Past Seminars

Host Professor: Francisco Junqueira

Check the paper's abstract

Moving to Opportunity (MTO) is a housing experiment designed to evaluate neighborhood effects on economic and social outcomes of disadvantaged households in the U.S. MTO randomly assigned three types of housing subsiding vouchers that play the role of instrumental variables for neighborhood relocation. Families decide among not relocating, relocating to a low-poverty neighborhood or to a high-poverty neighborhood. This paper investigates the non-parametric identification of the causal effects of neighborhood relocation. This quest belongs to the broad class of policy evaluation problems in which instruments cannot span all the potential pattern of treatment choices. I use revealed preferences analysis to identify causal effects of neighborhood relocation. I also explain the causal content of the widely reported Treatment-on-the-Treated parameter - voucher effect divided by compliance rate - in terms of relocation effects. I draw on the literature of Bayesian Networks to nonparametrically identify that average causal effects of neighborhood relocation. In contrast to most of previous literature, I find significant effects of neighborhood relocation on labor market outcomes.