Abstract
Tightly knit extended families, in which resources are shared across generations, characterize many developing countries. These intra-family flows mean that public policies may affect a very different group of people than the one they target. To assess the empirical importance of these effects, we study a cash pension program in South Africa that targets the elderly. Focusing on three-generation households, we use the variation in pension receipt that comes from differences in the age of the elder(s) in the households. We find a sharp drop in the labor force participation of prime-age men in these households when elder women reach 60 years old or elder men reach 65, the respective ages for pension eligibility. We also find that the drop in labor supply diminishes with family size and educational attainment. Other findings suggest that power within the family plays an important role: labor supply drops less when the pension is received by a man rather than a woman, and middle-aged men reduce labor supply more than younger men. This public policy seems to have had large effects on a group—prime-age men living with the old—quite different from the one it originally targeted.
Introduction
In many developing economies, the "nuclear family" is an insufficient model for understanding resource allocation. Instead, resources are shared across tightly-knit extended families. When a government designs a policy to help a specific group—such as the elderly—it often ignores how intra-family sharing might shift the actual impact of that policy. This paper investigates the "spillover" effects of a generous South African pension program, demonstrating that targeted transfers can significantly alter the work incentives of non-targeted household members.
The South African Pension Context
The study utilizes the 1993 Integrated Household Survey of South Africa, a collaboration between the World Bank and the South African Development Research Unit (SALDRU). The focus is on African households, where three-generation living arrangements are prevalent and the pension represents a massive income shock—often more than double the median per-capita income. At the time of the study, women became eligible for the pension at age 60 and men at age 65.
Methodology
To isolate the effect of pension income from other household variables, the researchers employ a robust quasi-experimental approach:
- Instrumental Variables (IV): Rather than correlating pension income with employment (which could be biased by unobserved household traits), the authors use age-eligibility rules as an instrument.
- Regression Discontinuity Logic: The study looks for sharp changes in labor supply exactly at the age thresholds (60 for women, 65 for men). If a 59-year-old woman in the house does not affect her grandson’s employment but a 60-year-old woman does, the effect is likely attributable to the pension.
- Sample Restriction: By focusing specifically on three-generation households, the authors control for the selection bias regarding who chooses to live with their elders.
Key Findings
1. The Impact on Labor Supply
There is a sharp, statistically significant decrease in the labor force participation of prime-age men when an elder in the household reaches pension eligibility. The authors estimate a labor supply elasticity of roughly -1, meaning a 10% increase in household income leads to a roughly 10% decrease in the probability of a prime-age male being employed. These individuals do not show up as "unemployed" (actively seeking work); they exit the labor force entirely.
2. The "Altruistic Grandma" Effect
The drop in male labor supply is significantly larger when the pension is received by a woman rather than a man. This suggests that female pensioners may be more altruistic toward younger relatives or have less bargaining power to retain the funds for personal use.
3. Power Dynamics and Clout
Middle-aged men (closer to age 50) reduce their labor supply more than younger men (in their 20s). This suggests that "clout" within the family hierarchy matters; older men may have more power to extract resources from the pensioner than their younger counterparts.
4. Ruling Out Caregiving
A common counter-argument is that younger family members stay home to care for the elderly. However, the study finds no effect on female labor supply. Since women typically provide the bulk of home care in this context, the fact that only men stop working points toward an income effect rather than a caregiving requirement.
Limitations and Considerations
While the results are striking, the authors acknowledge certain constraints:
- Cross-Sectional Data: The data is a snapshot in time. A panel dataset following the same families over several years would provide more granular detail on the exact timing of labor force exits.
- Migration Bias: While the authors tested for recent migration and found no evidence of it driving the results, there remains a theoretical concern that less employable men might move back home specifically because a relative has secured a pension.
- Human Capital Trade-offs: The study focuses on the negative labor supply of men but does not fully explore potential positive trade-offs, such as whether these transfers allow children to stay in school longer or improve household nutrition.
Conclusion
The South African pension program effectively acts as a redistribution mechanism that alters the economic behavior of the young. By providing a guaranteed income to the elderly, the policy inadvertently creates a disincentive for prime-age men in the same household to participate in the labor force. This highlights the necessity for policymakers to consider the extended family structure when designing social safety nets in developing nations.
Summary updated