Academic journal article Contemporary Economic Policy

Predicting the Effects of Changes in Welfare Payments on the Probabilities of Receiving Alternate Sources of Income: The Case of Homeless Persons in Los Angeles

Academic journal article Contemporary Economic Policy

Predicting the Effects of Changes in Welfare Payments on the Probabilities of Receiving Alternate Sources of Income: The Case of Homeless Persons in Los Angeles

Article excerpt


This article investigates the role that changes in welfare payments are likely to have on the earnings behavior of homeless persons. Using a cross-sectional random sample of 1,489 homeless persons in Los Angeles, the author analyzes the marginal effect of reducing public transfers on (a) the probability of earning and (b) the level of income from various traditional and nontraditional sources. This procedure allows the author to control for a number of important factors (including background, human capital, and social network variables) that may also influence the probability of earning income. Findings suggest that reducing government benefit income by $100 increases the probability of receiving income from traditional and nontraditional sources by 1.37% and 2.18% respectively. Among the latter are selling items on the streets and "other" (nonspecified) sources. It is concluded that welfare reform measures may in fact create additional societal costs as former welfare recipients turn to alternative forms of income. (JEL I30, J20, I38)


Recent changes to state and federal welfare programs, including those in Wisconsin and California as well as the historic U.S. Welfare Reform Act of 1996, have reflected a new approach to dealing with welfare benefits. In particular, there no longer exists an entitlement to cash welfare under Title IVA of the Social Security Act (U.S. House of Representatives, 1996). One of the groups most likely to feel the full effects of welfare reforms are the 250,000 to 2 million homeless persons who reside in the United States on any given night (see Burt [1996] for a summary of the "numbers debate"). A question that seems particularly relevant and has heretofore gone unanswered is: What impact might welfare reforms have on income-generating behavior of homeless persons?

Though a popular political emphasis of federal welfare reforms has been to move recipients to work, clearly some are not employable. The new program, which has been turned over to the states in the form of Temporary Assistance for Needy Families block grants, imposes a five-year ceiling on individual eligibility. Thus, the ultimate objective is to reduce lifetime welfare benefits paid out (U.S. House of Representatives, 1996). (How the recipients choose to play the eligibility game--either by continuing in the program until no longer eligible or by "saving" their eligibility as an income insurance--is not important for this analysis.) In any case, the first wave of recipients whose eligibility will expire under the new five-year limit will occur during the latter part of 2001.

Further compounding the effects of changes to the former Aid to Families with Dependent Children (AFDC) program are changes that impose restrictions on benefits from state and local welfare programs, such as General Relief (GR) or General Assistance (GA). In Los Angeles County alone, 6,500 persons lost GR benefits on July 1, 1998, with hundreds more each month expected to lose benefits (Oakes, 1998). Even a new GR "back to work" program implemented in Los Angeles County in December 1998 imposes a number of restrictions, including participation in job training and education, a "good-faith effort to find work" requirement, and limiting GR benefits to just six months. In short, regardless of the current emphasis on work, all recipients may ultimately find their benefits reduced to zero. Further, it may be argued that those who are least likely to feel the effects of welfare reform (i.e., those for whom the five-year time restriction does not bind) are those who were most likely to find a job on their own, witho ut the assistance of the new programs.

Because it has been well documented elsewhere that a large proportion of homeless persons receive public assistance (e.g., 58% in Schoeni and Koegel [1998]), what impact might changes to government programs have on the earnings behavior of this population? …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.