the food stamp program: income or food supplementation?

The Food Stamp Program:Income orFood Supplementation?January1977

Congressional Budget OfficeCongress of the United StatesWashington, D.C.

THE FOOD STAMP PROGRAM:INCOME OR FOOD SUPPLEMENTATION?

The Congress of the United StatesCongressional Budget Office

For sale by the Superintendent of Documents, U.S. Government Printing OfficeWashington, D.C. 20402 - Price $1.80

Stock No. 052-070-03870-5

PREFACE

At a time when the Food Stamp Program has come underclose public scrutiny, the 95th Congress will have theopportunity to consider the program's continued authoriza-tion. As the only national program providing assistance toall needy families, this program and future decisions madeabout it will have implications for debates on reform ofother welfare programs.

This report represents the first in the CongressionalBudget Office's series of Budget Issues Papers that analyzevarious social programs. The report reviews the develop-ment of the food stamp program, how it meets its statedobjectives, how it relates to other public assistance pro-grams, and who receives food stamps; in addition, it presentsvarious program alternatives for the future.

The report was prepared by G. William Hoagland of theHuman Resources Division, under the supervision of StanleyWallack and C. William Fischer. The author wishes toacknowledge the assistance of George Iden, John Korbel,June O'Neil, and Robert Reischauer of CBO. Richard Michael,formerly with CBO, also provided helpful comments.

A number of individuals outside CBO provided helpfulsuggestions. The author's special thanks go to MarcBendick, Katherine Bishop, Robert Fersh, Gar Forsht, RobertGreenstein, Stephen Hiemstra, Maurice MacDonald, JanicePeskin, Wendel Primus, Joe Richardson, James Rotherham,Charles Seagraves, James Springfield, Steven Storch, JamesStorey, Alair Townsend, and David Wright. The report wasedited and prepared for publication under the supervisionof Johanna Zacharias; Norma Leake typed the several drafts.

Alice M. RivlinDirector

January 1977

iii

TABLE OF CONTENTS

Page

Preface iii

Summary xiii

Introduction xxiii

Chapter I. Food Stamp Program — Origin andGrowth 1

Chapter II. Benefits from the Food StampProgram 23

Chapter III. Key Structural Elements of theFood Stamp Program 41

Chapter IV. Food Stamp Program Budget Options . . . 59

Appendix . 89

v

TEXT TABLES

Page

1. Three Hypothetical Families and Impact FoodStamp Program Participation Has on FoodPurchases and Income Supplementation 8

2. Food Price Changes, CPI Changes, Food StampAllotment, Food Stamp Income Eligibility,and Poverty Levels (1971 First Half to 1976First Half) 13

3. Profile of Food Stamp and Total U.S.Households 25

4. Sources of Income for Food Stamp Households . . 26

5. Distribution of Gross Monthly Income forFood Stamp Households and for allUnited States 27

6. Distribution of Food Stamp Persons by PovertyLevel (Pre-Food Stamp Bonus Transfer), Sexof Head of Household, Age of Head of House-hold, Household Size 28

7. Average Monthly Food Stamp Benefits forRecipients by Poverty Status and Sex ofHead of Household 30

8. Distribution of Food Stamp Recipients Pre-Food-Stamp-Transfer Poor and Post-Food-Stamp-Transfer Poor, by Household Characteristics. . 32

9. Impact of Bonus Food Stamps on Narrowing AFDCPayment Differentials Between Regions . . . . 33

10. Average Deduction by Household Size 47

Vll

Average Total Deduction by Poverty Statusand by Region 49

12. Itemized Deductions Per Household, UnitedStates and Regional Averages 50

13. Cost of Food at Home Estimated for FoodPlans at Four Cost Levels, July 1976 . . . . 54

14. Average Amount and Percentage of Gross andNet Monthly Income Spent on Food Stampsby All Program Households in September1975, by Type of Income and Gross MonthlyIncome 56

15. CBO Economic Assumptions, Consumer Price Indexand Food Sector Consumer Price Index . . . . 61

16. Summary, Food Stamp Program Outlays, CurrentPolicy Estimates for Fiscal Years 1977to 1982 63

17. Estimated Participation Rates for Food StampHouseholds, Simulated for April 1977 . . . . 78

18. Policy Criteria for Evaluating Major FoodStamp Program Alternatives 84

Appendix Table: Total Federal, State, andLocal Expenditures for Federal Food StampProgram, Fiscal Years 1969-1977 90

IX

FIGURES

1. Comparison of 1969 and 197.1 Food Stamp IssuanceSchedules for Four-Person Households . . . .

2. Family Food Assistance Participants andFood Stamp Participants 11

3. Family Food Assistance - Public AssistanceParticipants and Food Stamp - PublicAssistance Participants 15

4. Family Food Assistance - Nonpublic AssistanceParticipants and Food Stamp - NonpublicAssistance Participants 16

5. Total Federal and Non-Federal Cost(Obligations) of Food Stamp Program 20

6. Income Supplementation and Food ConsumptionImpact, Food Stamp Households, Age of HeadLess Than 65, 1973-1974 Consumer ExpenditureSurvey 38

7. Comparison Food Stamp Program - Current Program- Proposed Food Regulations - S.3136 FoodCoupon Allotment - Net Income Eligibility- Purchase Requirement Household of 4;48 States and D.C. - July 1976 67

XI

SUMMARY

The Food Stamp Program CFSP) has grown dramatically inrecent years. Today the program provides assistance to house-holds nationwide, including Guam, Puerto Rico, and the VirginIslands. Federal outlays in the program have grown fromless than $100 million in fiscal year 1965, serving fewerthan a million persons, to nearly $5.5 billion in fiscal year1977, serving over 17 million persons. Even though improvedeconomic conditions and stabilized food prices have recentlyslowed program growth, the program still is expected to costthe federal government about $5.4 billion in fiscal year1978.

The recent rapid growth of the food stamp program,charges of widespread irregularities and abuse, fraud andmismanagement, and allegations of high income recipientshave all generated increased public awareness and concernabout the program, its efficacy, its relationship to otherpublic welfare programs, and its future role in providingassistance to needy Americans. The program's legislativeauthorization expires with the 1977 fiscal year.

The food stamp program is designed to subsidize low-incomehouseholds in their purchases of food. Any household meetingthe basic eligibility requirements—an income test, an assettest, and a work requirement—may purchase food stamps at aprice below their market value. Stamps are redeemable forfood purchases at certified retail and wholesale food outlets.The difference between the market value (the allotment) andthe amount paid for the stamps (the purchase requirement)represents the federal transfer of in-kind benefits (referredto as bonus stamps). The federal government bears the entirecost of the bonus food stamps. State agencies assume respon-sibility for certifying eligible households and issuing stamps.All state administrative costs are shared with the federalgovernment on a 50-50 basis.

The basic philosophy behind the food stamp program issimply that every eligible household should have "an oppor-tunity to obtain a nutritionally adequate diet." Evolvingfrom earlier commodity distribution programs, the food stamp

Xlll

program has attempted to achieve the goal more effectivelyand efficiently and with increased food purchasing freedomfor the participant, Today, the food stamp program goes be-yond providing recipients the wherewithal to increase foodconsumption and, on the average, for every one dollar of bonusfood stamps transferred, approximately 43 cents is freed fornonfood purchases.

One of the key budgetary issues is whether the foodstamp program should be redirected to emphasize either of itstwo current goals of (a) increased food consumption or (b) in-come supplementation, or whether the current mix of emphasisshould be continued. Should the federal transfer be directedonly for food purchases? If not, how much general income sup-plementation should be allowed? If general income supplemen-tation is the goal, should the program be replaced with anunrestricted cash grant program? Most recent proposals havecalled for incremental modifications to the current programstructure.

The design of the program partially accounts for itsrecent growth. Administrative and legislative adjustmentsin 1969-1970 lowered the purchase requirements, set nationaleligibility income limits, and raised the basic allotment.Amendments in 1973 replaced the commodity distribution pro-gram with the food stamp program nationwide and in outlyingterritories. Finally, program growth resulted from the un-precedented world-wide food price inflation beginning in 1972,leading to both increased allotment levels and income eligi-bility standards. These events all coincided with unprece-dented unemployment rates and depressed family incomes.

The food stamp household had an average gross cash incomeof about $298 in September 1975, $3,576 on an annual basis,23 percent of the mean family income of all American familiesin 1975. Over 42 percent of the participating households re-ceived income from the Aid to Families with Dependent Children(AFDC) program, 17 percent had incomes from the Social Secu-rity Insurance program, and about 22 percent of the partici-pating households received some income from work. Of allthe food stamp benefits transferred, it is estimated that57 percent went to households that were receiving some otherform of public assistance, and 19 percent went to householdsclassified as working poor. This represents essentiallythe only federal income assistance to aid working poor house-holds.

xiv

The bonus transfer tends to even out some of the stateand regional differentials produced by varying state AFDCbenefit levels. It is estimated that the widest differentialin average AFDC benefits between the New England and Southernstates, a differential of almost $165 before food stamps, isreduced by over $50 after food stamp participation.

Of all households certified eligible for food stamps inSeptember 1975, 78 percent fell below the official ExecutiveBranch poverty threshold, _!/ and 90 percent fell below 125percent of this poverty line, Because bonus stamps aregreater for the lowest income households, almost 86 percent ofall the benefits went to households below poverty, and 95 per-cent went to households below 125 percent of the poverty line.Households with incomes above 125 percent of the poverty linereceived 5 percent of the total benefits. The average foodstamp household received approximately $78.83 in bonus foodstamps in September 1975, or an average of $948 in totalfederal transfers annually. On a per-person, per-meal basis,this represents an average transfer of about 25 cents.

Of all households estimated to be eligible for food stampsin the United States, on a monthly basis participation aver-ages about 55 percent, or 5.9 million households. This 55percent receives approximately 63 percent of the total poten-tial benefits. While the program was never specifically de-signed to remove participating households from poverty, theinclusion of the food stamp bonus benefit in a household'sincome definition does move over one-quarter of the pre-foodstamp poor households out of poverty. Assuming that a dollarin food stamps has the same value as cash to the recipient,currently it is estimated that nearly 4 million of the 14.4million pretransfer poor persons are moved out of poverty asa result of counting the food stamp benefit in their income.

The nutritional improvement that results from food stampparticipation is less clear. While it is estimated that, in

_!/ The official poverty definition is based on an incomethreshold that varies with family size, farm or non-farmresidence, and sex of head of household. For example,in 1975, for a non-farm family of four, $5,500 is thelevel below which that family is deemed to be in poverty.

xv

1973 on the average, food stamp households increased theiraverage food consumption by as much as $40,00 per monthover what they normally would have spent for food, increasedfood expenditures and nutritional improvements are not synon-ymous. Little evidence exists nationally concerning theeffect the program has had on raising the levels of nutri-tion among low-income households.

Basically five future budget options exist for the foodstamp program. These are summarized along with potentialbudget and recipient impacts:

Current Policy Status Quo. Since participation and costsof the food stamp program are sensitive to economicchanges, the recent trend of rapid growth in number ofrecipients and program costs is not inevitable. Basedon the increases in food prices relative to overallprice changes, and allowance of particular deductions,households could continue to be eligible for food stampswith incomes above the official poverty line. Federalcosts would be approximately $5.4 billion in fiscal year1978, increasing to $5.9 billion in fiscal year 1982.

Recent Legislative Reform Options. Separate bills ap-proved by the Senate and by the House Agriculture Com-mittee during the 94th Congress centered on modifyingprogram parameters such as income definitions, incomeeligibility limits, deductions, and purchase require-ments. Similar adjustments were included in regulationsproposed by the Ford Administration. Incremental reformproposals replace itemized deductions with a fixedstandard deduction. These proposals also retain the foodstamp purchase requirement, setting it at a fixed per-centage of income after deductions. Such proposalsusually result in about the same level of income supple-mentation as the current program. Because of relativelyhigh standard deductions and special deductions, benefitstend to be concentrated on the lowest-income groups.Such proposals also redistribute the proportion of foodstamp benefits away from northeastern to southern andmidwestern states. By setting a maximum income eligi-bility standard, they also eliminate participation byapproximately one million higher income recipients.

While these programmatic reforms are substantial, nosignificant reduction in federal outlays would occur.

xvi

No significant cost savings would result from removinghigh income recipients, students, and strikers, Thebill approved by the Senate (S. 3136) would have in-creased costs by $600 million in fiscal years 1978 and1982. The bill approved by the House Agriculture Com-mittee (H.R. 13613) would have resulted in minor reduc-tions in federal outlays in the short run, withadditional outlays of about $100 million in fiscal year1982; it would also havproximately $100 millioi.

Proposed Ford Administrreduced program costs bfiscal year 1978 andby eliminating over 3 mmajor costs reduction fextended accounting perincome was to be countedSuch a provision wouldon longer term needy hoprogram's current

tion regulations would haveapproximately $500 million in

)0 million in fiscal year 1982,llion current recipients. Theature of this proposal was an_od (three months) over whichfor determining eligibility,

result in concentrating benefitsiseholds, while reducing the

to immediate food needsresponsiveness

Food Consumption Emphasprogram costs could beto households below povof the bonus transferitems. Such proposalsrequirement from approxin the current programother in-kind food benedefinition, such as spechild nutrition programProgram (Women, Infantsbenefits in the elderlyreduce federal costs.

At least two proposalsin the 94th Congress:increased the food allo20 percent, Such propo3 and 4 million currenthouseholds below the pofederal in-kind benefitProgram costs in fiscalapproximately $800under H.R. 13375.

81-390 O - 77 - 2

increased state costs by ap-

is. Significant reduction inichieved by limiting participationrty and altering the proportion

which can be spent for non-foodvould increase the purchaseLmately 19 percent of gross incomebo 30 percent. The inclusion offits in the household's incomeial assistance payments in the, benefits received in the WICand Needy Children) and federalfeeding program, would further

along these lines were introduced1993 and H.R. 8145. S. 1993

tment for participants by overals would also remove betweenrecipients and limit benefits toverty line after including others in their definition of income.year 1978 would be reduced by

million under S. 1993 and $2.7 billion

•cvn

Income Support Emphasis Through Elimination of thePurchase Requirement, Such a proposal would removethe requirement that households purchase food stamps,Eligible households would receive only bonus foodstamps. Approximately 40 percent of all purchasedstamps in circulation would be removed from existence.Minor reductions in program costs associated with print-ing, handling and storing of purchased stamps wouldresult.

Eliminating the purchase requirement would remove onedisincentive for nonparticipation; it would also removeany programmatic incentive for increased food consump-tion over current levels. The proposal implicitlyalters the current program's objective of targeting thetransfer on food comsumption to one of more generalincome augmentation. For some households qualifyingfor the largest bonuses, the proposal would continue tosupport a high level of food consumption; for householdscurrently receiving small bonuses, greater choiceswould exist in their consumption patterns.

Increased participation would be accompanied by increasedcosts. Precise estimates of both increased participationand costs are difficult to make and rest on untestedchanges in the eligible population's behavior and valua-tion of the bonus stamps. Current estimates show thatparticipation could increase between 10 and 20 percentover present levels.

One bill introduced in the 94th Congress, S. 2451,represents a prototype of a program operating withoutthe purchase requirement. The proposal would have re-sulted in increased costs of $1.1 billion in fiscal year1978, increasing to possibly $1.8 billion in fiscal year1982, over current policy levels. Increased benefitswould have been concentrated on smaller size, more elderlyhouseholds. Proposals such as S. 2451, while retainingthe no-purchase requirement, must be accompanied bymodification in other program parameters such as lowerlevels of deduction and longer accounting periods inorder to limit costs.

Cashing Out Food Stamps. A fourth budget alternativewould involve replacing the bonus food stamps with anequivalent amount of cash. The House Agriculture

XVlll

Committee bill includes a cash-out of bonus stamps forcertain elderly, blind, and disabled persons. The cash-out option is not normally considered a food stampoption but should be considered in broader debatesassociated with various comprehensive welfare restruc-turing proposals, The proposal represents the finalalternative on the continuum from an in-kind welfareprogram designed to increase the consumption of aparticular good to an unrestricted cash assistance pro-gram with maximum freedom of choice.

Households receiving cash as opposed to bonus food stampswould use a portion of the grant for food purchases.What limited evidence exists does not suggest thathouseholds receiving cash have any better or worsequality of diets than households that receive foodstamps. Given that recipients value the current in-kindtransfers at less than government costs, cashed-outbenefit levels could be set below current levels of foodstamp assistance. Cashing-out food stamps would reduceadministrative complexities through consolidation withother cash-assistance programs.

If the basic guarantee and benefit reduction ratesremained the same under this option as in all others,and if the current prospective accounting system re-mained unchanged, it is estimated that costs wouldincrease between $2,2 and $2.7 billion over current policyestimates for fiscal years 1978 and 1982 respectively.Participation rates would nearly double under this pro-posal, but new recipients would qualify for lower averagebenefits; benefits would accrue to households in thehigher eligible income groups.

Summary Table 1 provides a generalized evaluation of theproposals compared to the current program.

On balance, the food stamp program issues that willface Congress in its deliberation on the fiscal year 1978budget will run the spectrum of proposals designed to tightlydefine how the taxpayers' transferred dollars are to be spentto more open-ended proposals providing maximum recipientchoice. Such a spectrum also covers proposals that wouldreduce federal outlays by as much as $2.7 billion or increaseoutlays by a similar amount. The choices and decisions

XIX

Congress makes will help shape not only the future directionof the food stamp program, but also the design and structureof any major reformulation of the country's total welfaresystem,

IMPACT OF FOOD STAMP PROGRAM BUDGET ALTERNATIVES ON PROGRAM GOALSAND EVALUATIVE POLICY CRITERIA RELATIVE TO CURRENT PROGRAM.*/

E l i g i b l e A v e r a g eProgram Partici- Benefits to Food-IncomeAlternatives pants Participants Tradeoff

I. Program Refinements Slight decrease in1. S. 3136, Senate food consumption,passed Reform Act. Increase Increase slight increase in

income supplementation

2. H.R. 13613, HouseAgriculture CommitteePassed Bill Increase Increase Unchanged

3. Administration Reduction Unchanged UnchangedRegulations

•II. Food ConsumptionSpecific Increase food con-1. S. 1993, Buckley sumption, decrease2. H.R. 8145, Michel Reduction Increase income supplementation

III. Elimination PurchaseRequirement1. S. 2451, Dole-McGovern Decrease food con-2. H.R. 10467, Heinz sumption, increase3. H.R. 10441, Hall Increase Increase income supplementation

IV. Cash-Out Bonus Increase Unchanged Decrease food con-sumption, increaseincome supplementation

XX

Distribution of Benefits VJorkIncentives

Budget CostsE./AdministrativeSimplicity FY 78 FY 82

(Billion $)Increase poverty reduction andcoverage, reduces benefits toworking poor, redirects benefits Someto larger size households. In- Increasecreases benefits to elderly,blind, and disabled.

Increase in poverty reduction,concentration on public assist-ance, elderly, and smaller size Somehouseholds redirects benefits Increaseaway from north and easternstates to southern.

Unchanged poverty reduction;concentration on public assis- Unchangedtance, elderly households, re-duces benefits to working poor.

Increased benefits for largersize households. Concentrationon elderly, public assistance; Decreaseunchanged poverty reduction.

Simplified $6.0 6.5(+.6) (+.6)

IncreasedComplexities 5.3 6.0

(0) (+.1)


(-.5) (-.8)


(-.8) (-.8)

Increased poverty reduction.Benefits distributed tohigher income eligible house- Unchanged Simplified 6.5 7.7holds, concentration on elderly, (+1.1) (+1.8)smaller size households.

Increased poverty reduction.Benefits distributed to higher Unchanged Simplified 7.6 8.6income eligible households. (+2.2) (+2.7)

a/ Consult Table 18 for a more specific review of the elements of this table.

b_/ Change from current policy estimates of $5.4 billion in fiscal year 1978and $5.9 billion in fiscal year 1982 shown in parentheses.

xx i

INTRODUCTION

Over the last decade, the Food Stamp Program (FSP)has undergone a dramatic increase in size. Total federaland state outlays in the program have grown from less than$100 million in 1965 to nearly $5.9 billion in fiscal year1976. In fiscal year 1977, the program is expected to costthe federal government slightly over $5.5 billion and toprovide assistance to over 17.4 million persons everymonth,

The rapid growth and increased visibility of foodstamp use have greatly increased public awareness of andconcern about the program's operation, its efficacy,its relationship to other public welfare programs, andits future role in providing assistance to needy Americans.Among other things, it has not been clear to the generalpublic whether the food stamp program is intended to pro-vide food for, or to supplement the income of, low incomefamilies. As the program actually operates, it does somemeasure of both.

The food stamp program subsidizes eligible householdsin buying food through regular retail stores. Any house-hold meeting the basic eligibility requirements—an incometest, an asset test, and a work requirement—may purchasefood stamps at a price below their market value. The dif-ference between the market value of the stamps (the allot-ment) and the amount paid for the stamps (the purchaserequirement), which varies by household size and net monthlyincome, represents the federal transfer (referred to asbonus food stamps). For most families under the program,the stamps provide the capacity to buy more food than theyotherwise would have been able to buy. And for many families,the stamps also free some income that would have been spenton food for other purposes. Thus, the program supplementsboth food and income.

Beginning in early 1975, a number of problems promptedreexamination of the food stamp program. Costs had sky-rocketed. There were also charges of widespread irregu-larities, abuse, fraud, and mismanagement, plus allegationsof food stamp issuance to "high income" recipients. All

xxiii

these factors prompted both the Legislative and Executivebranches to propose major modifications in program admin-istration, operation, and eligibility criteria, InFebruary of 1976, President Ford proposed administrativeregulations designed to reduce program costs. However,implementation of the regulations was blocked by the UnitedStates Court for the District of Columbia.

In April of 1976, the Senate passed the NationalFood Stamp Reform Act of 1976 (S. 3136). The HouseAgriculture Committee approved a bill (H.R. 13613) in lateAugust of 1976, but no action was taken by the full House.Thus, no food stamp bill was enacted by the 94th Congress.However, because legislative authorization for the foodstamp program expires in 1977, the program will expireunless the 95th Congress acts to renew it.

Much of the legislative debate in the 94th Congressfocused on methods of reducing federal costs. The impactof proposed modifications on current recipients and onthose eligible, but not participating, has been only asecondary consideration. Furthermore, little considera-tion has been given to the food stamp program's relation-ship to other welfare programs and to how program modifi-cations would affect this relationship.

The opening chapter of this paper outlines the historyof the food stamp program, particularly recent programgrowth. The second chapter describes the population servedby the food stamp program and the nature of the benefitsof the program. The third chapter explains the key struc-tural elements of the food stamp program. The finalchapter then analyzes five different potential legislativeapproaches to the food stamp program, examining theireffects on eligibility criteria, on program participants,on the program's primary objectives, and on the federalbudget.

xxiv

CHAPTER I FOOD STAMP PROGRAMS-ORIGINS AND GROWTH

Federal domestic food assistance programs go back somefive decades, but both their character and scope have changedmarkedly in the last five years, presenting major budgetissues for fiscal year 1978 and beyond: Should the foodstamp program emphasize food consumption or income supple-mentation, or should it continue to be a mix of both? Canits costs be kept under control?

The purpose of this opening chapter is to provide thereader with a brief account of the evolution of the foodstamp program, and particularly with an explanation of itsrapid change and growth in the last few years. An under-standing of this history is essential to a clear understand-ing of both food stamp budget issues and general welfareissues to be resolved in the years to come.

FOOD STAMP PROGRAMS UNTIL 1971

Commodity Distribution in the Depression

The farm depression of the 1920s and the general economicdepression of the 1930s led to the passage of the AgriculturalAdjustment Act of 1933, The primary objective of the Actwas to support farm prices and stabilize farm income by con-trolling agricultural production and removing excess suppliesfrom the market. Amendments to the Act in 1935 created theFederal Surplus Relief Corporation for the specific purposeof purchasing surplus farm products from the open market. I./

The 1935 amendments specified that once the federalgovernment had purchased price-depressing surplus foods, itwas to encourage "the domestic consumption of such commoditiesor products by diverting them . . . from the normal channelsof trade and commerce." Needy families and school lunch programs

_!/ Public Law 320, 74th Congress, August 1935. The Corporationwas funded by setting aside 30 percent of the nation's customrl n -f" -t <"» e-tduties.

became major outlets for commodities purchased under thislegislation. State and local governments established largedistribution systems to accept, store, and move surplusfoods to eligible families, schools, and charitable institu-tions. Nevertheless, it was only as an adjunct to agricul-tural programs designed to support farm income that thefederal government began to provide a part of basic foodneeds.

Early Food Stamp Plans

The first food stamp program grew out of the specificneed to increase the consumption of surplus agricultural com-modities. For 46 months between May 1939 and early 1943, afood stamp plan served approximately 4 million persons inselected areas of the country. The program never operatednationwide and was limited to families on relief, WPA workers,and other needy persons certified by certain relief agencies.The program was terminated when wartime mobilization reducedunemployment and increased the demand for U.S. food supplies.In fact, many forms of food were in such short supply thatthey were rationed.

Pilot Programs of the 1960s

The food stamp program as it exists today began on anexperimental basis in early 1961. President Kennedy's firstEconomic Report to Congress proposed a food stamp program forselected depressed areas of the country. The initial numberof pilot areas, eight, was later increased to 43, and partici-pation grew from 133,400 people in July 1961 to 392,400 inMarch 1964. The federal cost of the program reached $13million in fiscal 1962 and more than doubled to $29 millionin fiscal 1964. The experimental program, like the Agricul-tural Adjustment Act of 1933, removed price-depressing surplusgoods from the market, but it also emphasized the objectiveof improved nutrition.

The pilot food stamp program of 1961 closely followedthe design of the 1939-1943 program. Under the pilot pro-gram, certified families (for the most part families onpublic welfare rolls) exchanged "normal food expenditures"for food stamps of higher monetary value. Participatingfamilies were allowed to use all the stamps on whatever

food products they desired, except for certain importedfoods.

Studies of the pilot program indicated that between 85and 95 cents of every bonus dollar were spent to increasefood purchases; therefore, between 5 and 15 cents of thebonus dollar were used to supplement nonfood purchases. 2_/The early results also showed increased food purchases inspecific food groups—meats, poultry, dairy products, andfruits and vegetables. These food groups were considerednon-basic, perishable food items that were not truly "insurplus." Despite the program's inability to rid the marketof surplus commodities, its marked ability to increase thefood purchasing power of participating low-income familieswas cited as a benefit by proponents of the Food Stamp Act of1964.

Food Stamp Act of 1964

The Food Stamp Act of 1964 declared it to be a policy ofCongress "to raise the levels of nutrition among low-incomehouseholds" (emphasis added). The federal government estab-lished the value of the food stamp allotment in such an amount"as [would] provide such households with an opportunity morenearly to obtain a low-cost, nutritionally adequate diet."

The food stamp program established by the Act is an in-come-test program designed to subsidize eligible householdsin buying food through regular retail stores. Any householdmeeting the basic eligibility requirements—an income test, anasset test, and a work requirement—may purchase food stampsat a price below their market value. The difference betweenthe market value of the stamps (the allotment) and the amountpaid for the stamps (the purchase requirement), which variesaccording to household size and monthly income, represents thefederal transfer, also referred to as bonus food stamps.

The program under the Act was basically state-oriented.The Act placed the responsibility for establishing the foodstamp program in state agencies authorized to administer

2_/ "The Food Stamp Program, An Initial Evaluation of the PilotProjects," United States Department of Agriculture, Con-sumer and Marketing Service, C&MSI, April 1962.

assistance programs. The federal government established thepurchase requirements for the full stamp allotment and borethe cost of the bonus stamps, but individual states specifiedeligibility standards. Each state set monthly allowableincome cut-off levels by specific household sizes. Statesalso set resource and asset tests, usually consistent withthe state's public assistance criteria (e.g., the criteriaused in the Aid to Families with Dependent Children, AFDCprogram). Most important, eligibility was established primar-ily on income, making eligible those low-income, intact house-holds not receiving welfare.

Having states establish income and asset criteria andthe federal government establish national benefit levels pro-duced a number of problems. Most significant were widelyvarying eligibility standards and low participation rates insome food stamp areas. Low participation, inequitable allot-ment, and prohibitive purchase requirements precipitated thefirst major administrative program modifications in December1969, These were followed by legislative amendments inJanuary 1971,

Amendments of 1971

Shortly after taking office in 1969, President Nixonissued a message on hunger in which he declared: "The momentis at hand to put an -end to hunger in America itself for alltime." The message combined a series of recommendations, themost significant of which were:

1. Preference for the food stamp program over thedirect distribution of surplus commodities, andpriority for the expansion of the food stamp pro-gram in all areas of the country;

2. The provision of food stamps at no cost to thosein the very lowest income brackets;

3. The provision of food stamps to all others at acost no greater than 30 percent of income; and

4. The provision to families of enough food stamps topurchase a nutritionally adequate diet, later to bedefined by administration regulations as an "EconomyFood Plan."

With the opening of the White House Conference on Food,Nutrition, and Health in December of 1969, Secretary ofAgriculture Hardin announced modifications in the food stampprogram. The modifications included: (1) reduction of pur-chase requirements, particularly for lower-income families;and (2) setting total coupon allotments for all participatinghouseholds at a level that would enable families to purchasea diet that was more nutritionally adequate. _3/

The administrative modifications, along with additionallegislative provisions, were adopted by Congress in January1971 (Public Law 91-671) and represented the first major legis-lative modifications to the Food Stamp Act since its enactmentin 1964. The amendments explicitly established nutritionalobjectives. Food coupon allotments were changed from levelsthat provided recipients an "opportunity more nearly to obtaina low-cost nutritionally adequate diet" 4_/ to "an opportunityto obtain a nutritionally adequate diet." j>/ The cost of anutritionally adequate diet, and therefore the amount of thefood coupon allotment, would be determined by the Secretary ofAgriculture and adjusted annually based on changes in the priceof food. Later amendments required semiannual adjustments.

These changes were significant. In fact, in subsequentlitigation a federal court characterized the changes as

. , . a major shift in the policy of the Food StampAct, a shift from supplementing the diets of low-income households to guaranteeing those householdsthe opportunity for an adequate diet. Congressplainly intended the 1971 amendments to assure thatno eligible family need be malnourished; the govern-ment would provide all the opportunity to be health-fully fed. [Emphasis in the original.] 6/

3/ The White House Conference on Food, Nutrition and Health,Summary Report, Washington, D. C., December 1970.

4_/ Sec, 4(a) Public Law 88-525, August 31, 1964, The FoodStamp Act of 1964, 78 Stat. 703-709.

5/ Sec. 3(a) Public Law 91-671, January 11, 1971, The FoodStamp Act of 1964 (Amendments) 84 Stat. 2048.

6/ Rodway v. U.S. Department of Agriculture, 514 F.2d 809,820 (D.C. dr., June 12, 1975).

As a result of the 1971 amendments, food stamp allot-ments were increased substantially above the previous standards,especially among lower-income households. A comparison ofthe food stamp allotments for a family of four prior to andfollowing the amendments in 1971 is shown in Figure 1. Before1971, the stamp allotment had increased from $60 to $120 forfamilies of four with increasing income; after 1971, all

Figure 1

Comparison of 1969 and 1971 Food Stamp IssuanceSchedules for Four Person Households. l

DOLLARS

120

100Qoo

I 80o

ECD

20

I—J

J

J I

Food Stamp PurchaseRequirement

J I

MAXIMUM -ALLOWABLE

NET MONTHLYINCOME _

'69

MAXIMUMALLOWABLE

NET MONTHLYINCOME

^'71

100 200 300 400

NET MONTHLY INCOME (DOLLARS)

1 Monthly coupon allotment and purchase requirements for a family of four in 1971, applies to 48 statesand District of Columbia. Monthly coupon allotments and purchase requirements for a family of fourin 1969, applies to Northern States Issuance Schedule. A Southern States Issuance Schedule had aslightly lower allotment level.

Source: Developed by CBO, from USDA Issuance Tables, effective February 1969 and January 1971.

eligible families of the same size received the same amount(e.g., $108 for a family of four).

In addition to changes in the basic allotment, the amountrequired to purchase the full stamp allotment was lowered.The 1971 amendments reduced the purchase requirement froman amount "determined to be equivalent to a household's nor-mal expenditures for food" 7_/ to an amount representing"a reasonable investment on the part of the household'sincome." 8/ The four-person household with $60 monthlyincome previously had to pay $20 for $65 worth of stamps; fol-lowing the amendments it paid $10 for $108 worth of stamps.In addition to these changes, special deductions from a house-hold's gross income were allowed, such as medical expenses andhousing costs, under the assumption that the program's purchaserequirement should not restrict a household from participat-ing when other important human social needs made legitimateclaims on the household budget.

The amendments specifically established national uniformincome and resource eligibility standards, work registrationrequirements for able-bodied adults, periodic adjustments infood stamp benefit levels, an increased federal involvement inadministration, and partial federal support of administrativecosts.

The 1971 amendments established the major structural.characteristics of the food stamp program as it operates today.The changes brought about increased benefits, increased parti-cipation, and, consequently, increased federal outlays.

Income Supplementation

With the 1971 modifications, the food stamp program be-came the first universal, national welfare program with na-tional eligibility standards based on need and not on partic-ular household characteristics. The effects of this broadening

7/ Sec. 7(b), Public Law 88-525, August 31, 1964, The FoodStamp Act of 1964, 78 Stat. 703-709.

8/ Sec. 5(b), Public Law 91-671.

of the scope of the program are discussed in the next partof this chapter.

The 1971 amendments had another effect--increased in-come supplementation. While some income supplementationcould have occurred previously, the modifications provided agreater opportunity for some households, (i.e., those thathad been spending more for food than the food stamp purchaserequirement) to purchase food stamps for less than they hadbeen spending. The difference between the normal food expendi-ture and the lower purchase requirement represented freedincome that could be spent for nonfood items. £/

Table 1 illustrates how this phenomenon can occur. Threehypothetical households have identical gross incomes andsimilar deductible expenses but, prior to participating inthe program, the three had different demand schedules forfood. Family 1 committed about 23 percent of its income to

TABLE 1. THREE HYPOTHETICAL FAMILIES AND IMPACT FOOD STAMPPROGRAM PARTICIPATION HAS ON FOOD PURCHASES ANDINCOME SUPPLEMENTATION ._

Category Family 1 Family 2 Family 3

Gross IncomeFood Expenditures(prior to FSP)

DeductionsNet FS Income

FS AllotmentPurchase Requirement

Food PurchasesFreed Income

$20046

20180

13046

+840

200130

20180

13046

0+84

20080

20180

13046

+50+ 34

It is assumed that the household's total demand for foodfollowing the food stamp transfer does not exceed the totalfood stamp allotment. Should this not be the case, thensome proportion of freed income could also be spent on food,

8

food, Family 2 65 percent, and Family 3 40 percent, Uponbecoming eligible to participate in the food stamp program,all (with the same level of deductions) were required tospend $46 to purchase food stamps having a value of $130,Since the first family was already spending $46 on food, thetotal value of bonus food stamps supported increased foodpurchases of $84. Since Family 2 was already spending $130on food, paying $46 for food stamps worth $130 allowed the -family to free up $84 for other purposes in the family bud-get. Family 3 appears to be the most typical case, since itwas already purchasing $80 worth of food. By purchasing foodstamps worth $130 at a cost of $46, the family increased itsfood purchase by $50 ($130-$80) and still freed up $34 forother expenditures ($80-$46),

Income supplementation is one of the key budgetaryissues of the future. Should the federal transfer bedirected only for food purchases? If not, how much generalincome supplementation through food stamps should beallowed?

GROWTH IN THE CURRENT PROGRAM

The food stamp program grew rapidly following the 1971amendments, causing concern among budget watchers. But anexamination of the major factors determining this growthsuggests that the food stamp program should not continue togrow at an uncontrolled pace in the future—unless the pro-gram is liberalized or food prices once again explode.

Mandatory Nationwide Implementation of the Food StampProgram

The 1971 amendments to the Food Stamp Act were intendedto increase participation and increase nutritional benefits.Additional amendments in 1973 (Public Law 93-86) mandatedthat all areas of the country offer the food stamp programby the beginning of fiscal year 1975. This latter amendmentmeant that states in which government jurisdictions werecontinuing to provide family food assistance through directfood distribution (food distribution program, FDP) wouldbe required to transfer to the food stamp program (FSP).

9

81-390 O - 77 - 3

The resulting program switchover (from FDP to FSP)explained a significant portion of the food stamp program'sgrowth between 1971 and mid-1974. At the beginning of 1971,approximately 1,993 food stamp project areas in 46 stateswere operating. By the end of 1974, over 3,044 food stampprojects in every state were operating. Between 1971:1(calendar year 1971, first quarter) and 1974:111, familyfood assistance participation (FFAP, which represents thesum total of participation in both FDP and FSP) increasedfrom 13.8 million persons to 15.3 million persons, a 2.8percent annual growth rate.10/ During this same period(1971:1 to 1974:111), food stamp participation increasedfrom 10.1 million to 14.6 million persons, an 11 percentannual rate of increase. Therefore most of the real growthin food stamp participation resulted from the transfer ofjurisdictions from FDP to the FSP. (See Figure 2.)

Food Price Inflation

From late 1972 to mid-1975, the world experienced un-usually rapid inflation in food prices. In the United States,the Consumer Price Index (CPI) for food rose at an annualrate of over 16 percent in the first half of 1973; from thelast half of 1973 to mid-1975, food prices rose by another25 percent.ll/ Because the basic allotment in the food

10/ All figures represent seasonally adjusted quarterlyaverage participation. Seasonal adjustments weredeveloped from the X-II Variant of Census Method IISeasonal Adjustment Program. See Technical Paper No.15, U.S. Department of Commerce, Bureau of the Census,Feb. 1967. Seasonal variation occurs in the FSP,reaching a seasonal high during fall and winter monthsand having seasonal lows during spring and summer monthsMost of these seasonal factors relate directly toseasonal employment patterns affecting low income house-holds nationwide.

H/ Factors currently put forth for the rapid increases infood prices have been discussed elsewhere by CBO andinclude a reduction in government-held grain stocks, adiversion of cropland, poor weather, a decision by theSoviet Union to make up its domestic shortfall throughpurchases abroad, increases in worldwide per capitademand for food, and the stimulative effects on U.S.exports of a dollar devaluation. See: U.S. Food andAgricultural Policy in a World Economy, CongressionalBudget Office, April 26, 1976.

10

Figure 2

Family Food Assistance Participantsand Food Stamp ParticipantsMILLION PERSONS

Quarterly Averages Seasonally Adjusted

FFAP TOTAL

Food Stamp -. _^^ . ^M ̂

III | IV I I II I III 1 IV I I II I III I IV I I II I III I IV I I IIP I l l l p | IV

12.0

10.0

P = Preliminary QUARTERS

stamp program is set at the "cost of a nutritionally ade-quate diet adjusted semiannually," and the household incomeeligibility standard has been defined as three times thecost of the "nutritionally adequate diet," 12/ food price

12/ The cost of a nutritionally adequate diet used in thefood stamp program is the least costly of four food plansdeveloped by the U.S. Dept. of Agriculture. This plan,referred to as the "Thrifty Food Plan," specifies theamounts of foods of different types that families mightuse to meet their recommended daily nutrient require-ments. Retail prices of foods included in the plan areupdated monthly using changes in retail prices of foodsin U.S. cities as collected and published by the Bureauof Labor Statistics.

11

inflation not only resulted in increased benefits, but alsoincreased income eligibility standards. This meant thatmore families became eligible for food stamps.

Table 2 shows the growth in food prices on a semiannualbasis between the first half of 1971 and the first half of1976. Because of lagged semiannual adjustments in the foodcoupon allotment levels, the dramatic increases in foodprices beginning in 1973 did not result in a major adjust-ment in the allotment levels and income eligibility levelsuntil the beginning of 1974. In January 1974, the monthlyfood coupon allotment level for a family of four increasedfrom $116 to $142, a 22.4 percent increase. Net incomeeligibility standards also increased by a similar amount,going from an annual level of $4,644 to $5,676 for a familyof four.

Table 2 also shows the official poverty line for anonfarm family of four. The poverty line, as defined by theOffice of Management and Budget (under Section 625 of theEconomic Opportunity Act of 1964), is revised at annual inter-vals, "or less as deemed desirable," based on changes in theoverall CPI. The widening divergence between the CPI forfood and the overall CPI, beginning in 1972, accounts forthe fact that income eligibility standards in the food stampprogram for a family of four today are approximately 14percent higher than the estimated poverty line for thatfamily in 1976. Because different methods are used to adjustthe food stamp eligibility standard and the poverty line fordifferent family sizes, the gap between different familysizes also varies. For example, a family of six in the foodstamp program has a net income eligibility standard nearly25 percent higher than the official poverty standard forthat size family.

Recession

Almost coincidental with the adjustments in incomeeligibility standards was the dramatic growth in unemploy-ment. Unemployment rose from 5.3 percent of the laborforce in August 1974 to a high of 9.1 percent in Marchof 1975. The impact of the recession on the food stamp pro-gram is best illustrated by disaggregating food stamp (FSP)and family food assistance (FFAP) participants into public

12

TABLE 2. FOOD PRICE CHANGES, CPI CHANGES, FOOD STAMP ALLOTMENT, FOOD STAMPINCOME ELIGIBILITY, AND POVERTY LEVELS (1971 FIRST HALF TO 1976FIRST HALF)

Calendar Percentage change fromYear: Half preceding period at

compound annual rate

Family of FourFood Stamp Food Stamp Net OfficialAllotment Income Eligi-, Poverty

bility Level-7 Threshold-/

Food All-Items Annual Annual

71:71:

72:72:

73:73:

74:74:

75:75:

7676

III

III

;I

: II

I: II

: I: II

: I:II

33

44

1620

149

87

1-

.2

.8

.2

.7

.2

.7

.7

.2

.0

.9

.2-

43

33

59

1012

87

4-

.0

.6

.0

.6

.9

.2

.6

.9

.3

.3

.9-

$108108

108112

112116

142150

154162

166166

$360360

360373

373387

473500

513540

553553

$44

44

44

56

66

66

, 320,320

,320,476

,476,644

,676,000

,156,480

,636,636

$4,4,

4,4,

4,4,

5,5,

5,5,

5,5,

137137

275275

540540

050050

500500

815 (est.)815

Eligibility in the FSP is based on a net income concept, meaning that house-holds may deduct certain expenditures from a gross income higher than thislevel and still qualify for food coupons.

The official poverty line is based on a gross income concept. Officialpoverty threshold levels are published in the spring for income levelsof the previous year. Therefore, the official federal poverty thresholdpublished in April 1976, $5,500 for a family of four, applies to incomein 1975. Threshold levels for 1976 are estimated on July 1976 economicassumptions developed by the Congressional Budget Office.

13

assistance households and nonpublic assistance house-holds. 1J3/ Over the period from 1971 through the thirdquarter~of 1976, the number of public assistance foodstamp recipients (FS-PA) increased from about six millionrecipients to approximately 8.5 million, a 6.3 percentannual rate of increase. (See Figure 3.) However, mostof this growth came from the switchover from the FDP to theFSP. Public assistance households enrolled in either theFSP or the FDP (FFAP-PA) showed hardly any growth, goingfrom 8.1 million recipients to 8.5 million recipients, a1.0 percent increase compounded at an annual rate. Duringthe same period, the number of AFDC recipients increasedfrom 10.0 million recipients to approximately 10.8 millionrecipients, a 1.3 percent annual increase, approximatelythe same annual increase as the FFAP-PA growth rate.

The growth of nonpublic assistance recipients (FFAP-NPA, FS-NPA) was much more dramatic, as it was highly

13/ A food stamp public assistance participant (FS-PA) isdefined as a person residing in a household in whichevery member of that household is a recipient of afederally aided public assistance or general assistancegrant and is also receiving food stamps. Householdsreceiving Aid to Families with Dependent Children(AFDC) or General Assistance (GA) would be classifiedas public assistance food stamp households and undercurrent rules would be categorically eligible for aminimum amount of food coupons regardless of post-transfer income and resources of the household members.It is important to note that, while public assistancerecipients constituted approximately 46 percent of thetotal food stamp population, the definition of publicassistance for food stamp purposes understates thetotal number of actual public assistance persons re-ceiving food coupons. While people who receive Supple-mental Security Income (SSI) are eligible for foodstamps under the same provisions discussed above forpublic assistance households (e.g., if they live aloneor with other SSI recipients or with a public assis-tance household, they are categorically eligible forfood coupons), SSI recipients are officially recordedin the administrative program statistics as nonpublicassistance recipients (FS-NPA).

14

Figure 3

Family Food Assistance • Public Assistance Participants andFood Stamp - Public Assistance ParticipantsMILLION PERSONS


10.0

8.0

6.0

4.0

FFAP-PA

J i I I Il | II MM I IV I I 'I I "I I IV nTin I iv TH I HI I iv I n 1 in | iv I iip| i

1971 1972 1973 1974 1975 1976P = Preliminary QUARTERS

sensitive to the general economic conditions. There wasvirtually no growth in FFAP-NPA from the beginning of 1971until late 1973. High unemployment rates, increased util-ity costs (thereby increasing housing cost deductions),and increasing income eligibility standards generated bythe food price spiral, all coming at about the very timethe FSP replaced the old FDP, resulted in nonpublicassistance participation increasing from 4.7 million

15

persons in 1973:111 to nearly 10,6 million persons in1975:111, a 42 percent growth rate, (See Figure 4.)

Figure 4Family Food Assistance - Nonpublic Assistance Participantsand Food Stamp • Nonpublic Assistance ParticipantsMILLION PERSONS

10.0


4.0

P = Preliminary QUARTERS

The FFAP-NPA peaked at the same time as the unem-ployment rate in mid-1975 and has declined graduallyto approximately 9.3 million in 1976:111. This recentexperience suggests that a one percentage point change

16

in the overall unemployment rate will cause the number ofFFAP-NPA to change by approximately 628,000 persons.14/

Puerto Rico's Entrance into the FSP

The 1971 amendments to the Food Stamp Act of 1964 ex-tended the program to include Guam, Puerto Rico, and theVirgin Islands. The Commonwealth of Puerto Rico main-tained an active food distribution program up to the endof fiscal year 1974, reaching over 600,000 persons. Overthe first half of fiscal year 1975, Puerto Rico graduallybegan to phase in the food stamp program. Between July1974 and January 1975, food stamp rolls in Puerto Ricoincreased from 30,000 to 1 million persons. By the endof fiscal year 1976, participation had stabilized at about1.5 million, nearly half of Puerto Rico's population.

While the basic monthly food allotment in PuertoRico has not been significantly different from that ofthe contiguous 48 states, and therefore net income eligi-bility standards have been comparable to those of thestates, lower incomes result in higher average food couponbenefits and higher participation. In fiscal year 1976,the average food stamp recipient in Puerto Rico receivedapproximately $28.60 worth of bonus stamps monthly, 22percent higher than the average monthly benefit receivedby food stamp recipients in the continental United States($23.35), Beginning in fiscal year 1977, food price in-flation on the Island (probably resulting from the foodstamp program) resulted in higher food stamp allotmentsthan in the contiguous 48 states. The monthly allotmentfor a family of four stood at $172 in Puerto Rico, com-pared to $166 in the 48 states in July 1976. Net income

14/ CBO estimate based on functional relationship, esti-mated between FFAP-NPA, unemployment rates, personalincome and net food stamp income eligibility standardsThe estimate compares to that reported by the USDA tobe about 500,000-750,000, as discussed in a report tothe U.S. Senate in accordance with Senate Resolution58. Food Stamp Program, A Report in Accordance withSenate Resolution 58, USDA, Food Nutrition Services,July 2.1, 1975, p. 5.

17

eligibility standards were therefore four percent higherin Puerto Rico than in the 48 states.

Outreach

The 1971 amendments also established the requirementthat state agencies were to undetake "effective action,including the use of services provided by other federallyfunded agencies and organizations, to inform low-incomehouseholds concerning the availability and benefits of thefood stamp program." A series of court actions calledthe Department of Agriculture to task for failure to imple-ment the 1971 amendments. Altogether, 18 court suitshave been brought against states for failure to adequatelyimplement the outreach requirements mandated by the 1971amendment. In response to this situation, in April 1975USDA established one full-time outreach coordinator ineach state to report to USDA on the state's outreachactivities. Outreach efforts were further to be directedtoward the special needs of, among others, the elderly,the disabled, migrants, persons residing in rural areas,and ethnic groups.

Also, the Community Food and Nutrition Program (CFNP),administered by the independent Community Services Admin-istration, the successor agency to the Office of EconomicOpportunity, has as one of its purposes the focusing ofthe food stamp program (and other federal nutrition pro-grams) on the most needy poor. Federal grants are providedpublic and/or private organizations to carry out projectswhich include outreach and information. It is not possibleto determine the specific impact of increased outreachefforts on the growth of the food stamp program from offi-cial published statistics.

SUMMARY

Federal food assistance has moved almost entirely outof the direct purchase and distribution of surplus agri-cultural products to indirect food assistance to needyfamilies. From the direct provision of food commoditiesfrom outside the market with minimum recipient choice,federal programs have shifted to the indrect provision of

18

food through stamps used in the market by recipients whohave some freedom of choice.

At the same time, programs originally constrained bybeing subject to the appropriations process have becomeopen-ended entitlement programs. Appropriations now aremade to finance assistance on the basis of demand, a factwith great fiscal implications.

The 1971 amendments altered the nature of the foodstamp program; further they greatly increased the scopeof entitlement to participate in the program. The amend-ments were followed by changes in economic conditions thatgreatly magnified the impact of the programmatic changes.

From 1969 to 1976, the average number of monthlyparticipants increased from 3 million to over 18 million,as did all concomitant costs. Total program costs (in-cluding both state and federal outlays) increased from$272 million in fiscal year 1969 to approximately $5.9billion in fiscal year 1976. (See Figure 5.) The averagemonthly transfer for an individual increased from slightlyover $6.60 in 1969 to approximately $23.90 in 1976. Ad-ministrative costs went from $22 million (7.8 percent ofthe total cost) to nearly $370 million (6.2 percent oftotal cost) over the same period.

The growth in food stamp program participation andcosts over the last five years should not have beentotally unexpected. The growth developed out of the spe-cific legislative amendments of 1971, the action of thefederal courts in insuring that the legislative intent^was carried out, and the combined effects of worldwidefood price inflation, program design, and a generallydepressed economy. To the extent that these phenomenaare predictable, program growth can also be forecastwith some degree of reliability. Unfortunately, most ofthe factors that brought about the upsurge of food pricesbeginning in 1972 (i.e., worldwide drought and unfavor-able weather conditions, the 1973 oil embargo by the OPECcountries, the large grain purchases by the Soviet Unionin 1972, the poor anchovy catch in South America, and cornblight in the United States) were not entirely predict-able. Given that such circumstances will continue, long

19

Figure 5

Total Federal and Non-Federal Cost (Obligations)off Food Stamp Program

BILLIONS OF DOLLARS7.0

6.0

5.0

4.0

3.0

2.0

1.0

Non-Federal Administrative Costs

Federal Administrative and Other Program Costs

Food Stamp Bonus Costs

1969 1970 1971 1972 1973 1974 1975 1976

FISCAL YEAR

Source: Congressional Budget Office, see Table A1, Appendix.

20

range projections of the food stamp program must remaino/~»mcni7V» a + c?v*o/"*n~lQ-l-nT7o 1 ̂ /somewhat speculative.15/

15/ Chapter V will discuss long range projections undera current policy budget option given an optimisticeconomic condition.

21

CHAPTER II BENEFITS FROM THE FOOD STAMP PROGRAM

Who benfits from the food stamp program? How andhow much do they benefit? The primary beneficiaries,obviously, are the recipients or the users of food stamps,and another beneficiary is the American farmer. Thischapter gives a picture of the population using foodstamps, then analyzes the various benefits that thatpopulation receives from the food stamp program. Finally,the chapter explains how the food stamp program is ofdirect benefit to the American farmer, continuing one ofthe original purposes of domestic food assistance programs.

Profile of Food Stamp Households Today

A recent Department of Agriculture survey of eligiblehouseholds provides a snapshot of the current food stamprecipient population.!^/ That survey estimates that

16/ "Characteristics of Food Stamp Households Septem-ber 1975," Program Development Branch, Food StampDivision, Food and Nutrition Service, USDA: May 14,1976.

In May 1976, the U.S. Dept. of Agriculture re-leased official results from a survey of food stamphouseholds certified eligible for the FSP in Septem-ber 1975. The September 1975 food stamp survey con-tains the most current data available for improvedprogram analysis, estimates of the eligible pop-ulation, and information describing FSP participants.The survey included information on 11,327 eligiblefood stamp households receiving both public assistanceand nonpublic assistance income. Income, demographic,and program information was collected (or imputed)for all households in the survey. The observationswere appropriately weighted to represent the totalnumber of households officially reported as partic-ipating the the program in September 1975. The de-tailed, itemized deduction information collected fromthis survey represents the first such informationreported since November 1973.

23

approximately 55 percent of the estimated 10.4 millioneligible households participate in the food stampprogram. These eligible households represent approx-imately 31 million persons. Participation rates varyby public assistance status; nearly 93 percent of thepublic assistance households, but only 50 percent ofthe eligible nonpublic assistance households, partic-ipate in the program.

In September 1975, food stamp households averagedabout 3.3 persons (3.2 persons excluding Puerto Rico).Over 64 percent of all households are headed by females,and 45 percent of all households consist of one or twopersons. Seventeen percent of the households containan elderly individual, age 65 or over. The majorityof food stamp households reside inside central cities.National distribution is weighted toward the South.(See Table 3).

Nearly 60 percent of the heads of food stamp house-holds were reported as not being in the labor force in1975. This compared to a national figure of 26.8 per-cent. About 22 percent of the participating householdsreported having earned income. Households with earnedincome less that 125 percent of the poverty level andwith no public assistance benefits constituted about13.5 percent of the participating households. (Comparedto the Aid to Families with Dependent Children (AFDC)program, which reported 16.3 percent of its familiesreceiving earned income in 1973, the food stamp populationdoes not appear to be weighted as heavily toward the"working poor" as has often been suggested.)17/

17/ The term "working poor" as used here refers only tohouseholds with earned income less than 125 percentof the poverty line. It does not include householdsincluded in the labor force but at the time of thesurvey unemployed.

24

TABLE 3. PROFILE OF FOOD STAMP AND TOTAL U.S. HOUSEHOLDS

Food Stamp U.S.

July-Sept . 1975 March 1975Percent PercentDistribution-' Distribution^'

Household SizeTOTAL

OneTwoThreeFourOver Four

Region CensusTOTAL

NortheastNorth CentralSouthWest

Residence

Nonf armMetropolitanInside Central CitiesOutside Central CitiesNonmetropolitanFarm

Family TypeTOTAL

Male headFemale head

Age of Head of HouseholdTOTAL

Over 65Under 65

RaceTOTAL

WhiteNon-White

Employment Status of Head—TOTAL

EmployedUnemployedNot in Labor Force

100.0

24.020.517.114.523.9

100.024.421.738.615.2

98.264.744.819.933.51.8

100.035.664. 3

100.017.282.8

100.062.737.3

100.028.812.558.6

a./ Distributions developed from CBO tabulationsUSDA September 1975, Foodregion, employment statussource Characteristics of

Stamp Survey tape.

100.0

19.630.617.415.616.8

100.022.526.732.718.0

96.268.631.437.231.43.8

100.076.423.6

100.020.179.9

100.088.511.5

100.068.34.926.8fromRace ,

, and residence distributionHouseholds Purchasing Food

Stamps, Current Population Reports, Series P-23, No.61, July 1976.

b/ Data based on household information for March 1975,source: Household Money Income in 1974 and SelectedSocial and Economic Characteristics of Households,Current Population Reports, Series P-60, No. 100,August 1975. Region data based on family data,source Money Income and Poverty Status of Familiesand Persons, Current Population Reports, SeriesP-60, No. 103, September 1976.

(}/ Employment status refers to employment during weekprior to interview.

81-390 O - 77 - 4

The major source of income received by food stamphouseholds was the AFDC program; 41.7 percent receivedincome from this source in September 1975. Nearly 67percent of food stamp households had income from basicpublic assistance programs, either AFDC, SupplementalSecurity Income, or state general assistance programs.Approximately 21.4 percent of the food stamp householdsalso received social security benefits. (See Table 4.)

TABLE 4. SOURCES OF INCOME FOR FOOD STAMP HOUSEHOLDS(SEPTEMBER 1975)

Percent of householdsreceiving income

Source of Income from source

Aid to Families with DependentChildren (AFDC) 41.7

Earnings 22.4Social Security 21.4Supplemental Security Income (SSI) 17.1General Assistance 8.2Veterans benefits 3.4Self-employment 1.2Railroad retirement and pensions 0.9Student Aid 0.8Other (unemployment compensation, etc.) 20.9

SOURCE: CBO tabulations from USDA September 1975Food Stamp Survey tape.

A major concern in the food stamp program has beenthe number of "high income" recipients receiving foodstamps. Based on the September 1975 data, the meanmonthly gross income for food stamp households wasestimated to be $292.35 (including only cash income, noin-kind income such as medicaid, social services, orpublic housing assistance). This estimate includes theCommonwealth of Puerto Rico and other outlying terri-tories. If these outlying territories are excluded fromthe tabulations, the average gross income increases to$298. The mean annual income of a food stamp household

26

($3,576) was approximately 23 percent of the mean familyincome in the United States in 1975.i§/ The food stamppopulation's income distribution is weighted heavily to-wards the lower income levels as compared with the pop-ulation as a whole. (See Table 5.)

TABLE 5. DISTRIBUTION OF GROSS MONTHLY INCOME FOR FOODSTAMP HOUSEHOLDS AND FOR ALL UNITED STATES

HOUSEHOLDSIncome Class Food Stamps

50 states & D.C. For all U.S.(Monthly Income) (Percentage of (Percentage of

Households^ Households)USDAa/

Under$ 100200300400500600750

$1000

$100tototototototoor

$199299399499599749999over

7262518105410

.1

.6

.6

.4

.1

.6

.3

.9

.4

CENSUSb/7242617116311

.1

.8

.7

.8

.3

.1

.2

.9

.1

25777791229

.8

.7

.4

.1

.6

.1

.6

.3

.3

aySOURCE:CBO computer tabulations from USDA September1975 Food Stamp Survey tape. Gross monthly incomein the USDA tabulations was defined as gross sala-ries, wages and training allowances (before taxes),roomer and boarder payments, self-employment income,student loans, grants, scholarships, AFDC, GA, SSI,Social Security, veterans, Railroad Retirement, andall other monetary payments such as workmen's com-pensation and unemployment benefits.

b/ SOURCE: Characteristics of Households PurchasingFood Stamps July 1975, Current Population Reports,Series P-23, No. 61, July 1976. Tabulations were onthe basis of monthly money income defined to includeessentially all sources as listed above in footnote 1.All income data were collected before taxes. U.S.totals were based on monthly income in July 1974.

18/ The average mean income of all families in 1975was $15,546 as reported in Money Income andPoverty Status of Families and Persons CurrentPopulation Reports. Series P-60, No. 103, Sept.1976. Given that FSP eligibility is based onmonthly income basis, the 23 percent estimateprobably represents a lower end figure. Seefootnote 19.

27

TABLE 6. DISTRIBUTION OF FOOD STAMP PERSONS BY POVERTY LEVEL (PRE-FOOD STAMP BONUS TRANSFER),SEX OF HEAD OF HOUSEHOLD, AGE OF HEAD OF HOUSEHOLD, HOUSEHOLD SIZE (SEPTEMBER 1975)

Poverty LevelCategory Less than

Total Number of Persons(Percent of Total)

Sex of Head of HouseholdMale-HeadedFemale -Headed

TOTAL fL/

Age of Head ofHousehold:

Less than 2525-3435-4445-5455-6465+

TOTAL a/

Household Size:1234567+

14,438(77.9)

5,6508,738

14,438

1,8554,3833,7502,0341,1481,266

14,433

1,1821,7022,1062,4732,0041,6573,258

Poverty 1.01-1.25 1.26-1.49 G.T. 1.50 Poverty(persons in thousands)

2,375 957 762(12.8) (5 .2) (4.1)

(39.1)(60.9)(100.0)

(12.9)(30.4)( 26 . 0 )(14.1)(8 .0)(8.8)(100.0)

(8 .2)(11.8)(15.0)(17.1)(13.9)(11.5)(22 .6)

1,0971,2782,375

242768646367168185

2,375

123357417408393290388

(46.2)(53.8)

516441

(100.0)957

(10.2)(32.3)(27 .2 )(15.5)(7.1)(7.8)

92370271

957058

(100.0)957

(5.1)(15.0)(17.6)(17.2)(16.5)(12.2)(16.3)

3512017320417994

153

(53.8)(46 .2)(100.0)

(9 .6)(38.7)(28.3)(9 .9 )(7 .3)(6.1)(100.0)

(3.7)(12.4)(18.1)(21.3)(18.7)(9.8)(15.9)

410352762

85356174

923520

762

191411581951116968

(53.8)(46.2)(100.0)

(11.1)(46.7)(22.8)(12.1)(4 .6)(2.6)(100.0)

(2 .5 )(18.5)(20 .7)(25.6)(14.6)(9.1)(8.9)

Total

18,532(100.0)

71018

254211

18

1o

23223

,672,860,532

,274,877,841,587,421,529,532

,358,320,914,279,686,110,866

(41.4)(58.6)(100.0)

(12.2)(31.7)(26.1)(14.0)(7 .6)(8.3)(100.0)

(7.3)(12.5)(15.7)(17.7)(14.5)(11.4)(20.9)

TOTAL a/ 14,438 (100.0) 2,375 (100.0)957 (100.0) 762 (100.0)

SOURCE: CBO tabulations from USDA September 1975 Food Stamp Survey tape,

a/ Figures may not add to exact totals because of rounding.

18,532 (100.0)

Another measurement of high income recipients,however, is the official poverty line, which takesinto consideration differing income needs for varyingfamily sizes. Of all the households certified for foodstamp participation in September 1975, 77.9 percent hadgross monthly incomes which fell below the poverty line,and 90 percent fell below 125 percent of poverty.l^/Between 22,000 and 62,000 households with incomes over$12,000 participated in the program during mid-1975.(See Table 6.) For food stamp households whose grossincome exceeded the poverty threshold, participationtended to be skewed toward larger, male-headed house-holds .

In general, food stamp benefits decline with in-creasing income. Benefits averaged nearly 2.4 timeshigher for households whose gross monthly income wasbelow the poverty line as compared to those alreadyabove poverty. Of all food stamp bonus dollars dis-tributed in September 1975, 86.5 percent went to the77.9 percent of the households whose income fell belowpoverty. (See Table 7.) Recipients whose income ex-ceeded 150 percent of poverty received only about 2percent of the total benefits, or approximately $7.9million, in September 1975.

19/ The survey of Food Stamp Household Characteristicsfor September 1975 was analyzed using the follow-ing specific poverty thresholds for different familysizes:

1 person - $2,590 5 persons - $5,8702 persons - 3,410 6 persons - 6,6903 persons - 4,230 7 persons - 7,5104 persons - 5,050 Each additional

person +820

These poverty income guidelines represent theofficial OMB guidelines published in April 1975,effective for income in 1974 and assumed operativeat the time of the food stamp survey. Since re-cipient income was available only on a monthlybasis, the monthly income of the recipients wascompared to these poverty levels adjusted to amonthly basis. The procedure probably understatesthe number of high income recipients when income iscounted on an annual basis because of intra-yearvariations in income.

29

TABLE 7. AVERAGE MONTHLY FOOD STAMP BENEFITS FORRECIPIENTS BY POVERTY STATUS AND SEX OFHEAD OF HOUSEHOLD (SEPTEMBER 1975)

Poverty Sex ofLevel Head

Less thanPoverty

TotalMaleFemale

TotalBenefits(Millionsof Dollars)

$354.2138.9215.4

Percentof TotalBenefits

86 . 5%33.952.5

AverageBenefitperHousehold(Dollars)

$81.4690.4676.54

AverageBenefitperPerson(Dollars)

L

$25.0625.1125.04

1.01-1.24Total 35.8 8.7

MaleFemale

16.219.7

4.04.8

51.0255.8447.64

15.4015.0815.74

1.25-1.50Total

MaleFemale

Greater than 1.50Total

MaleFemale

11.76.05.7

7.93.94.0

2.91.51.4

1.90.91.0

43.8447.9340.17

34.4044.5527.92

12.4811.8813.19

10.589.7211.59

SOURCE: CBO tabulations from USDA September 1975 Food StampSurvey tape.

Poverty Reduction Effects

One measure of the food stamp program's effective-ness is its impact on reducing poverty. Bonus food stampdollars are not counted in a household's income when

30

determining poverty counts. Adding bonus dollars to ahousehold's income reduces the number of households(persons) officially counted as poor.20/

Table 8 examines only those recipients whose pre-food-stamp incomes were below the poverty line. About30 percent of the pre-food-stamp-transfer poor personswould be lifted above the poverty line if the food stampbonus were counted as income; this represents a reductionof approximately 420,000 persons. The reduction of thepoverty count appears to be greater for male-headed house-holds (32.4 percent) than for female-headed households.The transfer also removes a greater proportion of theyounger-headed households from poverty than elderly-headed households, and appears to be less significantin removing single-person households from poverty (10.3percent).

AFDC Gap Filler

The food stamp program has in some instances becomean equalizer in state welfare payments. State differencesin both need and payment standards in the AFDC programare narrowed by the food stamp program with its nationallyestablished allotment levels. AFDC benefits are countedbefore determining food stamp benefits. The differencein average AFDC payments in September 1975 between theMid-Atlantic region (including high paying states suchas New York, New Jersey, t.nd Pennsylvania) and the South-east region (including low paying states such as Alabama,Georgia, and Mississippi)

concentration

20/ While this procedurejust how far personssuggest thepoor persons. Morecost of the bonus tosame by the recipientand Nutrition, (Aprilaverage recipientcent of its face valof the bonus stampsoverstate the poverty

values

was over $164. After AFDC

admittedly does not indicatemove out of poverty, it does

of benefit dollars amongimportant, it assumes that thethe government is valued the

Clarkson, in Food Stamps1975), has estimated that the

the bonus at about 33 per-ue. Therefore, the full valuein the household's income mayreduction.

31

TABLE 8. DISTRIBUTION OF FOOD STAMP RECIPIENTS PRE-FOOD-STAMP-TRANSFER POOR AND POST-FOOD-STAMP-TRANSFER POOR, BY HOUSEHOLDCHARACTERISTICS

Category

Total Numberof Persons

(Percent ofTotal FSParticipants)

Sex of Head ofHousehold :Male-Headed

Pre-TransferPoorFS Parti-cipants(persons

14,438

(77.9)

5,650Female-Headed 8,788

Age of Head ofHousehold :Less than 2525-3435-4445-5455-6465+

Household Size:1234567+

1,8554,3833,7502,0341,1481,266

1,1821,7022,1662,4732,0041,6573,258

Post-TransferPoorFS Parti-cipants

in thousands)

10,236

(55.2)

3,8146,340

1,3432,8882,6851,436879

1,011

1,0601,4601,3801,5441,2531,2242,557

PovertyReduction[(Col. 1-Col.4 Col. 1]

29.1%

32.4%27.8%

27.5%34.1%28.4%29.4%23.4%20 . 1%

10 . 3%14 . 2%36 . 2%37.5%37.4%26 . 2%21.4%

SOURCE: CBO tabulations from USDA September 1975 FoodStamp Survey tape.

TABLE 9. IMPACT OF BONUS FOOD STAMPS ON NARROWINGAFDC PAYMENT DIFFERENTIALS BETWEEN REGIONS,SEPTEMBER 1975

USDA AdministrativeRegion—/

ALL AREAS AVERAGE

New EnglandMid-AtlanticSoutheastMid-WestWesternWest Central

AFDCGrant

$333.43

276.07291.43126.69256 . 36272.64149.73

BonusFood Stamps

$84.07

73.1069.95121.3178.7670.48119.83

AFDC PlusFood Stamps

$417.50

349.17361.38248.00335.12343.12269.56


a./ REGIONS: New England: Connecticut, Maine, Mass-achusetts, New Hampshire, Rhode Island,Vermont

Mid-Atlantic: Delaware, District ofColumbia, Maryland, New Jersey, New York,Pennsylvania, Puerto Rico, Virginia, VirginIslands, West Virginia

Southeast: Alabama, Florida, Georgia,Kentucky, Mississippi, North Carolina,South Carolina, Tennessee

Midwest: Illinois, Indiana, Iowa, Kansas,Michigan, Minnesota, Missouri, Nebraska,Ohio, Wisconsin

Western: Alaska, Arizona, California,Guam, Hawaii, Idaho, Nevada, Oregon,Samoa, Trust Territories, Washington

West Central: Arkansas, Colorado,Louisiana, Montana, New Mexico, NorthDakota, Oklahoma, South Dakota, Texas,Utah, Wyoming

33

households in these regions participated in the foodstamp program, the difference was narrowed by nearly$52. (See Table 9.) The widest AFDC regional dif-ferential, a ratio of almost 2.3 to 1, is reduced toabout 1.5 to 1 as a result of the food stamp program.

To the extent that state governments maintaintheir current payment levels in the AFDC program,mandated increased federal food stamp allotmentsprovide a mechanism by which states are relieved ofsome financial burden over time. Recipients mayexperience no reduction in real benefit levels asa result of this shift in financing.

Nutritional Effectiveness

Probably more important as a measure of the foodstamp program's effectiveness is its ability to meetits stated goal of improving the diet of low incomehouseholds. An important distinction must be made be-tween "nutritional" effectiveness and simply increasedfood expenditures. The latter may be a necessary con-dition for the former, but not a sufficient condition.While a number of studies have investigated the increasedconsumption of food as a result of food stamp partic-ipation, there have been few investigations to deter-mine how effective food stamps have been in improvingthe nutritional standards of the recipients.fLL/ Withreference to food intervention programs, the NationalAcademy of Sciences in late 1975 concluded:

Little or no effective evaluation of theimpact of these programs [e.g., food fort-ification, nutritional labeling, nutritioneducation, and supplemental feeding programssuch as food stamps, surplus food distribution,school lunches and women and infant children

21/ Most recently, the 1971-1972 First Health andNutrition Examination Survey (HANES) has collecteddietary, biochemical, and anthropometric data ona national sample of households, including foodstamp receipients and nonparticipating low incomehouseholds. At the time of this report, the datahave not been released.

34

feeding] on the nutritional well-being ofthe target groups has been carried out. Thecost-effectiveness of these programs versusother means of expanding purchasing powerhas not been evaluated for their relativeimpact on nutritional goals. As a consequencethere is little information with which toassess the continuing value of these programs. 2J2/

The few evaluation studies that do exist are limitedprimarily to selected counties scattered throughout theUnited States. A study of two Pennsylvania counties in1969 and 1970 attempted to uncover the nutritional impactof shifting from the older food distribution program tothe food stamp program.,23_7 Findings indicated that,among other things households participating in the fooddistribution program had no better diets than thosesimilar households that did not participate. With re-spect to the food stamp program, the study concludedthat food stamps provided some improvement in diets offamilies experiencing temporary shortages of funds, i.e.,when more than two weeks had elapsed since the familyreceived income from its last major source of pay. Butwhen households had received income within two weekspreceding their dietary interview, the impact of foodstamps on the households' nutrient intake was foundto be insignificant. After the major allotment andpurchase requirement changes in 1970, little, if any,dietary improvement was shown for households who con-tinued to participate in the food stamp program.

22/ World Food Nutrition Study, Enhancement of FoodProduction for the United States, Report of theBoard of Agriculture and Renewable Resources,National Academy of Sciences, November 1975, p. 61.

23/ J. Pattrick Madden and Marion D. Yoder, ProgramEvaluation: Food Stamps and Commodity Distributionin Rural Areas of Central Pennsylvania, Penn-sylvania State University, Agricultural ExperimentStation Bulletin 780, June 1972.

35

A similar nutritional evaluation study was con-ducted in Kern County, California, between August 1972-and February 1973.24y The survey concluded that foodstamp households had higher intakes for some nutrients.The study further concluded that participants in eitherthe food distribution program or the food stamp programspent lower percentages of higher total incomes onfood, indicating some income was "freed" from normalfood expenditures because of the receipt of free foodor bonus food stamps. The study found that the averagemonthly value of food available for participating house-holds was $144 compared to $126 for nonparticipants.As this difference of $18 was only about half of thevalue of bonus food stamps ($43.70) available to anaverage participating household, it appears that someportion of freed income went for increased food con-sumption and some for other, nonfood goods.

Even when food stamps do result in increased foodexpenditures for participating households, it is notclear that these higher levels of food expenditurestranslate into an overall higher level of nutritionalstatus. The Consumer Expenditure Diary Survey foundthat, on the average, food stamp households spentapproximately four times as much for nonalcoholic bev-erages (excludes fresh whole milk) than did nonfoodstamp households. By its very construction, the couponallotment for a nutritionally adequate diet, as basedon a carefully constructed food plan, requires a con-sumption pattern consistent with the consumption pat-terns of low-income households in 1965-66. Given thatlow-income households now have different food consumptionpatterns, it has been suggested that it is more thepurchasing behavior of the households using food stamps,and not the basic food coupon allotment or the increasedfood expenditures, that determines the nutritive successof the program. Modification of participant behavior,along with adequate allotment levels, would be requiredin any proposal designed to increase nutritional effec-tiveness of the program.

24/ Sylvia Lane, Food Distribution and Food StampProgram Effects on Nutritional Achievement ofLow Income Households in Kern County, California,University of California at Davis, forthcoming.

36

Food Consumption - Income Tradeoff

Preliminary data from the 1972-1973 Consumer Ex-penditure Diary Survey substantiate expenditure find-ings from these studies and provide a basis foranalyzing the general income substitution effects.Based on weekly food expenditures for food stamp house-holds and nonfood stamp households with similar incomeand household sizes, the Survey estimated that on theaverage approximately 43 percent of the bonus transferis used to free income for nonfood purchases. The re-maining 57 cents out of every $1.00 transfer goes towardfood expenditures.

Figure 6 clarifies this phenomenon. It shows theestimated weekly expenditures for food by food stamprecipients (age of household head less than 65) and theestimated expenditures by income classes for food in theabsence of the program. The data indicate that, betweenJuly 1973 and June 1974, food stamp households on theaverage spent approximately $27.00 per week on food,while nonfood stamp households with the same average .income would have spent $17.63 weekly on food..2-5-/ Giventhat the average food stamp household's purchase require-ment was $10.50 per week, the household was able to freeup to $7.13 per week in income (i.e., the $17.63 thatwould have been spent on food minus the $10.50 purchaserequirement) and still was able to increase food con-sumption by as much as $9.37 (i.e., the $27.00 averagefood consumption for food stamp household minus the$17.63 that would have been spent on food regardless of

25/ The relationship between weekly food expenditureswas adjusted to an annual basis, and annual incomewas fit to nonfood stamp households, and then wasevaluated at the average annual income of foodstamp households to determine the approximate ex-penditures for food for similar nonfood stamphouseholds.

WHERE: ANFDHO = Weeklyexpenditures for foodby nonfood stamp house-holds with income up to

ANFDHO = 553.4 + .0845Y(t-statistics) (10.893)

R2 = .908 D.W. = 1.958$9,000 multiplied by 52.

Y = Reported average annual income.

37

Figure 6

Income Supplementation and Food Consumption Impact,Food Stamp Households, Age of Head Less Than 65,1973-1974 Consumer Expenditure Survey

WEEKLY EXPENDITURE FOR FOOD AT HOME$50

40

30

20

10

Represents additional food purchasesresulting from food stamps.

Represents income freed for non-food purchasesas a result of food stamp participation, thisamount plus purchase requirement is estimatedhousehold expenditure on food in the absenceof food stamps.

Represents average purchase requirement,actual recipient cost.

$1- $1,800- $3,000- $4,200- $5,400- $6,600-1,799 2,999 4,199 5,399 6,599 8,999

ANNUAL INCOME

Source: Estimates prepared by CBO based on data from the Consumer Expenditure Diary Survey, 1973-1974 DiaryComponent of Survey: Special Tabulations, Bureau of Labor Statistics, August 1976. Estimates of expendituresfor food in absence of food stamp program based on weekly food expenditures for households not participatingin program, but falling within same income class.

38

the program). The proportion of the federal bonustransfer going to supplement incomes for general pur-chasing power through substitution in other portionsof household budgets was nearly 43 percent (i.e.,$7.13 freed up income -r $16.50 federal transfer).

It should be noted that this brief analysis makesa crucial assumption that probably biases the estimatedincome supplementation downward. It was assumed that,in the absence of the food stamp program, participatinghouseholds would spend approximately the same amount onfood as those currently nonparticipating householdswith similar gross money income. To the extent thathouseholds participate because they have a higher feltdemand for food than nonparticipants, such householdswould probably have a higher weekly food expenditure inthe absence of the program and, therefore, they have agreater proportion of income freed by participating.^/

26 / The CBO finding of income supplementation throughfood stamp usage (43 percent) approximates previousestimates by other researchers. Bemus et. al.in "Determinants of Household Food Consumption,"The Review of Economics and Statistics, May 1975,reported that household units, surveyed between1968-72, used approximately 14 percent of foodsubsidy income to increase their general purchas-ing power; the marginal propensity to consume(MPC) from this source of income was .86. Hymansand Shapiro, "The Allocation of Household Incometo Food Comsumption," in Five Thousand AmericanFamilies - Patterns of Economic Progress, Vol. II,Institute of Social Research, University of Michigan,1974, found greater income substitution rangingfrom 45 to 65 percent, MPC ranged from .35 to .54.Robert B. Reese, J.G. Feaster, and Gary B. Perking,in Bonus Stamps and Cash Income Supplements,Marketing Res. Rept. 1034, Economic ResearchService, USDA, Oct. 1974, estimated that approx-imately 45 percent of the food stamp bonus wasspent on nonfood items.

39

Impact on Agriculture and the General Economy

In calendar year 1975, the peak year of food stampparticipation and associated federal bonus outlays, thetotal value of food stamps in circulation representedabout 4.4 percent of the total national expenditure onfood ($185 billion). Approximately $4.9 billion inbonus food stamps were issued in 1975, representing animplicit increase in food expenditures not availablebefore the food stamp program. But since it is es-timated that 57 percent of the income freed by bonusfood stamps is used to purchase additional food, theincreased demand for food generated by the food stampprogram was approximately $2.8 billion. The farmer'sshare of retail food sales in 1975 was 42 percent.Therefore, the addition to farm income in 1975, as aresult of the food stamp program was $1.1 billion.This additional income represents 1.1 percent of thegross farm income in 1975.

Analyzing the food stamp program as it interactswith the total economy (e.g., changes in business re-ceipts among industry sectors, gross national product(GNP), and the number of jobs), a recent study concludesthat the food manufacturing sector business receiptswere higher by $589 million in 1972 and by $809 millionin 1974 under the food stamp program than they wouldotherwise have been in those years. The study furtherestimates that the GNP increased by $311 million and$427 million in those two years as a result of thefood stamp program, adding .between 56,000 and 77,000new jobs to the economy. 2̂ 7/

27/ Paul E. Nelson and John Perrin, Economic Effectsof the U.S. Food Stamp Program Calendar Year 1972and Fiscal Year 1974, U.S. Department of Agriculture,Economic Research Service, Agricultural EconomicReport No. 331, July 1976.

40

CHAPTER III KEY STRUCTURAL ELEMENTS OF THEFOOD STAMP PROGRAM

The food stamp program has three major components:income eligibility standards, including definition ofincome and deductions from income; purchase requirements;and allotment levels. Adjustment to any of these keyelements will affect program costs and several programobjectives, including adequacy of benefits delivered, popu-lation eligible for benefits, work incentives, and fairnessin treatment of the population served. This chapter explainsthe key structural elements of the food stamp program sothat the reader may better understand the potential impactof the program alternatives discussed in the next chapter.

CURRENT PROGRAM STRUCTURE

The basic design of the current food stamp program isone common to all income maintenance programs of the nega-tive income tax variety in which a household's benefit isreduced proportionately with increasing income. The foodstamp allotment varies only by household size, while theproportion of the allotted stamps a household must purchaseincreases with increasing income. The bonus stamps re-present the difference between the allotted and purchasedstamps. In general, for every $3.00 increase in income,the household's purchase requirement increases by $1.00.In other words, bonus food stamps are reduced by $1.00 forevery $3.00 increase in income (i.e., the marginal tax rateis approximately 30 percent). Purchase requirements arebased on a household's net income, that is, income afterallowable deductions are subtracted.28/

28/ The most general structure of the FSP is defined as:BS = AS - r (Y-Dg), where:Bg = bonus dollars transferred for household of size s\Ag = basic food coupon allotment or guarantee for

household of size sY = gross countable household incomeDg = deductions from gross income for household of

size sr = marginal tax rate on income or the benefit reduc-

tion rate

41

81-390 O - 77 - 5

The interaction of these components establishes the"break-even" income point, that is, the income level wherea household's benefits are reduced to zero, A 30 percentmarginal tax rate and an allotment of $166 for a family offour mean that any four-person household with less than$553 a month in net income is eligible.29/

Certain limitations in addition to income are placedon eligibility. Households may not have liquid resourcesexceeding $1,500 ($3,000 for households of two or morepersons with at least one member over 60 years of age).Further, any able-bodied person who is not employed orcaring for children must register for and accept suitableemployment; refusal to do so would result in the entirehousehold of which he or she is a member becoming ineli-gible. Modification of any one of these components wouldalter the focus of the program's implicit objectives andoverall costs.

Gross Countable Income

The mean monthly gross income of food stamp householdsin September 1975 was $292. Countable household incomeincludes all income received, or expected to be received,during the period a household is certified eligible toparticipate in the food stamp program. Households arecertified eligible for the program on the basis of theiranticipated income in the forthcoming month, and any change

29/ Administrative regulations establish net income eligi-bility for households based on the higher of the twoconditions: (1) the point at which the coupon allot-ment for that household size equals 30 percent ofincome or (2) the Secretary of Agriculture's incomepoverty guidelines. Recently, for one and two personhouseholds the poverty guidelines have been higherthan that arrived at by multiplying 3.3 times the oneand two person allotment schedule. For example, theallotment for a one-person household was $51 in July1976, so that net income eligibility based on a 30percent marginal tax rate would have been $165 a monthThe poverty guideline for one person households was$245.

42

in income from the certified levels must be reported. Thephilosophy behind this prospective accounting system issimply to make the food stamp program highly responsive tothe food needs of the applicant.

Income as defined for food stamp purposes includes allthe following:

o Compensation for services performed as an employee.

o Net income from self-employment, i.e., gross incomeless the cost of producing that income.

o Payments from annuities, pensions, retirement anddisability benefits, veterans' benefits, workmen'scompensation, unemployment compensation, socialsecurity, and strike benefits.

o Public assistance benefits.

o Payments made on behalf of a household by a personnot considered a member of that household.

o Support and alimony.

o Scholarships, educational grants, fellowships, andveterans' educational benefits.

o Rents, dividends, interest, and royalties.

Several forms of income are explicitly excluded frombeing counted for food stamp purposes:

o Money earned by a child under 18 who is a studentat least half time.

o Irregular income in small amounts, unless it addsup to more than $30 in a three-month period.

o Any gain that is not in money.

o Monies received from insurance settlements, saleproperty (except property related to self-employ-ment), retroactive lump-sum social security or rail-road retirement pension payments, income tax refunds

43

o Loans, except educational loans.

o Payments received under the WIC-Child NutritionProgram.

o Income received by volunteers for services per-formed in various ACTION programs.

The definition of income is a crucial factor in design-ing any reform option. For example, excluding other fed-erally subsidized nutrition benefits would result in anunderstatement of the available food resources to a house-hold. However, attempting to count such benefits wouldsubstantially increase administrative complexities.

Exclusions also result in providing unequal food stampbenefits to otherwise similar households. For example,different benefits are provided households with similargross incomes but with different withholdings on theirfederal or state taxes. A household over-withholding wouldqualify for larger benefits (because of the allowance oftax deductions) and would not have the tax refund countedin income upon repayment of the over-withholding. Theamount would be counted against the asset test. However,such a household would likely have a high in-out partici-pation rate, complicating determination of the proper levelof food stamps benefits.

The definition of countable income also alters theprogram's effect in increasing food expenditures or in-creasing the proportion of income freed for non-food pur-hases. Including in-kind nutrition benefits in thehousehold's income definition reduces the amount of thefood stamp transfer used to free income for non-foodpurchases.

44

Deductions

Households may deduct a number of household expend-itures before determining eligibility benefits. Currently,households are allowed to deduct the following:

o Ten percent of their income from compensation forservices performed as an employee, or trainingallowances, not to exceed $30 per household permonth .

o Mandatory deductions from earned income that arenot elected at the option of the employees, suchas local, state, and federal taxes, social securitytaxes, union dues.

o Payments for medical expenses when costs exceed$10 per month.

o Payments for child care or adult care whennecessary for the household member to accept orcontinue work.

o Unusual expenses associated with disaster orcasualty losses.

o Tuition and mandatory fees assessed by an ed-ucational institution.

o Payments for rent or mortgage plus utilities thatare more than 30 percent of a household's incomeafter all other deductions.

The allowance or disallowance of specific income de-ductions affects the adequacy of benefits, participationrates, food-income tradeoffs, target efficiency of thebenefits, and, most important, program costs. It has beenaccepted, usually without debate, that eliminating thecurrent practice of itemizing special deductions and re-placing them with a standard or flat deduction would reduceadministrative complexities and therefore costs.

45

11-390 O - 77 - 6

The 1975 USDA survey of food stamp households foundthe average level of all deductions to be approximately$74. This represents the sum of all deductions averagedacross all households, regardless of whether the house-hold actually claimed deductions. For the 82 percent ofall households that actually claimed deductions, theaverage deduction was $91.

For households with an elderly person (age 65 or over),the average level of itemized deductions was approximately$46. Excluding elderly households with zero deductionsincreased the average deduction for elderly householdsclaiming deductions to $62. With constant income andhousehold size, deductions tend to decrease with in-creasing age of the household head.±[0/

30/ The cross-sectional data from the September surveywere fit relating deductions to household character-istics. The coefficient on gross income, in thisrelationship, indicated that for every one dollarincrease in income, deductions increased by over 27cents. A negative coefficient on the age variablein this equation suggested that for every year in-crease in the age of the head of the household, deduc-tions decline about 77 cents. Further specificationto improve explanatory power of this equation couldslightly change absolute value of the coefficients,but it is not anticipated that the sign of the co-efficients would be altered.

The equation follows:Deductions = 55.80 + 7.86 HHS - .14 HHS2 - .77 AGE + .27 GY(+-statistics) (7.22) (1.36) (18.97) (62.04)WHERE: HHS = Household Size

AGE = Age of Head of HouseholdGY = Gross Income

R2 = .2986 S.E. = 72.25 F RATIO = 1195.0

46

Deductions by household size increase in a nonlinearfashion. (See Table 10.) While smaller households tendto have smaller deductions, deductions tend to level outaround three- and four-person households and even showsigns of declining thereafter. Since the majority ofparticipating households are one and two persons, andsince the average deduction for these smaller householdsis less than the overall average, any single standard de-duction set at or above the overall average will tend toredistribute benefits away from larger households and re-direct them toward smaller, more elderly households.

TABLE 10. AVERAGE DEDUCTION BY HOUSEHOLD SIZE,USDA SURVEY (SEPTEMBER 1974)

Average DeductionHousehold Size Overall

1 $48.862 70.243 85.134 90.865 87.656 86.187 86.968 90.999 72.47

10 or more 55.75

Overall $74.12

SOURCE: CBO Tabulations from USDA September 1975 FoodStamp Survey tape. Includes households withdeductions greater than gross income.

47

Itemized deductions vary by income of the household;they also vary markedly between regions of the countryfor households with the same income. For all householdswhose gross income was less than the poverty level, de-ductions averaged $55.36. (See Table 11.) This rangedfrom a high of $81.70 for poor households in the NewEngland states to a low of nearly $46.00 for poor house-holds in the West Central states. Almost all of thisvariation within the same income class is explained byvarying shelter costs in the regions. For poor householdsin the New England region, shelter costs reached over$155 in September 1975, in the Southeast region, thecomparable figure was only $82.

Setting a flat standard deduction, based on theaverage as it exists over all participating incomeclasses, would redistribute benefits to households belowpoverty on the average and redistribute benefits frompoor households in the Northeast to households in theSoutheast and West Central regions.

Households with larger gross incomes now qualify forthe program on the basis of large income deductions. (SeeTable 12.) Average total deductions for households withincome above 150 percent of the poverty line reachednearly $242 a month, 4.4 times the level for poor house-holds and 3.3 times the national average. (See Table 11.)

Proponents of indexing standard deductions for priceincreases argue that deductions inflate over time, andthat, in order to maintain the same relative impact as inthe current program, deductions should also be inflatedfrom the September 1975 levels. The difference betweenthe overall avsrages for September 1975 ($74.12) and aprevious USDA survey conducted in November 1973 ($56.05)—/represents growth in deductions brought about by inflationfactors, the changing profiles of the participating pop-ulation claiming deductions, and varying commodity specificinflation rates. Table 12 provides a breakdown of theoverall average deduction in September 1975 by type ofitemized deductions.

31/ "Food Stamp Program - A Report in Accordance withSenate Resolution 58," FNS, USDA, July 21, 1975, p. 13.

48

TABLE 11. AVERAGE TOTAL DEDUCTIONS BY POVERTY STATUSAND BY REGION, SEPTEMBER 1975

OverallRegion Average

ALLAREASAVERAGE

NewEngland

Mid-Atlantic

Southeast

Mid-West

Western

West-Central

$74

123

71

70

66

87

58

.12

.62

.78

.27

.31

.99

.72

PERCENT OF POVERTYLess Than 1.0- 1.26-Poverty 1.25 1.50 1.50+

$55.

81.

52.

54.

52.

58.

45.

36

70

07

99

09

15

97

$103.

126.

101.

113.

89.

94.

117.

51

36

73

63

80

35

25

$158.43

175.48

149.68

158.84

152.18

159.54

167.70

$241

321

216

260

218

224

228

.61

.27

.92

.74

.84

.43

.73

SOURCE: CBO tabulations from USDA September 1975 FoodStamp Survey tape. Regions are USDA, FoodNutrition Service, regions.

Mid-Atlantic:

Southeast:

Midwest;

Western:

West Central:

New England:

Delaware, District of Columbia, Maryland,New Jersey, New York, Pennsylvania,Puerto Rico, Virginia, Virgin Islands,West Virginia

Alabama, Florida, Georgia, Kentucky,Mississippi, North Carolina, SouthCarolina, Tennessee

Illinois, Indiana, Iowa, Kansas, Michigan,Minnesota, Missouri, Nebraska, Ohio,Wisconsin

Alaska, Arizona, California, Guam, Hawaii,Idaho, Nevada, Oregon, Samoa, TrustTerritories, Washington

Arkansas, Colorado, Louisiana,Montana, New Mexico, North Dakota,Oklahoma, South Dakota, Texas,Utah, Wyoming

Connecticut, Maine, Massachusetts,New Hampshire, Rhode Island, Vermont

TABLE 12. ITEMIZED DEDUCTIONS PER EOUSEKOLD, UNITED STATES___ AND REGIONAL AVERAGES

Overall Hew Mid- WestAverage England Atlantic Southeast Midwest Western Central

1. Shelter $46.50 77.85 49.16 36.76 44.86 59.54 28.87

2. Mandatory 10.09 21.79 9.62 9.71 8.35 11.12 7.69(taxes,etc.)

3. Medical 7.42 10.58 4.04 12.72 5.28 5.14 10.73Expenses

4. Work 5.72 7.60 5.46 6.84 4.36 5.68 6.08Allowance

5.

6.

7.

8.

9.

Child Care

SchoolTuitionand Man-datoryFees

Alimony-Support

CasualtyLosses

Other

TOTAL

2.

1.

0.

0.

1,

74

30

21

.35

. 31

.12

.12

.99

3.22

1.47

.05

.07

123.62

1.64

1.39

.23

.20

.04

71.78

2.22

.55

.60

.07

.80

70.27

2.04

.71

.02

.37

.82

66.31

4.43

1.57

.02

.35

.14

87.99

2.67

1.34

.34

.55

.45

58.72DEDUCTIONS a/

SOURCE: CBO tabulations from USDA September 1975 Food Stamp Survey tape.Tabulations include Puerto Rico and outlying territories. Numbersmay not add up to totals because of rounding.

a/ Figures may not add to exact totals because of rounding.

The shelter deduction constitutes the largest shareof all itemized deductions, making up nearly 62 percentof the total in September. Work-related deductions (e.g.,taxes, work allowances, and child care expenditures) re-presented the second largest proportion (24 percent) oftotal deductions claimed.

Shelter deductions make up the largest percentage ofall deductions claimed by households falling below poverty(77 percent), while only 27 percent of all deductionsclaimed by persons with gross income over 150 percentpoverty are attributable to the shelter deduction. Eighty-seven dollars, or 35 percent of the total deductionsclaimed by high income recipients, is attributable tomandatory payroll deductions.

50

In total, the level of deductions increased 33.8 per-cent between November 1973 and September 1975, representingan annual growth rate of 16.4 percent. During this period,the overall Consumer Price Index increased at an annualrate of about 9.5 percent. Deductions, therefore, increasedabout 70 percent faster than the overall CPI. While thismay overstate the actual increase in deductions because ofchanges in populations surveyed, it is clear that deductionshave risen over time. Therefore, proposals to replaceitemized deductions with a non-price-inflating, standard-ized deduction could result in a redistribution of benefitsover time and a reduction in both the eligible populationand the number of participants.

Allotment Level

The food stamp allotment level establishes the maximumbenefit level for a family. The allotment varies only byfamily size according to the cost of a "nutritionally ade-quate diet." The USDA develops four family-food plans:a thrifty plan, a moderate plan, a low-cost plan, and aliberal plan. The plans represent amounts of foods ofdifferent types that families might buy to provide nutri-tious diets at different levels of cost. Before January1976, the USDA established the food stamp allotment levelbased on the retail price of a food plan known as the"Economy Food Plan." Beginning in January 1976, this waschanged to the thrifty food plan. The economy food planwas a basis for establishing the original provery guide-lines as developed in 1964.32/

As a basis for setting food coupon allotments, theeconomy food plan was subject to numerous complaints andmajor law suits. In the most recent of these law suits (Rod-way v. United States Department of Agriculture, 514 F. 2d.809, D.C. Cir., June 12, 1975), the court ruled that the sys-tem of establishing coupon allotments would be sustained only

32/ Mollie Orshansky, "Counting the Poor: Another Lookat the Poverty Profile," Social Security Bulletin,U.S. Dept. of Health, Education and Welfare, January1965.

51

if the Secretary could show that the system deliveredcoupons to substantially all recipients in amountssufficient to allow them to purchase a nutritionallyadequate diet. Following this ruling, the Secretaryproposed three possible systems for establishing couponallotments. A final decision was made in December 1975,and the new system of coupon determination became effec-tive with the normal semiannual adjustment in January1976.

The revised coupon allotment system is based on thenew thrifty food plan, with different economies of scalefor larger size families. The increase in benefits tolarger size households followed from the Rodway case,which called for the delivery of coupons sufficient toallow all recipients to purchase a nutritionally adequatediet.

Costs of the USDA plans vary by family type. (SeeTable 13.) Specifically, the cost of the thrifty foodplan for a family of four with children between 6-8 and9-11 years of age is now used to establish the basicfood stamp allotment for four-person households. Forother size households, the four-person household baselevel is adjusted by various household-size, economy-of-scale factors.

One proposal has been advanced that would increasethe allotment levels equivalent to the low-cost foodplan. Such proposals increase the adequacy of the bene-fits but result in large benefit notches. A "benefitnotch" appears when one additional dollar of earningsresults in more than one dollar loss in benefits. Thisproposal will be discussed in the following chapter.

Low income households of similar size, but otherwisedissimilar because of a different sex-age composition and/or with special nutritional needs arising from infections,metabolic disorders, chronic diseases and abnormalities,are treated similarly. In order to assure nutritionaltreatment comparable to nutritional need, allotmentscould be varied by a household's sex and age composition.If the age-sex specific characteristics of food stamphouseholds in September 1975 were used to develop allot-ments instead of the hypothesized four-person household

52

now used, the allotments would decline to $155, nearly$12 less per month than the current program. Providingapplicant food stamp households an allotment level deter-mined by the specific sex and age characteristics of thehousehold would significantly reduce federal bonus costsand presumably come closest to targeting benefits onspecific nutritional needs.

Benefit Reduction Rate

Amendments to the Food Stamp Act (Public Law 91-671)in 1973 required that the purchase price represent a"reasonable investment" on food rather than "normal foodexpenditures." The distinction has never been made be-tween whether this was to be a reasonable investmentmeasured by gross income or by net income standards. Ingeneral, for every three dollars increase in income,benefits are reduced by one dollar..33/ Purchase require-ments are set according to income ranges or steps; re-quirements vary widely when calculated as a percentage ofa household's net income. Because of varying deductionsby income class, they also vary when calculated on thebasis of gross income. Purchase requirements for house-holds with the same net income are slightly higher forlarger households. For each household size, there is amaximum purchase price guaranteeing a minimum benefit.

The 1971 amendments allowed a household to buy eitherall, three-quarters, half, or one-quarter of its totalmonthly allotment. The amount of bonus stamps receivedis prorated according to the amount purchased.

33/ This low marginal tax rate of 30 percent finds itsbasis not in a desire to forestall unacceptable workresponse normally thought to occur with high marginaltax rates, but in a desire to set the food couponpurchase requirement at a level representing whatCongress felt was the maximum amount of money a house-hold should be required to spend on food. Food StampAct of 1964, Section 7(b) U.S.C. 2016(b), as amendedby the Act of January 11, 1971, Public Law 91-673, 84Stat. 2048.

53

TABLE 13. COST OF FOOD AT HOME ESTIMATED FOR FOODPLANS AT FOUR COST LEVELS, JULY 1976,U.S. AVERAGE */

Sex-age groups Thriftyplan b/Dollars

Cost forLow-costplanDollars

1 monthModerate-cost jilanDollars

LiberalplanDollars

FAMILIESFamily of 2: £./20-54 years.... 97.7055 years andover 87.20

Family of 4:Couple, 20-54 yearsand children--1-2 and 3-5years 138.306-8 and 9-11years 166.70

INDIVIDUALS /̂Child:7 months to1 year1-2 years3-5 years6-8 years9-11 years....

Male:12-14 years...15-19 years...20-54 years...55 years andover

Female:12-19 years . . .20-54 years...55 years andover

PregnantNursing

19.7022.4027.1034.6043.30

46.3050.9048.90

43.30

41.3039.90

36.0049.9053.00

127.80

113.20

179.40

216.70

24.3028.8034.4044.7055.80

59.6065.8064.30

56.40

53.2051.90

46.5064.2068.10

160.50

140.50

224.00

272.20

29.7035.5042.6056.1070.20

74.7082.6081.20

70.20

66.0064.70

57.5079.2084.80

193.30

168.50

269.50

327.50

35.2042.4051.4067.4084.40

89.7099.6098.20

84.70

79.0077.50

68.5094.50

101.20

54

Notes for Table 13.

SOURCE: U.S. Department of Agriculture, Consumer andFood Economics Institute, Agricultural Re-search Service, CFE (Adm.) 326.

ji/ Assumes that food for all meals and snacks is pur-chased at the store and prepared at home. Estimatesfor each plan were computed from quantities of foodspublished in the Winter 1976 (thrifty plan) and Winter1975 (low-cost, moderate-cost, and liberal plans)issues of Family Economics Review. The costs ofthe food plans were first estimated using pricespaid in 1965-66 by households from USDA's House-hold Food Consumption Survey with food costs atfour selected levels. These prices are updated byuse of "Estimated Retail Food Prices by Cities" re-leased monthly by the Bureau of Labor Statistics.

b/ Coupon allotment in the Food Stamp Program based onthis food plan.

£/ Ten percent added for family size adjustment.~ See footnote d.

d./ The costs given are for individuals in 4-person~ families, taking into account food-buying economies

of large families. For individuals in other sizefamilies, the following adjustments are suggested:1-person—add 20 percent; 2-person—add 10 percent;3-person—add 5 percent; 5-or-6-person—subtract 5percent; 7-or-more-person—subtract 10 percent.

55

Actual purchase requirements, as a percent of netincome, vary from 5 percent for one-person householdswith low monthly income to 29 percent for large sizehouseholds approaching net income eligibility cut-offlevels. Overall, the average purchase requirement wasabout 25.6 percent of net income in September 1975. Asa percentage of gross income, the average purchase re-quirement was about 19 percent. (See Table 14.)

TABLE 14. AVERAGE AMOUNT AND PERCENTAGE OF GROSS ANDNET MONTHLY INCOME SPENT ON FOOD STAMPS BYALL PROGRAM HOUSEHOLDS IN SEPTEMBER 1975,BY TYPE OF INCOME AND GROSS MONTHLY INCOME

GrossMonthlyIncome G:

Under

100 -

200 -

300 -

400 -

500 -

600 -

750 -

1,000

$100

199

299

399

499

599

749

999

+ 1

OVERALLAVERAGE

Type of Incomeross Net

(dollars)

39

159

247

346

445

545

660

835

,170

292

24

115

185

268

343

414

497

616

826

220

PurchaseRequire-ment

1

25

46

71

96

114

142

178

214

56

Proportionof IncomeSpent onFood StampsGross Net(percent)

2

15

18

20

21

20

21

21

18

19

.6

.7

.6

.5

.6

.9

.5

.3

.3

.4

4

21

18

26

28

27

28

28

25

25

.2

.7

.6

.5

.0

.5

.6

.9

.9

.6


56

For the population in general, it is estimated thatthe average family spends about 20 percent of gross in-come for food. The lowest income families (under $3,000a year) spend up to 50 percent of their income on food.It has generally been accepted that widely varying pur-chase requirements should be replaced with a standardrequirement.

All major alternatives recently advanced have in-cluded a standard purchase requirement ranging between25 and 30 percent of gross income. Standard rates willinteract with deductions and allotment levels to changethe income supplementation effect of the current programmarkedly.

57

CHAPTER IV FOOD STAMP PROGRAM BUDGET OPTIONS

Food stamp program issues for fiscal year 1978 andbeyond center on whether the program should be designedto emphasize broad objectives of (a) increasing foodconsumption, (b) providing some minimum level of incomesupport effectively and efficiently, or (c) continuingthe current blend of these objectives. The various foodstamp reform proposals before Congress generally aredesigned to serve one of these objectives.

Within the context of these broad program objectives,the following important policy criteria are used toevaluate proposed changes. They are, of course, givendifferent weights by different decision-makers. Thesecriteria include: assuring adequacy of benefits to allparticipants; concentrating benefits on the most needy;fairness in treating similar households similarly; fair-ness in treating dissimilar households differently; theassurance of work incentives; maintenance of adminis-trative simplicity; and limiting budget costs.

In general, the major budget options relatedirectly to the different perceptions of program ob-jectives. The options cover the following broad areas:

o Maintaining the status quo. Continuation ofcurrent program design and administration isa possible alternative. In the reform debate,however, it has usually been assumed a priorithat some change should take place.

o Current program refinement. This alternativeentails what some refer to as marginal changes.The basic program design remains unaltered.Modifications of basic eligibility standardsand tightening of program rules result in nosignificant change in the current program'smix of food consumption and income objectives.

o Food consumption emphasis. In the continuumof alternatives, this approach maintains thehighest degree of food consumption from dollars

59

transferred to participating households. Thisoption also significantly reduces program costsby both narrowing the eligible population andlimiting participation.

o Income support emphasis through elimination ofthe purchase requirement. This alternativecontains an element of increased food consumption.The option results in some participants havingthe ability to use relatively more freed incomefor nonfood purposes. The option removes a majordisincentive to non-participation in the currentprogram.

o Cash-out food coupons. In the continuum ofalternatives, this proposal provides recipientswith the greatest personal freedom to purchaseany goods and services within the household'spersonal preference, constrained only by theamount of the cash transfer and other householdincome. The proposal alters the program froman in-kind welfare approach to a direct cashapproach. While increasing the cash income ofthe recipient the proposal also eliminatesprogrammatic incentive for increased food con-sumption. The cash-out option is not normallyconsidered a food stamp option (except to theextent that it suggests a complete eliminationof the program), but it should be considered inthe broader debates associated with variouscomprehensive welfare restructuring proposals.Such options will be discussed in a forthcomingCBO paper concerning total welfare reform.

Maintaining the Status Quo: Current Policy EstimatesFY 1977-1982

To provide a general reference point from which tojudge the various reform proposals, CBO has developedlong range projections of the food stamp program. Usingits economic assumptions, 32/ CBO has developed a model

32/ "Five-Year Economic Assumptions," CBO EconomicAssumptions Panels, August 3, 1976.

60

of factors influential in determining past programgrowth in order to project future program growth.

Key variables influencing program growth areoverall price inflation and food price increases, un-employment rates, income growth, growth in other publicassistance programs, and overall participation amongeligible households. Table 15 summarizes the keyeconomic assumptions used in the development of thecurrent policy estimates.

TABLE 15. CBO ECONOMIC ASSUMPTIONS, CONSUMER PRICEINDEX AND FOOD SECTOR CONSUMER PRICE INDEXOverall CPI CPI FoodPercent Percent

Calendar Change From Change From UnemploymentYear Previous Year Previous Year Rate

1975 Actual

1976 Projected

1977

1978

1979

1980

1981

1982

9.1

5.7

5.4

5.6

5.4

5.5

5.7

6.0

8.5

3.1

2.2

2.8

5.3

5.5

5.8

6.0

8.5

7.3

6.4

5.8

5.3

4.9

4.5

4.5

SOURCE: "Five-Year Economic Assumptions," CBO EconomicAssumptions Panels, August 3, 1976.

As indicated, food price increases are assumed todecline from growth rates of the recent past. Food priceassumptions decline from an average annual increase ofslightly over 8 percent in 1975 to a 3.1 percent increasein 1976. A further decline is projected for 1977, witha moderate increase thereafter. Over the next seven years,food prices are assumed to increase at an average annual

61

rate of approximately 4.3 percent, a figure comparable tothe average rate of food price inflation over the period1965 to 1972. The economic assumptions used to projectthe number of food stamp recipients also include a de-clining unemployment rate, with the overall rate declin-ing to 4.5 percent by the end of 1980.

Given these assumptions, it is not surprising thatthe number of food stamp recipients shows a general de-cline throughout this period, almost entirely due to thedeclining number of nonpublic assistance food stamp re-cipients. Average monthly participation declines fromapproximately 18.5 million in 1976 to 15 million in 1982.(See Table 16.)

Because of declining food prices through 1978, andbecause of the declining number of recipients, the costof the food stamp program would be about $5.4 billion infiscal year 1978, down nearly $300 million from 1976levels.ji^/ With moderately increasing food prices after1978, the cost of the food stamp program begins to in-crease again despite declining participation, and reachesapproximately $5.9 billion by 1982.

Table 16 summarizes the status quo budget option.These numbers are indicative, not predictive. Long rangeprojections are highly speculative, given the vagaries ofweather, changing agricultural policies, and worldwideprivate speculation in U.S. and world food markets. Tothe extent that world and domestic food and agriculturalpolicies over the next five years can facilitate stabledomestic and international food prices, and to the extentthat normal weather conditions exist worldwide, food stampprogram growth and costs would stabilize under the CBOassumptions.

33/ The CBO current policy estimate of $5.383 billionin FY 1978 compares to the administration's es-timate of $5.756 billion under slightly higher un-employment and price inflation factors. CurrentService Estimates for Fiscal Year 1978, ExecutiveOffice of the President, Office of Management andBudget, November 1976.

62

TABLE 16. SUMMARY, FOOD STAMP PROGRAM OUTLAYS. CURRENT POLICY ESTIMATES. FOR FISCAL YEARS 1977 TO 1982

FiscalYear

AverageParticipation(millions

of persons)

Total BonusCosts

(billionsof dollars)

AdministrativeCosts

(billions ofdollars)

TotalCosts

(billionsof dollars)

1976P 18.498

197719781979198019811982

17.46516.50315.93415.44415.03014.989

$ 5.265

5.1805.0405.0385.0875.1985.534

$ . 366

.370

.343

.340

.347

.355

.382

$ 5.632

5.5505.3835.3785.4345.5535.916

P = Preliminary

SOURCE: Estimates developed from CBO economic assumptions of August 3, 1976.

63

Despite the potential for a lessening of programgrowth, the current program would continue to fostercertain inequities and inefficiencies. The programwould continue to allow participation by higher incomehouseholds.

The status quo reflects an average participationrate in the eligible population of approximately 50-55percent. This participation rate is not expected tochange dramatically, given that nationwide publicityand outreach efforts have already made the programknow virtually to all eligible persons, and given thenormal deterrents to participation associated with theparticular design of the program, consumer costs, re-cipients' felt needs, and program stigma.

Refinements of the Current Program

Three proposals would refine the current food stampprogram, retaining its basic characteristics. These areS. 3136, passed by the Senate in April 1976; H.R. 13613,approved by the House Agriculture Committee in August1976; and the Ford Administration revised regulations,now blocked by court injunction. All these proposalsattempt to improve program administration through simp-lified deductions and a redirection of current benefitsto the lowest income groups.

S. 3136. The National Food Stamp Reform Act of 1976would modify key program parameters but continue thepresent program mix of food consumption and incomesupplementation objectives.

The program refinement in S. 3136 would set a "cap"on net income eligibility at the official poverty lineafter a series of special deductions. It would alsoeliminate categorical eligibility provisions for AFDC andSSI recipients. The bill would set a flat 25 percent ofnet income as the purchase requirement. S. 3136 wouldeliminate the current program practice of allowing aseries of special deductions and replace them with aflat $100 a month deduction (except for households inoutlying territories, where the level was set at $60).

In addition to the standard deduction, householdscould deduct $25 a month if an elderly person (60 years

64

of age or older) resided within the household, $25 amonth if the household earned more than $150 a month,and all federal, state, and local income taxes andsocial security taxes. All deductions would be adjustedsemiannually based on the overall price change. Theproposal would not change the basic allotment. Finally,the proposal would require that the household's incomeover the previous 30 days, before applying for foodcoupons, be used to determine eligibility and benefitlevels .

H.R. 13613. The Food Stamp Act of 1976 (H.R. 13613)never received full House consideration before the 94thCongress adjourned. The proposal, similar to S. 3136,would eliminate categorical eligibility and would set netincome eligibility based on the official federal povertyguidelines after allowing a standard deduction based onhousehold

In addition, families with a member over 65 or blindor disabled would receive an additional $25 deduction.Working families could deduct up to $30 a month for workexpenses and up to $75 a month for dependent expenses.Additional deductions would include federal, state, andlocal taxes, and social security taxes.

H.R. 13613 would establish a purchase requirementof 27.5 percent of net income for all households inwhich every member was not either blind, aged, or dis-abled. These latter households would receive a check(as opposed to stamps) equivalent to the value of thedifference between their monthly food stamp allotmentand an implicit purchase requirement, also establishedat 27.5 percent of net income.

34/ H.R. 13613 would establish the following standarddeductions for various household sizes:1-person household . . . $452-person household . . . $553-person household . . . $654-person household . . . $755-person household . . . $806-person household . . . $85

65

Beginning in fiscal year 1978, states would be re-quired to assume 2 percent of the cost of bonus stampswithin their states (including the implicit stamps ascashed out to the elderly, blind, and disabled). H.R.13613 would render households containing a person in-volved in a labor dispute ineligible, unless the house-hold had been eligible for food stamps immediately priorto the dispute. Finally, the bill would make allstudents 18 years of age or older and enrolled in aninstitution of higher education ineligible, unless sucha person was head of a household.

Administration Regulations. The regulations pro-posed by the Ford Administration would establish netincome eligibility at the poverty line, a standard$100 deduction for all households ($125 if the householdcontains an elderly individual), and a purchase require-ment of 30 percent of net income. The regulations requirethat a household's income over the previous 90 days beaveraged before testing against the net income eligibilitylimits. However, enforcement of these regulations hasbeen enjoined Trump v. Butz, United States District Courtfor the District of Columbia, June 18, 1976.

Program and Budget Impacts. Figure 7 depicts themajor characteristics of these three proposals. For afour-person household in July 1976, all three proposalswould set net income eligibility limits at $458, downfrom $553 in the current program. Gross income eligi-bility, however, could vary from $558 for a workingfamily under the administration regulations to as highas $637 under H.R. 13613, between 21 and 40 percenthigher than the poverty cut-off level. One- and two-person households would be unaffected by the net incomeeligibility standard at the poverty level, since suchstandards currently apply for these households.

Because the average deduction in September 1975 was$75 for all households, any provision allowing a flat$100 standard deduction for all applicant households, aswould S. 3136 and the proposed regulations, would in it-self increase program costs and skew the proportion ofbenefits toward the lowest income brackets.

66

Figure 7Comparison Food Stamp Program—Current Program — ProposedFord Regulations — S. 3136 Food Coupon Allotment — NetIncome Eligibility — Purchase Requirement Household of 4;48 States and D.C. — July 1976

H rFood Coupon Allotment July 1976

Current Program Proposed RegulationsS. 3136 and HR. 13613

Ford Administration's ProposedCoupon Purchase Requirement hMaximum

Allowable| Net |

[Monthly (Monthly.Income (IncomeS. 3136, ICurrentHR.13613 Program

I ($553)

I MaximumAllowable

I Net

'pord[Regulations

S. 3136 CouponPurchase Requirement

Current Program CouponPurchase Requirement

i i i L_i i i$100 $200 $300 $400 $500

NET MONTHLY INCOME (GROSS MINUS DEDUCTIONS)$600

For the majority of households, under S. 3136 thefood stamp purchase requirement would be less than thatof the current program, higher under the regulations,and about the same under H.R. 13613.

Under S. 3136, at the net income eligibility cap($458 per month), a family of four would pay $114.50 forfood coupons worth $166, for a net transfer of $51.50.A similar four-person household making one more dollar($459) would receive zero benefits, a loss or a "notch"of $51.50 for the month, $618 for the year, or 11 per-cent of the household's pre-breakeven income. The notchis least under the regulations and about the same as thecurrent program under H.R. 13613.

67

At what point this notch decreases work incentiveis not clear. The current program has a $24.00 monthlyor $288 annual notch, 4 percent of the average four-person household's pre-breakeven income. S. 3136appears to lower a household's incentive to move out ofthe program until it can be assured, at a minimum, of anet increase in its income of about 7 percent that wouldoffset the loss created by the notch.35/

Any proposal, such as these three, which sets a capon income eligibility at the poverty line, but does notincrease the benefit reduction rate to at least 36 per-cent, will create a notch. (Increasing the benefit re-duction rate, and thereby increasing the coupon purchaserequirement, however, could reduce participation amongneedy households.) Further, because of the requirementthat the basic allotment be adjusted semiannually whilethe income eligibility standards be adjusted annually,the possibility exists for widening the notch at thesemiannual updates. More critical is the fact that thenotch will widen or decline depending on the relation-ship between food price inflation (which adjusts theallotment levels) and the overall Consumer Price Index(which adjusts the poverty levels).

Overall, S. 3136 would make ineligible approximately1 million current higher income recipients. For the re-maining participants, benefits would increase by over 12percent. The proportion of benefits going to householdsin poverty would increase from 86.4 percent to 88.8 per-cent. Increased benefits would be redistributed towardlower income households, public assistance households,and smaller size households. The proposal would increasethe proportion of benefits going to poor households inthe South, West Central, and Midwest regions and decrease

35/ It should be noted that higher notches would existin Hawaii, Alaska, and Puerto Rico under S. 3136,based on the allotment guarantees and the officialpoverty threshold levels in these areas. Theserange from $650 and $993 on an annual basis inPuerto Rico and Alaska, respectively, to over$1,082 in Hawaii.

68

the proportion of benefits received by poor householdsin the Northeast. It is estimated that approximately1.5 million (36 percent) of the pre-transfer^./ poorhouseholds would be made non-poor by the bonus transfers.

Relative to the current program mix of food con-sumption and income supplementation, participating house-holds on the average would find slightly more freed in-come for non-food purchases. Approximately 16.3 percentof the participating household's gross income would beused to purchase food stamps, down from 19.4 percent inthe current program.

Finally, the Senate proposal would increase federaloutlays by over $600 million, bringing the total cost ofthe food stamp program to nearly $6.0 billion in fiscalyear 1978 and to $6.5 billion by fiscal year 1982.

The House Committee bill (H.R. 13613) would make in-eligible approximately 138,000 students and 2,000 house-holds affected by strikes. In addition, about 1.2million current high income recipients would be removedfrom the program. However, an increase in participationof about 600,000 elderly, blind, and disabled personswould result from the bill's cash-out provision. Approx-imately 91 percent of the total benefits would be dis-tributed to households below poverty. Average benefitsfor current recipients remaining eligible would increaseby approximately 5.5 percent.

Because of the special cash-out provision, benefitswould be redirected toward more elderly, smaller sizehouseholds. The cash-out provision for the select house-holds would also provide an incentive for elderly house-hold members to form their own households. Excluding thecash-out provision results in a slight increase in benefitsgoing to larger size households. The proposal wouldchannel benefits away from households in the New Englandand Eastern states and toward households in the South,West Central, and Midwest regions. About 1.2 million(25 percent) of the pretransfer poor households would beremoved from poverty.

36/ Pre-food stamp but post-other cash income transfers.

69

Relative to the current program, no significantchanges would be observed in the proportion of benefitsgoing toward food consumption and income supplementation.There would be some increase in general income supple-mentation for the elderly, blind, and disabled qualifyingfor the cash-out provision.

Federal costs under the proposal would be reducedby approximately $34 million in fiscal year 1978, butstates would have increased outlays totaling about $100million. By fiscal year 1982, federal outlays would in-crease to $90 million, and state outlays would increaseby approximately $115 million.

The regulations proposed by the Ford Administrationwould reduce participation by nearly 3 million persons.For the remaining participants, benefits would increaseby over 6 percent. Approximately 94 percent of the totalbenefits would be redistributed to households below pov-erty. Benefits would be redirected toward the elderly,public assistance, and smaller size households. Theproposal would focus aid on the long-term needy becauseof the provision that would base eligibility on thehousehold's average income over the previous 90 days.The proportion of benefits would again be redistributedfrom the Northeast to South, Midwest and West Centralregions.

The proportion of income that recipients would berequired to spend for food stamps would remain unchangedfrom the current program.

The Administration's proposed regulations would re-duce program outlays by about $500 million in fiscal year1978 and $800 million in fiscal year 1982. While theinitial savings are significantly less than the estimatedsavings in the President's budget for fiscal year 1977,allowing for no inflation adjustment in the $100 standarddeduction, there is an increase in savings in outlyingyears. (S. 3136 and H.R. 13613 do index the standard de-duction. )

70

Food Consumption Emphasis Proposals

A third set of proposals stresses improving theprogram's food consumption cost-effectiveness. Asunder the already discussed options, the purchase re-quirement, the basic benefit level, and the definitionof household income would be modified.

Income would include other in-kind nutrition benefits,To make the program responsive to immediate food needs,consumption-emphasis options would include a prospectiveaccounting system for determining eligibility. And,despite greater administrative expense, eligibilitydetermination could be based on medically determinednutritional needs.

Adjustments to various program components, alongwith expanding nutrition education and limiting foodcoupon usage to selected foods, would result in improvingthe program's nutritional cost effectiveness for thosewho choose to participate. Balancing the food consump-tion impact of the bonus transfer against provisionsthat limit participation is a key issue under any food-consumption option. For those who do choose to partic-ipate, however, there would be an admission of feltnutritional need and an actual addition to the family'sfood expenditures.

Two major bills were introduced in the 94th Congressthat explicitly directed the focus of the program towarda more specific food-consumption objective and limitedprogram costs. H.R. 13375 and S. 1993 are prototypesof program modifications designed to limit programparticipation and costs and to increase its food-consumption effectiveness.

H.R. 13375 and S. 1993 would limit participation tohouseholds with net incomes at or below the officialfederal poverty levels. Both bills would eliminateitemized deductions, with the exception of providing a$25 a month deduction for households with at least oneelderly member. H.R. 13375 would also allow householdsto deduct all federal, state, local, and social securitytaxes. Both bills would also count as income all federalin-kind payments for food and housing.

71

S. 1993 would set the basic food stamp allotment atthe cost of the low cost food plan (approximately 30 per-cent higher than the thrifty food plan). Though notspecifically so stating, H.R. 13375 would apparently fixthe allotment as that cost determined to be the lowestcost diet consisting of highly nutritional foods, assumedhere to be the cost of the "Thrifty Food Plan."

Program modifications similar to S. 1993 and H.R.13375 would significantly reduce budget costs and makeineligible a large proportion of the current programparticipants. S. 1993 would make ineligible over 22percent of the currently participating households,approximately 4.2 million persons, while H.R. 13375would remove 17 percent of the currently participatinghouseholds, over 3 million persons. While H.R. 13375would reduce average benefits going to participatinghouseholds by over 17 percent, S. 1993, because itestablished the low cost food plan as the basic allot-ment, would increase the average benefit for the remain-ing households by over 24 percent.̂ ./

Both proposals significantly redistribute 98 per-cent of all program benefits to households classifiedbelow poverty. Both proposals also shift the proportionof benefits away from smaller size households to largerones. While under the current program over 22 percentof all benefits go to one- and two-person households,this figure would range between 16 and 17 percent underH.R. 13375 and S. 1993. Finally, it is estimated thatH.R. 13375 would reduce the relative share of benefitsgoing to elderly households and current public assistancehouseholds, but it would increase the proportion of bene-fits going to working poor households from 18.8 percent toabout 21.5 percent. S. 1993 would increase the proportionof benefits going to elderly households and public assis-tance households, but it would slightly decrease the rela-tive share going to the working poor.

36/ These estimates do not include the provisions ineach bill that would count other in-kind nutritionbenefits received in the definition of a household'sgross income; therefore, H.R. 13375's 17 percent re-duction in benefits is a minimum and S. 1993's 24percent increase in benefits is a maximum.

72

Judged strictly on the basis of their effectivenessin removing households from poverty, once the food stamptransfer is counted as income, the two proposals differsignificantly. The higher allotment level in S. 1993assures an increase in the number of poor householdsmade nonpoor following the transfer. Approximately 1.3million (29 percent) of the pre-food stamp poor house-holds would be classified as nonpoor under S. 3136. Incontrast, H.R. 13375 would make it more difficult forpoor households to move above poverty. With no majorallowances for deductions, only about 500,000 poor house-holds (11.2 percent) would be made nonpoor.

In total, including the provision for countingnutrition in-kind benefits in household income, it isestimated that, in fiscal year 1978, the modificationproposed to enhance the program food-consumption effect-iveness would result in outlays of approximately $2.7billion under H.R. 13375 and $4.6 billion under S. 1993.These would be reductions of nearly $2.7 billion and$800 million, respectively, from current policy estimates.

As indicated from these figures, savings in the foodstamp program can occur, but, as also indicated, with asignificant reduction in benefits to a large number ofcurrent recipients. Both proposals result in an increasein food stamp purchase requirements. Current participantswould increase their expenditures on food stamps, goingfrom an average 19 percent of gross income under the currentprogram to 30 percent under these proposals. It is thisfact alone that would result in a relative decline in in-come supplementation and an increase, although in fewerhouseholds, in the food consumption effectiveness of thetransferred federal dollars.

Income Support Emphasis Through Elimination ofFood Purchase Requirement

A major proposal advanced in the Senate (S. 2451)during the 94th Congress is likely to be reintroducedin reform proposals brought before the next session ofCongress. The proposal would eliminate the requirementthat households purchase any food stamps. For example,instead of paying $60 to receive $166 worth of foodstamps, a household would simply receive the actualfederal transfer of $106 in bonus food stamps.

73

Proponents of this proposal point to low participationrates among the eligible population, estimated under thecurrent program to be approximately 50 percent in any givenmonth. The elimination of the food stamp purchase re-quirement would remove one major disincentive to partic-ipation. While increased participation is one policycriterion for evaluation, it must be weighed againstpotentially increased program costs, a matter discussedlater in this section.

The elimination of purchased food coupons would re-move from circulation about 40 percent of all food stampsin circulation in fiscal year 1978. Such a reductionwould result in some cost savings (approximately $22 mil-lion) associated with printing, distribution, and storageof stamps. More important, eliminating the sale of foodstamps by program vendors, such as banks, post offices,and community organizations, would improve program integ-rity and accountability by eliminating the need to trans-fer monies from program vendors to the Federal Treasury.37/

The elimination of the purchase requirement wouldalter the food consumption focus of the program. Thosewho consider the basic aim of the program to be the im-provement of diets and the encouragement of a basicsubsistence diet, rather than the elimination of povertyper se, may see the proposal as working against theseobjectives. The proposal in essence would return to thepre-1970 program and would provide participating house-holds an opportunity to secure a nutritionally adequate

37/ The "Emergency Food Stamp Vendor Accountability Actof 1976," Public Law 94-339, was signed into lawJuly 5, 1976. This law is aimed at tightening thesystem of monitoring vendors and establishing criminalpenalties for vendor abuse. Elimination of delays byvendors depositing receipts in the Treasury wouldeliminate loss of interest by the federal government.However, return of food stamp receipts to the generalfunds of the Treasury has no direct impact on thefood stamp program account, a completely separateaccount.

74

diet. The proposal would provide recipients the abilityto purchase an adequate diet but no program incentive todo so. It can be argued, however, that the current pro-gram, while providing a program incentive for increasedfood purchases, has nevertheless not been shown to havesignificant impact on the level of nutrition.

The proposal would provide a greater cash incometransfer to all except the lowest income recipients (whomust use the stamps to augment food purchases). Such atransfer would move the program more towards general in-come assistance. Under the present program, the purchaserequirement provides some assurance that the participatinghousehold will spend a greater proportion of its incomefor food than it would in the absence of the food stampprogram. If the purchased stamps are eliminated, therecipient household would have to continue to purchasefood at its pre-food stamp transfer level to achieve afood consumption level believed to be necessary for anutritionally adequate diet.

It is more likely, however, that in the absence ofpurchased food stamps, the bonus food stamps would sub-stitute for pre-food stamp food expenditures. As a con-sequence, rather than increasing food purchases, therecipient household would continue to purchase food atthe same level as before the transfer of bonus stampsand thereby generate freed income. The value of thisfreed income would be the difference between what thehousehold would normally spend for food and the bonusfood stamps, again assuming the bonus food stamps com-pletely substitute for normal food expenditures.

Income supplementation now exists for those householdswhose purchase requirement is less than their normal foodexpenditures. These households would, except for trans-action costs and social stigma, currently participateregardless of a purchase requirement. Those eligiblehouseholds for which the current purchase requirement isgreater than their normal food expenditures would have noreason not to participate. It is this latter group whichwould have the largest proportion of freed income.

75

While not extensive and certainly not universallyacknowledged, some studies have reported that someeligible but non-participating households have consideredthe purchase requirement prohibitive. Since the currentfood stamp program has no purchase requirement for thosewith less than $30 net monthly income, and since someproposed modifications would raise this to a minimum of$100 monthly, the impact of eliminating the purchaserequirement would be to increase participation amonggroups whose income exceeds any proposed standard de-duction. If participation among lowest income groupsis restricted by the purchase requirement, then elim-inating this barrier would also have the greatest foodconsumption impact on these households which wouldqualify for some of the largest benefits. If the soleobjective of this provision is to increase participationamong the lowest income groups, improvement in the admin-istration of the program's variable purchase option andadjustments in the level of proposed standard deductionscould be considered as alternatives. Such adjustmentswould, however, have little food-consumption impact.

For households qualifying for the largest bonus(currently lowest purchase requirement), the proposalwould continue to force an established food consumptionlevel because so much of their disposable income must gofor food. For all other poor households, a greaterchoice would exist in their consumption pattern. It canbe argued that if freedom of choice in. how the governmenttransfer is to be spent is the goal, providing all poorhouseholds that freedom (or providing none) is to be pre-ferred over granting it to a select few.

Most reform packages proposing to eliminate thepurchase requirement include modifications of otherprogram parameters, changing the overall focus of theprogram. While increased participation and thereforeincreased budget outlays would follow from eliminatingthe purchase requirement, adjustments to other programparameters could be made that would offset the increasedcosts. These adjustments, however, would not be withouta potential loss in benefits to current low-income re-cipients .

76

As an example, the philosophy behind itemized de-ductions has been in some instances based on the existenceof a purchase requirement. It was believed that house-holds with other special human needs (e.g. medical careand housing) should not be restricted in their purchaseof food stamps. Therefore, such expenses were deductedfrom gross income in calculating eligibility for foodstamps. Eliminating the purchase requirement lessensthe original need for special itemized deductions or,for that matter, for any standard deduction. Partic-ipation in the food stamp program would not suffer be-cause households were forced to make a choice betweenbuying medical care or buying food stamps. Food stampswould be a free good.

The S. 2451 proposal increased budget outlays notsolely because it increased participation by doing awaywith the purchase requirement, but also because it in-creased the eligible population by setting the net in-come eligibility standard at the current program levelsand provided a flat $125 a month standard deduction forall households ($150 a month for households with anelderly person). High standard deductions could havebeen lowered with no measurable effect on the benefitsto the current recipients while still reducing programcosts.

Also, the current prospective accounting system isdesigned to be highly responsive to the food consumptionneeds of people. Given that the food consumption ob-jectives would be lessened under an option that wouldeliminate the purchase requirement, consideration couldbe given to recasting the accounting procedures to aslightly less responsive, less costly system. Sincemonies would not be tied up for purchasing stamps, house-holds could still be responsive to their immediate house-hold needs.

In order to limit cost increases under any proposalthat would remove the purchase requirement, a trade-offmust be made that reduces average benefits for the cur-rently participating food stamp households, includingthose households that currently fall below the officialpoverty threshold. Unless such a trade-off is made,food stamp program costs would increase.

77

81-390 O - 77 - 8

The key to determining the budgetary costs ofeliminating the purchase requirement is to determineparticipation rates by income and benefit levels inthe present food stamp program and to adjust theseparticipation rates to reflect decreased transactionalcosts for the eligible recipient. Little data existon food stamp participation rates by specific incomegroups. One recent study, however, provides estimatesof simulated eligible households and estimated partic-ipants for April 1977 under the rules of the currentprogram..38 / These rates and the approximate benefitsassociated with each income group provide a generalassessment of the potential costs and beneficiary im-pact associated with the proposal.

TABLE 17. ESTIMATED PARTICIPATION RATES FOR FOOD STAMP HOUSEHOLDS, SIMULATED FOR APRIL 1977

MonthlyIncome ClGross( dollarsmonth)

Less than

1-99

100-199

200-299

300-399

400-499

500-599

600-749

750-999

1000+

OVERALLAVERAGE

SOURCE :

1per

0 8.9

57.2

55.0

45.2

36.3

38.5

26.4

—

—

__

43.4

Household Size

2

15.

83.

76.

63.

46.

46.

41.

47.

6.

53.

3

8

1

3

2

9

2

8

3

2

7

16.

42.

90.

78.

72.

47.

42

31.

15

8.

52

8

.9

.7

2

.7

.9

.5

.1

.1

.3

.4

Harold Beebout, Pat Doyle, and Allenfor 1977, A Microsimulation Approach

4

15.6

50.3

85.7

80.4

87.5

74.4

5'' 4

32.9

14.6

6.7

49.5

Kendall,

5

3.2

34.1

80.0

78.5

93.2

92.4

57.0

27.4

9.3

20.5

56.7

Estimation, Mathematica Policy

6

19.7

50.1

100.0

68.8

74.0

90.4

76.8

42.5

17.6

20.2

51.9

of FoodResearch

7

32

100

100

93

74

95

82

52

22

4

57

StampInc. ,

.2

.0

.0

.2

.5

.0

.0

.8

.0

.5

.2

8

21.9

84.2

69.7

76.6

63.1

55.4

82.5

63.9

31.7

18.4

47.9

ParticipationJuly 28, 1976

OverallAverage

14.9

66.7

64.7

56.2

56.9

62.6

50.2

37.8

16.2

12.9

49.6

and Cost

38/ Harold Beebout, Pat Doyle and Allen Kendall,Estimation of Food Stamp Participation and Costfor 1977: A Microsimulation Approach, MathematicaPolicy Research Inc., July 28, 1976.

78

Table 17 presents estimated participation rates inthe current program by household size and gross monthlyincome. As indicated, participation rates range fromcomplete coverage in lower income groups to very lowparticipation rates in the higher income groups. Over-all participation averages about 50 percent. Theseparticipants receive approximately 55 percent of thetotal possible benefits. It should be noted that forthe income group with zero gross income no purchase re-quirement would exist under the current program. Ex-cluding this zero income class, overall participationrises to about 53 percent, receiving nearly 63 percentof the total benefits available to them. If completeparticipation were to result from this proposal, it isestimated that, in fiscal year 1978, program outlayscould increase nearly $2.7 billion. Based on currentparticipation rates, the majority of these benefitswould go to households with incomes in excess of $300a month. It is unlikely, however, that such highparticipation rates would result.

Current participation in the regular AFDC program(i.e., excluding unemployed fathers) is estimated to beabout 93 percent_..?_9./ Using a similar participation rateunder this option, budget outlays would increase bynearly $2 billion in fiscal year 1978. Since the averageAFDC family benefit is nearly three times the averagefood stamp benefit, and since AFDC participation pro-vides additional in-kind benefits such as categoricaleligibility in the food stamp program and medicaidprogram, higher participation rates in the AFDC programcan be expected. Therefore, it is unlikely that partic-ipation in the food stamp program would reach those levelsnow estimated to be occurring in the AFDC program.

Few clues are left to judge the real budget impact.Relationships between participation rates and averagepurchase requirements, participation rates and purchaserequirements as a percentage of gross and net incomes,

39_/ Kevin Hollenbeck, An Analysis of the Impact of Un-employment and Inflation on AFDC Costs and Caseloads,Mathematica Policy Research, February 13, 1976.

79

and participation rates as a function of the ratio ofbonus stamps to purchase stamps—all point to the simplefact that participation is more a function of the benefitthan a function of the purchase requirement. Of course,one is the reciprocal of the other, making it even moredifficult to ascertain the real budget impact. At aminimum, however, most estimates show an increase inparticipation of between 10 and 20 percent as a resultof eliminating the purchase requirement .40.7 Such in-creases would result in increased program outlays ofbetween $600 million and $1.2 billion in fiscal year1982.

Cash-Out Food Stamps

Cash-out food stamp proposals eliminate completelythe in-kind food transfer mechanism and replace it withan unrestricted cash assistance program. Past nationalanalyses of the welfare system, such as that done by thePresident's Commission on Income Maintenance Programs in1968 and the studies conducted over the period from 1972to 1974 by the Subcommittee on Fiscal Policy of the JointEconomic Committee, proposed replacing the food stampprogram and the AFDC programs with a federally financedand administered cash grant system. However, other majorproposals, such as the Family Assistance Plan in 1969,the Family Assistance Act of 1970 as passed, and morerecently a proposal to modify the unemployment compen-sation program as a means for achieving welfare reform,continue to retain food stamps as a supplement to thebasic cash grant. Under less than a comprehensive cash

40/ See CBO Cost Report, Senate Report No. 94-697,March 13, 1976.

80

assistance program, the goal of restricted food con-sumption could justify the retention of the food stampprogram as an in-kind transfer.£!/

Further, maintaining less than a complete cash allow-ance system, supplemented by food stamps and providingthe opportunity for recipients of the cash grant to usepart of that grant to purchase food stamps, results inlower total budget costs. This is because it is likelythat, as in the current program, participation in thefood stamp program would be less under such an optionthan under a complete comprehensive cash program. How-ever, since recipients tend to value the in-kind foodstamp transfer at less than government costs, lower cashgrants could be proposed that would reduce total budgetcosts at no loss of welfare to the recipient population.

Setting a cash guarantee level equivalent in dollarsto current allotment levels would provide every partic-ipant an opportunity to consume a minimum diet. Some ofthe transferred cash income would continue to be spent onfood. Would the nutritional standards of the participantsdecline? It does not necessarily follow that families re-ceiving cash assistance, as opposed to food stamps, wouldhave any better or any worse diets. One study examinedthe effects of the cash guarantee on the nutritionalquality of diets for a sample of families drawn from theRural Income Maintenance Experiments.^/ It found con-flicting results. Small but statistically significant

4_1/ A forthcoming article to appear in Public Choice,"Donor Optimization and the Food Stamp Program," byGiertz and Sullivan, suggests that the consumptionof food by the relatively poor has an especiallypowerful external effect on donors (taxpayers).Therefore they conclude that complaints that thefood stamp program provides an incentive for re-cipients to consume "inefficiently" in the usualwelfare sense miss the point: the donor's utility,not the recipient's, is being maximized.

O'Connor et al., The Negative Income Tax as a Meansof Improving Nutritional Levels Among Low IncomeFamilies, U.S. Dept. of Health, Education and Welfare,June 1975.

81

improvements were found in the quality of diets offamilies receiving the cash grant in one state in thestudy (North Carolina), but not in another (Iowa). Nocomparison was made between the diets of families thathad received food stamps before and after the cashtransfer.

Assuming that the basic guarantee in the cash-outoption would remain equivalent to the dollar value ofthe current food stamp allotment level; that the mar-ginal tax rate would remain at 30 percent; and thatthe current prospective accounting system would continue,with an assumed 90 percent participation rate, the costof cashing out food stamps could increase to between $2.2and $2.7 billion over current policy estimates for fiscalyear 1978.4_3/ This represents the total cash transferand does not include savings due to administrative simp-lification. If variations in the cash-out option areconsidered, such as maintaining a cap on income eligi-bility at the poverty threshold, transfer costs wouldincrease about $1.5 billion in fiscal year 1978 overcurrent policy levels.

In sum, the cash-out option, like the elimination ofthe food purchase requirement option, would significantlyincrease costs. However, unlike that option, cashing outfood stamps would probably not be undertaken without majormodifications in the other cash-assistance programs. Ifthis were the case, the total net cost to the federalgovernment would be different.

43/ These estimates were based on the March-April 1975matched Current Population Survey tape, whichcollected annual income for 1974 and requisitefamily information. Adjustments to the simulatedannual costs in 1974 were made to reflect adjust-ments for deductions and intra-year variation inincome flow. All figures were adjusted to fiscalyear 1978 dollars.

82

Overview of Options

Table 18 presents an overview of the various legis-lative options discussed in this chapter. Each optionhas been reviewed in terms of the various policy criteriaused in the debate of the food stamp program: adequacyof benefits for the participating population; impact onfood/income tradeoffs; impact of transfers on targetgroups; work incentives; and administrative simplicity.

The various proposals the 95th Congress will con-sider present widely ranging choices as to what directionthe future food stamp program will take. Such choiceswill each bear differently on the various policy criteriaoutlined above. The multiple objectives and wide rangeof federal outlays associated with each choice willchallenge the new Congress, not only in relation to thefood stamp program, but also as those choices interactand relate to the total question of income security forall needy Americans.

83

TABLE 18. POLICY CRITERIA FOR EVALUATING MAJOR FOOD STAMP PROGRAM ALTERNATIVES

LEGISLATIVE ALTERNATIVESAND ESTIMATED FEDERALBUDGET COSTS

AVERAGE BENEFITS FOOD/INCOME TRADEOFF

1. CURRENT PROGRAM Basic guarantee set atFY 1978: $5.4 billion level based on consump-

tion pattern of low-incomeFY 1982: $5.9 billion households in 1965-66,

meeting the RecommendedDietary Allowance for a fourperson household consistingof a couple aged 20-54, withtwo children aged 6-8 and9-11. Guarantee adjustedsemi-annually based on foodprice changes. Guaranteesfamily of four, regardlessof sex-age composition, withzero income following bene-fits:FY 1978 1982Monthly $172 $207Annually $2,064 $2,480Meal/Person 47<? 57<?

Estimated that on theaverage approximately57£ of $1 federal trans-fer spent on food.Average food stamphousehold increased foodexpenditures $40.00 permonth (1973), over whatnormally would have beenspent. No evidence ofimproved nutritionalstandards through part-icipation. Currentproportion of totalincome spent on stamps,estimated 19.4%.

2. PROGRAM REFINEMENTS: Maintains current guaranteeS.3136, Senatepassed Reform ActApril 8, 197694th Congress, 2ndsessionFY 1978: $6.0 billion

FY 1982: $6.5 billion

levels, but increases aver-age transfer to eligiblecurrent recipientsapproximately 19%.

Relative to current pro-gram reduces slightlyproportion of incomerequired to purchasestamps to 16.3%. Re-sults in decline infood consumption per$1 federal transfer.

3. PROGRAM REFINEMENTS:H.R. 13613, HouseCommittee on Agri-culture, September

94th2nd

1, 1976.Congress,sessionFY 1978: $5.3 billion


Maintains current guaranteelevels, but increasesaverage transfer toeligible current recip-ients by 19%.

Maintains approximatelycurrent program foodconsumption impact,reduces direct foodconsumption impact forelderly, blind anddisabled.

84

DISTRIBUTIONOF BENEFITS

WORK INCENTIVES ADMINISTRATIVESIMPLICITY

Reduces potential partic-ipation of working house-holds, due to 90 dayaccounting system. Re-directs benefits toelderly and smallerhouseholds. Estimated29% of participatingpoor households re-moved from poverty.

Standard deduction resultsin low implicit tax rate.No deduction for taxesor work expenditures.

Simplifies adminis-tration, through fixeddeduction levels andflat purchase require-ment; complicates ad-ministration with 90day accounting systemand monthly reportingsystem.

Concentrate 98% of ben-efits to households be-low poverty, increasesproportion of benefitsgoing to elderly house-holds and public assis-tance households, re-duces benefits going toworking poor households.

Decreased work incentivesthrough non-allowance ofwork related expenditures,higher average tax rateminor impact on work in-centive. Food stamp ben-efit notch approximately$940 per family of fourat net breakeven income.

Simplifies adminis-tration through elim-ination of deductlbles.Complicates adminis-tration through revisedasset tests and monthlyreporting system.

Increases participationprimarily among higherincome groups. Expandseligible population.For current recipientsredistributes benefitsto lower income, smallermore elderly size house-holds.

Maintains current workincentives, eliminatesbenefit notch, allowingnormal phase-out andspecial deductions forwork related expend-itures.

Simplifies programadministration, reducesapproximately 40%coupons in circulation.Simplifies adminis-tration through uniformdeduction and benefitreduction rates.

Increased participation,in higher income groups.Redistribution of totalbenefits shifted towardhigher income groups.

Assuming 30% marginaltax rate, currentdeduction for work-related expenditures,relative to currentprogram no change inwork incentives.

Simplifies programadministration; reducesissuance, purchasing andredemption process;provides for easierprogram integration.

85

TABLE 18. (Continued)

LEGISLATIVE ALTERNATIVESAMD ESTIMATED BUDGETCOSTS

AVERAGE BENEFITS FOOD/INCOME TRADEOFF

4. PROGRAM REFINEMENTS:Administrationregulations en-joined June 18, 1976FY 1978: $4.9 billion


Maintains currentguarantee levels, in-creases average transferto households remainingeligible after 90 dayretrospective accountingperiod by approx. 7%

Maintains approximatecurrent program foodconsumption impact,approx. 20% of grossincome would be usedto purchase stamps.

NUTRITION SPECIFIC Increase basic benefitsS.1993, Buckley to low-cost food plan,H.R. 8145, Michel approx. 30% higher thanas introduced in current program. Stand-94th Congress ard family of four re-FY 1978: $4. £• billion mained unchanged for

pricing higher food.FY 1982: 35.7 billion Average transfer to

eligible current recip-ients increases approx-imately 24%.

Increases proportionof transfer dollarsspent on food byparticipants. Approx-imately 30% of grosscash income would beused to purchasestamps.

6. ELIMINATION-PURCHASEREQUIREMENTS.2451, Dole-McGovern H.R.10467, HeinzH.R. 10441, Hallas introduced in94th Congress

Maintains currentguarantee levels, in-creases average trans-fer to eligible,current recipients byapprox. 15%.

FY 1978: >.5 billion

Reduces proportion oftransfer dollar spenton food; reduces im-plicit purchase re-quirement to 17% ofgross income. Increasesfood expenditures forneedy households notparticipating becauseof purchase requirement.


7. CASH-OUT FOOD Assumes current programSTAMP guarantee level, andFY 1978: $7.6 billion current accounting

system. ProvidesFY 1982: $8.6 opportunity for house-

billion holds to secureadequate diet; doesnot guarantee adequatediet.

Increased food ex-penditures per $1transfer reduced to20-30i?. Marginalpropensity to consumefrom cash less thanfrom coupons.

86

DISTRIBUTION OFBENEFITS

WORK INCENTIVES

Estimated 50% of eligiblepopulation._in. an?L giysnmonth participate.Par-ticipation highest amonglowest income groups,and households receivingother public assistancebenefits. Approx. 87%benefits distributed to78% of participatinghouseholds below pov-erty; 22% of benefitsto working poor house-holds. Estimated 26%participating poorhouseholds removed frompoverty.

ADMINISTRATIVESIMPLICITY

Low tax rate on earnings(approx. 30%), and lowbasic guarantee, re-sults in good work in-centive by itself. Inconjunction with otherprograms, however, poten-tial work incentives prob-lems. Estimated that37% of households re-ceiving AFDC income facemarginal tax of approx.75%. Food stamp benefitnotch $288 family offour at net breakevenincome.

Administratively com-plex, involves coor-dination of over 3,000welfare certificationoffices; issuance ofauthorization to pur-chase (ATP) cards byover 14,250 gov't.agencies, post officesand banks; acceptanceof coupons by 270,000retail and wholesaleoutlets; redemption bybanks, Federal Reserveand special account inTreasury. Complexitiesof deductible and assetdetermination result inoverissuance and under-issuance of properbenefits.

Maintains high grossincome eligibilitystandards. Redistri-butes benefits awayfrom larger sizehouseholds tosmaller more elderly;increases benefits toworking poor. Estimated36% poor households re-moved from poverty.

Increased participa-tion among elderly,blind and disabled;91%_ of benefits go topersons below poverty.Redistributes benefitsaway from smaller sizehouseholds to largersize, reduces benefitsto working poor.Estimated 25% poorhouseholds removedfrom poverty.

Standard deduction lowersimplicit tax rate, increaseswork incentive. Providesspecial deduction for taxesand special deduction forhouseholds with earned in-come , increased expendituresfor work registration. Foodstamp benefit notch $618 forfamily of four at breakevennet income.

Relative to current programno significant change inwork incentives. Provideslow implicit tax rate,allowance for taxes, plus$30 for earned income,child care deductions.Food stamp benefit notch$480 at breakeven netincome.

Simplified eligibilitydetermination throughstandardizing deductions,uniform purchase require-ment. Maintains currentadministrative anddistribution system.

Increases administrativecomplexities, requireseligibility determinationfor blind and disabled,requires increasedagency coordination.Continues determinationverification mandatorydeductions, child careexpenditures.

87

APPENDIX

APPENDIX TABLE TOTAL FEDERAL, STATE, AND LOCAL EXPENDITURESFOR FEDERAL FOOD STAMP PRCQIAM, FISCAL YEARS1969-1977 (In thousands of dollars)

A.B.

C.D.

E.

Total Budget Authority a/Federal Obligations:1. Bonus Food Stamps2. Federal Share of Ad-

ministrative Costs3. Printing & Production4. Shipment of Coupons5. Processing of Food

Stamps6. Employment Registration7. Federal Administration

(including investiga-tions & Audits)Subtotal (2-7)Total Obligations

Total Federal Outlays c/Nonfederal Operating Costsd/Total Federal, State, andLocal Obligations

1969

280,000

228,587

6,5854,672

59

88—

10,41321.817250,403247,766

21,700

272,103

1970

596,963

550,806

9,4884,261

74

108

13,42627.357578,162576,810

27,200

605,362

1971

1,670,000

1,522,904

20,12115,370

258

222

21,39957,370

1,580,2731,567,767

53,900

1,634,173

a/ Budget authority for 1969 to 1976 includes regular and supplemen-tal appropriations. For 1969 through 1971, no carry-over fundingincluded. For 1972 through 1976 authority includes carry-overand unused funding (recoveries). Carry-in funds for transitionquarter of $201,403,000; new budget authority for transitionquarter of $1,237,441,000.

b/ Budget authority for 1977 includes regular appropriation of$4,794,400,000 under Public Law 94-351. Carry-in authorityfrom transition quarter of $112,170,000 and anticipated sup-plemental appropriation of $643,830,000.

90

Transition1972 1973 1974 1975 1976 Quarter 1977

2,289,214 2,500,000 3,000,000 5,300,236 5,874,876 1,438,844 5,542,468

1,842,466 2,135,655 2,727,658 4,395,983 5,305,985 1,230,202 5,180,000

26,87412,380

658

2808,500

25,04373,735

1,916,2011,909,166

30,65410,8371,028

59815,000

27,71485,831

2,221,4852,207,532

47,53623,9702,287

1,90417,371

32,152125,220

2,852,8782,844,815

206,77435,2031,997

4,38228,233

41,134317,723

4,713,7064,598,956

255,50738,8601,891

3,65327,200

43,093370,204

5,676,1935,631,954

72,1923,138

42

1,6797

12,42189,479

268,793

101,207

370,0001,326,675 5,550,0001,325,159 5,508,933

74,300 111,000 137,100 287,000 255,507 72,192 268,793

1,990,501 2,332,485 2,981,915 5,000,706 5,931,700 1,398,862 5,818,793

c/ Federal outlays are taken from Office of Management and Budget, The Budgetof the United States Government, 1969 through 1977. Transition quarteroutlays provided by USDA, Food & Nutrition Service.

d/ Estimates for 1969 through 1975 are those as developed by USDA reportedby Kenneth W. Clarkson, Food Stamps & Nutrition, Evaluation Studies 18,American Enterprise Institute for Public Policy Research, Washington,D.C., April 1975. CBO estimates for 1976 and transition quarter basedon estimated federal share of administrative costs.

91

U. S. GOVERNMENT PRINTING OFFICE : 1977 O - 81-390