Strong economy and welfare reforms contribute to drop in food stamp rolls - Welfare Reform and Food Assistance - Brief Article

In fiscal 2000, the Food Stamp Program provided benefits to 17.2 million low-income Americans, a level lower than any year since 1979. Just 6 years earlier, in fiscal 1994, program participation peaked at over 27 million Americans. According to recent studies, the decline in participation was due in part to a strong economy and in part to 1996 welfare reform legislation. This new law, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), made some food stamp participants ineligible and redesigned the cash welfare system in ways that may have further reduced food stamp participation.

The rapid decline in program participation reflects major life changes for millions of low-income individuals and families. Each family or individual that has left the program, or that has not needed to apply in the first place, has its own story. Many of these stories center on good news--a new job or a raise--brought by the unusually strong economic expansion in the second half of the 1990's. Increased earnings lifted incomes for many families and reduced their need for food stamp benefits.

Other stories reflect changes in the Nation's social safety net, especially following major welfare reform legislation enacted in 1996. PRWORA imposed a 3-month time limit on able-bodied adults without dependents to receive food stamps, unless they worked or participated in an approved work-related program at least 20 hours per week, or lived in areas granted waivers because of high unemployment rates or insufficient number of jobs. A limited number of cases were exempted at the State's discretion. According to the General Accounting Office, the number of able-bodied adults without dependents participating in the Food Stamp Program dropped from about 1.1 million in 1996 to fewer than 0.4 million (362,000) in 1999.

PRWORA also made most legal, noncitizen immigrants ineligible for food stamps (illegal or undocumented aliens have always been ineligible). The new law made exceptions for legal, noncitizen immigrants with a substantial work history those admitted as refugees, and those who were U.S. veterans and their dependents. Congress later reinstated eligibility for those elderly disabled, and child-age immigrants who were already in the United States when the welfare reform law was passed. This reinstatement, however, still left most legal immigrants ineligible. Furthermore, recent data indicate that large numbers of eligible U.S.-born children of legal immigrants no longer participate in the program. USDA's Food and Nutrition Service (FNS) estimates that the participation rate in the Food Stamp Program by eligible U.S.-born children of permanent resident aliens dropped from 64 percent in 1996 to 38 percent in 1998, just a year after many of these children's parents and other adult family members became ineligible.

A July 2000 report by USDA's Economic Research Service (ERS) weighed the effects of both a booming economy and changes in welfare programs on Food Stamp Program participation. ERS found that 35 percent of the caseload decline from 1994 to 1998 was associated with new employment growth and reduced unemployment, while less than 5 percent of the decline appeared to be associated with changes in program rules. As is typical for this type of statistical analysis, about half of the caseload decline could not be explained. If some of the "unexplained" caseload decline was due to welfare reform, the effect of PRWORA may be higher than the statistical analysis found.

Food Stamp Participation Fluctuated During the 1980's and 1990's

The Food Stamp Program is the largest Federal food assistance program and a mainstay of the Federal safety net. The program paid out almost $15 billion in food stamp benefits in fiscal 2000, an average monthly benefit of $73 per participant. The maximum benefit is the amount of money needed to purchase a nutritionally adequate diet as defined by the Federal Government's Thrifty Food Plan. The benefits, in the form of either coupons or electronic benefits transfer (EBT) payments, may be used to purchase food and nonalcoholic beverages in authorized stores. The Federal Government then reimburses stores for the value of the food.

To qualify for the program, a household without an aged or disabled member must have gross income less than 130 percent of the Federal poverty level. Effective through September 2001, a family of four must have gross monthly income less than $1,848 to qualify All households must have net incomes (gross income minus certain deductions) less than the poverty level. Finally with some exceptions, the household must meet asset limits of $2,000 for most households or $3,000 for households with a member over age 60.

Most people who receive cash assistance through Temporary Assistance for Needy Families (TANE) also receive food stamps, but it cannot be said that most food stamp participants receive TANF. Seventy-three percent of food stamp households received no assistance from TANF in 1999. TANF serves primarily single-parent families with children (and a smaller number of two-parent families with children), while the Food Stamp Program serves many elderly and disabled people living alone, single adults, and two-parent working families who are not eligible for TANF (see box). Individuals who apply for TANF or Supplemental Security Income (SSI), the major cash assistance programs for people with disabilities, are permitted to simultaneously apply for food stamps, so participation patterns for these assistance programs are somewhat related.

Food stamp participation rose during the recession of the early 1980's and then declined during the middle and late 1980's. Participation rose again to new heights during the recession of the early 1990's, before declining again after 1994. Food stamp participation and the national unemployment rate have followed a roughly parallel track during much of the last 20 years (fig. 1), which suggests that economic conditions have a plausible role in fluctuating food stamp participation.

Food stamp participation and the unemployment rate diverge during some periods, which suggests that factors other than the economy may also affect food stamp participation. In the early 1980's, for example, program participation had already declined by the time unemployment peaked. In the early 1990's, program participation continued to rise for 2 years after the recession ended while unemployment began to fall.

Several major policy changes during the last two decades may have affected the number of people receiving food stamps (fig. 2). The Omnibus Budget Reconciliation Act of 1981 applied new eligibility requirements and lowered some deductions, perhaps reducing program participation in the years that followed. The Food Security Act of 1985 expanded eligibility by increasing the resource limits and designating categorical eligibility to households in which all members participate in either Aid to Families with Dependent Children (AFDC) or SSI. The Hunger Prevention Act of 1988 and the Mickey Leland Childhood Hunger Relief Act of 1993 raised benefits slightly and effected some other modest changes that eased program eligibility restrictions. During the early 1990's, Congress enacted several laws that expanded Medicaid eligibility, which may have indirectly increased food stamp participation: People often apply for Medicaid at the same local office where they apply for food stamps, using the same set of enrollment fo rms.

In the late 1980's and early 1990's, the Federal Government began to allow States more leeway in applying for temporary "waivers" from Federal regulations for administering AFDC, the predecessor to TANF. Waivers allowed States to experiment with such new policies as time limits and work requirements in AFDC, which may have reduced participation in the Food Stamp Program as well: Some people may have left the Food Stamp Program at the same time they left AFDC. even if they continued to be eligible for food stamp benefits.

Major policy changes and dramatic economic growth both occurred at about the same time that program participation fell steeply in the middle and late 1990's. Because it is difficult to pinpoint the cause of the recent decline in food stamp participation on the basis of national trends alone, researchers at the University of Oregon, the University of Florida, and ERS used State-level data to investigate the causes behind the recent caseload declines. By 1998. almost every State experienced declining unemployment and changes in both the Food Stamp Program and cash assistance programs, but the timing of these changes varied from State to State. Using statistical models and State level data, the researchers found that the largest share of the Food Stamp Program caseload decline--35 percent--was related to the strong economy. Changes to the rules for cash assistance programs and the introduction of TANF appear to be associated with a much smaller share of the caseload decline--about 5 percent in one statistical mo del and even less in two other models. (Changes in cash assistance do not include the direct changes to the Food Stamp Program that were made in the 1996 law, such as those that affect immigrants and able-bodied adults without dependents.) The full effect of the 1996 reforms could be somewhat higher due to changes in Food Stamp Program administration and practices that were not picked up by the policy measures used in this study

Other recent studies found similar results. Researchers at Mathematica Policy Research, Inc., under contract to ERS, used a more detailed classification of welfare policy changes (such as time limits and work requirements) to look at how these changes and economic factors influenced Food Stamp Program participation. Mathematica found that economic growth caused about 40 percent of the caseload decline. Detailed policy changes appeared to have little effect--from 0 to 2 percent of the caseload decline--but the timing of another 23 percent of the decline coincided with the implementation of the 1996 welfare reform, without being linked to a specific policy change that the researchers were able to measure.

What Happened to People Who Left the Food Stamp Program?

The experiences of people who left the program also help explain how economic conditions and program changes affected food stamp participation. Studying these food stamp "leavers" is not the same as studying caseloads in general. Caseload changes depend not only on how many people leave the program, but also on how many people enter the program. Nevertheless, two recent studies in Illinois and Arizona sponsored by ERS shed light on what happens to people when they leave the Food Stamp Program. (Two other reports, in Iowa and South Carolina, were not completed at the time of this writing.)

Researchers for the Illinois and Arizona studies used two methods to track the experiences of one-time food stamp participants who left the program in 1997, just as the welfare reform law was being implemented. First, the researchers used information from program administrative records and from Federal Government records on unemployment insurance. These unemployment insurance records report earnings from work, but with some gaps. For example, earnings that are paid in cash may not be reported to the Government. These records, however, still provide useful information about earnings of people who left the Food Stamp Program. Second, in 1999, the researchers surveyed a sample of people who left the Food Stamp Program about their employment situation and their general economic well-being.

The unemployment insurance data showed that household earnings increased fairly rapidly after households left the Food Stamp Program. On average, earnings increased 21 percent in Illinois and 17 percent in Arizona in the first 2 years after leaving the program. These results offer encouraging evidence that many leavers improve their economic situation by working. The proportion of households that are working did not increase rapidly in either State, however, indicating that most of the earnings growth occurred with households that had already been working in some capacity. Moreover, even households with earnings typically did not earn more than the poverty level. As noted in previous Food Stamp Program studies, many leavers return to the program within a year. In Illinois, 40 percent of households that left the Food Stamp Program returned within 12 months (which may be compared with 42 percent who returned within a year in a study using national data from the early 1990's). Some of these program returns are s hort-lived--the fraction of leavers in 1997 who participated in the program 1 year after leaving was only 25 percent in Illinois and 13 percent in Arizona.

In a followup survey interview in Illinois, 55 percent of people who left the program cited employment or increased income as their reason for leaving. Another 13 percent were sanctioned, meaning that their benefits had been cut or eliminated because of failure to follow program requirements. Twelve percent cited administrative-related reasons such as the difficulty of reapplying for program benefits. (The Arizona study also conducted a followup interview but did not ask specifically why the respondent had left the Food Stamp Program.) Like the findings from unemployment insurance data, the survey results suggest that favorable employment conditions are a primary factor in an individual's decision to leave the Food Stamp Program, but they are not the only factor.

Are Growing Numbers of Eligible People Going Unserved?

The decline in Food Stamp Program caseloads raises concerns about those who are eligible for the program but who do not participate. Do people who leave cash assistance programs incorrectly assume they are no longer eligible for food stamps? Or do people simply choose not to participate in the Food Stamp Program because their economic outlook is favorable and they know they would not be eligible for a long period?

The July 2000 ERS report investigated how many people with annual incomes below 130 percent of the poverty line received food stamps in 1998 versus 1994, Having income below 130 percent of the poverty line is one of several requirements for Food Stamp Program eligibility ERS found that 55 percent of the decline in participation from 1994 to 1998 was associated with decreased use of food stamps by individuals in households with incomes less than or equal to 130 percent of the poverty line. Twenty-six percent of the decline was associated with people leaving the Food Stamp Program as their annual incomes rose above 130 percent of the poverty line. (The remaining 19 percent of the decline was associated with decreased use of food stamps by people with annual incomes above 130 percent of poverty-most of whom presumably had incomes below this level for some fraction of the year, making them eligible in some months but not others.)

These participation patterns are corroborated by a recent report from FNS, which used a more elaborate method for measuring the number of people eligible for the Food Stamp Program. The FNS report found that 70.8 percent of people eligible for the Food Stamp Program participated in 1994, but only 59.4 percent of eligible people participated in 1998. The number of eligible people fell from 37.0 million in 1994 to 30.6 million in 1998, but the number of participants fell even faster, so the evidence shows a growing number of eligible people do not participate in the Food Stamp Program.

In July 1999, USDA announced several actions to reduce barriers to participation and make all eligible Americans aware of their eligibility, including a public education campaign, an information hotline, a new toolkit for State and local out-reach efforts, and new rules that allow States to simplify income-reporting requirements for program participants.

ERS is involved in several studies using national and local surveys to investigate the causes of food stamp caseload declines. One study is examining how decisions to participate in the Food Stamp Program are influenced by such factors as customer service at local welfare offices or, perhaps, heightened stigma associated with welfare reform. Understanding the decisions households make about participating in the Food Stamp Program helps predict how caseloads will fluctuate in the future and also helps ensure that the program serves all eligible people who want to participate.

The author is an economist with the Food and Rural Economics Division, Economic Research Service, USDA.

References

Castner, Laura. Trends in FSP Participation Rates: Focus on 1994 to 1998. Washington. DC: Mathematica Policy Research, Inc., November 2000.

Figlio, David, Craig Gundersen, and James Ziliak. "The Effects of the Macroeconomy and Welfare Reform on Food Stamp Caseloads," American Journal of Agricultural Economics, Vol. 82, No. 3, August 2000, pp. 635-641.

Gleason, Phil, Carole Trippe, and Scott Cody. The Effects of Welfare Reform on the Characteristics of the Food Stamp Population. Princeton, NJ: Mathematica Policy Research, Inc., forthcoming.

Gundersen, Craig. "Economic Growth, Welfare Reform, and the Food Stamp Program," FoodReview, Vol. 21, Issue 1, January-April 1998, pp. 23-27.

McConnell, Sheena. The Increase in Food Stamp Participation Between 1989 and 1990: A Report to Congress. Washington, DC: Mathematica Policy Research, Inc., August 1991.

Mills, Gregory, and Robert Kornfeld. Study of Arizona Households Who Leave the Food Stamp Program. Cambridge, MA: Abt Associates Inc., 2000.

Rangarajan, Anu, and Philip Gleason. Food Stamp Leavers in Illinois: How Are They Doing Two Years Later? Princeton, NJ: Mathematica Policy Research, Inc., 2000.

U.S. General Accounting Office. Food Stamp Program: Implementation of the Employment and Training Program for Able-Bodied Adults Without Dependents (GAO-01-391-R). February 2001.

Wilde, Parke, Peggy Cook, Craig Gundersen, Mark Nord, and Laura Tiehen. The Decline in Food Stamp Program Participation in the 1990's, Food Assistance and Nutrition Research Report No. 7. U.S. Department of Agriculture, Economic Research Service, July 2000.

Zedlewski, Sheila, and Sarah Brauner. Declines in Food Stamp and Welfare Participation: Is There a Connection? Discussion Paper 99-13. Washington, DC: The Urban Institute, October 1999.

[Figure 1 omitted]

[Figure 2 omitted]

RELATED ARTICLE: Characteristics of Food Stamp Households in 1994 and 1999

From 1994 to 1999, while the number of Food Stamp Program participants declined, the characteristics of food stamp households also changed. Well over half of all food stamp households contain children, although this proportion declined slightly from 1994 to 1999 (see table). About 20 percent of all food stamp households contained an elderly person in 1999, up 4 percentage points from 1994, The raw number of households containing a disabled person actually grew from 1994 to 1999, even as the total number of food stamp households declined, so the proportion of food stamp households containing a disabled person has grown rapidly (although, as the notes to the table observe, part of the apparent increase is due to a change in the definition of "disabled").

The main sources of cash income for food stamp households have also changed. The proportion of food stamp households that received Aid to Families with Dependent Children (AFDC) or Temporary Assistance for Needy Families (TANF) declined sharply from 38 percent in 1994 to 27 percent in 1999. Meanwhile, this period saw growth in the proportion of food stamp households that received earned income, Supplemental Security Income (SSI), and Social Security. While the Food Stamp Program still supplements the resources available to low-income single-parent households, a larger share of its benefits help low-income two-parent working families, the elderly, and the disabled.

Fewer Food Stamp Participants Receive AFDC/TANF, While More Are Elderly
or Disabled

Household type                     Participating food stamp households
                                          1994        1999        1994
                                         Millions               Percent

Total                                     11.1         7.7       100.0
Containing children                        6.8         4.3        61.1
Containing elderly person(s)               1.8         1.5        15.8
Containing disabled person(s)              1.4         2.0        12.5
Receiving earned income                    2.4         2.1        21.4
Receiving AFDC/TANF                        4.2         2.1        38.1
Receiving SSI                              2.4         2.3        21.4
Receiving Social Security                  2.0         1.9        18.0

Household type                 Participat
                                ingfood
                               stamphouse
                                 holds
                                1999
                                Percent

Total                          100.0
Containing children             55.7
Containing elderly person(s)    20.1
Containing disabled person(s)   26.5
Receiving earned income         26.8
Receiving AFDC/TANF             27.3
Receiving SSI                   30.2
Receiving Social Security       24.7

AFDC = Aid to Families with Dependent Children: TANF = Temporary
Assistance for Needy Families; SSI = Supplemental Security Income.

Notes: The definition of "disabled" was modified in 1995, which caused
the proportion of households defined as disabled in that year to
increase by 5.6 percentage points, from 13.3 percent under the old
definition to 18.9 percent under the new definition. Thus, this
definition change caused some but not most, of the increase in disabled
households from 1994 to 1999 reported here. Percentages do not sum to
100 percent because households may belong to more than one type.

Source: Ross, Randy, and Catherine Palermo. Characteristics of Food
Stamp Households: Fiscal Year 1999 (Advance Report). U.S. Department of
Agriculture, Food and Nutrition Service, July 2000.

COPYRIGHT 2001 U.S. Department of Agriculture
COPYRIGHT 2004 Gale Group