Explaining and Forecasting Results of The Self-Sufficiency Project

This paper models the Self-Sufficiency Project (SSP), a controlled randomized experiment concerning welfare. The model of household behavior includes stochastic labor market skill, job opportunities, and value of non-labor market time. All the variation within and between treatment groups, jurisdictions (provinces), demographic groups, and subexperiments is derived from four underlying sources: policy variation, endogenous selection into the experimental samples, the SSP treatments themselves, and different mixtures over 4 underlying types. Using the variation within the treatment group is quantitatively important for identifying the complex model: At the efficient GMM parameter estimates the standard errors of many parameters explode when based on only moments from the control groups. The model tracks the primary moments well except in the entry sample, and it matches out-of-sample outcomes not available for estimation. Predictions of the estimated model are computed for different welfare reform experiments. Counterfactuals suggest the SSP+ treatment has the most potential for generating long-run impacts.