Reanalyzes the New Jersey-Pennsylvania Negative Income Tax Experiment and finds a large, statistically significant labor supply withdrawal of white male heads of household. Estimates are markedly higher than in previous analyses for the same group and suggest the President's welfare reform package costs are seriously underestimated. Methodological differences are that participating eligible families are distinguished from nonparticipating eligible families and that this empirical model explicitly controls for state welfare programs for intact households. This experimental labor supply response does not necessarily indicate the size of response to an economy-wide program. The estimates may be unique to the New Jersey-Pennsylvania accounting system, and the response in a permanent program may be quite different from that in a three year experiment. Even so, the small responses estimated earlier shaped policymakers' views about costs and this analysis indicates true costs will probably be much higher. 57 pp. Bibliog.