Seminarios

Abstract

This paper investigates how individuals’ choices of housing investments are affected by social norms in their spatial network. I exploit variation from a field experiment that randomly assigns improved housing to slum dwellers in Estado de Mexico, Mexico. The program was randomly assigned at the household-level within slums; hence treatment and control households are co-residents within each slum. As the proximity to treated units varies by household location, the experimental design provides exogenous variation on the average housing quality surrounding the immediate location of slum dwellers, on top of random variation in program delivery. This setting allows me to test whether average treatment effects of the housing program on subsequent housing investments vary with the degree of proximity to treated units, and thus estimate housing spillover effects. Sixteen months after the intervention, I find that untreated households located relatively close to treated units report lower subjective valuation of potential material improvements than their untreated counterparts living farther from treated units, thus failing in their aspirations to improve their housing conditions. Hence, they end up investing less in housing quality. Conversely, treated households located relatively close to other treated units did not fail in their housing aspirations and substantially increased their housing investments compared to their treated counterparts located farther from other treated units. Moreover, while the latter group substituted housing quality investments (better floors, walls, and roofs) with lower access to water and sewerage services, treated households located relatively close to other treated units continued investing in basic utilities. This led to large and significant reductions in diarrhea episodes (53%) for children under five years old, as well as to sizable yet insignificant reductions in the prevalence of respiratory diseases (21%). The evidence shows large spillover effects of housing investments and ties them to health outcomes.