• E C O N O M I C A N D B U D G E T I S S U E B R I E F

E C O N O M I C A N D B U D G E T I S S U E B R I E F
CBO
A series of issue summaries from
the Congressional Budget Office
NOVEMBER 20, 2006
Agricultural Trade Liberalization
Summary
Since the end of World War II, the United States and major developed countries and most developing coun-
other countries have benefited from agreements reduc- tries, would also see a net economic benefit.
ing tariffs and subsidies that distort international trade
in manufactured goods. Similar liberalization for agri- If all policies worldwide that distort agricultural trade
cultural products would provide additional benefits to were phased out in this decade, the likely total annual
U.S. consumers and producers and to taxpayers by economic benefit to the world by 2015 would be
reducing federal spending. roughly $50 billion to $185 billion, which is about 3
percent to 13 percent of the value added by world agri-
U.S. policies that distort agricultural trade are modest culture. Those estimates account for only the efficiency
by international standards: tariffs and export subsidy gains and increased investment resulting from liberal-
rates are quite low on average, while rates of trade- ization. In studies that also incorporate effects on pro-
distorting domestic subsidies are somewhat higher. ductivity growth rates, the benefits are 50 percent to
Those facts suggest that U.S. agriculture has more to more than 100 percent larger.
gain from liberalization in terms of increased exports
than it has to lose in terms of increased imports. Roughly 80 percent to 90 percent of the benefit from
liberalization would come from eliminating tariffs
According to studies surveyed by the Congressional and similar import restrictions. However, even rela-
Budget Office, the United States and, to a lesser ex- tively modest exceptions for selected tariffs could sub-
tent, its agricultural sector would benefit economically stantially reduce the benefits that would otherwise be
from reductions in agricultural tariffs and subsidies realized.
worldwide. Many other countries, including all
Over the past 50 years, the United States and many other The Congressional Budget Office (CBO) has published
countries have concluded a series of multilateral trade several papers assessing current policies that distort world
agreements that have improved the economic well-being agricultural trade and surveying studies that examine the
of their residents. For the most part, the agreements have effects of reducing or eliminating such policies.1 This is-
reduced tariffs and other policies that distort trade in sue brief summarizes the conclusions of those papers.
manufactured products. Although removing the remain-
ing barriers to such trade has the potential to produce ad-
ditional benefits, trade barriers in agricultural products
are higher than those for manufactured goods and present
a greater opportunity for future gains. That opportunity
motivated member countries of the World Trade Organi-
1. See Congressional Budget Office, The Effects of Liberalizing World
zation (WTO) to initiate the Doha Round of multilateral Agricultural Trade: A Review of Modeling Studies (June 2006); The
trade negotiations, and although those talks have been Effects of Liberalizing World Agricultural Trade: A Survey (Decem-
suspended, liberalizing trade in agricultural products is ber 2005); and Policies That Distort World Agricultural Trade: Prev-
likely to be a priority in the future. alence and Magnitude (August 2005).
 E C O N O M I C A N D B U D G E T I S S U E B R I E F
2 CONGRESSIONAL BUDGET OFFICE
Policies That Distort World Figure 1.
Agricultural Trade Average Tariff Rates for
In the rubric of international trade negotiations and the Agriculture and Processed
talks concerning liberalization, policies that distort agri-
cultural trade fall into three major categories: market ac-
Foods, 2005
cess, which refers to policies such as tariffs and tariff-rate (Percent)
quotas that regulate the access of imports into a country’s Korea and Taiwan
domestic market; domestic support, which refers to vari- India
ous forms of assistance to domestic producers, such as Vietnam
Japan
production subsidies and price supports that raise the Thailand
price of agricultural products; and export subsidies Turkey
(sometimes called export competition). European Union and EFTA
Russia
Middle East and North Africa
Market Access Bangladesh
Tariffs on agricultural goods remain substantially higher Rest of Latin America and Caribbean
than those on manufactured goods almost everywhere China
Mexico
around the world. According to a 2006 World Bank Canada
study, the average agricultural tariff for the world as a South Africa
whole in 2001 was more than three times the average for Argentina
all merchandise trade (16.7 percent versus 5.2 percent).2 Brazil
Indonesia
For high-income countries, the ratio was even greater at Australia and New Zealand
5.5 (16 percent versus 2.9 percent). Middle- and low- United States
income countries had lower ratios at 1.9 and 1.4 (16.5
0 10 20 30 40 50 60
percent versus 8.9 percent, and 22.2 percent versus 15.9
percent), respectively. Only a few countries, most of them Source: Kym Anderson, Will Martin, and Dominique van der
major agricultural exporters, had agricultural tariffs as Mensbrugghe, “Market and Welfare Implications of Doha
Reform Scenarios,” in Kym Anderson and Will Martin,
low as their average tariffs for all merchandise trade. For
eds., Agricultural Trade Reform and the Doha Develop-
the United States, tariffs protecting agriculture were, on ment Agenda (New York: Palgrave Macmillan and the
average, 1.3 times those for all merchandise. World Bank, 2006), Table 12.3, p. 345.
Notes: Tariff-rate averages are weighted by imports.
U.S. agricultural tariffs—along with those of Australia EFTA = European Free Trade Association.
and New Zealand—are generally low in comparison with
those of other countries (see Figure 1). According to that In many countries, very high tariffs—much higher
same World Bank study, the United States’ average rate than the average for all agricultural products—and
for agriculture and processed foods in 2005 was 2.4 per- tariff-rate quotas (which generally have substantial tariffs
cent. Mexico’s and Canada’s averages were 10.3 percent for imports that exceed the quota) on a few selected prod-
and 9.0 percent, respectively. By comparison, the average ucts are a significant feature of agricultural protection.
for the 25 member countries of the European Union For example, according to another World Bank study, the
(EU) coupled with the four member countries of the
European Free Trade Association (EFTA) was 13.9 per-
cent.3 Most developing countries had averages that were
higher still. 3. The European Union is effectively one country for purposes of
international trade. There is free trade among its members, its
members have a common trade policy toward the rest of the
2. Kym Anderson, Will Martin, and Dominique van der Mensbrug-
world, and the union itself (rather than its individual member
ghe, “Market and Welfare Implications of Doha Reform Scenar-
countries) is a member of the World Trade Organization.
ios,” in Kym Anderson and Will Martin, eds., Agricultural Trade
Reform and the Doha Development Agenda (New York: Palgrave The European Free Trade Association consists of Iceland, Norway,
Macmillan and the World Bank, 2006). Switzerland, and Liechtenstein.
E C O N O M I C A N D B U D G E T I S S U E B R I E F
AGRICULTURAL TRADE LIBERALIZATION 3
highest agricultural tariff for the EU is 506 percent.4 For Figure 2.
the United States, the comparable number is 350 percent;
and for Korea, 917 percent. Less extreme but still signifi- Average Annual Rates of Reported
cant are the figures for Japan and Brazil, which are 50 Domestic Support, 1998 to 2005
percent and 55 percent, respectively. Although relatively (Percentage of the value of agricultural output)
few in number, very high tariffs and tariff-rate quotas in-
European Union
sulate a substantial portion of agricultural production
Japan
from international competition. Fifty percent of Eastern
United States
European production is protected by tariff-rate quotas, as
Korea
is 39 percent of EU production and 26 percent of U.S. Canada
production. On average, 28 percent of agricultural pro- Taiwan
duction in the major industrialized countries of the world South Africa
is protected in that way.5 Mexico
Australia
Domestic Support China
Total Support
Domestic farm subsidies are pervasive around the world. Chile
Of the 80 countries reporting on their policies to the New Zealand Support Limited by
World Trade Organization, 68 indicated that they Bangladesh Reduction Commitments
granted subsidies of some kind in at least one of the years Turkey
from 1998 through 2005. Subsidies of all kinds reported
0 5 10 15 20 25 30 35 40
to the WTO total more than $200 billion per year, or
roughly one-sixth of the $1.2 trillion total value added in Source: Congressional Budget Office based on subsidy data
the agricultural sector worldwide. reported to the World Trade Organization (WTO) by the
countries in question as of July 31, 2006, and dollar-
A few countries dominate the total dollar value of subsi- denominated value-of-production data from Producer and
Consumer Support Estimates, OECD Database 1986-2003,
dies granted. The EU and the United States grant about on the Web site of the Organisation for Economic Co-
one-third of the world total each—the EU somewhat operation and Development, and exchange-rate data from
more than the United States because its agricultural sec- International Monetary Fund, International Financial Sta-
tor is slightly larger—and Japan grants almost 12 percent. tistics.
In contrast, Australia, a major agricultural producer, Notes: “Support limited by reduction commitments” refers to the
accounts for less than one-half of 1 percent of the world category of trade-distorting support that was limited and
total. that countries made commitments to reduce in the Uruguay
Round Agreement on Agriculture. It is often referred to as
amber-box support.
The EU and the United States have such large subsidies
in part because they and their agricultural sectors are Some countries’ most recent reports of their subsidies are
for some year earlier than 2005. Of particular interest, the
large. Even a small subsidy rate can result in a large total most recent reports by the European Union and the United
subsidy when it is applied to a large output. Ranking States are for 2001. That year is before the European Union
countries by subsidy rates paints a somewhat different expanded from 15 member countries to 25.
picture (see Figure 2). The EU, Japan, and the United Some small developed countries have substantially higher
States have had subsidies averaging 37 percent of their rates of domestic support for their agricultural sectors than
any of the countries shown in the figure.
4. World Bank, Global Economic Prospects: Realizing the Development
Promise of the Doha Agenda, 2004 (Washington, D.C.: World respective agricultural outputs since 1998. The rates for
Bank, 2003). Mexico and Canada are 18 percent and 13 percent,
respectively. Australia and New Zealand have averaged
5. World Bank, Global Economic Prospects. The figure for the major
industrialized countries is for the members of the Organisation for 5 percent and 2 percent, respectively.
Economic Co-operation and Development (OECD), which con-
sists primarily (although not exclusively) of the major industrial- The Uruguay Round Agreement, which preceded the
ized countries of the world. current Doha Round and created the WTO in 1995,
 E C O N O M I C A N D B U D G E T I S S U E B R I E F
4 CONGRESSIONAL BUDGET OFFICE
contained the first significant liberalization of agricultural Export Subsidies
trade. That agreement divided domestic agricultural sub- Export subsidies are granted by substantially fewer coun-
sidies into two broad categories depending on whether tries than are domestic subsidies. Under current interna-
they were deemed by the negotiators to significantly dis- tional trade rules, a WTO member country may provide
tort trade. The subsidies deemed not to distort trade were such subsidies only if it has made a commitment to re-
exempted from controls under the agreement. In addi- duce them.7 Twenty-five countries, accounting for over
tion, certain of the trade-distorting subsidies—roughly 40 percent of agricultural exports worldwide, have made
one-third of them by dollar value worldwide—were ex- such commitments. The EU makes the greatest use of
empted as well for various reasons. The rest of the trade- such subsidies, providing 85 percent to 90 percent of the
distorting subsidies had limits imposed, and countries export subsidies reported by the 25 countries. The
made specific commitments to reduce them.6 World- United States has accounted for between 1 percent and 2
wide, about one-third of agricultural subsidies fall into percent. The EU’s export subsidies have averaged 6.6 per-
that last category. Although the United States and the EU cent of the value of its exports; the United States’, about
provide comparable amounts of subsidies in total, the 0.05 percent.
United States provides only about one-third as much sup-
port in the trade-distorting category limited by reduction
commitments. The EU provides well over half of the The Effects of Liberalizing World
world total of such subsidies, the United States about Agricultural Trade
one-fifth, Japan between 8 percent and 9 percent, and ev- Countries typically adopt trade-distorting agricultural
ery other country substantially less. policies to benefit their domestic agricultural producers.
In doing so, however, they generally impose costs on their
As a percentage of agricultural output, U.S. subsidies in consumers, who as a result must pay more for agricultural
the category are high enough to significantly distort pro- products protected by tariffs; on their taxpayers, who
duction and trade but much lower than those of some must pay for any subsidies; and on competing foreign
other countries, most notably the EU. The EU’s subsidies producers, who lose sales. The costs to domestic consum-
in the category have averaged 19 percent of the value of ers and taxpayers alone are usually greater in dollar terms
agricultural output since 1998 (but were a little lower, at than the benefits to domestic producers. Therefore, elim-
15.9 percent, in 2001, the most recent year for which the inating those policies generally yields a net economic
EU has reported). The United States’ subsidy rate, at 7.7 benefit, although the agricultural sector may be harmed.
percent, has been less than half of the EU’s. Mexico has
averaged just over 1 percent and Australia only 0.5 The Effects of Full Liberalization
percent. Almost all of the studies examined by CBO predict that
the United States would gain from full liberalization of
agriculture (that is, complete elimination of all tariffs and
subsidies) by all countries. The one exception (which uses
an alternative assumption about the workings of nonagri-
cultural markets) nevertheless predicts that U.S. agricul-
6. The net result of the provisions was five categories, or “boxes.” ture would benefit. Moreover, almost all of the studies
The green box is for subsidies that were deemed by the negotiators
predict that U.S. agriculture as a whole would benefit, al-
to have little or no distorting effect on trade. The blue box is for
certain direct payments under production-limiting programs. though to a lesser extent than would the country as a
Tying subsidies to requirements that recipients limit production whole. It is difficult to draw reliable conclusions from the
is a means some countries use to reduce the distorting effect of studies’ estimates of the effects of liberalization on pro-
their income-support programs. The special and differential box is ducers of individual agricultural products, some of whom
for certain development subsidies granted by some developing
would lose subsidies or tariff protection.
countries. De minimis support consists of subsidies that are below
specified percentages of the value of production that the negotia-
tors deemed low enough so as not to be a cause for concern. 7. The Uruguay Round Agreement made a temporary exception for
Finally, the amber box is for all support not falling into any of the developing countries without reduction commitments to provide
other four boxes. Such support, often called nonexempt trade- certain kinds of export subsidies during the initial nine-year
distorting support, was limited, and countries made commitments implementation period of the agreement. However, that period is
to reduce it. now over.
E C O N O M I C A N D B U D G E T I S S U E B R I E F
AGRICULTURAL TRADE LIBERALIZATION 5
The studies generally also find that all developed coun- Developing countries as a group would benefit more
tries would benefit from a liberalization agreement and from the liberalization of their own policies, which di-
that most developing countries—including China, India, rectly affect both their exports and their imports, than
and Brazil—would gain as well. Countries whose agricul- they would benefit from the liberalization of developed
tural sectors would probably benefit the most include countries’ policies, which directly affect only the exports
Australia, New Zealand, Canada, Brazil, and Argentina. of developing countries. To the extent that developing
Countries whose agricultural sectors would probably be countries are harmed by developed countries’ policies
harmed the most include members of the European that distort trade, the evidence suggests that the Euro-
Union and the European Free Trade Association and pean Union and high-income Asian countries are much
high-income Asian countries. The United States is some- larger sources of harm than the United States.
place in the middle—as are China and India, whose agri-
cultural sectors are predicted to be little affected. Considerations Concerning Partial Liberalization
A multilateral trade agreement might not eliminate all
For the world as a whole, one can conclude on the basis policies that distort agricultural trade. Instead, it might
of the studies CBO examined that if all policies distorting eliminate some such policies, reduce others, and leave still
agricultural trade were phased out by the end of this de- others in place. The studies that CBO surveyed support
cade, the likely total annual economic benefit by the mid- several observations about the effects of such partial
dle of the next decade from resulting efficiency gains and liberalization.
investment growth would be in the range of $50 billion
The Significance of Tariffs. Tariffs and tariff-rate quotas
to $185 billion, or 0.1 percent to 0.4 percent of the value
are by far the most costly of the policies that distort world
of world output of all goods and services, or roughly 3
agricultural trade as they account for 80 percent to 90
percent to 13 percent of the value added by world agri-
percent of the total cost. Moreover, extremely high tariffs
culture. In the studies that also include effects of liberal-
on a few selected products cause a disproportionately
ization on the rate of productivity growth, the benefits
large percentage of the economic burden of agricultural
are 50 percent to more than 100 percent larger. If the
tariffs. Correspondingly, an agreement that does not en-
other studies that CBO examined had also incorporated sure that many of those high tariffs are significantly re-
such effects and found them to be of similar magnitude, duced is not likely to have much beneficial effect. The
the range for the estimated total benefit from all three ef- plan in the Doha Round negotiations was to allow each
fects of liberalization—efficiency gains, investment country to choose a few sensitive products and each devel-
growth, and productivity growth—would be roughly 0.2 oping country to choose a few additional special products
percent to more than 0.8 percent of global output, or 7 whose protective tariffs would be cut less than others.
percent to more than 25 percent of the value added by The 2006 World Bank study modeled variations on a
world agriculture.8 prototype scenario for agricultural liberalization in the
general range of those under discussion in the Doha
Agricultural trade liberalization would most likely in- Round. That study found that allowing countries to des-
crease the real wages of both skilled and unskilled workers ignate as few as 2 percent of their tariff lines as protecting
and, to a slightly lesser extent, interest earnings and prof- sensitive products and developing countries to designate
its in almost all countries, with larger effects in percentage an additional 2 percent as protecting special products
terms for less-developed countries. Whether the wages of could eliminate 80 percent of the economic gain that
unskilled workers increased by more than the wages of would otherwise result from the prototype liberalization
skilled workers or vice versa for a given country would scenario.
most likely depend on whether the growth of the coun-
try’s agricultural output increased or decreased as a result The Effects of Subsidies. Domestic subsidies are the sec-
of liberalization—with increased growth of output result- ond most costly of the policies distorting agricultural
ing in greater relative growth of unskilled workers’ wages. trade, followed by export subsidies. Unlike tariffs, which
tend to harm all countries, subsidies tend to benefit the
8. The latter range of estimates assumes that agricultural value added countries purchasing the subsidized products and to
maintains the 3.5 percent share of world output that it had in harm the countries granting the subsidies (although their
2003—the most recent year of available data. agricultural sectors benefit) and the countries that are
 E C O N O M I C A N D B U D G E T I S S U E B R I E F
6 CONGRESSIONAL BUDGET OFFICE
competing agricultural exporters. Because most subsidies ported subsidies in the category were almost 25 percent
are granted by developed countries, export subsidies tend below their allowable level, while Japan’s were 84 percent
to benefit developing countries and to harm developed below theirs.
countries; and, to a lesser extent, the same pattern is true
for domestic subsidies. Special and Differential Treatment. In international trade
negotiations, developing countries are often given special
Actual Tariffs and Subsidies and Their Allowable Values and differential treatment, under which their required cuts
Under Agreements. Agreements to reduce tariffs will pro- in tariffs and subsidies are smaller than those for devel-
duce economic benefits only if they lower the tariffs actu- oped countries. An irony is that such concessions won by
ally in effect. Many agricultural tariffs currently in effect developing countries are more costly to them than they
are substantially below their maximum allowable values are to developed countries. In the 2006 World Bank
that were agreed to in the Uruguay Round Agreement.
study, eliminating special and differential treatment from
For example, India’s actual tariff on cattle is 39 percent
its prototype scenario increased the estimated benefits to
even though its allowable tariff is 100 percent.9 There-
high-income countries by 21 percent, to middle-income
fore, an agreement to reduce its allowable tariff by 50 per-
countries by 37 percent, and to low-
cent would have no effect. For the world as a whole, the
average agricultural tariff rate is less than one-half of the income countries by 64 percent.
average allowable rate. Further, for least-developed coun-
tries, the average tariff is more than 80 percent below the
Liberalization Beyond the Agricultural Sector
average allowable tariff. Multilateral trade agreements The effects of liberalization in the agricultural sector
usually specify reductions in allowable tariffs, not actual would be influenced by liberalization in other sectors
tariffs. Because of the substantial differences between the such as manufacturing or services. Of particular interest,
two, reductions in the allowable tariffs must be quite developing countries’ exports of agricultural products to
large to effect a meaningful reduction in actual tariffs. developed countries would increase more if developing
countries reduced their own barriers to imports of manu-
Similar differences between trade-distorting subsidies and factured goods. Similarly, developed countries’ exports of
their allowable amounts mean that reductions in those manufactured goods to developing countries would in-
limits also must be substantial to have much effect. For crease if developed countries reduced their own agricul-
example, the EU=s most recently reported subsidies in tural tariffs and subsidies. The reason is that such liberal-
that category were less than one-half of their allowable ization would make it easier for countries with a
value, so a 50 percent reduction in the allowable value comparative advantage in agriculture to export agricul-
might have no effect. The United States’ most recently re- tural goods in exchange for imports of manufactured
goods from countries with a comparative advantage in
9. Benjamin Buetre and others, “Agricultural Trade Liberalization: manufacturing. Moreover, the 2006 World Bank study
Effects on Developing Countries’ Output, Incomes, and Trade,”
found that adding liberalization of trade in manufactured
Australian Bureau of Agricultural and Resource Economics
Project 110039 (paper presented to the Seventh Annual Confer- goods to its prototype agricultural liberalization scenario
ence on Global Economic Analysis, Trade, Poverty, and the Envi- increased the benefit for developing countries more in
ronment, Washington, D.C., June 17-19, 2004). percentage terms than it did for developed countries.
E C O N O M I C A N D B U D G E T I S S U E B R I E F
AGRICULTURAL TRADE LIBERALIZATION 7
Bibliography Burfisher, Mary E., ed., Agricultural Policy Reform in the
Some of the studies of agricultural liberalization that WTO—The Road Ahead, Agricultural Economic
CBO surveyed for the papers summarized in this issue Report No. 802 (U.S. Department of Agriculture,
brief are the following: Economic Research Service, Market Trade Econom-
ics Division, May 2001).
Anderson, Kym, and Will Martin, eds., Agricultural Trade
Reform and the Doha Development Agenda (New Fontagne, Lionel, Jean-Louis Guerin, and Sebastien Jean,
York: Palgrave Macmillan and the World Bank, Market Access Liberalisation in the Doha Round: Sce-
2006). narios and Assessment, Working Paper No. 2003-12
(Paris: Centre d’Etudes Prospectives et d’Informa-
Beghin, John C., David Roland-Holst, and Dominique tions Internationales, September 2003).
van der Mensbrugghe, Global Agricultural Trade and
the Doha Round: What Are the Implications for North Roberts, Ivan, and others, Reforming World Agricultural
and South? Working Paper 02-WP 308 (Ames, Iowa: Trade Policies, Australian Bureau of Agricultural and
Iowa State University, Center for Agricultural and Resource Economics Research Report 99.12 and
Rural Development, June 2002). Rural Industries Research and Development Corpo-
ration Publication No. 99/96 (September 1999).
Brown, Drusilla K., Alan V. Deardorff, and Robert M.
Stern, Computational Analysis of Multilateral Trade World Bank, “Envisioning Alternative Futures: Reshap-
Liberalization in the Uruguay Round and Doha Devel- ing Global Trade Architecture for Development,”
opment Round, Discussion Paper No. 489 (Ann Chapter 6 in Global Economic Prospects and the Devel-
Arbor, Mich.: University of Michigan, School of oping Countries, 2002 (Washington, D.C.: World
Public Policy, Research Seminar in International Eco- Bank, 2002).
nomics, December 8, 2002).
This brief was prepared by Bruce Arnold. It and
Buetre, Benjamin, and others, “Agricultural Trade Liber- other CBO publications are available at the agency’s
alization: Effects on Developing Countries’ Output, Web site (www.cbo.gov).
Incomes, and Trade,” Australian Bureau of Agricul-
tural and Resource Economics Project 110039 (paper
presented to the Seventh Annual Conference on Glo-
bal Economic Analysis, Trade, Poverty, and the Envi- Donald B. Marron
ronment, Washington, D.C., June 17–19, 2004). Acting Director