Academic journal article Economic Inquiry

Why Many Eligible Individuals Choose Not to Go on Welfare

Academic journal article Economic Inquiry

Why Many Eligible Individuals Choose Not to Go on Welfare

Article excerpt


It has been puzzling to economists why many low-income individuals who are eligible for welfare benefits nevertheless choose not to go on welfare. In the United States, for example, over 20% of eligible individuals choose not to participate in the Aid to Families with Dependent Children (AFDC) and the food stamp program (e.g. Moffitt 1987; Ranney and Kushman 1987; Robins 1990). Many attribute this phenomenon to welfare stigma, which offsets the value of welfare benefits. Among researchers who take this view, Moffitt (1981, 1983) considers both a benefit-dependent and a flat component of the welfare stigma and finds empirical evidence for the latter. These studies are carried forward, mostly empirically by, for example, Ranney and Kushman (1987), Bassi (1990), and Besley and Coate (1992).

Previous work on welfare participation typically uses a one-period static analysis where agents are differentiated by their wages or assets. Little attention has been paid to the possible role of life-cycle concerns in welfare participation decisions, such as on-the-job wage-rising potential and consumption smoothing. In the real world, life-cycle wage growth can result from workers' gains in experience, learning by doing, and on-the-job training, as argued in the empirical study of Brown (1989). (1) The prospect of on-the-job wage-rising potential may motivate young workers with low wages eligible for welfare to continue working for a better future.

Furthermore, when focusing only on one period, the previous static analysis does not capture the restriction on consumption smoothing across life stages of the means testing of welfare programs that aim at a guaranteed income level. (2) This guaranteed level of income sets an upper limit on welfare recipients' levels of consumption in each stage of life. By contrast, nonrecipients can smooth their consumption over their life cycle according to their lifetime earnings, at least via retirement savings if borrowing constraints are present. Due to its restriction on consumption smoothing, the means testing reduces the gain in utility for those who live on welfare at a young age and work at the same time for higher wages later, or who live on welfare only in old age and work in all working ages.

This article offers a new explanation to the puzzling question of why many welfare-eligible agents do not participate. We do this using a life-cycle analysis with realistic wage profiles over lifetime, in the presence of borrowing constraints for both participants and nonparticipants. (3) We show that on-the-job wage-rising potential and consumption smoothing through retirement savings may be the key factors that motivate working and nonparticipation. These factors can substantially reduce the importance of the welfare stigma used in the literature in resolving this puzzling issue, because welfare participation diminishes the wage-rising potential in later life stages, and its restriction on income and assets prevents agents from consumption smoothing over lifetime.

We find that although those with very low earnings and little wage-rising potential are typically welfare recipients, agents with income low enough to make them eligible for welfare but with better wage-rising potential may choose to work, participate only in old age, or never participate at all. Nonparticipation remains the best choice for eligible agents with large wage-rising potential even if stigma-free universal old-age social security is available. These results help explain why young agents with income low enough to make them eligible for welfare may not participate in welfare programs like the AFDC in the United States. We will also apply this model to a comprehensive welfare system in Hong Kong, in which there have been dramatic increases in the level of benefits per recipient in recent years.

While departing from the existing literature on welfare participation by taking into account life-cycle concerns, our model shares some features with previous studies for the sake of comparability. …

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.