AEI Agricultural Policy Series The 2007 Farm Bill and Beyond · 2019-01-18 · THE 2007 FARM BILL...

AEI Agricultural Policy Series

The 2007 Farm Bill and Beyond

Project Directors

Bruce L. Gardner and Daniel A. Sumner

The AEI Press

Publisher for the American Enterprise Institute

W A S H I N G T O N , D . C .

Cover images.Aerial View of Tractor in Field © Rim Light/PhotoLink/Photodisc Green/Getty Images.Cow at Trough © B Drake/PhotoLink/Photodisc Green/Getty Images.Grain Silos © S Alden/PhotoLink/Photodisc Green/Getty Images.

© 2007 by the American Enterprise Institute for Public Policy Research,Washington, D.C. All rights reserved. No part of this publication may be usedor reproduced in any manner whatsoever without permission in writing fromthe American Enterprise Institute except in the case of brief quotationsembodied in news articles, critical articles, or reviews. The views expressed inthe publications of the American Enterprise Institute are those of the authorsand do not necessarily reflect the views of the staff, advisory panels, officers,or trustees of AEI.

Printed in the United States of America

iii

OVERVIEW AND ACKNOWLEDGMENTS V

KEY FINDINGS AND POLICY RECOMMENDATIONS 1

U.S. AGRICULTURAL POLICY REFORM IN 2007 AND BEYOND 5Bruce L. Gardner and Daniel A. Sumner

JUSTIFICATIONS FOR EXISTING POLICY 21

Does the Economic Situation of U.S. Agriculture Justify Commodity Support Programs? Bruce L. Gardner 23

Farm Subsidy Tradition and Modern Agricultural Realities Daniel A. Sumner 29

EFFECTS OF EXISTING POLICY 35

Who Really Benefits from U.S. Farm Subsidies? Julian M. Alston 37

Money for Nothing: Acreage and Price Impacts of U.S. Commodity Policy for Corn, Soybeans, Wheat, Cotton, and Rice Bruce A. Babcock 41

U.S. Sugar Policy: Analysis and Options John C. Beghin 47

U.S. Dairy Policy: Analysis and Options Joseph V. Balagtas 53

Specialty Crops and the 2007 Farm Bill Mechel S. Paggi 57

Agricultural Policy and the U.S. Livestock Industry Gary W. Brester and Vincent H. Smith 63

The Distribution of U.S. Agricultural Subsidies Barrett E. Kirwan 69

Double Indemnity: Crop Insurance and the Failure of U.S. Agricultural Disaster Policy Joseph W. Glauber 75

Contents

THE 2007 FARM BILL AND BEYOND

iv

INTERNATIONAL TRADE AND POLICY ISSUES 81

Lessons from Agricultural Policy Reform in Other Countries Julian M. Alston 83

The Impact of the WTO and Bilateral Trade Agreements on U.S. Farm Policy Timothy E. Josling 87

The Farm Bill and WTO Compliance Daniel A. Sumner 93

U.S. International Food Assistance Programs: Issues and Options for the 2007 Farm Bill Christopher B. Barrett 97

CONSERVATION AND ENVIRONMENTAL POLICY 103

Land Retirement for Conservation: History, Analysis, and Alternatives Ralph E. Heimlich 105

Payments for Ecosystem Services and U.S. Farm PolicyJohn M. Antle 111

Public Policy Solutions to Environmental Externalities from Agriculture Nicolai V. Kuminoff 115

Biofuel Incentives and the Energy Title of the 2007 Farm BillJohn A. Miranowski 121

RURAL DEVELOPMENT 127

Rural Development Policy Maureen R. Kilkenny and Stanley R. Johnson 129

Infrastructure Investment and Rural Development Mitch Renkow 135

v

Congress reconsiders federal agricultural policyevery few years when it drafts a new farm bill. Inmost years, this is a relatively uneventful process,but 2007 might prove to be different.

Budget deficits and PayGo rules are forcing law-makers to closely examine all spending programs.Concerns about the security of oil imports and highenergy prices are driving calls for expanded ethanolsubsidies. International trade rulings and negotia-tions also call the structure of current farm programsinto question.

To help inform these debates, the American Enter-prise Institute asked two noted agricultural econo-mists and former top United States Department of

Agriculture officials, Daniel A. Sumner of the Univer-sity of California at Davis and Bruce Gardner of theUniversity of Maryland, to direct a project examiningthe wide range of policies that comprise Americanagricultural policy. They brought together a collectionof eighteen noted economists to study these issuesand offer policy recommendations.

This document is a collection of short summariesof their research. The full papers may be found atwww.aei.org/farmbill.

This publication, and the research and conferencethat led to it, were funded by the American EnterpriseInstitute’s Inez and William Mabie Endowment forAgricultural Policy Research.

Overview and Acknowledgments

Title 1 Commodity Subsidies

• Elimination of farm subsidies for corn, wheat,and soybeans would have little effect on farmproduction or commodity prices. These Title 1subsidies are in effect “money for nothing.” (Bab-cock, p. 41)

• Farm households earn an average of 20 percentmore than the average American household; theaverage income of commercial-scale farms(which receive the largest share of subsidies) ismore than three times that of the average house-hold; and the median net worth of farm house-holds in 2005 was close to five times greater thanaverage. (Gardner and Sumner, p. 7; Gardner, p. 24)

• Elimination of farm subsides would not harmvulnerable farm families. Only 3 percent of allfarm households have both low-income and low-net worth and these receive few payments.(Gardner, p. 26)

• Eliminating subsidies would negatively affectowners of cropland, but the losses in land valuewould be less than the annual value of croplandprice appreciation in recent years. (Gardner andSumner, p. 10)

• Loan deficiency, direct, and countercyclical pay-ments may be in conflict with current WTO lim-its and these programs thwart progress in open-ing global markets. (Josling p. 87; Sumner, p. 95)

• Agriculture has continued to flourish in coun-tries, such as Australia and New Zealand, whichhave made significant, market-friendly reformsto their agricultural policies. (Alston, p. 83)

• Payment limits on recipients of subsidies havebeen evaded because they were based on indi-vidual payments, rather than farm payments.The result has been needless restructuring of theownership profile of farms. (Kirwan, p. 71)

Crop Insurance

• Taxpayers currently pay $3 billion a year to sub-sidize crop insurance, in addition to $2 billionfor ad hoc disaster payments. In some casesfarmers collect twice on the same loss. (Glauber,p. 79)

• Farmers receive between $2.06 and $2.73 forevery $1 they paid in crop insurance premiums.But it’s a remarkably inefficient program: the

1

Key Findings and Policy Recommendations

government pays $1 for every 60 cents farmersreceive. (Glauber, p. 77)

Sugar, Dairy, and Livestock

• The sugar program costs consumers hundreds ofmillions of dollars each year by keeping U.S.sugar prices at about double the world level.(Beghin, p. 49)

• Current dairy policy also transfers millions ofdollars from consumers to dairy farmers; the costto consumers and taxpayers far exceed the bene-fits to dairy farmers. (Balagtas, p. 56)

• The livestock industry is seriously disadvantagedby biofuel policies, which drive up the price ofmany feed grains. (Brester and Smith, p. 63)

International Food Assistance

• U.S. food aid programs are hugely wasteful.Logistical costs eat up 60 percent of the U.S. foodaid budget; in contrast, Canada only spends 32percent on logistics. These high costs are a resultof requirements that food be purchased andtransported from the United States, and ship-ping, bagging, and processing be undertaken byapproved contractors. (Barrett, p. 99)

• Such mandates can produce absurd inefficien-cies. For example, the 1986 Farm Bill stipulatesthat Great Lakes ports should retain their 1984shares of food aid cargoes. But because no shipslegally permitted to transport food currentlyserve Great Lakes ports, the stipulated share ofaid must be moved by truck or rail through theGreat Lakes region before it is then transportedby land to another U.S. port for overseas ship-ment. (Barrett, p. 100)

Key Policy Recommendations

Title I Subsidies

• Eliminate Title I subsidies. American agricultureis financially healthy; farmers do not need subsi-dies to prosper; current subsidy programs arevulnerable to successful challenge before theWTO; and these programs serve no useful pub-lic purpose. (Gardner, p. 26; Sumner, p. 32 andp. 93; Antle, p. 111)

• Well-designed payments to farmers for ecosys-tem services could replace traditional subsidiesand be WTO compliant. (Antle, p.112)

• As an alternative to a complete end to subsidies,Title I subsidy programs, disaster assistance, andcrop insurance programs could be integrated intoa more cost effective, nonduplicative, and trans-parent safety net for farmers. (Babcock, p. 42)

• Proposed subsidy payment limitations mustclearly define a “farmer” and more closely scruti-nize farm reorganizations to avoid wasteful farmreorganizations. (Kirwan, p. 71)

• Use of explicit and powerful transparency insti-tutions, and temporary and limited adjustmentassistance measures, may facilitate significantpolicy change. (Alston, pp. 84–85)

Crop Insurance

It makes little sense to have both recurring disasterprograms and an expensive crop insurance pro-gram. Either eliminate the crop insurance program,or reform it to shrink program costs and reduce theneed for ad hoc disaster payments. (Glauber, p. 78)


2

Sugar, Dairy, Livestock, and Specialty Crops

Dairy: Current dairy programs—the price supports,milk marketing orders, and the MILC program—should be repealed. Or, at the very least, subsidyprograms should be replaced with direct paymentsthat do not distort market prices or production.(Balagtas, p. 55)

Sugar: The sugar program should be eliminated. Ifpolitically necessary, sugar producers could becompensated for their losses, which would alsomake transparent the huge magnitude of the cur-rent transfer from consumers to a handful of largeproducers. (Beghin, p. 51)

Livestock: Be wary of increasing subsidies for bio-fuels; set a high standard of proof in unfair compe-tition claims, and impose stringent sunset provi-sions on retaliatory actions where countervailingmeasures are justified; and explore whether amandatory or voluntary animal identification pro-gram satisfies concerns about animal health (e.g.,“mad cow disease”). (Brester and Smith, pp. 63–67)

Specialty Crops: Support the shift of funds in thegeneral direction suggested by these industries, butclosely examine new requests for assistance fromthe specialty crop industry to ensure they pass apublic interest test. Increased funds for researchand development, protection from invasivespecies, and programs that promote specialty cropconsumption are most likely to pass such a test.(Paggi, p. 58)

Food Aid Programs

End requirements that food purchased with U.S.dollars be domestically produced, packed, andprocessed, and that only U.S.-flagged carriers shipfood aid; permit the use of cash for local andregional food purchases. (Barrett, p. 101)

Conservation, Bioenergy and the Environment

Conservation: Redesign land retirement programsto achieve environmental goals in addition towildlife habitat protection, and to permit non-intensive agricultural uses while protecting envi-ronmentally sensitive land. (Heimlich, p. 109)

Bioenergy: Tax fossil fuels or institute a cap-and-tradesystem for greenhouse gas emissions instead ofexpanding ethanol subsidies. Reconsider currentethanol mandates and tariffs that bias fuel choices anddrive up food and livestock feed costs. Alternatively,adopt subsidies, alternative fuel mandates, and researchand development programs that do not favor ethanol,permitting the market to identify the most cost-effec-tive alternatives to fossil fuels. (Miranowski, p. 125)

Environment: Solve budget limits on farm environ-mental programs by taxing those management prac-tices that negatively impact the environment ratherthan paying farmers to reduce polluting practices.Alternatively, if payments continue, link payments tomeasurable environmental outcomes and requirefarm subsidy recipients to document their environ-mental management practices. (Kuminoff, p. 117)

Rural Development: Federal investment in watertreatment projects appears difficult to justify, butfederal investments in broadband deploymentscould be justified if they have positive net socialbenefits. (Renkow, p. 137) Federal policymakerscould also move rural development funds throughthe rural commercial banking system, find ways topromote rural liquidity as subsidy-inflated returnsto farm land are reduced in future years, and stopproviding outright grants and require rural com-munities to match federal funds for infrastructure.(Kilkenny and Johnson, p. 132)

These papers show that current agricultural policyis wasteful and often counter to national welfare.American consumers and taxpayers would benefitfrom a more market-friendly and cost-effectiveagricultural policy.

KEY FINDINGS AND POLICY RECOMMENDATIONS

3

As U.S. agricultural policy is revisited in the 2007Farm Bill debate, the occasion is ripe to reconsiderwhether our commodity programs and related legis-lation best serve agriculture and the economy, andwhat policies might do better. The studies in thisworking paper series demonstrate that the currentprograms serve neither the national interest nor the interests of rural people or agriculture as anindustry. Congress and the president would do thecountry an economic favor by substantially reform-ing these policies.

We recommend a shift to a much reduced gov-ernment role in markets now influenced by agricul-tural programs, along with an end to the substantialtransfers to producers and other resource ownersassociated with the supported commodities. This is a less radical reform than it may appear to be. Alarge fraction of U.S. agriculture already operates in an approximately free market environment.Drawing upon a series of papers on particular policyissues commissioned by the American EnterpriseInstitute (AEI) in 2006, we provide evidence thatthe programs now in place have costs far in excessof the benefits they can reasonably be argued to pro-vide, and that U.S. agriculture would prosper no less than at present, and with better long-termprospects for both farmers and the public generally,in the absence of these programs. Before developing

ideas for policy change, we outline some of the pur-poses that advocates have proposed for farm subsidyprograms and provide some factual context aboutmodern agriculture in the United States.

Purposes of Agricultural Policy

Agricultural policy can serve two broad objectives:the efficiency and prosperity of agriculture as anindustry, and the well-being of farm and rural peo-ple.1 These purposes are closely related but distinct,in that agriculture as an industry is important to thenation as a whole: to consumers of food, to earnersof income from industries related to agriculture, andto taxpayers, in addition to farmers themselves. Thewell-being of farm households is related to agricul-ture as an industry, but is broader in that farmhouseholds receive a large fraction of their incomesfrom sources other than farming, and farming iseven less important to rural welfare more broadly(see Kilkenny and Johnson 2007).

Broad national aspects of agriculture as an indus-try that have been advocated as giving purpose toagricultural policy are food security, food affordability,nutrition, competitiveness in international trade, andagriculture as a contributor to prosperity and incomegrowth in rural areas. For American agriculture to

5

U.S. Agricultural Policy Reform in 2007 and Beyond

Bruce L. Gardner and Daniel A. Sumner

contribute to these purposes, farms must be econom-ically viable, farmland must generate returns suffi-cient to keep it from conversion to other uses, returnsto investment must be sufficient to keep the capitalstock in agriculture at the technological frontiers, andinvestment in new technologies and the skills toimplement them must be maintained. At the sametime, with relatively open borders and competinguses for rural land and other resources, these broaderpurposes are not served by maintaining farms in par-ticular commodity markets or in particular regionswhen market forces would otherwise signal a shift.

Federal commodity programs fit into these pur-poses through their contribution to the economicviability of farms balanced against their costs to con-sumers, taxpayers, the broader agricultural and non-farm economy, and other national objectives, suchas the provision of rural environmental quality. Thethreats to farm viability that have made such pro-grams attractive are low market prices of farm prod-ucts, and variable prices and farm income because ofweather shocks and market fluctuations. But if theseprices and returns reflect market realities, what isthe justification for commodity programs sendingdifferent signals to producers?

Answers given are market failures in the form ofimperfect competition among the corporate buyersof farm products and sellers of inputs to them, aswell as the actions of foreign governments thatrestrict the market access of U.S. products or subsi-dize their own production of competing products.Other market failures, which lie behind excessivesoil degradation and environmental damages associ-ated with unregulated agricultural production, arealso seen as defects that can be remedied with com-modity programs (for example, by requiring conser-vation practices by recipients of commoditysupport) or by conservation programs per se (forexample, by idling environmentally sensitive acreageor subsidizing environmentally friendly practices)(see Heimlich 2007; Kuminoff 2007).

Poverty in rural America was a major force behindthe original establishment of commodity programs inthe 1930s. Rural poverty remains a blight in theAmerican economy to this day, and the idea that

higher commodity prices could reduce that povertyattracts support to these programs. Evaluating farmpolicies therefore must be considered in light of theireffects on farm and rural incomes, and especially onthe effectiveness at dealing with rural poverty com-pared to other approaches to this concern.

Among the central questions addressed in theAEI studies are whether current U.S. agriculturalcommodity programs constitute good policy forachieving any of these purposes. To be good policy,a commodity program must, at least, generate bene-fits to U.S. society that exceed the costs incurred topursue these goals.

For a fuller treatment of modern agricultural policy,it is important to go beyond the standard commodityprograms to consider crop and revenue insurance,international trade, conservation and environmentalconcerns, bioenergy, and rural development. In eachof these policy areas, the AEI studies ask whether thebenefits of current policies and alternatives to themgenerate benefits that exceed the costs.

The Economic Situation in Agriculture Today

It is important to distinguish the economic situationof households that have some farming activities(classified in the data as farm households) and theeconomic situation of farming as an industry. Theseare quite different issues, although the answer inboth cases is that they are doing quite well in gen-eral, and that pockets of concern are not those thatcan be remedied with commodity programs.2

For most of the twentieth century, farm householdincomes averaged far below the incomes of non-farmhouseholds. Post-World War II prosperity spilled overto the farm economy and farm incomes rose signifi-cantly from the Depression-era lows, even relative tonon-farm household incomes. Nonetheless the aver-age farm household income was still less than 60 per-cent of the average of all U.S. households from 1945to 1960, according to historical comparisons compiledby the U.S. Department of Agriculture (USDA) (seefigure 1). In the 1960s farm incomes, relative to non-farm incomes, began a trend increase that continues to


6

the present. By 1990 farm incomes had gained equal-ity with non-farm incomes, and in recent years farmincomes have been about 20 percent higher (Coveyand others 2005). Given the variability of farm prices,this income picture has had its ups and downs, with aspike up in the early 1970s and a dip below trend inthe early 1980s. An additional trend is the increasingimportance of income from off the farm for farmhouseholds. In recent years 80 to 90 percent of farmhousehold income has come from non-farm sources(Covey and others 2005). Indeed for most farm house-holds, that is the income that matters most.3

Small farms make up the bulk of all farms, but amuch smaller percentage of farm output. Farmswith gross revenue of less than $25,000 per yearcomprise 70 percent of all farms but contribute onlyabout 3.5 percent of U.S. farm output and report anegative net income from farming, according to the2002 Census of Agriculture (USDA 2004, table56).4 Yet the average income of the householdsoperating these farms was well above the average ofnon-farm household income because of off-farmearnings—although about 3 percent of farms, classi-fied by USDA as limited resource farms, report very

low average household incomes: a median of$10,300 in 2004 (Covey and others 2005).

Family farms classified by USDA as commercialscale operations (those that have $250,000 or morein sales) earned an average of $145,300 net incomefrom farming in 2004, plus $46,038 from off-farmsources, for a total of $191,338: higher than almostall non-farm households—notwithstanding thatgiven farm price and yield variability, 10 to 20 per-cent of the farms in this sales category have negativeincomes each year (see USDA 2005, p. 26; USDA2004, table 56). These farms produced three-fourths

of U.S. farm output, according tothe 2002 Census of Agriculture.

Net wealth data reinforce theincome picture. Farmers tend tohave low debt to asset ratios (anaverage of 13 percent debt toassets in 2005) and the smallerfarms most often carry no debt(Covey and others 2005). Farmhousehold wealth in all salesgroups is much above the wealthof other American households(Hoppe and Banker 2006).

To consider the economichealth of agriculture as an indus-try, one must focus on the sourcesof agricultural revenue, which aredominated by the 300,000 farmsthat produce almost all the farmoutput in the United States. Inthat context, is there any evidence

that agriculture is a troubled industry with funda-mental economic problems that need governmentintervention? As noted, the farms that produce thegreat bulk of U.S. farm output are on average prof-itable, with net income of more than $145,000 percommercial farm. In addition, farmland prices con-tinue to increase, reflecting wealth growth for own-ers. This growth in farmland prices has occurred inall regions, including those with little or no influ-ence from non-farm land uses. Therefore the growthin land prices reflects continued confidence in thefuture profitability of farming.

U.S. AGRICULTURAL POLICY REFORM IN 2007 AND BEYOND

7

FIGURE 1FARM HOUSEHOLD INCOME AS A PERCENTAGE OF

U.S. HOUSEHOLD INCOME, 1930–2005

SOURCE: Covey and others (2005).

0

20

40

60

80

100

120

140

160

1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

Percent

Farm output in the United States continues toexpand, even as farm prices continue a seculardecline relative to prices in the economy as a whole.The index of agricultural output has grown at 2 per-cent per year over the past fifty years (CEA 2007,table B-99). This growth of output continues despitegradual reductions in the amount of cropland andagricultural labor input. USDA’s index of aggregateinput quantities is unchanged between 1950 and2002, implying that overall total factor productivitygrowth has proceeded at about 2 per year. This pro-ductivity growth has allowed agricultural exports to continue to expand at the same time that the U.S. population has increased and U.S. demand forfood has expanded. Agricultural imports have alsogrown, reflecting a globalization of the food economy.Import data indicate that much of the expansion ofagricultural imports is in highly processed products(beer, wine, processed specialty foods) imported fromEurope, along with fruits and vegetables importedduring off-seasons for U.S. produce.

The result of growing agricultural productivity, acompetitive farm industry, and openness to importshas been continuing low food prices and a decliningshare of consumers’ income spent on food. Thus thesuccess of agriculture as an industry has providedbenefits to producers and consumers alike.

The question remains however of the extent towhich the economic success stories of commercialagriculture and farm households have been fosteredby commodity support programs and other farmpolicies, and are now dependent upon them. As wewill see, farm commodity programs have beenaimed at a narrow group of commodities and haveno particular linkage to farm productivity thatunderlies the success of agriculture or to alleviatingAmerica’s remaining farm poverty.

What Would Happen Without the CommodityPrograms: Short-term Impacts

Consider first the complete and rapid elimination ofcommodity programs. What would this entail? Thegovernment would cease a variety of activities that

differ greatly for different commodities. For mostcommodities (such as fruits and vegetables, hay,meat products, ornamentals), there is little govern-ment involvement or income support and little toeliminate. In the commodities where the govern-ment is active, its main roles are: intervention tosupport domestic commodity markets by adding todemand for products; regulation to support com-modity prices by limiting supplies; payments madedirectly to commodity producers; and regulation ofinternational trade in agricultural commodities,either to keep other countries’ products out or tomarket more U.S. products in foreign countries.

Interventions by the U.S. government to acquirecommodities when prices are low—either for laterresale, as with classical grain stockpiling, or forredistribution through food aid programs—wereonce a central device of commodity support. Whileauthorization of government commodity purchaseremains for dairy products and for purchases of sur-plus commodities for food distribution by USDA,this approach has been abandoned as a means ofsupport for the major crops. Even for dairy, its use ismuch reduced. The hoped-for benefits of price sta-bilization for both producers and consumers nevermaterialized. Instead, the accumulation of stockpilesgenerated intolerable costs as the government’s buy-ing prices tended to be too high, and producersobjected to the price-depressing effects of govern-ment sales.

Regulation of production, through acreagerestrictions or production quotas, began in the1930s for the major crops and persisted until the lastdecade for a few products such as peanuts. But thisapproach to commodity support has now beenabandoned, again because of perceptions by bothcommodity buyers and producers that the costswere great but the gains small. Production restric-tions create obvious costs to those who must buy athigher prices or are rationed out of the market. Itoften entailed idling productive land and otherresources, with obvious waste and impacts on com-petitiveness. Producers were convinced by the grow-ing response of competing agricultural exportingcountries that expanded production as the United


8

States held its output in check. Congress removedthe authority to implement annual acreage restrict-ing programs in the “Freedom to Farm” Act of 1996.

Payments to producers of programs crops andthose who have a history of such production havebeen increasing in importance in recent decades.Since 1985 for cotton and rice, and the early 1990sfor grains and oilseeds, “marketing loan” paymentshave gradually replaced government acquisition as amechanism to guarantee minimum prices to farm-ers. This approach gives producers price protection,but without accumulating commodity stocks andwithout choking off market demand through highermarket prices. Instead the costs of support are borneby taxpayers, who provide the funds for the pay-ments. Previously existing support payments variedinversely with market price, were tied directly tocurrent production, and required annual acreageidling. These were replaced in the 1996 Act by payments—renamed Direct Payments under thecurrent 2002 Act—that are based on past acreageand yield of the program crop and do not vary from year to year. This form of payment goes someway to reduce the criticism of earlier price supports that they generated overproduction of supportedcommodities.

The Countercyclical Payments Program, intro-duced in 2002, is intermediate between the marketingloans and the “direct payments.” The countercyclicalpayments are tied to the history of production of theprogram crop and vary inversely with market price of that crop; however, like the direct payments, theyare not tied to current production.

The many complexities of current programs andtheir national and international market conse-quences are discussed in the AEI papers by Alston(2007a), Babcock (2007), Balagtas (2007), Barrett(2007), Beghin (2007), Brester and Smith (2007),Kirwan (2007), Paggi (2007), and Sumner (2007b).The central issue we consider here is the implica-tions for U.S. agriculture if these programs were toend. This issue is clouded by the fact that the con-sequences of an end to commodity support looksquite different for the immediate future than theconsequences would have been in the immediate

past. The reason is that the immediate past includesseveral low-price years for major commodities from 1999 to 2005, when marketing loan programsas well as payments under the CountercyclicalPayment Program were substantial. Market pros-pects for the next five years, at least, are much morefavorable, according to the estimates of both USDAand private prognosticators (USDA 2007a; FAPRI2007). Accordingly, payments under current pro-grams would be much lower in the future than theyhave been in the past: reduced by roughly half,according to the baseline published by the U.S.Congressional Budget Office (CBO) in spring 2007(CBO 2007).

Babcock (2007) considers the market situationfor program crops without the full complement ofcommodity programs, under 2002–05 market con-ditions. When all programs are removed simultane-ously, Babcock estimates that there would berelatively little readjustment in acreages for thegrains and oilseeds and little overall reduction inproduction. The price of corn would have been 1.1percent higher; the price of soybeans, 0.4 percentlower; the price of wheat, 0.6 percent higher; andthe price of rice, 1.7 percent higher. Only for cottonwould the effects remain relatively large: an increasein price of 9.7 percent. Overall, these are smalleffects for payments that averaged $16 billion peryear in this period, amounting to about 25 percentof the market value of the commodities supported.Prices rise as production falls in most cases, with aslight price decline for soybeans because of a shift ofacreage from corn. The effects are small because ofthe presumed minimal supply response to them inthe analysis carried out. This means that buyers ofcommodities that receive payments will not paymuch more if commodity programs are ended.Taxpayers will pay much less, and the losses willaccrue in the supply side of the markets.

On the supply side, losers include farmers,landowners, and owners of other resources used infarm production. Direct payments and countercycli-cal payments are directly tied to specific farms with program base. Therefore landowners would beexpected to reap gains from these payments.


9

Subsidies to output are expected to accrue to sup-pliers of inputs according to the effects on thedemand and supply of these inputs. Direct evidenceof these incidence effects is not available, but fromanalysis of agricultural input markets Alston(2007a) estimates that 20 to 45 percent of outputsubsidies ends up in the pockets of landowners,many of whom are also producers. The findings ofKirwan (2007) on the effects on land rental rates ofpayments tied to land are consistent with this range:suggesting that, at least in the short run, each of themain forms of commodity program paymentsaccrue significantly to producers and suppliers ofother farm inputs, as well as landowners.

These considerations are important because theydetermine who will feel the economic pain were thecommodity support programs to end. It is clear thatoperators that grow program crops and landownerswith program crop base will bear the greatest losses,as they are the chief gainers from the existence of the programs. Would their losses trigger wider eco-nomic problems in rural areas? The possibility hasbeen raised of bankruptcies among farmers and con-sequent problems for rural banks and others withwhom farmers do business. Several factors indicatethat the risks of financial calamity with an end tocommodity programs are actually quite low.

First, the ratio of debt to assets in U.S. agricultureis now at a historical low, with total farm debt at 13percent of farm assets in 2005, as mentioned earlier.Farm real estate debt is $120 billion: just 7.5 percentof the $1.63 trillion value of farm real estate. If wesuppose that $8 billion annually (half of the $16 bil-lion in program payments from 2002 to 2005)would translate to a decline in the rental value ofland, and that this loss was capitalized at ten timesthe rental value, the $1.63 trillion would be reducedby $80 billion to $1.51 trillion: the kind of lossesthat occur to some business sector every year with-out triggering a financial crisis.

Of course, much farm real estate is not croplandthat has program base, nor is it in the part of thecountry where farm payments are most important.That suggests that any problems caused by the lossof farm real estate values would be regional. But,

even here the problems are limited. For example, inIowa much of the cropland has program base. Iowaaveraged $1.3 billion in payments per year in2002–05, which, using the ratios above, wouldtranslate into a loss of land values of about $6.5 bil-lion. The value of Iowa’s farmland was about $102billion in 2006. So the estimated loss in land valuewould be about 6 percent. Even if the loss weretwice that estimate, land values would decline byonly a little over 10 percent of the 2006 level.Furthermore, given that land values in Iowa haverisen by over 50 percent between 2002 and 2006,the loss would simply offset a single year’s gains dur-ing those years.5

Second, the majority of cropland is not owned bythe farm operator who grows crops on it. Thus thelosses would be spread over a substantially largerpopulation than the farmers who own cropland.And, the farmers who own the smallest percentageof the land they farm are the larger commercial grainproducers. These are farmers who in some cases dohave large debts: for machinery and operatingfunds, more than for real estate acquisition. But lossof payments tied to land would be a smaller loss forthem. Losses in the value of land they own would beoffset by reduction in their payment for land theyrent from landlords.

Third, small farms, which are the ones where thefarmer is predominantly also the landowner, havestill less dependence on farm income and land as anasset in their total wealth position, and have less realestate debt, than the larger, commercial farms. So anend to commodity programs would not triggerfinancial stress for them either.

Fourth, it is important to remember that the farmcommunity will not lose nearly as much as taxpayersgain when payments end. In part this is because mar-ket prices will rise when surplus production inducedby the program ends. Moreover, the government’scosts of operating the agencies that operate the pay-ment and support programs and the crop insuranceprograms are approximately $3 billion per year(Glauber 2007; USDA 2007b). This is spending thatputs nothing in the pockets of farm operators,landowners, or other agricultural interests, and so will


10

not be missed by any of them. Still further, raising thetaxes (or increasing the deficit that will have to becovered by future taxes) to fund commodity pro-grams imposes costs on the economy through effi-ciency losses that are just as substantial for decoupledpayments as for any other program spending. These“deadweight losses” of taxation at the margin havebeen estimated at 20 cents per dollar of revenueraised as a conservative estimate (Ballard, Shoven, andWhalley 1985). Thus an end to $16 billion in annualbudgetary cost of farm programs (including cropinsurance) will generate an additional $3 billion inreal gains to the U.S. economy.

Fifth, the present time is propitious for an end toprograms because of the strength of commoditymarkets and the expectation that the strength willcontinue over the next decade. The increased finan-cial caution of farmers and other investors in agri-culture since the 1980s has meant that land priceshave only recently returned to their nominal levelsof thirty years ago. From current levels, the pro-jected strength of grain markets attributed toexpected future demands for bioenergy is likely toincrease the value of cropland. In this environment,an end to farm programs will not mean $16 billionless per year, but rather less than half that, becausethe only loss is that of direct payments. This losswould perhaps place a damper on the enthusiasmwith which current optimistic market expectationsare raising already high land prices even higher. Thislast point is perhaps most important in sealing thecase for reform now.

The Long Term with a Free-Market Agricultural Policy

Over the longer term (beyond the eight to ten yearsof the baseline projections discussed earlier), thedownsides of an end to commodity programs will belessened, and the upsides heightened, for allAmerican economic interests. First, the UnitedStates would be in a better position to obtainimproved access to foreign markets for U.S. agricul-tural products, a more promising source of income

for U.S. farming than government subsidies couldhope to provide—as argued in the papers of Sumner(2007b) and Josling (2007).

Second, while current programs have alreadyeliminated many of the inefficiencies generated bypast excesses in commodity market managementand supply control, these features persist in dairyand a few other commodity market interventions,and support prices still distort market signals sig-nificantly in sugar and cotton. An end to these relicsof past mismanagement would permit improvedefficiencies that would benefit consumers and tax-payers, and quite likely producers too. That U.S.farm commodity producers have nothing to fearfrom free markets has been demonstrated by thesuccess of those parts of the U.S. agriculturaleconomy that have never relied on commodity pro-grams, notably livestock, hay, and the collection offruits, vegetables, and other commodities that some-times go under the label of specialty crops (see Paggi2007). Farmers have responded to value-added andother sales opportunities such as organic productionthat have enabled them to prosper without any sig-nificant subsidies of the kind that commodity pro-grams have provided.

Table 1, on the following page, shows that farm-ing in states with high reliance on government payments—those states with agriculture productionconcentrated in grains, oilseeds, and cotton—is notmore profitable than farming in states with littlereliance on government payments. California andFlorida produce relatively little of the program cropsand receive little in government payments relative tothe size of their agricultural output. Agriculture inIowa, Mississippi, and North Dakota is devotedmuch more to grains, oilseeds, and cotton andreceives a much larger share of government pay-ments relative to the value of their farm output. But,agriculture is no better off in the states with highgovernment payments in terms of net farm incomerelative to the value of total agricultural output, or interms of net farm income per farm in the state.

Third, the funds now expended could bereturned to taxpayers or turned to other publicgoods that have benefits greater than their costs.


11

This could occur in the short run as well as thelonger run, but in the long run agricultural policyreform would be notable as an element in a broaderprogram of balancing the federal budget and return-ing to a more limited government that wouldinvolve reform in many areas of government spend-ing. Agriculture is only a small part of the federalbudget—but it is representative of, and a long-standing contributor to, a political culture of waste-ful spending directed at special interests. The recentIraq spending bill, which depended for passage inthe House of Representatives on the inclusion ofbillions of dollars of special funding for peanut stor-age, Midwest drought payments, and spinach grow-ers, among other agricultural interests, is an exampleof how individually small programs coalesce to amultibillion dollar aggregate of economic mischief.A principled end to commodity programs wouldover the longer term go a long way toward curingthis economic ill.

Finally, the experience of other countries, asdescribed in the paper by Alston (2007b), is heart-ening in its demonstration that the almost completeend of formerly highly protectionist policies inAustralia and New Zealand has worked out wellafter a period of adjustment. These reforms havebeen in place for thirty years now in the case of New Zealand. Despite being implemented in aperiod of farm financial crisis, exacerbated byEuropean and U.S. export subsidies depressing the

world markets for their farm products, the experi-ence in both New Zealand and Australia has beenpositive. Consumers and taxpayers have clearlygained, and producers’ incomes have grown, andthey are not asking for a return of governmental pro-tection from market forces. This is further supportfor the belief that after U.S. commodity support pro-grams are gone for a time, our producers will notwant them back either.

Answers to Some Objections

It is natural that proponents of current farm pro-grams are plentiful, since payment recipients facepotential losses if these programs were ended—atleast in the short run. But, whether fueled by self-interest or public interest concerns, other partici-pants in the political process have also voicedarguments in favor of retaining current U.S. farmprograms. Sumner (2007a) outlines a dozen ratio-nales for offered by farm program supporters; a fewof these were introduced in the first section in thisoverview. They can be summarized as a list of allegedmarket failures: persistent characteristics of the mar-ket environment that warrant government interven-tion, some argue. These failures would negate theefficiency gains that this overview has suggestedwould flow from ending the programs, supporters offarm programs argue.

12

TABLE 2

GOVERNMENT PAYMENTS AND NET VALUE ADDED, SELECTED STATES, 2000–05 AVERAGE

California Florida Iowa Mississippi North Dakota

Share of value of ag. production 0.06 0.01 0.43 0.24 0.54from cotton, grains, and oilseeds

Ratio of government payments to 0.02 0.02 0.12 0.16 0.19gross value of production

Ratio of net farm income to gross 0.25 0.36 0.21 0.32 0.25value of production

Net income per farm ($thousand) 95 60 33 31 33

SOURCES: USDA, Economic Research Service, Farm Income Data Files, http://ers.usda.gov/data/FarmIncome/finfidmu.htm. USDA, 2002 Census ofAgriculture, http://www.nass.usda.gov/Census_of_Agriculture/index.asp.


Supporters of the status quo argued the followingpoints:

• Farmers cannot be expected to cope with theuncertainty and variability that unregulated mar-kets generate.

• Food has special characteristics as a good neces-sary to health and survival that calls for a govern-ment regulatory role.

• Farmers cannot earn fair returns in unregulatedcommodity markets because of special character-istics of farming, the regulations they face, or the market power of companies that buy farmproducts.

• Foreign governments through their policiesimpose costs on U.S. agriculture that call for aU.S. policy response.

• In unregulated markets, only large, industrialfarming would survive and there is social value inpreserving small-scale, diversified, low-technologyfarming.

These objections are taken up in turn.

Dealing with Variability and Risks in Agriculture.There is no doubt that agricultural prices and yieldsare variable from year to year. Farmers well under-stand such variability and take measures to reducethe implied income variability, including recognizingthat low yields often engender higher prices—so formany farmers, market conditions create a naturalhedge. Nonetheless, uncertainty and variability areundoubtedly serious issues for farmers for two rea-sons. First, variable returns can cause mistakes inallocating resources. Second, variable incomes cancause problems for household consumption. Thequestion is whether remedies through farm policiesare a better solution than private markets can pro-vide. Those farm programs aimed at price variabilityhave focused on government acquisition and storageof commodities at guaranteed minimum producerprices, idling acreage in years when commoditystocks have become large, and providing paymentsthat establish a floor for producers, while letting

market prices decline to equate supply and demand.The first two of these approaches have already beendeemed a failure by a sufficient political coalition ofbuyers and producers of farm products, for reasonsdescribed earlier. Payments to farmers in years of lowprices are better accepted politically, but as dis-cussed, their actual effect on stabilization of agricul-ture is limited and their economic costs areunacceptably high.

Insurance programs aimed at crop failure andother forms of output risk can be better targeted atrisk management as opposed to income transfer, but, as discussed in depth in Glauber (2007), theseprograms have huge costs for the small risk reduc-tion they have achieved. In 2005, farmers received$3.1 billion in indemnity payments and paid $2 bil-lion in insurance premiums, so they experienced anet gain of $1.1 billion in protection against risk fromthese programs. However, the government’s pre-mium subsidies, payments to insurance companies,and administrative costs added up to $3.0 billion. Sothe taxpayers paid $3 for every $1 in protectiongiven to producers. This experience is typical ofrecent years and can be expected in the future.

Not only are these programs hugely inefficient,but there are private insurance markets for manyagricultural risks—and there would be more if gov-ernment-subsidized and regulated insurance did notprovide unbeatable competition. In addition, for-ward pricing and futures markets provide price riskprotection that many farmers use. Perhaps the mosteffective risk management tools in commercial agri-culture today are contracting arrangements with farm product buyers—prevalent in broilers, eggs,processed and fresh fruits and vegetables, andincreasingly, in hogs and cattle—where growers arepaid agreed-upon prices or service payments, plusincentive payments for delivery of specified com-modities. Finally, there is no reason to believe thatgovernmental provision of subsidized risk manage-ment is more appropriate or necessary for agriculturethan for any other industry.

Food Security and Price Stability. The argumentthat food is special is associated with the fact that


13

U.S. farmers produce the raw materials for a low-cost, reliably supplied, and generally safe supply offood for consumers. This sterling performance is afact, but the association of this success with com-modity programs is not. Again there is abundantevidence, by contrast, between the large segments ofU.S. agriculture that do not have program protec-tions and the commodities that do. Supplies ofunregulated crops and meats are no less reliable insupply or quality than the subsidized commodities.Furthermore, there is no evidence that importedfoods are any less secure than domestically producedfood. The United States has not been plagued withtainted coffee, bananas, or winter vegetables, nor areimported Canadian wheat or Mexican tomatoes anyless reliable than supplies from North Dakota orCalifornia. The mantra of the United States in urgingother nations to open their food markets to ourexports is that imports are just as reliable as domes-tic production, and diversified imports are probablymore reliable.

In addition, farm costs account for less than 10 percent of the retail value of food for the subsi-dized crops. The feed grains, such as corn and barley,and oilseeds, such as soybeans, enter the food supplyonly as feed for livestock or in highly processed foodproducts. Wheat, the most important food grain,goes though the baking, pasta, or processed cerealindustries before reaching the retail market and, ofcourse, the cotton subsidy has no relation to foodconsumption. Finally, most of the consumer foodbudget is spent on foods that receive little or nosupport—and there is no more concern about secu-rity of access to oranges or pork chops than there isabout access to corn flakes.

The general surge in farm prices in the mid-1970sraised alarms for some about the potential jump infood prices. What did the commodity programs—which at that time were more geared to stabilizationand stockpiling than today, and so made more differ-ence in the stability of food prices than today’s pay-ment programs—accomplish in this situation? Verylittle. The stocks did not appreciably slow the pricerises, and U.S. policies quickly turned to grain exportcontrols to keep commodities at home rather than

being shipped abroad. This had little effect, too,according to USDA’s assessment after the fact (USDA, ERS 1986), but the farm community’s spir-ited objection to policies that would attempt to hold prices down prevailed to an extent that “nograin embargoes” become a mantra of candidates for national political office for thirty years afterwards.In short, U.S. agricultural policy has never beeneffective at food price stabilization, and simply can-not do this job. We have learned not to rely on gov-ernment to keep prices low when market conditionscall for them to be high; now it is time for us to stoptrying to keep prices high when market conditionscall for them to be low.

Farmers’ Lack of Market Power. A longstandingconcern of farmers has been that agricultural productprices are lower than they should be because of themarket power of processors and others who buyfarm products. This concern has been recognized inthe past by special allowances for farm marketingcooperatives and antitrust action against meatpack-ers and investigations of big grain companies. Bresterand Smith (2007) review the area where these con-cerns have been expressed most intensely in recentyears: in livestock slaughter and contacting of pro-cessing firms with livestock growers. The number ofcompeting buyers for the output of a typical livestockproducer is declining at many locations and forseveral categories of animals, but Brester and Smithshow that the lower costs of larger firms offset anyincrease in marketing margins that lessened compe-tition might have engendered. Nonetheless, there is alegitimate role for USDA’s Grain Inspection andPackers and Stockyards Administration, along withthe U.S. Department of Justice and the Federal TradeCommission to play in this area. It is not a promisingapproach, however, to have commodity supportprograms for livestock on this basis, and producershave no interest in such support. The existing com-modity programs have no ability to target imperfectcompetition in the products they cover.

Harm to U.S. Agriculture Due to Other Countries’Actions. Foreign governments, more obviously than


14

domestic food processors, have taken actions thathave harmed the interests of U.S. farmers by reduc-ing market outlets for their products. They do thisthrough import restrictions against products fromthe United States and domestic and export subsidiesto favor their own. Combating these efforts is animportant task for the U.S. government, as discussedfurther below, but maintaining our own restrictionsand subsidies generates costs greater than the bene-fits, taking a bad situation and making it worseinstead of better. Removing U.S. subsidies wouldremove one rationale that other countries use tojustify their own subsidies and trade barriers.Furthermore, as other countries have challenged theprice-depressing impacts of U.S. programs, theUnited States is also free to use World Trade Organi-zation (WTO) procedures to challenge programs ofother countries deemed to be in violation of theirinternational obligations.

Promoting Small-scale Farming. The idea of agri-cultural policy to promote traditional small-scale fam-ily farming is often repeated but is unrelated to thefarm programs that now operate in the United States.Hence even if one accepts the appeal of a return tosome version of the historical roots of family farming,this line of objection to free market agriculture doesnot improve the case for current programs. As shownabove and, for example in Kirwan (2007), the currentfarm programs reward farms roughly in proportion totheir output and in most cases are roughly neutral tofarm size and technology. They do tend to reduce theincentive for traditional livestock and crop mixedfarming, and by leaving out specialty crops, reducethe further diversification of farms eligible for pro-gram payments. Of course, many participants in com-mercial agriculture would object strongly to farmprograms that supported what they see as less effi-cient small-scale farm operations to the exclusion ofthe larger commercial farms.

Payment Limits and Means Testing. An oftenheard objection to current payment programs is thatpayments go to people who are rich and to farmsthat garner very large payments. This issue has been

a staple of farm policy debates for four decades. The2007 Farm Bill proposal from the USDA and severalprominent Representatives and Senators have pro-posed various forms of means testing or paymentlimits as reforms of existing programs. The reformhas obvious appeal if one considered the proper roleof commodity programs to be welfare programs tohelp the needy. But, as noted above, farm programpayments would need to be completely recast to tar-get the farm poor or the farms that are in severefinancial risk. From an efficiency perspective, itmakes no sense to support high-cost failing farmsand implicitly penalize successful operations.Clearly, a policy to foster a strong industry and effi-cient resource use would try to encourage low-costproduction—not high-cost production.

Furthermore, as documented above, given theincome and wealth of farm households both large andsmall, no welfare objective for farm program pay-ments makes sense. But a policy to reach the farmpoor could use income categories, as in the tax code.Indeed, an appropriate policy reform from this per-spective would be to end commodity supports anduse some of these funds to provide payments alongthe lines of earned income tax credits for farm house-holds with very low incomes. From that perspective,the cut off for payment eligibility would be in therange of $20,000 to $30,000 per household: not the$200,000 adjusted gross income limit that Agri-culture Secretary Mike Johanns has proposed. Fromthe welfare perspective, why should farmers betreated differently from anyone else? That considera-tion leads back to agricultural policy as having indus-trial policy aims, not income redistributional aims. Ifthe purpose of farm programs is to regulate indus-tries, a focus on means tests or payment limits is alame reform. The only sensible reform is a real onethat eliminates the payments completely.

Some More Productive Roles for Government in Agriculture

Let us turn now to productive roles for governmentin agriculture, ones that have better prospects of


15

generating economic benefits that exceed their costs.Such efforts are now largely crowded out by theattention given to farm subsidies and the budget allo-cated to transfer payments.

Agent of U.S. International Trade Interests. Someother countries have hampered U.S. agriculture byplacing restrictions on sales of U.S. farm products.The U.S. government has the crucial and difficulttask of eliminating these barriers wherever possible.Steps in this direction have been taken in WorldTrade Organization negotiations and in bilateraltrade agreements between the United States andcountries such as Chile, and in regional agreementssuch as the North American Free Trade Agreement(NAFTA) with Canada and Mexico, and the CentralAmerican Free Trade Agreement (CAFTA). How-ever, these agreements in total have taken only babysteps toward free trade as compared to what poten-tially could be accomplished, and they complicatebroader free trade efforts. Of course, improvementsin international trade arrangements cannot beaccomplished without the United States being will-ing to remove or reduce its subsidies and allow for-eign access to U.S. markets, notably the long-standing sugar import restrictions and the morerecent tariff on imported ethanol.

Among the concerns are trade restrictions basedon dubious claims of health or safety risks to plants,animals, or humans, such as bans on imports of beefand cattle from countries with minimal risk of madcow disease and bans on genetically modified prod-ucts in Europe and in developing countries. Theseare paralleled by U.S. insistence on modified domes-tic labor or environmental policies in other countriesas a condition for importing their products into theUnited States. Reforms in these areas are important,but can be done only by improved federal policy, nota retreat from governmental involvement.

Conservation and Environmental Improvement.In the areas of conservation of soil and other naturalresources such as endangered species, and environ-mental protection, the studies by Antle (2007),Heimlich (2007), and Kuminoff (2007) examine

aspects of these complex matters. Evaluation of theConservation Reserve Program, currently the largestconservation and environmental program, show thatit has generated substantial benefits whose estimatedvalue exceeds the program’s considerable costs. Whatis perhaps surprising is that the main benefits are not reduction of soil erosion or improvements inwater quality, but rather the provision of wildlifehabitat. Heimlich (2007) develops a set of recom-mendations for targeting funds better to achieveenvironmental goals. The recommendations focus onmore precise identification of land parcels and prac-tices in the context of surrounding ecological condi-tions, unifying the current splintered set of programs,and better judging whether acreage retirement oraltered practices on cropped acreage is most appro-priate. With respect to land retirement, the issue isone of transition from the existing 10- or 15-yearcontracts to a more lasting structure of conservation,perhaps through purchase of permanent easementsby which farmers give up rights to grow crops on cer-tain acreage.

Other environmental concerns—such as poorwater quality in the Chesapeake Bay, the SacramentoRiver delta, and the Gulf of Mexico; and broad-basedissues such as global climate change—also relate toagriculture and are less well addressed by idlingcropland. In general, the measurement of positive ornegative environmental services from agriculture,and how they should be subsidized or taxed, is stillan area of conceptual, empirical, and political uncer-tainties. Considerable additional analysis of both thebenefits and costs is needed to inform public policychoices. Antle (2007) argues for a system of pay-ments for ecosystem services that would replacecurrent subsidy approaches. This would move thebasis for policy beyond the interest-group politics ofredistribution, to take advantage of the desire offarmers and other citizens to do the right thing withrespect to conservation and environment whiledirecting such efforts in the most productive ways.

Advancing Bioenergy. Bioenergy has become a cen-tral focus of the 2007 Farm Bill and is likely to be anagricultural policy issue for years to come. As shown


16

in the paper by Miranowski (2007), the existingethanol and biodiesel subsidies are not promising as aremedy for either the environmental or energy secu-rity objectives that have been used to justify them.Research and development spending in this areacould prove productive, but should be undertakenwith caution. The concern is that a new research anddevelopment focus on bioenergy would drain fundsfrom research areas of demonstrated success. More-over, short-run applied research and demonstrationprojects are too likely to devolve into support fordubious prospects that would garner private invest-ment if they had a strong likelihood of success.

Promoting Rural Development. Rural develop-ment, in the form of maintenance or growth of bothagriculture-related and non-agricultural enterprises,has been an important contributor to the historicalrise in farm household incomes. Studies by Renkow(2007) and Kilkenny and Johnson (2007) haveinvestigated the possibilities for public investmentor other policies aimed at fostering rural develop-ment. Existing subsidy and investment programshave dubious prospects for meeting a benefit-costtest, as Renkow’s paper shows. But Kilkenny andJohnson see scope for more productive federalefforts if they leverage rather than substitute for localfunding of local public goods, which may encouragethe small and most remote rural communities torecombine to achieve and sustain critical mass levelsof population.

The possibilities for promoting traditional agricul-ture have been enlarged in the discussion of ruraldevelopment to include direct sales of farm productsto retail outlets or farmers’ markets or buying clubs,value-added production on farms in which farmerstake on some processing tasks, organic production,and increased roles for farmer cooperatives. Farmspursuing these paths are thriving in many parts ofthe country where consumers support the goods andservices they provide. Should these sorts of activitiesbe subsidized or promoted under federal farm pol-icy? If reforms of current policies were to undertakethese forms, the policies would not fit well undereither of the two broad policy objectives stated ear-

lier: the prosperity of agriculture as an industry, andthe well-being of farm people. The tradition-mindedobjectives have more in common with policy for thearts or historical preservation: cultural policy bettersuited to local or state preferences, or if federal,through grants to states or local institutions to pro-mote the traditions they hold dear.

Enhancing Long-term Agricultural Productivity.Some of the most cost-effective investments in U.S.agriculture have been in the areas of research anddevelopment and other services to protect andenhance agricultural productivity. Yet these invest-ments have tended to be underfunded: in partbecause farm subsidy programs have captured thelion’s share of the agriculture budget. Moreover, suchareas are unlikely to receive adequate investmentfrom the private sector: precisely because they haveelements of public goods. Within the narrow politi-cal calculus of the Farm Bill reauthorization, the spe-cialty crops industries have made additionalinvestments in protection from invasive species andadditional funding for research the central features oftheir claims to redress the balance of Farm Bill atten-tion across commodities, and a more general casealong these lines can be made. A positive farm policyfor the United States would improve agricultural pro-ductivity and competitiveness by cost-effective fund-ing of basic infrastructure that would not receiveadequate private sector investments. Such a policycould benefit consumers, producers, and the economyat large.

Conclusion

The AEI studies have reviewed the evidence on com-modity programs and other key elements of U.S.agricultural policy, and how these programs and poli-cies fit in with the objectives of agricultural policyand the economic situation in rural America today.We conclude that while there are productive roles forgovernment to play in U.S. agriculture, the currentfocus on commodity programs is misplaced as asolution to problems that exist, and has imposed


17

costs on taxpayers far in excess of the benefits theydeliver to either rural America or the nation as awhole. The 2007 Farm Bill could take a significantstep toward improving the nation’s welfare by start-ing the process of eliminating those programs, andby adopting prudent reforms in crop insurance, con-servation, bioenergy, and rural development.

Notes

Bruce Gardner is professor, Department of Agricultural andResource Economics, University of Maryland. Daniel A.Sumner is director, University of California AgriculturalIssues Center; Frank H. Buck, Jr. Professor, Department of Agricultural and Resource Economics, University ofCalifornia, Davis; and member of the Giannini Foundation.Paper prepared for American Enterprise Institute project,Agricultural Policy for the 2007 Farm Bill and Beyond,directed by Bruce Gardner and Daniel A. Sumner. Theviews expressed here are those of the authors and not thoseof any institution with which they are affiliated.

1. For a fuller discussion and analysis of the purposes ofagricultural policy, see Sumner (2007a).

2. For further details and discussion of the economicsituation in U.S. agriculture, see Gardner (2007).

3. A farm is defined as an operation that produces orwould normally produce $1,000 worth of agricultural out-put, so clearly very small operations are included in the dis-tribution.

4. For detailed sources, see Gardner (2007). 5. Data from Iowa State University’s Farmland Value

Survey, which estimated the average value per acre at$2,083 in 2002 and $3,204 in 2006.

References

Papers cited in this overview that were prepared for AEI Agricultural Policy Series: The 2007 Farm Bill andBeyond. American Enterprise Institute. http://www.aei.org/farmbill

Alston, Julian M. 2007a. “Who Really Benefits from U.S.

Farm Subsidies?”

———. 2007b. “Lessons from Agricultural Policy Reform in

Other Countries.”

Antle, John. 2007. “Payments for Ecosystem Services and

U.S. Farm Policy.”

Babcock, Bruce A. 2007. “Money for Nothing: Acreage and

Price Impacts of U.S. Commodity Policy for Corn,

Soybeans, Wheat, Cotton, and Rice.”

Balagtas, Joseph V. 2007. “U.S. Dairy Policy: Analysis and

Options.”

Barrett, Christopher B. 2007. “U.S. International Food

Assistance Programs: Issues and Options for the 2007

Farm Bill.”

Beghin, John C. 2007. “U.S. Sugar Policy: Analysis and

Options.”

Brester, Gary W., and Vincent H. Smith. 2007. “Agricultural

Policy and the U.S. Livestock Industry.”

Gardner, Bruce. 2007. “Does the Economic Situation

of U.S. Agriculture Justify Commodity Support

Programs?”

Glauber, Joseph W. 2007. “Double Indemnity: Crop

Insurance and the Failure of U.S. Agricultural Disaster

Policy.”

Heimlich, Ralph E. 2007. “Land Retirement for Conserva-

tion: History, Analysis, and Alternatives.”

Josling, Tim. 2007. “The Impact of the WTO and Bilateral

Trade Agreements on U.S. Farm Policy.”

Kilkenny, Maureen, and Stan Johnson. 2007. “Rural

Development Policy.”

Kirwan, Barrett. 2007. “Distribution of U.S. Agricultural

Subsidies.”

Kuminoff, Nicolai V. 2007. “Public Policy Solutions to

Environmental Externalities from Agriculture.”

Miranowski, John. 2007. “Biofuel Incentives and the Energy

Title of the 2007 Farm Bill.”

Paggi, Meckel S. 2007. “U.S. Specialty Crops and the 2007

Farm Bill.”

Renkow, Mitch. 2007. “Infrastructure Investment and Rural

Development.”

Sumner, Daniel A. 2007a. “Farm Subsidy Tradition and

Modern Agricultural Realities.”

———. 2007b. “The Farm Bill and WTO Compliance and

U.S. Farm Programs.”

Other references cited in this overview:

Ballard, C.L., J.B. Shoven, and J. Whalley. 1985. “General

Equilibrium Computations of the Marginal Welfare Costs

of Taxes in the United States,” American Economic Review

75: 128–38.


18

CBO (U.S. Congressional Budget Office). 2007. The Budget

and Economic Outlook: Fiscal Years 2008 to 2017. U.S.

Congress (January).

CEA (Council of Economic Advisers). 2007. Economic Report

of the President. Washington, DC: U.S. Government

Printing Office.

Covey, T., R. Green, C. Jones, J. Johnson, M. Morehart, R.

Williams, C. McGath, A. Mishra, and R. Strickland. 2005.

Agricultural Income and Finance Outlook. USDA, Economic

Research Service (November).

FAPRI (Food and Agricultural Policy Institute). 2007.

Downloads available at http://www.fapri.org

Hoppe, R. A., and D. E. Banker. 2006. “Structure and

Finances of U.S. Farms.” Economic Information

Bulletin No. 12 (May). USDA, Economic Research

Service.

USDA (U.S. Department of Agriculture). 2004. 2002 Census

of Agriculture, Volume 1, Part 51, United States Summary.

AC-02-A-51 ( June).

_____. 2007a. USDA Agricultural Projections to 2016. Long-

Term Projections Report OCE-2007-1 (February).

_____. 2007b. 2008 Budget of the U.S. Department of

Agriculture. http://www.obpa.usda.gov/budsum/fy08bud

sum.pdf

USDA, ERS (U.S. Department of Agriculture, Economic

Research Service). 1986. Embargoes, Surplus Disposal, and

U.S. Agriculture. Agricultural Economic Report No. 564

(December).

U.S . AGRICULTURAL POLICY REFORM IN 2007 AND BEYOND

19

JUSTIFICATIONS FOR EXISTING POLICY

There are two alternative views of the economicsituation of American agriculture. One is that U.S.farming, despite periodic commodity booms, is inchronic economic trouble and government supportis necessary to keep the sector viable; the other isthat U.S. farming is basically strong and is receivingassistance it does not need. The preponderance ofevidence favors the strong-agriculture view.

Income for all farm households is about a thirdgreater than that of non-farm households; incomefor the large-scale farm households who receive thebulk of farm subsidy payments is over three times aslarge as non-farm households. Farm households arealso an average of five times wealthier than theirnon-farm counterparts.

The implication is that current U.S. farm policiesare satisfying political needs, not economic ones,and ending them would not be damaging to agri-culture as some fear.

Economic Trends and the Situation in Agriculture

Farm Output, Inputs, and Productivity Growth.U.S. agriculture has become ever more productive as new technology and capital have substituted for labor. While labor use on farms has declined

substantially since 1950, cropland and traditionalcapital are roughly constant over the longer term,and the use of material inputs (fertilizers, fuels, pur-chased feed additives) has about doubled. Agri-cultural output has continued to grow at a steadyclip so that total factor productivity shows animpressive trend rate of 1.9 percent, as continuingadvances in technology have kept the real cost ofU.S. farm products on a pronounced downwardpath. There is no evidence of a slowdown in thistrend in recent years despite energy price shocks,environmental constraints, and concerns aboutexhaustion of productivity gains attributable to ear-lier technological innovations.

These cost declines have been largely reflected inlower prices of farm products and hence lower costsof raw materials for foods. The average price of farmproducts rose 40 percent between 1978 and 2005,an annual rate of increase of about 1.2 percent.However the overall price level rose at an annualrate of 3.2 percent. Thus the real price of agriculturaloutput fell by an average of 2.0 percent during thisperiod—essentially the same as the rate of produc-tivity growth.

Farm Income and Wealth. The parallel declines in farm prices and cost decreases indicate that farmers have had to have greater than average cost

23

Does the Economic Situation of U.S. Agriculture Justify Commodity Support Programs?

Bruce L. Gardner

reductions in order to profit from farm productivitygrowth. This has occurred for larger, commercial-scale farms, but not for small farms, on average. Yetthe incomes of people living on both large and smallfarms have in fact grown in real dollar terms.Moreover, farm incomes on average have grownfaster than the incomes of the non-farm population,and now exceed non-farm incomes regardless offarm size.

From the 1930s Depression/Dust Bowl yearsthrough the 1960s, farm households could be rea-sonably categorized as a low-income population.Farm family incomes reached sustained equality withtheir non-farm counterparts by the 1990s, however,and since 2000 farm household incomes have beensignificantly higher than nonfarm incomes.

A key factor in this pervasive income increase isthe economic integration of farm households into thenon-farm economy. In recent years off-farm incomesources account for 80 percent or more of averagefarm household incomes. In 2004, the mean incomeof farm households was $14,000 from farming and $67,000 from off-farm sources, for a total of$81,000. The comparable figure for non-farm house-holds: $60,000.1

These data suggest that income from farming itselfmay indeed be quite low. However, a full understand-ing of the farm income data requires consideration ofdifferences between farms of different sizes. A major-ity of farms, 87 percent of them in 2004, have lessthan $25,000 annually in sales. The costs of farmingat this scale are such that these farms on averageearned only $1,020 from farming, and more than halfare estimated to have losses from their farm enter-prises.2 Nonetheless, the average household incomeof these farms is $71,500 thanks to off-farm income.

It might be thought that this situation describes atroubled agriculture rather than a prosperous one—farmers can’t make money farming so they have totake off-farm jobs. This is a fair characterization forsome mid-size farms, but not for these small-scalefarms. There is no practically conceivable reductionin costs or rise in commodity prices that could makethese farms profitable enough to generate householdincomes equal to the U.S. average. These farms are

more properly considered, for the most part, asavocational farms. The people who live on them arededicated to farming and the rural life, but do notwant to expand to the scale and commercial orienta-tion necessary to achieve an enterprise that will itselfsupport a household at the U.S. average incomelevel. The benefit of off-farm earning opportunities isthat they permit such avocational farms to surviveand at the same time provide the operator’s familywith an income that, on average, exceeds that of theaverage non-farm household.

Family farms classified by USDA as commercialscale operations (those which have $250,000 ormore in sales) were even healthier. Households onthese large-scale farms earned an average of$145,300 from farming plus $46,038 from off-farmsources—over three times that of the average non-farm household.

With respect to the balance sheet of farm enter-prises, the financial health of the farm sector is alsoexcellent. The overall ratio of debts to assets in 2005is estimated by USDA at 13.4 percent, and lower inthe 2000–2005 period than in any other period inwhich comparable estimates have been made. About60 percent of farms were debt-free in 2005, com-pared to 40 percent in 1985. The median net worthof farm households, including the smallest farms,was estimated at $460,000, close to five times the$92,000 estimated for all U.S. households.

Relation of Farm Commodity Programs to the Farm Economy

U.S. commodity programs were introduced in the1930s when farm incomes were low. How much ofthe improvement since then can be attributed tothese programs? And, if the programs were to begreatly cut back or eliminated, what would happento farm household incomes and wealth?

USDA payments to producers averaged $16 bil-lion annually during 2002–2005. These paymentsare concentrated on a subset of commodities.Commodities accounting for 58 percent of the valueof U.S. farm output in these years received no


24

payments at all. The top four commodities inreceipts of payments (corn, wheat, cotton, and rice)received 79 percent of all commodity paymentsalthough they accounted for only 18 percent of thevalue of farm output.

Payments are also concentrated on relatively fewfarmers. In the 2002 Census of Agriculture, onlyone-third of farm operators reported receiving anygovernment payments at all, and 3 percent of thefarms received 52 percent of the total payments.

Table 1 shows payments by type of farm. Grain/oilseed and cotton farms get far more governmentassistance than the other three categories of cropfarms. However, they do not have higher net cashincomes than other farms. Dairy farms do get higherincomes than beef and hog farms, and have highernet cash incomes on average, but poultry farms havestill higher incomes with the least government sup-port among livestock farms. Overall these data giveno indications that production of the supportedcommodities is more profitable than production ofnon-supported commodities.

This is not to say that the removal of current pro-grams would have no effects upon the economic sit-uation of the farms that now benefit from thoseprograms. Other papers in this project provide evi-dence on that issue.3 The conclusion is that ownersof land who now get commodity program paymentsfor the grains, cotton, and rice would experiencelosses of wealth, as land values would decline,which could be as much as 20 percent of a farmer’swealth for farmers whose assets are very largelyundiversified and tied up in program cropland. Butincome flows to producers would not be greatlyaffected. And the wealth losses would not generateappreciable increases in the risk of bankruptcies orsevere financial stress because of the low debt-to-asset ratios that now prevail in U.S. agriculture.

Not surprisingly, eliminating current Title 1 pro-grams would have different effects on different farm-ers in different situations. Those who are part-timefarmers relying primarily on off-farm sources ofincome undoubtedly experience some loss of rev-enues when payments are reduced or market prices

fall. But they are only rarely at riskof bankruptcy or foreclosure onmortgaged land, and when they areat severe financial risk, their lowshare of commodity returns in theirtotal financial picture means thatno conceivable level of price sup-port could make them solvent.Larger, highly leveraged farmsappear more likely to be at risk offinancial harm. But their costs aretypically lower, so they have moreof a cushion to take some reductionin commodity returns without gen-erating negative net income.

The variety of economic situa-tions of U.S. farms is the basis foran analysis by the EconomicResearch Service of USDA basedon the joint distribution of incomeand wealth of farm operatorhouseholds on family farms. ERSconsidered four categories of

DOES THE ECONOMIC SITUATION JUSTIFY COMMODITY SUPPORT PROGRAMS?

25

TABLE 1GOVERNMENT PAYMENTS AND NET CASH INCOME ($)

Per Farm

Net cash Payments

NAIC category* Payments income Value of sales per dollar sales

Grain and oilseeds 8,404 28,146 107,567 0.078

Cotton 26,552 56,552 261,379 0.102

Vegetables 2,919 121,705 379,913 0.008

Fruits and nuts 721 35,695 140,951 0.005

Greenhouse and nursery 157 76,134 233,944 0.001

Beef (ranching) 1,136 2,186 28,883 0.039

Beef (feedlots) 2,739 17,261 412,523 0.007

Dairy 9,724 71,559 313,628 0.031

Hogs 2,926 37,359 366,113 0.008

Poultry 943 161,878 552,285 0.002

SOURCE: 2002 Census of Agriculture, U.S. Summary, Table 59.NOTE: North American Industry Classification System. To qualify, half of a farm’s sales mustbe from product category.


26

farms: those with high income and high wealth,high income and low wealth, low income and highwealth, and low income and low wealth. The crite-rion for “high” versus “low” is being above or belowthe U.S. average (both farm and non-farm) levels ofincome or wealth. The farms with low income arethe prime candidates for financial stress if commod-ity subsidies were to end. But those with low incomeand high wealth are better positioned to adjust to an absence of subsidy support. Data for this groupare shown in the first column of Table 2. Theremoval of support would have an income effect anda wealth effect, reducing the value of land that qual-ifies for payments.

Note first that the removal of $1,000 would notmean that net farm income would decline by$1,000. Taking that much from commodity rev-enues would cause less production, and that wouldmean higher commodity prices.

Consider the low income, high wealth categoryof farms, which received $3,643 on average ingovernment payments. These farms account for 24 percent of the value of U.S. agricultural produc-tion, which is about $50 billion as of 2004, or$62,000 per farm. The elimination of paymentsamounts to 5.9 percent of revenues. With reason-able market price adjustments, the reduction in netfarm income would be about $1,700, or 7 percentof the median farm household income of the low-income group. On the wealth side, the capitalizationof the loss depends on the relevant discount rate.Supposing a 10 percent discount rate, the averagehousehold in this category would lose $17,000, or

4 percent, of wealth, leaving the households with a $380,000average net worth.

The situation for the lowincome, low wealth group is dif-ferent. They get less governmentpayments, $1,300 on average,and have only slightly lowermedian household incomes, sotheir income would decline by asmaller percentage, about 2 per-cent; but their losses of wealth

would be more noticeable, perhaps 15 percent, or$7,000. But here, too, it is hard to see how manyfarm households would be placed at higher risk ofsevere financial stress. Note also that the low income,low wealth group is a small fraction of U.S. farmhouseholds, 3.1 percent of them.

The higher income groups would see slightlylarger losses, but have much higher initial situationsof both income and wealth, so the financial painwould be significantly less for them.

Conclusion

If farm commodity program payments were to end,farm households would experience losses, but theselosses would be far less than the costs to taxpayerswho underwrite these payments and have on aver-age lower incomes than the farm households whoreceive them. An end to payments would not gener-ate significant risks of financial failure among farmenterprises or damage to rural economies. The pre-ceding conclusions are based on conditions of2002–05, when payments averaged $16 billionannually. The consequences of reform would beeven less for the conditions projected in 2007–2015,when current programs would cost about half thatmuch. The main effects of an end to the commodityprograms in the 2007 Farm Bill would be a sub-stantial permanent gain to taxpayers and consumersat the cost of substantially smaller and transitorylosses to farmers and non-farm owners of agricul-tural assets.

TABLE 2FARM HOUSEHOLD INCOME/WEALTH CATEGORIES

Low income Low income High income High incomeCategory high wealth low wealth low wealth high wealth

Median income ($) 23,171 13,461 69,533 85,781

Median wealth ($) 397,700 52,031 42,675 571,823

Number of farms (th.) 803 63 40 1,154

Ave. payment ($) 3,643 1,301 3,537 5,846

SOURCE: Tovey et al., 2005, Table 9.

Notes

This policy brief draws on the author’s longer paper “Does the Economic Situation of U.S. Agriculture JustifyExisting Farm Programs?” available at www.aei.org/farmbill. The longer paper includes background supportand citations not included here. Gardner is a distinguishedprofessor in the Agriculture and Resource EconomicsDepartment, University of Maryland. He is co-director ofthe AEI agricultural policy project. The views expressedhere are those of the author and not those of any institutionwith which he is affiliated.

1. USDA, Economic Research Service, AgriculturalOutlook Tables, ERS website.

2. The data underlying this and subsequent informa-tion about farm household economics are developed bythe Economic Research Service of USDA, using their

Agricultural Resource Management Survey of about10,000 farms annually. Estimates reported here use datafrom Covey et al., 2005.

3. Alston, Julian. 2007. “Who Really Benefits from U.S.Farm Subsidies?” AEI Agricultural Policy Series: The 2007Farm Bill & Beyond. American Enterprise Institute.http://www.aei.org/farmbill. Babcock, Bruce. 2007.“Money for Nothing: Acreage and Price Impacts of U.S.Commodity Policy for Corn, Soybeans, Wheat, Cotton andRice.” AEI Agricultural Policy Series: The 2007 Farm Bill &Beyond. American Enterprise Institute. http://www.aei.org/farmbill. Balagtas, V. Joseph. 2007. “U.S. Dairy Policy:Analysis and Options.” AEI Agricultural Policy Series: The2007 Farm Bill & Beyond. American Enterprise Institute.http://www.aei.org/farmbill. Beghin, C. John. 2007. “U.S.Sugar Policy: Analysis and Options.” AEI AgriculturalPolicy Series: The 2007 Farm Bill & Beyond. AmericanEnterprise Institute. http://www.aei.org/farmbill.

DOES THE ECONOMIC SITUATION JUSTIFY COMMODITY SUPPORT PROGRAMS?

27

Over the decades, farm bills have become ever morecomprehensive and complex, yet the commodityprograms remain the core of the legislation and con-tinue the tradition of subsidy begun with the NewDeal. The federal government has sought to increasefarm incomes through price and income supportsfor selected program crops (and dairy) since 1933.Advocates have proffered many rationales for whysuch transfers from consumers and taxpayers mightbe justified, but these are simply not consistent withthe facts of modern agriculture.

For example, it is impossible to explain the cur-rent farm program payments as helping needyfarmers when most of the payments go to the larger and more successful farms and when farmhouseholds have higher incomes and far higherwealth than non-farm households. Furthermore,most farmers do not receive subsidies: farm subsi-dies are distributed roughly in proportion to out-put of program crops, and these program cropscomprise only about 30 percent of U.S. agriculture.Finally, the economic evidence shows that U.S.agricultural production overall would be onlyslightly reduced by elimination of the programs,while farmers shifted some acreage and otherresources from that part of agriculture that getsmost of the support toward those enterprises thatare largely unsubsidized.

Despite the lack of convincing rationales for theircontinuation, after seven decades, farm programsare well entrenched politically. For decades skepticshave advanced arguments for fundamental policychange, and yet the subsidy programs remain.Nonetheless, there are reasons to think the 2007Farm Bill may be different. Drivers of policy changeinclude federal budget pressures, heighteneddemands for farm conservation and environmentalprograms, growing claims for more attention fromgrowers of unsubsidized crops, pressures to complywith international trade rules, and forecasts ofunusually high commodity prices over the next fiveyears. Hence, 2007 may well offer some uniqueopportunities for substantial reform.

The Long Tradition of Farm Commodity Programs

Major agricultural policy innovations occurred in1862, during the administration of AbrahamLincoln, with the founding of land grant universi-ties, the establishment of the U.S. Department ofAgriculture (USDA), and the implementation ofthe Homestead Act. The federal government con-tinued to stimulate agricultural productivitygrowth by creation of agricultural experiment

29

Farm Subsidy Tradition and Modern Agricultural Realities

Daniel A. Sumner

stations in 1887 and agricultural extension servicesin 1914.

This emphasis on enhancing productivitychanged with the Agricultural Adjustment Act(AAA) of 1933, which created supply controls, pricesupports, and income supports to respond to a col-lapse of commodity prices and farm incomes. Butthe New Deal programs for agriculture did not solvelow price problems. Agriculture remained depressed—with some price relief because of bad weather—untilWorld War II shifted out demand and curtailed sup-ply. Despite the massive government intervention, ittook World War II to bring cash farm income backto the (already depressed) 1929 level. Despite theirgeneral inability to enhance the general prosperity of U.S. agriculture, Farm Bill commodity programshave continued for 75 years.

During this long history, U.S. agriculture has con-tinued to evolve substantially even while subsidyprograms have supported dairy, grains, oilseeds, cotton, and a few other crops, accounting for about30 percent of agricultural output, with most live-stock products, fruits, vegetables, and many othercrops largely outside the subsidy web. Agriculture’sshare of the total economy has declined to about 1 percent of gross domestic product as labor shiftedout of farming. Productivity growth has occurredfaster in agriculture than in the rest of the economy,and farm prices have fallen relative to other prices.The share of farms deriving most of their incomefrom farming has declined to well less than half ofthe total. For commercial-sized farms, which pro-duce the great bulk of agricultural output, the grosssales per farm have grown rapidly. One of the mostimportant changes in the past half century is thatincomes of farm operators have increased fromabout 40 percent of the national average to about120 percent of the national average.

Reasons, Rationales, and Rationalizations forFarm Commodity Programs

Attempts to explain or justify farm programs includerationales that have become clearly outmoded, such

as the claim that farm income subsidies are neededto support farm incomes and combat rural poverty.Others raise issues for which current programs areclearly ill suited, such as the claim that farm subsi-dies are needed to combat market power by com-modity buyers. Some rationales raise economicarguments that do not fit the economic data, such as the argument that farm subsidies are responsiblefor keeping the American food supply affordable.Subsidies for grains, oilseeds, and cotton do reducemarket prices of bulk commodities by a few per-centage points (and, indeed this is a major com-plaint of international competitors, especially inpoor countries), but import protection programs fordairy products and a few other commodities raiseprices. The net effect on retail prices is very small, inpart because the share of bulk commodity prices inthe retail food budget is so small and in part becausemost of U.S. agricultural and most imported foodproducts are outside the reach of U.S. farm subsi-dies. Table 1 shows the relationship between farmprogram payments and value of agricultural outputby commodity. Differences are stark and reinforcethe point that farm programs are specific to a rela-tively small share of U.S. agriculture.

We should also ask: Why would it be appropri-ate to use tax funds to lower the market prices ofselected commodities, rather than allow food pricesto be determined by market forces? Food stampsand other food assistance programs respond to con-cerns of consumption of the poor. Agriculturalresearch and development has been a much moreeffective tool for lowering food prices for Americanand others over many decades because they increaseagricultural productivity. Hence, to continue thelong-term path of lower food prices suggestsexpanded investments in agricultural R&D, perhapstailored to products that improve nutrition, ratherthan continued subsidy of program crops.

Farm programs shift income from taxpayers andconsumers to farmers and other owners of farmresources that are especially useful in production ofprogram crops. If farm programs were removed,recipients would lose income and asset values thathave been propped up by government transfers


30

would fall. But these losses are actually quite small.For example, the price of land in Iowa would likelyfall by a few percentage points, much less than theincrease in land prices that has occurred in just thepast year due to the ethanol-fueled corn price boom.A similar point applies to the economic health ofrural communities. Farm income is such a small

share of rural incomes in almostall areas that loss of farm programswould have tiny impacts. Formany reasons, farm subsidies arenot targeted and are not effectiveas rural development policy.

A few points resonate throughall the economic rationales sug-gested for commodity programs.First, none of the arguments forsubsidy programs account for thecurrent distribution of supportacross commodities. If they applyat all, each of the rationales wouldseem to apply to many non-program commodities just asmuch as to the program crops.Second, even if they suggest arationale for some governmentinvolvement in markets, none ofthe arguments account for theform of price and income supportprograms now used in the UnitedStates. Third, neither commodityprograms nor arguments for themsuggest that a goal is long-termproductivity or long-term healthof the industry. In fact, there is no evidence that the subsidizedcommodity industries are moreinnovative or successful on anydimension than those parts ofagriculture with little or no sub-sidy. Thus, whatever the appropri-ate rationale for commodityprograms, it is hard to claim thatthey solve any long-run problemsof farm commodity industries.

So, why do the familiar farm commodity pro-grams continue? The strongest rationale may be thatwe continue to have farm programs mostly becausewe have had them for three or four generations. Forthe supported commodities, the programs havebeen so thoroughly imbedded into all aspects of theindustry that producers and others find it hard to

FARM SUBSIDY TRADITION AND MODERN AGRICULTURAL REALITIES

31

TABLE 1SHARES OF U.S. CASH RECEIPTS AND PROGRAM PAYMENTSa

FOR SELECTED AGRICULTURAL COMMODITIES, CROP YEARS

2002–2005 AVERAGE (percent)

Share of individual Share of total value commodity payments in

of production total outlays

Upland cotton 1.9 22.3

Rice 0.6 7.3

Wheat 3.0 9.5

Corn 8.7 43.5

Soybeans 7.2 5.5

Other grains/oilseedsb 1.3 4.2

Horticultural cropsc 21.3 ~0.0e

Meat animalsd 37.8 ~0.0e

Dairy 10.8 5.1f

Other commoditiesg 7.4 2.5

Total 100.0 100.0

SOURCE: U.S. cash receipt data are available from the USDA Economic Research Service, FarmIncome Data, accessible at: http://www.ers.usda.gov/Data/FarmIncome/finfidmu.htm. Thecommodity payment data are available from the USDA’s Farm Service Agency, BudgetDivision, “Commodity Estimates Book for FY 2007 President’s Budget” (for crop years 2002and 2003 data) and available at http://www.fsa.usda.gov/dam/bud/CCC%20Estimates%20Book/estimatesbook_PresBud.htm, and the “Commodity Estimates Book Material for FY2007 Mid-Session Review” (for crop years 2004 and 2005) and available at http://www.fsa.usda.gov/dam/bud/CCC%20Estimates%20Book/estimatesbook_MSR.htm.

NOTES:

a. Included in the total are production flexibility contract payments, direct payments, counter-cyclical payments, loan deficiency payments, marketing loan gains, and certificate exchangegains. For the dairy sector, the figure consists of payments under the Milk Income LossContract (MILC) Program.

b. Includes barley, oats, sorghum, millet, flaxseed, peanuts, sunflowers, safflower, and miscel-laneous oil seeds.

c. Includes fruits, tree nuts, vegetables, melons, and greenhouse/nursery.

d. Includes cattle/calves, hogs, sheep, lambs, and poultry/eggs.

e. Program payments for the meat animal and dairy sector are very small and given here asapproximately zero.

f. The data for the Milk Income Loss Contract Payment are available only by fiscal year. Theshare given is based on the average payment budgeted during fiscal years 2003–06.

g. Includes figures for tobacco, sugar, honey, wool, and mohair.


32

imagine how the industry would adjust to a marketwithout the programs. Adjustment is complex.When support has been a part of an industry for,literally, longer than anyone can remember, it is notsurprising that industry participants would resistchange. Eliminating farm programs may be appro-priate policy reforms for consumers, taxpayers, non-subsidized producers, and the economy as a whole,but will naturally be resisted by those who currentlyreceive the benefits.

Current Forces for Farm Bill Reform

Throughout 2006 and early 2007, farm organizationsand other groups have laid out Farm Bill positions,with most organizations representing subsidizedfarmers suggesting that the farm commodity subsi-dies be extended, perhaps with modifications thatwould increase benefits for their group, for as manyyears as possible. Given the long history of policy suc-cess by commodity interests, are there forces forchange in 2007? Several current drivers for market-oriented changes in farm subsidy programs can beidentified.

First, the budget deficit and congressionalbudget rules will limit projected spending on farmsubsidies, and there are many demands on thefunds allocated for the Farm Bill by congressionalbudget committees. However, with current fore-casts of high grain and oilseed prices, projected out-lays from current farm programs are low relative tothe recent history and this relieves some of thebudget pressures on the programs because eveneliminating the price-contingent subsidies wouldsave little in projected outlays.

Second, environmental interests have pushed toreplace commodity subsidies with support for farm-provided conservation and environmental services.The Conservation Security Program created by the2002 Act was a small and awkward step in thatdirection that tied payments to approved practiceson farms.

Third, most payments go to those who producemost of the program commodity output, and so,

by design, most farm program payments go to acomparatively small number of relatively wealthyfarm operators and owners. Those who wish farmprograms were targeted to aid the poor expressdismay at the obvious fact that current programscannot possible serve this purpose.

Fourth, growers of horticultural crops havebegun to argue aggressively for more support thattargets research, nutrition education, protectionfrom invasive species, and opening internationalmarkets. These groups directly challenge spendingsuch a high proportion of the agriculture budget onincome transfers that do little to improve productiv-ity or competitiveness.

Fifth, Congressional leadership and the Secretaryof Agriculture have pointed out that rulings in theWTO dispute over the U.S. upland cotton programsuggest that other U.S. farm programs also may bevulnerable to WTO challenges, and therefore it isprudent to modify the program now before changeis mandated by external forces. Finally, economistsand advocates for market forces stress the negativeconsequences of farm subsidy programs for resourceallocation and note that some of the most dynamic,innovative, and successful agricultural industries arefound among those that are not encumbered by sub-sidy programs.

Conclusion

Farm commodity programs have long been anestablished part of the American agricultural land-scape. However, it is difficult to find any convincingrationale for continuing the current farm programs,other than that they are popular among those whoreceive the benefits.

Once commodity programs are gone, other pro-grams affecting agriculture and rural affairs can beconsidered on their individual merits. Agricultureinvolves some broad public goods and industry col-lective goods that would not be supplied appropri-ately without some government involvement, andperhaps funding. Examples include agriculturalresearch, information services, efforts to improve

international market access, and control of harmfulinvasive species. There is also widespread supportfor using public policy to respond to rural environ-mental externalities.

The 2007 Farm Bill is being developed during aperiod of particular attention to the problems of thecurrent programs and unmet goals that could beaddressed by alternatives. Whether these alterna-tives replace or significantly alter the historic com-modity support programs is the primary issue beforeCongress and the public.

Notes

This policy brief draws from the author’s longer paper, “FarmSubsidy Tradition and Modern Agricultural Realities,” avail-able at www.aei.org/farmbill. The longer paper includes back-ground support and citations not included here. Sumner isthe director of the University of California Agricultural IssuesCenter and Frank H. Buck, Jr. Professor, Department of Agri-cultural and Resource Economics, University of California,Davis. The author thanks Robert Thompson, Julian Alston, andother AEI project participants for useful comments. The viewsexpressed here are those of the author and do not representthose of any organization with which he may be affiliated.

FARM SUBSIDY TRADITION AND MODERN AGRICULTURAL REALITIES

33

EFFECTS OF EXISTING POLICY

While a variety of reasons for farm commodity pro-grams have been put forward, their fundamentalresult is to transfer income from other sectors to agri-culture. In the case of commodities protected byimport barriers, like beef, sugar, and dairy products,the transfers come at the expense of domestic andforeign consumers of the food and fiber productsthose commodities are used to make; in the case ofcommodities supported by subsidy programs thetransfers are financed from government revenues,ultimately at the expense of U.S. taxpayers.

When subsidies to farmers result in changes infarm input use and production, and in prices of farminputs and outputs, some of the benefits will beshifted from farmers to suppliers of farm inputs, con-sumers of farm outputs, and other agribusiness inter-ests. The ultimate distribution of benefits (and costs)of farm subsidies thus depends on the nature of thesubsidies, the extent to which they are coupled toproduction and influence inputs and outputs, and theconditions of commodity supply and demand. Thepolicies also involve deadweight losses, in the sensethat the costs to taxpayers and consumers exceed thesum of benefits to farmers, landowners, and otheragribusiness interests, but the redistributive conse-quences generally are measured on a larger scale.

These distributional and economic efficiencyconsequences of policies are typically not directly

observable. Instead, they must be inferred usingeconomic models, intuition, and insight, combinedwith data and measures of those aspects that can beobserved directly. This policy brief presents ideas andevidence about the distribution of benefits from U.S. farm subsidies. Particular attention is paid to twopopular but competing ideas: (a) that the benefitsfrom farm subsidies accrue to farm operators, (b) thatthe benefits from farm subsidies ultimately accrue tolandowners (and not to farmers).

Many economists have argued that all farm subsi-dies are ultimately capitalized in land values. Thispaper shows, both theoretically and empirically, that this is not so, although there is much room fordisagreement as to the precise shares that accrue tolandowners, farmers, and consumers. Careful reviewof the econometric and other evidence leads to theconclusion that perhaps 40–60 percent of subsidypayments accrues to landowners, 15–35 percentaccrues to farmer operators, and 20 percent accruesto consumers, leaving a 5 percent deadweight loss tothe economy as a whole. Given that farmers ownabout half the land they farm, perhaps 20–30 percentof the subsidy accrues to non-farmer landowners, and45–55 percent accrues to farmers.

Using 2005 values, the paper estimates that theelimination of subsidies for program crops—in par-ticular corn, soybeans, upland cotton, wheat, and

37

Who Really Benefits from U.S. Farm Subsidies?

Julian M. Alston

rice—would result in a decline in program cropoutput by about 7.3 percent, and would save a deadweight loss of about $400–800 million per year.The full social costs of distortions from the subsidyexpenditure of $16.5 billion may be five to ten timesgreater than this amount, in the neighborhood of $4 billion per year. The opportunity cost of suchexpenditure (i.e., forgone investment alternatives) is likely to push the total societal cost of U.S. farmsubsidies even higher still.

Why Do Economists Disagree About Farm Program Impacts?

Support among economists can be found for both ofthe extreme positions, as well as the in-between ideathat both landowners and farm operators benefitfrom subsidies. This mixture of views reflects thefact that any estimates of the effects of policiesinvolve measurement error and many of the deter-minants of findings themselves have to be estimated.But it also reflects more fundamental factors. In conducting analysis of farm programs and measur-ing the consequences of a change in a policy or setof policies, economists have to make a lot of choices,including which factors to hold constant in theanalysis and which ones may vary, with the appro-priate choice depending on the purpose of theanalysis. Different studies may report differentresults because they are using different modelingapproaches, assumptions, or data, or because theyare applying their analysis to a different question(and therefore holding different things constant). Itis important, then, to make these aspects clear whendiscussing results from policy models.

This brief refers to analysis of the likely conse-quences from a comprehensive, fully anticipated,and permanent elimination of U.S. farm programpolicies; holding the policies of all other countriesconstant; in a medium- to long-term scenario (i.e.,allowing several years for adjustment to a change inpolicy); looking forward from a baseline consistentwith 2005; and keeping the Conservation ReserveProgram in force. The qualifications in this paragraph

illustrate just a few of the things that may differamong studies, giving rise to different results. Inaddition, findings depend on producers’ and con-sumers’ responsiveness to price changes, and econo-mists have different views about relevant behavioralparameters, which cannot be measured with perfectprecision. Further, the distribution of benefitsdepends on specific details about the subsidies andhow they are represented in the analysis.

A Theoretical Analysis

Simple theoretical models can be used to illustratehow different types of subsidies have different inci-dence. Analysis with such models indicates that weshould expect a fully decoupled payment attached to land to be reflected entirely in land rents (one dol-lar of additional land rent paid and earned per dollarof additional subsidy payments). Under extremeassumptions (such as a fixed supply of land) the samewould be true of an input subsidy on the use of land.More generally, however, even a subsidy on land willhave some effects on input combinations and output,and thus the incidence will be shifted partly to sup-pliers of non-land inputs and consumers. A subsidyon output is expected to have even less of its inci-dence on land and more of it on consumers andsuppliers of non-land inputs, but still it will have adisproportionate incidence on landowners as thesuppliers of the least elastic factor of production.

The longer paper reports results from a simple, butpowerful and useful, static model in which land andother inputs are used to produce a single aggregateoutput. Application of this model, which is conven-tional and has been widely applied in agriculturaleconomics, indicates that, depending on specific val-ues for the key parameters but allowing some elastic-ity of supply of land to agriculture, landowners wouldreceive the following approximate shares of benefitsfrom subsidies: (a) decoupled direct payment tied toland, 100 percent; (b) land input subsidy, 45–77 per-cent; (c) output subsidy, 19–44 percent. The specificdetails of actual policies matter for incidence. Real-world policies are typically not pure output subsidies


38

or pure input subsidies, often combining multipleinstruments together. Still, the theoretical analysis ofstylized policies provides some guidance as to therange of incidence outcomes we might expect fromreal-world policies, and it gives a basis for interpreta-tion of the results from empirical work with models ofland markets.

A Sector Model of U.S. Agriculture

U.S. agricultural policies involve a mix of subsidiesthat are decoupled to different degrees. To measurethe consequences we must make assumptions aboutthe extent to which farmers respond to the pay-ments. The impacts of subsidies on program crops(including corn, soybeans, upland cotton, wheat,rice, and other program crops) were quantified in asimulation model, with different forms of subsidiesrepresented as equivalent fractions of a pure decou-pled payment and a fully coupled output subsidy.The fractions were 40 percent coupled for “directpayments,” 50 percent coupled for “countercyclicalpayments,” and 100 percent coupled for “loan pro-gram payments.” Applying those fractions in a modelof U.S. program crop production as a whole in 2005,the total subsidy of $16.52 billion was equivalent to a decoupled transfer of $5.56 billion, 100 percentof which accrues to land, combined with a pure out-put subsidy of $10.96 billion, 30 percent of whichaccrues to land. The overall incidence was thereforeabout $8.85 billion on land and $7.65 billion onconsumers and suppliers of non-land inputs, includ-ing farm operators who supply labor and manage-ment among other things.

Taking this approach, the subsidy spending worth28.6 percent of the total value of program crop pro-duction had incentive effects equivalent to an outputsubsidy of 19.0 percent. Applying the subsidy rate of19.0 percent in the simple static model, the impliedeffect of eliminating the programs would be a reduc-tion in production of program crops in aggregate by7.3 percent. If the CRP were eliminated along withcrop subsidies, the effects would be smaller—an out-put reduction of around 5.0 percent.

The Australian Bureau of Agricultural andResource Economics (ABARE) has recently publishedresults from an analysis of the consequences of elimi-nation of U.S. farm commodity programs, using amuch more detailed model that included equationsfor supply and demand for individual commodities inother countries, trading with one another and theUnited States. They predicted consequences overtime, allowing full adjustment to complete elimina-tion of the policies by 2016. Their estimates of effectson production ranged from 2.9 to 13.9 percent for theprogram crops considered in the simple model, butwere only 2.9 and 3.8 percent for soybeans andmaize, which represent two-thirds of the value of pro-duction. Thus, the implications are similar: the totaloutput effects of elimination of subsidies would bemodest, even in the most-subsidized parts of the agri-cultural industry; as would be the deadweight lossesassociated with subsidy-induced distortions in com-modity markets (or net gain from eliminating them).

Synthesis

Combining the results from a review of econometricmodels in the literature, multi-market simulations,and the application of a sector model of U.S. agri-culture yields a range of results about the share ofsubsidy payments going to land. The truth probablylies in between the results from the static theoreticalmodels with full adjustment and the general run of the econometric evidence: that a significant shareof even the so-called “decoupled” transfers goes to farmers rather than landowners, and that bothlandowners and farm operators receive a signifi-cant share of the net benefits from subsidies. Thesimulation results presented here correspond to an in-between case, perhaps reflecting an intermedi-ate run scenario with incomplete adjustment to sub-sidies. They also reflect the inherent uncertaintyabout the issue.

To make matters concrete, in broad terms theresults presented here suggest that 40–60 percent ofsubsidy payments accrues as benefits to landowners,20 percent accrues as a benefit to consumers, and

WHO REALLY BENEFITS FROM U.S. FARM SUBSIDIES?

39

15–35 percent accrues as a benefit to farm operatorsper se, with a modest amount, roughly 5 percent,wasted as a deadweight loss (more if we count bene-fits to foreign consumers as a loss to the UnitedStates). In round figures, then, perhaps 75 percent ofthe subsidy expenditure accrues as a benefit to farmoperators and landlords. Given that farmers collec-tively own about half the land that they farm, farmersreceive about half of the total that goes to landowners,leaving 20–30 percent to non-farmer landlords. Thus,45–55 percent of the total subsidy expenditureaccrues as a benefit to farm operators.

In short, for every dollar of government spendingon farm subsidies, farmers receive about 50 cents,landlords receive about 25 cents, domestic and for-eign consumers receive about 20 cents, and 5 centsis wasted. Additional amounts are wasted collecting

the taxes to finance the spending and in administer-ing the policies—perhaps another 20 cents. If thepurpose is to transfer income to farmers, the mech-anism is very inefficient, with less than half of theamount taken from taxpayers ending up with theintended recipients.

Note

This policy brief draws on the author’s longer paper“Benefits and Beneficiaries from U.S. Farm Subsidies,” available at www.aei.org/farmbill. The longer paperincludes background support and citations not includedhere. Alston is a Professor in the Department ofAgricultural and Resource Economics, University ofCalifornia, Davis. The views expressed here are those of the author and not those of any institution with whichhe is affiliated.


40

Changes to commodity programs in the 2002 Farm Bill reversed a trend of declining governmentintervention in U.S. agriculture. This trend beganwith the 1985 Farm Bill, which brought a freeze onprogram acres and yields along with dramaticallylower guaranteed prices. Some thought that theUruguay Round Agreement that formed the WorldTrade Organization together with the 1996 Farm Bill signaled the beginning of the end to U.S. sup-port of agriculture. But low prices from 1998 to2001 motivated farm supporters in Congress to givefarmers emergency, supplemental payments (MarketLoss Assistance payments), thereby signalingCongressional intent to switch direction with the2002 Farm Bill.

The 2002 Farm Bill reversed direction in a num-ber of ways. Minimum price floors for feed grainsand wheat were increased. Price floors were createdfor pulse crops (peas and lentils), and a new pro-gram that gave out payments when prices were lowwas adopted for all program crops. And instead ofallowing the fixed, declining payments to disappear,the payments were frozen through the life of the Bill.All of these changes worked to significantly increasesupport for producers of program crops.

The changes in farm policy were projected to givefarmers about $5 billion in additional payments peryear. Critics of U.S. farm policy include those who

believe that our farm programs impoverish farmersin poor countries, cause obesity, and support largeagribusiness corporations. These criticisms rely onthe presumption that U.S. farm programs decreasethe market prices of supported commoditiesbecause they induce U.S. farmers to produce morethan they otherwise would have. Farm bill support-ers add credence to these criticisms when they jus-tify payments by the need to keep farmers inbusiness and the reliance of rural economies onmaintenance of planted acreage.

This analysis obtains insight into the extent towhich U.S. farm programs actually have the effectsthat both its supporters and critics say it has:namely, whether production levels are higher andcrop prices are lower than they would be withoutthe program. The approach taken is to assume thatfarm programs and the evolution of U.S. agriculturewere what they were until 1998, at which time farmprograms were eliminated. The resulting picture ofagriculture from 2002 to 2005 is then compared towhat actually happened during this period to meas-ure the impacts of farm programs.

The analysis finds that production and prices forthe three largest program crops—corn, wheat, andsoybeans—would have been virtually unchanged ifall major farm programs had been eliminated in1998. Rice production was significantly larger, and

41

Money for Nothing: Acreage and Price Impacts of U.S. Commodity Policy for Corn, Soybeans,

Wheat, Cotton, and Rice

Bruce A. Babcock

prices slightly lower, than would otherwise havebeen the case. The greatest price effect was found for cotton, the most heavily subsidized crop; worldcotton prices would have been 10 percent higher inthe absence of U.S. subsidies.

Given political realities, a status quo farm pro-gram seems the most likely outcome for this year’sfarm program. Nonetheless, a more ambitious objec-tive would be for Congress to pass a new set of com-modity programs that would integrate traditionalcommodity programs, disaster assistance, and cropinsurance into a more cost-effective, nonduplicative,and transparent safety net for farmers.

Modeling Program Supply Effects

How Main Farm Programs Work. The first step toestimating the impact of farm programs on produc-tion levels is to understand how the main programswork. Three payment programs—direct payments,countercyclical payments, and marketing loans—are designed to protect crop-specific target prices.

The first tier of support for farmers is direct pay-ments. The level of these payments is determined bymultiplying a crop-specific direct payment rate by85 percent of a farm’s direct payment base acres andpayment yields. Payment rates, base acres, and pay-ment yields are fixed. Hence farmers know preciselytheir payment levels each year. Furthermore, pay-ments are unaffected by farmers’ planting decisions,by production levels, or by the level of market price.

The second tier of support for farmers is the countercyclical payment program (CCP). Paymentsunder this program are triggered whenever the sea-son-average price, as measured by the NationalAgricultural Statistics Service (NASS) is less than theeffective target price. The maximum payment rate isachieved when the season average price falls belowthe “loan” rate (defined next) for each crop. Theamount of payment equals 85 percent of the productof the payment rate, a farm’s program yield, and afarm’s program acreage. Thus the CCP providesfarmers with support on a fixed amount of produc-tion, with a cap on the amount that can be paid out.

Farmers also receive countercyclical payments regard-less of what they plant.

The third tier of support is the marketing loanprogram. Loan programs in the form of non-recourse loans have long been part of U.S. farm pro-grams. With non-recourse loans, producers have toput their harvested production under loan: that is,they must fill out a loan application and take a loanfrom the government. The amount of the loanequals the product of the loan rate and harvestedproduction. If market price does not rise above theloan rate during the loan period, then the producerdoes not pay back the loan. Rather, the crop as col-lateral is forfeited to the government.

Marketing loans differ from standard non-recourseloan programs in two ways. First, a producer still hasthe option of taking out a loan from the governmentat harvest, pledging harvested production as collat-eral. But marketing loans offer another option. If mar-ket price does not rise above the loan rate during theloan period, the producer can choose either to forfeitthe crop or repay the loan at the loan repayment rate.The producer will choose to pay back the loan at the repayment rate if the market price is greater thanthe repayment rate because the producer can receivethe difference between the two. If the repayment rateis greater than the market price, then the producerwill forfeit the crop.

Marketing loans are designed to guarantee thatfarmers can always receive a price equal to the loanrate for their crop without the need for the govern-ment to actually take possession of crops. Thismeans that market prices are free to adjust down-ward to clear domestic and international markets.Thus in bumper crop years, market prices are free tofall to whatever level is needed to balance supplyand demand. In short crop years, market prices willrise to higher levels because traders know that there are no government stocks that are waiting to be sold. With marketing loans, USDA writes checks tosupport the loan rate rather than taking ownershipof crops.

Table 1 reports average annual marketing loan,countercyclical, and direct payments by crop fromthe 2002 through the 2005 marketing year. Also


42

shown are the average paymentsper acre and the average pay-ments expressed as a percent ofmarket revenue. Corn farmersreceived the most aggregate pay-ments, followed by cotton, wheat,rice and soybeans. On a per-acrebasis and as a percent of marketrevenue, rice and cotton farmersreceived by far the most pay-ments. Figure 1 shows the shareof payments received for eachcrop from the three programs. Asshown, almost all the support forwheat and soybeans came fromdirect payments. Direct paymentsalso make up the largest share ofcorn payments. But marketingloan and countercyclical pay-ments for corn also make up sig-nificant shares. Rice and cottonhave significant payment sharesfrom all three programs.

Two main conclusions can bedrawn from table 1 and figure 1.First, the subsidies for wheat and soybeans have been relativelysmall in size and the subsidiesthey have received are primarilydirect payments. This suggeststhat any positive acreage effects offarm programs on soybean andwheat are likely small, and theyare positive only to the extent that farmers increaseplanted acreage in response to direct payments.Second, cotton payments are large and mostly com-prised of the payments that likely induce farmers toincrease planted acreage. Hence significant acreageimpacts are likely to come from cotton, rice, andperhaps corn.

The model used to estimate what agriculturewould have looked like without the program pay-ments was originally developed in the early 1980s byresearchers at Iowa State University and the Univer-sity of Missouri to simulate how future agricultural

prices and quantities would change under alternativepolicy regimes. The current model is updated annu-ally by FAPRI (Food and Agricultural Policy ResearchInstitute) analysts at the University of Missouri, but itno longer serves as the primary policy simulation toolfor forward-looking policy analysis. This model wasupdated late in the fall of 2005. Hence it provides up-to-date information about market conditions throughthe 2005 crop year.

The model assumes that farmers make their plant-ing decisions based on information available to themwhen those decisions are made. Thus, expected price

MONEY FOR NOTHING

43

TABLE 1AVERAGE ANNUAL PROGRAM PAYMENTS, 2002–2005

Marketing Counter- Average annual Payment as a loan cyclical Direct aggregate percent of mar-

Crop ($ million) ($ million) ($ million) payment ($/acre) ket revenue (%)

Corn 1,904 1,436 2,116 55 24

Soybeans 83 0 608 7 4

Wheat 52 0 1,151 16 17

Cotton 953 1,125 633 156 59

Rice 368 150 431 240 62

Source: Food and Agricultural Policy Research Institute. 2006. Agricultural Outlook. StaffReport 1-06 University of Missouri, Columbia, MO, and author’s calculations.

FIGURE 1RELATIVE IMPORTANCE OF DIFFERENT PROGRAM PAYMENTS,

2002–2005 (percent)

Source: Author’s calculations.

0%

20%

40%

60%

80%

100%

Corn Soybeans Wheat Cotton Rice

CCP LDP DP

and farm program payments ratherthan actual prices and farm programpayments drive planting decisions. Thismeans that in 2004 corn farmers didnot know that they were going to har-vest a record crop that would drivedown market prices and drive up gov-ernment payments. And corn farmersin 2005 did not know that HurricaneKatrina would damage Gulf grain han-dling, thereby temporarily drivingdown harvest time prices and creatingwindfall government payments.

Effects of Program Elimination onPrices and Output. Tables 2 and 3show the impact of program elimina-tion on acreage and prices. The acreageimpacts of farm programs for corn,soybeans, and wheat have been quitemodest. The programs changed corn,soybean, and wheat acreage by an aver-age of less than 1 percent from 2002 to2005. The reason why the acreageeffects are so modest is that expectedmarketing loan payments for these threecrops were relatively modest in size, andthe decoupled countercyclical anddirect payments had little impact oncrop acreage. Furthermore, becausethese three crops compete with oneanother for acreage, elimination of themarketing loan program reduces netrevenue for all three crops (especiallycorn and soybeans), which left expectedrelative net returns largely unchanged.Comparing the average corn and soy-bean estimates, the programs slightlyfavored corn, so corn acreage was a bithigher than it otherwise would havebeen but for the programs, and soybean acreage was alittle lower than what it would have been.

The largest impacts of the programs were on cot-ton and rice acreage. But for the farm programs,cotton acreage have been 12.1 percent lower on

average across this time period, and rice acreagewould have been 21.2 percent lower. As has beenstated so many times by the cotton industry, cotton isa costly crop to grow and cotton farmers need farmprograms to cover their costs.


44

TABLE 2SIMULATED PLANTED ACREAGE WITH PROGRAM ELIMINATION

(MILLION ACRES) AND PERCENT DIFFERENCE

FROM HISTORY, 2002–2005

Crop 2002 2003 2004 2005 Average

Corn 77.5 78.3 81.2 81.0 79.5-- 1.8% -- 0.3% 0.3% -- 0.8% -- 0.7%

Soybeans 73.7 73.9 74.7 72.7 73.8-- 0.2% 0.8% -- 0.6% 0.7% 0.2%

Wheat 60.1 61.0 59.5 58.0 59.6-- 0.3% -- 1.7% -- 0.3% 1.3% -- 0.3%

Cotton 10.0 12.8 14.0 11.6 12.1-- 26.9% -- 3.6% 4.5% -- 16.5% -- 10.8%

Rice 2.2 1.9 2.9 3.2 2.6-- 30.8% -- 37.9% -- 14.4% -- 3.6% -- 21.2%

Total 223.5 227.9 232.2 226.6 227.6-- 2.8% -- 1.0% -- 0.1% -- 0.8% -- 1.2%

11-crop 248.1 251.6 252.2 246.2 249.5-- 2.7% -- 1.1% -- 0.2% -- 0.9% -- 1.2%

SOURCE: Author’s calculations.

TABLE 3SIMULATED CROP PRICES WITH PROGRAM ELIMINATION

AND PERCENT DIFFERENCE, 2002–2005

Crop 2002 2003 2004 2005 Average

Corn ($bu) 2.37 2.44 2.05 1.89 2.191.9% 0.9% -- 0.2% 1.7% 1.1%

Soybeans ($bu) 5.57 7.28 5.73 5.24 5.950.7% -- 0.8% -- 0.2% -- 1.1% -- 0.4%

Wheat ($bu) 3.59 3.44 3.41 3.37 3.451.0% 1.2% 0.2% 0.1% 0.6%

Cotton ($lb) 0.54 0.65 0.40 0.55 0.5422.1% 5.7% -- 2.7% 14.1% 9.7%

Rice (cwt) 4.68 8.43 7.20 8.01 7.084.1% 4.3% -- 1.8% 0.8% 1.7%

SOURCE: Author’s calculations.

A Look Forward

The recent sharp increase in the demand for corn byethanol plants has dramatically increased corn and soy-bean prices. These high prices are projected to continuethroughout the period covered by the 2007 Farm Bill.Projected high prices are having a number of ramifica-tions for the new Farm Bill. Projected expenditures forthe marketing loan and countercyclical programs forcorn, soybeans, wheat, and rice over the 2008 to 2012period are less than $300 million per year. In contrast,expenditures for cotton are projected to be about $1.3billion per year. This disparity in projected expendi-tures has led the soybean and wheat commodity asso-ciations to call for adjustments in target prices that favortheir farmers. Corn growers have proposed a programthat would replace marketing loans and countercyclicalprograms with a new revenue insurance approach thatis integrated with the crop insurance program.

The problem facing these groups is that anyadjustment in target prices or program reform willlikely result in an increase in projected program costsbecause program costs are projected to be so low.New Congressional budget rules require proponentsof cost increases to pay for the increases by offsetssomewhere else in the budget.

The most straightforward way of paying for pro-gram modifications would be to reduce direct pay-ments. Another source of funds that could be tappedis the crop insurance program. However, both pro-grams have strong advocates, so the odds are growingthat the politically easy path of not changing pro-grams for the 2007 Farm Bill will be taken. By defini-tion, a simple extension of current programs willrequire no budget offsets. Because of high projectedprices, an extension of 2002 farm programs for corn,soybeans, wheat, and rice largely represents a freemarket outcome because projected marketing loanand countercyclical payments are so low.

However, such an extension does not mean that a better outcome is not possible. Continuation ofdirect payment amid high prices really does meanthat farmers will be receiving money for nothing.Furthermore, cotton loan rates and target prices con-tinue to encourage cotton plantings, although USDA’s

March prediction of a 21 percent drop in intendedcotton acreage suggests the beginning of an adjust-ment in U.S. cotton. And the WTO compliancepanel may rule this summer that the United Statesneeds to make further changes to the cotton pro-gram to bring it into compliance with the WTO cot-ton decision.

Another area of farm policy that is in need ofreform is the recurring intervention of Congress inoffering farmers disaster payments, despite recordparticipation in the crop insurance program. Themost common justification for the buildup in tax-payer support for the crop insurance program is tomitigate the need for ad hoc disaster assistance.Inclusion of yet another agricultural disaster bill in the recently vetoed supplemental appropriations billfor the Iraq war is yet another example of the failureof an expanded crop insurance program to meet thisbasic policy objective. It makes little policy sense tohave recurring disaster programs and an expensivecrop insurance program that is soon projected to costmore than the direct payment program. It also makeslittle sense not to account for commodity programswhen considering how a rationalized crop insuranceand disaster program would work.

Although a status quo farm program seems the mostlikely outcome for this year’s farm program, a moreambitious objective would be for Congress to pass anew set of commodity programs that would integratetraditional commodity programs, disaster assistance,and crop insurance into a more cost-effective, nondu-plicative, and transparent safety net for farmers.

Note

This policy brief draws on the author’s longer paper“Money for Nothing: Acreage and Price Impacts of U.S.Commodity Policy for Corn, Soybeans, Wheat, Cotton,and Rice,” available at www.aei.org/farmbill. The longerpaper includes background support and citations notincluded here. Babcock is a professor of economics and thedirector of the Center for Agricultural and RuralDevelopment, Iowa State University. The author is gratefulfor comments by Howard Leathers and others. The viewsexpressed here are those of the author and not those of anyinstitution with which he is affiliated.

MONEY FOR NOTHING

45

The U.S. sugar program is inefficient and benefitssugar producers and growers at the expense of sugarusers by keeping U.S. sugar prices at about doublethe world level. Studies estimate sugar producersgain hundreds of millions of dollars per year underthe program, while sweetener users lose over onebillion dollars per year. Net welfare losses to societyare significant but much smaller than are the trans-fers. Thus, the U.S. sugar program induces substan-tial waste of resources and large transfers from U.S. sugar users to U.S. sugar producers, sugar-cropgrowers, and select foreign producers.

The sugar program is also a costly way to trans-fer income to sugar exporters in developing coun-tries because under the current system of bilateralimport tariff rate quotas (TRQs), the sugar programencourages sugar exports by high-cost producersrather than low-cost producers, a form of tradediversion.

The current sugar program is endangered by sugarimports progressively entering the U.S. marketthrough global and regional trade agreements. Theseimports will either induce large sugar inventories, orcontract domestic production to unpalatable low levels. The program is also a potential target for reformbecause of likely reductions in Amber Box limitsunder a potential Doha Agreement of the World TradeOrganization (WTO). The current WTO accountingfavors abandoning the current program.

There are two ways to reform the current programshort of simply abolishing it. The current sugar pro-gram could be bought out, under a system that com-pensates all or some current sugar productions fortheir losses under a program repeal. Alternatively, itcould be changed to a standard crop program,removing the domestic supply controls (allotments),lowering the loan rate, implementing countercyclicaland direct payments, but preserving significant tradeprotection. Both alternatives would induce large fis-cal outlays, but would produce significant gains forU.S. consumers. In a buy-out, developing countrieswould also gain via expanded sugar exports to theUnited States.

Background on the U.S. Sugar Program

The sugar program has been functioning as a pricefloor with the so-called loan program. A loan rate guar-antees a minimum price to sugar producers, which isshared between processors and sugar-crop farmers(beet and cane). Controls on domestic production andtrade actually keep prices slightly above the loan rate.This system of price supports is made possible by atight restriction on imports of sugar through a set ofbilateral TRQs managed by the Office of the U.S. TradeRepresentative and the U.S. Department of Agriculture(USDA), as described below.

47

U.S. Sugar Policy: Analysis and Options

John C. Beghin

Domestic supply is managed with “allotments,”or production quotas, such that the U.S. marketprices of raw cane sugar and beet sugar remainabove the loan rates. USDA then avoids having tobuy sugar forfeited under the loan program.Equilibrium in the U.S. sugar market is reachedwhen U.S. quantity demanded at a price just overthe loan rate is equal to the sum of the productionallotment, imports under the TRQ system, NAFTA-authorized out-of-quota imports from Mexico, and change in stocks. Out-of-quota imports fromMexico, when they take place, determine the U.S.market price of sugar at the margin.

The current sugar program under the 2002 FarmBill maintains loan rates to processors of 18¢ perpound for cane sugar and 22.9¢ per pound for refinedbeet sugar. For reference, the world price of raw sugarhas averaged less than 10¢ per pound in the last 10 years. The provision of previous sugar programswhich forbids direct federal outlays was reinstated in2002 to avoid a likely unsustainable expansion of pro-duction generated under the 1996 Farm Bill. Supplycontrols by means of a marketing allotment systemwere reinstated. The allotment can only be appliedwhen imports for domestic consumption are less than 1.532 million short tons raw value (strv), the so-called trigger level. This way, sugar processors buy andprocess no more sugar than they can sell domesticallyat prices no lower than the loan rate. In recent years,the allotment has been set at around 9.35 million strv.

Under the 1994 Uruguay Round Agreement onAgriculture (URAA) of the WTO, the U.S. agreed to allow market access to a minimum quantity of1,117,195 metric tons (mt) of raw sugar each mar-keting year (October–September), and 22,000 mt ofrefined sugar at a preferential near zero tariff (the in-quota tariff rate). Many countries avoid the smallduty because of the General System of Preference or Caribbean Basin Initiative (CBI) programs. U.S.imports have systematically exceeded this minimummarket access commitment since 1994, but declinedin the late 1990s to accommodate increases in domes-tic production. The raw sugar TRQ has been allocatedto 40 quota-holding countries based on their histori-cal export shares during the 1975–81 period, when

trade was relatively unrestricted. The WTO commit-ment on market access put a lower bound on theimport volume allowed under the TRQ system.Hence, import-contraction has run its course, as nofurther contraction could take place while meetingthe WTO market access commitment.

The duty above quota on raw and refined sugarreached 15.36¢/lb and 16.21¢/lb, respectively, in2000. These tariffs remain prohibitive, except forimports coming from Mexico under the NorthAmerica Free Trade Agreement (NAFTA). The rele-vant sugar tariff for Mexican sugar has been fallingsince 1995, and is scheduled to fall to zero in 2008.As a result, Mexican sugar exports will be able tocompete with U.S. sugar. Out-of-quota importsfrom Mexico have already been occurring since1998–99, depending on the relative attractiveness ofthe Mexican and U.S. markets.

The prospect of competition from Mexican sugaris limited by Mexico’s significant trade barriers andhigh domestic prices of sugar, which reduce theattractiveness of exports to the world markets. How-ever, Mexico will provide reciprocal unlimited duty-free access for High Fructose Corn Sweetener(HFCS) from the United States starting in 2008.Depending on the price of corn as a feed stock for HFCS, about 250 thousand metric tons (tmt) to300 tmt of U.S. HFCS could go to Mexico and dis-place Mexican sugar, which would then likely beexported to the United States. This would doubleMexican sugar exports to the U.S., from roughly 250 tmt to 500 tmt. There is the credible threat thatU.S. sugar could be exported to Mexico, especially if Mexican prices are higher than U.S. market prices.Policymakers in both countries are aware of thesearbitrage opportunities and their potential to disturbcurrent market conditions.

Several recent opportunities to open the sugarmarket to freer trade were missed because U.S. sugarinterest groups were effective at opposing the tradeopenings such as in the U.S.-Australia Free TradeAgreement. Under the Central American-DominicanRepublic Free Trade Agreement (CAFTA-DR), theU.S. has established additional TRQs, starting at acollective gain in access of only 107 tmt, with the


48

gain growing to just over 151 tmt in year 15, there-after growing by 2 percent a year (simple growth)into perpetuity. The United States also establishes aquota for specialty sugar goods of Costa Rica, but inthe amount of a minuscule 2 tmt annually. Provi-sions will ensure that only net surplus exportingcountries in the region have increased access, andprovisions have been made to allow alternativeforms of compensation to be established to facilitatesugar stock management by the United States.

There are several free trade agreements beingconsidered by the U.S. government, including theFree Trade Agreement of the Americas (FTAA), andagreements with Brazil, Colombia, Thailand, andothers. The potential for expanded trade from LatinAmerican exporters is large in the context of an FTAAthat would include Brazil, or even in the simpler con-text of a bilateral agreement with Colombia. Braziland Thailand are major world sugar exporters andlarger countries than CAFTA member countries. Theyare less dependent on the United States, and lesslikely to bow to pressure to accommodate U.S. sugarinterests. These FTAs represent the best opportunityto drive a change in sugar policy. However, the failureof the U.S.-Australia FTA to allow even tiny gains inU.S. sugar imports is a bad omen for any sugar reformin future FTAs.

Winners and Losers under The U.S. Sugar Program

The sugar program and associated trade restrictionshave significant effects on national income in theUnited States. They greatly tax U.S. users of sugar(final consumers and food processors). U.S. users ofsugar have been paying between two and three timesthe world price of sugar, depending on world anddomestic market conditions. Transfers have beenlarger than the associated deadweight losses, as is often the case in agricultural markets where consumerdemand is relatively unresponsive to price. Studiesestimate that U.S. sugar producers gain hundreds of millions of dollars each year from the program,while U.S. consumers would save between $1.4 and

$1.9 billion annually if the program were eliminated.The net deadweight loss of the program is about $500 million and growing as U.S. population grows.

U.S. sugar production would decrease, but notdisappear, in unfettered markets. Research showsthat beet sugar production would be more affectedthan cane sugar, with larger price and output con-tractions if markets were freed.

The historical import quota allocation providesanother criticism of the sugar program. The importquotas do not allow the most efficient sugarexporters (Brazil, Thailand, Australia, among others)to exploit their low-cost potential to export to theUnited States. Furthermore, world prices aredepressed because of the import TRQ system. Thesystem is not an efficient way to transfer income toTRQ-right holders, as high-cost holders cannot sellthese exporting rights to low-cost exporters. Theexporting country under the TRQ could be mademuch better off by renting or selling its TRQ rightsto Brazil or another low-cost producer, and have itssugar producers face a more relevant scarcity signalthan the inflated U.S. prices.

The U.S. is a large country in world sugar marketsand its imports influence world sugar prices. Hence,any change in the U.S. sugar program resulting inchange in U.S. sugar imports would affect worldprices. For example, following the removal of the U.S.sugar program and associated TRQ, the world price ofsugar would increase by 14 percent to 17 percent. (Ofcourse, even this new, higher world price represents asubstantial reduction in the U.S. sugar price.) Further,under U.S. leadership global liberalization of sugarmarkets would be likely. Under global trade liberal-ization, the world price of sugar would increase by 40 percent to 60 percent depending on the modelingapproach and would lower the adjustment cost facedby the U.S. sugar growers and producers.

Policy Options, Possible Reforms, and Their Effects

The Doha WTO Round could lead to changes inmarket access, export subsidies, and domestic

U.S. SUGAR POLICY: ANALYSIS AND OPTIONS

49


50

support, but based on current proposals, reform ofdomestic support is the only one that could lead tochanges in the current U.S. sugar program. Likelyreductions in domestic support as measured by theWTO agreements are likely to induce reforms of thesugar program, as they would constitute a conven-ient way to meet new WTO commitments ondomestic box support. Removing the domestic com-ponent of the sugar program would eliminate thecontribution of high sugar prices to the U.S.’s overalldomestic support of agriculture while leaving tradebarriers in place. The impact would be only moder-ate effects on actual market prices and producerincomes as long as border protection remains high.

One way to accomplish this is to shift sugar to astandard farm payment scheme with loan deficiency,direct and countercyclical payments. This has beenconsidered a way to make the sugar program less ofan outlier and to ensure that it is sustainable withinNAFTA and a potential WTO Doha agreement, whilepreserving much of the income gains to producersand processors. However, shifting sugar policy to anon-budget government payment scheme creates asignificant problem with payment limits, especiallyfor the cane sugar industry, because farms are part ofvery large integrated firms. Even if delayed, futureWTO agreements will eventually lead to meaningfulmarket access expansions.

The contribution of sugar to the aggregatedomestic support limits in the WTO agreementfrom 1995 to 2001 hovered around $1.1 billion.Estimates for 2005 and 2006 give roughly compa-rable figures ($1.2 to 1.3 billion). If the sugar pro-gram became a standard commodity program and ifNAFTA trade did not surge, the potential contribu-tion of sugar to this measure of domestic supportwould be zero. Higher imports from Mexico underNAFTA liberalization in 2008 translate into largegovernment stock accumulation under the currentsugar program. Under a standard farm subsidypayment scheme these large Mexican import sup-plies would decline with lower prices, but triggeronly modest countercyclical payments (CCPs) andloan deficiency payments (LDPs). If sugar were eli-gible for direct payments (which do not vary with

market prices) as tailored for other program crops,the outlays would be large—$463 million, butthese may qualify as only minimally trade-distortingand not affect the U.S. maximum commitment fordomestic farm support under the WTO agreement.

A radical reform of the sugar program would beto simply buy out the allotments and open the bor-der. To leave producers and processors as well off asthey are now, the buyout scheme would need tooffer at least as much discounted income gain as thecurrent program offers. One difficulty in designinga buyout of the sugar program is the absence of anestablished market price for rental or purchase ofmarketing allotments. To gauge the cost of a buy-out, consider a limited increase of duty-free importsby about 1.3 million strv. That would induce thedomestic price to fall by about 1.9¢ per pound. Theresulting annual value of protection lost under thisscenario is $355 million, leading to a discountedlump sum value of $1.887 billion for the2002–2007 period of the 2002 Farm Bill (using a 5percent discount rate). A buyout of 25 years of antic-ipated producer revenue loss has an annual cost of$647 million if spread over 10 years. The cost of amore complete buyout allowing free trade would bemuch more expensive. Full buyouts would be morefeasible in a context of rising world prices. Thesehigher prices could be induced by meaningful mul-tilateral trade liberalization in future WTO rounds.

A partial buyout on a regional basis would allowcompensation of the least-efficient sugar beet pro-ducers by enticing them to exit. This approachmight be politically acceptable, as beet growers aresmaller than and not as visible as the large canegrowers of the Southeast. Larger imports could takeplace and the most efficient producers could sur-vive, although they are likely to oppose such policy.This path of reform would resemble the recentreform of EU sugar policy.

Conclusions

The current sugar program is slowly becomingunsustainable in its current form, because sugar

imports are entering the U.S. market throughregional and bilateral trade agreements. UnderNAFTA, both U.S. and Mexican governments arejointly managing trade integration in sweetener mar-kets. After 2008, trade will be unrestricted, but theU.S. and Mexico have little to gain if trade surgesoccur. A convergence of producer protection levelscould occur. What happens to sugar in any potentialFree Trade of the Americas deal would be muchmore consequential.

Further, the sugar program in its current form isa potential target for reform because of likely reduc-tions in domestic support limits under a potentialWTO agreement. A possible response would be tochange the current sugar program into a standardfarm subsidy program, removing the domestic sup-ply controls, lowering the loan rate, implementingthe payment schemes available for program crops,but keeping the current trade protection nearlyintact. Current WTO proposals are unlikely tochange the border protection enjoyed by U.S. sugarproducers. Payment limitations under a standardprogram would antagonize large cane producers.Some initial shift of high-cost farms out of sugarwould likely occur, as prices would be lower.

A full buyout of the sugar program would beexpensive. There is an intertemporal trade-off whenconsidering a reform of the sugar program. One couldwait for the current program to crumble under

increasing pressure from trade integration, or onecould think of a buyout in the next farm bill. The highbudget costs of a complete program buyout may helpfocus resources on a partial sugar buyout.

A partial buyout represents an opportunity toremove high-cost farms and processors from theindustry. The most efficient farms could compete atprices close to recent world prices (12¢–14¢ perpound). U.S. sugar user and consumer gains wouldbe sizable. However, even a partial buyout that fullycompensates those who leave the industry is likelyto be more expensive in budgetary terms than shift-ing sugar to a standard farm subsidy program.

Of course, the program that makes most economicsense in terms of maximizing net benefits to societyat large is simple elimination of the sugar program,including both border barriers and domestic pricesupports and allotments.

Note

This policy brief draws on the author’s longer paper “U.S. Sugar Policy: Analysis and Options,” available atwww.aei.org/farmbill. The longer paper includes back-ground support and citations not included here. Beghin isdirector of the Food and Agricultural Policy ResearchInstitute at Iowa State University and Marlin Cole Professorin the Department of Economics, Iowa State University.The views expressed here are those of the author and notthose of any institution with which he is affiliated.

U.S. SUGAR POLICY: ANALYSIS AND OPTIONS

51

Public policy plays a larger role in U.S. dairy mar-kets than in any other agricultural market. Centralelements of current dairy policy date back to theNew Deal farm legislation of the 1930s, andalthough the original programs have evolved overtime, they remain largely unchanged in their maineconomic implications as new programs have beenadded over time.

Meanwhile, the U.S. dairy economy has beentransformed. U.S. dairy farms have become morespecialized and productive, as technological advanceshave facilitated increased milk production per cow,rapid expansion of dairy farming in the West, andlower costs of transporting milk. The number offarms has declined and remaining farms have gottenlarger, and individual dairy cooperatives have grownand merged so that the largest of them now representsignificant shares of national milk production.Consumption patterns have also changed, as thelargest share of milk no longer goes to fluid con-sumption but to manufactured dairy products whichare increasingly traded in global markets. In short,today’s dairy sector bears little resemblance to that ofthe 1930s, or even to that of the 1960s. Yet dairy poli-cies designed for that earlier era are still in effect today.

In this policy brief I describe the economic effectsof the main elements of current U.S. dairy policy,including the Milk Price Support Program (MPSP),

federal and state milk marketing orders, and the MilkIncome Loss Contract (MILC) program. In short, themain effect of these programs is not to remedy someobvious market failure, but to transfer income to U.S.dairy farmers. As is the case with similar policies forother agricultural commodities, consumers and tax-payers bear significant costs for these programs.Moreover, cost-benefit analyses show that the coststo consumers and taxpayers exceed the benefits todairy farmers.

Ironically, current dairy policy is not unambigu-ously beneficial for U.S. dairy farms. The MILC pro-gram and marketing orders actually harm clearlyidentifiable subgroups of dairy farms, and the MPSPreduces exports and innovation, with detrimentallong-term effects for all U.S. dairy farms. The mostdirect solution to the problems with current policy isto repeal current legislation and to allow dairy mar-kets to work. Clearly, consumers and taxpayers wouldbenefit from such deregulation. In addition, deregula-tion can contribute to a healthier, more competitiveU.S. dairy sector, and improve the outlook for manydairy farms—particularly the more efficient farms—that are harmed by the current policy regime.

Of course, many observers have pointed out thatU.S. dairy policy is costly and often ineffective, yet thegovernment continues to intervene in U.S. dairymarkets. Thus, if wealth transfers to some dairy farms

53

U.S. Dairy Policy: Analysis and Options

Joseph V. Balagtas

continue to be politically popular, I explore ways inwhich government support for dairy farms can beachieved more effectively and at lower cost to con-sumers and taxpayers. The preferred policy is a directpayment that does not depend on market prices orproduction. Such a policy would reduce market dis-tortions and improve program transparency.

Costs and Benefits of Current Dairy Policies

Central elements of current U.S. dairy policyinclude the Milk Price Support Program (MPSP),milk marketing orders, and the Milk Income LossContract (MILC). In addition, U.S. dairy marketsare insulated from world markets by tariff-ratequotas that apply to a long list of dairy products.

The Milk Price Support Program (MPSP). TheMPSP employs government purchases of wholesalebutter, nonfat dry milk (NFDM), and cheese to estab-lish minimum prices for these products. The primaryeconomic effect of the MPSP is to raise the averagewholesale prices of butter, NFDM, and cheese, therebyincreasing production of these products. The MPSPbenefits dairy farms indirectly by increasing proces-sors’ demand for milk used in supported products.Benefits to dairy farms come partly at the expense ofconsumers, who face higher prices for dairy products.Of course, the MPSP also imposes costs on taxpayers.Government expenditures for MPSP purchases aver-aged $410 million per year during years 2000–2003,a figure that does not include significant administrativecosts and costs of storage. The taxpayer and consumercosts of the MPSP exceed the benefits to dairy farms,generating a net loss to the U.S. economy.

Moreover, the MPSP is not unambiguously benefi-cial to dairy farms. The MPSP distorts dairy marketsin a way that harms the long-run viability of the U.S.dairy sector. Artificially high U.S. prices for supporteddairy products reduces domestic consumption as wellas U.S. exports. The MPSP also stifles innovation asU.S. dairy manufacturers continue to produce thebulk commodities supported by the policy. Partly as aresult, the U.S. dairy sector lags major exporting

regions (New Zealand, for example) in the develop-ment of high-value dairy ingredients, such as concen-trated milk protein, where growth opportunities exist.

The MPSP also poses a significant obstacle tomultilateral trade negotiations. The MPSP contributesmore to the U.S. Aggregate Measure of Support (usedby the WTO as a measure of trade-distorting farmpolicies) than any other U.S. commodity policy.Removal of the MPSP would help the United States tooffer much more far reaching reductions in domesticsupport in WTO negotiations.

Milk Marketing Orders. Milk marketing ordersregulate the sale of farm milk throughout most of theUnited States, setting minimum prices that processorsmust pay within specific geographic regions. Thedetails of marketing order regulation have changedoccasionally since they were first implemented in the1930s, but the main policy instruments—price dis-crimination and revenue pooling—have remainedmostly unchanged over time, as have the main eco-nomic effects.

Marketing orders raise the average producer priceof milk by enforcing minimum prices for milk used in fluid and soft products (e.g., yogurt). U.S. dairyfarmers as a group benefit from marketing order regulation, but the costs to consumers—in theform of higher prices for fluid milk and other dairyproducts—exceed farmer benefits. That is, marketingorders generate a net loss for the U.S. economy. More-over, the consumer costs are borne disproportionatelyby families with children and low-income households.Nutrition also suffers, because beverage demand shiftsto low-nutrition, low-priced alternatives.

Moreover, milk marketing order regulation actuallyharms dairy farms in low-cost regions. Milk prices inlow-cost regions, including the Upper Midwest andthe West, are kept artificially low by marketing orderregulations that encourage production in high-cost regions such as the South and the Northeast and discourage milk shipments across regions. Also,the central justification for milk marketing orderregulations—that the structure of dairy marketsallowed processors to exercise market power againstdairy farms—is not obviously relevant today because


54

of changes in dairy market structure. Taking advan-tage of the antitrust exemption granted to U.S. farms,dairy farmers have organized into dairy cooperativesthat market more than 80 percent of all the milk pro-duced in the United States. The large market sharecontrolled by dairy cooperatives arguably makesmarketing order regulation obsolete.

Milk Income Loss Contracts. The 2002 Farm Billintroduced a new payment scheme for U.S. dairy, theMilk Income Loss Contract (MILC), which provides adeficiency payment to dairy farmers. Payments aretriggered in months when milk prices are low, and aremade on current monthly production up to a quantitylimit per farm of 2.4 million pounds per year (theannual output of about 120 cows). Because the targetprice is so high, the program has made paymentsfrequently (in 39 of 61 months from December 2001through 2006). The USDA paid out $2.0 billion inMILC payments over the period FY2003–2005.Naturally, farms with 120 cows or more tend to receivelarger total payments, but MILC payments are a moreimportant source of revenue for smaller farms becauseof the quantity cap on payments per farm. DuringFY2003–2005, MILC payments were the equivalent of1 percent of milk revenue on California farms, com-pared to more than 4 percent of milk revenue in theUpper Midwest, where average farm size is smaller.

Indeed, MILC payments actually reduce incomesfor larger dairy farms, which account for 60 percentof the milk production. The payments induce moremilk production from smaller farms, and more milkproduction lowers milk prices. The lost milk revenuedue to lower milk prices exceeds the MILC paymentfor larger farms. Moreover, like the MPSP, the MILCprogram may prove to be an obstacle to freer agricul-tural trade because it is clearly a production enhanc-ing and trade distorting subsidy that may bechallenged in the WTO.

Reforming U.S. Dairy Policy

A report by the USDA showed that the combinedeffect of the various components of U.S. dairy policies

on the farm price of milk is nearly imperceptible; U.S. dairy policies raise average milk prices by about1 percent and dairy farm revenue by about 3 percent.Thus, the benefits to U.S. dairy farms as a whole aremodest. The same USDA report showed that con-sumer and government costs exceed the benefits toproducers. Thus, U.S. dairy policy fails the cost-benefit test.

As discussed above, the aggregate benefits to dairyfarms understate the benefits to some farms, andoverstate the benefits to others. Yet there is no com-pelling reason for a regional bias in federal dairy policy, nor is it obvious that federal policy shouldfavor smaller farms. In short, the social costs of U.S.dairy policy exceed the benefits, and regional andfarm size biases inherent in specific dairy policyinstruments are unjustified. Absent any obvious mar-ket failure, a reasonable approach to fixing theseshort-comings is to eliminate the current programs.Elimination of the status quo programs would yieldsavings for consumers and taxpayers, and wouldmake some farms (those in the Upper Midwest andthe West, as well as larger farms) better off. Elimi-nation of the current programs would also make theU.S. dairy sector more competitive in export markets,and would encourage innovation.

Elimination of the dairy policies discussed heremay not be politically feasible in the short run.However, the discussion above suggests that subsidiesfor dairy could be achieved in a less costly way thanthe status quo. Mainly, the policy should deliver rev-enue directly to dairy farmers, while minimizing mar-ket distortions. As an example of a policy that couldincrease returns to dairy farming while satisfying thisrule of thumb, consider replacing the current set ofpolicies with a direct payment per farm such that itdoes not depend on milk prices or production. Sucha payment does not eliminate the supply response tothe policy, as the payment will keep some dairy farmers in business who might otherwise exit. Butremoving the link between the payment and currentproduction and prices would dampen the supplyresponse compared to the current programs andlower overall costs for consumers and taxpayers. Byraising dairy farm income, such a payment would

U.S. DAIRY POLICY: ANALYSIS AND OPTIONS

55


56

replace the income rationale for the MPSP and mar-keting orders, which achieve enhanced dairy farmincome with more market distortions. In addition,the elimination of the MPSP and milk marketingorders eliminates the need for import restrictions,because at market-determined prices the U.S. dairysector can be competitive in world dairy markets. Inaddition to being less distorting, such a policy is alsomore transparent, facilitating cost-benefit analysis.

Conclusion

The conceptual and empirical evidence on the eco-nomic effects of U.S. dairy policy is consistent andclear: the main elements of U.S. dairy policy trans-fer income from consumers and taxpayers to dairyfarmers. Milk marketing orders and the MILC pro-gram also transfer income from some dairy farmersto others. Moreover, dairy policy distorts marketsso that the net effect is to make society as a wholeworse off. In light of these findings, this brief

identifies the potential benefits and costs of dereg-ulating U.S. dairy markets. Dairy consumers andtaxpayers clearly stand to gain. But some dairyfarms—those in the Upper Midwest and the West,as well as larger dairy farms—also stand to gain.

Of course, some dairy farms would be worse off ina world without subsidies. Thus, I also discuss waysto continue to subsidize dairy farms at lower cost andwith less market distortion. The preferred policy is adirect payment that does not depend on marketprices or production. Such a policy would reducemarket distortions and improve transparency.

Note

This policy brief draws on the author’s longer paper “U.S. Dairy Policy: Analysis and Options,” available atwww.aei.org/farmbill. The longer paper includes back-ground support and citations not included here. Balagtas isan assistant professor in the Department of AgriculturalEconomics, Purdue University. The views expressed hereare those of the author and not those of any institution withwhich he is affiliated.

Specialty crops comprise a large share of U.S. agri-culture but have been largely outside the major farmcommodity programs that are a major focus of theFarm Bill. U.S. law defines the specialty crops asfruit, tree nuts, vegetables, and nursery crops. In2005, production of these crops accounted for only3.2 percent of the harvested cropland acreage butgenerated farm revenue of around $50 billion—about 42 percent of the total crop revenues.

Of course, specialty crops do benefit from a num-ber of farm policies such as research and development(R&D), conservation programs, and crop insurance,among others. The U.S. Department of Agriculture’s(USDA) January 2007 Farm Bill proposal includedincreased financial support for specialty crops with-out creating any new subsidy programs. AgricultureSecretary Mike Johanns pointed to “broad, new effortsin the area of research and marketing and purchasingsupport to specialty crops . . . we identify $5 billionover the next decade.” An additional $0.5 billion per year in dedicated funding would represent asignificant increase in federal support for these crops,but does not begin to rival the $10 to $25 billion per year devoted to subsidies for program crops (e.g., corn, wheat, soybeans, and other grains; cotton;and oilseeds).

Specialty crops are produced throughout the United States, with the largest value of production in

California. Specialty crops cannot be treated as a singleindustry, like wheat or soybeans, because of the myriadof crops, marketing channels, seasonality, and local andfarm characteristics. Their economic interests are alsodiverse, but these commodity industries do share ageneral antipathy to direct government subsidy pay-ments and an interest in enhancing government fund-ing for what they see as public goods to supportdemand enhancement and productivity growth.

Two major sets of policy questions surround the role of specialty crops in the 2007 Farm Bill. First, ifthe rules governing subsidies for program crops arechanged to permit program crop growers to plantspecialty crops on land eligible for program cropsubsidies, should specialty crop growers receive new federal support to help level the playing field? If so, where should that extra money come from?Second, which of the many requests specialty cropgrowers are advancing for additional support are in thepublic interest?

Studies are mixed as to how much specialty cropgrowers would be harmed by removal of the specialtycrop planting restriction. Some suggest a significant,multibillion-dollar impact, while others suggest theimpacts will be small. Furthermore, recent projec-tions of large reductions in program crop subsidieslimit the ability to use these funds to compensate spe-cialty crop growers.

57

Specialty Crops and the 2007 Farm Bill

Mechel S. Paggi

R&D and programs that promote specialty cropconsumption are two proposed areas of support thatlikely pass a public interest test. Other proposedoutlays are less likely to do so.

Current Government Policy and Specialty Crops

While the traditional farm subsidy programs do notapply to specialty crops, the federal governmentundertakes a number of agricultural policies thatsupport these crops. Such policies include subsidizedcrop insurance, general government-supported R&D, management of invasive species, government-facilitated promotion programs, purchases for schoollunch and related food programs, government-sponsored international promotion programs, andoccasional disaster assistance.

Indirect Support. Federally subsidized crop insur-ance applies for all the major program crops (mainlygrains, oilseeds, and upland cotton). Crop insurancealso applies to many of the specialty crops in at leastsome regions of the country. For example, govern-ment-subsidized insurance policies are utilized byaround 44 different specialty crops. The subsidies for these crop insurance programs totaled about $300 million, or 21 percent of the total crop insur-ance subsidy, in 2006.

Marketing assistance to specialty crop producerscomes from (a) marketing orders and commoditycheck-off programs to fund domestic demand pro-motion, (b) funding to assist in the promotion ofexport sales, (c) direct government food purchases,and (d) increased sales resulting from public nutritionassistance programs. In some cases the governmentassistance programs and the related taxpayer orconsumer costs are indirect and difficult to quantify.For example, marketing orders for certain fruits andvegetables allow industries to maintain quality andquantity controls and each of these has complexeffects on producers and consumers. In other casesthe subsidy is straightforward and government out-lays are well documented. For example, the USDA,

Foreign Agricultural Service, Market Access Programprovided around $49 million (about 35 percent of thetotal) to promote export marketing efforts of 38 U.S.specialty crop commodity groups and related organi-zations in FY 2005. It is unclear, however, how muchthis program and others actually increased demandfor the commodities affected and thus how much pro-ducers gained.

As with other food commodities, specialty cropsalso benefit from direct government purchases, orsupport for purchases, through a number of domesticfood and nutrition assistance programs. The USDAadministers 15 domestic food assistance programs forchildren and low-income adults which either providefood or the means to purchase food. The five largestprograms—the Food Stamp Program, the NationalSchool Lunch Program, the Special SupplementalNutrition Program for Women, Infants, and Children(WIC), the Child and Adult Care Food Program, andthe School Breakfast Program—account for about 95 percent of the USDA’s total expenditures fordomestic food assistance. Food and nutrition programsaccounted for about 59 percent of total USDA spend-ing in 2005 ($50 billion). It is, however, difficult todetermine the exact amount which industry benefitsfrom these programs. A 2002 report by the Food andNutrition Service reported a total of around $7 billion(20 percent) from all food assistance programs wentto support the consumption of fresh and processedfruits and vegetables, but the degree to which thisfunding stimulated demand expansion is not clear.

Agricultural research and extension sponsored byUSDA also provide benefits to the specialty cropindustry. About $280 million in USDA research andextension funding is devoted to specialty crops, but ofcourse, specialty crops also benefit from generalresearch and extension that is not dedicated to thesecrops. Many of the activities of the USDA’s Animaland Plant Health Inspection Service (APHIS) also pro-vide benefits to the specialty crop industry. As withother components of government assistance to theindustry, the amount of public funding devoted tospecialty crops is not available. One exception is thefruit fly exclusion and detection program, with anannual appropriation of $60 million.


58

Direct Support. Recently, a few small ad hoc pro-grams have been specifically aimed at specialtycrops. For example, the Emergency AgriculturalAssistance Act of 2001 distributed about $160 mil-lion through block grants for state-chosen pro-grams to improve the competitiveness of U.S.specialty crops. The Specialty Crop Competitive-ness Act of 2004 was designed to provide substan-tial additional support for programs in research,marketing, education, pest and disease manage-ment, and food safety. The proposed legislationcalled for a mandatory appropriation of $470 mil-lion annually; however, Congress appropriatedonly $7 million in FY 2006.

Policy Issues for the Specialty Crop Industry

The Specialty Crop Farm Bill Alliance, composed ofmore than 70 organizations representing fruits, veg-etables, tree nuts, nursery plants, and other specialtyproducts, urges permanent and mandatory fundingin the 2007 Farm Bill. One recent product of theAlliance’s efforts is the proposed “EquitableAgriculture Today for a Healthy America Act,” or the “EAT Healthy America Act,” which proposesincreased support for specialty crop block grants,R&D, invasive, disaster assistance, conservation,international trade promotion, pest and disease con-trol, nutrition, and more. The Alliance claims thisnew support is all aimed at supporting the compet-itiveness of the specialty crop industry. The Alliancehas been clear that it does not have interest in directpayment or price support programs, such as is pro-vided for program crops.

A major issue for specialty crops in the 2007 FarmBill relates to a specific provision of the commodityprograms for grains, oilseeds, and cotton. Beginningwith the 1990 Farm Bill, commodity program partic-ipants were allowed to plant other crops on a portionof their program base acres, but were generally pro-hibited from planting fruits, tree nuts, vegetables,including dry edible beans and potatoes, and (in alater amendment) wild rice. This disincentive to pro-gram participants has created a kind of supply control

measure for the specialty crop industry. Existingspecialty crop producers benefit from the restrictionon program base land entering specialty crop produc-tion; however, those benefits are gained throughhigher specialty crop prices at the expense of mar-keters and consumers of specialty crops.

In 2005, the WTO decision in the disputebrought by Brazil against the U.S. cotton programpointed to these planting restrictions in ruling thatdirect payments would be counted in the overallU.S. support for cotton. This indicated to many ana-lysts that direct payments for program crops couldnot be counted as only “minimally trade distorting”because they restricted the subsidized base landfrom planting fruits, tree nuts, vegetables, melons,or wild rice. One proposed solution to this problemis to continue the direct payments to program cropsbut remove the planting restriction from programcrop rules.

Specialty crop producers see this as a threat andhave argued that a potential shift of program baseacreage into specialty crops would depress prices fortheir crops, which are planted on relatively few acrescompared to the program crops from which theacreage would be released. They also argue that itwould be simply an unfair advantage if programbase holders were allowed to switch into specialtycrops and maintain their direct payments.

There is no question that if restrictions were lifted,some acreage would shift from program crops, orother payment-eligible land uses such as hay, to therestricted crops. Measuring the magnitude of theprice impacts on individual crops, however, is quitedifficult, and no consensus has emerged from thefew studies that have taken a partial look at theissue. Much of the program crop base acreage isoutside the regions where most specialty crops aregrown. Moreover, where program base land doesoverlap with agronomic conditions suitable for highvalue per acre specialty crops, such as the CentralValley of California, much of the former base landhas already shifted to specialty crops. Nonetheless,there are large potential price effects from even asmall share of the 266 million acres of programcrops shifting to individual small-acreage specialty

SPECIALTY CROPS AND THE 2007 FARM BILL

59


60

crops, where even a few thousand acres of newproduction can have large price impacts.

Evaluation of Policy Issues and Options for the 2007 Farm Bill

There are two major issues in the debate surroundingthe 2007 Farm Bill of particular import to the specialtycrop industry. First, what is the source of money tofund additional programs demanded by specialty cropadvocates? Second, what, if any, additional supportshould be provided the specialty crop industry if thefruit and vegetable planting restrictions are eliminated?In addition, any proposed new programs in support ofthe specialty crop industry should be evaluated ontheir own merits relative to competing uses of thefunds, including reducing the federal budget deficit.

With regard to budgetary considerations, it isunlikely that additional funds for specialty crops cancome simply from diverting money from programcrops. The USDA recently projected mandatoryspending on farm programs to decline from a record$32.3 billion in 2000 to $11.7 billion in 2008, if the2002 Act were to continue (which typically estab-lishes the outlay baseline). Because prices for themajor program crops are expected to remain high,baseline funds available for spending on (Farm BillTitle I) commodity programs were estimated to beabout 42 percent less from 2008 to 2017 than wasprovided in the 2002 Farm Bill. In the absence ofsome additional funds, spending constraints of thismagnitude would require diverting much of theremaining subsidy funding from existing programs(and their recipients) if significantly increased fund-ing for specialty crops were to occur.

In addition, specialty crop producers will expectadditional funds as compensation for their potentiallosses if planting restrictions on program crop baseare eliminated. The amount of such losses has beenestimated by some studies to initially be as high as$2.4 billion, while others suggest overall marketimpacts are likely to be quite small.

Regardless of whether new funds for the specialtycrop industry results from shift in policy direction or

from compensation for elimination of plantingrestrictions, any new spending should be evaluatedfor its contribution to the public good. For somepolicies, support of the specialty crop industry canbe readily identified with such a goal. For example,agricultural R&D outlays have consistently shownhigh social rates of returns, and funds to supportbetter environmental performance of specialty cropsand additional effort to reduce vulnerabilities toinvasive species may also have positive socialreturns. Programs that stimulate the consumption offruits, vegetables, and nuts by those who would oth-erwise consume less than recommended amountswould contribute to a healthier overall diet, with thepublic benefit of potentially reduced national med-ical costs.

Other proposed outlays are less clearly in thepublic interest. For example, funds to promoteexport sales may provide benefits to growers of indi-vidual commodities, but they have at best uncertainreturns to taxpayers. Industry-specific promotionprograms that increase consumption of one spe-cialty crop product at the expense of another alsohave questionable social benefits. Some R&D fund-ing may also have questionable domestic social ratesof return if, for example, beneficiaries are largely for-eign consumers. Generally, proposed spending pro-grams for specialty crops must be evaluated on theirmerits, not simply because they are labeled as spe-cialty crop programs.

Summary and Conclusions

The debate surrounding a Farm Bill involves a largecast of characters with many different interests. In2007, among the more notable voices in this groupare representatives of the specialty crop industry—apart of agriculture that has traditionally receivedmodest federal assistance compared to producers ofcommodities such as grains, oilseeds, and cotton. Asdescribed above, the specialty crop industry hasalways received some support from federal govern-ment programs. The total value of federal support forspecialty crops is difficult to calculate, but that

support is mostly indirect and is quite small (relativeto their market value of output) compared to subsi-dies for program crops.

As suggested by the comments of AgricultureSecretary Johanns, and provisions of the USDA’s 2007Farm Bill proposals, increasing government supportfor specialty crops is on the policy agenda. But, it is not a sufficient rationale for increased spending to observe that federal support for specialty crops is modest compared with support for program crops.If Congress decides to use taxpayer dollars to sup-port specialty crops, then it makes sense to support

services that would not be provided by the privatesector and that achieve positive net public benefits.

Note

This policy brief draws on the author’s longer paper “U.S.Specialty Crops: Industry Structure and Linkages toFederal Farm Policy,” available at www.aei.org/farmbill.The longer paper includes background support and cita-tions not included here. Paggi is director of the Center forAgricultural Business, California State University, Fresno.The views expressed here are those of the author and notthose of any institution with which he is affiliated.

SPECIALTY CROPS AND THE 2007 FARM BILL

61

Livestock production accounts for about one-half oftotal U.S. farm cash receipts, yet apart from milk andwool, livestock producers have not been direct recipi-ents of agricultural support payments. However, manyelements of U.S. agricultural policy do have conse-quences for livestock producers and meat consumers.This brief reviews those policies, assesses their conse-quences, and recommends areas for reform.

Recommendations for Reform

Biofuels: Congress should be aware of the adverseeffects of subsidizing biofuels (ethanol) on the live-stock sector. Biofuel production subsidies are likelyto result in higher prices for consumer meat prod-ucts that will have the largest adverse effects on thepoorest U.S. households.

Grazing fees: Congress should consider changes inthe allocation of grazing fee permits that allow agri-cultural and other interests to compete for the use ofpublic lands.

Waste discharge: The environmental conse-quences of animal waste disposal have increased inrecent years because of the relocation of some house-holds to more rural areas. Congress should, therefore,closely examine who pays for agricultural compliancewith waste discharge and other EPA mandates, and

consider paying for at least some mitigation costswhere appropriate. Funding for the research anddevelopment of technologies that can economicallymitigate these externalities could be an important pol-icy contribution of the 2007 Farm Bill.

Tariffs and duties: Many industries, including theU.S. livestock sector, have tried to protect them-selves from foreign competition by requestingantidumping tariffs and countervailing duties.While these actions may be appropriate in some sit-uations, Congress should set a high standard ofproof of unfair competition, and also impose strin-gent sunset provisions on retaliatory actions wherecountervailing measures are justified.

Animal identification: Congress should explorewhether a mandatory or voluntary animal identifica-tion program satisfies domestic and foreign concernsabout animal health (e.g., BSE1 or “mad cow disease”).

Price reporting: Given that the fixed costs associ-ated with mandatory price reporting (MPR) havealready been incurred and variable costs are modest,reinstating MPR may be a reasonable course of action.

Review of Specific Policies

Feed Grain Policy. Between 2000 and 2005, U.S.feed grain production averaged 11.1 billion bushels,

63

Agricultural Policy and the U.S. Livestock Industry

Gary W. Brester and Vincent H. Smith

of which 95 percent (10.3 billion bushels) consistedof corn. Over the same period, the U.S. importedless than 1 percent and exported about 19 percent oftotal domestic feed grain supplies. About 56 percentof domestic feed grain production was used domes-tically by the livestock industry and 26 percent bythe food, industrial products (mostly ethanol), andseed industries.

In 2006, 2.15 billion bushels of corn (about 20 percent of the 2006 crop) were used to producefuel alcohol (ethanol). New ethanol plants under con-struction will add an additional 2.1 billion gallons ofcapacity and, in 2010, ethanol production is expectedto reach 7 billion gallons and use 25–30 percent of U.S. corn production.

Clearly, increased demand for corn by ethanol pro-ducers will increase corn prices, and increase bothlivestock production costs and consumer prices formeat products. Thus, energy policies that impact thefeed grain industry should not be crafted within avacuum by ignoring concomitant effects on the live-stock sector.

Expansion of Crop Insurance to LivestockProduction. The 2000 Agricultural Risk ProtectionAct (ARPA) expanded the scope of subsidized cropinsurance by requiring the USDA Risk ManagementAgency (RMA) to develop insurance productsspecifically targeted toward livestock producers.Several such products have been developed (e.g.,Livestock Risk Protection, Livestock Gross Margin)to provide insurance against downward future live-stock price movements or increases in input costs.RMA also offers subsidized forage insurance prod-ucts across the United States. In addition, a series ofpilot area yield rangeland insurance products will beoffered in parts of fourteen states in 2007. The pre-miums for these area yield products are subsidizedbetween 55 and 64 percent, depending uponselected coverage levels. If offered on a nationalbasis, these programs have the potential to exposethe federal government to substantial subsidy out-lays, perhaps in excess of $2 billion annually.

Moral hazard and adverse selection problems areendemic in crop insurance programs. Consequently,

only a few crop insurance products have been offeredby the private sector in the absence of federalsubsidies. However, subsidized insurance programsgenerate well-known inefficiencies, inappropriateincentives, and fraudulent activities. Therefore,increasing the number of commodity offerings and/or the level of insurance subsidies is likely to causeadditional misallocations of resources and waste.

Wool Program. The National Wool Act of 1954 andits amendments supported wool (and mohair)prices until 1995. Although the Act authorized thesupport of wool prices through loans, purchases,payments, and other means, the primary form ofsupport actually consisted of direct payments towool producers using a complex, parity-based for-mula. During 1990–1993, the value of U.S. woolproduction averaged almost $54 million annually,while over the same period, direct wool supportpayments to producers averaged almost $122 mil-lion annually and represented about 18.5 percent ofthe total value of sheep, lamb, and wool production.

The 1954 Act was repealed in 1993, and the pro-gram was phased out by 1996. However, followingclaims from domestic producers that imports werematerially injuring their industry and the abandon-ment of tariffs on lamb imports, the 2002 FarmSecurity and Rural Investment Act re-introduced non-recourse marketing assistance loans and loan defi-ciency payments for wool and mohair productionwhich had about the same producer price-supportingeffect as the National Wool Act.

The lamb and wool industry is small relative to the beef, pork, and poultry industries. Nonetheless,the lamb industry has succeeded in lobbying forSection 201 trade resolutions, direct payments, andtarget prices. A standard economic welfare analysiswould conclude that the wool program generatesdeadweight losses in several domains and transferseconomic welfare from consumers and taxpayers to producers. Political economy models explain the re-emergence of support for wool prices. Individ-ual consumers, who on average consume only 1.1 pounds of lamb meat each year and increasinglyprefer non-wool textiles, bear only small costs. Sheep


64

production operations receive large benefits, and theaggregate cost to taxpayers is also small (a veritabledrop in the 2.3 trillion dollar federal budget bucket).Therefore, it seems likely that a combination of targetprices, marketing assistance loans, and LDP programswill continue in the future. The wool program pro-vides a classic example of the efforts that are likely tobe pursued by a variety of “small” agricultural indus-tries throughout farm legislation discussions.

Land Use Policies and Public Lands Grazing.Nearly 35 percent of the total land area in the UnitedStates is at least partially used for grazing. The 1934Taylor Grazing Act, the basis for livestock grazingpolicies on public land, created exclusive livestockgrazing rights on many tracts of land. Recreationistsand conservationists have increasingly advocated thatpublic lands be managed for increased non-grazingactivities. However, the two groups disagree to someextent about how public land use should be changed.

Recreationists generally want public lands man-aged to increase wildlife and public access. In con-trast, conservationists generally want less grazing and reductions in all types of human activities onpublic lands. These issues were explicitly addressedby the 1976 Federal Land Policy and ManagementAct, which added the goals of sustaining grazingyields and multiple land use to the management ofpublic lands.

Public land use conflicts are likely to intensify,given competing interests of ranchers, recreationists,conservationists, and government agencies. Establish-ing bidding rights for the use of public lands for all of these groups may provide appropriate incentivesfor land stewardship and reduce conflicts over theallocation of these lands among competing uses.

Environmental Policies. Livestock production gen-erates two primary environmental externalities—odor and surface water discharges—that involveboth point-source and nonpoint-source pollution.The Clean Air Act (CAA) of 1970, as amended in1990, established limits on particulate discharges.Confined Animal Feeding Operations (CAFOs) pro-duce a variety of air pollutants including: (1) odiferous

gases that contain a variety of compounds, (2) hydrogen sulfide, (3) methane, and (4) ammo-nia. To regulate these emissions, the U.S. Environ-mental Protection Agency (EPA) established the AirQuality Compliance Agreement program underwhich Animal Feeding Operations (AFO) emissionsare monitored.

Water pollution from CAFOs is treated as point-source pollution and regulated under the Clean WaterAct of 1972. CAFOs larger than a minimum size aregoverned by the National Pollution DischargeElimination System (NPDES). They must obtainNPDES permits by specifying waste managementplans including nutrient management plans for theapplication of manure to land.

The Coase theorem provides no moral guidance asto who should pay for the mitigation of a negativeexternality. This is especially the case with respect toagricultural waste management, where, for example,markets for tradable pollution rights may be difficultto establish and may cause substantial regional pollu-tion issues. In addition, if transactions costs are large,most economists agree that government interventionis necessary to establish property rights or provideregulatory oversight. Additional policies need to bedeveloped that allow livestock feeding operations tointernalize the negative externalities associated withodor and nutrient runoff associated with animalwaste. Funding for the research and development oftechnologies that can economically mitigate theseexternalities could be an important policy contribu-tion of the 2007 Farm Bill.

Animal Identification. In response to the discoveryof BSE in the United States in December 2003, theSecretary of Agriculture declared that the USDAwould develop a National Animal IdentificationSystem (NAIS) for all livestock species. In April2004, the Secretary announced a framework forimplementing an NAIS for premise verification ofdiseased animals.

Three overarching issues have surfaced in terms ofthe need for such systems. The first involves the needto protect public health by ensuring that safe meatproducts are provided for human consumption. The

AGRICULTURAL POLICY AND THE U.S. LIVESTOCK INDUSTRY

65

second involves access to overseas markets, whichincreasingly require that source countries operateanimal identification systems. The third concerns mit-igating the potential for bio-terrorist activities thatcould affect the livestock sector. However, some live-stock producers are concerned about the intrusivenature and costs of such systems. One importantproblem involves data management, data ownership,and potential product liability.

The livestock industry’s need to maintain domesticconsumer confidence and access to export marketshas provided an impetus for the establishment of vol-untary and/or private animal identification systems.However, government action may be necessary toaddress “free-rider” and credence problems to main-tain the competitiveness of the U.S. livestock industry.

Antidumping/Countervailing Duties. Over thepast decade, U.S. cattle producers have sought protection from foreign competition, but with lim-ited success. Even when successful, as in the casebrought by the Ranchers-Cattlemen Action LegalFund (R-CALF), advocates sometimes suffer buyer’sremorse as a result of the unintended consequencesof such policy pursuits. Indeed, once a tariff is estab-lished, removing it is often politically difficult.Further, industries often claim economic losses fromdumping when, in fact, such effects are the result ofchanges in global and local market conditions.

In general, it seems wise to require that an indus-try establish an indisputable body of evidence thatdumping is taking place before any countervailingtrade restriction is enacted. Further, antidumpingpolicies should be subject to sunset provisions thatrequire frequent reviews of whether or not theexporter is continuing the practice.

Country-of-Origin Labeling. Concerns about theadverse effects of U.S. meat and livestock imports ondomestic livestock prices have also increased inter-est in country-of-origin labeling (COOL) legislation.Proponents of this legislation argue that: (1) con-sumers have the right to know and choose thesource of their meat products, (2) COOL wouldenhance food safety and quality, and (3) COOL

would increase the demand for domestically pro-duced products and improve domestic livestockprices. Opponents argue that implementation ofCOOL would be prohibitively expensive because ofproduct blending, the number of ownershipexchanges that occur in commodity livestock andmeat markets, and the complexity of the meat sup-ply chain.

The 2002 Food Security and Rural Investment Actadded a new subtitle to the Agricultural MarketingAct of 1946 that instituted voluntary COOL in 2002to be followed by mandatory COOL in 2004 forunprocessed fresh, frozen, and ground beef and pork.But since then, mandatory COOL has been delayedtwice (except for wild and farm-raised fish andshellfish) and is now scheduled to be implemented in 2008.

Given that the U.S. meat industries are mature,expansion of the U.S. meat sector will likely hinge onforeign markets. The imposition of U.S. import tradebarriers is likely to inhibit the industry’s ability toaccess foreign markets. In addition, such actions mustbe viewed within the context of WTO obligations.

Mandatory Price Reporting. The use of spot mar-kets has disappeared in the poultry industry, greatlydiminished in the hog industry, and also becomeless important in the cattle and lamb industry overthe past 15 years. Traditionally, spot market priceswere voluntarily reported to the Agricultural Mar-keting Service’s (AMS) Market News by buyers andsellers and often formed the basis for other negoti-ated and formula sales. Price, quantity, and qualityinformation regarding alternative marketing arrange-ments were often considered proprietary. In 1999,the Livestock Mandatory Price Reporting Act (MPR)required that beef, pork, and lamb prices bereported over the 2001 to 2005 period by large meatpackers and importers, including premiums anddiscounts for quality characteristics. Over 90 percentof commercial slaughter cattle prices were beingreported when the MPR Act expired in 2005. Pro-ducer groups are seeking to reinstate MPR on the basisthat it provides price and quantity information that:(1) can be readily understood by market participants,


66

(2) facilitates price discovery, (3) improves USDA pricereporting services, and (4) encourages competition.

Most research indicates that the MPR Act providedmuch needed data for the analyses of market pricingefficiency within the livestock sector. Given that thefixed costs of developing the information technologyfor such reporting have already been incurred andthat the variable costs of reporting are reasonablysmall, it appears that reinstating MPR would be areasonable course of action.

Market Structures. The beef, pork, poultry, andlamb industries have experienced considerablestructural change over the past 30 years. In general,each sector became more concentrated as firmsadopted low-cost or differentiation business strate-gies. The extent of meat packing concentrationvaries among beef, pork, poultry, and lamb pro-cessing. Currently, the largest four packers in thebeef industry currently slaughter and process about 80 percent of all cattle marketed in the UnitedStates compared to 40 percent in 1980. In the hogindustry, the four largest packers slaughtered over64 percent of all hogs in 2004. In the poultry pro-cessing industry, the four largest packers slaugh-tered 54 percent of all poultry in 2004. The lambpacking industry has also experienced substantialincreases in concentration. The four-firm concen-tration ratio has increased from 25 percent in 1980to 70 percent in 2005.

Increased packer concentration may be the resultof scale economies obtained by larger plants, or oper-ational efficiencies of multiple plants gained from theconsolidation of existing assets, or both. If theseeffects are relatively large, then increased concentra-tion is likely to lower consumer meat prices and/orraise livestock prices. However, increased concentra-tion may also increase oligopoly selling power and/oroligopsony buying power. If this is the case, thenincreased concentration would increase retail meatprices and reduce livestock prices.

Over the past two decades, livestock producershave targeted meat packing concentration throughlegislative and judicial means. Nonetheless, mostresearch suggests that the use of anti-trust regulation

to reduce packer concentration may be counterpro-ductive, at least from the perspective of livestock pro-ducers and consumers. Any gains in the form ofhigher input prices and lower consumer prices fromreduced packer concentration are likely to be offset byincreased processing costs.

Marketing Arrangements and Captive Supplies.Marketing arrangements refer to the methods bywhich livestock and meat are transferred throughsuccessive production and marketing stages and themethod by which prices are determined for eachtransaction in a vertical supply chain. Historically,most of the meat industry relied upon negotiatedagreements or spot (auction) markets for price dis-covery. Over time, the variety, complexity, and use ofalternative livestock and meat marketing arrange-ments have increased. Currently, most livestock aretransferred among production stages using a portfo-lio of forward contracts, formula prices, productioncontracts, marketing contracts, integrator alliances,and packer-ownership of livestock rather than auc-tion markets. The use of these “alternative marketingarrangements” raises several questions about theireffects on economic efficiency and welfare distribu-tions among producers and consumers.

Food processor market power is detrimental toconsumers and input suppliers if competition isreduced, entry barriers are increased, and industryoutput declines. However, vertical integration/coor-dination may or may not increase consumer pricesand reduce livestock prices. Increased cost efficien-cies, reduced uncertainty, and improved productquality may offset negative effects of increased mar-ket power. Vertical integration or coordinationbetween adjacent upstream and downstream firmswhere each has market power can be welfareenhancing if market distortions between two adja-cent firms are internalized.

Conclusion

In summary, the livestock sector, like most othersectors of U.S. agriculture, is affected by a plethora

AGRICULTURAL POLICY AND THE U.S. LIVESTOCK INDUSTRY

67

of agricultural and other government policies—some of which are favorable while others are detri-mental to livestock producers. Some involve incometransfers that impose efficiency losses on markets,while others may improve economic efficiency byaddressing distortions associated with externalitiesand/or market power. Clearly, considerable scopeexists for policy analyses that improve economic effi-ciency. In this respect, the U.S. livestock sector hasmuch in common with other components of U.S.agriculture and, for that matter, other sectors of theU.S. economy.

Note

This policy brief draws on the authors’ longer paper“Agricultural Policy and the U.S. Livestock Industry,”available at www.aei.org/farmbill. The longer paperincludes background support and citations not includedhere. Brester and Smith are professors in the Departmentof Agricultural Economics and Economics, Montana StateUniversity. The views expressed here are those of theauthors and not those of any institution with which theyare affiliated.

1. BSE is an abbreviation of Bovine Spongiform Ence-phalopathy.


68

Equity concerns motivate most policies that aim toredistribute resources among the population. Equityconcerns played a key role in the adoption of farmsubsidies in the 1930s when urban interests joinedfarmers and rural legislators in order to give “agricul-ture a fair share in the national income.” By the 1960s,when the farm population had dwindled and subsi-dies were ballooning, equity concerns prompted criesto limit, reduce, or even eliminate farm subsidies.That debate over equity came to a head in the 1970Farm Bill when Congress enacted payment limits onagricultural subsidies for the first time. Recently, inlight of greater transparency in the subsidies receivedby individuals and against the backdrop of the WorldTrade Organization Doha Development Round nego-tiations, concerns about equity in farm subsidies havereemerged as a major policy issue. Indeed, U.S.Secretary of Agriculture Mike Johanns has repeatedlycalled for “equitable” farm policy.

An effective discussion of equity must considerwho receives agricultural subsidies, who benefitsfrom the subsidies, and how effective are regulationsaimed at changing the distribution of agriculturalsubsidies. This paper presents evidence on each ofthese three equity considerations. It demonstratesfirst that, among subsidy recipients, the concentra-tion of subsidies has changed little in 40 years, and subsidies reduce inequality among recipients.

Second, contrary to conventional wisdom amongeconomists, landlords do not extract the full subsidythrough higher rental rates. In fact, landlords extractonly 25 cents from the marginal subsidy dollar.Finally, limits on total subsidy receipts appear to beineffective as farms reorganize in order to maximizetheir subsidy payments.

Who Receives Farm Payments?

U.S. agricultural subsidies aim to increase the viabil-ity of U.S. farms by subsidizing qualified agriculturalland (e.g., Direct Payments and CountercyclicalPayments) and the production of major commodities(e.g., Loan Deficiency Payments and Marketing LoanGains). Consequently, large, productive farms receivelarger payments than small, less productive farms. Infact, 58 percent of subsidies go to just 8 percent offarms. Conversely, a majority of farms (57 percent) donot grow subsidized crops or farm subsidized land.These farms receive no farm payments at all.

The equity debate often focuses on the wellbeingof farm households, specifically subsidized farmhouseholds, relative to the general population.Historically, farm incomes have been far below thoseof the general population. Recently, the income distri-bution among farmers more closely resembles the

69

The Distribution of U.S. Agricultural Subsidies

Barrett E. Kirwan

overall income distribution in theUnited States. Prior to receivingsubsidies, the median (average)farm-household income from bothfarm and non-farm sources in2003 was $45,118 ($63,844),while the median (average) U.S.income in 2003 was $43,300($59,200).1 Subsidies contributesignificantly to recipient income.Prior to receiving subsidies, subsi-dized farm households had alower median income ($43,656)than farm households that wouldreceive no subsidy ($47,282).Accounting for subsidies increasesthe median income of recipienthouseholds to $48,520, and themedian income of all farm house-holds to $47,620, resulting infarm-household income that is 10 percent higher than the medianU.S. household income.

At least since the 1960’s, theconcentration of subsidies amongthe largest farms has been a fea-ture of U.S. agricultural policy.This fact is most effectively illus-trated using Lorenz curves.Lorenz curves, commonly used toillustrate the degree of incomeinequality within a country, illus-trate the cumulative share ofincome across a population, mov-ing from poorest to richest. Adiagonal line illustrates perfectequality, while the farther thecurve bows from the diagonalline, the greater the inequality inthe income distribution. The Lorenz curves in figure1 illustrate the subsidy distribution among subsidyrecipients in 1968 and 2002. These curves revealhigh subsidy concentration in 1968 and very littlegrowth since then. Although not illustrated here,Lorenz curves over all farms, not just those receiving

subsidies, portray a much greater degree of inequal-ity because, as noted above, most farms receive nopayments.

A criticism commonly heard in the media is thatthe U.S. system of subsidizing farmland and pro-duction leads to a situation where small farms that


70

FIGURE 1SUBSIDY LORENZ CURVES

SOURCE: The 1968 Lorenz curve comes from Paulsen, Arnold. 1969. “Payment Limitations: TheEconomic and Political Feasibility.” American Journal of Agricultural Economics 51(5) 1237–42.The 2002 Lorenz curve is the author’s calculation using data from the USDA–NationalAgricultural Statistics Service’s “U.S. Census of Agriculture.”

FIGURE 2

SOURCE: Author’s calculations using data from USDA–National Agricultural Statistics Service’s“U.S. Census of Agriculture,” 1987, 1992, 1997, 2002.

1

.9

.8

.7

.6

.5

.4

.3

.2

.1

0%

of S

ubsi

dies

Culmulative population proportion

0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1

Line of perfectequality

2002

1968

Ave

rage

Sub

sidy

Rat

e

.3

.2

.1

0

--.15 10 15 20

Assets (log dollars)

1987

1997

1992

2002

might most benefit from subsidies receive the fewestsubsidy dollars and seem to be treated unfairly.However, looking at subsidies relative to wealth orassets, much as we discuss taxes relative to income,reveals that small, subsidized farms receive highersubsidies relative to their assets than do large farms.Figure 2 illustrates the relationship between subsi-dies as a proportion of assets and total farm-levelassets in 1987, 1992, 1997, and 2002. The down-ward sloping lines depict a negative relationship: thelower a farm’s assets, the higher their subsidies arerelative to those assets. In other words, agriculturalsubsidies appear to be progressive, in spite of adesign agnostic to recipient wealth.

Who Benefits from Farm Subsidies?

Traditionally, economists have regarded landowners(including farmers that own their own land) as theultimate beneficiaries of most agricultural subsidies.The story often told is that potential tenants competeagainst each other, bidding up the rent for subsidizedland, until the rental rate fully reflects the subsidy. Thelandlord receives the subsidy through higher rentalrates, and the tenant is in the same position as if therewas no subsidy. Since non-farmers own 42 percent ofall U.S. farmland and nearly all rented farmland, if theconventional wisdom holds true, nearly half of farmsubsidies ultimately benefit non-farmers.

Recent analysis, however, reveals the conventionalwisdom to be unfounded. Agricultural subsidies havea much smaller effect on rental rates than previouslybelieved. According to recent evidence, landlordsextract only 25 percent of the subsidy through higherrental rates, while tenants take home the remaining75 percent.

These findings suggest that, for the most part, agri-cultural subsidies increase the cash-flow of recipientfarmers. Although asset values might be inflatedsomewhat due to subsidies, eliminating agriculturalsubsidies will not have the dire consequences onewould expect if the entire subsidy had been capital-ized into farmland values. Removing subsidies willdecrease farmers’ cash-flow, but it will not result in

large capital losses in farmland values, even in regionswith a high share of land devoted to program crops.

How Effective Are Regulations Aimed at Changing the Distribution of

Agricultural Subsidies?

In the name of fairness, the 1970 Congress enactedlimits on the total subsidy any single farmer couldreceive. Additionally, equity concerns prompted Con-gress to begin income-testing agricultural subsidies in2002 by denying subsidies to individuals with anadjusted gross income greater than $2.5 million.Current proposals seek to tighten payment limits and further decrease the income test. For example,the USDA has proposed excluding farmers withadjusted gross incomes greater than $200,000 frommost payments.

In spite of these efforts, critics claim that loopholes,lax enforcement, and a vague definition of a “farmer”allow farms to effectively avoid payment limitationsand collect virtually unlimited subsidies. Figure 3, onthe following page, presents evidence consistent withthese claims. It illustrates that payment limitationsmay constrain the amount of subsidy an individual“farmer” receives but not the amount received by the farm.

Figure 3 illustrates the distribution of Direct Pay-ment for wheat and rice in the 2004 crop year.Panels A and C illustrate the distribution of pay-ments across FSA farms for wheat and rice, respec-tively. The crop-specific subsidy distributions overFSA farms are smooth and symmetric. Super-imposed on each of the panels is a solid, vertical linedepicting the $40,000 payment limit for DirectPayments. At the FSA farm level, it appears thatrelatively few wheat farms receive subsidies abovethe cap. Yet a considerable portion of the rice-farmdistribution lies above the payment cap.

Of course, payment limits apply to members ofthe farm organization, i.e., “farmers,” not to the farmitself. Panels B and D, therefore, depict the distribu-tion of direct payments across farm members forwheat and rice, respectively. The distribution for

THE DISTRIBUTION OF U.S. AGRICULTURAL SUBSIDIES

71

wheat farm members in panel B is smooth and sym-metric, reflecting the distribution across FSA farms,with relatively few people receiving payments abovethe limit.2

The rice subsidy distribution across farmers inpanel D, however, tells an entirely different story.Rather than having a standard bell-curve shape, thisdistribution is bimodal, with a sharp peak justbefore the payment limit. The irregular shape of thisdistribution, the coincidence of the distributionpeak for rice farm members just before the paymentlimit, and the relatively high proportion of FSA ricefarms affected by the payment limit suggest thatconstrained rice farms restructure their organizationin order to, essentially, bypass payment limits.

Implications for the 2007 Farm Bill

An effective discussion of fairness in the farm bill mustconsider who receives agricultural subsidies, who ben-efits from the subsidies, and how effective are regula-tions aimed at changing the distribution of agriculturalsubsidies. The answers to these questions presentedabove have the following policy implications.

The structure of agricultural subsidies reveals theyare meant to subsidize farm production. In contrast,much of the equity debate focuses on subsidizingfarms/farmers, per se. The research presented hereunderscores the long-heralded concentration of farmsubsidies, yet it demonstrates that farm subsidiesreduce inequality among recipients. In spite of this

72

FIGURE 3COMPARISON OF PAYMENT DISTRIBUTIONS ACROSS FARMS AND FARMERS FOR WHEAT AND RICE

SOURCE: Author’s calculations; data obtained through Freedom of Information Act Requests, USDA-Farm Services Agency.

b. Wheat FarmersDirect Program Payments

Wheat—2004

a. Wheat FarmsDirect Program Payments

Wheat—2004

c. Rice FarmsDirect Program Payments

Rice—2004

d. Rice FarmersDirect Program Payments

Rice—2004

.3

.25

.2

.15

.1

.05

0

0 1 2 3 4 5 6 7 8 9 10 11 12

.3

.25

.2

.15

.1

.05

0

0 1 2 3 4 5 6 7 8 9 10 11 12

.3

.25

.2

.15

.1

.05

0

0 1 2 3 4 5 6 7 8 9 10 11 12

.3

.25

.2

.15

.1

.05

0

0 1 2 3 4 5 6 7 8 9 10 11 12

Den

sity

Den

sity

Den

sity

Den

sity

Subsidy Amount (Log $)

Subsidy Amount (Log $) Subsidy Amount (Log $)

Subsidy Amount (Log $)


finding, it stands to reason that policy would betteraddress equity concerns directly rather than as anunintended side-effect of production subsidies.

A precipitous decline in farmland value is an oft-cited barrier to the removal of farm subsidies. Recentevidence suggests that the positive correlationbetween subsidies and farmland values is not causal,and removing farm subsidies will not cause large,widespread capital losses.

Finally, if payment limitations are meant to haveteeth, policy must more clearly define a “farmer” andmore closely scrutinize farm reorganizations in theface of binding payment limits. Otherwise, paymentlimits should be removed in order to avoid wastefulfarm reorganizations.

Notes

This policy brief draws on the author’s longer paper “The Distribution of U.S. Agricultural Subsidies,” avail-able at www.aei.org/farmbill. The longer paper includes background support and citations not included here.Kirwan is an assistant professor at the University ofMaryland, College Park. The views expressed here arethose of the author and not of any institution with whichhe is affiliated.

1. Substantial year-to-year variation in farm incomeresults in extremely high and low income in any given year.The Winsorized mean, a measure of central tendency thataccounts for extreme outliers, of farm income in 2003 is$62,146.

2. Payments above the cap are possible through the“three-entity” rule.

THE DISTRIBUTION OF U.S. AGRICULTURAL SUBSIDIES

73

For the past 30 years, Congress has struggled withhow best to provide protection against natural disas-ters to agricultural producers. In 1980, Congressreplaced a standing disaster program with subsidizedcrop insurance delivered by the private sector.

Today, 27 years after the passage of the 1980 Actand two reform bills later, crop insurance hasbecome a major fixture in the farm safety net, and bymost measures of participation, crop insurance is amajor success. Yet, despite participation rates forcrop insurance that vastly exceed what was envi-sioned in 1980, Congress continues to pass ad hocdisaster bills.

Subsidies for crop insurance have averaged morethan $3 billion a year since 2002, and annual disas-ter payments have averaged more than $2 billion.More-over, much of the disaster assistance goes toproducers who are also receiving crop insuranceindemnity payments. The result, as the title of thispaper suggests, is “double indemnity.” For many pro-ducers, disaster assistance allows them to collecttwice on the same loss.

The time is ripe to reconsider the federal cropinsurance program, given its high costs, its failure toprevent disaster assistance, its potentially distortingeffects on production and input use, and its relativeinefficiency at transferring benefits. The followingdiscussion considers five potential options: eliminate

the crop insurance program; increase crop insuranceparticipation by making crop insurance purchasescompulsory for producers who receive commodityprogram benefits; continue the current program, butprovide supplemental coverage based on area-yieldand revenue insurance plans; reduce program costsby reducing delivery costs; and allow the private sec-tor companies who actually sell the insurance tocompete more directly on rates.

The Rationale for Government Intervention

The failure of private agricultural insurance marketsis a primary justification for government interven-tion. However, it is far from clear that this justifica-tion is sound.

Advocates of subsidized insurance often note thatcrop insurance faces problems of moral hazard,adverse selection, and a correlation of risk acrossfarmers. Moral hazard occurs when an insuredproducer can increase his or her expected indemnityby actions taken after buying insurance, a practicewhich insurers typically combat through deduct-ibles, co-payments, or other loss-sharing mecha-nisms. Adverse selection occurs when a producer has more information about the risk of loss than doesthe insurer, and is better able to determine whether

75

Double Indemnity: Crop Insurance and the Failure of U.S. Agricultural Disaster Policy

Joseph W. Glauber

premium rates are fair. Finally, correlation of risksmeans that insurers cannot easily diversify their risksacross space, and would have to hold large reservesin the event of a large crop loss, thereby making suchinsurance more expensive to purchase.

The problems of moral hazard, adverse selection,and correlated risks are certainly not unique to cropinsurance. Other lines of insurance face similar prob-lems, yet private markets exist. However, it is possiblethat the costs of addressing these problems in the cropinsurance context are high enough to make the over-all costs of insurance too high to support a viable mar-ket. Thus, the debate over the last 70 years has focusednot on whether Congress should provide assistance tofarmers who suffer disaster-related crop losses, butrather on the form such assistance should take.

Crop Insurance versus Disaster Assistance: Why Not Both?

The purpose of the 1980 Act was to make crop insur-ance the primary form of catastrophic loss protectionfor producers, but its success depended upon highparticipation rates. Yet, despite certain changes the Actmade to the crop insurance program—barring receiptof aid from standing disaster programs if crop insur-ance were available in a producer’s county, creatingfederal premium subsidies, and putting the delivery ofcrop insurance in the hands of private companies—participation rates grew very slowly. By 1988, only 25 percent of the eligible farm area was enrolled—farbelow the 50 percent Congress envisioned.

By the end of the 1980s, it was clear to policy-makers that the subsidy levels provided were not suf-ficient to achieve 50 percent participation withouteither making insurance purchases compulsory orincreasing the subsidy level. Congress enacted bothprovisions in 1994, creating a heavily subsidizedcatastrophic risk protection policy that producerswere required to purchase if they participated incommod-ity price support and other farm programs.The 1994 Act also provided additional subsidies for coverage levels greater than 50 percent (buy-uplevels). Responding to producer criticism, however,

Congress repealed the compulsory purchase require-ment in 1996. To further encourage enrollment inhigher coverage levels,1 Congress provided supple-mental premium subsidies in the 1999 and 200 cropyears, and in 2000 increased subsidy levels for mostbuy-up levels. These provisions worked: by 2005,over 80 percent of eligible acres were enrolled, andover 64 percent of these were enrolled at 70 percentcoverage levels or higher.

Increased subsidies and participation have resultedin a four-fold increase in the costs of the currentprogram compared to average outlays in the early1980s. From 1981 to 1993, annual federal crop insur-ance outlays (including delivery charges) averaged$559 million. From 2001 to 2005, they have averagedalmost $3 billion.

Yet, as the last decade has shown, higher participa-tion rates are no guarantee that Congress will refrainfrom passing supplemental legislation to provide dis-aster assistance. Since the passage of the 1980 Act,almost $24 billion has been provided to producers inthe form of disaster assistance. Two years after the passage of the 2000 Act, with participation rates near 80 percent and over 50 percent of acreage insured at coverage levels of 70 percent or higher, Congressprovided $2.1 billion in supplemental disaster assis-tance. Disaster costs from FY 2001 to 2005 totaled $9 billion.

Why is there continued pressure for disaster assis-tance when participation is so high? While theMidwest and Northern Plains states tend to insure athigh levels of coverage, the South and Southwesthave tended to insure at lower coverage levels.Pressures for disaster assistance continue to comefrom regions where participation and coverage arelow. Disaster payments have also tended to be con-centrated, however, in the same Plains states as cropinsurance payments. Texas, North Dakota, Kansas,South Dakota, and Nebraska were the top recipientsof disaster aid and crop insurance indemnities from2001 to 2005, accounting for 38 percent of totalcombined disaster and crop insurance payments. Farfrom substituting for crop insurance, disaster assis-tance outlays have been highly correlated with insur-ance indemnities. This suggests that much of the


76

disaster assistance goes to producers who are alsoreceiving crop insurance indemnity payments.

Cost and Benefits of the Crop Insurance Program

Do the current crop insurance program’s benefitsexceed its costs, both to the taxpayer and to societyat large? While there are different ways to measurethis question, all of them show the answer is “no.”

The most straightforward method is to treat insur-ance payments as a net transfer measured as the dif-ference between what producers receive and whatthey pay through premiums. Over the last 25 years,producers have received an average $2.056 inindemnity payments for every $1 paid in premiums,for a total producer gain of $19.3 billion.

A drawback of the net indemnity measure is that itignores delivery costs and thus probably understatesbenefits. In the absence of government subsidies, pro-ducers would pay a premium equal to the risk pre-mium plus an expense load to cover delivery costsand an appropriate rate of return. One publicly dis-cussed measure is similar to this, adding total indem-nities and delivery costs, divided by premiums paidby producers. According to this measure, from 1981

to 2005, producers received $50.6billion in benefits and paid $18.5billion in premiums, or $2.73 forevery $1 paid in premiums. Thismethod, though, likely overstatesbenefits, since it is unlikely manyproducers would purchase unsub-sidized insurance.

Neither of these methodologiesconsiders the effects of insuranceon crop production or input use. Anumber of studies show that cropinsurance likely does encouragesmall increases in planted acreagein some eligible crops. Most studiesthat examine the question of inputuse conclude that the effects ofcrop insurance on input use are

negative, which suggests that the resulting effect oncrop yields is also negative. Whether this effect is largeenough to offset any positive effect insurance has oncrop acreage is an open question.

When comparing net producer gains to totalcosts, crop insurance is relatively inefficient. From1981 to 2005, producers have received on average$0.60 in benefits for every $1 of government cost.This is due, in part, to the fact that delivery costs forcrop insurance—like other property/casualty lines ofinsurance—tend to be high.

What are the net social costs of the crop insuranceprogram? Over the past five years, the combinedannual costs of crop insurance and disaster assistancehave averaged $4–5 billion. Assuming that the secondorder production effects are negligible and surplusgains to insurance providers are minimal, net producergains can be approximated by multiplying the transferefficiency (60 percent) times the combined costs.Subtracting total costs from producer gains would leavea net social cost of roughly $1.6 billion annually.

Program Alternatives

Given the high costs of federal crop insurance, itsfailure to prevent disaster assistance, its potentially

DOUBLE INDEMNITY

77

FIGURE 1CROP INSURANCE AND DISASTER COSTS, FY 1981–2006

SOURCE: Federal Crop Insurance Corporation; Commodity Credit Corporation.

Mill

ion

Do

llars

Fiscal Year

7000

6000

5000

4000

3000

2000

1000

0

19811987

19951999

20032005

19911985

19891983

19971993

2001

Disaster Crop Insurance

distorting effects on production and input use, andits relative inefficiency at transferring benefits, whatare potential options for policymakers? This briefconsiders five options.

Eliminate the Crop Insurance Program. Under cur-rent baseline projections, elimination of the programcould save taxpayers some $4–5 billion a year.Program elimination would undoubtedly increase theprobability of disaster assistance; however, as noted,disaster assistance has already been provided mostyears anyway. And, despite criticisms of the program,crop insurance remains popular with producers andothers in the sector, such as lenders who are poten-tially adversely affected by crop shortfalls.

Make Crop Insurance Mandatory for ProducersWho Receive Commodity Program Benefits.Making crop insurance purchase mandatory for partic-ipants in commodity programs would likely ensurehigh participation rates, and could reduce incentivesfor Congress to provide ad hoc disaster assistance. Thiscould also help mitigate adverse selection problems.

This has not, however, been popular with pro-ducers, because participation rates are already high,and requiring producers to purchase insurance athigh coverage levels would impose substantial costs,even at subsidized rates. Moreover, many producers,such as fruit and vegetable growers, receive no com-modity program benefits and would hence beexempt from this requirement.

Continue Current Program, but Provide Supple-mental Coverage Based on Area-Yield and RevenueInsurance Plans. . Under this option, producers couldpurchase a supplemental area yield policy to covertheir crop insurance deductibles. This would addressthe “hole in the safety net” criticism that producersoften have to absorb large losses before receiving cropinsurance payments. By basing payments on areayields, additional coverage could be supplied withoutencouraging the attendant moral hazard problems.

Under area yield insurance, producers receiveindemnity payments and pay premiums based oncounty yield outcomes, not individual ones. Because

producer actions can have only limited impact on thecounty yield, moral hazard problems are negligible andadverse selection problems are considerably lessened.

Purchasing high area yield coverage with low levelsof individual yield coverage has been shown to be acost-effective strategy of providing risk protection, par-ticularly for producers in high risk areas where indi-vidual yield coverage is costly. Supplemental coveragecould most likely pay in widespread loss events thataffect the county average yield – the very events that aremost likely to result in ad hoc disaster assistance.

However, because the risk-reducing effectiveness ofarea yield insurance depends largely on the correlationbetween the individual and county yield, supplemen-tal coverage may be less effective for producers in cov-ering losses where individual yields are weaklycorrelated with county yields. Moreover, area yield cov-erage is currently available only for a limited number ofcrops and counties for which the National AgriculturalStatistics Service collects a sufficient amount of yielddata. These crops include corn, soybeans, cotton, grainsorghum, wheat, and peanuts. Extending area yieldcrop coverage to other crops, while possible, wouldraise program costs and potential actuarial issuesbecause of the lack of adequate historical data.

Reduce Program Costs by Reducing DeliveryCosts. Between 35 and 40 percent of every tax dol-lar spent on crop insurance goes toward programdelivery. One option to reduce delivery costs wouldbe for the government to retain all underwriting risksand simply reimburse companies and agents for thesale and service of policies, such as what is currentlydone under the National Flood Insurance program.However, if the current risk-sharing system wereeliminated, companies would likely seek higherA&O reimbursement rates to compensate for theloss of underwriting gains. Similar rates of return canbe found in other property/casualty insurance lines.

Allow Companies to Compete More Directly onRates. One option would be to allow companies tobuild delivery costs into their premium rates, withpremiums subsidies offered in the form of voucherswhich producers could use toward the purchase of


78

insurance policies. Companies could then competewith one another for business based on premiumrates. It is likely that rate competition would result inlower rates in areas such as the Midwest, where under-writing gains have been high. For less profitableregions, supplemental area yield coverage could pro-vide producers a less costly risk management tool andinsurers a more profitable underwriting base.

Conclusion

Despite growth and improvements in the program,federal crop insurance has failed to replace disasterassistance—the stated goal of the 1980, 1994, and2000 Acts. Increased subsidies have encouraged pro-ducers to enroll in the program and to purchasehigher levels of coverage, but even with high partici-pation rates Congress continues to pass supplemen-tal disaster bills.

The marginal costs of adding more acreage orincreasing coverage through increased subsidies arehigh, and costs have increased substantially. In thelate 1970s, the annual costs of standing disaster assis-tance were less than $500 million. By the late 1980s,the combined annual costs of disaster and crop

insurance were $1.5 billion; by the late 1990s theywere almost $2.5 billion; and from 2001 to 2005,they were almost $5 billion. Proposed legislation toestablish standing disaster legislation in addition tocrop insurance would likely push these costs higherstill. Moreover, by indemnifying losses twice—oncethrough insurance and once through disaster—thereare concerns that combined programs will poten-tially overcompensate losses and distort productiondecisions. Congress has many options for reform itcould embrace; failure to change the program willensure that ever-increasing costs and double indem-nity for producers will be with us for years to come.

Notes

This policy brief draws on the author’s longer paper“Double Indemnity: Crop Insurance and the Failure of U.S.Agricultural Disaster Policy,” available at www.aei.org/farmbill. The longer paper includes background supportand citations not included here. Glauber is deputy chiefeconomist at the United States Department of Agriculture.The views expressed here are those of the author and notthose of any institution with which he is affiliated.

1. “Coverage level” refers to the amount of a lost cropwhich the insurance policy will pay for.

DOUBLE INDEMNITY

79

INTERNATIONAL TRADE AND POLICY ISSUES

During the past 20–30 years many high-incomecountries have significantly reformed their agricul-tural policies. Australia and New Zealand haveessentially eliminated farm subsidies. Canada hassignificantly reduced the rate of support to farmers,and has changed the form of support, away frommarket distorting policies to individual incomestabilization accounts. The European Union hasmaintained relatively high rates of support, asmeasured by Producer Subsidy Equivalents, but has radically reformed its forms of support, todecouple payments from production. The differentreforms reflected some common influences, such asevolving market conditions and pressures frominternational trade agreements, but they weremostly done to serve domestic economic interests.In a number of instances reforms were catalyzed by budgetary pressures or other “crises.” In thecountries that have made the greatest progress inreforming their agricultural policies, importantelements included (a) a sector-wide or economy-wide approach to policy, (b) open, public discussionof policy, perhaps through explicit transparencyinstitutions, and (c) adjustment assistance measuresor explicit compensation provided to producersharmed by a change.

Background

Agricultural policy has evolved in different direc-tions and at different speeds in different countries,even among countries that seem very similar inmany ways. Australia and New Zealand are held upas examples of countries that have been able to weantheir farmers off government support. Australia progressively eliminated its agricultural policies over a 30-year period, beginning in the 1970s, as anelement of an economy-wide process of industrialpolicy review and reform. New Zealand made majorchanges suddenly, in the mid-1980s, as an elementof an economy-wide policy reform to address anational financial crisis.

In the late-1980s through the mid-1990s, theUnited States appeared to be on a path to join thesecountries that had largely phased out agriculturalsupport programs. In the Uruguay Round of GATTnegotiations, the United States joined countries likeAustralia and New Zealand in pressing other high-income countries, like those in the EU and Canada, toreduce their production- and trade-distorting agricul-tural subsidies. Consistent with that position, in the1996 Farm Bill the U.S. government substantiallyreformed its policy to decouple the main subsidy pay-ments from agricultural production. Since then, how-ever, the direction of U.S. agricultural policy has

83

Lessons from Agricultural Policy Reform in Other Countries

Julian M. Alston

reversed, with a return to higher rates of supportand more distorting instruments of support and aweakening of the U.S. stance on free trade in agri-cultural products.

In contrast to the United States, during the past tenyears the trend in Europe and Canada has been toincreasingly decouple support from production.Canada has preserved support for supply-managedcommodities, dairy and poultry, but has otherwisesignificantly reduced the rate of support to farmers,and has changed the form of support, away from market distorting policies to individual income stabi-lization accounts. The European Union has main-tained relatively high rates of support, as measured by Producer Subsidy Equivalents, but has radicallyreformed its forms of support over the past 15 yearsto replace market-distorting subsidies with decoupledpayments based on historical production. This EUpolicy reform is very important to the United Statesbecause (a) it is a large change, (b) the countriesinvolved produce and trade a large share of the totalquantity of temperate-zone agricultural products, and(c) it is expected to endure.

Lessons for the United States

Policy change in Australia, New Zealand, Canada,and the EU may offer lessons for the United States. Allfour of these countries have changed their instru-ments of support in the direction of reducing theextent to which support for agriculture is coupled toproduction and trade distorting. The different reformshave reflected some common influences. Evolvingmarket conditions and pressures from internationaltrade negotiations and agreements have been com-mon to all countries, including the United States. For the most part, reforms have been driven more by domestic interest rather than external pressure,although external pressures have helped at times. In a number of instances reforms were catalyzed bybudgetary pressures or other “crises.”

In the countries that have made the greatestprogress in achieving enduring reforms to their agricultural policies, change was facilitated by

complementary policies or institutional arrange-ments, including (a) a sector-wide or economy-wideapproach to policy, (b) open public review processes,perhaps through explicit transparency institutions,and (c) adjustment assistance measures or explicitcompensation provided to producers harmed by a change.

Economy-Wide Approach

Change in agricultural policies in Australia and New Zealand was facilitated by the fact that the policychange processes were not singling out agriculture, orindividual agricultural industries. Rather, they weredeliberate, economy-wide microeconomic reformprocesses, taking an economy-wide perspective on theimplications of industry-specific policy change. Partlyas a result of this context, a greater awareness wasdeveloped of inter-industry connections and how dis-tortions in one part of the economy had consequencesfor other parts, which in turn had implications forpositions taken by interest groups. Agricultural inter-ests became divided because their interests differed,and this division helped make change possible.

In the U.S. context, agricultural economics argu-ments and policy debates typically have not high-lighted the fact that within U.S. agriculture there aremany losers as well as gainers from U.S. farm com-modity programs, particularly comparing producersof non-program crops and program crops. The intra-sectoral distributional consequences are tricky toresolve, given the complex nature of agriculture andthe commodity programs themselves.

Awareness of these issues and demand for infor-mation on distributional aspects are growing, espe-cially from producers of unsubsidized non-programcommodities who are arguing for a redirection of farmprogram support—for instance, specialty crop pro-ducers as reflected in the administration’s Farm Billproposal. What may be most lacking is an independ-ent, credible source, with enough resources to be ableto provide the kinds of information that would bemost useful, and with an interest in doing so—i.e., atransparency institution.


84

Transparency Institutions

Policy change in Australia was facilitated by itsIndustries Assistance Commission (now the Produc-tivity Commission), a transparency institution thatwas established to systematically review industrialpolicies—including farm commodity programs andother, especially border, policies—and recommenddirections for change and processes of change. In theUnited States there is no real counterpart, though anumber of institutions serve as sources of informationabout farm program consequences and thus providesome services related to transparency. Movementtoward economically rational agricultural policycould be facilitated if the United States had a trans-parency institution that played a role like that of theIndustries Assistance Commission in Australia—at least to provide public, independent analysis of the consequences of policies in a context in whichcompeting ideas and information could be aired.

Adjustment Assistance

In both Australia and New Zealand the governmentprovided explicit measures to facilitate farmers toadjust to conditions of financial hardship (whethercreated by reform or the market) and to leave agri-culture and pursue alternative occupations. Theseadjustment assistance measures included debtreconstruction and credit subsidies to facilitateamalgamation of uneconomic farms into more economic-sized units, welfare support for farm fam-ilies, support for retraining, and farm financial andrural family counseling services. The idea that farmor rural policies and programs can be used to facili-tate adjustment of resources and not to subsidizeresources to remain in the sector was developed indetail in the work of D. Gale Johnson and other U.S. agricultural economists in the 1940s. However,the United States has never implemented majorpolicies of this sort, and many U.S. farm programsseem designed to have the opposite effect and retainresources in agriculture in spite of market signalsthat they should leave.

Compensation and Buyouts

Adjustment assistance can be seen as an implicitform of compensation. But in some instances, gov-ernments have introduced alternative policiesexplicitly to compensate agricultural interests forsome of their losses resulting from policy change.Examples include the Canadian elimination of theWGTA transportation subsidies, and some aspectsof the EU reforms. In cases when explicit compen-sation was paid to individuals suffering losses as aresult of the elimination of programs that involvedrights to produce or sell a farm commodity—i.e.,quotas or allotments—the government effectivelybought out the programs. Unlike all the other agri-cultural protectionist policies that were phased outwithout any explicit compensation, in every signifi-cant instance when quotas were eliminated fromAustralian agriculture, explicit compensation waspaid. Similarly, in the United States, in two caseswhere quotas (or allotments) were eliminated,peanuts in 2002 and tobacco in 2004, quota ownerswere compensated. This approach may involve largebudgetary commitments but appears to be one wayto achieve an enduring change.

Decoupled Transfers

Canada and the EU have moved substantially in thedirection of decoupling support from productionand trade, reflecting, among other things, responsesto both budgetary pressures and past or prospectiveinternational trade agreements and trade disputes.In Canada the most distorting subsidies werereduced or eliminated, and whole farm counter-cyclical income support was introduced. In the EU, market-related subsidies have been to a greatextent replaced with decoupled transfers based onhistorical production and transfers, but with somecountercyclical elements provided using borderprotection. In both of these cases, the extent towhich the more-newly introduced policies aredecoupled from production is not clear; nor are the implications of the changes for the distribution

LESSONS FROM AGRICULTURAL POLICY REFORM IN OTHER COUNTRIES

85

of benefits from the policy or the efficiency of the transfers.

Costs of Transfers

Decoupled transfers are likely to have significantlydifferent economic incidence from the kinds of poli-cies that have been phased out in Canada and theEU, and the Canadian income stabilization subsidieswill have different incidence again. As well as chang-ing the distribution of benefits, such changes ininstruments involve changes in distribution of costs.Elements of costs that should not be neglected inthese comparisons are the deadweight losses associ-ated with taxation to finance transfers, and the costsof administration, enforcement, and compliance.The Canadian income stabilization policies in par-ticular seem likely to involve significant costs of this latter type that may well outweigh the socialcosts of resource distortions associated with the un-decoupled policies that they have replaced.Deeper investigation is warranted before we can say

clearly whether the real-world “decoupled” transfersare preferable on efficiency grounds or distributionalgrounds. A related question is whether a shifttoward decoupled transfers is a transitional steptoward elimination of policies altogether or a step inthe reverse direction, enshrining certain transfers ina more permanent fashion. The approach in Austr-alia and New Zealand, of phasing out subsidiesrather than substituting toward more decoupled(and supposedly less-distorting) policies, might becheaper in the short run and may offer more endur-ing reform in the long run.

Note

This policy brief draws on the author’s longer paper“Lessons from Policy Reforms in Other Countries,” avail-able at www.aei.org/farmbill. The longer paper includesbackground support and citations not included here.Alston is a professor in the Department of Agricultural andResource Economics, University of California, Davis. Theviews expressed here are those of the author and not thoseof any institution with which he is affiliated.


86

U.S. farm policy exists within the constraints of theWorld Trade Organization (WTO), and the UruguayRound Agreement on Agriculture (URAA), in partic-ular. The URAA restricts the amount that can bespent by nations on export subsidies and trade-distorting domestic support. In addition, the URAAobliges countries to notify the WTO of their levels ofdomestic support, and these notifications are open toscrutiny and objections by trading partners.

The longer paper finds that WTO rules (thoughnot the rules in regional and bilateral trade agree-ments) have had a noticeable impact on the directionof U.S. domestic farm policy. They have changed theset of instruments used and the level of supportgiven. They have modified the scope for such instru-ments and the way in which farm policy works. Andthey have influenced the political and economicobjectives of farm policy, and the strategic options forachieving those objectives. However, they have onlybeen allowed to operate within the space defined bydomestic legislation.

The paper argues that it is necessary to take abroad look at the impact of trade agreements on U.S. farm policy. To be sustainable, farm policy mustbe compatible with other aspects of U.S. trade andeconomic policy, or at least not be so incompatible asto put achievements in those other areas in jeopardy.If the United States decides to keep price-related

subsidies such as loan deficiency payments andcountercyclical payments, then conflicts between theWTO payment limits will become a frequent occur-rence. If, however, the 2007 Farm Bill continuesdown the road of the 1996 Bill, separating incomepayments from commodity market conditions, thenthe WTO limits will not be a major constrainingfactor in the future.

The Uruguay Round Agreement on Agriculture: A Policy Framework

Although the General Agreement on Tariffs andTrade (GATT) covered agriculture, it provided excep-tions to normal restrictions on trade barriers. First, itallowed quotas and licenses in cases where thedomestic market was being managed by the state,and second, it exempted export subsidies for pri-mary commodities from the constraints placed onmanufactured goods export subsidies. Since agricul-tural sectors largely escaped effective scrutiny anddiscipline by the GATT, by the 1980s world marketswere being distorted by the competitive subsidiza-tion of farm products. The U.S., and other exportingcountries, therefore pushed for more enforceable andcoherent constraints on agricultural policy in theUruguay Round of GATT negotiations (1986–94).

87

The Impact of the WTO and Bilateral Trade Agreements on U.S. Farm Policy

Timothy E. Josling

The URAA, which came into effect in 1995 as partof the agreement creating the WTO, is built aroundthree “pillars”: market access, domestic support, andexport competition. The new market access rulesrequired all forms of quotas and restrictive licenses tobe converted into tariffs, with the tariffs being incre-mentally reduced over time. Export competitionrules prohibited new export subsidies and required areduction in existing subsidies according to volumeand expenditure. Domestic support rules governeddirect payments which went to farmers in addition totransfers from consumers through border policies.These included deficiency payments, direct incomesupplements, administrative price systems, subsidiestied to research and extension, conservation compli-ance payments, and other programs that benefitfarmers directly. These payments were then classifiedaccording to their distortionary effects on trade.“Amber Box” measures (subsidies tied to price or out-put levels) were to be reduced by 20 percent relativeto a base period (1986–90), subject to de minimisamounts excluded from the commitment. “Blue Box”subsidies (those tied to supply control programs)were restricted to payments based on fixed acreageand yield, while “Green Box” subsidies (those unre-lated to price and output) were not constrained.Together, these form the framework for the restric-tions that all WTO members accept, and the ruleswith which U.S. farm policies are to comply if theyare not to risk challenge at the WTO.

Monitoring of compliance is implemented throughthe Agricultural Committee of the WTO, anddepends on “notification” by members. WTO mem-bers accepted an obligation to notify annually thepayments made on domestic support, but have beenincreasingly tardy in exchanging information—impairing the usefulness of the committee as a placefor timely challenges.

U.S. Domestic Policy and the WTO Constraints, 1995–2005

Although the United States had to overhaul its instru-ments of farm policy in the light of the adoption of the

URAA, this did not result in any major reduction inthe level of support given by farm policy, and largelypassed without significantly disrupting stakeholderinterests. It converted many quantitative importrestrictions into tariffs, though domestic prices werenot markedly affected. It introduced limits on exportsubsidies, including those pertaining to coarse grains,wheat, oilseeds, milk, beef, pork, poultry, and occa-sional assistance to fruits and vegetables. However,export subsidies are no longer a major aspect of U.S.policy, with the exception of dairy products, and theeconomic impact of the residual spending on suchsubsidies has not been extensive. Disciplines on U.S.food aid were also largely ineffective, as the URAAmerely clarified existing practices.

The URAA had its greatest impact on U.S. policiesthrough the domestic support restrictions. The URAAintroduced an Aggregate Measure of Support (AMS)as a basis for binding trade-distorting subsidies andfor scheduling future subsidy reductions. The 20 per-cent reduction in domestic support was derived froma base level of $23.9 billion (1986–88) for the UnitedStates, and was reduced by 2000 to $19.1 billion.

Yet, by starting at a particularly high base, andmoving many direct payments from the Blue toGreen Box in the 1996 FAIR Act, U.S. farm programswere not much constrained by the cap on direct sup-port. While unlimited Green Box support was atabout $50 billion from 1995 to 2001, payments inthe constrained categories never exceeded $16.9 bil-lion in that period. However, with an increase in out-lays since 2004 reflecting loan deficiency payments(LDPs) and countercyclical payments (CCPs)—both linked to price levels, and counted as AmberBox subsidies—it became possible to conceive of abreach of the AMS ceiling of $19.1 billion. Addi-tional federal government programs to reduce thefinancial consequences of uncertainties in weather,yields, prices, and global markets, have also beennotified to the WTO as Amber Box. This subsidizedinsurance increases revenue and reduces its variance,and thus encourages planting. However, as with theCCPs, the subsidy element in crop insurance hasbeen covered by the non-product-specific de minimisprovisions.


88

It is currently unclear whether Direct Payments(DPs), introduced in the 2002 Farm Bill as a replace-ment for Production Flexibility Contract payments,belong in the Green Box or the Amber Box. Therationale for including these in the Green Box is thatthey are based on historical yields and acreage andnot related to current price. However, the WTO panelexamining the Brazilian cotton case came to the con-clusion that DPs had a link with production becauseof the restriction on DP recipients from planting fruitsand vegetables. If DPs (which have ranged frombetween $3.9 billion to $5.3 billion annually since2002) were to be included in the Amber Box, theAMS limit would have been breached in recent years.

Future direct payments could possibly exceed theWTO limits, under certain crop and price outcomes,if no reforms are made to make them fully consistentwith the Green Box definition. This consistencywould be enhanced by moving to payments that werenot based on commodity prices. Such a move wouldalso make the farm policy more responsive to newneeds, such as conservation and rural development,and less focused on the production of a small set ofcommodities. Although the threat of sanctions byother nations may be unlikely to coerce the UnitedStates into conformity with WTO rules, the desire topreserve the integrity of a system of trade rules thatbenefits the United States (and its agricultural sector)should do so. WTO constraints thus lead U.S. farmpolicy in a direction in which it should go.

Bilateral and Regional Trade Agreements

Regional and bilateral trade agreements have gener-ally had less impact on U.S. domestic farm policiesthan have the provisions of the WTO, since prefer-ential trade pacts usually avoid obligations thatdirectly impinge on domestic farm programs.Nonetheless, the desire for gains from trade in agri-culture has produced agreements to increase marketaccess. By increasing the amount of farm goodsentering the U.S. market as a result of tariff cuts andquota increases, these agreements do, however, influ-ence the viability of domestic support programs. At

the same time, bilateral agreements are careful tostate explicitly that the parties retain their rightsunder the WTO, and make clear that they aredesigned to be in conformity with WTO rules onFTAs and customs unions.

Having long been a champion of the multilateraltrading system and of nondiscrimination, for a com-bination of geopolitical and economic reasons, theUnited States has now become an active supporteror regional and bilateral trade agreements as a com-plement to its WTO commitments. A Free TradeArea (FTA) with Israel in 1985, as an expression ofpolitical and economic support, was followed byone with Canada in 1986, which was expanded toinclude Mexico in the North American Free TradeAgreement (NAFTA) in 1994. Since 2002, theUnited States has began to negotiate more bilateralagreements, reflecting a policy of “competitive liber-alization,” with market access being liberalizedbetween the U.S. and a number of geopolitical alliesand geographic neighbors.

All the U.S. FTAs have provisions for tariff reduc-tions that affect food and agricultural goods.However, the agreements control trade in a range ofproducts considered politically sensitive in one orboth partners. For the United States, these sensitivi-ties include sugar, citrus fruits, peanuts, and dairyproducts. For the partners, the list often includesproducts such as corn and beans along with rice.

During the NAFTA negotiations, the United Stateslooked to gain improved market access in Mexico forgrains and livestock exports, while Mexico wanted tobe able to increase the sales of fruits and vegetablesin the United States. For trade with Mexico, a boldermove to liberalize trade was planned—with no per-manent exemptions after a 15-year transition period.Since this trade is small relative to the U.S. domesticmarket, the impact of NAFTA on the United Stateshas been minor. However, the integration of marketshas heightened awareness of differences in marketinginstitutions, and the activities of the Canadian WheatBoard have become a source of tension associatedwith NAFTA.

The Central American Free Trade Agreement(CAFTA), which includes the United States, Costa

THE IMPACT OF THE WTO AND BILATERAL TRADE AGREEMENTS ON U.S. FARM POLICY

89

Rica, El Salvador, Guatemala, Honduras, and Nica-ragua, took effect in 2006, and was intended to helpfoster economic growth in the Central Americanregion by reducing barriers to trade and invest-ment. The agreement does not regulate domesticsubsidies, but increases market access for agricul-tural goods in both directions. While agriculturaltrade barriers in the Central American nations arehigher than those for manufactured goods, CAFTAphases in provisions for increased market access.Nonetheless, the impact of market access in theUnited States will likely be limited, since mostCAFTA nations have already enjoyed permanentduty-free access from previous agreements. Indeed,approximately 99 percent of CAFTA exportsalready enter the U.S. market duty-free, with excep-tions largely in the cases of sugar, dairy, cotton,meats, and peanuts.

Future Impacts of Trade Agreements onDomestic Farm Policy

At present, the prospects for further multilateral liber-alization, through the Doha Round of world tradetalks, look somewhat remote. But if agreement wereto be reached on a significant cut in bound tariff rates,a sharp reduction of trade-distorting domestic sup-port payments, and the elimination of export assis-tance through food aid and credit guarantees, theDoha Round would have a more profound impact onU.S. farm policy than did the Uruguay Round.

While the changes agreed in a Doha Round pack-age will have to be phased in over a period of years,so long as its main parameters are known before the2007 Farm Bill is drafted, flexibility to accommodateits demands could be built into the legislation.Should the United States face tight limitations on theamount of Amber Box support it could provide, it islikely that the price support programs for these com-modities would have to be modified. Price supportlevels would have to be reduced in order to keepwithin the total AMS limit when market prices arelow, and CCPs might have to be subject to limitationsto stay within the Blue Box limit.

If the Doha Round results in increased marketaccess, the ability of the United States to supportprices for certain commodities by controlling imports(most notably dairy and sugar) might be constrainedregardless of any separately agreed limits on subsidies.Should export subsidies be ended under a successfulDoha Round of world trade negotiations, the subsi-dies in the Dairy Export Enhancement Programwould need to be phased out. The subsidy compo-nents of “in kind” credit guarantees would also needto be removed. If the WTO talks were to mandate thatfood aid be limited to cash grants, the basis for U.S.food aid policy would be profoundly undermined.U.S. farm exports, however, would benefit from areduction in the implicit subsidy given by the single-desk seller status of the Canadian Wheat Board—although it is possible that the Canadian governmentwill in any case end the CWB’s export monopolybefore long.

Although offering the prospect of expandingmarkets with Latin America, enthusiasm in the U.S.for a Free Trade Area of the Americas (FTAA) haswaned. Economies in that region have been success-ful agricultural exporters in the past decade, andpose a challenge to the competitiveness of U.S. pro-duction. The reaction of U.S. policymakers willtherefore be conditioned by the extent to which theUnited States can expect to compete in such marketsas citrus, oilseeds, beef, rice, sugar, and corn. On theother hand, broader competitive pressures for tradeliberalization are likely to grow. Markets withinSouth America are being consolidated through MERCOSUR, and since the EU could soon concludean agreement with MERCOSUR that would eventu-ally include preferential access for farm products,U.S. exporters in all sectors of the economy wouldface a disadvantage if the FTAA did not materialize.

Conclusion

The impact of trade agreements on U.S. domestic farmpolicy is both elusive and pervasive. On the one hand,farm policy instruments are staunchly defended intrade negotiations and are rarely “put on the table.”


90

On the other hand, trade agreements have a corrosiveeffect on the ability of domestic policy to manage mar-kets to the advantage of domestic producers. Indeed,the Blair House pact with the EU (at which the agri-cultural modalities for the Uruguay Round wereagreed) was intended in large part to ensure that tradeconstraints did not force domestic policy changes.

Sometimes the full impact of a trade agreement isnot felt until market conditions change. In the case ofthe URAA, the restraints on U.S. policy began to benoticed when prices receded from their high levels in1996 and 1997. The impact on domestic subsidypolicies comes when future legislation is crafted toaccommodate the URAA. The effects of litigation atthe WTO can be more immediate, but may again bereflected in policy revisions in future years.

The underlying trends in farm policy are closelylinked with the restraints built into the WTO URAA.The United States was insistent that domestic policiesof all countries be restrained by the rules negotiatedin the Uruguay Round. To the extent that Europe’sCommon Agricultural Policy has changed since1992, the URAA can be attributed some of the credit.But, the United States farm policy has also increas-ingly become the target at the WTO of developing

country exporters, such as Brazil, which consideredtheir own agricultural growth to be hindered.

WTO rules have modified the scope of domesticpolicy instruments and changed the strategic optionsfor achieving political and economic objectives,albeit within the space defined by domestic legisla-tion. If the 2007 Farm Bill continues down the roadof the 1996 Bill, by separating income paymentsfrom commodity markets, then the WTO limits willnot be a major factor in the future. If the UnitedStates decides to keep price-related subsidies, such asloan deficiency payments and countercyclical pay-ments, then conflicts between the WTO paymentlimits will become a more frequent occurrence.

Note

This policy brief draws on the author’s longer paper “TheImpact of the WTO and Bilateral Trade Agreements onU.S. Farm Policy,” available at www.aei.org/farmbill. Thelonger paper includes background support and citationsnot included here. Josling is professor emeritus, FoodResearch Institute, and senior fellow, Freeman-SpogliInstitute for International Studies, Stanford University.The views expressed here are those of the author and notthose of any institution with which he is affiliated.

THE IMPACT OF THE WTO AND BILATERAL TRADE AGREEMENTS ON U.S. FARM POLICY

91

Secretary of Agriculture Mike Johans, along with keymembers of Congress, rates compliance with WorldTrade Organization (WTO) commitments as amongthe most important drivers for change in the 2007Farm Bill. This is a new development. In earlier farmbill debates, international obligations were either notsignificant, or, in 1996 and 2002, not expected tohave any serious implications for farm subsidy pro-grams. The successful challenge of the U.S. uplandcotton program by Brazil in a wide-ranging WTOdispute disrupted the complacency with whichmany viewed the U.S. ability to satisfy its WTO obli-gations. Other challenges to U.S. farm programs arenow routinely discussed and evaluated, and Canada,together with a dozen other WTO members, hasfiled a challenge to the U.S. corn program and theU.S. commitments regarding its aggregate measureof farm subsidies.

D. Gale Johnson noted the “inconsistency”between farm subsidies and U.S. advocacy of freeinternational trade as a source of economic growthand secure international relations at the foundation of the GATT in the late 1940s. But, before the adventof the WTO in 1994 this tension was rhetorical anddid not involve international agreements or explicitobligations. This policy brief explores the tensionsbetween U.S. farm subsidy programs and U.S. obliga-tions under WTO agreements. It shows that U.S. farm

subsidies are clearly in conflict with the traditionalthrust of U.S. trade policy, including agricultural tradepolicy. Furthermore, U.S. farm subsidy programsremain vulnerable to successful challenge as inconsis-tent with the letter and spirit of WTO agreements ofwhich the United States was a leading advocate.

The most direct way to remedy this conflict wouldbe to remove subsidies tied directly or indirectly tocommodity production. However, if some kinds ofaid to farm producers or rural areas were justified onbroad social welfare grounds, such programs could bereadily designed to be consistent with obligations notto distort global commodity markets or otherwiseviolate international commitments. This generallyimplies focusing support on public goods such asenvironmental protection or rural poverty in waysthat minimize linkage to commodity production.

WTO Agreements, Compliance, and Dispute Settlement

Members (countries or groups of countries such asthe European Union) join the WTO if they accept itsobjective, which is to open markets and facilitatetrade. To meet this objective, the WTO plays tworoles. First, it facilitates the negotiation of multilateraltrade agreements that lower trade barriers and set

93

The Farm Bill and WTO Compliance

Daniel A. Sumner

rules that reduce other policy impediments to tradebetween nations. Second, the WTO administers themultilateral trade agreements and facilitates compli-ance through settlement of disputes. Under rulesadopted in 1994, disputes at the WTO are nowresolved, even given time for a full appeal process,more rapidly than in U.S. courts, for example.

But, because its members are sovereign govern-ments, the WTO cannot compel compliance with itsrulings. At the end of the process, countries maychoose to suffer retaliation from other membersrather than change their own policies. But, moreimportant than retaliation, lack of compliance withWTO rulings may have large costs in terms of cred-ibility with trading and negotiating partners. Thislatter cost is a particular problem for the UnitedStates, which has long urged fealty to the rule of lawin trade relations.

The WTO Agriculture Agreement of 1994 was thefirst to include binding commitments on agriculturalsubsidy programs. In the agriculture agreement mem-bers pledged to list their government programs thatsubsidized agricultural production and to ensure thatan “aggregate measure” of this support did not exceedthe established limit set for each year. This process ofdetermining which programs to include and howthey affect the aggregate measure has been complexand contentious, and U.S. farm subsidy programshave been challenged as being inconsistent with thespirit and legal obligations of the WTO agreement.

Even more important than the specific commit-ments of the WTO Agricultural Agreement, in 1994WTO members also agreed to bring agricultural pro-grams thoroughly under WTO rules relating to subsi-dies more broadly. That is, rather than being generallyoutside the general WTO provisions, since 1994 agri-culture has been subject to these provisions exceptwhere specifically provided for. The general WTOprohibition on export subsidies and subsidies thatprovide special privileges for local goods over imports(once import tariffs have been paid) now also apply toagricultural goods. Furthermore, the general WTOprovisions against using subsidies that have adverseeffects on the economic interests of another membermean that farm subsidies that distort market prices to

a degree that seriously harms the industry in anothermember country are subject to challenge, even ifthese subsidies are in line with the rules of the agri-culture agreement.

The Cotton Case and the New WTO Corn Case

In 2004 and 2005 the WTO dispute settlement paneland appellate body found, in a dispute brought byBrazil, that the U.S. cotton program (and in the caseof export credit guarantees, programs for many othercommodities) included prohibited export subsidiesand local content subsidies. They also found that thecotton program caused price suppression on theworld market for cotton that was significant enoughto cause serious prejudice to the interests of Brazil.Furthermore, in the process of its decisions, the WTOpanel and appellate body found that the “so-called”decoupled payment programs introduced by theUnited States with the 1996 Farm Bill constitutedsupport for cotton because they limited the crops thatfarmers could grow while maintaining eligibility forthe payments. After missing initial deadlines, inresponse to these rulings the United States madesome adjustments to its export credit guarantee pro-gram and eliminated a special “step 2” feature of thecotton program that subsidized buyers of U.S. cotton.The United States did not remove or adjust other pro-grams, and Brazil has gone back to the WTO claim-ing that the United States has failed to fullyimplement the original rulings. That implementationcase is underway, with initial rulings expected in June2007, with the potential for appeals after that.

After much speculation about cases being pre-pared against U.S. programs for wheat, rice, corn,dairy, and other crops, in January 2007, Canada,joined by eight other nations, requested formal WTOconsultations with the United States as the first step ininitiating a case against the U.S. corn subsidy programthat seems modeled after the cotton case. In addition,in the same WTO consultation request Canadaclaimed that the United States has been misreportingits “so-called” decoupled programs and may have


94

exceeded its allowed limits on the aggregate measureof support.

Implications for the 2007 Farm Bill

The United States was the leader in bringing agricul-ture under the full WTO umbrella and strengtheningthe dispute resolution process. These accomplish-ments promise to yield important economic benefitsfor the United States and U.S. agriculture for decadesto come. But seeming U.S. violations of its obligationsand apparent unwillingness to play by the rules it cre-ated undermine the credibility of the WTO and of theUnited States. To be sure, not every farm program vio-lates the text of the WTO agreements, and the UnitedStates should defend its policies against unwarrantedlegal challenge. Nonetheless, it is also important torecognize when obvious contradictions between rhet-oric and action in farm trade policy harm the strongU.S. agricultural interests in opening markets andreducing subsidies globally. There are times when full,immediate, and forthcoming compliance with WTOdispute settlement rulings will reap significant, broadstrategic gains. Furthermore, the U.S. should considerunilaterally eliminating or altering its key agriculturesupport programs to preclude future WTO chal-lenges, and secure the benefits of free trade for itsfarmers. The 2007 Farm Bill can be the vehicle toachieve all these ends.

Given high projected commodity prices, 2007 isan ideal year to deal with long-term inconsistencies ofU.S. farm programs with WTO agreements. Elimi-nating the marketing loan and countercyclical pro-grams for grains, oilseeds, and cotton and the MILCprogram for dairy would remove the largest impedi-ments to meeting WTO obligations. These programsclearly stimulate production and affect markets,arguably in violation of the WTO. Short of programelimination, reducing loan rates and target pricesenough that these programs are unlikely to affect pro-duction would reduce the likelihood of successfulchallenge against these programs. But, the underlyingrationale of these programs, to offset low marketprices for producers, is fundamentally at odds with

the principles that the United States has urged on theWTO.

The direct payment programs for grains, oilseeds,and cotton also ought to be eliminated or significantlyreformed. Like the countercyclical programs, theseprograms allow considerable planting flexibility anddo not tie payments to commodity prices. However,they do restrict what can be grown on program landand they provide substantial funds to producers andlandlords for use on farm investments and as collateralfor farm loans. Further, if growers expect that the farm-specific land use and yield data on which they arebased may be updated, the program provides incen-tives to maintain area and enhance yields. Thus theseprograms too are not immune from WTO challenge.

The cotton panel and appellate body rulings sug-gest that WTO rules count direct payments as part oftrade distorting support. A simple way to meet thecurrent WTO ruling is to remove the planting restric-tions for payment base land. However, that would notrespond to the other production incentives imbeddedwithin the direct payment program. Here again, elimi-nating the program is the simplest way to assure com-pliance with the letter and spirit of WTO obligations.That would also remove the objections raised by tra-ditional growers of the “restricted crops” that theywould face unfair competition and lower marketprices if growers receiving direct payment and coun-tercyclical payments were allowed to plant their cropson payment-eligible land.

In contrast, payments to farmers for environmen-tal objectives pose many fewer WTO concerns. Thegeneral rule is that such programs only face WTOlimits if they also stimulate production. So, for exam-ple, programs that pay farmers on a per acre basis forproduction of a specific crop so long as they meetminimal environmental standards would rightly beseen to stimulate production of the crop. Such pro-grams are not prohibited, but they would contributeto distorting international markets and to the aggre-gate measure of support. The WTO rules for consid-ering environmental programs as only minimallytrade distorting allow for support that pays for envi-ronmental benefits supplied by the farm withoutovercompensating the grower.

THE FARM BILL AND WTO COMPLIANCE

95


96

Conclusion

Compliance with WTO obligations constrains U.S.farm subsidies, but these obligations are consistentwith the interest of the United States for an agri-culture policy that responds more to market sig-nals rather than government subsidy. Using the2007 Farm Bill to move vigorously in this directionwould avoid international conflict and, indeed,exhibit the economic leadership that the rest of theworld should expect from the United States.

Note

This policy brief draws on the author’s longer paper “U.S.Farm Policy and WTO Compliance” available at www.aei.org/farmbill. The longer paper includes background supportand citations not included here. Sumner is director of theUniversity of California Agricultural Issues Center and FrankH. Buck Jr. Professor in the Department of Agricultural andResource Economics, University of California, Davis. He is co-director of the AEI agricultural policy project. The authorthanks David Orden and other AEI project participants forsuggestions. The views expressed here are those of the authorand not those of any institution with which he is affiliated.

For over fifty years, international food assistanceprograms—commonly known as “food aid”—haveenjoyed broad political support from agribusiness,shipping, foreign policy, and charitable organizations.Yet these policies have become a point of controversyin the World Trade Organization (WTO) Doha Roundnegotiations, many perceive the programs to bewasteful, and calls for significant reforms havebecome widespread.

This paper argues that current programs are builton outdated and often flawed assessments aboutwhat food aid can accomplish, and should be remod-eled around objectives for which food aid is demon-strably effective: rapid, invaluable humanitarianresponse to food emergencies in low-income coun-tries, and, to a lesser degree, longer-term economicdevelopment in low-income countries. Many restric-tions on and requirements of U.S. food aid policyshould be abolished or reformed. Primary amongthese are the “buy-U.S.” requirements that food dis-tributed as aid be grown, bagged, and processedentirely by U.S. suppliers, as well as mandates thatrequire shipping food aid in U.S. carriers. Becausetimely response is essential to meeting emergencyneeds, a revised program would include cash forlocal and regional purchases, as well as exemptionsfrom any remaining requirements that delay and addcost to emergency response.

Once food aid programs concentrate solely onhumanitarian and development objectives, it maymake sense to reorganize the remaining programsinto two programs: one for emergency response, theother for non-emergency, development program-ming. This would achieve focus and save on unnec-essary bureaucratic expense, and would also enablethe United States to feed far more hungry people pertaxpayer dollar.

U.S. Food Aid Program Background

In the period before the 1990 Farm Bill, generousfarm price support programs generated massivegovernment-held stocks of grain, and substantial tar-iff and non-tariff barriers in farm products restrictedglobal commodity trade. The original food aid law—Public Law 480 (PL 480), passed in 1954—allowedthe government to dispose of commodity surplusesin markets that were separated from the U.S. market,and so would not drive down domestic prices, whileattempting to serve commercial, geopolitical, andhumanitarian goals.

The U.S. government currently operates sevendistinct food aid programs, two of which are dor-mant (see table 1 on the following page). The over-whelming majority of U.S. food aid has always fallen

97

U.S. International Food Assistance Programs: Issues and Options for the 2007 Farm Bill

Christopher B. Barrett

under the three titles of PL 480, which are fundedthrough annual and supplemental appropriations bythe Congress and authorized in regular Farm Bills.The Food for Progress, McGovern-Dole InternationalFood for Education and Child Nutrition Program,and Bill Emerson Humanitarian Trust are all likewiseauthorized by Farm Bills and covered by annualagriculture appropriations.

How Food Aid Has Changed

Concerns about contemporary U.S. food aid arise inresponse to changes that have occurred in the back-ground conditions that originally informed thedesign of food aid programs over fifty years ago. Thefive major changes concern:

1. Disposal of government grain surpluses and useof price supports to help U.S. farmers;

2. Use of food aid to create new commercial exportmarkets;

3. The dramatic shift from program to emergencyfood aid;

4. Reduced cash resources for food security pro-gramming in developing countries; and

5. Changed food aid policies among operationalagencies and other donors.

Helping U.S. Farmers? For many years, food aidprovided a means to dispose of surplus commoditystocks at a low taxpayer cost. Now that price supportprograms are no longer a primary source of revenuefor American farmers, government grain stocks aregone. The government now purchases food specifi-cally for distribution as aid, but pays above openmarket prices, as restrictions on specific bagging andprocessing services force it to procure commoditiesfrom a limited competitive tender.

The widespread belief that food aid benefitsAmerican farmers financially is now mistaken. Little ofthe food purchased comes from average farmers;instead, it is mostly sourced from larger-scale agribusi-ness. At the same time, the $654 million spent on foodaid commodities in 2005 is merely a drop in the oceanof the nearly $1 trillion U.S. food economy, and pro-vides only about 1 percent of net farm income in theUnited States. Food aid programs are simply too smallto move U.S. market prices for food.

Creating New Export Markets? In 1954, many foodaid backers believed that food aid might help developexport markets for U.S. agriculture by establishingnew distribution channels. Yet U.S. food aid has actu-ally displaced commercial agricultural exports to thetargeted countries. Once one controls for other factorsat play, food aid does not appear to build up a tasteamong foreign consumers for U.S food exports. Nordoes it stimulate sufficient income growth in recipient

98

TABLE 1U.S. FOOD AID PROGRAMS

Bill Emerson McGovern-PL 480 PL 480 PL 480 Food for Section Humanitarian Dole

Program Title I Title II Title IIIa Progress 416 (b)a Trust IFECNb

Year begun 1954 1954 1954 1985 1949 1980 2003

Managing agency USDA USAID USAID USDA USDA USDA USDA

2005 actual funding 44 1,413 0 198 76 377 87($ million)

2006 estimated 32 1,138 0 205 0 Not reported 99($ million)

SOURCE: White House budget proposal for FY 2007.NOTES: a. Dormant program.

b. International Food for Education and Child Nutrition.


markets to induce enough expansion of demand forU.S products to compensate for the commercialexports displaced by the free food aid shipments.

The Shift from Program to Emergency Food Aid.The nature of food aid has changed significantly.Over the past forty years, there has been a dramaticshift away from government-to-government conces-sional sales to food aid donations through NGOs,cooperatives, and the World Food Program. U.S.food aid now largely responds to emergencies,which, in the modern era, are less remote and affecta much larger population. Title I funded food aid(sales to foreign governments) now comprises only6.6 percent of the total U.S. food aid program, downfrom 62.6 percent in 1980. Title II funded food aid(donated food for emergencies) constituted 77.7 per-cent of total U.S. food aid contributions in FY 2005,more than double its 1980 share.

Reduced Funding for Food Security Drives“Monetization.” At the same time, non-food aidfunds for food security projects in developing coun-tries have decreased sharply. This has induced NGOsand cooperatives working to end hunger andinvolved in food aid to adopt more controversialpractices, notably “monetization,” wherein they sellthe food aid they receive in local markets to raisecash for food security programming. As a result,more than half of Title II non-emergency resourceshave been monetized each year since 1999.

Increased Effectiveness of Food Aid Practices.Food aid practices have changed significantly in thelast twenty years, becoming much more effective inthe process. With improvements in early warningsystems, emergency needs assessment methods, andsupply chain management systems, the effectivenessof emergency food aid has increased over the past twenty years, and has often helped to prevent wide-spread famine. One manifestation of this is theincreased attention paid to the nutritional content ofcommodities distributed. Since fortified foods are notcommonly available through commercial channels inmany poor countries, micronutrient deficiencies are

widespread, and dietary diversity is generally limitedamong low-income households. Rations designed tomeet these needs can work wonders.

Donors have responded to these improved prac-tices by channeling their aid increasingly to theWorld Food Program and other NGOs that focus on assessed recipient need. However, these success-ful groups have increasingly purchased food in recip-ient countries’ local or regional markets. The UnitedStates remains the only major donor that does notpermit its funds to be used for local and regionalfood purchases. Aside from U.S. food aid, over 60 percent of global food aid flows are procured out-side the donor country, in developing countries, astriking shift in just the past several years.

What Remains the Same: Politically DrivenRestrictions on Food Aid

Despite these changes, U.S. food aid programs havecontinued to labor under politically driven restric-tions that limit programs’ flexibility to respondappropriately to rapidly growing emergency needs.These restrictions also unnecessarily drive up costs.In FY 2005, USAID purchased $654 million in food,but spent nearly $1 billion in transport, storage, andadministrative costs: only 40 percent of expenditureswent for food. By contrast, in FY 2003, 68 percent ofCanada’s food aid budget was spent on food.

A significant share of these costs are attributable tothe many restrictions placed on U.S. food aid, partic-ularly those with respect to shipping, bagging, andprocessing. Other restrictions, such as the require-ment that the food provided through U.S. food aidprograms must be grown in the United States, alsohurt the ability to respond quickly to emergencies.In emergency medicine, the “Golden Hour” is thefirst 60 minutes after an accident or the onset ofacute illness, the window during which the chancesof saving a patient are greatest. The internationalhumanitarian community has generally internalizedthe principle of the Golden Hour: Rapid response isessential. Food aid procurement restrictionsnonetheless stand in the way. The average delay in

U.S. INTERNATIONAL FOOD ASSISTANCE PROGRAMS

99

delivering emergency food aid is nearly five monthsbecause of current rules requiring all U.S. food aid tobe grown and procured in and shipped from theUnited States. The GAO found that, had the U.S.authorized regional food purchases by the WorldFood Program, it could have saved 120 days in thedelivery time for food aid rations to Afghanistan.Delays cost money and lives.

Shipping restrictions on food aid are especiallycontentious. Partly the result of the desire to main-tain a viable U.S. merchant marine fleet that can becalled upon in time of war, the 1985 Farm Bill mandated that at least 75 percent of the gross ton-nage of U.S. food aid be shipped on privately owned,registered U.S.-flag commercial vessels. Another pro-vision of the 1985 Bill mandates that at least 75 per-cent of the non-emergency minimum tonnage befortified, bagged or processed, and that 50 percent ofnon-emergency grain shipments be bagged.

But these U.S.-flagged ships need not be U.S.-owned. Indeed, foreign-owned lines operating U.S.-flagged ships carried 58 percent of USAID food aidby weight in FY 2005. Moreover, dry bulk ships areof negligible value to the U.S. military, so the GAOhas repeatedly concluded that cargo preference onfood aid shipments does not advance the objectivesfor which they are designed. It estimates that cargopreference provisions inflate freight costs by 69 per-cent without advancing national security interests inmilitary sealift capacity.

In 1996, Congress created a new Maritime Secur-ity Program (MSP), providing subsidies of $2.6 mil-lion per ship a year, to give the Pentagon the legalright to use ships and crews for military operationswhen necessary. There thus now exists a distinct pro-gram explicitly for the purpose for which cargo pref-erence was intended. Shipowners can now legallydouble dip, collecting MSP payments while alsoenjoying cargo preference mark-ups on many foodaid shipments. This is wasteful.

Food aid volumes are a tiny share of global ship-ments, and so it does little to keep the U.S. merchantmarine afloat. Yet, a few ports depend heavily on foodaid shipments and benefit from cargo provisions thatdistort shipping patterns. For example, food aid

makes up one-third of the tonnage shipped from theGulf of Mexico port at Lake Charles, Louisiana.Analysis of USAID data shows that 45.1 percent of allfood aid in 2005 was shipped through Houston-areaports, providing many times those ports’ share ofcommercial exports of the same commodities. Themost ridiculous example of distorted logistics arisesfrom the 1986 Farm Bill stipulation that Great Lakesports should retain their 1984 shares of food aid car-goes. Because no cargo preference-eligible carriersserve Great Lakes ports today, these restrictions arenow met entirely by overland cargo transport to facil-ities in the Great Lakes region where they are movedto a different mode of transport (truck or rail) to moveto another U.S. seaport for ocean shipment.

Recommendations for Reform

Historically, food aid has been proven effective atmeeting only two objectives: food surplus disposaland providing humanitarian assistance. Since sur-plus disposal is unlikely to be permissible under anyfuture WTO agreement, a reformed food aid pro-gram ought to focus on how best to provide human-itarian assistance.

The U.S. food aid budget has shrunk by nearlyhalf in real terms since 1980. Since increasinglyscarce resources make it more important to maxi-mize value per dollar, it would be wise to removerestrictions on emergency assistance, so that aid canfollow best practices in situations where food is mostneeded. Congress should also consider increasingannual food aid appropriations to $2 billion a year,an increase from the $1.6 to $2.0 billion range thathas prevailed since 2002 once supplemental appro-priations are taken into account.

Three other reforms would increase the amount offood which NGOs can provide to recipients: repeal therequirement that 75 percent of non-emergency foodaid be fortified, bagged, or processed; eliminate thecap on Section 202(e) cash made available to opera-tional agencies to cover administrative, transport, stor-age, and handling costs; and reform the cargopreferences that drive up shipping costs.


100

The argument for eliminating the value-addedminimum is simple: this mandate is rarely if ever metand it has no basis in advancing food security objec-tives. Repeal of the Section 202(e) cash spending capwill also be necessary if there are limits placed on theability to monetize non-emergency food aid. But themost contentious reform will be over cargo prefer-ence in ocean freight.

Congress should direct agencies to streamlinecommodity procurement and freight contracting byinsisting on using prevailing best practices in com-mercial supply logistics, and harmonize the listing ofvessels eligible for cargo preference with that forwhich the Maritime Administration (MARAD) partlyreimburses USAID and USDA. Ending the GreatLakes port provisions could likewise eliminateunnecessary freight costs that do not improve theefficiency or effectiveness of food aid programs.

Another important avenue for reform is theprospective use of cash to make local and regionalfood purchases to respond to food emergencies indeveloping countries. By law, the food providedthrough U.S. food aid programs must be grown in,processed in, and shipped from the United States.This “tying” of food aid to domestic procurement setsthe U.S. apart from other donors, which have partlyor wholly abandoned the practice.

The economic logic supporting local and regionalpurchases is impeccable: quicker, lower-cost deliverythat is less likely to disrupt recipient country markets.This is not uniformly the case: some markets are toothin to absorb a significant increase in commercialfood demand without driving up prices, and in otherplaces quality control and transport capacity concernslimit the ability to make these purchases. But mostoften, local and regional purchases are cheaper andquicker than shipments of U.S. food on U.S. shipshalfway around the world. A detailed study of globalfood aid transactions recently found that local foodaid procurement was 66 percent less expensive thanshipments directly from donor countries.

Cost savings notwithstanding, the Golden Hourprinciple is perhaps the most compelling argument forlocal and regional purchases. Rapid response saveslives, limits ill health, keeps children in school, and

prevents precipitous and sometimes irreversible assetloss by desperately poor people forced into distress saleof their land, livestock, or even persons (via the abhor-rent practice of debt peonage). Further, when food aidarrives late, it fails to deliver promised humanitarianbenefits in full and can often prove disruptive to localcommercial farming and marketing systems.

USAID has made notable efforts to improve timelyresponse through creative rerouting of cargo alreadyat sea and, most notably, through pre-positioningwarehouses in Dubai and Djibouti, nearer to someregions (such as the Horn of Africa) where emergen-cies frequently arise. While this can, in principle, cutresponse time to two to four weeks, from the nearlyfive months averaged by U.S. emergency food aid,the volume and variety of commodities available aresharply limited, and maintaining this capacity addsto overall program costs because of facility limita-tions. Thus, the timeliness and efficiency of emer-gency response are sharply limited under the currentconfines of “buy America” provisions.

Since the 1990 Farm Bill relaxed rules on sales offood aid for local currencies in developing countries,“monetization” has led to the widespread transfor-mation of non-emergency food aid out of direct dis-tribution and into open market sales to raise cashresources for development projects. Many observersconsider monetization a terribly inefficient way touse taxpayer dollars, while others fear that it distortsmarkets, displaces imports, and could therebyimperil trade agreements.

Although it is tremendously wasteful for the gov-ernment to buy food in the United States, ship itabroad to an NGO or cooperative, only to sell it offagain in a nation where food is cheaper, NGOsdefend monetization on the grounds that it providescash resources that are not available any other wayfor overseas development programs. To a certaindegree, therefore, debates about monetization are amisplaced struggle over the appropriate funding lev-els for development assistance. In the event of aWTO agreement, however, new trade rules mayforce explicit limits on monetization rates, since sev-eral WTO member states perceive monetization to bea de facto export subsidy.

U.S. INTERNATIONAL FOOD ASSISTANCE PROGRAMS

101

Conclusion

Reform of U.S. food aid programs is long overdue.The U.S. government no longer holds food sur-pluses, and alternative, better ways exist to assist themerchant marine than through cargo preference onfood aid shipments. In emergency situations, wherefood is extremely scarce and markets do not functionfreely, food aid often proves more valuable than itsdelivered cost. But current programs are built on out-dated and flawed assessments about what food aidcan accomplish. U.S. food aid programs can andshould be refocused on effective response to foodavailability emergencies in low-income countries.

As recognition of the changed landscape for foodaid spreads, support for a range of potential reformproposals will continue to grow. Given the smallshare of farm and shipping revenues that comes fromfood aid programs, the cost of reforms would be neg-ligible, while the improvement in response tohumanitarian disasters and chronic food insecuritywould be considerable.

A clear demarcation of emergency from non-emergency food aid programs would provide greaterstability for essential investments necessary to head

off future food emergencies, and improve the focuson the most efficient and effective uses of scarceresources. This would streamline the administrationof food aid programs, reduce the transaction costs forthose implementing programs supported by U.S.food aid, and facilitate further reforms in commodityprocurement and freight contracting by reducing thenumber of programs involved.

The complex web of procurement restrictions andbureaucratic duplication currently makes rapidresponse to humanitarian crises unnecessarily difficultand U.S food aid programs far more expensive perbeneficiary than other donors’ food aid programs. Thetime is ripe for food aid reform in Washington.

Note

This policy brief draws on the author’s longer paper“U.S. International Food Assistance Programs: Issues and Options for the 2007 Farm Bill,” available atwww.aei.org/farmbill. The longer paper includes back-ground supporting data and citations not included here.Barrett is International Professor of Applied Economicsand Management, Cornell University. The viewsexpressed here are those of the author and not those ofany institution with which he is affiliated.


102

CONSERVATION AND ENVIRONMENTAL POLICY

The Conservation Reserve Program (CRP) andother land retirement programs have been some ofthe most successful conservation strategies inUSDA’s portfolio of conservation programs, pop-ular with landowners, environmentalists, sportgroups, and farm organizations. Land retirementhas not been without its critics, however. Someissues that have persisted throughout the course of modern land retirement programs since 1985include:

• Potential benefits have not been as fully realizedas possible;

• Land retirement is not an effective way toachieve many environmental objectives;

• The transition to less intensive land use has not been achieved or has been too costly;

• Agro-energy could undo 20 years worth of con-servation retirement.

CRP reform could meet many of these concerns. A reformed land conservation program would bestreamlined to avoid the multiplicity of programs,confusion, and inefficiencies of the current system. Itwould also focus on environmental goals which it canachieve and target lands that will provide the highestaggregate benefits.

More extensive reforms, such as permitting moreextensive grazing and forestry uses while the land isunder contract, can help further achieve CRP’s origi-nal goal of facilitating transition from heavily farmedland to more environmentally friendly agricultureuses. Another way to accomplish this goal is to movethe CRP from a land rental program to one in whichthe government purchases permanent easements,which would convey in perpetuity the right to intensively crop CRP land while permitting more ecologically friendly uses. Extensive use of permanenteasements would also prevent re-entry of currentlyretired land into the production of program cropsafter the current traunch of contracts expires, whichis quite likely given current corn prices and the desireto find more land with which to produce corn-basedethanol. The last thing Congress should do is simplyauthorize another round of 10-year contracts with noclear plan for transition.

Estimates of Economic Costs and Benefits

Modern CRP has been evaluated several times, priorto and during implementation and again prior toreauthorization. Net social benefits were estimatedranging from slightly less than $1 billion per year to more than $11 billion over 10 years. The costs

105

Land Retirement for Conservation: History, Analysis, and Alternatives

Ralph E. Heimlich

of CRP over 1985–2005 are estimatedat $21.8 billion, less enumerated sav-ings in government costs for commod-ity programs of $11 billion, giving anet cost of $10.7 billion (table 1). Apartial accounting of estimated naturalresources benefits totals $23 billion in net present value over the period,resulting in a net social benefit of$12.2 billion. Two categories are leftout: the change in farm income andthe change in consumer prices.

CRP rental payments undoubtedlyhelped keep many farms afloat in the1980s and solvent in the 1990s, butthey are transfer payments and areproperly considered a cost, rather than a benefit. The effect of CRP on cropprices for all farmers would be a benefit,but the amount of the effect depends onbeing able to estimate the net differencein world prices with and without CRP, taking intoaccount slippage. Changes in consumer prices, bothdomestic and international, are even more problem-atic, since much of the production on CRP land wasfor crops exported to world markets that moved frommassive surpluses to more balanced market conditionsand back over the 1985–2005 period.

While it is likely that natural resource benefits ofcontinuing CRP after 2007 would remain the same,or even increase if the program were better targetedto the benefits it best produces, the benefits of sup-ply control savings, consumer costs, and changes infarm income due to changes in commodity priceswould all likely differ from the past.

Obstacles to Achieving Greater Benefits

Of all the USDA agricultural conservation programs,CRP has the most explicit focus on achieving themost benefits for the dollars spent. It is the only pro-gram that has an explicit index for measuring theeffectiveness of each application considered relativeto its cost, the Environmental Benefits Index (EBI).

However, there remain technical and institutionalbarriers to achieving greater benefits in the program.These include a clearer focus on which goals areimportant, better targeting to achieve those goals,and the wisdom of splintering land retirement intomany subprograms.

CRP was originally focused on a single goal—retiring highly erodible cropland. Since 1990, however, the program has addressed multiple envi-ronmental goals, including soil erosion (both sheetand rill and wind), water quality (both surface andground), and wildlife habitat. The implicit premiseof the EBI is that any parcel offered contributes to all of these goals, and that the best parcels to enrollwould score highly on all of them. In reality, anygiven parcel will score well on one of these attrib-utes, at most, and either moderately or poorly on theothers. This is especially true when the attributes aremutually exclusive, such as sheet and rill erosionversus wind erosion, surface water runoff versusgroundwater infiltration, or habitat needs of gamewildlife versus T&E species.

While we would all savor the “win-win” situationof having every parcel contribute significantly to all


106

TABLE 1SUMMARY OF COSTS AND BENEFITS, CRP, 1985–2005

1985–2005Million dollars, Million dollars, undiscounted NPV at 3%

average per year discount rate

Direct costs (rent, incentives, $1,520 $21,799establishment cost, technical assistance and administration)

Supply control savings $783 $11,052

Net cost to the government $736 $10,747

Soil productivity $202 $3,003

Water quality $543 $8,078

Wind-blown dust $96 $1,427

Wildlife habitat $704 $10,474

Partial natural resources subtotal $1,545 $22,982

Net social benefit $809 $12,235

SOURCE: Agricultural Conservation Economics.

goals, more often we face the situation where aparcel that scores moderately well on several attrib-utes prevails over one that is stellar on only one,especially if it is less productive land with a lowrental rate. If, instead of accepting acreage based onthe total EBI score, the best scores on wildlife (N1),water quality (N2), and soil erosion (N3) wereaccepted for one-third of the acreage, each, higheraverage and minimum scores would have beenachieved in past signups.

In addition to examining how to best use the EBI, we need to ask what goals are best served byland retirement. Given modern agricultural meth-ods, retiring land from production is the best way to create more wildlife habitat, but is it the best wayto reduce non-point source water pollution orreduce soil erosion? Since cost considerations biasCRP toward taking the least productive cropland, it is likely to be land with the least application offertilizers and pesticides that contribute to waterpollution. The highly erodible land definition wasdiluted to increase the pool of lands eligible to bid in the mid-1980s, resulting in considerable amountsof land accepted in CRP for which water quality and soil erosion problems could be addressed usingconservation practices under EQIP, CSP, or otherprograms without retiring the land. While givingequal weight to wildlife, water quality, and erosionin the EBI seems superficially fair, wildlife scoresshould be given more attention since they cannot beadequately addressed without retiring land.

After the program has focused on the goals it canbest achieve, the next biggest barrier to getting greaterbenefits is the ability to target the lands that will pro-vide the highest benefits. Targeting land retirementhas advanced incrementally from totally untargetedprograms in the 1930s and 1950s, to the broad targetof highly erodible land in the 1980s, and the assess-ment and selection of bids on multiple environmen-tal criteria in the EBI in the 1990s. However, theindicators used in the EBI are still relatively bluntinstruments. Particularly lacking is the ability to focuson a parcel’s location and relationship to its landscapesetting. While the EBI provides specific informationon the parcel itself, GIS capabilities and data can

provide equally detailed information about the par-cel’s landscape setting. USDA’s Farm Service Agencyis already moving in this direction, but additionalsteps are needed. Better landscape assessment isurgently needed to choose between competingparcels offered for retirement, and to judge whetherany benefit is likely to accrue at all.

A final obstacle to achieving maximum benefits inland retirement is the increasingly splintered mosaicof land retirement programs facing landowners. Firstthe Wetland Reserve Program (WRP) was made aseparate program, followed by differentiating the con-tinuous signup and its derivative, the ConservationReserve Enhancement Program (CREP). Recent splin-terings include the Farmable Wetlands Pilot program(FWP) and the Grassland Reserve Program (GRP).These offshoots seek to carve out recognition of theirindividual constituents, and avoid competing withinthe general CRP rubric of EBI and soil rental rate. The multiplication of “reserves” is confusing tolandowners, however, and weakens the coherenceand larger constituency for the overall land retirementeffort. This is compounded by splitting responsibilityfor the programs across agencies (CRP, continuous,CREP, and FWP are in FSA; WRP and GRP in NRCS). It would be far clearer if all land retirementwere done under a single program in which all typesof land were offered for bid and more individualizedterms could be extended to the landowner. New andexpanded criteria for continuous or CREP enrollmentcould be implemented, and whole-field enrollmentcould also be made “continuous” by comparison to anEBI score standard fixed for a given period of time.Guidance would have to be provided regarding theoverall objectives and scope, and the criteria to applyin retiring land under the various subcategoriescorresponding to the many now-existing programs.

The Best Use of Land Retirement

The 2002 Farm Bill increased the funding for “work-ing lands” conservation programs five-fold, redressinga perceived imbalance between funding available toretire land from production and funding to install

LAND RETIREMENT FOR CONSERVATION: HISTORY, ANALYSIS AND ALTERNATIVES

107

conservation practices and systems on lands remain-ing in production. As a general rule, land retirementworks best when there are multiple resource prob-lems or opportunities that can be addressed by retire-ment (even if the selection is made based pre-dominantly on one, as above), and where few or noneof them can be addressed by changing farmingmethods. For example, land on which changes incrop rotation or nutrient application can reduce non-point water pollution concerns to acceptablelevels would most likely best remain in production,while land that is so highly erodible or so vulnerableto leaching or runoff that no cropping system caneconomically reduce soil or nutrient losses (assumingthose losses actually reach affected waters) would bea good target for retirement.

Applying this logic to wildlife habitat needs tendsto favor land retirement because little good can bedone for wildlife, particularly for sensitive threat-ened and endangered species, while engaged inintensive cultivation or retiring only small parcels.However, common wildlife species, such as smallmammals and pheasants, often can benefit fromrelatively minor changes in cropping systems, orfrom land retirement in a relatively small part of theoverall landscape.

Transitioning to Permanent Change

As originally envisioned, land retirement programswere intended to provide a transition from intensivecrop production to more extensive grazing or forestryuses for land that was environmentally sensitive. Wedon’t really know how successful modern CRP will beas a transition program, because nearly all of the acresenrolled since 1985 are still in the program. Surveysand contract extensions give a glimpse of the ultimatedisposition of CRP land enrolled. After the renewaland extension (REX) process FSA initiated in 2005,some CRP land will be entering its third 10-year rentalcontract. Some of this land may have also been idledunder previous programs in the 1950s and 1930s.Under REX, about 84 percent of contracts expiring in2007 accepted the renewal/extension offer.

Since 1985, some of the provisions of land retire-ment programs that stood in the way of a true tran-sition have been eliminated, but others remain. Theoriginal enrollments under the 1985 Farm Actrestored crop base acreage if the land reverted tocrop production and if a substantial reduction in soilerosion was accomplished by the required conserva-tion compliance plan. Sodbuster provisions apply-ing to new cropland require achieving the tolerablesoil loss level (T value), a more stringent target. The2002 Farm Bill once again made commodity pro-gram payments dependent on crop acreage base,although with nearly complete planting flexibilityand base protection for expiring CRP land.

It is no longer necessary to re-crop expiring CRP to retain base and collect direct (decoupled) pay-ments, but the incentives offered by countercyclicaland loan deficiency payments still push landownerstoward program crop production. Emergency hayingand grazing and managed haying and grazing provi-sions enacted in 1996 allow producers to use someCRP forage during declared disasters or to improvethe quality and vigor of the stand in exchange for areduction in rental payments. However, the restric-tions involved preclude using CRP forage for animportant component of any grazing operation.

The 2007 Farm Bill is yet another opportunity tomove CRP and other land retirement programs closerto true transition programs by permanently retiringcrop acreage base on enrollment, requiring the samestrict standard of soil erosion for returning CRP as forsodbuster, and allowing more liberal grazing andforestry use during contracts. These may require arental schedule that declines over time, and maymean that fewer landowners would be interested inparticipating, but those who do would be more com-mitted to actually changing use over the long term.

Outright Purchase or Easement

Adjusting for inflation, total payments for landretirement in CRP-like programs since 1933 are$48.7 billion in 2006 dollars, not counting WRP.This is equivalent to $1,730–$2,596 per acre, which


108

is greater than the 1996–97 U.S. average croplandvalue of $1,270 per acre. Repeated rental of land inten-year contracts costs as much as purchasing it aspublic land in 1996 when CRP was reauthorized.

However, there is another alternative to outrightpurchase. Permanent easements convey the right tointensively crop land in perpetuity, without restric-tions on compatible uses like grazing or forestry,while retaining private ownership. Purchase of per-manent easements faces a number of difficulties thatannual rental payments do not. Foremost, the con-gressional budget process does not distinguish theup-front payment needed to purchase such rightsfrom the year-by-year expense of a rental agreementin the way the two are scored for deficit reductionpurposes. Because expenditures in out-years of thebudget cycle do not count against deficit reductioncaps, rental is preferred over appropriation of fundsfor permanent easements that must be paid right now.

Potential for Conversion of CRP to Cropland

For most of the time since 1985, there was littlequestion that keeping land in CRP helped controlcommodity supplies and increase crop prices.However, in the last quarter of 2006, the price ofcorn shot up 50 percent to $3.12 per bushel, a pricenot seen since 1996. Unlike recent previous cornprice increases, this one is being driven not by sup-ply shortages, but expectation of greatly increaseddemand for corn, primarily for ethanol production.

How much more current CRP acreage would beprofitable to crop at these prices? More than 10 mil-lion of the 16 million acres set to expire in the nextfew years (63 percent) would be profitable to crop atprices recently observed. Most of this (70 percent)would be in corn, and another 23 percent in wheat.Agro-energy is the early 21st century’s equivalent ofthe Russian wheat deal: threatening to undo the last20 years of conservation effort and unleash a newburst of “fencerow to fencerow” enthusiasm inAmerica’s heartland. Agro-energy production hasbeen enticed into the market with a graduallyincreasing array of subsidies, crowned by the pro-

duction mandate in the 2005 Energy Act, and addi-tional subsidies are being discussed as part of apotential energy title to the 2007 Farm Bill.

Conclusions

Land retirement is the United States’ largest conser-vation program. Motivated in equal parts by supplycontrol and conservation purposes, judgments on itssuccess vary as conditions change. The first traunchof modern CRP, from 1985–1996, succeeded inhelping curb commodity production without moreexpensive paid land diversions, and achieved con-servation success, narrowly defined in terms of soilerosion and wildlife habitat.

The second traunch, from 1997 to the present,was undertaken despite improving conditions in theagricultural economy, and with a much more ambi-tious and ill-defined set of conservation goals. In amore global agricultural marketplace, with no clearneed for domestic supply control, but continuinghigh levels of government payments, land retire-ment’s performance is more questionable. Thediverse conservation goals and the splintering ofland retirement programs to address specific envi-ronmental constituencies make the question of con-servation performance less clear, as well. While agreat deal of good has been achieved, it is likely thatmore precise targeting would produce better resultsthan land retirement has accomplished to date.

Whether land retirement programs have facili-tated a transition to less intensive land use remainsan open question that will not be definitivelyanswered until contracts expire. While it is likelythat some covers (hardwood and softwood trees inCRP, wetlands in WRP) and instruments (permanenteasements in WRP and GRP) are likely to remain outof cultivation, failure to allow transition use undercontract and to permanently remove crop base onmost acreage enrolled make re-cropping a distinctlikelihood, especially if prices are favorable whencontracts expire.

Have we lost all the benefits of CRP if the land isre-cropped? All environmental externalities have

LAND RETIREMENT FOR CONSERVATION: HISTORY, ANALYSIS AND ALTERNATIVES

109

aspects of both flows and stocks to them. Soil erosiondamages can be turned on and off as we crop or idleland, but at some point, erosion damages are so severethat soil productivity cannot be regained. Certainly,what we are calling “long term” retirement is less thanthe blink of an eye on the geological scale of soilbuilding. Opportunities for “farm” wildlife are proba-bly flows that can be interrupted with little damage,but the biodiversity represented by threatened andendangered species can vanish from the Earth forever.Carbon sequestration is an ecological service flowoccurring over time, but at what point do damagescaused by global climate change become irreversiblereductions in the stocks that support all life?Intervention to achieve a more appropriate land usepattern seems reasonable, especially where the bene-fits at risk are not well represented in markets.

In retrospect, a program of permanent easementsmight have avoided some of the problems nowfacing USDA land retirement programs, particu-larly CRP. Permanent easements to remove croppingrights would have guaranteed a land use transition,would have allowed for less intensive uses immedi-ately, and would have cost less than the recurrent10-year rental contracts since 1985. However, landretirement’s evolution over the last 20 years leavespolicy makers in a quandary. Terminating CRP nowleaves many of the incentives for re-cropping inplace, especially if another lucrative farm bill like the

2002 Farm Bill is passed in 2007 or market pricesare high. Reauthorizing CRP for another traunch of10-year contracts, while permanently retiring baseand making sodbuster apply to CRP land, wouldremove incentives for re-cropping and allow newsignups to better target environmental objectives,but would add to rental dollars already expended.

Recasting CRP as a smaller, tighter permanenteasement program would correct the problematicprogram design of the past and allow for bettertargeting, but would probably require paying forcropping rights that have not been exercised in atleast 20 years, in addition to 20 years of rentsalready paid. In any case, some blending of theseapproaches with a more explicit balancing of generaland continuous/CREP signup modes is probably inorder. The last thing Congress should do is simplyauthorize another round of 10-year contracts withno clear plan for transition.

Note

This policy brief draws on the author’s longer paper “Land Retirement for Conservation: History, Analysis, andAlternatives,” available at www.aei.org/farmbill. The longerpaper includes background support and citations notincluded here. Heimlich is the owner of AgriculturalConservation Economics. The views expressed here arethose of the author and not those of any institution withwhich he is affiliated.


110

There is widespread sentiment among both policyanalysts and public interest groups that U.S. farmsubsidy policy is flawed and in need of reform (Agri-culture Task Force, 2006). Existing subsidy policiescannot be justified on either equity or efficiencygrounds. Subsidies based on acres produced of“program” crops not only distort domestic and inter-national markets; they also disproportionatelybenefit large, commercial farms owned by corpora-tions and individuals with above-average incomesand wealth.

The longer paper argues that there are soundscientific and economic reasons for an agriculturalpolicy based on the provision of ecosystem servicesby agriculture. There is a great deal of scientific datashowing that agricultural land use and managementimpact a wide array of goods and services associatedwith ecosystem function that are not usually valuedby or traded in markets. Economic theory tells usthat under these circumstances markets based onprivate benefits and costs do not allocate resourcesefficiently. Left to their own devices, markets will tendto over-produce market goods and under-produceecosystem services. Thus, there is both a scientificand economic basis for an agricultural policy thatwould provide incentives for farmers to supply theappropriate combination of market goods andecosystem services.

An efficient agricultural policy could be imple-mented using a policy mechanism that I call pay-ments for ecosystem services, or PES. A PES systemrewards farms for increasing the quantity of eco-system services they supply above and beyond theamount that would have been provided absent suchrewards. A key feature of PES is that they are notsubsidies; they are financial incentives provided tofarmers who bear costs to increase the supply ofecosystem services valued by society. Thus, PESdiffers in important legal and economic respectsfrom existing commodity subsidies. A PES policyalso differs in important ways from most existing“conservation” or “environmental” programs. A keydifference is that most programs prescribe conserva-tion and environmental practices that farmersshould use. Experience with environmental regula-tion shows that this prescriptive approach is far lessefficient than policies based on performance stan-dards and incentives. Performance standards andincentives are one way that a PES system could beimplemented, and recent research has confirmedthat performance-based policies would be a far moreefficient way to achieve policy objectives than exist-ing practice-based policies.

The United States has a long history of govern-ment programs related to conservation and environ-mental management. Over time U.S. policy has

111

Payments for Ecosystem Services and U.S. Farm Policy

John M. Antle

shifted from maintaining on-farm productivitytoward programs that reward farmers for resourceconservation and provision of ecosystem services.Thus, the United States has already traveled somedistance down the path toward policies that resem-ble a PES system, although existing policies fall farshort of what an efficient PES system would be. Anefficient policy would aim to maximize the net ben-efits to both farmers and society, taking into accountthe value of marketed commodities and non-marketgoods and services. Most of USDA’s soil conservation,conservation compliance, and conservation reserveprograms are inefficient for several reasons:

• Farmers are paid for adopting certain practices,regardless of the amount of ecosystem service they provide, rather than being paid per unit ofvalued services.

• Policies often have multiple, conflicting goals, andare implemented with politically determinedbudget constraints that do not reflect efficientregional resource allocation.

• Conservation programs are often more costly thannecessary because they require farmers to takeland out of production, even though it is oftenpossible to continue using the land productivelywhile providing environmental services with alter-native management practices.

Despite these shortcomings, U.S. conservationand environmental programs already have somefeatures similar to a PES system. Issues that neededto be addressed to move the existing systemtowards a more efficient one include:

• Policy decision makers will need better informa-tion to design efficient mechanisms for the provi-sion of ecosystem services from agriculture. Policydecision makers will need estimates of the envi-ronmental and economic effects and budgetimpacts of a PES system and how it will differ fromthe existing commodity and conservation pro-grams. To implement a more efficient PES system,additional investment will be needed in the sci-ence needed to quantify ecosystem services, takinginto consideration issues such as the interactions

between different kinds of services. In addition,further research needs to address how to designand implement contracts for ecosystem services.

• Farmers will need better information and decisiontools to help them assess when they should enterinto contracts to provide ecosystem services. Thesetools need to help farmers estimate the quantitiesof services they can produce and the costs of alter-native management practices.

• The compatibility of a PES system with WTO rulesneeds to be further investigated, and if necessaryWTO rules may need to be modified to allow anefficient PES system. PES based on provision ofecosystem services, not market goods production,should be compatible with the WTO rules.However, current WTO rules do not allow pro-ducers to be compensated in excess of their privateopportunity costs, thus possibly calling into ques-tion the acceptance of a PES system that rewardedfarmers based on social value that exceeds oppor-tunity cost.

Much of the data and science needed to imple-ment an efficient PES currently exists, but there aregaps that need to be filled. To implement a PES pol-icy successfully, data and models are needed to esti-mate farmers’ willingness to participate in contracts,the quantity of services that would be supplied, andthe budgetary costs of programs. Substantial progresshas been made in developing this capability for someecosystem services related to water quantity andquality and greenhouse gas mitigation, but less well-defined services, such as those related to biodiversity,need further work. Information is also needed tosupport farmers’ contract participation decisions.Recent work on carbon sequestration provides a tem-plate for how information to support other ecosys-tem services could be made available to farmers.Existing USDA and land-grant university researchand outreach organizations are well-positioned toprovide the currently available information to policydecision-makers and to farmers, and to design andimplement the research programs needed to supple-ment existing data and knowledge.


112

Note

This policy brief draws on the author’s longer paper“Payments for Ecosystem Services and U.S. Farm Policy,”available at www.aei.org/farmbill. The longer paper

includes background support and citations not includedhere. Antle is a professor in the Department of AgriculturalEconomics and Economics, Montana State University. Theviews expressed here are those of the author and not thoseof any institution with which he is affiliated.

PAYMENTS FOR ECOSYSTEM SERVICES AND U.S. FARM POLICY

113

The production of food and fiber by commercial agri-culture has both positive and negative effects on thesurrounding environment. Farms can provide wildlifehabitat and scenic views for their urban neighbors,while the carbon sequestered by vegetation grown on that land can help to mitigate global warming. At the same time, noise, dust, and odors produced bynormal farming operations can annoy urban resi-dents, and runoff of pesticides, fertilizer, and animalwaste can lead to water pollution downstream.

The economic values that society places on theseoff-farm byproducts of agricultural production (i.e.,externalities) are rarely captured fully by private mar-kets for farmland or commodities. When marketprices fail to reflect the impact of farming on the sur-rounding environment, farmers are left with littleincentive to incorporate the value of off-farm envi-ronmental quality into their management decisions.This type of market failure is often a rationale forgovernment action such as regulation, taxes, subsi-dies, and the redefinition of property rights.

For the past twenty years, government interven-tion has primarily consisted of voluntary conservationpayment programs funded by farm bill legislation.These programs pay farmers to manage their land inways that reduce erosion and runoff, while increasingthe provision of wildlife habitat and other environ-mental benefits. In addition to voluntary conservation

programs, the federal effort to address agriculturalexternalities also includes compliance provisions ofthe farm bill and regulations on chemical use, emis-sions of air and water pollutants, and the private useof endangered species habitat. The burden that theseenvironmental laws place on farmers has been high-lighted by recent debates on amending endangeredspecies legislation and expanding regulations on con-fined animal feeding operations (CAFOs).

This paper describes the current regulatory envi-ronment and suggests opportunities for reforming the Farm Bill conservation programs to more effi-ciently achieve environmental goals. Overall, produc-tion agriculture faces far less regulation than othersectors of the economy. Current laws addressing airpollution, water pollution, and the use of toxic chem-icals implicitly or explicitly exempt all but the largestconfined animal feeding operations from federal reg-ulation. This lack of regulation underscores the role ofthe Farm Bill conservation programs in addressingenvironmental issues.

While current programs generate benefits thatexceed their implementation costs, federal budgetlimits prevent them from being extended to morethan a small share of farmland. If policymakers were willing to tax the management practices thatnegatively impact the surrounding environment,instead of paying farmers to change those practices,

115

Public Policy Solutions to Environmental Externalities from Agriculture

Nicolai V. Kuminoff

environmental issues could be addressed on a greaterportion of farmland at a lower cost to taxpayers.

Alternatively, if opportunities for reform are lim-ited to changing the way farmers are paid, there are atleast three ways to make the current conservationprograms more efficient. First, there is potential forefficiency gains from coordinating the differentprograms. Second, linking payments to measurableoutcomes would help to align farmers’ financialincentives with public environmental goals. Finally,requiring the recipients of farm bill payments toreport their management practices would create newopportunities to address environmental issues with-out additional government intervention.

Environmental Regulation and Conservation Payments

Environmental regulation in the United States is oftendescribed as following a “polluter-pays-principle.”That is, firms have to pay in order to legally emit pol-lutants into the air and water by purchasing a permitfrom the government or by installing pollution abate-ment equipment that satisfies minimum technologystandards. While the agricultural sector of the econ-omy is no exception to the polluter-pays-principle,farmers receive preferential treatment.

Table 1, at the end of this brief, summarizes themajor environmental regulations and their exemptionsfor farmers. Current laws regulating air pollution, waterpollution, and the use of toxic chemicals implicitly or explicitly exempt all but the largest concentratedanimal feeding operations (CAFOs) from federal regula-tion. Together, these exemptions give farmers the rightto discharge unlimited quantities of pesticides, fertilizer,and animal waste produced as a byproduct of normalfarming operations. Only a small share of the largest ani-mal feeding operations—less than 1 percent of all farmsthat account for approximately 10 percent of gross farmrevenue—are required to obtain federal permits in orderto discharge pollutants into U.S. waters. The authorityfor regulating emissions from the other 99 percent offarms is delegated to the states, which mostly rely onvoluntary or incentive-based programs.

This leaves three major regulations with the poten-tial to constrain farmers’ management practices: EPA’spesticide registration program, endangered species leg-islation, and the compliance provisions of the FarmBill. For most farmers, the actual burden imposed bythese regulations appears minimal. Cancellation of akey pesticide certainly has the potential to decreaseprofits for farmers who depend on it. But the larger theeconomic burden, the greater the likelihood of a specialexemption, as demonstrated by the recent exemptionsfor methyl bromide. The Endangered Species Act alsohas the potential to impose a regulatory burden onfarmers by prohibiting normal farming practices thatharm threatened or endangered species. The high pro-file conflict between farmers and fishermen over waterallocations in the Klamath River Basin has draw atten-tion to this issue. But there appear to be few compara-ble examples. Finally, while compliance provisions ofthe Farm Bill can be used to withhold payments tofarmers who convert wetlands or farm highly erodiblecropland, enforcement is limited by provisions of the1996 Farm Bill and an appeals process that routinelywaives most penalties.

Production agriculture clearly faces less environ-mental regulation than other sectors of the economy.However, it would be incorrect to claim that farmersare unaffected. Compared to developing countries, theUnited States tends to place more emphasis on usingthe pesticide registration process to protect food safety, the health of pesticide applicators, and wildlifepopulations, which may put domestic growers at acompetitive disadvantage in international markets.Furthermore, farmers may be affected by environmen-tal regulations on upstream and downstream firms. For example, firms that manufacture pesticides orprocess farm products do not enjoy the same regula-tory exemptions as farmers. Part of the burden fromthese regulations may get passed on to farmers throughhigher pesticide prices and lower farm gate prices.

Rather than addressing agricultural externalitiesthrough regulation, past farm bills have fundedvoluntary conservation programs. The ConservationReserve Program (CRP) and the Wetlands ReserveProgram (WRP) pay farmers to retire environmentallysensitive land. The Environmental Quality Incentives


116

Program (EQIP), the Conservation Security Program(CSP), and the Wildlife Habitat Incentives Program(WHIP) pay farmers to implement conservation prac-tices on working farmland. Finally, the Farm andRanchland Protection Program (FRPP) and theGrasslands Reserve Program (GRP) provide fundingto purchase easements on farmland to prevent it frombeing converted to more intensive uses. Overall,USDA spent nearly $4 billion on these programs in2005. While these expenditures may seem large, theywere only sufficient to enroll a small share of farm-land. Together, CRP, WRP, CSP, WHIP, FRPP, and GRP funded activities on approximately 46 millionacres—at most 10 percent of U.S. cropland.

Opportunities for Policy Reform

The coupling of conservation payments with minimalregulation has been dubbed a “pay-the-polluter”approach to addressing farm externalities, in contrastto the “polluter-pays-principle” that applies elsewherein the economy. This preferential treatment is at leastpartly due to the administrative costs of regulation.While it is feasible for EPA to regulate small groups offarmers like CAFOs, extending permit and enforce-ment programs to the entire farm sector wouldrequire an order-of-magnitude expansion.

If policymakers were willing to tax the manage-ment practices that generate negative externalities,instead of paying farmers to change those practices,the environmental impacts of production agriculturecould be addressed on a greater portion of farmlandat a lower cost to taxpayers. The monitoring andenforcement costs of operating such a program couldbe at least partly funded by the tax revenue it gener-ates. Of course, such a drastic departure from currentpolicy may be politically infeasible. If realistic oppor-tunities for reform are limited to changing the waypolluters are paid, there are at least three ways toimprove the efficiency of current programs.

1. Coordinating Conservation Programs. Currently,farmers self-select into different conservation programsthat seek to address the same set of externalities. This

situation may fail to exploit the potential efficiencygains from having a mix of policy instruments. Forexample, land retirement is probably better suited topreserving habitat for sensitive wildlife species, while itmay be more efficient to address runoff of farm chemi-cals by paying a farmer to reduce applications of fertil-izer or to install buffer zones adjacent to waterways.Retirement and working land programs also differ intheir ability to create tax savings by reducing the size ofpayments made through price support programs.

Policymakers could exploit the diversity of conser-vation instruments by coordinating their enrollmentprocesses. One possibility would be to have farmerssubmit a single application that outlines both landretirement and working land options, knowing thatprogram officials would choose which (if either) con-tract to offer them. Program officials could then allocatefunds between land retirement and working land pay-ments in a way that exploits each instrument’s relativeadvantages at addressing externalities and creating taxsavings. If merging or coordinating the programs is notfeasible, a simpler approach would be to adjust theweights on the indices used to select contracts for eachprogram to emphasize their relative strengths.

2. Linking Payments to Environmental Perform-ance. The working land programs (EQIP, CSP, andWHIP) pay farmers to implement conservation prac-tices regardless of the environmental outcome. Thisstrategy fails to align farmers’ financial incentives withpublic environmental goals. For example, a farmercould collect WHIP payments for implementing a proj-ect that is believed to have a beneficial effect on habitatfor the endangered grey wolf, regardless of whether anywolves ever visit the farm. A performance-based ver-sion of the same program might pay farmers in pro-portion to the number of wolves that visit the farm ormake additional payments for wolf pups born on thefarm. In this example, linking payments to the behav-ior of the wolf population provides a stronger incentivefor the farmer to enhance wolf habitat. More generally,linking payments to environmental outcomes rewardsfarmers for developing new, cost-effective ways ofaddressing externalities. This incentive for innovation ismissing from the current programs.

PUBLIC POLICY SOLUTIONS TO ENVIRONMENTAL EXTERNALITIES FROM AGRICULTURE

117


118

It would be more challenging to develop a practi-cal way of linking payments to environmental out-comes for externalities that are widely dispersed ordifficult to measure, such as water pollution and ecosystem services like biodiversity and carbonsequestration. One possibility would be to base pay-ments on the predictions of agricultural ecosystemmodels that use chemical and biophysical relation-ships to explain how environmental outcomesdepend on farm characteristics, climate variables, andland management. For example, the EnvironmentalPolicy Integrated Climate (EPIC) model can predicthow different tillage methods affect carbon sequestra-tion, soil erosion, and nitrogen runoff. However,because models like EPIC abstract from the complex-ity of real ecosystems, their predictions will be impre-cise. Furthermore, using the predictions from thesemodels as the basis for payments would requiredetailed information on the physical characteristics ofindividual farms and farmers’ management practices.

3. An Information Compliance Provision. Addingan information compliance provision to the Farm Billwould require that, each year, farmers who receive anypayments must report some information about thegeographic characteristics of their farm and their landmanagement practices. In the near term, these datawould provide an important input to the continuingresearch effort to understand the links between farm-land management and environmental outcomes. Inthe longer run, the data could be used together withagricultural ecosystem models to make performance-based payments to farmers. Finally, releasing some ofthe data to the public would create new opportunitiesfor private parties to negotiate solutions to externalitieswithout additional government intervention.

A key challenge in developing the compliance pro-vision would be to certify the accuracy of self-reportedinformation. EPA currently faces this challenge with itsToxic Releases Inventory, which is based on firms’ self-reported production, use, and releases of numerouschemicals. Misreporting has been an ongoing concern.EPA addresses this concern by randomly auditing asmall number of firms as well as visiting firms thatreport unusual activity, such as the largest year-to-year

changes in chemical use. Firms are fined for misre-porting information or failing to submit the requiredinformation. If farmers were found misreporting theirmanagement practices, one possible enforcementoption would be to remove their right to collect futurepayments from Farm Bill programs.

Conclusion

The paucity of environmental regulations on farm-ing means that the 2007 Farm Bill will play a keyrole in the production of agricultural externalitiesover the next few years. Replacing conservationpayments to polluters with taxes on managementpractices that generate negative externalities wouldallow policymakers to address environmental issueson a greater portion of farmland at a lower cost totaxpayers. If this proposal is infeasible, there arethree ways in which the current programs can bereformed to address externalities more efficiently.First, there is some potential for efficiency gainsfrom coordinating the conservation programs.Second, linking payments to measurable environ-mental outcomes would help to align farmers’ finan-cial incentives with public environmental goals.Finally, requiring the recipients of farm bill pay-ments to report their management practices wouldprovide new information that could help researchersunderstand the links between management prac-tices and environmental outcomes, serve as the basisfor a future system of performance-based payments,and create new opportunities for private parties tonegotiate solutions to externalities.

Note

This policy brief draws on the author’s longer paper“Public Policy Solutions to Environmental Externalitiesfrom Agriculture,” available at www.aei.org/farmbill. Thelonger paper includes background support and citationsnot included here. Kuminoff is assistant professor in theDepartment of Agricultural and Applied Economics, Vir-ginia Tech. The views expressed here are those of the author and not those of any institution with which heis affiliated.

PUBLIC POLICY SOLUTIONS TO ENVIRONMENTAL EXTERNALITIES FROM AGRICULTURE

119

TABLE 1SUMMARY OF FEDERAL ENVIRONMENTAL REGULATIONS AND KEY EXEMPTIONS FOR FARMERS

Area Regulation Key Exemptions/Limitations Outcome

SOURCE: Author’s calcuations.

Point Source Wastewater Permits

Point sources must satisfy technol-ogy and water quality standardsto obtain a permit to legally dis-charge pollutants into U.S. waters.

Dredge or Fill Permits

Permits are required to fillwetlands.

Nonpoint Source Pollution

States must develop plans toaddress pollution from nonpointsources in waters failing to meetambient quality standards.

State Implementation Plans

Each state must develop anenforceable plan to meet nationalambient air quality standards, orbe regulated by EPA.

Pesticide / Fertilizer Registration

Registration, determination ofapproved uses, and limitationson who can apply them.

Monitoring, reporting, and liabil-ity for storage / disposal of toxicchemicals.

Prohibits “takings” of threatenedand endangered species, andmigratory birds.

Farmers who convert wetlands orfail to apply conservation systemson highly erodible land cannotcollect payments.

Point sources include confinedanimal feeding operations(CAFOs) in general, but exempt“agricultural stormwater dis-charges and return flows fromirrigated agriculture.”

Excludes “normal farming” activi-ties with incidental discharges ofdredged or fill material.

1) Federal funding and enforce-ment are limited.

2) States determine which nonpoint sources to regulate.

Regulations emphasize “majorsources” that emit thresholdlevels of pollutants. Thresholdsimplicitly or explicitly excludefarmers.

Subject to EPA approval, statesmay register additional pesticideuses or temporarily use an unreg-istered pesticide to address pestemergencies.

Exempts registered pesticidesand agricultural use of fertilizers.

Unclear whether intent must bepresent in the case of poisoningof migratory birds.

Provisions only apply to a smallshare of current recipients of farm program benefits. Moreover,enforcement is questionable.

1) Approx 4,100 CAFOs havepermits.

2) All other farms may legally discharge soil, animal waste,fertilizer, and pesticides into US waters without a permit.

In some cases, farmers can con-vert wetlands to crop productionwithout a permit.

Some states exempt farmers whileother states promote voluntaryadoption of best managementpractices. Direct regulation bystate or local officials is rare.

Individual farms are not regulatedunder the Clean Air Act.

EPA determines which pesticidesand fertilizers farmers can use,but special exemptions have been allowed for methyl bromideand others.

EPA does not regulate, track, orreport farmers’ use of registeredpesticides and fertilizers.

Legal actions have been takenagainst farmers and ranchers who “take” threatened or endan-gered species.

Farmers receiving payments have an incentive to comply.Other farmers do not.

WaterPollution

AirPollution

ChemicalUse

WildlifeHabitat

Farm Bill

Biofuel Basics

Biofuel is a renewable energy source that can beused to substitute for petroleum imports, improveoxygenate levels in fuels and hence improve airquality, and reduce the amount of carbon dioxidereleased into the atmosphere. Ethanol from corn is the most commonly used biofuel. When cost-competitive enzymatic processes are developed forbreaking the cellulosic fibers in corn stover, grasses,and woody plants into component sugars, biomassmay be used to supplement corn grain as an ethanolfeedstock. Other conversion processes, such as pro-ducing bio-gas or bio-oil and refining the productinto a biofuel (e.g., bio-butanol), are also underdevelopment. But even under an optimistic scenario,such technologies will not be available on a com-mercial scale before 2015.

Biodiesel is a liquid fuel with combustion propertiessimilar to petroleum diesel fuel. It can be made fromvegetable oil, animal fat, or recycled grease via a processwhere methanol is used to convert base oil into methylester. In the U.S., biodiesel is primarily produced fromsoy oil, animal fats, and waste cooking oil.

Ethanol output has expanded steadily since 1980to an estimated 7.5 billion gallons in 2007. Thisexpansion has been heavily driven by policies,notably subsidies granted through federal and state

tax credits, and ethanol as a gasoline additive toachieve air quality regulations in the Clean Air ActAmendments. Recently, increases in oil prices havegiven an overwhelming boost to biofuels as a substi-tute for petroleum-based fuels.

Currently, the primary feedstock for ethanol iscorn. Over 13 percent of the 2005 U.S. corn crop wasused as ethanol feedstock, with projections that over25 percent of the corn crop could go into ethanol pro-duction by 2009. Corn prices have nearly doubled inthe last year alone, and corn futures contract prices fordeliveries into 2010 are above $4.00 per bushel. Thisis turning long-standing farmer concerns with lowprices and surpluses to concerns from other interestgroups over the growing competition between food,feed, and fuel. Livestock producers are concerned overhigh feed costs and disruptions in livestock markets,humanitarian groups over potential increases in globalfood prices, and environmental interests overincreased water quality and greenhouse gas emissionas corn production expands to meet ethanol demands.

The 2007 Farm Bill will be developed in an erapromising higher commodity prices, where marketprices will be well above support levels, and loan deficiency payments will not be required to make up the difference between market price and thesupport price, and countercyclical payments willalso decrease.

121

Biofuel Incentives and the Energy Title of the 2007 Farm Bill

John A. Miranowski

Costs and Benefits of Policy Intervention

The main policies influencing the ethanol market are:1) every gallon of ethanol that is blended receives a taxsubsidy of 51 cents per gallon, and 2) the Clean Air ActAmendments generate increased market demand forethanol. In 2006, the tax subsidy was over $2.5 billion.If the industry expands to 14–15 billion gallons annu-ally, the tax subsidy will climb to $7.0–7.5 billion peryear. Do the benefits of these policies exceed the costs?

Three sources of benefits from expanded biofuelproduction are:

• Energy security. About 60 percent of U.S. oil con-sumed is imported, much of it from politicallyunstable parts of the world. To the extent thatthese “less reliable suppliers” create a threat to theU.S. economy and national security, substitutingdomestically produced biofuel sources forimported oil will reduce that risk and possibly thelevel of national defense expenditures.

• Environmental quality. Blending ethanol with gaso-line reduces carbon monoxide and other harmfulair emissions. The Clean Air Act Amendments(1990), which mandated the use of reformulatedgasoline and oxygenated fuels in regions with seri-ous mobile source air quality problems, weredesigned specifically for this purpose. In addition,substituting biofuels for petroleum reduces theamount of greenhouse gas (GHG) emissionsreleased into the atmosphere.

• Reduced federal budget exposure. Higher corn pricesdue to increased ethanol demand result in reducedgovernment commodity program payments tocorn and soybean producers. Also, rural employ-ment growth may eventually reduce the need forrural development funding.

The issue of benefits relative to costs is easiest to quantify for the ethanol subsidy. In terms of 2006values, do the sum of the energy security gains, envi-ronmental quality (air quality and GHG emissionreduction) gains, and reduced commodity programcosts attributable to ethanol expansion exceed the$2.5 billion in ethanol subsidies?

With respect to energy security, the benefits arenearly impossible to measure short of attributing aportion of defense expenditures to energy securitycosts, as a few groups have done. Instead, it is impor-tant to note that corn ethanol supplied about 3.5 per-cent of the 142 billion gallons of gasoline consumedin 2006, which is likely an imperceptible contribu-tion. Even if we use 15 billion gallons of biofuel by2015, that would account for less than 10 percent ofgasoline consumption and 6–7 percent of U.S. motorfuels and likely be a small contribution to energysecurity at a cost of $7.5 billion.

Using biofuel reduces greenhouse gas (GHG)emissions relative to fossil fuels. Including the life-cycle energy use of corn-based ethanol, GHG emis-sions are reduced by about 12–13 percent relative tothe same Btu-equivalent amount of gasoline. Theactual GHG emission reduction per mile driven isabout 1 percent with an E10 blend. With respect to Clean Air Act Amendment gains from blendingethanol, the major gains are in non-attainmentregions with air quality problems. Most of these gainshave been or soon will be realized. Benefits fromblending ethanol in attainment regions are morelimited for carbon monoxide and volatile organic car-bons but may increase nitrous oxide emissions.1

Budgetary savings due to commodity priceincreases are primarily a matter of countercyclicalpayments for corn, since prices have been and areexpected to continue to exceed the loan-programsupport price either with or without an ethanol sub-sidy. The subsidy of 51 cents per gallon is about 25 percent of the price of ethanol. Using estimatedsupply and demand functions for ethanol based onrecent market data, removal of that subsidy wouldreduce the production of ethanol by 5 percent in theshort run.2 This would reduce the demand for cornby about 1 percent, since about 20 percent of corn isused in ethanol production, and this would reducethe price of corn by about 3 percent or 8 cents perbushel in the short run. Under 2002–04 market con-ditions, countercyclical payments for corn were beingmade even with the rate of ethanol use that theethanol subsidy then generated. Thus, without theethanol subsidy countercyclical payments for corn


122

would have been 8 cents per bushel higher. With 8 billion bushels of corn receiving payments, thebudgetary cost savings due to having the ethanol sub-sidy are therefore estimated at $640 million.

Calculating the benefits and costs at the margin,adding a gallon of ethanol at a subsidy cost of 51 centsgenerates budgetary gains of 6.4 cents per gallon ofadditional ethanol stimulated by the subsidy. Thus, ifthe subsidy were to be justified by benefits as large asthe costs, the energy security benefits plus clean airbenefits would have to be almost 45 cents per gallonat the margin, or additional gallons of ethanol the sub-sidy induces compared to the absence of the subsidy.3

Alternative Estimates

The preceding estimates were made on the basis ofsupply and demand functions for ethanol based onmonthly data, so they show a short-run effect ofremoving the subsidy. Over the longer term, the sup-ply of ethanol would be more responsive to theabsence of a subsidy because processing capacity,which is largely fixed in the short run, would be morevariable. Demand would be more responsive to price,also. The long-run break-even price that a modernplant can pay for corn is estimated at about $4.25 perbushel. In the absence of the subsidy, the breakevenprice decreases to about $2.75 per bushel. Some haveargued that these responses are substantial and that,were it not for the subsidy, fuel ethanol would notexist until crude oil prices were sufficiently high ($60per barrel crude oil) and would not expand beyond7.5 billion gallons, the RFS.

Additional Costs and Benefits of Biofuel Expansion

In addition to direct costs of production, biofuelexpansion can have a number of other costs. As cornacreage increases in response to higher corn pricesand increased ethanol production, nutrient use andsoil loss will likely increase. Corn acreage will bepushed unto more erodible soils, more erosive

practices will be utilized in corn production, and thederived demand for fertilizer nutrients will increase.Nutrients and sediment from soil runoff are the twomajor sources of water quality deterioration origi-nating in Midwest agricultural areas. Further, use ofmore nutrients and more erosive production prac-tices will reduce or eliminate the GHG emissionreduction associated with corn grain ethanol.

A frequently cited benefit of the ethanol boom inthe Midwest is rural development. The corn ethanolindustry creates well-paying jobs in rural communi-ties, which have local income and job multiplierimpacts that ultimately lead to community develop-ment. It is argued that these community benefits maybe further enhanced by local investment as opposedto outside investment.4 As the industry grows, thesepotential benefits are becoming less significant for anumber of reasons. First, earlier dry mill plants weresmaller, employed more workers per unit of output,had a larger share of local investment, and had a lim-ited impact on local corn markets and livestock feedcosts. Today, new dry mill ethanol plants haveincreased annual production capacity from 40 mil-lion gallons or less to over 100 million gallons. Newplants employ under 0.5 workers per million gallonscapacity, whereas earlier plants employed up to 2.5workers per million gallons capacity. Roughly 40percent of ownership in earlier plants in Iowa waslocal and relatively little investment is local in newplants.5 For the Midwest and the U.S., the addedbenefits of ethanol industry expansion have likelyexceeded the added costs of industry expansion todate. The corn ethanol industry has demonstratedthat biofuel can make a modest contribution in meet-ing air quality requirements, GHG emission reduc-tions, energy security, and reducing farm programpayments. Such industry expansion can also bringabout local economic development without destruc-tive impacts on livestock industry growth and thequality of the rural landscape.

Continued expansion of the corn ethanol indus-try will add benefits, but eventually the added costswill exceed added benefits. At what ethanol outputlevel the added benefits exceed the added costs is uncertain. Continued industry expansion will

BIOFUEL INCENTIVES AND THE ENERGY TITLE OF THE 2007 FARM BILL

123

produce less new employment per unit of capacityand may eventually create a net job loss. Localindustry expansion may create environmental costsif groundwater supplies are drawn down and con-sumption is disrupted in some local areas. Althoughcorn ethanol has helped meet certain CAAA require-ments, significant expansion of the industry willcontributed to decreasing reductions of GHG emis-sions if more intensive production practices andmore erodible lands are used to produce corn.

Alternatives to Biofuel Subsidies

On balance, do the benefits of biofuel incentivesoutweigh the costs? The basic economic question tobe addressed with respect to economic policy andthe welfare impacts of biofuel incentives involvespolicy objectives. If the goals are to improve energysecurity and reduce dependence on foreign oil,reduce GHG emissions, and reduce federal budgetexposure on farm commodity programs, what is theleast distorting or least-cost approach to achievethese objectives?

The economic response to the GHG emissionsreduction objectives would be to impose a tax onpetroleum and other fossil fuels equal to the mar-ginal value of the negative externalities created byGHG emission costs. Alternatively, we could estab-lish a cap and trade system for GHG emissions andlet the private sector pursue the least-cost way ofachieving a given reduction in GHG emissions. Byletting the private sector find the least-cost solution,resources would be allocated more efficiently inachieving the GHG emissions reduction. Our nationhas not demonstrated the political will to tax petro-leum and other fossil fuel sources or create a cap andtrade system except in the case of sulfur emissionsfrom power plants. Similar provisions could be usedin the case of energy security costs. A tax on petro-leum and other fossil fuels would increase conserva-tion, reduce imports, and improve energy security.

Historically, the U.S. has pursued a “cheapenergy” policy through the use of extensive tax sub-sidies to the fossil fuel energy sector, including many

provisions contained in the Energy Policy Act-2005. In the ongoing policy effort to maintain low petro-leum prices, prices do not serve as an effective sig-naling mechanism for alternative energy and biofuelmarkets to develop substitute fuels. Instead, policymakers have chosen fuel forms, such as ethanol andbiodiesel or electric cars in California, to receive taxsubsidies and mandates in an effort to create a “morelevel playing field” relative to petroleum fuels.

Even if the nation chooses to pursue second bestpolicy options, some second best options may beless inefficient than other options. For example,since there is no “silver bullet” to solve the so-called“energy problem,” a more efficient solution wouldbe to provide comparable tax subsidies to all formsof low carbon or renewable fuels, including hydro-gen and nuclear energy. Alternatively, mandatescould be more fuel inclusive to let the market deter-mine which fuel forms would achieve the mandateat lowest cost. Such approaches would stimulatemore efficient or competitive fuel forms to be devel-oped and marketed, would encourage adoption ofcarbon reduction technologies in refining and pro-cessing petroleum and other fossil fuels, and wouldencourage different fuel utilization technologiesfrom improved internal combustion engines to elec-tric and hydrogen powered vehicles. The alternative,having policymakers and the government “pick thewinners,” is not an efficient alternative.

As a second best solution, Kopp (2006) argues,“Subsidies and mandates are better suited to com-mercialization, while policies focusing on R&D are better suited [to] pre-commercialization.” Moretypically, both incentives are provided to new tech-nologies. The recently announced winners of U.S.Department of Energy (DOE) grants for six cellu-losic-ethanol biorefineries are an illustration of thisapproach (Science, 2007). In addition to the 51 centper gallon ethanol subsidy, the six biorefinery grantrecipients will receive up $80 million to research,develop and commercialize cellulosic ethanol.

Several biofuel provisions have been proposedfor the upcoming 2007 Farm Bill processes. Theserange from provisions to extend or eliminate theethanol subsidy to R&D incentive for biomass


124

feedstock and biomass ethanol provision. Optionscurrently being discussed for the Energy Title of the2007 Farm Bill:

Expand Federal Direct Market Intervention toSupport Renewable Energy:

• Raise the level of the Renewable Fuel Standard(RFS).

• Extend renewable energy tax credits to 2015 or later.

• Reduce biofuel tax credits when they are noteffective in increasing biofuel supply or are notneeded.

• Provide accelerated depreciation on renewableenergy equipment and facility investment.

• Use more land enrolled in the CRP for biomassharvesting.

• Refocus the CCC Bioenergy Program to supportbiomass used in bioproduct processing.

Expand Federal Indirect Support for RenewableEnergy:

• Expand the national cellulosic ethanol researchinitiative.

• Expand creative financial engineering to supportdevelopment of the biobased economy beyondgrants, loans, and loan guarantees.

• Bridge the gap between federally funded basicresearch and industry-funded applied researchand development.

Conclusions

Energy-related provisions were included in the 2002Farm Bill under the Rural Development Title. Giventhe increased interest in biofuels and other biore-newables, a separate Energy Title is being consid-ered for the 2007 Farm Bill. We need to carefullyweigh the added benefits and costs of such govern-ment intervention. We have conducted an interesting

social experiment over the last three decades withethanol. The federal government and some stategovernments as well have provided incentives toincrease both corn grain ethanol production andconsumption to improve local air quality, reduceGHG emissions, and provide a substitute fuel fromrenewable resource that could serve to improveenergy security. The experiment was successful, butin large part because the price of crude oil increased.Pushing this experiment further will eventually havethe added costs outweighing the added benefits.Further expansion is typically justified on groundsof moving to biomass-based ethanol, which is pur-ported to have even large environmental and devel-opmental benefits. However, economic analysisraises serious questions about the potential of bio-mass ethanol as a sustainable biofuel source.

Notes

This policy brief draws on the author’s longer paper,“Biofuel Incentives and the Energy Title of the 2007 FarmBill,” available at www.aei.org/farmbill. The longer paperincludes background support and citations not includedhere. Miranowski is professor of economics and director,Institute of Science and Society, Iowa State University.The views expressed here are those of the author and notof any institution with which he is affiliated.

1. U.S. Environmental Protection Agency. RegulatoryImpact Analysis: Renewable Fuel Standard Program,EPA420-R-07-004, April 2007. http://www.epa.gov/otaq/renewablefuels/420r07004.pdf.

2. John Miranowski and Jittinan Aukayanagul. “ShortRun Econometric Model of the Ethanol Market,”Unpublished working paper, Iowa State University,2007.

3. The corn-price increases caused by the ethanol sub-sidy have other consequences as well, namely, highercorn costs to livestock producers and other buyers andhigher returns to owners of land suitable for corn pro-duction. But these effects largely cancel out, the gains ofone side of the market equaling the losses of the other, sothey do not affect the benefit-cost calculations. They donot exactly cancel out because, with the subsidy in place,the marginal cost of ethanol is greater than the marginalvalue of ethanol to buyers, by the amount of the subsidy. This implies a deadweight loss triangle whose area is 1/2*51 cents/gallon *million gallons = $X million per year.

BIOFUEL INCENTIVES AND THE ENERGY TITLE OF THE 2007 FARM BILL

125

4. John Miranowski and Leandro Andrian,“Community impacts of local investment in biofuel plants.” Unpublished working paper, Iowa State Univer-sity, 2007.

5. John Miranowski and Mark Imerman. “Changingstructure of the ethanol industry and implications forfuture rural development.” Unpublished working paper,Iowa State University, 2007.


126

127

RURAL DEVELOPMENT

Farm employment is not the long-run economic base ofrural America, in the sense that it does not lead ordetermine the rest of employment in rural America.Over the past two centuries the farm share of U.S.population has fallen from 70 percent to 2 percent,and the share that is rural has fallen from 90 percentto 25 percent, but the share of the population that isrural but not farmers has stayed remarkably stable atabout 22 percent (figure 1 on the following page).Modern studies have found, if anything, negative rela-tionships between dependence on farm sector supportand rural development (Goetz and Debertin, 1996;Drabenstott, 2005). The negative externalities ofurban congestion, high costs of urban space, andattractive rural amenities (e.g., Deller et al., 2001), arethe modern determinants of rural development in theUnited States.

The piecemeal process of agricultural adjust-ment has been successful. The geographic and sec-toral mobility of farmers has been facilitated by thefact that farmland has been a good asset and it isrelatively easy to liquidate. Increased farm size andproductivity growth in the agricultural sector havebeen rewarded by the market. Households whoremain in the farm sector now have higher incomeson average, $81,000/year, compared to averageU.S. household income of around $60,000(USDA/ERS, 2006a).

In contrast, the process of rural adjustment to the changes in transportation and production tech-nologies that render the smallest and most remote(rural) communities redundant continues. Figure 2,on the following page, shows the spatial gradient innet domestic migration. Cities grow bigger at thefastest rates. Non-metro towns adjacent to metroareas also grow at positive rates. Both are growing theexpense of small towns that are not adjacent to cities.This is spatial rationalization.

Main Policy Issue

A segment of our population bears the costs of ruraladjustment to spatial rationalization. Unlike farmers;non-farm rural home and business owners do nothave liquidatable assets. Geographic mobility is not anindividually rational option. Even if there are buyers,spatial rationalization is not achieved by replacing oneset of residents and owners in a declining town withanother set. Unlike in the farm sector, where reduc-tions in the number of farmers/increases in farm sizereward remaining farmers with higher incomes, reduc-tions in the number of rural citizens per town penalizethe remaining rural residents and business owners.

Numerous market failures exist and justify rural public good provision, public investment to

129

Rural Development Policy

Maureen R. Kilkenny and Stanley R. Johnson

encourage rural adjustment, and publicly fundedcompensation for the losers in the process of spatialrationalization. Outward signs that asset fixity, geo-graphic immobility, and other space-related marketfailures hurt rural people include higher rates of rural

underemployment, missing mar-kets (no buyers for rural property,high rural vacancy rates), factorprice distortions (lower pricespaid for the same factor of pro-duction in rural areas), under-provision of public goods(inequitable access to publicschool education), price distor-tions (rural prices above marginalcosts of production plus deliv-ery), and low rural wages.

And because rural propertyvalues and incomes are low, thereare compounding negative conse-quences. One, the rural tax base islower precisely where the per tax-payer costs of public services andinfrastructure (schools, roads,utilities) are higher. This fact aloneindicates that special effort andintergovernmental transfers torural communities may be neededto ensure every U.S. citizen,regardless of where they are born,has access to public educationthough high school, electricityand communications, clean water,and proper sewage systems.

The other unfortunate conse-quence is that rural householdshave lower financial leverage.As Jane Jacobs (1984) argues,stagflation—low wages and highprices—is a spatial phenomenon.Rural wages and rents are lower,while the prices of tradable goods(automobiles, college education)are the same (or higher) as inurban areas. The relevant conse-

quences of spatial stagflation are that it underminesthe rural tax base and increases the cost of geographicand sectoral mobility for rural people and businesses.Because localization of the money supply process isthe antidote to spatial stagflation, it is important that


130

FIGURE 2AVERAGE NET DOMESTIC MIGRATION RATE

SOURCE: CO-EST2005 estimates, Bureau of Census, 2000–2005 averages.NOTE: Domestic net out-migration rates are highest from small and remote communities.

FIGURE 1U.S. POPULATION SHARES

SOURCE: http://www.census.gov/population/www/documentation/twps0029/tab18.htm.NOTE: Two centuries of change in the share of the U.S. population that farms, but no change inthe share that is rural but does not farm.

21

Urban Farmers Non-Farmers

0

10

20

30

40

50

60

70

80

90

100

1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990

72

2

23

Per

cen

t

-- 0.7%

--1.0%

-- 0.5%

0.0%

0.5%

1.0%

1.5%

> 1

mil

250K

--1 m

il

50--

250K

20--

50K

adj

2.5-

- 20K

adj

<2.

5K a

dj

20--5

0K n

onad

j

2.5-

20K

non

adj

<2.

5K n

onad

j

0.1%

Per

cen

t o

f C

ou

nty

Po

pu

lati

on

in 2

000

funds flow through rather than around commercialbanks in rural areas.

The dilemma for U.S. rural development policyis that keeping declining towns on “life support”delays the inevitable. But mobilizing rural citizenspiecemeal exacerbates the cost of that “life support”by increasing the rate of decline. The current mixof “sector-,” “place-,” and “people-based” policiesmay even be making rural adjustment morepainful. Each person who exits a small town leavesremaining schools and businesses another studentor customer farther below minimum efficient scale.Each taxpayer who exits a small town increases the costs of local public good provision borne bythe remaining residents. Piecemeal mobility ofrural people reduces rural property values, damp-ens economic opportunity, and worsens thedependence of rural communities on intergovern-mental funds.

“Place-based” policies are those for which thelocation or spatial category of the beneficiary is a keycriterion for eligibility. Six commonly recognized pit-falls and shortcomings of place-based policies arethat they may (i) generate nothing but rents for theowners (potentially absentee) of property in targetedplaces, (ii) attract or retain (trap) poor people in poorareas, (iii) distort business as well as human migra-tion decisions, (iv) enable the postponement of nec-essary adjustments, (v) create dependencies, and are(vi) subject to abuse by place-based politicians(Kilkenny and Kraybill, 2004).

Nationwide policies also have at least two notableshortcomings. First, because of spatial heterogeneity,spatially uniform/economy-wide or “pan-territorial”sector-based or people-based policies do not neces-sarily have spatially neutral effects (Hurter andMartinich, 1989; Kilkenny and Huffman, 2003;Blank, 2005). Second, nationwide spatially neutralpolicies may help mobilize people out of low-income, low-vitality rural areas; but in doing so, theypush rural communities further below critical mass.In sum, nationwide and people-based policies havespatially differentiated effects, are likely to be ineffi-cient, can be excessively costly, and can make ruralproblems even worse.

What do we want from rural development policy?The USDA Rural Development Agency’s strategic plansays it nicely: “A rural America that is a healthy, safe, andprosperous place to live and work.” Farm policies helpedagriculture adjust to having fewer, larger, more pros-perous farms; run by a declining number of farmers.Rural development policy should help rural Americaadjust to having fewer, larger, more prosperous ruralcommunities. One big difference is that rural adjust-ment cannot be accomplished one household at atime. Another big difference is that the number of ruralcitizens consistent with spatial rationalization has, andshould continue to, increase.

Existing rural development programs emphasizecategorical eligibility and equal access. Any and everyrural town may obtain federal funds to help secure aminimum quality of life. Because budgets are limitedand because not all rural places are economicallyviable, attempting to sustain all rural communities isdone at the expense of achieving critical mass andsustainability in most rural communities. To leveragethe development of potentially sustainble rural com-munities we recommend the use of place-tailoredmatching funding rather than place-based categoricalfunding.

We suggest five principles for the 2007 Farm Billrural development programs. In order of increasingdifficulty of implementation, they are (1) take thelead for more interagency cooperation, (2) movefunds through, not around, the commercial bankingsystem, (3) initiate a modulation process, (4) leverageown-effort by emphasizing matching funding, and(5) facilitate rural adjustment by mobilizing wholecommunities.

The first principle is about interagency specializa-tion and trade (not nationwide program consolida-tion). Our country prefers the federations of towns,counties, districts, states, and national governmentcomprised of many specialized agencies over a singletop-down central government because pan-territorial,one-size-fits-all programs to provide public goods areless efficient than locally tailored programs. To lead,specialize, and trade there are at least four things theUSDA RD Agency should be doing: (i) coordinateand specialize, (ii) minimize redundancies, (iii) share

RURAL DEVELOPMENT POLICY

131

or merge some operations, and (iv) watchdog toensure equal opportunity for rural constituencies inother agencies’ programs.

The second principle of working through the com-mercial banking system is another no-cost recommen-dation. The USDA RD has already been expandingguaranteed loan program levels while reducing grantand direct loan program levels. This trend should con-tinue. In particular, we encourage partnering withrural commercial banks rather than subsidizing non-bank competitors of rural banks. When funds flowthrough the rural offices of banks they will augmentrural liquidity and will help avoid rural stagflation.

Also, commercial banks have much at stake regard-ing the mechanism of farm support. Farm loans arebacked by farm land collateral, at values inflated byfarm support programs. If those programs were termi-nated and farm land values collapsed, a process ofmultiple deposit contraction would occur. That hasthe potential to undermine the solvency of theMidwestern banking system and disrupt the country’smoney supply, not to mention the local money sup-plies in rural farm-belt communities.

The sensitivity of rural bank assets to farm sup-port policy motivates our third principle for the2007 farm bill: initiate a modulation process to sup-port an orderly transition to a market-oriented farmsector without strangling rural liquidity. Additionalpressure to reduce direct support to farmers comesfrom the World Trade Organization negotiations.Our suggestion is to mandate a bipartisan specialtemporary commission involving executive and leg-islative branch representatives and experts fromwithin and outside of government to figure out howto maintain rural liquidity even as subsidy-inflatedreturns to farmland are reduced.

Our fourth principle, leverage own effort, mergesa popular (private sector) place-competitivenessstrategy with federal strategy of the provision of pub-lic goods and rural quality of life. We suggest thatthis be done by requiring places to compete for fed-eral funding, and that disbursements leverage localfunding, not substitute for it. We also recommendproviding incentives to communities that plan, espe-cially those that coordinate consolidations with other

rural communities, and communities that makeprogress on their plans.

The fifth principle is to enable community-widespatial relocation to avoid piecemeal rural out-migration and chronic decline. The house-holds and businesses in rural towns that cannotmaintain even a portion of their infrastructureshould be eligible for community-wide relocationassistance simultaneously. Consolidations of two ormore towns into sustainable-size towns, and con-versions of rural satellite towns into suburbs ofcities, should be especially encouraged.

Spatial rationalization can be accomplished in avariety of ways. The hardest is piecemeal: householdby household and business by business throughrural out-migration, relocations, closures, or death.The easiest but lowest payoff way is jurisdictionalconsolidation. Jurisdictional consolidation does notrequire households or businesses to move, just anenlargement of jurisdictional boundaries to encom-pass the merging communities. Jurisdictional merg-ing reduces local public sector duplications. Itreleases local public revenues for other purposes,such as the match to be eligible for federal fundingof infrastructure if required.

Real spatial rationalization is achieved by phys-ical relocations of households and businesses.Towns in flood zones, for example, are relocated.The cost per household is the price of a compara-ble home in the destination small town. Instead ofpiecemeal decline, numerous declining communi-ties could be reorganized into sustainable newones. Instead of sinking funds again and again oninfrastructure, spend the funds once and for all inways that help citizens achieve a higher quality oflife elsewhere while preserving each communityand their legacies. Funds saved on infrastructureshould be allocated to pay for the relocations.Landmark buildings such as a school, church,town hall, or other building of sentimental valueshould accompany the bulk of the relocated popu-lation. Also, the merged town could be renamedwith a hyphenated version of both towns’ names,in the same way that children in Spain carry thesurnames of both parents.


132

Conclusion

Rural development policy can reduce the downsiderisks and increase the benefits of spatial rationalization.Rural development policy can also stop subsidizingspatial redundancy. If rural development policy doesboth, it will increase economic opportunity andimprove the quality of life for all rural Americans at afaster rate and lower cost to taxpayers.

Note

This policy brief draws on the authors’ longer paper “RuralDevelopment Policy,” available at www.aei.org/farmbill. Thelonger paper includes background support and citations notincluded here. Maureen Kilkenny is professor of resourceeconomics, and Stanley Johnson is distinguished professorEmeritus, Iowa State University. Both are currentlyemployed in the College of Agriculture, Biotechnology andNatural Resources, the University of Nevada, Reno. Theviews expressed here are those of the authors and not thoseof any institution with which they are affiliated.

References

Blank, Rebecca. 2005. “Poverty, Policy, and Place: How

Poverty and Policies to Alleviate Poverty Are Shaped by

Local Characteristics,” International Regional Science Review

28(4):441–464.

Cowan, Tadlock. 2006. An Overview of USDA Rural

Development Programs Congressional Research

Service (CRS) Report for Congress, Washington, D.C.;

June 22.

Deller, Steve, Tsung-Hsui Tsai, David Marcoullier, and

Donald English. 2001. “The Role of Amenities and Quality

of Life in Rural Economic Growth,” American Journal of

Agricultural Economics 83(2):352–365.

Dorr, Thomas, USDA Undersecretary for Rural Develop-

ment. 2003. “USDA Programs make our rural regions

stronger.” Iowa View Editorial. The Des Moines Register

July 25.

Dranbenstott, Mark. 2005. “Do Farm Payments Promote

Rural Economic Growth?” The Main Street Economist:

Commentary on the Rural Economy, Center for the Study of

Rural America, Federal Reserve Bank of Kansas City,

Kansas City, MO.

Goetz, Stephan, and David Debertin. 1996. “Rural Population

Decline in the 1980s: Impacts of Farm Structure and

Federal Farm Programs,” American Journal of Agricultural

Economics, 78(3): 517–29.

Government Accounting Office. 2005 “21st Century

Challenges: Reexamining the Base of the Federal Govern-

ment,” GAO-05-325SP. Washington, D.C.

Jacobs, Jane. 1984. Cities and the Wealth of Nations: Principles

of Economic Life. New York: Random House.

Kilkenny, Maureen. 1998. “Transport Costs and Rural

Development,” Journal of Regional Science 38(2) 293–312.

Kilkenny, Maureen, and David Kraybill. 2003. “The

Economic Rationales for and against Place-Based Policies,”

Paper presented at the AAEA-RSS Annual Meetings, July

27–30, Montreal, Canada.

Kilkenny, Maureen, and Sonya Huffmann. 2003.

“Rural/Urban Welfare Program and Labor Force

Participation,” American Journal of Agricultural Economics,

85(4): 914–27.

U. S. Department of Agriculture, Economic Research Service.

2006a. “Farm Household Economics and Well-Being:

Farm Operator Household Income in Perspective,” http://

www.ers.usda.gov/Briefing/WellBeing/incomein

perspective.htm¬#farm Updated February 8, 2006 (date

accessed 10/31/06).

———. 2006b. “2007 Farm Bill Theme Papers: Rural

Development,” www.usda.gov/documents/Farmbill07

ruraldevelopment.pdf (accessed 8/3/2006).

RURAL DEVELOPMENT POLICY

133

Farm bills dating back to the 1970s have committedfederal resources to various types of infrastructureinvestment programs targeted to rural communities.Title VI of the 2002 Farm Bill—the “Rural Devel-opment” title—authorized hundreds of millions ofdollars annually for water purification, wastewatertreatment, and solid waste management in ruraltowns and counties. Between 2002 and 2006, totalfederal outlays on water treatment and waste disposalamounted to $3.3 billion. More than 93 percent ofthis money took the form of grants; the balance wasdistributed as loan subsidies. Money was also author-ized for investment in rural telecommunicationscapacity, including programs related to broadbandaccess, local television access, distance learning, ande-commerce. Since 2004 the Rural Utilities Servicehas made $1 billion in loans (at subsidized rates) totelecommunications providers, primarily for broad-band deployment.

How do the returns to rural development pro-grams stack up against their costs? Unfortunately,economic studies that provide estimates of netbenefits are not available in the literature, mainlybecause the data needed to carry out such analysis—particularly the counterfactual of what would havehappened absent the public investment—have notbeen developed. We can consider, however, theextent to which infrastructure investments authorized

within the Rural Development title actually can belinked to various manifestations of what is com-monly, if imprecisely, termed rural development. Thispolicy brief summarizes the evidence on whether ornot such links exist.

Specifically, it concentrates on infrastructureinvestments made in small, relatively remote commu-nities having populations less than 20,000—the typesof communities that are targeted by the programs thathave been historically funded under the RuralDevelopment title. Development here will be taken toencompass sustained positive changes in the magni-tude and composition of economic activity withinthose communities, the amenities that they possess,and the range or quantity of government servicesmade available to their residents.

Two types of federal infrastructure investmentshave received the most discussion in the context of the Rural Development title—water treatmentinfrastructure and telecommunications infrastruc-ture. Neither type of investment appears to offermuch in the way of a “pump priming” stimulus tolocal economic activity. In both cases, the prospec-tive payoffs to infrastructure investment in ruralareas vie with the inherent difficulty in recoupingtheir large fixed costs in locations with low popula-tion densities. Moreover, the net benefits of thesesorts of investments are often substantially related

135

Infrastructure Investment and Rural Development

Mitch Renkow

to proximity of rural communities to urban employ-ment centers.

With respect to water treatment investments, itappears difficult to justify federal involvement ongrounds that such investments counter the unfundedfederal mandates with respect to drinking water andwastewater effluent standards. Strong evidence of fiscal substitution effects and local willingness to payfor improved water quality, coupled with the fact that the public goods generated by those investmentsare generally consumed locally, call into question the advisability of federal funding for these types of investments.

On the other hand, federal funding of broadbanddeployment into remote areas may be justified on thegrounds that access to up-to-date telecommunica-tions activity is an inherent “citizenship right” thatshould be afforded to all. However, this justificationrequires an assessment that the positive social netbenefits associated with universal (or at least, near-universal) provision of high-speed Internet access toremotely located citizens outweigh its high costs.

Water Infrastructure

The importance to different types of businesses ofthe availability and cost of water and sewer servicesis highly variable. Some types of industrial processesare more dependent on adequate supplies of waterthan others, and the lack of a local water system ina given rural community may well preclude locationof firms that utilize those processes intensively. Whatlittle research has been conducted on the economicimpacts of rural water and sewer systems suggeststhat their net benefits tend to be much more modestin rural communities than they are in more denselypopulated urban communities. This is fundamen-tally due to the substantial economies of scale anddensity in plant construction and deployment ofwater and sewer lines.

Evidence linking the availability of water treat-ment infrastructure in rural communities to thosecommunities’ ability to attract new businesses isweak. While some empirical research connects some

firms’ location decisions to the availability and qualityof local water supplies, most analysts of firm locationfind that other factors—particularly the quality andsize of the local labor force, the existing road network,and the nature of the local telecommunicationsinfrastructure—are much more important. More-over, the higher unit costs of water treatment in(smaller) rural systems generally translate to higherunit prices for water customers; this in turn tends tolower a community’s attractiveness to new firmsconsidering relocation. These higher water chargesattenuate the positive effects offered by greater avail-ability of local water treatment capacity.

For many rural communities, changes in residen-tial development patterns may well be the mostprofound impact brought about by augmentation oflocal water treatment capacity. This is especially thecase for rural communities located in reasonably closeproximity to urban areas where deployment of waterinfrastructure facilitates exurbanization trends cur-rently observed at the urban-rural fringe. In terms oflocal economic development, local businesses—particularly those in construction, retail, and servicesectors—generally will be positively affected byincreases in demands for their goods and servicesarising from population growth. On the other hand,residential development associated with exurbaniza-tion often leads to substantial fiscal pressures on localgovernments (and taxpayers) as they pay for theincreased level of public services associated with resi-dential development.

Unfunded Mandates. Stringent federal treatmentstandards for both drinking water and wastewatereffluent, in conjunction with devolution of fiscal,regulatory, and enforcement responsibilities to stateand local governments, represent an important setof unfunded federal mandates facing local commu-nities. Rural Development title funds channeled torural communities for drinking water treatment andwastewater treatment would thus appear to repre-sent a counterbalancing cushion to the costs ofunfunded federal mandates. Just how large thatcushion is would of course vary widely acrosscommunities, system types, and the extent to which


136

federal requirements raise costs above what thecommunity would have incurred absent thoserequirements.

However, there are several reasons why federal aidto water infrastructure investment might represent a less-than-optimal use of public resources. First,there is evidence that greater local cost shares in con-struction of water treatment facilities leads to moreeconomical choices—simpler treatment technologies,more careful oversight of project costs, and shorterconstruction periods. Second, when water treatmentis funded locally, communities have considerablystronger incentives to promote water conservation inorder to delay the installation of additional treatmentcapacity. Third, in many circumstances, fiscal substi-tution effects will prevail such that each dollar offederal funds will not result in an additional dollar oftotal expenditure on some socially desirable outcome(such as safe drinking water or suitably treated waste-water discharges). That is, federal funds may simplydisplace resources that local governments would havespent to achieve those outcomes, even in the absenceof federal mandates. Finally, to the extent that thepublic goods created by federal investment in ruralcommunities’ water treatment capacity are consumedlocally, one can reasonably question those investmentson efficiency grounds. A number of studies indicatesubstantial local willingness to pay for water qualityimprovements. This suggests that in many cases localfinancing of water treatment infrastructure will be fea-sible, in addition to being more efficient due to thebeneficiaries of the services provided by the infrastruc-ture paying for the generation of those services.

Benefits versus Costs. According to Office ofManagement and Budget data, annual expenditureson water treatment and wastewater disposal systemsmade through the Rural Community AdvancementProgram during the period 2002–2006 averaged$660 million. Over 90 percent of these funds werein the form of grants. Two factors are important indetermining whether and when such investmentsare likely to pass a reasonable benefit/cost test.

First, the ability of water infrastructure invest-ments to provide a strong “pump priming” stimulus

to local economies is often limited in the sense thatthe services provided by infrastructure may in manycases be necessary for sustained economic growth,but they are probably never sufficient. This meansthat assessing relative benefits will depend heavilyon environmental services and associated amenitiesassociated with water treatment. But even if the netbenefits of those environmental services outweighthe costs associated with their provision, it remainshighly debatable whether federal underwriting ofthose investments is sound public policy. The ampleevidence of fiscal substitution effects and local will-ingness to pay for improved water quality, coupledwith the fact that the public goods generated bythose investments are generally consumed locally,call into question the advisability of federal fundingfor these types of investments on both efficiency andequity grounds.

Second, economies of scale in plant constructionand economies of density in the installation of water lines mean that benefits are more likely to outweigh costs in areas in which the level anddensity of populations served are highest. For thisreason, investments in water treatment infrastruc-ture in rural communities located within the com-mutershed of expanding urban labor markets arelikely to yield the highest net returns. On the otherhand, the probability of costs outweighing benefitsfor any particular project will be greatest in the most remote and least populous communities—which are of course precisely the kinds of commu-nities that Rural Development title programs aredesigned to serve.

Telecommunications Infrastructure

Properly implemented telecommunications capacityreduces the cost of physical distance to consumersand to businesses in the communities served.Hence, telecommunications investments in ruralareas can yield substantial positive impacts on theproductivity and profitability of existing firms, aswell as on the location decisions of potential newfirms. And rural consumers stand to benefit from

INFRASTRUCTURE INVESTMENT AND RURAL DEVELOPMENT

137

the broader array of goods and services choicesavailable to them in an electronic marketplace.

In striking contrast to the situation that existed atthe end of the twentieth century, rural communitiesand rural dwellers are now connected to the internetto nearly the same degree as their urban counterparts.Where a rural-urban digital divide persists, however,is in access to high-speed data transmission throughbroadband connectivity. Despite some narrowing ofthis gap over the past few years, access to broadbandcontinues to lag in rural areas—especially in sparselypopulated rural areas and remote communitieslocated far from urban centers.

The benefits of broadband over dial-up connectiv-ity are well chronicled. By reducing the effective costof distance, the high-speed data transmission enabledby broadband has significant potential for ameliorat-ing key constraints on economic performance of busi-nesses in rural places—small markets, high transportcosts, and physical isolation. E-commerce, healthcare, and education are areas in which broadbandoffers significant potential benefits to rural residents.

Ironically, precisely the same physical remotenessand low population densities that make broadbandparticularly desirable in rural areas also render itsdeployment expensive. All broadband technologiesrequire very large up-front financial outlays by serviceproviders. For this reason, rural small businessesgenerally pay more than do their urban counterpartsfor high-speed internet access. Moreover, demand-side constraints have proven to be important factorslimiting deployment of broadband in rural areas. Thisis because broadband carriers generally require aminimum number of customers before they will offerservice in an area.

Benefits versus Costs. Since 2004 the RuralUtilities Service has made $1 billion in loans (at sub-sidized rates) to telecommunications providers,primarily for broadband deployment. The rates ofreturn on broadband deployment are inverselyrelated to the density and remoteness of populationsserved. This, of course, limits the effective rate ofreturn on telecommunications investments madeunder the auspices of the Rural Development title,

since those investments are specifically targeted tomore remote and less populated areas.

Broadband availability appears likely to expandmore rapidly into rural communities located nearer to urban areas than into more remote locations. Thelion’s share of the narrowing of rural-urban gaps in access to high-speed data transmission that hastaken place in recent years is accounted for by deploy-ment of DSL and cable service in rural communitieslocated near urban centers. However, some level of rural-urban digital divide may in fact be sociallydesirable—i.e., that the costs of bringing broadbandto particularly remote areas may greatly exceed thebenefits that universal service would bring.

On the other hand, there is substantial historicalprecedent for federal investments in communica-tions infrastructure that provide close to universalaccess to a dominant mode of communication insociety (such as mail service or telephone service).Presumably, such investments reflected an assess-ment by policymakers that the public goods createdby deployment of these sorts of integrative infra-structure were sufficiently large to outweigh therelatively steep costs of their providing those infra-structure services to remote consumers. The desir-ability of continued or expanded federal funding of programs promoting broadband deployment intoremote rural areas depends on the extent to which acomparable assessment of these positive social exter-nalities exists today.

Conclusion

Net payoffs to investments in both water treatmentand telecommunications infrastructure are related tothe size and density of the population served. Thisrepresents a formidable barrier to those sorts ofinvestments passing a reasonable cost/benefit test inthe remote and less-populated areas targeted by fed-eral programs under the Rural Development title ofthe Farm Bill. With respect to water treatmentinvestments, it further appears difficult to justifyfederal involvement on grounds that such invest-ments counter the unfunded mandates attributable


138

to stringent federal treatment standards for both drink-ing water and wastewater effluent. Strong evidence offiscal substitution effects and local willingness to payfor improved water quality, coupled with the fact thatthe public goods generated by those investments aregenerally consumed locally, call into question theadvisability of federal funding for these types of invest-ments on both efficiency and equity grounds. In short,federal underwriting of those investments appears tobe unsound public policy.

On the other hand, federal funding of broadbanddeployment into remote rural areas may be justifiedon the grounds that access to up-to-date telecom-munications capability is an inherent “citizenshipright” that should be afforded to all. However, thisjustification requires an assessment that the positivesocial benefits and public goods associated with uni-versal (or at least, near-universal) provision of high-speed internet access to remotely located citizensoutweigh its high costs.

INFRASTRUCTURE INVESTMENT AND RURAL DEVELOPMENT

139

Board of TrusteesBruce Kovner, ChairmanChairmanCaxton Associates, LLC

Lee R. Raymond, Vice Chairman

Chairman and CEO, Retired Exxon Mobil Corporation

Tully M. Friedman,TreasurerChairman and CEOFriedman Fleischer & Lowe, LLC

Gordon M. BinderManaging DirectorCoastview Capital, LLC

Harlan CrowChairman and CEOCrow Holdings

Christopher DeMuthPresidentAmerican Enterprise Institute

John V. Faraci Chairman and CEOInternational Paper Company

Christopher B. GalvinChairmanHarrison Street Capital, LLC

Raymond V. GilmartinChairman and CEO, Retired Merck & Co., Inc.

Harvey GolubChairman and CEO, Retired American Express Company

Robert F. GreenhillChairman and CEOGreenhill & Co., Inc.

Roger Hertog

Martin M. KoffelChairman and CEOURS Corporation

Marc S. LipschultzPartnerKohlberg Kravis Roberts & Co.

John A. Luke Jr.Chairman and CEOMeadWestvaco Corporation

L. Ben LytleManaging PartnerRincon Advisors, LLC

Robert A. PritzkerPresident and CEOColson Associates, Inc.

J. Joe RickettsChairman and FounderTD Ameritrade HoldingCorporation

Kevin B. Rollins

Edward B. Rust Jr. Chairman and CEOState Farm Insurance Companies

D. Gideon SearleManaging PartnerThe Serafin Group, LLC

Mel SemblerFounder and ChairmanThe Sembler Company

William S. StavropoulosChairman EmeritusThe Dow Chemical Company

Wilson H. TaylorChairman EmeritusCIGNA Corporation

William H. Walton Managing MemberRockpoint Group, LLC

James Q. WilsonPepperdine University

Emeritus TrusteesWillard C. Butcher

Richard B. Madden

Robert H. Malott

Paul W. McCracken

Paul F. Oreffice

Henry Wendt

OfficersChristopher DeMuthPresident

David GersonExecutive Vice President

Jason BertschVice President, Marketing

Henry Olsen IIIVice President, Director of theNational Research Initiative

Danielle PletkaVice President, Foreign andDefense Policy Studies

Council of AcademicAdvisersJames Q. Wilson,ChairmanPepperdine University

Gertrude HimmelfarbDistinguished Professor ofHistory Emeritus

City University of New York

R. Glenn HubbardDean and Russell L. CarsonProfessor

Graduate School of Business Columbia University

Samuel P. HuntingtonAlbert J. Weatherhead III University Professor ofGovernment

Harvard University

William M. LandesClifton R. Musser Professor ofLaw and Economics

University of Chicago Law School

Sam PeltzmanRalph and Dorothy KellerDistinguished Service Professor of Economics

Graduate School of BusinessUniversity of Chicago

George L. PriestJohn M. Olin Professor of Law and Economics

Yale Law School

Jeremy RabkinProfessor of GovernmentCornell University

Murray L. WeidenbaumMallinckrodt DistinguishedUniversity Professor

Washington University

Richard J. ZeckhauserFrank Plumpton RamseyProfessor of Political Economy

Kennedy School of GovernmentHarvard University

Research StaffGerard AlexanderVisiting Scholar

Joseph AntosWilson H. Taylor Scholar in HealthCare and Retirement Policy

Leon AronResident Scholar

Claude BarfieldResident Scholar

Roger BateResident Fellow

Walter BernsResident Scholar

Douglas J. BesharovJoseph J. and Violet Jacobs Scholar in Social Welfare Studies

Edward BlumVisiting Fellow

Dan BlumenthalResident Fellow

John R. BoltonSenior Fellow

Karlyn BowmanSenior Fellow

John E. CalfeeResident Scholar

Charles W. CalomirisVisiting Scholar

Lynne V. CheneySenior Fellow

Steven J. DavisVisiting Scholar

Mauro De LorenzoResident Fellow

Thomas DonnellyResident Fellow

Nicholas EberstadtHenry Wendt Scholar in PoliticalEconomy

Mark FalcoffResident Scholar Emeritus

John C. FortierResearch Fellow

Ted FrankResident Fellow; Director,AEI Liability Project

David FrumResident Fellow

David GelernterNational Fellow

Reuel Marc GerechtResident Fellow

Newt GingrichSenior Fellow

James K. GlassmanSenior Fellow; Editor-in-Chief, The American magazine

Jack L. GoldsmithVisiting Scholar

Robert A. GoldwinResident Scholar Emeritus

Scott Gottlieb, M.D.Resident Fellow

Kenneth P. GreenResident Scholar

Michael S. GreveJohn G. Searle Scholar

Christopher GriffinResearch Fellow

Robert W. HahnResident Scholar; Executive Director, AEI-Brookings JointCenter for Regulatory Studies

Kevin A. HassettSenior Fellow; Director, Economic Policy Studies

Steven F. HaywardF. K. Weyerhaeuser Fellow

Robert B. HelmsResident Scholar

Frederick M. HessResident Scholar; Director,Education Policy Studies

Ayaan Hirsi AliResident Fellow

R. Glenn HubbardVisiting Scholar

Frederick W. KaganResident Scholar

Leon R. Kass, M.D.Hertog Fellow

Herbert G. KleinNational Fellow

Marvin H. KostersResident Scholar Emeritus

Irving KristolSenior Fellow Emeritus

Desmond LachmanResident Fellow

Michael A. LedeenFreedom Scholar

Adam LerrickVisiting Scholar

Philip I. Levy Resident Scholar

James R. LilleySenior Fellow

Lawrence B. LindseyVisiting Scholar

John H. MakinVisiting Scholar

N. Gregory MankiwVisiting Scholar

Aparna MathurResearch Fellow

Mark B. McClellan, M.D. Visiting Senior Fellow, HealthPolicy Studies and AEI-Brookings Joint Center forRegulatory Studies

Allan H. MeltzerVisiting Scholar

Thomas P. MillerResident Fellow

Joshua MuravchikResident Scholar

Charles MurrayW. H. Brady Scholar

Roger F. NoriegaVisiting Fellow

Michael NovakGeorge Frederick Jewett Scholar in Religion, Philosophy, and Public Policy

Norman J. OrnsteinResident Scholar

Richard PerleResident Fellow

Alex J. PollockResident Fellow

Sarath RajapatiranaVisiting Fellow

Michael RubinResident Scholar

Sally Satel, M.D.Resident Scholar

Gary J. SchmittResident Scholar; Director, AEI’s Program on AdvancedStrategic Studies

Joel SchwartzVisiting Fellow

Daniel ShaviroVisiting Scholar

Kent SmettersVisiting Scholar

Christina Hoff SommersResident Scholar

Samuel ThernstromManaging Editor, AEI Press;Director, W. H. Brady Program

Bill ThomasVisiting Fellow

Fred ThompsonVisiting Fellow

Richard VedderVisiting Scholar

Alan D. ViardResident Scholar

Peter J. WallisonSenior Fellow

Ben J. WattenbergSenior Fellow

John YooVisiting Scholar

The American Enterprise Institute for Public Policy ResearchFounded in 1943, AEI is a nonpartisan, nonprofit research and educationalorganization based in Washington, D.C. The Institute sponsors research,conducts seminars and conferences, and publishes books and periodicals.

AEI’s research is carried out under three major programs: Economic PolicyStudies; Foreign Policy and Defense Studies; and Social and Political Studies.The resident scholars and fellows listed in these pages are part of a networkthat also includes ninety adjunct scholars at leading universities throughoutthe United States and in several foreign countries.

The views expressed in AEI publications are those of the authors and do not necessarily reflect the views of the staff, advisory panels, officers, or trustees.

AEI Agricultural Policy Series The 2007 Farm Bill and Beyond · 2019-01-18 · THE 2007 FARM BILL...

Documents

Transcript of AEI Agricultural Policy Series The 2007 Farm Bill and Beyond · 2019-01-18 · THE 2007 FARM BILL...