effect of teen childbearing including sibling fixed effects and quasi-experimental methods or instrumental variables procedures.

Empirical evidence on the socioeconomic consequences of teen childbearing in South Africa is scarce. Madhavan and Thomas (2005) find a negative relationship between early childbearing and schooling using South Africa census data, but point out that it is difficult to give the results a causal interpretation. Marteleto et al. (2008) use data from the Cape Area Panel Study to investigate how early life characteristics affect young-age childbearing and the factors facilitating school enrolment after childbearing. They find that young mothers who were weaker students prior to giving birth were less likely to enrol in school subsequent to giving birth. However, a significant proportion of young mothers managed to continue with their studies, particularly African girls.