We discuss the role played by discrete labor supply (leisure consumption) choice in" affecting measures of the welfare cost of labor supply tax distortions. We construct comparable" continuous and discrete choice models, each calibrated to have similar aggregate" (uncompensated) labor supply elasticities. In the former, there is a single representative" consumer; in the latter there is a distribution of individuals across preference parameters. In the" discrete model, taxes induce a large response from a subset of the population of the population shows unchanged behavior. Welfare costs of similar taxes in continuous" models can substantially exceed those in discrete models or vice versa formulation used. Experiments are also reported for a two labor type household model with one" continuous variable (secondary labor) and one discrete variable (primary labor) are also made using an empirically based model specification calibrated to UK data. Model" results clearly show that discrete choice matters in the assessment of the cost of labor supply tax" distortions.