Time on welfare: why do people enter and leave the system?

American Journal of Economics and Sociology, The,  Jan, 1995  by Peter J. Leahy,  Terry F. Buss,  James M. Quane

• E-mail
• Print
• Link

I

Introduction

Despite federal and state spending on anti-poverty programs, nearly one in seven Americans fell below official poverty levels in 1992. Progress in reducing the incidence of poverty ended during the 1970s, and the incidence was reversed in the early 1980s. The most important reasons for the reversal were: (1) the entry of women into the labor force and the growth of female-headed families; (2) revisions in the measurement of the poverty rate in 1980 and 1982; (3) the persistence of lower earnings among women relative to men and (4) cuts in federal spending on welfare during the Reagan administration (Sawhill, 1988). The last factor has been estimated to account for about 25% of the increase of the poverty rate since 1980.(1)

Although 25 to 30 million people lived below poverty at any time during most of the 1980s, a much smaller number were poor for prolonged periods. While as many as one person in four experiences a period of poverty during any decade (Corcoran et al., 1985); for most it is a brief experience precipitated by divorce, illness, or job loss. Nevertheless, a significant minority, variously estimated at between 4 and 12 million people, remain poor for long periods, sometimes a lifetime (Duncan, 1987; Nathan, 1988; Ruggles and Marton, 1986; Sawhill, 1988). These are labelled the "underclass", "hard-core poor", or the "chronically welfare dependent (GAO, i.e. General Accounting Office, 1990)."

Changes in marital status or family composition explain much of the dynamics of poverty. For example, the largest increase in poverty since the late 1970s has been in families with young children headed by females (Sawhill, 1988). The fall in the average real wage between 1975 and 1985, lower wages for women, and lack of spousal support make single parent female-headed families prone to poverty.

This paper addresses the relationship between welfare dependency and family status, using personal interview data from a mid-western city. The following questions were asked:

1. What are the characteristics of the dependent welfare poor?

2. What factors best explain how long they stay on welfare?

3. What events explain entry and exit from welfare?

4. How do female-headed families differ from two-parent families on welfare and in reasons for entry and exit.

II

Literature Review

Panel studies show that a high proportion of first-time Aid for Families with Dependent Children (AFDC) spells last two years or less. One recent study (Ellwood, 1986), examined data from the Panel Study of Income Dynamics (PSID) for the 15-year period 1968-1983, and found that only 16.7 percent of all welfare recipients lived in families that received welfare eight or more years in their first spell of welfare (see also Hutchinson, 1981). The median spell on welfare was only two years, although because many welfare recipients experienced more than one spell on welfare, the median total time on welfare was four years. Ellwood reports that 40 percent of those who leave a first spell return for a subsequent spell. Thus, welfare "careers" tend to be made up of multiple welfare spells and, when returns are included, welfare careers can be lengthy. Nearly 31 percent of welfare recipients received welfare for eight or more years over the 15 year span. Also using PSID data, Duncan et al (1984) found that 27 percent of non-aged white female welfare recipients received welfare for eight or more years, as did 44% of non-aged (18-62) black female welfare recipients.

Overall, while most people escape welfare relatively quickly, a sizeable proportion of welfare users do not. These are the "dependent poor," and they account for most of the money spent on public assistance. While there is disagreement about the size of this group, reasonable estimates place it at perhaps four to seven percent of the population (Ellwood, 1989). Since this group tends to have multiple problems (poor education, bad health, etc.) and to require multiple services, its members are responsible for a large share of social service expenditures.

Research on welfare dependency associates lack of education or work experience and having children out of wedlock with long-term dependency.(2) Race, taken by itself, is less important when other factors are taken into account (Rank, 1988). A woman who enters the welfare system when young is likely to stay on welfare longer (Bane and Ellwood, 1983; Murray, 1986).

III

Research Methods

Evidence to chart welfare dependency has come from the results of national surveys such as the Panel Study of Income Dynamics, the National Longitudinal Surveys, the Current Population Survey, and, as noted, more recently from the Survey of Income and Program Participation (see Kimeny, 1991, for a critique of these datasets). These studies are nationwide and offer longitudinal records. Panel studies have inevitably found more dynamism in the poverty population than have crossectional studies revealing more people experiencing shorter spells of poverty and fewer people experiencing continuous prolonged spells. (Bane and Ellwood, 1983, 1986; Blank, 1986; Duncan and Hoffman, 1988; Ellwood, 1986, 1987; Hoffman, 1987; Lerman, 1987; Murray and Laren, 1986; O'Neil, et al., 1984).