For helpful comments on earlier drafts, we thank, without implicating, Moses Abramovitz, Yves Balcer, John Bishop, Alan Blinder, Richard Burkhauser, Michael Darby, Irwin Garfinkel, Alan Gustman, Daniel Hamermesh, Martin Holmer, George Jakubson, Robert Lampman, Paul Menchik, Robert Moffitt, Michael Murray, Joseph Quinn, Timothy Smeeding, Eugene Smolensky, Barbara Wolfe and two anonymous referees.

President Nixon's 1969 proposal for a Family Assistance Plan (FAP) established welfare reform as a major social policy goal. In his first year in office, President Carter also placed welfare reform-the Program for Better Jobs and Income (PBJI)-high on his legislative agenda. The FAP and PBJI were both variants of the negative income tax. They shared several common elements, including a national minimum income guarantee, an extension of benefits to persons who were categorically ineligible under existing programs, a concern with maintaining work incentives by keeping marginal benefit reduction rates on earnings well below 100 percent, and a belief that a reformed welfare system could control the growth in costs and case loads. Both also generated fatal political opposition and harsh criticism from welfare analysts who pointed out that the triangle-the tradeoffs among income guarantees, work incentives, and total costs-made their goals mutually inconsistent. President Reagan also placed welfare reform at the center of his social policy agenda in his first year in office. Unlike his predecessors' plans, his reform was successful. By October 1981, a drastic fiscal retrenchment had been proposed, legislated, and implemented in the largest cash welfare program, Aid to Families with Dependent Children (AFDC). The Reagan reform, incorporated in the Omnibus Budget Reconciliation Act of 1981 (OBRA), does not confront the iron triangle of negative income tax plans like FAP or PBJI. It does not attempt to reduce poverty by altering income guarantees or extending eligibility. It does not attempt to encourage work effort by lowering marginal tax rates on recipients. Supply-side logic notwithstanding, it reduces costs and case loads by raising the tax rate on welfare recipients' earnings to 100 percent and by establishing more restrictive gross income limits. President Reagan has reformed welfare by cutting the budget. He has clearly reduced welfare dependency in the short run, as the number of AFDC recipients has declined in most states by between 10 and 15 percent in the last year. A complete evaluation of the long-run effects of the OBRA reforms on the economic well-being and work effort of welfare recipients must await data on behavioral responses that have only recently been induced. Nonetheless, an analysis of the redistributive effects of welfare in recent years can provide a basis for estimating how reduced welfare dependency will affect economic well-being in the short run.

William J. Wilson (1978, 1985, 1986) has hypothesized that the combination of increased spatial concentration and increased inequality of income among blacks has caused adverse behavioral consequences for poor blacks and contributed to the development of an underclass. Lessening segregation and the general rise in black economic well-being in the postwar period enabled middle-income blacks to move out of segregated inner-city neighborhoods. As a result, low-income blacks in these areas now rarely come in contact with middle-class blacks, who had previously influenced social organizations and community institutions, and provided role models of economic and social success. Wilson hypothesizes that poor blacks have changed their labor force and family behaviors because of the social and economic consequences of this selective outmigration. Wilson's hypothesis has both an empirical and a causal component. In this paper we focus primarily on the former. In the first two sections, we review changes in the level and distribution of male earnings, and in the spatial concentration of poverty. The trends for blacks are compared to those for whites. The third section discusses the links between the empirical evidence and the causal component-did these changes lead to behavioral responses that contributed to the development of an underclass?

We analyze the impact of recent immigration on the employment and wages of less educated workers during the 1990s, a period of heightened geographic diffusion of immigrants across the nation. We focus on men residing in metropolitan areas, who are between the ages of 25 and 62 and are from the three major racial/ ethnic groups: white non-Hispanic, black nonHispanic, and Latino (hereafter referred to as race groups). Theory predicts that immigration will increase the wages of native workers who are complements to immigrants and decrease the wages of natives who are substitutes. Because immigrants have low education relative to natives, low-educated natives are likely to be substitutes, and high-educated natives are likely to be complements. We find negative effects of recent immigration on the employment, and especially the wages, of low-skilled workers. The wage effects are largest for Latinos, followed by blacks.

Discussions about the antipoverty effects of economic growth in the United States have largely been predicated on John Kennedy's metaphor that a rising tide lifts all boats. But the magnitude of these effects has been a subject of debate since the inception of the War on Poverty (see Lowell Gallaway, 1965, and Henry Aaron, 1967). This debate has public policy as well as academic implications-the greater the antipoverty effectiveness of growth, the less the need for special programs or income supplements during economic expansions. Elsewhere, we have shown that increased real income need not be associated with a decline in poverty (see our 1984 and 1985 papers). In fact, poverty rates did not fall from 1982 to 1983, even though real median income increased. And in 1984 the official poverty rate was about the same as it was in 1967, while real median family income was 7.1 percent above its 1967 level.' If a rising tide was lifting all boats, the tide was late in many harbors. In this paper we examine the relationship between macroeconomic conditions and poverty. Section I argues that several factors now limit the effectiveness of growth in reducing poverty. Section II differentiates the effects of secular economic growth from those of cyclical recoveries. The next section presents our interpretation of the data, followed by a brief conclusion. We show that growth had a large antipoverty effect through the early 1970's, but that the more recent experience has been different because growth rates have slowed and inequality has increased.

In the September 1975 issue of this Review, Morton Paglin posits a new summary statistic, the Paglin-Gini, which he uses to measure and analyze the level and trend of income inequality in the United States. This measure intends to recognize only income differences among families unrelated to the observed age-income profile as contributing to meaningful income inequality. This is to be contrasted with the standard Lorenz-Gini, in which all differences in income among living units contribute to inequality. Paglin's measure indicates substantially less inequality in the United States than is indicated by the Lorenz-Gini. Moreover, Paglin-inequality falls considerably in the post-World War II period while other measures indicate very little change in inequality over this period. Paglin is addressing an important problem. For the postwar period when the age distribution, the length of time spent in school, and the propensity of the young and the old to form their own households have all changed rapidly, the trend in the conventional Gini coefficient has no obvious normative interpretation. However, it is often given one. Unfortunately, Paglin's proposed index of inequality does not meet its objective and, even if it did, its normative content would be no clearer than that of the standard measure. Indeed, no single indicator is sufficient to capture trends in normatively relevant inequality without further analysis. Like any other time-series, a time-series of inequality must be approached with a well-specified, multivariate model. After describing Paglin's procedure in the next section, we challenge his measure in Section II. In Section III, Paglin's interpretation of policy-relevant inequality is questioned, and in Section IV the implications of the Paglin-Gini for the equity of the transfer system are analyzed. Section V is a summary of our critique.

Sheldon Danziger, Jacques van der Gaag, Michael K. Taussig, Eugene Smolensky, The Direct Measurement of Welfare Levels: How Much Does it Cost to Make Ends Meet?, The Review of Economics and Statistics, Vol. 66, No. 3 (Aug., 1984), pp. 500-505

This paper formulates and estimates a model of the determinants of female household headship. Headship responds to variations in the levels of well-being a woman can expect if she marries or if she heads her own household. We measure the opportunity cost of female headship and the effects of welfare benefits and women's work in the market on female headship and poverty. We find that if welfare benefits were reduced, there would be small reductions in the proportion of women heading households for whites and nonwhites, but a substantial increase in poverty for nonwhites. We also find that wives' work in the market reduces poverty and female headship for nonwhites, and reduces poverty, but increases headship for whites.