Abstract:
Discrete-choice models provide a tractable method and a simple way to represent utility-maximizing labor supply decisions in the presence of highly nonlinear and possibly non-convex budget constraints. Thus, it is not surprising that they are so extensively used for ex-ante evaluation of tax-benefit reforms (see Van Soest, 1995, Hoynes, 1996, or Blundell et al., 2000, among others). The question asked in this paper is whether it is possible and desirable to get still more flexibility by relaxing some of the usual constraints imposed on household preferences and rationality. By embedding the traditional structural approach in a more general specification, it is first shown that the restrictions on underlying well-behaved leisure-consumption preferences are rejected. It is often the case that structural models gain some flexibility through additional refinements (e.g. costs of work) whose ad hoc interpretation seems unnecessary. We argue instead for the use of a fully flexible model which remains agnostic on the role of the covariates. More fundamentally still, the standard approach - the assumption of unitary households optimizing statically - is strongly rejected when tested against a general model with price- and income- dependent preferences. In a static environment, this result is a strong rejection of the unitary approach. Restrictions from structural and standard models also imply important discrepancies in the simulated predictions of behavioral responses to a tax reform. This results encourage further effort toward more general approaches including dynamic and collective models.