Concentrating on Aid to Families with Dependent Children (ADC), the paper first explores the theoretical implications of the complex economic relationships between family stability, the demand for welfare, and the labor supply of men and women. Then, using 1960 Census data for New York City, regression equations are developed to test empirically: (1) the relationship between the proportion of female-headed households and ADC incidence; (2) the effect of ADC incidence on male and female labor force participation rates. Finally, the results are discussed and conclusions drawn regarding the important policy dilemmas and alternatives which welfare programs pose. Most of the recent research by economists into the relationship between work and welfare has concentrated on General Assistance, the welfare program which appears to present family units with a choice between working and receiving welfare, even though the choice is constrained by requirements that all able-bodied males work or look for work [1, 3, 7, 13]. Yet, by far the largest program of public assistance in the United States is the Aid to Families with Dependent Children (ADC); in 1971, 10.5 million people received aid on this program, while less than one million received General Assistance and another three million received Old Age Assistance, Aid to the Blind, or Aid to the Disabled. Thus, an understanding of the reasons why The author is Assistant Professor of Economics at New York University. * The author is indebted to Professor Jacob Mincer of Columbia University and William Johnson of RAND Washington for help in developing the theoretical and regression models as part of her Ph.D. dissertation at Columbia University. The dissertation findings are summarized in this paper. Funds from New York City's Human Resources Administration and Economic Development Administration and the Ford Foundation made the study possible. The Journal of Human Resources * VIII * Supplement This content downloaded from 157.55.39.104 on Sun, 19 Jun 2016 06:46:15 UTC All use subject to http://about.jstor.org/terms 104 1 THE JOURNAL OF HUMAN RESOURCES families seek ADC is fundamental to explanations of the demand for public assistance in general, before the possible effects of the welfare alternative on work incentives can be evaluated. Indeed, no discussion of the implications of suggested welfare reforms can hope to provide appropriate guidelines if it does not take account of the reasons why ADC caseloads rose so precipitously in the past decade. In earlier work, I examined descriptively the trends in the important factors contributing to family dependence on public assistance in New York City from 1957 to 1967 [6]. This paper reports on the findings of a complementary study of ADC incidence in New York City using Census data by health area in 1960. This data base enabled me to test analytically the more complex relations between family formation and ADC incidence and their subsequent effects on the labor force participation rates of men and of women, which limitations of the time series data precluded. Public assistance is available on the ADC program for all families with children in need and deprived of a male supporter either through death, disability, divorce, desertion, institutionalization, or, in the case of the unmarried mother, when there never has been a legal father. Many of these situations are obviously not those in which any economic choice is involved, but to the extent that some are, specifically those which concern the decision to live together or separately, the family formation choices may be affected by the welfare alternative. In this case, the circumstances under which a family may decide to separate form an integral part of the decision to seek ADC; this paper explores these connections theoretically and empirically. Secondly, the paper examines the subsequent effects on labor supply which the existence and nature of the ADC program and the family formation decisions suggest. It is often argued that there is little connection between employment conditions and the numbers on ADC because families on ADC are, by definition, without a working father. However, such an argument avoids the more fundamental question of the inability of adults to support their families through labor market earnings. The puzzle still remains: Why has there been such a startling increase in the number of families not supported by family members, but instead seeking government assistance? The last section will discuss the conclusions to be drawn from the results and relevant policy implications. THE THEORETICAL MODELS The models developed in this paper serve two purposes: (1) to explore the relationship between family stablity and the demand for ADC, and (2) to trace the subsequent relationship between welfare demand and labor supply. In focusing on the family decision to seek income support from the governThis content downloaded from 157.55.39.104 on Sun, 19 Jun 2016 06:46:15 UTC All use subject to http://about.jstor.org/terms