Estimating and testing a model of welfare participation: The case of supplementary benefits in Britain

Abstract

Evidence of incomplete participation in welfare programmes is often displayed. This paper analyses how such apparent failures to participate can be explained by the presence of random errors in assessing benefit eligibility, by income underreporting, by benefit misreporting, and by participation costs. We apply the model to the receipts of supplementary benefits in Britain. A bivariate extension of the use of generalized residuals in testing the robustness of limited-dependent-variable models is provided and fails to reject a number of our stochastic assumptions.