By Simon Franklin, Spatial Economics Research Centre and London School of Economics
This paper looks at the link between housing conditions and household income and labor market participation in South Africa. The author uses four waves of panel data from 2002-2009 on households that were originally living in informal dwellings. He finds that households receiving free government housing later experienced large increases in their incomes. This effect is driven by increased employment rates among female members of these households, rather than other sources of income.
The paper also takes advantage of a natural experiment created by a policy of allocating housing to households living near new housing developments. Using rich spatial data on the rollout of government housing projects, the paper generates geographic instruments to predict selection into receiving housing. It then uses housing projects that were planned and approved, but never actually built, to allay concerns about non-random placement of housing projects.
The fixed effects results are robust to the use of these instruments and placebo tests. The paper presents suggestive evidence that formal housing alleviates the demands of work at home for women, which leads to increases in labor supply to wage-paying jobs.