A considerable body of research has examined the associations between participation in the Food Stamp Program and a range of economic and nutritional outcomes, including labor supply, food spending, nutrient availability, dietary quality, and food security. More recent studies have extended the set of outcomes to include consumption stabilization and other indicators of family and child well-being.
Virtually all of this research focuses on current Food Stamp Program participation. What happens to food stamp recipients after they leave the program or over the long run is largely unknown. Two studies have looked at factors associated with returns to participation after exits, tracking individuals up to 30 months after they left the program, but no research has examined other or longer term outcomes. The research reported here extends the body of work on the economic effects of the Food Stamp Program by examining young mother food stamp recipients for a period after they have their first child and again over a 5- to 15-year period after they reach age 26. This information will be able to distinguish between the shortrun and longrun effects of receiving food stamps.
Such information is critical to understanding the full range of program effects. For example, presumably underlying the Food Stamp Program’s immediate task of achieving a sufficient diet among those at risk of undernutrition is the broader goal of helping people lead healthier and more productive lives. Its success in achieving this goal can be assessed only by examining the longer term outcomes of program participants. Such information may also address the concerns of many who believe that policies intended to alleviate poverty instead intensify economic problems for the poor by making the poor less self-reliant.
In this study, the Panel Study of Income Dynamics (PSID) is used to examine young mothers when they first become mothers for a 4-year period before age 25 and to examine them over 5-year periods after age 25 until they are age 40. For a second sample, the same criteria as the first are used, but the initial period is increased to age 30, with outcome periods from ages 31 to 35 and 36 to 40. Because many of the young mothers also received Aid to Families with Dependent Children (AFDC) or Temporary Assistance for Needy Families (TANF), these programs are examined as well. However, food stamp participation thresholds were/are typically higher than welfare eligibility thresholds. Thus, while the overlap in program participation is substantial, it is not total, which allows an estimation of food stamp effects apart from AFDC effects. The percentage of time spent using food stamps is then examined with no AFDC (food stamps alone group), AFDC with no food stamps, food stamps and AFDC, below 150 percent of the Federal poverty line (FPL) with no food stamps or AFDC, and all others. The reference group included mothers below 150 percent of the FPL with no food stamps or AFDC. This study was mainly concerned with the effects of receiving food stamps alone relative to being in poverty without receiving food stamps.
Four dependent variables are used for the economic outcomes. These include the mother’s family income-to-needs, the percentage of time with income at or below 150 percent of the FPL; the maximum hours of work for either the head of household or the wife; and AFDC, food stamp, and other welfare income (which the study will refer to as transfer income), all measured after the ending period.
For two of the dependent variables, the log of family income-to-needs and hours of work, ordinary least squares regression analysis with robust standard errors is used, clustered by family status for the women while they were growing up. For the percentage of time with income below 150 percent of the FPL and income from food stamps, AFDC, and other welfare income, Tobit models with robust standard errors are used. For all four of these dependent variables, sister fixed-effect (FE) models are also used.
The results indicate that if one were to look at only the entire 10- and 15-year outcome periods, one might conclude that the effects of government assistance are adverse and far reaching. Study data indicate that this conclusion may well be faulty and premature. By looking at 5-year periods after the initial young parenting period, one finds that only the immediately subsequent 5-year period that generally is affected by time on food stamps. If other longer periods after the initial period showed any effects at all, these effects generally diminished as time passed. The results hold when using either standard regression models or sister FE models.
The study was also able to clarify the differences of the effects of assistance to married versus single mothers. While some differences between the two groups were found (such as younger single mothers were more negatively affected by time receiving food stamps than were younger married mothers), the differences were not great for the effects of food stamps alone. Thus, overall, the analysis found that time on food stamps alone had some negative effects on short-term outcomes but found little evidence that time on food stamps had long-term negative consequences relative to time with low income without either food stamps or AFDC.
These results support the notion that food stamps negatively affect some outcomes, but these effects are generally short lived. More specifically, the evidence presented for hours of work shows strong negative effects for joint participation over the course of the outcome period but little evidence that this is the case for time with food stamp participation alone. Other outcomes, such as income-to-needs, time at or below 150 percent of the FPL, and government assistance income, generally show that the effects of using food stamp income alone last only a short time. Further, the study’s FE models show some evidence that this is the case for joint participation as well. For older married mothers, the FE models show some negative income consequences for all outcome periods.
In general, the study does not find that time receiving food stamps improves the economic situation of participants relative to those with low income without transfer assistance. However, because of the generally short-lived nature of the effects of food stamps, the study finds that the long-term impact of time on food stamps is neither positive nor negative for the economic outcomes examined relative to time with low income without government assistance. Thus, if the use of food stamps is leading to positive outcomes in other or noneconomic areas, such as nutrition, consumption stabilization, or physical and emotional well-being, these alone may well justify continued support of the Food Stamp Program.
Direct inquiries about this study to the Project Contact listed above.