7. "Title VII Survey" (April 1976)

In this study, the staff of the former Senate Select Committee on Nutrition and Human Needs estimated that, on the average, 30 percent of the participants in the nutrition program for the elderly-authorized by title VII of the Older Americans Act of 1965-also received food stamps.

8. "Toward Efficiency and Effectiveness in the WIC Delivery System" (April 1976) This study was prepared for the Food and Nutrition Service by the Urban Institute and was based on an April 1975 survey of 96 WIC clinics and 3,600 WIC participants. The study showed that about half the sampled WIC recipient households received food stamps. Using program data from February 1977, we inferred from this estimate that about 5 percent of all food stamp households also participated in WIC.

The study also indicated that 31 percent of the WIC households in the survey had one or more children being served either free or reduced-price lunches at school, and 7 percent of the households in the survey had at least one child being served free or reduced-price breakfasts at school.

9. "Characteristics of Food Stamp Households, September 1975" (May 1976)

On the basis of September 1975 data for a nationwide random sample of 10,855 food stamp households living in the continental United States or Hawaii, the Food and Nutrition Service reported that 42 percent of the households surveyed received AFDC income and 17 percent received SSI income. The study, however, did not cover participation in other Federal food assistance programs.

10. "Multiple Welfare Benefits in New York City" (August 1976)

This study, prepared by the Rand Corporation under a grant from the Department of Health, Education, and Welfare (HEW), drew its findings from a 17-percent sample of all AFDC cases on the New York City welfare rolls in December 1974 (42,450 cases surveyed). The study stated that:

The average AFDC case received $6,088 in total annual income from all identifiable sources in 1974. Nearly $3,400 was received from the basic AFDC assistance payment and shelter allowances, $1,600 from medicaid benefits, $500 from food stamp bonuses, $128 in social services, and $460 from nonwelfare income.

66 percent of the AFDC cases studied had been on the rolls continuously for 3 or more years, 26 percent had been on continuously for less than 3 years, and 8 percent were on welfare intermittently during the 3-year period studied.

The receipt of cash and inkind income raised the average AFDC case in New York City over the Federal poverty level. However, 10 to 20 percent of the sampled cases failed to achieve this level.

Data from similar studies in Michigan and California showed that New York City welfare benefits and other income were slightly higher than in Michigan but about in line with those received in California.

One of the study's principal conclusions was that the provision of multiple benefits provided the means for many, but not all, welfare families in New York City to live at an income level above the federally defined poverty level. The study did not cover Federal food assistane programs other than food stamps and AFDC.

11. "Overlapping Food Programs-Alternatives" (December 1976)

This unofficical paper, prepared by Food and Nutrition Service staff and submitted to the Office of Management and Budget and the Congressional Research Service, stated that duplicate or overlapping benefits between the food stamp program and other Federal food assistance programs should be minimized to (1) eliminate double costs to the taxpayer and (2) end the benefit inequities among certain households with equal needs. The paper said there currently were no good official estimates of the amount of overlapping food benefits. But it estimated, using data from the Service's September 1975 survey of the characteristics of food stamp households, that there were 6.8 million school-age children in food stamp households with monthly incomes below the income eligibility limit for free or reduced-price meals under the school lunch program.

The paper listed the following alternatives to remedy the problem of benefit overlaps between food stamps and the other feeding programs administered by the Service (for example, school lunch, school breakfast, summer feeding, child care food, and WIC):

Eliminate the other Service feeding programs that cause the overlap with food stamps.

Eliminate the other Service feeding programs but increase food stamp benefits for certain families.

Count the value of food received under the other Service feeding programs as income for food stamp purposes.

Reduce the food stamp benefits by the value of the additional food consumed as a result of participation in other Service feeding programs.

Offer a choice of programs or mix of benefits for food stamp households that wished to participate in other feeding programs-with a corresponding reduction in food stamp benefits for households that take part in other programs. Eliminate free Service feeding programs by requiring all children to pay something for their meals, but allow children of food stamp households to pay for their meals with food stamps.

Do nothing now, maintain the present system (with its overlaps), and wait for the development of an Administration policy on welfare reform.

STATEMENT OF ROBERT GREENSTEIN, DIRECTOR, PROJECT ON FOOD ASSISTANCE AND POVERTY

I am pleased to have this opportunity to appear before you today. I am Robert Greenstein, Director of the Project on Food Assistance and Poverty. Most recently, I served as Administrator of the Food and Nutrition Service in USDA. The Project on Food Assistance and Poverty is a project of the Field Foundation, a small foundation with a long-standing interest in the problems of hunger and poverty. It was this foundation which sponsored some of the original medical research teams that found hunger and malnutrition in many parts of the U.S. in the late 1960's.

The administration's safety net

Before discussing specific proposals for revision of the food stamp program, I should like briefly to address the question of whether "truly needy" persons have been protected by the Administration's budgetary proposals. The Administration has maintained that the truly needy will not be harmed, and rests it case largely on its claim that seven programs which it says constitute the "social safety net" have been exempted from major cuts. These programs are Social Security, Medicare, veterans programs, SSI, Headstart, summer jobs for youth, and free school meals. We have completed some basic research on this matter. Our findings, based on survey data from the Bureau of the Census and the relevant Federal agencies, are as follows: Of the Federal dollars involved in these seven programs, 79 percent goes to benefit persons with incomes above the poverty line.

Almost 85 percent of the Federal outlays for these seven programs is concentrated in two programs-Social Security and Medicare. These two programs are not meanstested; the large majority of beneficiaries of these programs are not poor. For example, 86 percent of those covered by Medicare are above the poverty line, and over half have income exceeding twice the poverty line.

The third largest of the seven programs are the veterans programs. The majority of persons benefitting from the veterans programs have incomes above the poverty line. Two-thirds of all the beneficiaries of these seven programs are not poor. If free school meals (which do not constitute a safety net by themselves) are excluded, 80 percent of the beneficiaries are not poor.

In contrast, most Americans who are poor benefit little if at all from any of these seven programs. Of the nearly 25 million Americans who are below the poverty line, 16 million persons, or 64 percent, either receive no benefits from any of the seven programs or, at most, receive free school meals. Six million persons-nearly a fourth of those in poverty-do not even receive free school meals. These 6 million are not touched in any way by these seven programs.

These findings make clear that these seven programs do not maintain a safety net under persons in poverty. These programs primarily protect the elderly-most of whom are not poor-and afford little protection to non-elderly persons below the poverty line.

The real "means-tested" safety net and what the administration proposals mean for

it

The real safety net for the poorest people in this country consists of a cluster of largely means-tested programs which, while not inexpensive, are far less expensive than Social Security and Medicare, precisely because they are targeted. These programs are the Aid to Families with Dependent Children (AFDC) program, the Supplemental Security Income (SSI) program, food stamps, Medicaid and various housing programs, low income energy assistance, etc. With the sole exception of SSI, every one of these programs has been targeted for sharp cuts by the Administration.

Many of these cuts in different programs hit the same families. The combined impact of these various cuts on many families far below the poverty line will be

severe.

Current benefit levels

Current benefit levels in these true safety net programs are not generous. There is not a single state in the U.S. where combined AFDC and food stamp benefits equal the poverty line. About 12 states provide benefit levels which are less than 65 percent of poverty, and over half of the states have combined benefit levels which are less than 75 percent of poverty.

In some states, AFDC payments remain remarkably low. In Mississippi, the maximum AFDC payment for a mother and three children who have no other income is $120 a month or $1,440 a year. Even with food stamps, this family still falls below 50 percent of the poverty line.

The situation is similar in a number of other States, especially in the South. The maximum AFDC payment for a family of four is $140 in Texas, $148 a month in Alabama and Tenessee, $158 a month in South Carolina, $188 a month in Arkansas. Nor are food stamp benefits generous either. The average food stamp benefit is now 45 cents per person per meal. A destitute family with zero disposal income receives a maximum food stamp benefit of 64 cents per person per meal.

Many of these families are already significantly worse off than several years ago Although many persons may believe that low-income recipients of public assistance are the group of Americans that have made out best in recent years because of the indexing of benefits, this is not really the case. To be sure, recipients of Social Security and SSI have been protected. But the non-elderly poor have fallen far behind.

AFDC benefits are automatically indexed in very few states. In most states, these benefits fall far behind inflation. A study conducted last year by the Department of Health and Human Services found that during the 1970's AFDC benefits fell about 20 percent when measured in real dollars. The situation is now worsening. In 1980 and so far in 1981, five states instituted across-the-board decreases in AFDC pay

ments.

Data on the income of food stamp recipients shows the same trend. USDA surveys show that between September 1976 and November 1979 the average gross incomes of food stamp participants rose 17 percent while the CPI rose 32 percent. During this same period, per capita income for the nation as a whole rose 40 percent. Current status of the food stamp program

In examining the Administration's budget proposals to determine the impact on poor families, the food stamp cuts are of particular importance. The Food Stamp Program is the only basic assistance program available to all poor families and individuals. AFDC is limited to families with children (and in half the states to single parent families with children). SSI is only for the aged, blind, and disabled. The recipients of the food stamp program are poor. The latest USDA survey data shows that over 50 percent of all food stamp households have gross incomes of less than $300 a month. Nearly 90 percent have gross incomes below the poverty line during the period they are receiving food stamps. 93 percent have liquid assets of less than $500, while over half have no liquid assets at all. Over half also do not own a car or other vehicle. Of the persons receiving food stamps, 78 percent are children, elderly or disabled persons, or single parent heads of households.

The survey data also show that of the nearly 8 million households receiving food stamps, 34 percent contain one or more persons receiving SSI or Social Security. Few students remain on the program since Congress eliminated most of them last year. USDA estimates that students now comprise two-tenths of one percent of all recipients. The small number of remaining students consists of students with children to support (primarily AFDC mothers whose education can help them find jobs and get off welfare and food stamps in future years), elderly and disabled students, and students who work half or full time. Middle income students are ineligible for food stamps and do not receive benefits. None of the Administration proposals saves money by cutting students.

Nor do the Reagan proposals saves much by cutting households with higher incomes-because there are few such households on the program. The Reagan proposal to place a gross income limit at 130 percent of the poverty line saves about 2 percent of program costs (and this is accomplished primarily by eliminating lowincome working families between 130 percent and 150 percent of poverty with high shelter or child care costs, and elderly persons receiving Social Security). According to the Administration's own figures, this proposed reduction accounts for 15 percent of the total savings in the Administration's food stamp proposals.

What this indicates, of course, is that 85 percent of the savings come from households below 130 percent of the poverty line. And in fact, since fewer than 9 percent of the recipients have gross income between 100 percent and 130 percent of poverty, the overwhelming bulk of the savings comes from persons below the poverty line itself. For example nearly 90 percent of the savings from the proposal to reduce food stamp benefits for families with children eligible for school lunches comes from families living below the poverty line.

This reflects, among other things, the changes made in the program in recent years. In the 1977 and 1980 Food Stamp Acts, Congress significantly tightened the eligibility requirements for this program. The USDA data show that the reductions in the income and assets limits made by these laws have reduced the number of persons eligible for food stamps by over 6 million, and have eliminated between 1 and 11⁄2 million actual participants from the program. It is true that participation has expanded, but this is due to increased unemployment and because many poor families who had always been eligible entered the program after the elimination of the purchase requirement.

It is striking that the two main groups impacted by elimination of the purchase requirement (EPR) were the rural poor and the elderly poor. The USDA study "Effects of the 1977 Food Stamp Act" found that the rate of increase following EPR was five times greater in rural areas than in the big cities, and that in absolute terms, the majority of new entrants to the program were rural. The study also found that the rate of increase was 60 percent greater among the elderly than among the non-elderly, and that the new participants had low incomes (with most at the bottom to middle of the food stamp eligibility scale).

Finally, the 1977, 1979, and 1980 Acts contained a number of other tightening provisions. Households owing expensive cars were made ineligible, households voluntarily quitting a job were disqualified for a period of time, persons found to have committed fraud were disqualified, new provisions providing for photo-I.D.'s and computer wage matching were added, states were made subject to sanctions for failing to make sufficient progress in reducing high error rates and rewarded for achieving low error rates, extra Federal funds were provided for fraud investigations and prosecutions and for computerization, and substantial sums were saved by adjusting benefits once a year rather than twice.

Food stamp costs

Food stamp costs have risen in recent years. The number of new participants following elimination of the purchase requirement was somewhat larger than had been expected (particularly in rural areas and among the elderly). In addition, these new participants were poorer than had been expected and qualified for larger benefits. However, these are not the primary causes of the program's increased costs. The principal cause of cost increases is the economy.

USDA estimates that each 1 percent increase in food prices adds $148 million to the cost of the program. Each 1 percent increase in unemployment adds 1.25 million participants to the program at a cost of about $580 million per year.

The Food Stamp program is now estimated to cost about $11 billion in fiscal year 1981. If food price inflation and unemployment had held at the levels officially predicted when the 1977 Act was passed, the cost of the program in fiscal year 1981 would be $7.2 billion, and there would be 3.3 million fewer persons receiving food stamps.

The administration's food stamp proposals

The Administration has announced eight different proposals to cut the Food Stamp Program. Several of these cuts hit some of the poorest families in the country.

The most severe of the Administration's proposals is the proposal to sharply reduce food stamp benefits for families with school children, on the grounds that poor children who receive a school lunch as well as food stamps are receiving excessive benefits. This proposal would reduce food stamp benefits about $12 per month for each low-income child. This is more than 25 percent of the average monthly per person food stamp benefit.

Most of those affected will be single AFDC mothers with children, who are among the poorest food stamp households. In states such as Mississippi, Texas, Alabama and Arkansas, Georgia, South Carolina, and Tennessee, an AFDC mother with three school children would lose $36 a month or nearly 10 percent of her total combined AFDC and food stamp income during the school year. In all of these states, the maximum AFDC payment amounts to only 15-30 percent of the poverty line. (Even if food stamp benefits are added in, combined AFDC and food stamp benefits fall below 60 percent of the poverty line in all these States.) A loss equal to nearly 10 percent of combined AFDC and food stamp benefits will cause serious hardship.

USDA's own data demonstrates the degree to which this proposal affects the poorest families:

The poorest families receiving food stamps are those with gross incomes below 50 percent of the poverty line; 43 percent of all such households would have their benefits cut under this proposal.

More than 2,150,000 households with gross incomes below the poverty line would suffer benefit losses under this proposal. 87 percent of all those affected by this proposal live below poverty line.

The USDA data also indicates that the impact is the starkest in the South, the region with the largest proportion of its population in poverty.

Fifty-five percent of all food stamp households in the East South Central census region (Alabama, Kentucky, Mississippi, and Tennessee) who have gross incomes below 50 percent of the poverty line would have their benefits reduced.

Fifty percent of all households in the West South Central census region (Arkansas, Louisiana, Oklahoma and Texas) who are below 50 percent of the poverty line would suffer benefit cuts.

Forty-six percent of all households in the South census region (Delaware, Florida, Georgia, Maryland, North Carolina, South Carolina, Virginia, West Virginia, and the District of Columbia) who are below 50 percent of the poverty line would face

cuts.

The effect of this proposed cut would be heightened because many of the lowincome families affected by it would simultaneously be hit by cutbacks in other programs on which they depend. Cuts proposed by the Administration in AFDC, Medicaid, housing assistance and low-income energy assistance would affect many of the same families whose food stamp benefits would be reduced under the food stamp/school lunch offset proposal.

It is important to look at the combined impacts of these cuts on poor families with children. The Center for the Study of Welfare Policy of the University of Chicago has examined the combined impact of the food stamp/school lunch proposal and of the Administration proposal to reduce AFDC benefits for those AFDC mothers who work. The chart on the next page which presents the findings of the University of Chicago Center, per selected states, shows that the combined impact of these food stamp and AFDC proposals is to push the average AFDC working mother and her children deeper into poverty.

In addition, there is the fact that AFDC benefits are not adjusted for inflation in most states. The total income loss next year for poor AFDC mothers and children— when measured in terms of "real income"-would be even greater than the chart indicates.

The Administration has defended the food stamp/school lunch offset proposal by contending that food stamps plus school lunches constitute a duplication. This is not a strong argument. As noted, the average food stamp benefit is 45 cents per person per meal, and the maximum benefit for families with no disposal income is 64 cents per meal. These benefits are based on the USDA's Thrifty Food Plan, which is the least expensive food plan the Federal Government has devised. The recent USDA National Household Food Consumption Survey found that 91 percent of all families whose food expenditures were at the level of the maximum food stamp allotment did not receive the Recommended Daily Allowances (RDA's) for the basic nutrients and that two-thirds of the families spending at this level did not even receive 80 percent of the RDA's. The survey also found that most food stamp families spend more than this for food. Since the average gross household income in the Food Stamp Program is but $325 a month (which annualizes to $3,900 a year), this means that families with very limited incomes are digging deeper into these small incomes to provide additional funds for food because food expenditures at the level of the Thrifty Plan do not provide enough food to last the entire month.

Food expenditures at the level of the Thrifty Food Plan simply do not provide three adequate meals a day for an entire month for most low-income families. Even with school lunches available, most poor families on food stamps find that they must spend more on food at home than the Thrifty Food Plan costs. Yet the Administration proposal would give poor families with school children significantly less for food at home than the cost of the Thrifty Food Plan, on the dubious theory that the availability of the school lunch makes the dollar equivalent of the Thrifty Plan an excessive amount to spend on food.

Moreover, food stamp benefits are not actually based on the current cost of the Thrifty Food Plan. Benefits in any calendar year are determined by the cost of the Thrifty Food Plan during September of the previous year. Thus food stamp allotments lag 4 to 15 months behind the actual cost of the Thrifty Plan. If food prices rise at a 10 percent annual rate, food stamp benefits at any point in time are 3 to 13