Brian Jacob

A key goal of many social policies is to improve the human capital and long-term productivity of program participants, not just address their short-term material needs. Little is currently known about such effects from means-tested housing programs, on which the U.S. spends $40 billion a year. This paper examines the causal effects of housing vouchers on children’s outcomes using data from a randomized wait-list lottery conducted in Chicago in 1997. We focus on families who were living unsubsidized in private-market housing when they applied for a voucher, and so our study identifies the effects of expanding the scale of housing programs on the margin. This is a different question from that of Moving to Opportunity, which offered vouchers to public housing families and so identifies the effects of changing the mix of families across programs. Using longitudinal administrative data, we are able to identify behavioral as well as schooling outcomes for up to 14 years after baseline. We find no effects of housing vouchers for girls, and mixed results for boys. Relative to the magnitude of the subsidy, even the modestly positive effects for boys seem quite small. The results suggest housing vouchers are unlikely to be nearly as cost-effective for improving child outcomes as other strategies that directly focus on developing children’s human capital.