Kinship Pressure and Employee Selection

Many microenterprises in lower-income countries rely on relatives or family members to work as employees in their business. A common explanation for this phenomenon is that in environments with weak institutions, hiring relatives can reduce concerns of moral hazard or adverse selection that arise when hiring non-relatives. However, a substantial literature in anthropology suggests that firms hire relatives not as the optimal response to contracting frictions, but rather as a result of kinship pressure or mutual insurance arrangements. This project will test this hypothesis through two RCTs using a sample of microenterprise owners in Zambia, a population for whom survey evidence suggests this pressure might manifest.

The first RCT will offer wage subsidies to business owners to hire a relative or non-relative either privately or publicly. A disproportionate change in the subsidy required to hire relatives privately as compared to publicly (as compared to the difference for non-relatives) can be interpreted as evidence for the social pressure cost to hire relatives. A second RCT will then seek to understand mechanisms and document the potential productivity losses from hiring relatives. Here, relative and non-relative employees will be randomly assigned to work with employers on a real effort task, over several days. The project will then measure how the match determines, i) verifiable measures of moral hazard (e.g. showing up late to work and absences) and ii) total output. These data will then be correlated with individual measures of social pressure to hire, to understand whether those business owners who hire relatives are those who face higher social pressure costs of hiring.

Understanding the market frictions constraining firms hiring choices has been considered a key policy lever to generate firm growth. Research has explored the possibility that missing information markets are a key constraint, and explored policies designed to alleviate moral hazard, or help potential employees signal their skills to employers. This project, however, might suggest that other missing markets are more important in constraining employment. For instance, if firms operate as a mechanism to provide insurance to kin, then fixing insurance markets may lead to better firm employee matches and enhance firm growth. Moreover, this experiment could help to provide some correlates for firms whose primary binding constraints in hiring are contracting frictions and for those for whom it is kinship pressure, which would allow policymakers to better target policies to each.


Nicholas Swanson

University of California, Berkeley