Front Cover
 Title Page
 Table of Contents
 Examination of proposals for tariffication...
 Special and differential treatment,...
 Nontraditional exports of developing...
 The impact of trade liberalization...
 Food security and compensation:...
 The impact of trade liberalization...
 Agricultural trade reform, price...
 Back Cover

Title: GATT, agriculture, and the developing countries
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00085359/00001
 Material Information
Title: GATT, agriculture, and the developing countries
Physical Description: Book
Language: English
Creator: Islam, Nurul ( Editor )
Valdes, Alberto ( Editor )
Publisher: International Food Policy Research Institute
Place of Publication: Washington, D. C.
Publication Date: September, 1990
Copyright Date: 1990
 Record Information
Bibliographic ID: UF00085359
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 23077334 - OCLC
isbn - 0-89629-314-9

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Title Page
    Table of Contents
        Table of Contents
        Page 1
        Page 2
        Page 3
    Examination of proposals for tariffication and disciplines on subsidies and quantitative controls currently under negotiation
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
    Special and differential treatment, agriculture, and the developing countries in the Uruguay round
        Page 11
        Page 12
        Page 13
        Page 14
    Nontraditional exports of developing countries: The case of horticultural exports
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
    The impact of trade liberalization on low-income, food-deficit countries
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
    Food security and compensation: The role of the GATT
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
    The impact of trade liberalization on domestic and international price instability
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    Agricultural trade reform, price stability, and the impact on developing countries
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
    Back Cover
        Page 54
Full Text


w N









ISBN 0-89629-314-9


1. Introduction 1
Nurul Islam and Alberto Valdes

2. Examination of Proposals for Tariffication and Disciplines on Subsidies
and Quantitative Controls Currently under Negotiation 4
Alberto Vald6s and Joachim Zietz

3. Special and Differential Treatment, Agriculture, and the Developing
Countries in the Uruguay Round 11
John Whalley

4. Nontraditional Exports of Developing Countries: The Case of Horticultural
Exports 15
Nurul Islam

5. The Impact of Trade Liberalization on Low-Income, Food-Deficit Countries 25
Thomas W. Hertel

6. Food Security and Compensation: The Role of The GATT 33
Odin Knudsen

7. The Impact of Trade Liberalization on Domestic and International
Price Instability 39
Rod Tyers

8. Agricultural Trade Reform, Price ability, and the Impact on
Developing Countries 47
Ammar Siamwalla



Nurul Islam and Alberto Valdes

IFPRI convened a seminar on "GATT Negoti-
ations on Agriculture and Developing Countries"
on May 30-June 1, 1990, in Montreux, Switzer-
land, with participation by a number of distin-
guished experts on GATT and developing-coun-
try officials engaged in the GATT negotiations.
The seminar's deliberations focused on issues of
negotiation of particular interest to developing
countries and options and strategies available to
the developing countries that would derive signif-
icant benefits from the liberalization of world
trade. The seven papers in this volume served as
the context for the deliberations.
The chapter by Valdes and Zietz examines
the proposals put forward by the larger players
on reforming agricultural trade in light of a gen-
eral agreement to reduce subsidies. The scope
for compromise among the EC, the United
States, the Cairns Group, and Japan is assessed
as is the effect of the outcome of negotiations on
developing countries. Whalley discusses the
issue of special and differential treatment (S&D)
for developing countries in GATT. He evaluates
the significance of S&D for the trade, growth, and
development of developing countries and how
these countries may extend its application in
agriculture. Islam looks at the ways tariff and
nontariff barriers affect the horticultural exports
of developing countries. He assesses the effects
of the elimination of these barriers on the earn-
ings of developing countries from horticultural
trade and the prospects for such liberalization. In
the next chapter, Hertel analyzes the effects on
price of trade liberalization, what the rise in price
will mean for the poorer, food-deficit countries,
their strategies for coping with the price
changes, and how GATT negotiations can allevi-
ate some of the short-run drawbacks. In his
chapter Knudsen examines the importance of
food security and compensation for developing
countries in the context of trade liberalization,
current mechanisms for dealing with these is-
sues, and the ways in which structural adjust-
ment can enable developing countries to benefit

from liberalization in the long run. Tyers dis-
cusses how domestic market insulation can lead
to international price volatility, the nature and
importance to governments of domestic market
insulation, and the national and global gains
from trade liberalization. In the final chapter
Siamwalla focuses on the effects of current agri-
cultural trade policies of the EC, the United
States, and Japan on price volatility, how these
approaches affect the negotiations in the Uru-
guay Round, and national and international ac-
tion to contain the effects of international price
variability on developing countries.
From the discussion on these issues in the
seminar arose agreement on the following key


The ingredients for successful negotiations
on agriculture are now present. The April 1989
midterm review concluded with a mandate for
substantial, progressive reductions in the levels
of agricultural support. Reforms are needed in
four major areas: domestic subsidies, import ac-
cess and border protection, export competition,
and sanitary and phytosanitary measures. In ad-
dition, all the proposals submitted for discussion
at GATT included provisions on special and dif-
ferential treatment for developing countries.
The real work now is to precisely and quickly
specify the key modalities for reducing the level
of government support for agriculture within a
framework that brings agriculture fully into the
GATT system.
There is a general and broad agreement on
the tariffication approach. In substance, how-
ever, there are many details to work out. These
include how to apply the tariffication principle,
what products should be covered by tariffication,

and what measures should apply during the tran-
sition period.
The Aggregate Measure of Support (AMS) is
generally seen as a useful monitoring and eval-
uation device, rather than a technique of negoti-
ations. But the real commitment will be under-
taken in terms of specific policy instruments.
Because it is difficult to compute AMS for all
agricultural commodities, a strict reliance on
what the AMS approach can measure runs the
risk of restricting negotiations to the main farm
products in industrial countries, without full rec-
ognition of the other products of importance to
developing countries.
The developing countries have a vital interest
in an agreement on strengthening the GATT
rules and disciplines, rather than accepting an
agreement around specific outcomes and symp-
toms in world commodity markets. The objective
is to seek rules that provide more certainty,
transparency, and stability to international trade.
The alternative is to agree on specific outcomes
in terms of trade flows, market shares, and the
like, which are perceived as symptoms of an
imbalance between supply and demand, and to
reexamine them in five or six years without an
explicit commitment to new rules. As small eco-
nomic actors, developing countries are particu-
larly interested in bringing agriculture more
prominently into the GATT framework, with an
international system that provides a framework
of norms, rules, and procedures, and with a
strong monitoring mechanism within GATT. The
alternative could be discriminatory protection
and bilateralism, which is clearly much less
promising for developing countries and small
economic actors in general.
In practical terms, this means looking beyond
agriculture and having a cross-sectoral view of
potential benefits of the negotiations. This also
suggests the importance of involvement in the
negotiations of governments of developing and
developed countries at the highest levels in
order to break the deadlock over sectoral issues.


The mandate of the Uruguay Round urges
special and differential treatment for the devel-
oping countries and takes into account their fi-
nancial and trade needs and special constraints
on development. However, there is a need for
quickly integrating into the GATT system fuller

participation of the developing countries. This
includes accepting reciprocity, particularly by
middle- and high-income developing countries.
Past experience with differential treatment for
developing countries has not been encouraging.
Nonreciprocity on the part of the developing
countries kept them outside the mainstream of
the GATT negotiations and consequently trade
liberalization bypassed the tropical commodities,
such as tropical and processed tropical prod-
ucts, which were of major interest to the devel-
oping countries.
Moreover, in recent years developing coun-
tries have undertaken trade liberalization either
unilaterally or as part of structural adjustment
programs or economic policy reforms under the
auspices of the World Bank and the International
Monetary Fund. They reduced their trade barri-
ers without reciprocal concessions from the in-
dustrial countries. In their current negotiations
for trade concessions under the auspices of
GATT, the developed countries should give
these developing countries "credit" for their past
liberalizations. This might best be accomplished
by granting developing countries concessions
now against the concessions given by them ear-
lier outside GATT.
Four measures deserve emphasis regarding
differential treatment for developing countries.
These are
* a longer transition period of adjustment for
developing countries;
* exclusion of development measures in the
coverage of the AMS for developing countries,
as suggested in the Cairns Group proposal;
* more flexibility in domestic price stabilization
measures for the major staples; and
* compensation for the erosion of trade prefer-


Many nontraditional exports such as horticul-
tural and floricultural products have high export
potential for many developing countries. A major
question raised on several occasions is where
do such products belong in the negotiations?
How can they be incorporated under the GATT
rule? Current negotiations under way in commit-
tees on agriculture, tropical products, tariffs,
nontariff barriers, and sanitary and phytosanitary

(SPS) measures all impinge on the final outcome
for these commodities.
A final agreement on SPS measures, negoti-
ated soon, will be significant. Harmonization of
national import regulations and barriers for horti-
cultural products based on international recom-
mendations and sound scientific criteria is a
long-term objective under the SPS negotiations.
They are essential ingredients of the package on
SPS and should be complemented by a system
of consultation, dispute, and settlement proce-
If negotiations succeed in tightening rules on
export subsidies and market access, developing
countries will want to make sure that SPS mea-
sures do not become loopholes for discrimina-
tion against exports. Furthermore, if SPS agree-
ments were to be restricted to those products
explicitly negotiated under the group on agricul-
ture, this could represent an outcome detrimen-
tal to developing countries, whose potential ex-
port growth often lies with nontraditional
horticultural products excluded from the agricul-
ture negotiations.


High on the negotiating agenda is identifying
an offsetting mechanism to help net food-import-
ing developing countries if world prices of the
principal staples were to rise as a result of trade
liberalization in OECD countries. This will re-
quire identifying changes in world prices attrib-
uted to the policy reforms following a GATT
agreement. This price change will probably be
small compared with the variability in world
prices resulting from other causes such as ex-
change rate fluctuations and weather variability.
The exact mechanism of compensation is still
uncertain. Alternatives include the IMF Compen-
satory and Contingency Facility, multilateral food
aid on a grant basis, and enlarged trade conces-
sions or market access for the exports of net
food-importing countries within the context of
GATT negotiations for a transitional period. The
last two alternatives seem more feasible than the
first. The possibility of reaching a food aid agree-
ment outside the GATT negotiations in time for
such an agreement to be accomplished simulta-
neously with the GATT needs further explora-
tion. There is a precedent in the case of the Food
Aid Convention negotiated as a part of the Inter-
national Wheat Agreement under the auspices of
the Kennedy Round.

Almost all of the empirical research on the
topic shows that the level of instability of world
prices is exacerbated by government policies
that shield domestic farmers, consumers, and
stockholders from fluctuations in international
The larger the number of countries that agree
to reduce the insulation of domestic markets, the
greater is the possible reduction in the instability
of world prices. But if some countries do reform
while others do not, the former will suffer from
instability created by the latter. The developing
countries need bargaining power to ensure that
major OECD countries reduce insulation suffi-
To the extent that price instability remains a
problem, developing countries need to under-
take measures to achieve relative stability, espe-
cially for food staples, which constitute a high
proportion of consumption expenditures of the
poor households. This can be done by govern-
ment open-market operations in domestic mar-
kets, by variable taxes and subsidies on external
trade, or by a combination of both, depending on
the particular circumstances of a country. This
would require that developing countries be
granted exemptions from the GATT rules. The
methods of operation of the necessary domestic
and border measures should be subject to sur-
veillance by the GATT so that they do not en-
courage overt protectionism.

Negotiations should not be split up into differ-
ent groups before clarification of a unifying
theme to serve as the principle of the negotia-
tions. It is necessary to ensure that cross-
sectoral concessions on trade liberalization
measures can be coordinated as the negotia-
tions in the different groups proceed. Splitting
into different committees agricultural commodity
groups that belong together, such as agriculture,
tropical products, and natural resources, creates
great uncertainty and imposes a heavy burden
on the limited negotiating capacity of the devel-
oping countries. The developing countries need
to ensure that the widest possible commodity
coverage should be provided in the GATT nego-
tiations, particularly for those commodities of in-
terest to the developing countries.


Examination of Proposals for Tariffication and
Disciplines on Subsidies and Quantitative Controls
Currently Under Negotiation

Alberto Valdes and Joachim Zietz

During the April 1989 meeting of the Trade
Negotiations Committee (TNC) of the General
Agreement on Tariffs and Trade (GATT) negoti-
ators agreed on a "substantial progressive re-
duction in agricultural support," with all forms of
agricultural protection up for negotiation. This
agreement is a first in the history of GATT. All
major participants in the group on agriculture
have since tabled proposals for reforms in keep-
ing with these new guidelines. Limited space
here allows only a summary of the submissions
of the large countries or groups of countries.

At the heart of the U.S. proposal are tariffica-
tion and strict disciplines on domestic subsidies.
In particular, to improve access for imports, the
United States proposes the rule of tariffs (tar-
iffication); the elimination of all grandfather
clauses, waivers, and the like; a ban on volun-
tary agreements for export restraint; a prohibi-
tion on export subsidies; and tightened rules for
food aid. To limit internal support, a "traffic light"
approach is suggested, consisting of a category
of policies that would be prohibited, a category of
policies to be permitted, and an amber category
of sector-wide policies that are subject to disci-
plines. An aggregate measure of support (AMS),
without its nominal protection component, is sug-
gested to implement the disciplines in the amber
category.1 All major changes are proposed to be
implemented within a period of up to 10 years
and to be applicable to developing countries.

Exceptions are envisioned only for the least de-
veloped countries.
The proposal of the Cairns Group is similar to
that of the United States in its emphasis on
tariffication (although with a view toward harmo-
nization), the elimination of all grandfather
clauses, exceptions, and so on, and the termina-
tion of export subsidies. The Cairns Group also
advocates a traffic light approach to limit internal
support, with the AMS concept applied to the red
and amber categories during a transition period
and the amber category thereafter. Included in
the permitted support payments (green category)
would be direct income assistance, decoupled
from production and marketing; resource rede-
ployment assistance; and non-commodity spe-
cific aid for the development of infrastructure. To
avoid indirect export subsidies via food aid, the
latter would be on a grant basis only, channelled
through international organizations. Tariff quo-
tas as well as gradual reductions in export sub-
sidies and AMS levels would be used during the
transition period. Developing countries would be
exempt from the limits on government support
for agricultural and rural development programs,
would be allowed higher amounts of support for
measures that distort trade, would be subject to
lower cuts in import barriers, and would be given
longer timeframes for implementing reforms.
The European Community's (EC) proposal
differs fundamentally from those of the United
States and the Cairns Group. It is grounded on
two points: agriculture is sufficiently different
from all other sectors to warrant special rules;2
and the focus of the GATT negotiations is not the

1 Compare this idea with Zietz and Vald6s (1988, 71-72).
2 Agricultural markets are considered inherently unstable in the absence of government intervention. That is, they are
characterized by "a succession of cyclical crises.

nature of the rules governing agricultural trade
after some transition period, but only on the tran-
sition period itself. The EC is not willing to com-
mit itself to a final goal other than the elimination
of severe market imbalances. Following this
logic, the EC suggests adjustments only for
those commodity groups for which excess sup-
plies exist or are likely to exist and only to the
extent required to eliminate the excess.
The EC proposes using AMS as the main
vehicle through which to reduce current market
imbalances. Unlike the Cairns Group proposal
the EC does not call for the elimination of spe-
cific policies that are subject to AMS disciplines.
Limited or modified tariffication is seen as open
for discussion only if tariffs can be harmonized
and only as an extension of AMS.3 The EC's plan
calls for flexible application of the rules in devel-
oping countries, in particular the least developed
ones. Flexibility applies to the types of products
under AMS discipline as well as to the magnitude
and timeframe of adjustments.
Japan proposes the use of AMS along lines
similar to those suggested by the EC, but with a
number of important government support mea-
sures excluded and with no additional restric-
tions placed on included measures. Japan sup-
ports an end to export subsidies but also favors
continuation of import restrictions to maintain the
domestic production of major staple foods in
amounts deemed consistent with food security
concerns. Special and differential treatment for
developing countries is taken into account with
respect to the timeframe and the degree of the
reductions in import protection and government
The four proposals can be summarized as
follows. The core of the U.S. and Cairns Group
proposals is tariffication, a ban on export subsi-
dies, and restraints on domestic subsidies. The
proposals by the EC and Japan are much less
comprehensive. They contain no commitment to
tariffication but rely mainly on reductions in AMS.
It is apparent that the EC is willing to consider
reductions in government support only to the
extent that they eliminate excess supplies, rather
than as a means of bringing agriculture closer to
the GATT framework. Developing countries are
generally provided with more flexibility in imple-
menting the rules. State trading is essentially not


From the perspective of integrating agricul-
ture into the GATT and thereby enhancing that
institution's credibility and usefulness, compre-
hensive tariffication (with broadly uniform rates)
and a ban on export subsidies are crucial ele-
ments in reform of the trade in agricultural prod-
ucts.4 Relying on reductions in AMS in lieu of
tariffication will not achieve these objectives for
several reasons.5 First, in contrast to tariffica-
tion, AMS is not applicable in its general form to
sectors outside of agriculture. Agriculture would
thus continue to be a special case. Second, the
trade effect of the AMS reductions can vary con-
siderably depending on what support measures
are applied. Third, AMS applications as envis-
aged by the EC and Japan allow for complete
stabilization of domestic prices. That price stabil-
ity will continue to be exported and induce insta-
bility in world prices. Even though tariffication is
the preferred solution to the problems of agricul-
tural trade, a number of issues, discussed below,
have to be overcome to achieve a breakthrough
on this point.

Price Stability, Food Security, and the
Special Nature of Agriculture

The stability of agricultural prices and food
security tend to be major political issues in many
countries. Both are at the center of the insistence
by the EC and Japan that agriculture is a sector
in need of special rules. However, this idea is
unacceptable to countries that rely as heavily on
agricultural exports as the EC and Japan do on
industrial exports. This approach amounts to
treating agricultural exporters less favorably
than industrial exporters, a situation that not only
is unfair, but is also inherently inconsistent with
GATT's underlying principle of nondiscrimina-
tion. This type of differentiated treatment is a key
reason the whole GATT system has lost credibil-
Treating agriculture as a special case is not
unique to advanced developed countries. As na-
tions move from an agriculture-based economy
to one founded on industry, agriculture generally
shifts from being a sector that is taxed to one that

3 Modified tariffication consists of limiting import protection and export subsidies to X percent of the AMS value (for example,
30 percent of 1.5) and reducing its equivalent absolute value (for example, 0.45) along with the AMS value.
4 See also Blackhurst (1986) for the simple logic of a tariffs-only rule in a most-favored-nation (MFN) framework.
5 A detailed analysis is provided in Zietz and Vald6s (1988, 62-68).

is protected. Unless a country continues to have
a comparative advantage in agriculture, border
measures tend to stabilize the domestic prices of
staple food products at levels in excess of the
average world price.6 (The Republic of Korea
provides a typical recent example of this type of
development.) Over time, a growing number of
developing countries can be expected to experi-
ence a similar phenomenon. If each one is free
to effect price stability and food security by insu-
lating their domestic market through border mea-
sures, key agricultural markets will continue to
face the present problems of chronic oversupply
as well as depressed and highly volatile prices.
Given the potential for these long-run prob-
lems it seems unwise to permit frontier measures
for the purpose of price stabilization. The only
exception that can be envisioned is special and
differential treatment (S&D) for developing coun-
tries whose income level is low or where food
accounts for a high share of household expendi-
tures. Once a country graduates, however, price
stability and food security have to be effected by
internal measures that are decoupled from pro-
Price stabilization measures are often consid-
ered a way to provide food security. This concept
is found in Japan's proposal, among others. Di-
rect stabilization of farm incomes, stockpiling,
and diversification of sourcing are alternatives
that are known to be less disruptive to agricul-
tural markets. However, even if none of them is
considered feasible or sufficient, there is still no
need to jettison the idea of tariffication and, with
it, the prospect for more stable world prices. To
ensure domestic production in a world founded
on bounded tariffs, a country may simply request
an above average level of long-run tariff protec-
tion on its main staple food that is the source of
the food security concerns (for example, rice in
Japan or white maize in East Africa).7 To make
this idea acceptable to its trading partners, a
country might have to "pay" for its high tariff by
providing better market access in other areas
inside or outside of agriculture.

Treatment of Domestic Subsidies

There appears to be some momentum in the
negotiating group on agriculture for the traffic
light approach to domestic subsidies. To imple-

ment this approach and at the same time avoid
special rules for agriculture, the existing subsi-
dies code would be applied to agriculture, with
the addition of some guidelines on the three
categories of subsidies.
It should be noted that the approach of mea-
suring the amber category with AMS (without the
nominal rate of protection component) is applica-
ble to all sectors, not just agriculture. Hence,
some joint sessions with the negotiating group
on subsidies would be useful, especially be-
cause numerous submissions in that group deal
explicitly with quantitative limits on subsidies
along AMS lines.
A note of caution is needed around fixing the
AMS levels so as to bind the subsidy equivalents
in the amber category.8 For one, subsidies are
easily hidden by governments. For another,
AMSs cannot be calculated for all products or
countries, they tend to be time-consuming to
estimate and update, and their impact on trade is
not always clear. These factors make negotia-
tions on AMS difficult. Valuable time and good-
will can be wasted in such negotiations, espe-
cially considering that restrictions on the amber
category of subsidies may not be needed in the
end because they are simply too costly to main-
tain in the absence of significant border mea-
sures (Snape 1987).

Political-Economic Obstacles to an
Accord on Agriculture

One key problem that may block final agree-
ment on agriculture is the inflexible negotiating
mandates of both the EC and Japan. For all
practical purposes, currently the agriculture min-
istries determine their countries' negotiating po-
sitions, and those ministries have become de
facto representatives of the farming lobby.
Therefore it is easy for the farming lobbies to
block any substantive change in the GATT rules
on agriculture.
There are few ways to circumvent this prob-
lem. In the case of the EC, it may be possible to
secure flexibility on tariffication and an end to
export subsidies in return for harmonization of
the EC rates of protection. Essentially, harmoni-
zation means lifting the zero tariff on oilseeds
and related animal-feed proteins that the United
States obtained from the EC during the Dillon

6 The observed simultaneous growth in income and agricultural protection is a natural consequence of the shift away from
agriculture and is readily explained within standard political economy models (Anderson and Tyers 1989).
7 Although this measure would break with the rule of uniform tariffs across broad commodity groups, it seems preferable to
schemes based on quantitative import restrictions or no agreement at all.
8 The problems are magnified if the broad version of AMS is applied, as in the EC proposal.

Round. At the moment, the United States ap-
pears to oppose such a deal. Moreover, it is not
clear to what extent tariff harmonization alone
would soften the EC's stance. In addition, little of
a similar nature can be offered to Japan to get it
to agree to tariffication in agriculture.
A modified negotiating approach that looks
beyond agriculture may be needed to achieve
substantive results. In particular, it is probable
that the dominant influence of the farming lob-
bies on the negotiations in agriculture can only
be reduced if agriculture is not dealt with sepa-
rately from the other sectors during the Uruguay
Round of negotiations. Cross-sectoral negotia-
tions, deals, or quid pro quos could provide a
useful way to diminish their influence. Consider,
for example, the potential for countervailing
power vis-a-vis the domestic farming lobby if the
industry lobby in the EC or Japan learned that it
had to "pay" for its government's intransigence
on agriculture with either better market access
for foreigners or no GATT deal in their area of
interest (Zietz 1989).

The Need for a Unifying Theme
for the Uruguay Round

Cross-sectoral deals, as suggested above,
are much easier to accomplish if there is a theme
that unifies liberalization efforts across all sec-
tors. Tariffication could be such a theme. It is
clear to most that the Uruguay Round badly
needs a unifying theme along the lines of the
across-the-board tariff cuts that emerged in the
Kennedy and Tokyo Rounds. Experience has
shown that, in the absence of a common theme,
each negotiating group wastes valuable time de-
ciding how to negotiate, what approach to use,
and how to coordinate its negotiations with those
of other closely related groups.9


In early 1990 the EC tabled a proposal that
supports country-specific safeguards under Arti-
cle XIX where imports from a limited number of
suppliers cause injury. Although it is helpful that
a proposal on selectivity has finally come into the
open after lingering in the background for so
long, the whole idea of selectivity is unaccept-
able, especially for developing countries. It is
totally at odds with GATT's fundamental princi-
ple of nondiscrimination and MFN treatment.10

Selectivity presents two major problems: it dif-
fuses the resistance of exporters and therefore
helps perpetuate trade restrictions; and it does
not speed domestic adjustment in the country
that initiates the action. It is possible to single out
politically or economically weak developing
countries without incurring the wrath of econom-
ically powerful trading partners that are also af-
fected. It is unfortunate that the United States
has found the EC proposal "interesting" and has
called for further discussions.
Clearly, a different reform of safeguards Arti-
cle XIX is needed. Zietz and Valdes outline one
possibility, based on the idea of extending tar-
iffication to Article XIX (1988,69-70). That is,
safeguard measures would be limited to tempo-
rary tariff increases or excess tariffs above the
long-run tariffs written into a country's tariff
schedules. These measures by definition would
apply to all imports, regardless of source, and
would be temporary and subject to automatic
phase-out over time. To avoid the problem of
prohibitive tariffs, the use of excess tariffs could
be conditioned on some minimum access provi-
sion similar in spirit to that currently written into
Article XI:2(c) of the General Agreement.
Excess tariffs might also be a solution to
phasing-in the tariffication. As compared with
tariff-quota arrangements or similar quantity-
based schemes for the transition period, from the
start excess tariffs are consistent with the rule of
tariffs. No new rules would be needed for the
transition period.


Is Active Participation and Reciprocity

In contrast to earlier GATT rounds, numerous
developing countries have participated in the
Uruguay Round. Increasingly these countries re-
alize there is little point in watching from the
sidelines as the main trading blocs make deals.
The unfortunate result of such a passive role is
well known from past GATT rounds: the areas of
particular interest to developing countries re-
ceive little if any attention. Participation, how-
ever, cannot involve simply asking for more spe-
cial preference schemes but accepting the idea

9 For example, the group on nontariff barriers is closely linked to the groups on agriculture, textiles, and natural
resource-based products, simply because nontariff barriers are most prevalent in the latter areas.
10 Zietz and Vald6s (1988, 25-30) provide an extensive discussion of the fundamental principles underlying the GATT.

of reciprocity. Having given up the disciplines of
reciprocity, many developing countries find that
they have at the same time dumped much of the
protection the GATT can provide them.11 Non-
reciprocity is backfiring, not only in terms of un-
predictable, whimsical trade barriers in industri-
alized countries, but also by retarding growth at
How is reciprocity to be interpreted? First, it
must not be understood too narrowly, in the
sense of a balanced give-and-take within a given
sector or, worse yet, a given product group. Sec-
ond, it does not mean that everything which ap-
plies to advanced industrialized countries will
apply equally to the poorest least developed
country. Exceptions are possible and their im-
portance is clearly understood and accepted by
all industrialized countries. In contrast, the idea
of reciprocity across sectors, although it has not
received universal applause, seems increasingly
to be appreciated as the Uruguay Round nears
its end. Third, reciprocity affords little leverage in
the negotiations unless the negotiating partner
can measure its effects. It is important for devel-
oping countries to form meaningful interest
groups that can make substantive offers to in-
dustrialized countries in return for specific re-

Is the Principle of Tariffication Acceptable?

Understandably, developing countries want
to outgrow their economic dependence on partic-
ular industrialized countries. Converting the
GATT to a system based on tariffication and
MFN treatment is a crucial step in this direction.
Currently, developing countries are largely de-
pendent on the economic goodwill of the coun-
tries in the Organisation for Economic Co-opera-
tion and Development (OECD), the generalized
system of preferences (GSP) being a case in
point. Developing countries have little certainty
about the benefits it generates and the countries
granting the preferences can remove them at
will. Further, the GSP does not apply to many of
the products of particular interest to developing
countries or to nontariff barriers. Furthermore,
GSP helps the advanced developing countries
considerably more than the less advanced ones.
Although the EC's Lom6 Convention is an im-
provement over GSP, it shares with the latter
some of the unfortunate limitations of preference

schemes. Developing countries are locked into
their traditional, low value-added exports. Al-
though they benefit from quota rents, they are
otherwise exposed to the whims of the agricul-
tural and manufacturing lobbies in the prefer-
ence-giving countries. This scenario is clearly
not what economic independence was meant to
be. Tariffication in an MFN world would provide
developing countries with a chance to decide
their own destiny rather than living within the
confines of a negotiated world where economic
and political power count and not comparative
To be a disciplining force in world trade, tar-
iffication has to be applied to all countries, re-
gardless of income level, after some reasonable
period of adjustment. Accepting tariffication,
however, need not imply to developing countries
that there is no room for S&D. S&D would just
take a different form that is consistent with the
GATT system rather than being a challenge to its
core principles. To be more specific, S&D could
mean fixing the long-run bound tariffs for eligible
countries at levels higher than those for OECD
countries, perhaps with the maximum allowable
long-run tariff pegged to the average rate applied
by OECD countries. The advantage of pegging
tariff rates would be twofold: tariff reductions
among OECD countries would automatically
lead to reciprocity by developing countries, and
developed countries would gain better access to
the markets of the developing countries: they
would simply lower their own tariffs.

What About Food Price Increases After
Trade Liberalization?

Many studies have shown that protection in
agriculture by OECD countries has led to both
artificially low world prices and increased price
instability (see, for example, Vald6s and Zietz
1980 or Anderson and Tyers 1990). Trade
liberalization is likely to raise world prices ceteris
paribus.13 Food importers are apprehensive
about higher food import bills; some exporters
worry they will lose quota rents. As a result, there
is little enthusiasm among food importers and
exporters covered by the Lom6 Convention or
similar agreements for changing the rules.
One important caveat in interpreting studies
that show price increases in the wake of liberal-
ization must be noted. The predicted price in-

11 See Curzon and Curzon (1989) for a clear elaboration of this idea.
12 See the discussion in Krueger, Schiff, and Valdes (1988).
13 Valdes (1987) provides a convenient summary of these results from the perspective of a developing country.

creases result from counterfactual ceteris pari-
bus exercises. That is, they do not translate into
actually observed price increases but rather are
deviations from a baseline. If the baseline itself
suggests a secular downward trend, then a pol-
icy-induced world price increase, phased in over
many years, would simply moderate the secular
price decline rather than lead to an increase in
observed prices (Zietz and Vald6s 1990).
Even in the event that observed food import
bills rise, the developing countries can be as-
sured of help. First, all four reform proposals

outlined at the beginning of the chapter contain
some provision for increasing shipments of food
aid. Second, the talks between the GATT's direc-
tor general with the World Bank and the Interna-
tional Monetary Fund (IMF) have shown that
both organizations stand ready to support devel-
oping countries that face rising food import bills
and similar adjustment problems. This tying to-
gether of the GATT reforms with financial assis-
tance by international organizations is a new and
profoundly important event.


Anderson, Kym, and Rod Tyers. 1989. Agricultural protection growth in advanced and newly industri-
alized countries. In Agriculture and governments in an interdependent world, ed. A. Maunder.
London: Dartmouth for the International Association of Agricultural Economists.
1990. How developing countries could gain from world food trade liberalization in the Uruguay
Round. In Agricultural trade liberalization: Implications for developing countries, ed. lan Goldin and
Odin Knudsen. Paris: Organisation for Economic Co-operation and Development and The World
Blackhurst, Richard. 1986. The new round of GATT negotiations: Rejuvenating the trading system.
EFTA Bulletin 27 (October/December).
Curzon, Gerard, and Victoria Curzon. 1989. Non-discrimination and the rise of "material reciprocity."
The World Economy 4 (December): 481-499.
Krueger, Anne O., Maurice Schiff, and Alberto Valdes. 1988. Agricultural incentives in developing
countries: Measuring the effect of sectoral and economywide policies. World Bank Economic
Review 2 (September): 255-271.
Snape, Richard H. 1987. The importance of frontier barriers. In Protection and competition in interna-
tional trade: Essays in honor of W. Max Corden, ed. Hendryk Kierzkowski, 215-232. Oxford: Basil
Valdes, Alberto. Agriculture in the Uruguay Round: Interests of developing countries. World Bank
Economic Review 1: 571-593.
Valdes, Alberto, and Joachim Zietz. 1980. Agricultural protection in OECD countries: Its cost to
less-developed countries. IFPRI Research Report 21. Washington, D.C.: International Food Policy
Research Institute, December.
Zietz, Joachim. 1989. Negotiations on GATT reform and political incentives. The World Economy 12
(March): 39-52.
Zietz, Joachim, and Alberto Valdes. 1988. Agriculture in the GATT: An analysis of alternative ap-
proaches to reform. IFPRI Research Report 70. Washington, D.C.: International Food Policy
Research Institute.
1990. International interactions in food and agricultural policies: Effect of alternative policies. In
Agricultural trade liberalization: Implications for developing countries, ed. lan Goldin and Odin
Knudsen. Paris: Organisation for Economic Co-operation and Development and The World Bank.


Special and Differential Treatment, Agriculture, and
the Developing Countries in the Uruguay Round

John Whalley

This chapter discusses the issue of special
and differential treatment in the General Agree-
ment on Tariffs and Trade (GATT) for developing
countries and how it may or may not apply to
whatever forms of disciplines on agriculture
emerge from the round.1 The chapter explains
what the term "special and differential treatment"
(S&D) means and, more importantly, its signifi-
cance for the various negotiating groups in the
Uruguay Round. It evaluates S&D as a negotiat-
ing objective for developing countries; how sig-
nificant it has been for their trade performance;
whether a more restricted form of treatment
would necessarily retard their growth and devel-
opment; and how developing countries might
seek to expand its application to agriculture.
The chapter suggests that the quantitative
studies, limited as they are, seem to point to the
conclusion that S&D has had only a marginal
effect on the economic performance of countries.
In the more rapidly growing economies such as
the Republic of Korea, Taiwan, Turkey, and oth-
ers, there is little evidence that S&D has played
much of a role in their strong performance. Until
a more complete set of studies on the enlarge-
ment of S&D for agriculture becomes available,
the null hypothesis from the research-oriented
literature seems to be that other, more focused
trade policy issues, such as product coverage
(sugar and meats) in agricultural liberalization,
may be of more significance.

The term S&D refers to various rights and
privileges given to developing country contract-

ing parties to the GATT but not simultaneously
extended to developed countries. Such treat-
ment involves two different things. First, devel-
oping countries have relatively more freedom to
protect domestic markets compared with devel-
oped countries. Second, preferential terms of
access to developed country markets are sup-
posedly granted to developing countries.
The term itself derives from the 1973 Tokyo
Round Declaration, which recognizes "the impor-
tance of the application of differential measures
to developing countries in ways which will pro-
vide special and more favorable treatment for
them in areas of negotiation where this is feasi-
ble and appropriate" (GATT 1974, 8). In GATT
terms developing countries are given special
treatment under Articles 18, 28:bis(3), Part IV,
and the 1979 Framework Agreement known as
the Enabling Clause.
Article 18, including amendments made at a
1955 GATT Review Session, was the GATT's
first attempt to accommodate developing country
concerns in the General Agreement. Article 18
has three major components. Article 18-A allows
developing countries to renegotiate tariff bind-
ings in order to promote the establishment of a
particular industry. A developing country using
this provision is expected to offer compensation
or face retaliation. Article 18-B is the balance-of-
payments escape clause for developing coun-
tries. The 18-B criteria for imposing import re-
strictions are less onerous than those that apply
to developed countries under Article 12. Article
18-C permits a developing country to apply
quantitative import restrictions for infant industry
purposes in order to extend new industries or
production structures. As with Article 18-A, Arti-
cle 18-C also provides for compensation and

1 This chapter draws on an earlier discussion of special and differential treatment in Whalley 1989 and forthcoming. I am
indebted to Patrick Low (on leave from the GATT Secretariat) for many lengthy discussions of these issues.
2 This section draws on Appendix C of Whalley 1989.

retaliation in the absence of a negotiated agree-
Article 28:bis(3), which made its first appear-
ance in the same 1955 Review Session, recog-
nizes that the needs of developing countries to
use tariffs for economic development and fiscal
purposes should be taken into account in tariff
negotiations. During the 1955 review, the con-
tracting parties were also encouraged to in-
crease capital flows to developing countries to
"facilitate the objectives of the General Agree-
ment by stimulating economic development of
these countries, while at the same time render-
ing it less necessary for them to resort to import
restrictions" (see GATT 1955, 49).
Part IV of the GATT, added in 1965 and enti-
tled "Trade and Development," was largely con-
sidered to be a reaction to United Nations Con-
ference on Trade and Development (UNCTAD) I
of 1964. It was also aimed at showing that the
GATT is sympathetic to developing country con-
cerns. Part IV adds three articles to the then
existing 35 GATT articles in Parts I to III.
Part IV did not provide a GATT Article 1 ex-
ception for trade preferences in favor of develop-
ing countries, as had been argued for at the time.
Only later, at UNCTAD II in New Delhi in 1968,
did the United States finally consent to partici-
pate in a global system of preferences, and then
on the condition that they "be limited to tariffs,
should be temporary, should be based on volun-
tary adherence, and should be extended by all of
the developed countries to all of the developing
countries on an MFN basis" (Stone 1984). The
generalized system of preferences (GSP) that
resulted was granted a 10-year waiver in the
GATT in 1971, although not until 1975 did the
United States implement its version of GSP.
In the Tokyo Round developing country activ-
ities were largely aimed at formalizing, codifying,
and extending these S&D provisions in the
GATT. At the beginning of the Tokyo Round in
1973, S&D for developing countries amounted to
Article 18, Article 28:bis(3), and Part IV. In addi-
tion, preferences extended under the GSP were
covered by a waiver of 10-years' duration from
GATT Article 25, which would have expired in
1981. Apart from one other waiver for a small
South-South preference agreement, the most-fa-
vored-nation (MFN) obligation of Article 1 was
still fully applicable to developing country trade
agreements. Any regional arrangements,
whether among developing countries only, or
between developed and developing countries,
were covered only under GATT Article 24.
Among the Tokyo Round results were four
Framework Agreements. Most important was the
Framework Agreement entitled "Differential and

More Favorable Treatment, Reciprocity and
Fuller Participation of Developing Countries."
Better known as the Enabling Clause, it is gen-
erally considered to be the centerpiece of the
results of S&D emerging from the Tokyo Round.
The Enabling Clause provided a permanent
legal basis for GSP schemes and similar status
to special and differential treatment provisions
under the Tokyo Round codes. It also covered,
under certain circumstances, regional and global
preferential arrangements among developing
countries not conforming to Article 24. It recog-
nized least-developed countries as a separate
category of countries, in need of even more fa-
vorable S&D. However, the Enabling Clause
also contained a concession to the developed
country point of view, the celebrated "gradua-
tion" provision: "...developing countries recog-
nize that as their economies grow stronger, they
would participate more fully in the GATT frame-
work of rights and obligations." The concept of
graduation has assumed a prominent place in
the debate on S&D.
The September 1986 Ministerial Declaration
launching the Uruguay Round reiterated that
S&D be accorded to the developing countries in
accordance with the terms of the 1979 Frame-
work Agreement, setting the stage for possible
further revisions in the treatment of developing
countries to emerge from the Uruguay Round.


S&D itself is not a matter for negotiation in the
Uruguay Round, since it is already part of the
system of trade rules embodied in the GATT and
its associated codes emerging from the Tokyo
Round. Rather, the issue in the Uruguay Round
is whether changes may be made to constituent
parts of S&D and how special and differential
treatment may or may not be applied to various
enlargements of the coverage of the GATT that
are supposed to occur in the round, in particular
in agriculture.
The declaration launching the round contains
a clear and unequivocal reaffirmation of S&D as
a principle of the trading system. However, the
developed countries have raised possible reform
of Article 18-B, and GSP graduation pressures
against developing countries (outside the round)
seem 'likely to continue. On the other hand, the
developing countries have made how to work out
new trade rules for agriculture in ways compati-
ble with S&D an issue, along with its application

to services, intellectual property, and invest-
In the agriculture group, developing countries
within the Cairns Group (especially Brazil) have
repeatedly insisted that any agricultural liberal-
ization occurring under new rules should take
place in such a way that development objectives
are not hindered. In practice they are calling for
weaker disciplines on themselves in this area.
What exactly these weaker disciplines would be
is not clear. Perhaps any change in Articles 11
and 16 of the GATT would allow existing treat-
ment for developing countries for a transitional
period, and negotiations on a surveillance-based
aggregate measure (such as a producer's sub-
sidy equivalent [PSE]) would be more lax as far
as developing countries are concerned.
Concerns over the special position of devel-
oping countries in agriculture have also arisen
among the group of agricultural net importers,
led initially by Jamaica but more recently by Peru
and Egypt. They argue that any form of global
liberalization will be disadvantageous to this
group of countries and therefore agricultural lib-
eralization (including that by developed coun-
tries) may be inconsistent with their development
On the access side, a form of S&D could be
argued for if any variant of the widely touted
tariffication proposals were implemented. Under
these, a portion of imports, which would grow
gradually, would be allowed at a lower bound
rate, while other imports would be subject to
higher rates. Over time the higher rates would
disappear, leaving trade only at the bound tariff
rates. The special and differential treatment pro-
vision for developing countries might be used to
argue that they immediately obtain access to

developed country markets at the lower bound
More generally, the application of S&D to ag-
riculture seems unlikely to involve extensions to
the right-to-protect, since Articles 18-B and
28:bis(3) already encompasses agriculture. It
seems more likely on the access side that some
form of preferential access arrangement of the
type outlined above be applied, or that preferen-
tial rules as to how developing countries might
be asked to deal with trade-distorting agricultural
subsidies in any multilateral effort to reduce or
even eliminate them be applied.
In the Tokyo Round the approach the devel-
oping countries followed was to elaborate and
enhance their S&D in the GATT in the two
senses of special rights to protect and preferen-
tial access. In the Uruguay Round the approach
thus far has been to defend S&D where possible,
and to seek to extend its principles to any new
areas coming under the GATT, but not to make
S&D the sole negotiating objective of developing
countries in the round (that is, to the exclusion of
agriculture, tropical products, textiles and cloth-
ing, safeguards, and other issues).
How this approach will work itself out in the
final stages of the round and what developing
countries will achieve remain to be seen. In spe-
cific areas within the round, such as agriculture,
the negotiating objectives of individual develop-
ing countries are quite different (exporters ver-
sus importers and exporters of grains versus
exporters of sugar versus exporters of meats).
As such, S&D, which is inherently a bloc-wide
negotiating approach by the grand coalition of all
developing countries, may prove more difficult to
sustain as a dominant negotiating objective of
the developing countries.

GATT (General Agreement on Tariffs and Trade). 1955. BISD. 3rd supp. Geneva: GATT.
1974. GATT activities in 1974. Geneva: GATT.
Stone, F. 1984. Canada, the GATT and the international trading system. Ottawa: Institute for Research
on Public Policy.
Whalley, J., ed. 1989. The Uruguay Round and beyond. The final report from the Ford Foundation-sup-
ported project on developing countries and the global trading system. London: Macmillan Press.
Forthcoming. Special and differential treatment for developing countries in the Uruguay Round.
Paper prepared for a symposium on the Uruguay Round. Economic Journal.


Nontraditional Exports of Developing Countries:
The Case of Horticultural Exports

Nurul Islam

The definition of nontraditional exports is not
very clearcut or unambiguous. The concept is
usually intended to convey the notion that these
exports are quantitatively small and are not im-
portant in trade. The rationale behind their being
a subject of serious attention is that they are
expected to grow in the future. However, what
one country defines as a nontraditional export
another may see as traditional. Moreover, the
comparative advantage of different countries in
different commodities changes over time.
When the agricultural exports of developing
countries as a whole are considered, the nontra-
ditional commodities are generally defined as
those that are less important than such commod-
ity groups as food, agricultural raw materials,
and tropical beverages and tobacco. For exam-
ple, historically fruits and vegetables have not
been important exports of developing countries.
Only in recent years have they assumed impor-
tance. In addition to horticultural products, flori-
cultural products, which include cut flowers, cut
foliage, and plants, are also increasing in impor-
During the period 1983 to 1985 food com-
modities, including cereals, livestock products,
sugar, oils, and oilseed, constituted 41 percent
of the total agricultural exports of developing
countries; tropical beverages and tobacco con-
stituted 24 percent, raw materials and other mis-
cellaneous commodities 22 percent. Fruits and
vegetables, on the other hand, accounted for 13
percent. Fruits and vegetables were, however,
relatively more important than agricultural raw
materials, even though raw materials such as
cotton, hard fibers, and rubber have historically
been important export items of developing coun-
tries. The horticultural exports of developing
countries amounted to an average of US$9 bil-
lion during the period 1983 to 1985, or 37 per-
cent of the world trade in these products.

World trade of floricultural products, that is,
cut flowers, cut foliage, and plants, amounted to
US$2.5 billion in 1985, following an increase of
15 percent between 1981 and 1985. Developing
countries supplied a rising share of total world
imports between 1981 and 1985. The share of
developing countries in total world trade was
15.4 percent for all floricultural products com-
bined in 1985, 23 percent for cut flowers, 16.4
percent for cut foliage, and 6 percent for plants.
The major markets were Western Europe and
North America. The European Community (EC)
accounted for 36 percent of the world market, the
United States for 22 percent, Canada 2 percent,
Switzerland 5 percent, and Japan 1.7 percent.
The market prospects of these groups of
commodities are likely to be bright partly be-
cause of a high income elasticity of demand for
them and partly because of a growing tendency
for food consumption patterns in developed and
middle-income developing countries to diversify.
The developing countries have a comparative
advantage in these commodities because of
their general labor intensity.
The horticultural exports of developing coun-
tries are constrained by tariff and nontariff barri-
ers in the importing countries. In general, the
tariffs on horticultural products vary by product,
season, and country of origin. In addition, they
differ depending on the packaging unit and the
content of sugar or other ingredients, as well as
the stage of processing. In the case of nonstor-
able horticultural crops such as fresh vegetables
and fruits, seasonal tariff rates are often applied.
That is, higher tariffs are imposed on imports
during those seasons when they compete with
domestic products. A wide range of nontariff bar-
riers affects the trade in horticultural products.
They include quotas, voluntary export con-
straints, variable levies, minimum price systems,
and countervailing taxes and duties, as well as
technical specifications, especially health re-

strictions and strict labelling and packaging
Table 4.1 provides a brief summary of the
effective tariff rates in selected developed coun-
tries that were prevalent in the post-Tokyo
Round period. However, the average rates
quoted in the table conceal a very wide disper-
sion of nominal tariffs across the various fruits
and vegetables and countries. As is evident from
Table 4.2, for example, in the case of fruit juices,
at the same time that the United States had a
tariff of 52 percent, Australia's was 29 percent,
Japan's 24 percent, and the EC's 7 percent.
Similarly, in the case of potatoes, tomatoes, and
onions, the United States had the highest duty
rates-at 17 percent, almost the same as Fin-
land-followed by the EC and Japan in that
order. The rates of duty not only varied widely
across the importing countries and different
commodities, but each importing country also

Table 4.1-Post-Tokyo round effective
tariffs in developed coun-
tries, 1985

Vegetables Fruits
Country Fresh Processed Fresh Processed


Japan 9.0 17.5 21.5 21.8
EC 6.7 15.1 7.7 16.6
Finland 14.0 14.6 10.7 8.5
Switzerland 4.9 11.4 7.4 13.7
United States 7.6 11.0 1.1 20.3
Australia 3.4 9.8 0.2 6.3
Austria 2.1 13.3 3.1 17.3
New Zealand 0.1 5.7 0.0 11.1
Sweden 5.1 6.2 0.6 0.7
Norway 3.4 6.6 1.0 3.1

Source: Alexander Yeats, "The Escalation of Trade Barri-
ers," in The Uruguay Round. A Handbook on the
Multilateral Trade Negotiations, ed. Michael J.
Finger and Andrzej Olechowski (Washington,
D.C.: World Bank, 1987), 119.
Note: Effective tariff rates measure the influence of pro-
tection on the value added in a production pro-

differentiated the rates of duty depending on the
source of the imports, as is evident from Table
Lower tariff rates were applied to countries
that had preferential arrangements with the
major importing countries. For example, the
United States had an agreement with the Carib-
bean countries and the EC had one with African,
Caribbean, and Pacific (ACP) states. A few de-
veloped countries placed preferential rates on
imports from a large number of developing coun-
tries under the generalized system of prefer-
ences (GSP).
A wide-ranging set of nontariff barriers affects
the horticultural exports of developing countries
(see Tables 4.4, 4.5, and 4.6). First, all the major
importers of horticultural products maintain a
system of marketing orders to regulate and pro-
tect their domestic production, even though the
system differs widely in its restrictive impact on
horticultural trade. Abrupt changes in quality and
packaging requirements imposed by an import-
ing country are sometimes used to reduce im-
ports, as was the case with exports of Mexican
tomatoes to the United States. Many countries
require licenses to import horticultural products,
a measure that increases the transaction costs
of import trade.
Second, quotas and voluntary trade restric-
tions are also quite often used in horticultural
trade as follows:



Pineapples, fresh
or dried
Other fresh
tropical fruit
Provisionally pre-
served fruit, jams,
prepared fruits
Fruit juices
Tomato paste from
Turkey (preferential


France, United Kingdom,
Italy, Portugal, Spain,

Spain, France, Portugal

Spain, France, Greece

France, Italy, Portugal


Japan imposes import quotas on citrus fruits,
fruit juices (orange and grapefruit), prepared
pineapples, dried peas and beans, and other
processed items.
Third, subsidies and price supports are also
provided for horticultural products, along with
other agricultural products, in the United States,
the EC, Canada, and Japan. For example, the
EC subsidizes various processed tomato prod-

Table 4.2-Average post-Tokyo tariff levels in major industrial countries on
imports of selected horticultural products from developing countries

United Canada EC Japan Australia New Austria Switzerland Finland Norway Sweden
Commodity Group States Zealand

1 Potatoes, tomatoes,
onions 17.1 6.7 13.2 5.8 0.5 0.4 0.3 18.8 2.4
2 Vegetables,
frozen 8.3 19.1 17.8 10.0 0.6 -- 21.4 -
3 Vegetables,
preserved 15.8 0.3 2.0 15.0 8.4 3.4 3.9 0.3
4 Vegetables, pre-
served by vinegar 4.4 16.0 3.6 12.7 12.0 10.0 21.7 6.7 30.8
5 Vegetables, pre-
served, n.e.s 11.3 10.3 8.5 21.6 14.9 0.7 11.4 3.1 15.5 1.2 12.3
6 Vegetables, dried 2.3 12.5 10.6 12.3 16.3 0.6 1.1 10.7 3.0
7 Beans, peas, etc. 1.8 0.3 0.2 3.3 7.8 0.3 4.5
8 Tropical fruits
and nuts 0.2 5.8 33.2 0.1 0.1 15.6 17.4
9 Citrus fruits 6.7 4.4 12.2 6.0 8.1 11.9 0.2
10 Raisins 11.3 5.0 2.4 6.5 -
11 Other nuts, n.e.s. 2.3 2.5 14.9 0.7 2.4 2.7 4.5
12 Apples and pears 0.7 1.1 3.6 12.4 2.9 4.2 10.2 3.2 0.1
13 Fruits preserved
by sugar 3.1 0.4 16.0 9.6 29.7 18.3 8.8 17.2
14 Jams and jellies 6.6 5.5 12.4 27.9 13.4 10.1 45.0 5.1 9.9
15 Nuts and fruits
roasted or pre-
served 1.8 0.4 6.2 27.0 4.8 28.9 9.3 12.5 0.3 0.2 0.6
16 Fruit juices
including orange 51.9 0.2 7.1 24.2 28.9 1.0 8.6 11.6 9.9 0.2

Source: United Nations Conference on Trade and Development (UNCTAD) Secretariat, Agricultural Trade Expansion and
Protectionism, with Special Reference to Products of Export Interest to the Developing Countries, TD/B/C.1/239
(Geneva, Switzerland: UNCTAD, January 1983), Annex, 6-7.
Notes: A blank indicates that the pre- and post-Tokyo round tariff averages were zero. n.e.s. = not elsewhere specified.

ucts, canned peaches and canned mixed fruits,
dried plums, dried figs, sultanas, and currants.1
Fourth, the phytosanitary and food health reg-
ulations are an important class of nontariff barri-

ers to horticultural trade. They are a complex
system of specifications enforced by national
and multinational plant quarantine and food san-
itation laws.2

1 See Bale (1986, 10-11). Many horticultural products in the EC are included in the Common Agricultural Policy and therefore
are subject to the whole set of interventions, as is the case with other agricultural products, ranging from variable levies and
reference prices to export and production subsidies. The production subsidies for certain processed fruits and vegetables have
increased supplies in EC countries at the expense of Third World exports. The application of the reference price system has
prohibitive effects on imports of cucumbers and tomatoes into the EC during the spring and autumn seasons.
2 The provisions of national food standards and regulations differ in a variety of ways, ranging from composition to labelling.
These differences often constitute an obstacle to the flow of international trade. Clearly, if these national
differences in import requirements could be harmonized to the point where exporters tend to comply with only one
set of common requirements, it would be very much to their advantage. If market conditions in one country were
more favorable than in another, they could transfer their exports quickly to that country without first having to
ensure whether their products would meet the import requirements. Also, reformulation of products, different
labels, and other matters involve extra costs which could be avoided. Furthermore, food standards and regulations
can be drawn up in such a way that third parties might regard them as unreasonably severe and placing unduly
onerous obstacles in the way of their exports. (GATT 1988,11).

oo o oo o 8
co o 00 COD C O CD .
T- ^- T- T-T-t

> o C.ooooo S
~ -'r qi dc 6 ppd .
LL N - - - r, 'It

0 00 00000000
0D 0O O Oi0 0-T-

o04 0 00
0 0000 0 0 0 0 -0
w No o(0ooooNw
N- (D O O oio- 0
V)) C/) V 6), 0 C j Cin )C l


c ao -0 2
6 6q. C -0

o0 t 0
0 0) 0 0 0) N0


t5 "
0) N,~ O
0) C0)Lf l5
(D c - o o o-
LL o r- o n-6

.02 0oi oioaoaoo.0
(D L - - -T


C4 0) (0D 0

00 0 oo 60

Cu n Np'

0 00 000000ooooo

NC\- -- NN-NN

+ +

0 000

od d
00 0 0 0 M 0 U) U 4FtC

W +W

0) 0)) 0) 0) 0) 0) 0)
0) 0)0) 0) 0)0) 0) 0)0)

0 N

+ +

LLd N-~ ~Vm~~



w w
0) 0) 0)U


1) C 6
LL ) C r,

00 (00(00
ddo ri in en

S00 C0 dd0 0-

o) .

-a -0) a 0

E p 5o 0L 3 C w ) i EE0w)E "i
(n o L) w a0 IC 00 fl So 0 CL V-oS F- M

LL a

0) 0)0)0)

a, 0o

a Q) 0) c)

2 m aL 5 <

a) 0
LL r-

0 0)0) 0)
'0) 0) <






()0) 0)
0 0 T
u_ U- LL

,- C( CO



0 o

(03 C(' CM j CM 0 Cj C\J CMj

CMV 0 00 a o
(Ni 1r 02 cdo -(0)
c\j LO (j 5' (if

iww wwww
LU~~~~~; L U llllO l

U) L) U- U) U) Ui U) L)
DD0 ff0D n f ) 0) D0

CM r- r^ T 0 I
CMj C\J 0 0 T T
w~~~3 6q > /
WWW l U) U) C) U
D D L n- D D"

S.o 6

o S
0\ 00 0- O

0) ()
0) 0) 0)


LOn00 LO)
SCj CM C()

0o 0)

-0 0
* -I
0)- C\j



00 0
00 Co
C\ Cj CMj

(D 0)) a)
S0) 0( 0)

0. 0. -o
o' o
01 r' co 0
0(Ni CM- (03

0) 0)
0) 0)
u_ u_

0)0)D0) 0



0 10

cm a)i
E a

i.- 00

a- ) 0 )

a)CD o<
0)0) 0

-. Ou "- -- *

I- | I-
coaxl CU V

=)VC ( 0) 0

0.S r ni ui Mpn- o-
a) U )2 = 0L ) T)

(0(0QUu 5i~ c~3E.02U)
0. G 0 a 2 m E M *r-
2'0. 0)0 U- UO (D CC 0
F- I

c t..: -; O a
Z = ,)
M o 2 )or
0).O V) -
'- 0 'SE D0
j E a) 0o .

0) 0)-o= ,.
(/3 0)0).^ 1
C U) L)
M: Mn a)- .
E._ 0E


Is? I i_. > III-"
ICOt : -.o
0) 0M E -0
(01^ *j 0) i
(D ) M0 Q -C0
co i n .= 0c =.-o E

M 0 Q
u~i -o) a) C .o a-D 7

CC) to >(7 C
C00L > 0

.2~ -i E2

.0 M2 -0 ( 0>
E, % -S-
,E' ( c E c o a)
0~ x M tz QoCL

- -0- = =0 J >o

*c M a) C: 0 c0
"gn o- c o.-or
.'-< o 0 c r o So r-
Mo 0 CO >0''

ll II. 8| Si( l
.--" "--

aic (O ur Sn .-S ^
L r

io5.3 o ==p g
a) 4-C ) CL o0

00C D : oc m c. < o uIn
E. 0
4) U) E~ -

ilD iz- .nll zi0
CO0. D C|0 | L -
"--0 0.. 0Q 0
"('n o L,- -r : E = o ,'.
(n 0) 0 o -
=< < = C-u : -Z- <0
0)r-.CM COC (0 C

v: (n !2 s LO -
O':0 U) O (
.6C, 3:(n -' CD~ L)f0)C UC

-).. -(T 0 0 .- C O
!E c in L -- o 0)
M ,O L M t-0 E (n
0 L_: ) co C

_0 C ,,- E ccu a" M ,, c
Zn cOo zCD CO;

) CO Z o )O -E M t M c
Qv "-, a M CL :' 0 'n in/'

(n e M,'o E > E
m o- E M .=- ,
.0 'c 00 .00 n)CD0t5.c 0c0)

a) 0 ) a)I (-.0 M Q) m ) ,0

< ':5oJ|1P0gl -S
0)0 :R a) a ) ILE ~C L) 0
Et 00 M :3a) -Cl a) U

a) (O CLU QCL .2 2-0 .0 tcC 0 3 *
'0 -. .5- V 0) 0 r- o M 15
0 -- c

CL c a))~ (D0 cuji E
Q 3 -2 CO M E 'n CO
E0 - 0 --2 m) E~ -
o ~ -10 U)O:OC0 3)
a)c 0 IZ )Q) 7 a .0

U) a.co M C 0
-0 00 0= )C
-0 z3, 0) (6 .2 U)0> M ) 3 CO Sa)0
a) -0 ~ >~ ~-C 3
>- M. -- CL

mO (D crn 'r-O. xU)DZ-

Table 4.4-Escalation of tariffs between fresh and processed products and
nontariff barrier (NTB) coverage ratio in industrial countries

Average Applied Tariff NTB Coverage
Product Pre-Tokyo Post-Tokyo Ratio

Fresh 13.3 8.9 39
Processed 18.8 12.4 48
Fresh 6.0 4.8 20
Processed 17.0 14.4 54

Source: Based on data from United Nations Conference on Trade and Development (UNCTAD), Trade Development Report
(Geneva: UNCTAD, 1984), 201.

Table 4.5-Nominal tariffs and the estimated ad valorem equivalent of nontariff
barriers (NTBs) in France, Japan, Sweden, and the United States in
the mid-1970s

Vegetables and Products Edible Fruits and Nuts
Nominal NTB-Price Nominal NTB-Price
Country Tariff Relative Tariff Relative

France 12 30 14 22
Japan 12 72 14 180
Sweden 7 80 4 51
United States 13 50 7 128

Source: Based on data from United Nations Conference on Trade and Development (UNCTAD), Trade Development Report
(Geneva: UNCTAD, 1984), 201. The tariff code for vegetables and products is CCCN-07 and for edible fruits and nuts
CCCN-08. These two codes, however, do not cover all fruits and vegetables; a few items are covered under other
codes such as Code 20.

It is very hard to distinguish unnecessary re-
strictions from real ones since standards and
safety concerns over food sanitation and dis-
ease infestation differ widely across countries.
Furthermore, national standards and regulations
are liable to change from time to time, some-
times abruptly or without prior notice. Exports of
a country oriented toward meeting the food stan-
dards of an importing country on the basis of
past practice and rules often become unaccept-

able and need to change to meet new standards.
This situation creates uncertainty in the export
It is expected that if national standards are
harmonized in accordance with the agreed
guidelines for international standards formulated
by consensus, the regulations and their pur-
poses are likely to become transparent so that
obviously protectionist components may be de-

> >
>0COQ 0





















00 000 P


00 00 0 0

0 PP

00 p


0 0



0 00

0 0


Q > Q


-' "- V '

co o > 8
a U) .
Na)> a) a

S >n CL C. a g
Su a ;6a)
7" ui .0 _
( ca (L. > > > > >


3 S

O V aU)
0f a) 0?
I a J Z
o *- U" E U '-
o 0) C 2 ? _.

o 0 0 ci, 003

ILL d)
*b p

I PI : I



co a a )
o. i

u. )E

< E

0 C. >
8Q 0

QD C:; a) S 62)
Co =- E

A a) c a)i

E --0 c :3 c m
co ^ u r -D

.9 c to Z a
UJ a) n o E o .2

0c, c>
I ,.. O

S0 0 m W C ." ". L
C5) u) co

"6 a a, m .

c .,


,E "" .2

0) 0 x a
Q) c u (

a) U a) m a)
M m c DN

,^0) ,01z U) C1 (
Q 0.

Z Z U) DD a)
-1 isj co

a) CLL Ca ) inS
II .1

0 m c -0 E

: Ear ^j c a)













Floricultural exports face tariffs and nontariff
barriers as follows:

Import Tariffs (percent)


Cut flowers 4.8 17-24 (15-24 GSP) 40
Foliage plants 7.5 10 (7 for GSP) 40

Selected plants

8-13 (6-12 GSP)

As the table shows, the United States allows
duty-free entry to designated beneficiaries of the
GSP and members of the Caribbean Basin Initia-
tive; the EC provides duty-free entry to countries
of the Lome Convention and to developing coun-
tries; and Canada provides free entry to large
numbers of countries that historically were mem-
bers of the British Preferential Tariff in Africa,
Asia, and the Caribbean.
So far as nontariff barriers are concerned, in
the United States cut flowers are subject to in-
spection for compliance for phytosanitary regula-
tions that are frequently revised. Further, the
importation of plants with soil/or soil attached is
prohibited except from approved nurseries. The
EC has laid down quality standards that many
other developed countries follow; it has also es-
tablished standards for packaging and presenta-
tion of certain types of flowers for their size and
length, and for other factors. Each unit of presen-
tation (such as bunch, bouquet, or box) must
contain flowers of the same genus, species, and
variety (cultivar) and of the same quality class
and must have reached the same stage of matu-
ration (ITC [UNCTAD/GATT] 1987, 2-10 and Ap-
pendix I-IV). Japan is alleged to use phytosan-
itary regulations to protect its fresh cut flower
industry from import competition. Inspection pro-
cedures are so exacting, take such a long time,
and involve such poor handling, that they cause
extensive damage to the imported products
(U.S. ITC 1989, 7-3 and 7-4).


The increase in earnings from horticultural
exports that would follow from the elimination of
tariffs by the United States, the EC, and Japan is
estimated to vary from 6 percent to 9 percent. If.
the nontariff barriers are totally eliminated, the
increase in export earnings is estimated to vary
from 24 percent to 36 percent (Table 4.7). The

percentage increase in export earnings is higher
for fruits than for vegetables. In part the reason
is the much higher incidence of nontariff barriers
on fruits and the higher initial quantity of exports
of fruits (Islam 1990).
The maximum increase in export revenues of
developing countries if only tariffs are eliminated
is estimated at US$850 million and if the nontariff
barriers are eliminated at US$3.3 billion.3
A recent study by the United Nations Confer-
ence on Trade and Development (UNCTAD) of
the impact of the elimination of tariffs on selected
horticultural products estimates an increase in
export earnings of an average of 15 percent
(UNCTAD 1985, 213). Another study estimated
an increase of 13 percent for a group of mostly
fresh products that have lower rates of duty
(Vald6s and Zietz 1980).


The prospects for the liberalization of trade in
horticultural products are linked with the success
of the Uruguay Round of the Multilateral Trade
Negotiations. The Uruguay Round of the trade
negotiations on seven agricultural commodities
is being carried out in two separate committees,
the Committee on Agriculture and the Committee
on Tropical Products. The latter explicitly in-
cludes negotiations on seven product groups in-
cluding tropical fruits and nuts, flowers, and spe-
cific plants. There is no mention of other fruits or
vegetables. For this reason, and because the
Committee on Agriculture also excludes them, it
is not clear whether negotiations on these com-
modities will take place. The implicit understand-
ing, at least in the early stages, was that the
category defined as tropical commodities was of
special interest to the developing countries. Fur-
thermore, developed and developing countries
were not competing much in the world trade in
most of these commodities, and therefore liber-
alization could be undertaken to a large extent
and at a relatively fast rate without adversely
affecting the exports or domestic production of
developed countries.
Two agreed-upon provisions of the GATT ne-
gotiating mandate on tropical products are rele-
vant in this context. First, "Negotiations shall aim
at the fastest liberalization of trade in tropical
products, including processed and semi-pro-
cessed forms, and shall cover both tariff and

3 The consequences of trade liberalization through a reduction in tariffs are estimated for four major categories of horticultural
exports: fresh vegetables, processed vegetables, fresh fruits, and processed fruits.

non-tariff measures affecting trade in these prod-
ucts" (UNCTAD 1988, 356). Second, "the con-
tracting parties recognize the importance of
trade in tropical products to a large number of
less developed contracting parties and agree
that negotiations in this area shall receive spe-
cial attention, including the timing of negotiations
and the implementation of the results" (ibid.).
Since most of the horticultural products are
not specified for negotiation by the Committee
on Agriculture, it is urgent that this issue be
clarified and that the developing countries seek
their inclusion under one of the committees, in-
cluding those on tariffs and nontariff barriers.
To make matters worse, concessions so far
granted by the industrialized countries in the
Committee on Tropical Products do not cover all
the commodities that have already been placed
under the purview of the committee.
What is most important for the growth of the
horticultural exports of developing countries is a
liberal trade regime and its stability and certainty
over time, since many of their exports are new or
have been introduced into world trade only in
recent years. They require investment in export
infrastructure, including marketing and distribu-
tion facilities. The prospects for market develop-
ment are likely to be brighter for the processed
than for the fresh products; in this regard, tariff
escalation by degree of processing is important.
Without a substantial liberalization of trade in the
processed horticultural products, the expansion
of exports will be limited.
The GATT negotiations on sanitary and phy-
tosanitary regulations are particularly important
for horticultural products. The GATT midterm re-
view in April 1989 did agree to develop
... harmonization of sanitary and phytosanitary regu-
lations and measures on the basis of appropriate
standards established by relevant international orga-
nizations, etc.; also, to ensure transparency and the
existence of an effective notification process for na-
tional regulations or bilateral agreements; to allow a
consultation process which ensures an opportunity for
the bilateral resolution of disputes; to improve the

Table 4.7-Results of trade liberaliza-
tion: tariffs and nontariffs
(percentage increase in
export earnings), 1983 to

Export Elasticity
Commodity Import Elasticity 0.5 1.0

Vegetables -0.5 4 5
(10) (12)
-1.0 5 7
(13) (18)
Fruits -0.5 8 8
(35) (32)
-1.0 10 11
(47) (49)
Total -0.5 6 7
(24) (24)
-1.0 8 9
(33) (36)

Source: Calculated by the author.
Note: The numbers in parentheses are increases result-
ing from the elimination of nontariff barriers.

effectiveness of multilateral dispute settlement pro-
cess; to provide necessary input of scientific expertise
and judgment, relying on relevant international orga-
nizations (GATT 1989).
An important commitment made in this re-
spect is to "... assess the possible effects on
developing countries of the GATT rules and dis-
ciplines for sanitary and phytosanitary measures
and evaluate the need for technical assistance"
(GATT 1989). This point is of particular impor-
tance to developing countries in view of the cru-
cial role of such regulations in determining the
flow of trade in horticultural products.

Bale, Malcolm D. 1986. Horticultural trade of the expanded European Community: Implications for
Mediterranean countries. Washington, D.C.: The World Bank.
GATT (General Agreement on Tariffs and Trade). 1988. Sanitary and phytosanitary regulations: Role
and status of work of selected organizations. Geneva. Mimeo.
1989. Gatt newsletter: Uruguay Round special issue. May.
ITC (International Trade Center [UNCTAD/GATT]). 1987. Floricultural products: A study of major
markets. Geneva.
Islam, Nurul. 1990. Horticultural exports of developing countries: Past performances, future prospects
and policy issues. IFPRI Research Report 80. Washington, D.C.: International Food Policy Re-
search Institute.
UNCTAD (United Nations Conference on Trade and Development). 1985. Trade and development
report. Geneva: UNCTAD.
1988. Uruguay Round: Papers on selected issues. Geneva: UNCTAD.
U.S. ITC (International Trade Commission). 1989. Competitive conditions in the U.S. and world markets
for fresh cut roses. Washington, D.C.
Vald6s, Alberto, and Joachim Zietz. 1980. Agricultural protection in OECD countries: Its cost to less
developed countries. IFPRI Research Report 21. Washington, D.C.: International Food Policy
Research Institute.


The Impact of Trade Liberalization on Low-Income,
Food-Deficit Countries

Thomas W. Hertel"

In some sense the issue of trade liberalization
in agriculture may be viewed as a debate over
how much of the global supply of food will be
produced in the industrial market economies and
how much in the lower income countries. The
current configuration of farm and food policies in
the wealthy countries has generally served to
retain productive resources in agriculture and
maintain-if not accelerate-the rate of technical
change in this sector. By contrast, poorer coun-
tries have tended to discriminate against the
farm sector (World Bank 1986). Although there is
considerable variability in the way developing
countries treat specific commodities, the overall
outcome, when both direct and indirect effects
are considered, has been to discourage agricul-
tural investment and innovation (Krueger,
Vald6s, and Schiff 1988). As a consequence,
quantitative analyses of trade liberalization typi-
cally predict a reallocation of food production
from North to South.
The mechanisms motivating this shift in food
production are analyzed below. For purposes of
this paper, these effects are divided into two
groups: those generating higher world prices for
food; and those that alter relative prices for food
within individual countries. The first of these ef-
fects has led to great concern on the part of
food-deficit countries, especially those with low
incomes and severe foreign exchange con-
straints. How will they pay the bill for more costly
food imports? What will be the consequences for
the poorest segments of their populations if do-
mestic food prices are permitted to rise? What
other imports must be displaced to maintain food
imports at their old levels? These concerns have
motivated low-income, food-deficit countries to
request compensation to offset the deleterious

*The author would like to thank Philip Abbott, Lowell Hardin,
helpful comments on an earlier draft of this paper.

effects of agricultural trade liberalization (GATT
Once the question of compensation is raised,
issues of the appropriate amount (if any) and
duration for such assistance necessarily follow.
These questions cannot be answered without
reference to the second mechanism by which
global food production is transferred from North
to South, namely, altering domestic prices for
farm and food products. To the extent the rela-
tive change in world prices is transmitted into the
domestic economy, it will be shown that the im-
pact of higher food prices on the food deficit
economy's import bill is dampened. In some in-
stances this situation leads to a decline in a
country's food import bill, as consumption is re-
duced and domestic producers expand produc-
tion. This effect is even more dramatic when
food-deficit economies simultaneously liberalize
their own policies, a step that stimulates addi-
tional food investment and innovation in their
food sector.


Defining the Scenario

The projected impact of agricultural trade lib-
eralization on world prices depends critically on
the nature of the liberalization scenario consid-
ered. First, what sectors are to be liberalized?
For example, research on the Australian econ-
omy has shown that unilateral removal of farm
support lowers agricultural output, unless nonag-
ricultural protection is simultaneously removed
(Higgs 1989). Because the nonfarm sector is

Marshall Martin, Philip Paarlberg, and Wallace Tyner for their

relatively more heavily protected, economywide
liberalization actually leads to an expansion of
the farm sector's output. Because of the dearth
of economywide studies, the focus here is only
on the results of studies that have analyzed ag-
ricultural liberalization alone.
A second question that must be addressed is,
What countries are engaged in the liberalization
process? If the scenario involves only liberaliza-
tion within the Organisation for Economic Co-op-
eration and Development (OECD) countries,
then the expectation is that prices will rise. How-
ever, it has been shown that liberalization in
much of the developing world could serve to
depress world prices (Goldin and Knudsen
1990). In most of this paper the emphasis is on
the question of OECD liberalization, since it is
where the empirical models are on firmest
Two more issues that are critical to answering
these questions involve the depth of the cuts in
farm support and the timeframe over which they
are to be implemented. Most empirical studies
analyze complete liberalization (100 percent
cuts in farm support) because it is the easiest
case to apply in a quantitative model. Analysts
operate on the presumption that the relative
magnitude of the price, quantity, and welfare
effects of partial liberalization would be similar,
provided that this liberalization is proportional
across commodities, countries, and instruments.
Regarding the timeframe, few studies presume
to be able to predict the very short run (less than
one year) consequences of trade liberalization.
Short-run movements in world prices tend to be
dominated by random weather shocks, specula-
tion, and changes in stocks. The long run, say,
more than five years, is also difficult for econo-
mists to analyze, since the impact of liberaliza-
tion on investment, economic growth, and tech-
nological innovation are all areas in which the
quantitative models are on relatively weak
ground. Consequently, most studies focus on a
period of one to five years.
Finally, there is the question of how the pro-
jected prices are to be interpreted. Most models
of agricultural trade report price changes relative
to what they would have been in the absence of
liberalization. However, these hypothetical price
changes are not likely to be the same as what
are observed in practice. The reason is that
many other factors are also likely to change. In
fact, as pointed out by Tyers in Chapter 7, the
historical variability of food prices is far greater
than the price changes projected by virtually all
models of trade liberalization. Furthermore,
some authors have argued (for example, Valdes
and Zietz in Chapter 2) that since the long-run

trend in the relative price of food has been down-
ward, gradual liberalization of farm programs
might not raise prices at all-but would simply
serve to dampen their decline. Because of the
difficulty of projecting future price levels, the
focus here is on the hypothetical price changes
following agricultural trade liberalization.

Assessing the Consequences of
OECD Liberalization

Given the enormous consumer and taxpayer
transfers to agriculture in the OECD countries-
an amount in excess of US$200 billion in 1988
(OECD 1989)-a quantitative model is hardly
needed to predict that policy changes leading to
the removal of these transfers would generally
raise the level of world food prices. However, to
assess the consequences of such reforms on
specific food-deficit countries, it is necessary to
have some idea of the magnitude of the price
increases for specific commodities. Unfortu-
nately, the task of simulating such a massive
reallocation of resources is not easy, and exist-
ing trade models often give very different esti-
mates (Goldin and Knudsen 1990, Table 17.1).
This section discusses a few of the major
sources of the discrepancies in estimated
changes in world prices. The examples chosen
are relevant to the grains markets, since grains
represent a substantial portion of the food import
bill for the low-income, food-deficit countries.
Choice of Base Year. In some cases, differ-
ences among the results obtained with different
models may be traced to fairly obvious differ-
ences in assumptions. For example, because of
the volatility of international prices and changing
national policies, the measured level of farm
support in OECD countries has varied consider-
ably from year to year-both across commodi-
ties and countries (OECD 1987, 1989). By con-
trolling for this difference, Magiera and Herlihy
(1988) were able to reduce dramatically the ap-
parent discrepancies in the predicted changes in
world price of several of the most widely cited
models (see also Dixit, Herlihy, and Magiera
Treatment of Policies. There are also other,
less obvious sources of discrepancies in the pre-
dictions of these trade models. In practice, farm
and food policies are extremely complex. Some-
times they are also contradictory in their effects.
However, the trade models are generally forced
to deal with these policies in a fairly simple
way-often boiling down the entire set of policies
into a single production subsidy equivalent

By way of illustration, consider the U.S.
grains programs. Participating producers are
asked to idle acreage, but they receive a "bribe"
for doing so. It comes in the form of a deficiency
payment, which is linked to the allocation of
acreage among alternative uses, which was his-
torically linked to the output decisions of produc-
ers. In practice, administrators use the acreage
reduction component of the grains programs to
dampen the production and budgetary effects.
For example, when expected deficiency pay-
ments are large, the requirement for idle acreage
is also large. Thus the effect of the U.S. grains
programs on world prices is ambiguous. Will the
returning acreage outweigh the disincentive of
reduced deficiency payments? Virtually all of the
trade model results published to date have as-
sumed a strong link between deficiency pay-
ments and the incentive to produce. However,
since 1985 this incentive effect in the United
States appears to have been weakened by the
imposition of a freeze on the yields used to cal-
culate the total deficiency payments. As a result,
most of the trade models have likely overstated
the increase in world grain prices following elim-
ination of the U.S. grains programs.
Reactions of Producers. Yet another difficulty
in projecting world price changes following trade
liberalization in agriculture is the question of pro-
ducer response to the new policy and market
environment. One particularly problematic area
has to do with the response of livestock produc-
ers to altered feedstuff prices. Since livestock
comprise such a large component of the total
demand for grains by the OECD, the response of
these producers is crucial. How much will pro-
duction change? If grain prices rise, to what
degree will producers substitute other feedstuffs
for higher priced grains? The latter question ap-
pears to be the most controversial.
If livestock producers respond to altered in-
ternal prices for grain by significantly altering
their feed mix, then the net effect of liberalization
on world markets will be much smaller. Thus it is
not surprising that the model which places the
greatest emphasis on feedstuff substitutability
(OECD's MTM model) shows the least propen-
sity for a rise in grain prices. By contrast, some
of the other models (such as the U.S. Depart-
ment of Agriculture's [USDA] SWOPSIM model)
do not permit feedstuff substitution to influence
the world demand for grains. Not surprisingly,
their simulation results show some of the largest
changes in grain prices. (See Goldin and Knud-
sen 1990, Table 17.1 for a comparison of re-
It is hoped this brief digression into the con-
sequences of OECD liberalization for world grain

prices has provided a flavor of why it is so diffi-
cult to assess the commodity-specific conse-
quences of such reforms. It is the author's judg-
ment that, while world grain prices may indeed
rise (especially for countries that have enjoyed
export subsidies in the past); the magnitude of
this increase may have been overstated in much
of the work reported to date.


In their October 1989 proposal to the Negoti-
ating Group on Agriculture, the group of food-
deficit countries presented some estimates of
the potential increase in net import costs for
basic foods following OECD agricultural liberal-
ization. By multiplying one set of projected price
changes (IATRC 1988) by their 1984 to 1986
average import expenditures, they arrived at the
following percentage increases in net import
costs: Egypt, 29 percent; Jamaica, 30 percent;
Mexico, 24 percent; Morocco, 28 percent; and
Peru, 33 percent. Even if the possible overstate-
ment of the hikes in grain prices noted above is
taken into account, these figures are staggering
for a group of countries that are already hard-
pressed to meet their foreign exchange require-
ments. In fact, any increase in the price of im-
ported foodstuffs will be difficult for these
countries to absorb-hence the request for com-
While the approach of multiplying projected
price changes by a country's net import bill pro-
vides a useful initial assessment of the potential
impact of such a shock, it suffers from an impor-
tant shortcoming. It implicitly assumes that the
importing country in question does not alter the
quantity of food it consumes and produces-the
difference between these two being net imports.
Since the price shocks simulated by most trade
models refer to the medium to long run, and
since all reform proposals involve a gradual
phaseout of OECD subsidies (for example, over
10 years), the potential for adjustment on the
part of food importers would seem to be consid-
erable. How might this adjustment affect their
import bill?

Domestic Adjustments and the Cost
of Imports

In a study conducted on behalf of the U.S.
Agency for International Development (USAID),
Quizon, Gardner, and Quinn (1988) address this
very issue. They begin by postulating world price
increases for the individual commodities dis-

played in Table 5.1. The size of these increases
is given in the first column of the table (they are
taken from earlier studies of Tyers and Anderson
1986, Valdes and Zietz 1980, and Zietz and
Vald6s 1986). (As noted above, there are many
reasons to question any given set of price projec-
tions, but for present purposes the only critical
thing is that they are all positive numbers.) The
authors then proceed to simulate the response of
a large number of individual USAID-assisted
countries to this price shock.
A critical assumption in their analysis is that
these proportional changes in world prices are
fully transmitted into the domestic economy.
However, this process does not mean that the
new domestic prices are equal to world prices.
For example, if food grain prices are initially half
the world price and the latter increases by 10

percent, then Quizon, Gardner, and Quinn's as-
sumption is that the domestic price also in-
creases by 10 percent. Of course, there are
many reasons why food-importing countries
might attempt to insulate their consumers from
this price increase (this issue is explored in detail
below). For now, it is useful to follow through with
the consequences of full price transmission of
the food import bills of these countries.
When world price Changes are passed
through to domestic consumers, they respond by
buying less food. However, since food is a ne-
cessity, there is a limit to how much consumers
can cut back on consumption. In their study,
Quizon, Gardner, and Quinn assume that a 10
percent increase in domestic prices reduces de-
mand for most commodities by 3 to 4 percent.
This cutback may involve switching to other food

Table 5.1-Summary of the short- to medium-run impact of OECD liberalization
on the trade status of a selection of food-importing developing
countries: full price transmission assumed

Number of
Price Shock Number of Reversals
(Percentage Importers Higher Lower (Become
Commodity Change) Region Considered Import Bill Import Bill Exporters)

Sugar 16.7 Latin America 5 1 1 3
Asia 15 5 8 2
Africa 16 8 5 3

Wheat 12.7 Latin America 10 9 1 0
Asia 15 7 5 3
Africa 23 21 2 0

Maize 11.7 Latin America 7 2 4 1
Asia 6 3 2 1
Africa 14 1 6 7

Rice 10.0 Latin America 10 2 5 3
Asia 7 3 1 3
Africa 19 10 9 0

Soybean oil 3.0 Latin America 11 11 0 0
Asia 10 10 0 0
Africa 6 5 1 0

Palm oil 2.9 Latin America 0 0 0 0
Asia 11 10 1 0
Africa 7 7 0 0

Coffee 2.3 Latin America 0 0 0 0
Asia 0 0 0 0
Africa 0 0 0 0

Source: J. Quizon, B. Gardner, and L. Quinn, The Consequences of Agricultural Trade Liberalization for Developing
Economies Assisted by AID, a report to USAID prepared by Wharton Econometrics and Louis Berger International,
Table 9, Table 17, and Appendix. The authors considered only USAID-assisted countries in their analysis.

sources, or it may involve a reduction in total
caloric intake. To the extent that all food prices
rise, the opportunity for switching to lower cost
foods will be limited and the reduction in demand
for any given commodity will be smaller (but
nutritionally more damaging) than that predicted
in their study.
Producers also respond to increased domes-
tic prices. In fact, for most commodities in most
regions, here is where the greatest potential for
reducing imports lies. Of course, in the very short
run (that is, less than a year for annual crops),
producer supplies are not responsive to price.
Even beyond that period the supply response of
the aggregate farm sector is somewhat limited
because of the relative fixity of land, labor, and
capital (see Binswanger 1990 for an extensive
discussion of this point). In calculating the ef-
fects shown in Table 5.1, Quizon, Gardner, and
Quinn (1988) assume "short- to medium-run"
supply elasticities for individual crops. In most
cases these still imply greater adjustment in sup-
ply than demand.
The net effect of the postulated price in-
creases on a selection of food importers is re-
ported in the last three columns of Table 5.1.
Note that in some cases the projected domestic
response to these price shocks is sufficient to
turn net importing countries into net exporters.
For example, only 6 of the 27 maize importers
considered are projected to experience an in-
crease in import expenditures on maize, despite
the fact that the world price of maize rises by
11.7 percent. As to some of the other commodi-
ties, 22 out of 36 sugar importers, 11 out of 48
wheat importers, and 21 out of 36 rice importers
actually experience a decline in their import bill.
Since a relative increase in domestic food
prices may be expected to stimulate additional
investment and innovation, the long-run pros-
pects for reducing the food import bill are poten-
tially much greater than those shown in Table
5.1. For example, in their analysis of the aggre-
gate agricultural supply response in Chile,
Coeymans and Mundlak (1989) estimated that
the first year's response to a 10 percent agricul-
tural price hike is only about 2 percent, but the
long-run response was in excess of 14 percent.
Anderson and Tyers (1990) have attempted
to incorporate this long-run flavor in their most
recent analysis of trade liberalization in agricul-
ture. Some of their results are presented in Table
5.2. For the moment look at the first two columns
of the table. These report the ratio of food pro-
duction to consumption in the OECD as a whole
and in a selection of developing countries before
and after liberalization of the OECD's food poli-
cies. They project a dramatic reallocation of food

production, with the OECD's self-sufficiency
ratio falling from 109 percent to 83 percent. As a
result of these authors' assumption of full price
transmission, the self-sufficiency ratio increases
for virtually all non-OECD countries, with the
developing countries as a group becoming net
suppliers of food products after OECD liberaliza-
All of this adjustment comes at a cost. As the
second-to-last column in Table 5.2 indicates,
even in the long run OECD liberalization leaves
a number of developing countries, and the whole
of Africa, worse off. That is, gains to food produc-
ers are still not as large as losses to consumers.
Consequently, net welfare falls.

Implications for Compensation

The results in Tables 5.1 and 5.2 indicate
that, over time, the need for compensation might
be reduced, if food importers were to convey the
changes in international prices into their domes-
tic economies. In fact, to the extent that the
OECD reforms are announced in advance and
phased in gradually, adjustment of the sort por-
trayed in these tables might be achieved in a
fairly smooth manner. However, adjustment will
still involve higher food prices domestically, a
change that will certainly be opposed by con-

sumer groups. Even a small increase in the price
of food can cause great hardship for low-income
households. Thus there may be a considerable
incentive to avoid achieving this type of adjust-
ment by failing to transmit the higher interna-
tional food prices into the domestic economy.
Low-income countries will be constrained in
the degree to which they can insulate the domes-
tic economy from higher world prices. As was
seen above, such an effort might cost as much
as one-third of some countries' food import bills.
Consequently, in the absence of compensation
these countries may have no choice but to pass
the price increases through to the domestic
economy. However, if international compensa-
tion is implemented to pick up the higher cost of
food imports, then the pressure to pass the price
increases through to the domestic economy may
be reduced. This likelihood is particularly true if
the compensation is viewed as more than a tem-
porary measure.
In summary, liberalization of OECD farm and
food policies can be expected generally to raise
the level of food prices worldwide. The degree to
which this process raises the import bill of food-
deficit countries depends critically on their will-
ingness to transmit this price signal into their
domestic economies. If they do relay this signal
to producers and consumers, their compensa-

Table 5.2-Long-run effects on individual countries of both OECD and OECD plus
developing country food policy liberalization

Change in Economic Welfare
Food Self-Sufficiency Within the Food Sector, Constant 1985
(Production as a Percent of Consumption) US$ Billion per Year for
Liberalization in Liberalization in
Countries Preference OECD OECD + LDCs OECD OECD + LDCs


OECD Total 109 83 74 51.1 61.0
Bangladesh 61 70 70 -0.0 0.9
China 104 117 122 3.2 15.2
India 84 99 98 -1.0 0.2
Indonesia 103 112 60 -0.4 2.2
Korea, Republic of 65 73 14 -1.3 6.5
Pakistan 84 116 141 0.9 2.6
Philippines 104 114 119 0.2 0.2
Taiwan 81 88 42 -0.2 0.8
Thailand 156 173 159 0.5 -0.5
Other Asia 63 71 116 0.1 4.7
Subtotal Asia 2.0 32.8

Argentina 173 213 345 3.9 7.3
Brazil 156 186 215 4.9 3.7
Mexico 95 120 115 0.4 1.4
Other Latin America 182 226 283 6.2 4.4
Subtotal Latin America 15.4 16.8

Egypt 30 38 43 -1.1 1.0
Nigeria 48 57 27 -0.6 0.6
South Africa 62 78 59 -0.5 -0.2
Other Sub-Saharan Africa 45 57 113 0.6 6.6
Other North Africa
and Middle East 46 58 48 -4.3 -1.3
Subtotal Africa -5.9 6.7

Total developing
countries 92 111 118 11.5 56.3
World total 100 100 100 65.0 120.3

Source: K. Anderson and R. Tyers, "How Developing Countries Could Gain from Food Trade Liberalization in the Uruguay
Round," in Agricultural Trade Liberalization: Implications for Developing Countries, ed. I. Goldin and O. Knudsen
(Paris: OECD and the World Bank, 1990).

tion requirements will diminish over time. Thus
there is a natural tension between the low-in-
come, food-deficit countries and the donor coun-
tries that would be providing compensation for
the increases in world food prices. The former
group may have an incentive to insulate their
domestic economies from the higher prices,
while the latter will likely have an interest in
encouraging adjustment to the new state of af-
fairs. One possible compromise might involve
using the donor's compensation to supplement
the income of the poorest households, perhaps
through public works programs or even food
stamps-provided the market price of food is
permitted to adjust upwards. This sort of mea-
sure would soften the blow to consumers and

concomitantly preserve the incentive to adjust
and hence reduce the food import bill over time.

The Role of Food Importer Policies

To this point, the possibility of farm and food
policy reforms outside the OECD countries has
not been discussed. Rather, it has simply been
assumed that existing price disparities remain in
place. Thus, in the event of full price transmis-
sion, the higher world food prices are simply
"tacked on" to the old price structure in develop-
ing countries.
What would happen if these countries simul-
taneously liberalized their agriculture? The likely
result would be gains in efficiency, as domestic

resources are reallocated to reflect their produc-
tivity better at world prices. Anderson and Tyers
(1990) have estimated the food sector welfare
effects of this simultaneous policy liberalization.
These are reported, along with the associated
self-sufficiency ratios, in the right-most columns
of Table 5.2. Note that developing countries as a
group further increase their agricultural output
relative to consumption. In addition, the net
gains in food sector welfare increase substan-
tially over the OECD liberalization case. This
result is particularly true for Asia, where a num-
ber of food-deficit countries now become net
welfare gainers. Finally, note that the welfare
gains of the OECD countries are further en-
hanced by the simultaneous participation of de-
veloping countries in liberalization of the food


There is little doubt that the short-run effect of
the higher food prices expected to accompany
the OECD's agricultural liberalization will hurt
low-income food importers. Even though the
magnitude of this shock can be minimized by
spreading it over a number of years, the fact
remains that a product for which they are cur-
rently net importers will become more expen-
sive, at least relative to what it would otherwise
have been.
The long-run consequences of this price
shock depend very much on how the individual
countries handle it. If food importers choose to
insulate their domestic markets from the price
shock, then the short-run and the long-run con-
sequences are the same-more expensive im-
ports. If the price signal is instead transmitted to
domestic consumers and producers, it is con-
ceivable that some food-deficit countries may
eventually revert to a surplus position. In any
case a significant reallocation of food production

from North to South can be expected. If, in addi-
tion, low-income food importers take this oppor-
tunity to reassess their domestic policies and
remove existing distortions, then in the long run
net welfare gains are quite likely.
Since many of the world's poor reside in rural
areas and rely on agriculture and related indus-
tries for their livelihood, a healthy farm economy
will help put income in the hands of many who
need it most. However, the inexorable migration
of population to urban areas is an indication of
the fact that farming cannot provide an adequate
livelihood for all rural residents. This fact is es-
pecially relevant to the food-deficit countries,
many of which do not enjoy a comparative ad-
vantage in farming. Thus it seems that the inter-
ests of the low-income, food-deficit countries in
the current round of the GATT negotiations
would be well served by seeking an agreement
whereby they accept liberalization of their agri-
cultural policies in exchange for improved ac-
cess to the nonagricultural markets in the OECD
countries. This strategy would create additional
jobs and income in the nonfarm economies.
After all is said and done, the hunger problem
is ultimately a problem of insufficient income.
Matthews (1985, 8-9) has calculated that the
cereal equivalent of the aggregate calorie deficit
for the world's malnourished population is less
than 4 percent of total production. At the present
time the world appears to have the capacity to
produce sufficient food. The critical issue is
where this food will be produced and who can
afford to purchase it. Will it continue to be pro-
duced disproportionately in the OECD coun-
tries-at considerable expense to both taxpay-
ers and consumers and to Third World farmers?
Or will farmers in the developing countries re-
ceive improved production incentives, so that
they increase their share of global food produc-
tion and income? The answers to these ques-
tions will depend importantly on the outcome of
the current round of GATT negotiations.

Anderson, Kym, and Rod Tyers. 1990. How developing countries could gain from food trade liberaliza-
tion in the Uruguay Round. In Agricultural trade liberalization: Implications for developing countries,
ed. lan Goldin and Odin Knudsen. Paris: Organisation for Economic Co-operation and Development
and The World Bank.
Binswanger, H. 1990. The policy response of agriculture. In the Proceedings of the World Bank Annual
Conference on Development Economics, 1989. World Bank Economic Review 4:231-258.
Coeymans, J. E., and Y. Mundlak. 1989. Agriculture and sectoral growth: Chile, 1962-82. International
Food Policy Research Institute, Washington, D.C. Mimeo.
Dixit, P. M., M. T. Herlihy, and Steven L. Magiera. 1989. Global implications of agricultural trade
liberalization. In Agricultural-food policy review: U.S. agricultural policies in a changing world,
Chapter 16. USDA/ERS Agricultural Economic Report No. 620. Washington, D.C.: U.S. Department
of Agriculture.
GATT (General Agreement on Tariffs and Trade). 1989. Ways to take account of the negative effects
of the agriculture reform process on net food importing developing countries. Proposal submitted by
Group W/74 to the Negotiating Group on Agriculture. Geneva, October 25.
Goldin, I., and O. Knudsen. 1990. The implications of agricultural trade liberalization for developing
countries. In Agricultural trade liberalization: Implications for developing countries, ed. lan Goldin
and Odin Knudsen. Paris: Organisation for Economic Co-operation and Development and the World
Higgs, Peter J. 1989. The taxation of Australian agriculture through assistance to Australian manufac-
turing. In Macroeconomic consequences of farm support policies, ed. Andrew B Stoeckel, David
Vincent, and Sandy Cuthbertson, Chapter 10, 293-329. Durham, N.C., U.S.A.: Duke University
IATRC (International Agricultural Trade Research Consortium). 1988. Assessing the benefits of trade
liberalization. Summary of the Symposium of the International Agricultural Research Consortium on
"Bringing Agriculture into the GATT," Annapolis, Md., U.S.A., August.
Krueger, Anne O., Maurice Schiff, and Alberto Vald6s. 1988. Agricultural incentives in developing
countries: Measuring the effect of sectoral and economywide policies. World Bank Economic
Review 2 (September): 255-271.
Magiera, Stephen L., and Michael T. Herlihy. 1988. Comparing world price changes from trade
liberalization models. Background paper for the Symposium of the International Agricultural Re-
search Consortium on "Bringing Agriculture into the GATT," Annapolis, Md., August.
Matthews, A. 1985. The common agricultural policy and the less developed countries. Dublin: Gill and
Macmillan, Ltd.
OECD (Organisation for Economic Co-operation and Development). 1987. National policies and
agricultural trade. Paris: OECD.
1989. OECD agricultural policies, markets and trade: Monitoring and outlook 1989. Paris: OECD.
Quizon, J., B. Gardner, and L. Quinn. 1988. The consequences of agricultural trade liberalization for
developing economies assisted by AID. A report to USAID by Wharton Econometrics and Louis
Berger International.
Tyers, Rod, and Kym Anderson. 1986. Distortions in world food markets: A quantitative assessment.
Background paper for the World Development Report. World Bank, Washington, D.C.
Vald6s, Alberto, and Joachim Zietz. 1980. Agricultural protection in OECD countries: Its costs to less
developed countries. IFPRI Research Report 21. Washington, D.C.: International Food Policy
Research Institute.
World Bank. 1986. World Development Report 1986. New York: Oxford University Press.
Zietz, J., and A. Vald6s. 1985. The costs of protectionism to less-developed countries: An analysis for
selected agricultural products. Prepared for The World Bank, Washington, D.C., January. Mimeo.


Food Security and Compensation:
The Role of The GATT

Odin Knudsen


The issues of food security and compensa-
tion have entered the negotiations on agricul-
tural trade in the Uruguay Round: food security
through several proposals that, if accepted,
would permit countries to use extraordinary mea-
sures (nontariff barriers and subsidies) to ensure
adequate domestic production of certain basic or
strategic food commodities; and compensation
through discussions on financially buffering from
the impact of a possible rise in food prices those
developing countries accustomed to importing
low-cost and subsidized foods from industrial-
ized countries.
Despite the apparent incompatibility of these
issues, they are important politically, and per-
haps economically, to the outcome of the Uru-
guay Round. Governments are sensitive to even
the appearance that the food security of their
nations has somehow been compromised in the
multilateral negotiations. Furthermore, if costs
from higher food import bills are realized as a
result of the round, policymakers will be asked
what compensatory gains they achieved for their
country. Some form of direct compensation for
price rises would enhance the political accep-
tance of an agreement. In addition, the discus-
sion of compensation opens the broader issues
of the rights and obligations of developing coun-
tries, of special and differential treatment, and of
the role of the General Agreement on Tariffs and
Trade (GATT).
Addressing the issues of food security and
compensation in the context of the GATT is im-
portant if an agreement is to be reached in the
Uruguay Round. In addressing these issues

here, the next section examines whether food
prices could actually rise as a consequence of
agricultural trade liberalization by reviewing the
predictions of simulation models. The existing
buffers against risk and the possible role of food
aid as a compensation mechanism are looked at
next. The true long-run solution-structural ad-
justment and market access-is discussed in the
next to last section. The chapter concludes with
a brief summary.


In the world of the blind, the partially sighted
have power. In a sense, economists and policy-
makers have gained this sight from the simula-
tion models on agricultural trade liberalization.
From the models they have gotten a framework
for promoting different proposals in the Uruguay
Round. This situation is particularly true for the
food-importing countries that are hearing fore-
casts that food prices in world markets will rise if
agricultural support and protection are elimi-
nated. All the models except the MTM model of
the Organisation for Economic Co-operation and
Development (OECD) Agricultural Directorate
predict that wheat prices will increase; all but the
MTM and the Zietz-Vald6s model forecast that
the prices of coarse grains will rise.1 For the
other commodities, all models predict price in-
creases but of considerably different magni-
tudes. On the other hand, if the developing coun-
tries liberalize their agricultural trade, the price
rises predicted by some of the models are con-
siderably muted; the MTN and Zietz-Vald6s re-

1 These rises should be qualified, however, in that the increases are above a declining trend in international prices
attributable to technological advances. That is, productivity gains through technology could offset these price rises.

suits indicate a quite dramatic fall in the pre-
dicted price.
At the same time, these results should be
treated with caution. First, all the models have
technical problems. Many of the estimates of
supply and demand responses (price elastici-
ties) are best-guesses or estimates from other
countries. This caveat is particularly true for the
estimates of cross-elasticities, which are used to
trace the impact of a price change in one com-
modity market on the price in another market.
Second, the protection and support have been
crudely entered into the models as border equiv-
alent measures (in other words, the producer
subsidy equivalents are converted to equivalent
tariffs). For some interventions, this procedure is
wrong. For example, the deficiency payments in
the United States are not just supportive of pro-
duction, but have a mixed effect because land
has to be diverted to qualify for payments. Third,
some of the base data are old, from the 1970s,
especially those in the general equilibrium mod-
els such as the International Institute of Applied
Systems Analysis (IIASA) and Rural-Urban
North-South (RUNS) models. Even in the partial
equilibrium models, which use 1982/85 or 1986
data bases, the bias is toward an upward move-
ment in prices because of the very substantial
subsidization that occurred in those years. Fi-
nally, the most important aspect of the liberaliza-
tion-the food and feed interaction with livestock
production-has been only crudely modeled. It is
in this interaction that the MTM model gets lower
increases in food grain price because of the
assumption of higher cross-elasticities between
food and animal feed, particularly pasture land.
Nevertheless, it is remarkable that there is such
a broad consensus across models (with a few
exceptions) that prices will rise as a result of
Although the long term has been modeled,
little attention has been given to the transition. In
the short term, prices could be unstable. Expec-
tations can dramatically move short-term prices,
and speculative bubbles may occur. Further, if
international commodity stocks are low, as they
are now, prices can move dramatically with small
changes in production. In addition, public stock-
holding will likely decline as a consequence of
reduced intervention by governments; however,
little is known about how quickly or to what ex-
tent private stockholding will substitute for public
In addition, the supply response of the major
grain-producing nations has changed. When the
United States used its deficiency payment pro-
gram to divert large amounts of acreage, the
country's supply response could be quite rapid

and significant. Now, however, acreage is being
held in conservation reserve programs under
long-term contracts. Although the U.S. secretary
of agriculture could release this acreage, politi-
cally it would be very difficult to do so. With the
European Community, where measures to con-
trol supply are being implemented for grains and
dairy, Europe's supply response could also be
slow and constrained. Although a country such
as Argentina could respond significantly to
higher world prices, many such countries are
politically and financially strapped. Their govern-
ments are desperate for tax revenues, and ex-
port taxation remains as one of its few sources.
This taxation means, however, that the rises in
prices will not be fully passed on to farmers or,
for that matter, consumers.
For policymakers and negotiators, these
price uncertainties translate into political risk.
Even if a dramatic rise in food prices occurs that
is simply coincidental with a negotiated outcome,
politicians will be under pressure to explain their
past positions and possibly (and worst of all) to
reverse agreements. Short-term price instability
could threaten the sustainability of any reforms
out of the Uruguay Round, whether resulting
from or accompanying a negotiated outcome co-
incidentally. Although the model results give
long-run assurances that the price effects will
not be too dramatic, they provide only muted
comfort as to what might happen in the short run.
Addressing this price risk and presenting pos-
sible buffer mechanisms, particularly for the
large food importers, are legitimate issues for
policymakers and anyone else who wants sus-
tainable reform of agricultural policies.


For balance of payments relief in the case of
a price rise, the importing country either has to
suppress imports of other commodities in prefer-
ence to food or seek international financing. The
obvious source of financing is the International
Monetary Fund (IMF), in particular its Compen-
satory and Contingency Financing Facility. In the
case of an importer, the cereals part of this facil-
ity is relevant. Under the cereal provision of the
Compensatory Facility, a country can finance the
excess costs of cereal imports that are short-
term and largely attributable to circumstances
beyond its control. If the change is long-term,
that is, the commodity price rises because of a
structural change-for example, a GATT agree-
ment whereby prices are structurally altered as
subsidies are reduced-the Compensatory Fa-

cility cannot be used. Even where the excess
cost of imports is attributable to short-term
changes in prices, the IMF needs to be satisfied
that the member is fulfilling the conditions of
cooperation with the IMF on policy issues. Fur-
thermore, the facility provides a loan not a grant,
at standard IMF interest rates; the funds are
therefore not concessionary. An additional twist
is that the reference base for the facility is the
12-month period preceding the request com-
pared with a five-year average centered on the
base period.2 In other words, the lowly past is
averaged with the highly priced future to calcu-
late the shortfall. Under normal import require-
ments (in terms of quantity), a country must be
well into the higher price period to make the
compensation significant. Moreover, the amount
of the financing will need to be balanced against
any surge in export earnings.3
Thus, the compensatory facilities may not be
a buffer. An alternative is to use food aid as a
possible offsetting measure. The proposals be-
fore the GATT negotiators on food aid legitima-
tize its role as a bona fide instrument. For exam-
ple, the U.S. proposal excludes food aid from the
prohibited list of interventions. The Cairns Group
proposes that food aid be a legitimate form of
assistance, if it is provided as grants through a
multilateral agency to avoid its use as an export
This legitimization of food aid is both good
and bad news. The good news is that the GATT
will not ban food aid as illegal. The bad news is
that unless carefully handled, its legitimization
opens the door for its use as an export enhance-
ment-an inducement to purchase commercially
from a food donor country. Furthermore, the
availability of food aid is likely to be less.
To prevent food aid from being used as a
"gray-area" export subsidy, the members of
GATT will have to adopt some tests of legitimate
aid or restrict direct bilateral food aid in prefer-
ence to multilateral aid. A related issue is how

food aid is procured-could it be used as hidden
production support for a donor country?
Members will be keeping a careful watch on
whether food aid is being used as a domestic
support mechanism. Where agricultural trade re-
form results in lower public stockholding to sup-
port prices, food aid would suffer from both lower
availability and stricter restrictions on its use. If
the Uruguay Round establishes rules that define
clearly the circumstances when food aid does
not displace commercial trade, then it would be-
come more difficult for many countries to ob-
To prevent food aid being used as a hidden
export subsidy, it could be handled multilaterally,
with purchases being made on a competitive bid
basis out of the international markets. Donor
countries would make cash donations and then
would be only indirectly involved in the purchase
and distribution, mainly through the governing
boards of the international institutions. However,
unless pressure can be brought to induce large
donations of cash, food aid is less likely to come
under a multilateral framework.
Most donors would favor the provision of con-
cessionary "food aid" through the dumping of
surplus stocks than cash donations to a multilat-
eral agency. Nevertheless, it is conceivable that
the Uruguay Round could arrive at long-term
pledges of cash-equivalent food aid to a multilat-
eral institution as part of the bargain with devel-
oping countries for their support for an agree-
ment on the agricultural negotiations. It will,
however, require hard bargaining to establish
significant levels of aid, especially since many
governments are experiencing budgetary con-
straints. A compromise would be to permit dona-
tions of either cash or food.5
The food aid commitments under a multilat-
eral program could be distributed as grants to
food deficit, least developed countries as long as
they had bona-fide, targeted-to-the-poor food
distribution systems. Higher income countries
with large food deficits could buy the food at a

2 The base period is the 12-month period for which actual data are available, provided the lag between the end of the shortfall
year and the time of consideration of the request by the IMF Board does not exceed six months. Data may be estimated for up
to 12 months of the shortfall year for cereal imports.
3 There is an anomaly in the cereal provision of the Compensatory Facility. When the facility is used to counterbalance
shortfalls in export earnings alone, changes in imports are not used to determine eligibility. It is only with the cereals part of the
facility that a netting out of exports and imports comes into play. The member may request a purchase of an amount equal to
the net shortfall in its exports, calculated as the sum of its export shortfall and the excess in its cereal import costs. For three
years from the date of a member's request, any purchases to cover export shortfalls are calculated as net shortfalls in exports.
4 The bulk of food aid today (80 percent) is provided on a bilateral basis. Some of it is distributed for genuine humanitarian
motives, but part is a convenient and publicly justified way to get rid of surpluses or promote other national interests, some of
which are commercial.
5 There is already a precedent for this approach through the Food Aid Convention (established as part of the Kennedy

fixed world price set to the 1986-90 averages
under the allocation quotas. Even in the case of
concessionary purchases, assurances should
be established that the food aid will go largely to
the poor and that part of the money generated
from its sale will go to finance agricultural devel-
opment. Both the grant and concessionary parts
of the food aid program could be established for
a specified transition period, subject to renewal
after review.


Although food aid could be used as a buffer
in a transitional period, the real issue is the
long-term adjustment needs of developing coun-
tries. If the Uruguay Round is successful, signif-
icant structural adjustment in developing coun-
tries will be needed, not just in agriculture, but in
services, textiles, and other areas. It is in this
structural adjustment that the issues of food se-
curity, compensation, and the GATT really stand
True food security is only achieved through
growth and strong export performance. Enhanc-
ing growth and exports requires adjusting to new
relative prices and trade opportunities. Success
in the Uruguay Round will be measured by how
countries are able to take advantage of freer
trade in areas where trade has previously been
restrictive. However, many developing countries
will have difficulty adapting to the new relative
prices and taking advantage of freer access to
industrial markets. They have been reluctant to
take advantage of freer trade and instead have
pursued an inward orientation. As a result, most
developing countries are highly protected and
restrictive in their trade, particularly in agricul-
tural products. Not only has this inward orienta-
tion resulted in inefficient industries, it has also
established an anti-export bias through over-
valuing the real exchange rate. If the Uruguay
Round is successful, the trading opportunities
could be significant, and those countries that can
make the structural transition to an outward-ori-
ented economy can grow and achieve true food
This structural adjustment will require two
forms of assistance: financial, from both private
and public sources; and commercial, through
changing how GATT deals with special and dif-
ferential treatment and through greater market
access. The financial assistance can only come
in part from bilateral and multilateral aid and
loans because the required amounts are well
beyond the relatively limited resources of institu-

tional aid. To make the transition, capital will
have to come from private sources through both
commercial banks and return of flight capital.
However, for capital inflows to reoccur requires
the prospect of economic viability. For some
countries, this condition establishes what ap-
pears to be a vicious trap-viability is only pos-
sible with capital inflows. For still other countries,
structural adjustment is strictly in their hands:
they can create open economies that encourage
both domestic and foreign investment.
The second form of assistance is commercial
trade policy, the area directly related to the
GATT. Special and differential treatment and the
generalized system of preferences (GSP) have
been failures. Any benefits have been small and
specific to a few privileged countries, and the
costs have been high in terms of diverting trade
and discouraging meaningful trade concessions
and liberalization. Most special and differential
treatment for developing countries has weak-
ened the negotiating power of their governments
in the GATT. It has permitted developed coun-
tries to establish higher levels of protection for
commodities of immediate interest to developing
countries than for general industrial imports.
Equally serious, special and differential treat-
ment has encouraged developing countries to
take a back seat in the GATT. As a conse-
quence, they have conceded the use of gray-
area measures by industrial countries, in partic-
ular, the increasing utilization of voluntary export
restraints against developing and newly industri-
alized countries.
Developing countries have taken a prominent
position in the Punta del Este declaration. Unfor-
tunately, it appears the developing countries are
not getting far in the bargaining. This situation is
particularly unfortunate in that the expansion of
trade in the next century will depend on develop-
ing countries being fully integrated into the world
trading system. It would be disappointing if at the
end of the Uruguay Round about 80 percent of
GATT's members are on the sidelines, highly
protective in their trade policies and waiting for
special and differential treatment to be granted
to them. Freedom from the obligations of the
GATT has been detrimental to developing coun-
tries in terms of bargaining power and getting
true concessions. It has also limited their ability
to garner internal support for trade liberalization.
It is of interest to developing countries that Part
IV and Article 18B of the GATT be rescinded or
significantly modified in exchange for significant
reforms in agriculture, tropical products, textiles,
and the banning of gray-area measures.


Compensation to developing countries for
possible liberalization of trade in agricultural
commodities must be viewed more broadly than
from the perspective of the relatively narrow in-
terests of food-importing countries. True com-
pensation means that in the negotiations the
developing countries must win meaningful ac-
cess to developed countries-that is, access
with zero tariffs and the absence of quantitative
and other gray-area measures, in particular, vol-
untary export restraints. In turn, developing
countries should offer as reciprocity the opening
of their own economies-for example, tariffs
below a maximum of 20 percent and nontariff

coverage of production to less than 5 percent-
and their compliance in the rescission or modifi-
cation of the harmful articles to the GATT. Fur-
thermore, to reduce the perceived risks and to
cover the possible adverse effects of rises in
agricultural prices, food and cash aid (as part of
a renewed Food Aid Convention) should be do-
nated to a multilateral institution such as the
WFP for distribution or concessional sale to
countries in need that have adopted targeted
food distribution systems. In turn, the World
Bank and the regional development banks
should be prepared to assist in the structural
adjustments that will be necessary if developing
countries are to take full advantage of a new,
freer trading regime.

Burniaux, J. M., D. van der Mensbrugghe, and J. Waelbroeck. 1990. The food gap of the developing
world: A general equilibrium modeling approach. In Agricultural trade liberalization: Implications for
developing countries, Chapter 9. Edited by lan Goldin and Odin Knudsen. Paris: Organisation for
Economic Co-operation and Development and the World Bank.
Burniaux, J. M., J. P. Martin, F. Delorme, I. Lienert, and D. van der Mensbrugghe. 1990. Economy-wide
effects of agricultural policies in OECD countries: A GE approach using the Walras model. In
Agricultural trade liberalization: Implications for developing countries, Chapter 10. Edited by lan
Goldin and Odin Knudsen. Paris: Organisation for Economic Co-operation and Development and
The World Bank.
Frohberg, K., G. Fischer, and K. S. Parikh. 1990. Would developing countries benefit from agricultural
trade liberalization in OECD countries? In Agricultural trade liberalization: Implications for develop-
ing countries, Chapter 8. Edited by lan Goldin and Odin Knudsen. Paris: Organisation for Economic
Co-operation and Development and The World Bank.
Moreddu, C., K. Parris, and B. Huff. 1990. Agricultural policies in developing countries and agricultural
trade. In Agricultural trade liberalization: Implications for developing countries, Chapter 4. Edited by
lan Goldin and Odin Knudsen. Paris: Organisation for Economic Co-operation and Development
and The World Bank.
Parikh, K., G. Fischer, K. Frohberg, and 0. Gulbrandsen. Towards free trade in agriculture. Dordrecht,
Netherlands: Martinus Nijhoff Publishers, 1988.


The Impact of Trade Liberalization on
Domestic and International Price Instability

Rod Tyers

Salient characteristics of international food
markets in the past two decades have been a
continuing decline in average prices relative to
those of other tradable goods and a marked
increase in price fluctuations (Figure 7.1). In par-
ticular, since the early 1970s, real international
food prices have swept through the widest range
this century, from twice to half their 1980 values.
Although fluctuations in climate have always

made agricultural markets comparatively vola-
tile, this volatility has increased as more govern-
ments have come to use international food mar-
kets as sources of residual supply and as
dumping grounds for unplanned surpluses. Also
important has been the rise in macroeconomic
volatility since the early 1970s. Fluctuations in
exchange rates have affected prices directly,
while associated movements in interest rates

Figure 7.1-Real international food prices, 1900 to 1987



o 125-



1900 1910 1920 1930 1940 1950 1960 1970 1980 1990

Source: R. Tyers and K. Anderson, Disarray in World Food Markets: A Quantitative Assessment (Cambridge: Cambridge
University Press, Forthcoming), based mainly on price series from the World Bank's Economic Analysis and Projec-
tions Department.
Note: This index refers to export prices in U.S. dollars for cereals, meats, dairy products, and sugar, deflated by the U.S.
producer price index (primarily of industrial product prices), with weights based on the importance of each product in
global exports in 1977 to 1979.

have altered the cost of stockpiling, and surges
in inflation have affected the hedging demand for
commodity stocks.
Although the root causes of the volatility in
international food markets do not originate in
agricultural policies, it is clear that government
policies which shield domestic farmers, consum-
ers, and stockholders from fluctuations in inter-
national prices exacerbate the degree of the vol-
atility. Such policies insulate domestic markets
and reduce the incentives for adjusting the do-
mestic quantities produced, consumed, or
In the terminology of Blandford (1983), two
effects lead to greater international market vola-
tility. First is the "transmission" effect, whereby
short-run variability in supply and demand in a
particular country is transmitted to the world mar-
ket through variations in the country's trade vol-
umes. Second is the "absorption" effect, or the
degree to which a country does not share in the
short-run variability in the rest of the world, as
reflected in fluctuations in its border prices, be-
cause its trade volumes do not fully adjust. Both
these effects stem from policies that sever the
link between domestic and border prices, that is,
from policies that insulate domestic markets. Un-
less the original market disturbances are permit-
ted to affect the decisions of agents throughout
the world, the adjustments required of those who
remain exposed are larger, and hence the
changes in world prices required to elicit those
adjustments are correspondingly larger.
Of course, a government could ensure that its
domestic stockpile adjusted sufficiently to bear
the national share of the global burden. If, how-
ever, it used insulating trade and price policies to
do so, it would reduce the incentives for private
agents to hold stocks. When stocks are instead
held publicly, they tend to be managed so as to
stabilize domestic supply without regard to
changes in global scarcity (as reflected in inter-
national price fluctuations). This tendency stems
from governments' aversion to price risk on be-
half of domestic agents and with respect to pub-
lic expenditure and foreign exchange reserves.
Such risk aversion, combined with incomplete
markets for spreading income risk through insur-
ance, can make insulating price policies and
inward-looking stockpiling socially beneficial at
the level of the single country. But their imple-
mentation by many countries raises the risks
faced by the collectivity of all countries. As with
all common property resources, the "commons"
(in this case the world market for food products)
is overexploited for its risk-spreading benefits, a
situation that at least partially nullifies its useful-
ness for this purpose.

The rules of the General Agreement on Tar-
iffs and Trade (GATT) clearly offer a solution to
this problem. If a "critical mass" of participants
(Runge, et al. 1989) can be seen to be reducing
their market insulation, it could become welfare-
improving for other countries to reduce theirs
The next section of this paper briefly reviews
the nature of domestic market insulation, its "nat-
ural" origins in infrastructural costs, and its
sources in government policy. The third section
then examines the reasons why it is welfare-im-
proving at the national level to implement insu-
lating policies. The remainder of the paper offers
a premilinary analysis of the national and global
gains from a multilateral reduction in market in-
sulation. The fourth section quantifies the link
between market insulation and international
price volatility and offers a preliminary assess-
ment of the likely effects of multilateral "tariffica-
tion" on international and domestic market vola-
tility. The chapter ends with some brief


Market insulation is best measured using the
elasticity of price transmission (Bredahl, et al.
1979), or its complement, defined here as the
degree of insulation. The latter parameter holds
a value of one when no part of any change in a
border price is transmitted to the domestic mar-
ket and zero when trade is undistorted and all
international price changes are transmitted.
Table 7.1 presents estimated average values of
the degree of insulation for some industrialized
and developing countries. The pattern is clearly
one of substantial insulation by both groups of
countries in the short and long run, with the
food-exporting countries of the Americas and
Australasia alone bearing the brunt of the burden
of adjustment.
Even in the absence of intervention by gov-
ernments in domestic commodity markets, how-
ever, domestic prices may be only partially re-
sponsive to changes at the border. This situation
can arise when the marketing margins (and
hence the costs of domestic infrastructure) are
large in relation to the product price and when
substantial lags occur between the dates of pur-
chase and delivery of food imports. Neverthe-
less, a great deal of the observed market insula-
tion is the result of government intervention.
Even where differing trade policy instruments
give rise to the same average rate of protection,
their effects on the relative stability of domestic

Table 7.1-Degree of food market

Short Run Long Run

EC-10 83 62
Spain and Portugal 70 67
EFTA-5 91 60
Japan 76 53
United States 30 22
Canada 49 28
Australia 39 22
Argentina 39 30
Brazil 55 33
Mexico 78 63
Egypt 92 78
Nigeria 74 50
Korea, Republic of 89 72
Taiwan 63 31
China 81 52
Indonesia 87 50
Philippines 85 75
Thailand 68 41
Bangladesh 63 53
India 85 58
Pakistan 84 69

Source: Time series analysis of consumer and producer
prices for seven commodity groups (including
grains, meats, dairy products, and sugar) over the
period 1961-1983, as presented in R. Tyers and
K. Anderson, Disarray in World Food Markets: A
Quantitative Assessment (Cambridge: Cambridge
University Press, forthcoming).
Note: The degree of insulation refers here to the com-
plement of the average elasticity of price transmis-
sion (one minus that elasticity). The average price
transmission elasticity is weighted by the values at
border prices of the production and consumption
of each of seven commodity groups, including
grains, livestock products, and sugar.

prices may vary considerably depending on the
policy instruments used. Simple specific tariffs
and specific export taxes or subsidies, for exam-
ple, distort the proportional changes transmitted
to domestic markets in response to changes in
prices at the border. Their ad valorem counter-
parts, on the other hand, preserve proportional
changes but distort the absolute magnitudes of
the changes in border prices. Binding quotas
insulate the domestic market totally.

In general, the policy instruments applied to
agricultural markets are a great deal more com-
plex than these. Agricultural trade policies al-
most always employ instruments designed to be
adjusted over time. Notorious among such in-
struments is the system of variable levies and
export restitutions adopted as part of the Com-
mon Agricultural Policy (CAP) of the European
Community (EC). In this case the domestic mar-
ket price is determined by agreement among
agricultural ministers, and a levy is then imposed
on imports or a restitution is paid to exporters,
which is routinely adjusted so that consumers,
producers, and stockholders always face the
agreed domestic price. The extent to which this
policy insulates domestic markets is therefore
dependent first, on how responsive the political
process that determines domestic prices is to
changes in world agricultural prices and, second,
on associated changes in European exchange
Many countries employ "state trading,"
whereby government and semigovernment
agencies have monopoly control over interna-
tional trade in important commodities (Zietz and
Valdes 1988). These agencies typically use
trade as a means of managing domestic supply
and therefore maintaining domestic prices at de-
sired levels. In industrial countries these entities
include Japan's food agency and the wheat
boards of Australia and Canada. In developing
countries examples are the Food Corporation of
India, the National Logistics Agency of Indonesia
(BULOG), and the National Food Authority of the
Philippines. In all cases these agencies are em-
powered to maintain relatively stable domestic
prices, and the evidence in Table 7.1 attests to
their success.


In all countries some agents can be expected
to have stronger preferences for price stability
than others. Since market insulation occurs in
both developing and industrialized countries, it
can be hypothesized that in each case this pref-
erence would be strongest among the groups
with the most apparent influence over agricul-
tural policy: broadly, they are consumers and the
owners of industrial capital in developing coun-
tries and farmers in industrialized countries (An-
derson and Hayami, et al. 1986).
What, then, are the directions of the welfare
impacts of price stabilization on these groups?
An extensive literature on the theory of commod-
ity price stabilization yields few broad general-

izations about its benefits and their incidence
(Newbery and Stiglitz 1981). The preferences of
farmers, workers, industrialists, and government
treasuries for food price stabilization are there-
fore best derived from quantitative assessments
that are specific to individual countries. To illus-
trate this approach and to show the contrasts
between the welfare incidence of market-insulat-
ing policies in industrial and developing coun-
tries, the results from an elementary model of a
single open commodity market are reported
briefly. This model is detailed mathematically by
Tyers and Anderson (forthcoming) and by
Gibbard and Tyers (1990).
In both cases the results from the model show
that there are net gains nationally from the insu-
lation, a finding that supports the public interest
explanation for insulating policies. However,
these net gains are small in relation to national
income (less than 1 percent). Nevertheless, the
results bear out the hypothesis that in each case
the group that has the most to gain from market
insulation is the most politically influential one.
The gains to the owners of industrial capital and
to government revenue are dominant in the de-
veloping countries, where industry tends to be
protected at the expense of agriculture and
where the cost of collecting revenue by other
means is especially high. In the industrialized
countries, on the other hand, where agriculture
tends to be protected at the expense of other
sectors, farmers have the dominant interest in
price stabilization. Importantly, in neither the in-
dustrial nor the developing countries does any
group of agents in the domestic economy appear
to lose significantly. This fact explains why, de-
spite the smallness of the benefits, governments
tend not to find market-insulating policies politi-
cally costly to implement.
There is another important reason for the
prevalence of market insulation that stems from
two general characteristics of market-insulating
policies. First, such policies always separate do-
mestic from border prices and hence distort do-
mestic incentives, at least in the short run. Sec-
ond, because the current and future trend in
international market prices is uncertain, there is
no obvious and undisputed level at which do-
mestic prices should be set to achieve the objec-
tive of comparatively stable domestic prices. The
process by which the domestic price is set is
therefore subject to lobbying by vested interests.
Because governments can claim that the distor-
tion is temporary, pending the return of border
prices to "trend" levels, the cost of the substan-
tial distortion of domestic prices away from bor-
der prices is reduced. Since the lobbying and
propagandizing efforts of farmers in industrial

countries is stronger than that of the groups that
lose from high food prices, the farm interests
tend to dominate the process. Similarly, in devel-
oping countries the initially well-intentioned sep-
aration of domestic from international markets by
governments averse to food price and wage risk
reduces the political costs of policies that ensure
that the trend in domestic prices is below that at
the border.
Nevertheless, even where insulating policies
are not accompanied by protection, it would be
improper to leave the impression that such poli-
cies are desirable from the standpoint of world
welfare. They involve the shedding of risk
through international food markets. However,
when only some countries shed risk in this way,
the other countries bear the cost. Thus, for any
one country, the cost of market insulation poli-
cies is the burden of foreigners, the interest
group with the least voice in domestic politics.


To examine the global effects of market insu-
lation by individual economies, the Tyers-Ander-
son model of world trade in grains, meats, live-
stock products, and sugar is used (Anderson and
Tyers 1990b; Tyers and Anderson 1988 and
forthcoming). This model is used to simulate
changes in world market behavior over almost
two decades after insulating policies are re-
moved. The preliminary results are retrospective
in that both the protection and the insulation
policies are removed at the base period of the
model, which is 1980-82. For any particular sim-
ulation, after a few years roughly constant levels
of uncertainty in the forecast prices are ob-
served, and it is these that are compared in
Table 7.2. The index used is the coefficient of
variation of the simulated international price,
which is the percent deviation from the mean
within which that price can be found about two-
thirds of the time. Thus, as the coefficient of
variation becomes smaller, price volatility is re-
The results in Table 7.2 clearly show that
world food markets would be very much less
volatile were agricultural trade policies to be lib-
eralized, or changed so as to be noninsulating.
The latter could be achieved, for example, by
transforming existing policies so that the only
instruments used were ad valorem tariffs and
export taxes or subsidies. If such a change were
implemented in only the industrial market econ-
omies, the volatility in the food market would fall

Table 7.2-Effects of liberalizing food markets on international price instability
(coefficient of variation around trend levels)a

Coarse Ruminant ruminant Dairy Weighted
Wheat Grain Rice Meat Meat Products Sugar Averageb


Reference 58 53 38 24 8 26 36 34
Coefficient in the
absence of domestic
insulation by:
EC-12 39 45 32 15 8 13 28 26
Japan 54 51 33 9 7 18 33 28
United States 60 64 36 17 10 27 31 35
All industrial
market economies 33 47 28 7 8 11 25 23
Developing Asia 29 43 16 14 7 13 12 20
All developing
countries 17 23 10 6 6 8 8 12
All industrial and
developing countries 15 23 9 4 5 6 7 10

Source: Simulations of the Tyers-Anderson model of world trade in staple food commodities.
a Liberalization is some developing countries does not mean the removal of all market insulation, but only of that part
estimated to be the result of policies that affect domestic prices directly. Crude estimates suggest that in China this
part is 50 percent of the insulation, in Indonesia 54 percent, the Philippines 67 percent, Thailand 68 percent,
Bangladesh 51 percent, India 50 percent, Pakistan 55 percent, Brazil 71 percent, Egypt 66 percent, Nigeria 61
percent, and Sub-Saharan Africa 51 percent. In all other countries all market insulation is removed.
SBased on weights derived from the shares of the exports of each commodity group in total exports of grain, livestock
products, and sugar.

by a quarter. If it were carried out in the very
populous and highly insulated economies of
South and East Asia, the reduction would be
even more dramatic. A commitment by all indus-
trialized and developing countries to use non-
insulating policies would reduce volatility in the
global food market by more than two-thirds, to a
level smaller than the rates of domestic inflation
in most countries.
The domestic price implications of such a
liberalization are quantified in Table 7.3. To sim-
plify the exposition, the table includes only the
effects of liberalizing the international and do-
mestic prices of rice. The first column lists the
simulated coefficients of variation of producer
and consumer rice prices that stem from current
policies in selected industrial and developing

countries. All are smaller than those for the inter-
national price, with the tendency (consistent with
the arguments in the third section) for industrial
countries to insulate producer prices more than
consumer prices and for most developing coun-
tries to do the opposite. The other columns list
the corresponding coefficients of variation after
liberalization. The most striking result shown in
the table is that liberalization by all industrial and
developing countries would reduce the volatility
of international prices to a level almost as low as,
and in many cases lower than, that currently
enjoyed in insulated domestic markets. This re-
sult demonstrates strongly the collective gain
from a switch to noninsulating policies by as
many countries as possible. Furthermore, the
gain (reduced volatility in the international food

Table 7.3-Effects of liberalization on domestic price stability: coefficients of
variation of producer and consumer rice prices

After Liberalization by
Reference All Developing All All IMEs
Case IMEs Asia LDCs and All LDCs


International Price

Domestic Prices in:

United States

Korea, Republic of



South Africa
Other Sub-Saharan

25 (13)
33 (33)
6 (29)
6 (5)
4 (2)

3 (3)

15 (3)
10 (3)
3 (2)
21 (4)
26 (5)
7 (7)
14 (4)

20 (20)

6 (6)
29 (29)


28 (28)
28 (28)
28 (28)
28 (28)
28 (28)

3 (3)

12 (2)
8 (2)
3 (2)
21 (4)
6 (6)
11 (3)

6 (8)
11 (11)

5 (5)
8 (8)
23 (23)

8 (8)

12 (7)
3 (2)
2 (1)

16 (16)

12 (10)
11 (11)
14 (9)
10 (10)
12 (9)

9 (9)
4 (5)
7 (7)

3 (3)
5 (5)
13 (13)

5 (5)

7 (4)
9 (9)
2 (8)
2 (2)
1 (1)

10 (10)
10 (10)

7 (6)
7 (6)
7 (7)
9 (8)
8 (6)
6 (6)
7 (6)

10 (10)
8 (8)

7 (7)
7 (7)
10 (10)

7 (7)

9 (9)
9 (9)
9 (9)
9 (9)
9 (9)

9 (9)
9 (9)

7 (6)
7 (6)
7 (6)
8 (8)
8 (6)
6 (6)
7 (6)

9 (9)
7 (7)
9 (9)

7 (7)
7 (7)
9 (9)

6 (6)

Source: Simulations of the Tyers-Anderson model of world trade in staple food commodities.
Notes: IME = industrial market economies; LDCs = developing countries; consumer prices appear in parentheses.
aLiberalization in some developing countries does not mean the removal of all market insulation, but only of that part
estimated to be the result of policies that affect domestic prices directly. Crude estimates suggest that in China this
part is 50 percent of the insulation, in Indonesia 54 percent, the Philippines 67 percent, Thailand 68 percent,
Bangladesh 51 percent, India 50 percent, Pakistan 55 percent, Brazil 71 percent, Egypt 66 percent, Nigeria 61
percent, and Sub-Saharan Africa 51 percent. In all other countries all market insulation is removed.

markets) is more than twice as great if develop-
ing countries also adopt noninsulating policies
than if they do not.1


The analysis of the third section showed that,
for a given level of world price volatility, policies
which do no more than insulate domestic mar-
kets yield positive net economic benefits at the
country level. This is true even though the pri-
mary beneficiaries in industrial countries are not
the same groups that benefit when these policies
are applied in developing countries. More stable
domestic prices are therefore generally (rather
than specifically) beneficial. However, these net
benefits are small in relation to national income,
a result that suggests that market-insulating pol-
icies also serve other purposes. One such pur-
pose is to make pure protection easier to imple-
Notwithstanding the small net economic ben-
efits derived, market-insulating policies are un-
desirable for two reasons. The first is that they
are a step toward the implementation of protec-
tion policies (which are very costly to national
economies but which benefit powerful interest

groups). The second, which stems from the re-
sults presented in the fourth section, is that they
represent free riding on the risk-spreading ca-
pacity of global food markets. If all countries
refrained from using insulating policy instru-
ments, the volatility of the international food mar-
kets would fall by so much as to make most
domestic agents in most countries better off (by
virtue of reduced price risk) than they are at
The difficulty in realizing the benefits of better
global risk-spreading is, however, that any indi-
vidual country can only lose from a unilateral
switch to noninsulating policies, such as ad val-
orem tariffs and export subsidies. This situation
is a clear case of the "assurance game" (Runge,
et al. 1989), where a critical mass of countries
must first be bound to make such policy
changes. Thereupon the greater benefits are
available to all. This condition suggests a partic-
ularly important role for the tariffication compo-
nents of the proposals now under consideration
in the Uruguay Round (Anderson and Tyers
1990a). If nontariff barriers, variable levies, and
state trading are subject to greater disciplines,
the degree of market insulation in all GATT mem-
ber countries should fall, and so, therefore,
would the volatility of international food prices.

1 In a more extensive treatment of the same subject (Tyers 1990) it is shown that these gains would accompany reductions
in government expenditure risk in developing countries. In most developing countries they would also accompany reductions in
balance of payments risks.

Anderson, K., Y. Hayami, and others. 1986. The political economy of agricultural protection: East Asia
in international perspective. Sydney: Allen and Unwin.
Anderson, Kym, and Rod Tyers. 1990a. Effects of tarifficationn" of food trade barriers following the
Uruguay Round. Seminar Paper no. 90-02. Centre for International Economic Studies, University of
Adelaide, Adelaide, Australia, May.
1990b. How developing countries could gain from food trade liberalization in the Uruguay Round.
In Agricultural trade liberalization: Implications for developing countries, ed. lan Goldin and Odin
Knudsen, Chapter 2. Paris: Organisation for Economic Co-operation and Development and The
World Bank.
Blandford, D. 1983. Instability in world grain markets. Journal of Agricultural Economics 34: 397-392.
Bredhal, M. E., W. H. Meyers, and K. J. Collins. 1979. The elasticity of foreign demand for U.S.
agricultural products: The importance of the price transmission elasticity. American Journal of
Agricultural Economics 61 (1): 58-63.
Gibbard, P., and R. Tyers. 1990. The distributional incidence of commodity price stabilization in an
open economy. Seminar Paper no. 90-04. Centre for International Economic Studies, University of
Adelaide, Adelaide, Australia, June.
Newbery, K. M. G., and J. E. Stiglitz. 1981. The theory of commodity price stabilization: A study in the
economics of risk. Oxford, U.K.: Oxford University Press.
Runge, C. F., H. von Witzke, and S. J. Thompson. 1989. International agricultural policy: A political
coordination game. In Policy coordination in world agriculture, ed. H. von Witzke, C.F. Runge, and
B. Job, Chapter 4. Kiel, F.R.G.: Wiss.-Verl. Vauk.
Tyers, Rod. 1990. Agricultural trade reform and price risk in domestic and international food markets.
Seminar Paper no. 90-03. Centre for International Economic Studies, University of Adelaide,
Adelaide, Australia, June.
Tyers, Rod, and Kym Anderson. 1988. Liberalizing OECD agricultural policies in the Uruguay Round:
Effects on trade and welfare. Journal of Agricultural Economics 39 (2): 192-216.
Forthcoming. Disarray in world food markets: A quantitative assessment. Cambridge, U.K.:
Cambridge University Press.
Zietz, Joachim, and Alberto Valdes. 1988. Agriculture in the GATT: An analysis of alternative ap-
proaches to reform. IFPRI Research Report 70. Washington, D.C.: International Food Policy
Research Institute.


Agricultural Trade Reform, Price Stability, and
the Impact on Developing Countries

Ammar Siamwalla

The current agricultural trade negotiations in
GATT are expected to lead to changes in the
agricultural policies of the contracting parties.
Such changes will have an impact not only on the
price of agricultural commodities, but also on
their variability. Two questions of interest to de-
veloping countries are what impact the policy
changes will have on price variability and
whether any countervailing measures to insulate
the domestic economy from its consequences
could be introduced, either by the individual
countries themselves or through some interna-
tionally sponsored measures.


To answer the first of the two questions, it is
necessary to ask what policies the various coun-
tries are pursuing with the focus only on those
that are the subject of negotiations in the Uru-
guay Round and on the three major trading
blocs-the United States, the European Commu-
nity, and Japan. As the impact on developing
countries is examined, attention is placed pri-
marily on the cereal and sugar policies of these
countries and their side-effects on stability,
rather than on the broad policies for these com-
modities as such.

The United States

Cereals. The United States has always been the
central player in the world markets for most com-
modities, particularly cereals. Until the Food Se-
curity Act of 1985 it had also played the role of
granary to the rest of the world, holding large
amounts of stocks that, on most occasions, have
been a stabilizing influence on the world mar-
kets. To be sure, this role was acquired not as a
result of a conscious decision, but rather as a

result of the peculiar system of support that the
United States had built up for its farmers.
Briefly speaking, the traditional system of
support was for the government to set a price
floor (called the loan rate) at which the govern-
ment would procure all the output that farmers
produced. In addition the farmers received a
deficiency payment so that the proceeds from
their sale reached a target price that was set
higher than the loan rate. Both loan rate and
target price were set higher than the market-
clearing prices, so that there was a persistent
tendency for the government to accumulate
stocks. To prevent excessive stocks from being
held, the government also attempted to control
production through an acreage restriction pro-
gram. Sometimes farmers were paid to take land
out of production. Generally this acreage restric-
tion was not enough to prevent farmers from
producing an excess supply, which they then
turned over to the government to hold as stocks.
From the point of view of international trade
such a system of support placed the United
States in the role of residual supplier, and hence
of stabilizer of the world market. When there was
excess supply in the world market, the loan rate
that acted as a floor price in the United States
would make U.S. grain uncompetitive against
overseas suppliers, and the U.S. share of the
world market would dwindle. The government
would contribute to the decline by imposing more
stringent acreage restrictions to prevent further
accumulation of stocks. When there was excess
demand, the accumulated stocks of the U.S.
government would be flushed into the world mar-
kets and moderate the price increase. However,
because many key decisions are political and
made in response to fiscal pressures rather than
market movements, U.S. policies did not always
act as a stabilizing force.
In 1985, after seeing its policies result in sub-
stantial declines in the U.S. market share, the

U.S. Congress legislated measures that entailed
flushing out the U.S. grain stocks into the world
markets when they were already very soft. That
action of Congress, known as the Food Security
Act of 1985, also displayed a different style of
managing U.S. farm programs as compared with
previous farm legislation. In that act Congress
indicated that it no longer wished to depend on
storage as a means of propping up prices. For
cereals other than rice, the loan rates were to be
adjusted to keep pace with the world price
(farmers' welfare was to be protected through
the target price, which was reduced only
slightly). A new tool was introduced in the form
of marketing certificates, which allowed disposal
of government cereals at close to market prices.
Finally, an export enhancement program was
also introduced that provides export subsidies to
maintain U.S. competitiveness in the world mar-
kets. In combination, these measures imply that
Congress wished the government to hold sharply
reduced stocks. Balancing of current demand
and supply was to be the rule. Congress and the
then administration deemed this set of policies to
be more "market-oriented."
This shift toward a balancing of current de-
mand and supply was significant and is unlikely
to be reversed in view of the severe fiscal con-
straints the United States now faces. Indepen-
dently of the reforms to be agreed upon in the
Uruguay Round, it appears that the United
States will cease playing the key stabilizing role
it used to, particularly in the 1960s and early
Rice and Cotton. Interestingly, despite the rhet-
oric of the Food Security Act of 1985, which was
mostly aimed at the European Community (EC),
the strongest thrust of that act was on rice and
cotton, where the direct competitors of the
United States were developing countries. For
these two commodities a marketing loan was
introduced. This measure mandates that the
government release its stocks at world prices.
The immediate result of flushing out the
government's stocks of these commodities into
the world market in 1986 and 1987 was a severe
drop in world prices, which were already low.
Since then the United States has pursued a pol-
icy of year-by-year balancing of demand and
supply for these two commodities, with no at-
tempt to stabilize prices.
Sugar. Before 1985, the United States protected
its domestic sugar industry by restricting im-
ports, which were mostly from developing coun-
tries. Congress set a quota for each exporting
country, which then issued export licenses for
individual shipments. This modus operandi en-
sured that the exporting countries retained the

quota rent from the restriction. Moreover, in a
way the amount each country exported was
guaranteed, so that its foreign exchange earn-
ings from its sugar exports to the United States
were also stabilized.
As sugar consumption in the United States
declined, helped along by high domestic prices
that encouraged substitutes and as the govern-
ment again acquired large stocks from its do-
mestic operations, the simultaneous income
guarantee to foreign and domestic producers
became untenable. Congress therefore shifted
its sugar policies to be on a "self-financing"
basis. Essentially, this change entailed a drastic
cut in the import quotas-and if the fall in domes-
tic consumption continued (which it did), import
quotas were to be slashed further. Therefore the
income from sugar of those countries relying on
exports to the United States were no longer
guaranteed. Moreover, in 1987, because of an
excessive import quota in the previous year, the
United States found itself with excess stock that
it exported to the world market, a step that again
drove free market prices down further.

The European Community

Cereals. From its inception the Common Agri-
cultural Policy (CAP) has relied almost exclu-
sively on border measures both to support do-
mestic prices above world prices and to insulate
the former from fluctuations in the latter. The

basic tools used were variable import levies and
variable export restitutions (that is, subsidies).
True, the Community does intervene by first buy-
ing cereals (which are mostly exportables) and
putting them in storage, but only for a short
period, after which it puts them up for sale on a
tender basis to exporters.
Compared with the United States, which has
often changed its policy regime, sometimes
drastically, the EC has consistently pursued the
same policy regime for cereals for more than two
decades. The Community did not add policy
shocks that destabilized world prices, as the
U.S. Congress did in 1985. On the other hand,
the policy regime itself has been extensively
blamed for exporting the entire variation in out-
put and consumption within the Community to
the rest of the world and for not sharing at all in
the adjustments necessary in the rest of the
world when there has been a worldwide surplus
or shortage, as in 1972 to 1974. Indeed, it is
difficult to design a policy regime that contributes
more to world market instability than the CAP for
Sugar. In theory, the policy regime for sugar has
been somewhat more responsive to world prices

than that for cereals as a result of a system of
multiple pricing through production quotas.
Under quota A (the total size of which is about 60
percent of aggregate output), the producer is
guaranteed a price considerably higher than the
world price, which is subject to a relatively light
(2 percent) "co-responsibility levy" used to fi-
nance the export restitutions. Under quota B, the
producer receives a somewhat reduced price
(because he is subject to a higher rate for the
co-responsibility levy). The producer is in theory
allowed to produce any amount in excess of
these two quotas under quota C, but he can sell
that output only at the world market price. Under
this system, the Community allocates the quotas
to member countries, some of which sub-assign
them to processors, who then pass them on to
growers using the same principle. If this proce-
dure is the case, then at least the marginal ton of
sugar produced in those countries fetches the
world price, and production presumably re-
sponds to it. This responsiveness would contrib-
ute somewhat to stability in the world market.
Some member countries (Belgium, the Nether-
lands, and the United Kingdom), however,
choose to give their producers an average of the
prices for quotas A, B, and C. This approach
reduces that responsiveness (Australia, Com-
monwealth of 1985, 201).
For some developing countries an important
component of the Common Sugar Policy is the
agreement signed with the African, Caribbean,
and Pacific (ACP) countries, which allows them
to export sugar to the Community at higher than
world market prices-even though the Commu-
nity is now a net exporter of sugar. The amount
they can export to the EC is, however, subject to
a ceiling. Essentially, such arrangements make
the Community import quotas for sugar an eco-
nomic grant. It should be added that because
this grant is sanctified by a treaty (the Lome
Convention), it has a higher degree of stability
for the recipient than does conventional eco-
nomic aid or food aid, or indeed the U.S. sugar
quota system.


Rice. The system of support by the Japanese
government provides for farmers is to close its
domestic markets altogether from the world mar-
ket. As no trade is allowed, the market should in
theory be cleared by domestic demand and sup-
ply. However, because the government also
guarantees the price the farmers receive for their
rice and sets a somewhat lower price for urban
consumers, discrepancies can arise between
demand and supply. Such discrepancies are ad-

dressed by movements in the public stock and
acreage controls. In the past, attempts to control
acreage tended to lag behind the growth in pro-
duction, and the government found itself con-
stantly having to accumulate stocks. As the
stocks reached a high level, it was forced to
dispose of them overseas, a move that disrupted
the world market. Eventually in 1980 the United
States negotiated an agreement with Japan lim-
iting the latter's exports to 1.4 million metric tons
over the following four years. Because the Jap-
anese are now very much on the defensive as far
as their rice policy is concerned, they have not
reentered the export market since the expiration
of that agreement. That is, as a consequence of
these post-1980 developments, Japan no longer
adds to or subtracts from the stability of the world
rice market (Australia, Commonwealth of 1988).
Sugar. Even though Japan is the world's second
largest importer of sugar, it has a heavily pro-
tected domestic sugar industry, which in 1987
received US$2 per kilogram in subsidies, twelve
times the world price for that year. The system of
levies and surcharges that brings the cost of
imported sugar part way to the production cost of
Japanese sugar-the other part is covered by a
direct payment to sugar processors-is ex-
tremely complicated and does full justice to a
famously numerate nation (Australia, Common-
wealth of 1988, 207-210). Broadly speaking, it is
a partially variable levy that insulates domestic
producers and consumers somewhat from the
volatility of the world market, which in turn
means, as in the case of the EC cereal policy,
that the Japanese are able to export some of the
variability in their production and consumption
and to take up a small share of the variability in
world demand and supply.

An Assessment

The policies of the Organisation for Economic
Co-operation and Development (OECD) have an
impact on world price stability in two ways: one
is through the policy regime itself; and the other
is through changes in the policy regime. Tradi-
tionally the EC and to some extent Japan have
tended to have a stable policy regime that, how-
ever, maximizes the destabilizing impact on the
rest of the world. The United States, on the other
hand, tended to have policy regimes that contrib-
uted somewhat to price stabilization, at least
until 1985. However, U.S. policies have often
shifted direction abruptly, at great cost to com-
peting exporters and sometimes to importers.
The sugar regimes in the United States and
the EC have guaranteed the suppliers of these
markets a stable income, but at the expense of

destabilizing the market for nonparticipants. Be-
cause quotas are decided by a political process,
the benefits that quota holders receive are not
risk-free, as U.S. suppliers found out in 1986 to
There is another way in which current policies
of the major trading blocs contribute to stability.
The OECD countries have created a spare ca-
pacity that could be called upon when there is a
widespread crop failure as in 1972. To be sure,
this spare capacity needs to be deducted from
the output that would have been forthcoming
from non-OECD suppliers that could not afford to
outsubsidize the OECD countries. However,
even after netting the spare capacity out, it is
probably true that there is spare capacity in agri-
culture in the world right now, which minimizes
the risk of an upward surge in prices. If the fear
is of upside price risks only, as in the case of
staple food crops, then clearly the OECD subsi-
dies have to some degree contributed to the food
security of the poorer countries. Such asymmet-
rical treatment of upside and downside risks is,
however, justifiable only if the interests of food-
exporting developing countries are ignored. The
fact is that the excess capacity in OECD agricul-
ture is a bane on the markets for many other
export commodities of developing countries as
well, for example, sugar, cotton, and vegetable


The agenda for the current GATT round is
primarily an outcome of the dispute among the
three major trading blocs, with the loosely allied
Cairns Group of agricultural exporters also play-
ing a role. With the United States as a principal
source of demand, the objective of the negotia-
tions is to lower the level of support given to
agriculture by the OECD countries. The negotiat-
ing issues have resolved themselves into three
major points:1
* An orderly reduction of subsidies for agricul-
tural exports and reform of GATT's Article
XVI, which specifically permits such subsi-
* An opening up of markets ("market access"),
which are presently closed to imports, and
reform of GATT's Article XI, which allows quo-
tas to be imposed on imports of agricultural

goods. It would also mean tariffication of all
border measures. The variable import levy
system employed by the EC would, among
other measures, no longer be acceptable.
A revision of the GATT rules concerning the
type of internal subsidies that are to be al-
lowed on agricultural goods.
Each of the three major trading blocs relies on
different mechanisms, which would have to be
substantially modified if the round culminates in
complete success on each of the above three
points. The United States relies substantially on
internal subsidies to prop up farmers' incomes,
the European Community relies more on export
subsidies, and Japan denies foreign suppliers
market access to its consumers. It is therefore
not surprising that the disagreements among the
major trading blocs are so profound.
Alongside the negotiations on agriculture
there is a parallel set of negotiations on tropical
products, whose definition excludes items such
as sugar, rice, and tropical vegetable oil, all of
which would compete with the output of OECD
farmers. In previous rounds these items were
separated from the rest of agriculture for two
reasons: developing countries felt that items of
interest to them should not be embroiled in in-
tractable disputes among the OECD contracting
parties; and the issues involved were relatively
more straightforward, involving only negotiations
on reductions in tariffs and tariff escalation. Dur-
ing the current round, although the negotiations
are on separate tracks, the United States main-
tains that any agreement on tropical products is
contingent on an agreement on agriculture.
Developing countries themselves favor other
tactics beyond the GATT negotiations for achiev-
ing their aims on these items, for example, com-
modity agreements, which are arrangements to
cartelize or stabilize markets in these commodi-
ties. Most attempts to do so during the 1970s
and the 1980s have failed; the only surviving one
is for natural rubber.
Three groups can be distinguished among
the developing countries: those belonging to the
Cairns Group (a coalition of agricultural export-
ers that cuts across the North-South divide); a
few importing countries concerned with the issue
of food security that would like to retain the
freedom to regulate imports into their countries;
and the majority, which have tended to stay very
much on the sideline, on agricultural issues. All,
even those among the Cairns Group, have advo-

1 The whole question of phytosanitary regulations, also a subject of the negotiations, is ignored here because it is not germane
to the issues being discussed in this chapter.

cated special and differential treatment (S&D, in
GATT jargon) for developing countries. To date,
however, there is very little consensus on what
sort of S&D should be arranged, other than that
developing countries be given more time to
change their policies to bring them into line with
the new rules.


Developing countries have an interest in see-
ing less price variability for agricultural commod-
ities for two reasons: they wish to protect their
own producers of agricultural commodities from
income instability; and they wish to prevent up-
surges in the prices of staple food items that
would threaten the nutritional well-being of the
poorer sections of their populations. Both these
objectives arise because the producers and con-
sumers involved are poor and do not have ac-
cess to the credit market to tide them over during
transitory shortfalls in income when prices drop.
The GATT trade negotiations are relevant to
these essentially domestic concerns at two lev-
*At the national level each government can and
does insulate its producers and consumers
from fluctuations in the domestic markets up
to a point (see the second point below), for
example, by mimicking what the EC does for
its producers and consumers. It could be ar-
gued that producers and consumers them-
selves can and indeed do even out their con-
sumption over time by building assets in good
times and by running them down and even
borrowing in bad times (Knudsen and Parnes
1975; Deaton 1989). The justification for the
government to step in to help individuals
achieve their aim is that it has better access to
the credit market than do individual producers
and consumers. In the context of the GATT
round the issue facing the government of each
contracting party is what limitations there
-would be on its freedom of action to undertake
such actions. For example, if export subsidies
are outlawed, would developing country gov-
ernments be able to shore up the domestic
prices of their exportable goods by providing a

subsidy at times when international prices are
*The ability of the governments to insulate do-
mestic producers and consumers from vari-
ability in international prices is also con-
strained by their fiscal resources, particularly
in those countries whose economies depend
overwhelmingly on a few commodities. In
such cases the government's better access to
the credit markets also is limited. Developing
countries then need to be assured that the
outcome of the round will not exacerbate the
variability that already exists. If there is an
increase in variability, then each contracting
party would presumably like to see itself pro-
tected from its consequences. The issue fac-
ing the negotiators is what international action
can be taken to provide that protection.

International Action

This assessment of the current problems of
international agricultural trade indicates that the
major trading blocs have tended to exacerbate
price variability. Therefore, if developing coun-
tries are interested in reducing international
price variability,2 it is in their interest to push the
GATT to subject the border measures of the
OECD countries to greater discipline. Should
developing countries nonetheless feel that the
present trading system does not protect them
sufficiently from international market instability,
and that the modifications of the GATT rules
currently being discussed in the round would still
not provide them with adequate income security,
then there are other tools to help them meet their
objective. These are the various credit instru-
ments: the compensatory financing facility and
the food facility of the International Monetary
Fund and the STABEX scheme that the EC pro-
vides to ACP countries. Developing countries
may wish to explore the possibility of expanding
these facilities. An avenue that should not be
tried is a new buffer-stock scheme. The history
of such schemes has been uniformly unhappy.

National Action

National action to insulate domestic consum-
ers and producers from international market in-
stability would, under many of the proposed
rules, be disallowed under the GATT. Develop-

2 The other interests of developing countries, particularly those that are cereal importers and exporters of tropical products are
separate from what GATT calls "agriculture," are put aside here. These countries would not be very keen to see the GATT remove
the subsidies that support present excess capacity in agriculture, because doing so would increase the prices, even though it might
reduce variability, the concern of this paper.

ing countries face a dilemma here. On the one
hand, it is in their interest not to have OECD
countries (particularly the EC) export their insta-
bility to the rest of the world. On the other hand,
they would themselves like to be free to stabilize
domestic prices for their producers and consum-
ers. This is a common dilemma in trade negotia-
tions. Every government would like to be free to
do what it wants other governments not to do.
Indeed, as Chapter 7 by Tyers indicates, if all
countries (including developing countries)
forego domestic market insulation policies, the
degree of world price instability would then fall
below the current degree of instability under the
current (insulated) domestic prices.
Developing countries, particularly the very
poor ones, do, however, have a weak case for
being exempted under some sort of (S&D) ar-
rangements from rules that would limit their abil-
ity to insulate their domestic markets. Because
they are developing countries, they do not have
a sophisticated credit market to which their agri-
cultural producers and food consumers (particu-
larly the poor among them) have access. Conse-
quently, some intervention on the part of their
governments can be justified. Since fiscally the
most cost-effective intervention would be at the
border (Siamwalla 1986),3 GATT rules could be
written in such a way as to allow them to do so,
even though it is recognized that such S&D treat-
ment for developing countries exacerbates the
international price instability of the agricultural
commodities involved. For this reason develop-
ing countries would gain more from this proposal
in the case of those commodities for which they

collectively have a small share in the world mar-
kets (for example, cereals) than, say, for tropical
It is to be expected that OECD countries
would require that this S&D treatment for devel-
oping countries be subject to some discipline.
Developing countries could agree to notify the
contracting parties of the proposed system of
intervention (for example, a variable levy) that
they intended to pursue, including the rules guid-
ing intervention by the state trading corpora-
tions. In that case many of them would have to
replace the current practice of ad hoc interven-
tions with a more clearly articulated system. This
reform would probably be in the interest of the
producers, consumers, and traders in the devel-
oping countries and make for a more sophisti-
cated marketing structure. Where public and pri-
vate marketing systems coexist, such a move
also implies a more competitive regime that
would impose some discipline on the perfor-
mance of the public system.
A second discipline for developing countries
is to make the notified system of intervention
demonstrably neutral in the long run (or at least
the portion that is not neutral should be subject
to tariffication).4 Such neutrality implies that sub-
sidies and taxes to reduce domestic price fluctu-
ations should balance out in the long run.
Whether such a discipline is desirable for the
developing countries themselves in the long run
is beyond the scope of this paper, as it pertains
to the level rather than the stability of prices, an
issue addressed elsewhere.

3 This argument is necessary to counter the objection that income support measures that are decoupled from prices could be
devised. The problem with this objection is that it would not only be costly but also, for most developing countries, administratively
4 Interestingly the EC has under discussion a proposal to split its variable import levy into a tariffied part (called a franchise) and
a smaller variable part


Australia, Commonwealth of. Bureau of Agricultural Economics. 1985. Agricultural policies in the
European Community: Their origins, nature and effects on production and trade. Policy monograph
no. 2. Canberra: Australian Government Publishing Services.
Australia, Commonwealth of. Australian Bureau of Agricultural and Resource Economics. 1988.
Japanese agricultural policies: A time of change. Policy monograph no. 3. Canberra: Australian
Government Publishing Services.
Deaton, Angus. 1989. Saving in developing countries: Theory and review. Paper presented at the First
Annual World Bank Conference on Economic Development, Washington, D.C., April 27-28, 1989.
Knudsen, Odin, and A. Parnes. 1975. Trade instability and economic development. Lexington, Mass.,
U.S.A.: D.C. Heath.
Siamwalla, Ammar. 1986. Approaches to price insurance for farmers. In Crop Insurance for Agricultural
Development: Issues and Experience, ed. Peter Hazell, Carlos Pomareda, and Alberto Vald6s.
Baltimore, Md., U.S.A.: Johns Hopkins University Press.


Thomas W. Hertel is an associate professor in the Department of Agricultural Economics, Purdue Univer-
sity, Lafayette, USA.

Nurul Islam is senior research adviser at the International Food Policy Research Institute.

Odin Knudsen is principal economist at The World Bank.

Ammar Siamwalla is a program director at the Thailand Development Research Institute in Bangkok.

Rod Tyers is senior lecturer in economics at the Australian National University in Canberra.

Alberto Valdes is director of the International Trade and Food Security Program at the International Food
Policy Research Institute.

John Whalley is a professor of economics at the University of Western Ontario, Canada, and a research
associate at the National Bureau of Economic Research, Cambridge, USA.

Joachim Zietz is professor of economics at Middle Tennessee State University, Murfreesboro, USA.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs