The estimation of utility consistent labor supply models by means of simulated scores

We consider a utility consistent static labor supply model with flexible preferences, a non-linear and possibly non-convex budget set, and a wage equation. Three stochastic error terms are introduced to represent respectively optimization and reporting errors, stochastic preferences, and heterogeneity in wages. Coherency conditions on parameters and supports of error distributions are imposed for all observations. The complexity of the model makes it impossible to write down the probability of participation. Hence simulation techniques have to be used in estimation. The properties of the estimation method adopted are first investigated by means of Monte Carlo. After that the model is estimated for Dutch data. We compare our approach with various simpler alternatives proposed in the literature. It turns out that both in the Monte Carlo experiments and for the reqal data the various estimation methods yield very different results. Furthermore estimates are sensitive to the exact specification of the budget constraint.