The marginal rate implicit in the Aid to Families with Dependent Children program is the rate at which payments to families decrease as their incomes increase. The rates depend on the provisions of the states' programs regarding the deductions from income which are permitted in calculating payments and the methods which are used to limit payments. The rates estimated here are those facing AFDC families on average at various levels of income. They reflect both these provisions of the states' programs and the number of families at each income level who benefit from these provisions. The rates on earnings and on unearned income are estimated for twentythree states using data collected by a survey of AFDC families in 1971. The estimated rates on earnings are low , rarely higher than 50 per cent. The estimated rates on unearned income are considerably higher, over 80 per cent in half of the states. ^ID to Families with Dependent Children (AFDC) is the nation's largest welfare program, paying $ 6.9 billion in 1972 to an average of 1 1 million people. The program is operated by the states which, as a condition of Federal financial support, are required to conform to Federal legislation and regulations describing administrative procedures, conditions of eligibility, and the structure of benefits. But the states are left with complete control over the level of benefits given to families with no income and substantial control over the amount by which benefits are reduced as income increases. As a result, in 1972 benefits given to a family of four with no income ranged from $60 per month in Mississippi to $375 in Alaska. The amount by which a family's ♦Union College. The research reported here was supported by funds granted to the Institute for Research on Poverty at the University of Wisconsin-Madison by the Office of Economic Opportunity pursuant to the provisions of the Economic Opportunity Act of 1964. The author is solely responsible for the conclusions. She thanks Thad Mirer and Robert Lampman for their very helpful advice. benefits are reduced as its income increases, or the tax on income, also varied considerably. The extent of the variation is unknown, however, due to inadequate information about the rate at which income is taxed in each state.1 One of the most persistent criticisms leveled at the AFDC program is that rates are very high and that welfare recipients therefore have little incentive to work.2 Other critics, in contrast, believe that the rates are very low and that consequently the break-even points, or the inCQmes at which payments fall to zero, are too high.3 Whether either of these criticisms is correct, or whether the truth lies somewhere in between, influences evaluations of the AFDC program. The equity of the program, both vertical and between recipients in the various states, depends in part upon the rate at which the state programs income. Any attempt to measure the effect of AFDC on the incentive to work must obviously require a measure of the rates on earnings. Finally, knowledge about the 1W. Joseph Heffernan has estimated the marginal rates implicit in Vermont's AFDC program in Variation in Negative Tax Rates in Current Public Assistance Programs: Example of Administrative Discretion, Journal of Human Resources, Vol. 8, Supplement, 1973. J. Donald Rowlatt estimated the marginal rates implicit in the welfare program of Alberta, Canada in An Estimate of the Tax Rate in a Public Assistance System, The Canadian Journal of Economics, February 1972, V, no. 1. Their methodology is different from that used here, in part because of differences in the available data. Leonard Hausman has estimated a single average rate on earnings for nine states in Cumulative Tax Rates in Alternative Income Maintenance Systems, Studies in Public Welfare, Paper No. 4, Joint Economic Committee, Congress of the U.S., December 22, 1972. His estimates are based in part on the published aggregates of the data used here but, being based on aggregates, are derived by a different procedure. 2President Nixon, in his message on welfare reform of August 1969, expressed this view. 3This belief was expressed on page 95 of Summary of the Principal Provisions of H.R. 1 as determined by the Committee on Committee on Finance, United States Senate, June 13, 1972.