S ome recent papers have estimated labor supply functions for the working poor. 1 All have a single objective: To estimate the labor force effects of a negative income tax or related income program. The basic argument is simply that the increase in unearned income leads to an income reduction in the labor supply. This, because leisure is a superior good. Also, most negative income tax schemes entail an increased marginal tax rate, that is the same as a reduced marginal wage rate and this, substitution effect, induces a further contraction in labor supply. This paper presents new estimates of labor supply functions based upon a different set of data from those used in other studies. Our estimate of withdrawal effects is smaller than most,2 even though our measure of nonemployment income includes unemployment compensation, public assistance payments and other components of nonemployment income that result from the hours worked decision. Presumably, the inclusion of these endogenous factors would overstate the true income effect. This underscores a result found by both Fleisher-Porter and Greenberg-Kosters, that the estimated income effect is very sensitive to the income definition used.3