Bounding the Labor Supply Responses to a Randomized Welfare Experiment: A Revealed Preference Approach

We study the short-term impact of Connecticut's Jobs First welfare reform experiment on women's labor supply and welfare participation decisions. A nonparametric optimizing model is shown to restrict the set of counterfactual choices compatible with each woman's actual choice. These revealed preference restrictions yield informative bounds on the frequency of several intensive and extensive margin responses to the experiment We find that welfare reform induced many women to work but led some others to reduce their earnings in order to receive assistance. The bounds on this latter "opt-in" effect imply that intensive margin labor supply responses are nontrivial.