Green Box

International
Food & Agricultural Trade

Policy Council
IPC Discussion Paper

March 2007
Should the Green Box be Modified?

by david blandford and timothy josling
This paper was made possible with support from the William and Flora Hewlett Foundation.

1616 P Street NW, Suite 100 Washington, DC 20036 / USA / tel 1 202 328 5056 / fax 1 202 328 5133 / www.agritrade.org
IPC finds practical solutions that support the more open and equitable trade of food & agricultural
products to meet the world’s growing needs.
© 2007 International Food & Agricultural Trade Policy Council

All rights reserved. No part of this publication may be reproduced by any means, either electronic or mechanical, without
permission in writing from the publisher.
Published by the International Food & Agricultural Trade Policy Council

Layout and Design: Yvonne Siu
Membership of the International Food & Agricultural Trade Policy Council
Piet Bukman (Chairman), The Netherlands Marcelo Regúnaga (Vice-Chairman), Argentina

Allen Andreas, United States Tim Groser, New Zealand Michel Petit, France
Bernard Auxenfans, France Carl Hausmann, United States Per Pinstrup-Andersen, Denmark
Malcolm Bailey, New Zealand Jikun Huang, China Henry Plumb, United Kingdom
Joachim von Braun, United States Rob Johnson, United States Eugenia Serova, Russia
David Blackwell, United States Hans Jöhr, Switzerland Hiroshi Shiraiwa, Japan
Csába Csáki, Hungary Timothy Josling, United Kingdom Jiro Shiwaku, Japan
Pedro de Camargo Neto, Brazil Mike Mack, Switzerland James Starkey, United States
Luis de la Calle, Mexico Rolf Moehler, Belgium Jerry Steiner, United States
Leonard Condon, United States Raul Montemayor, Philippines Robert L. Thompson, United States
H.S. Dillon, Indonesia Nestor Osorio, Colombia M. Ann Tutwiler, United States
Cal Dooley, United States Carlos Perez del Castillo, Uruguay Ajay Vashee, Zambia
Franz Fischler, Austria C. Joe O’Mara, United States
Michael Gifford, Canada J.B. Penn, United States

March 2007
Should the Green Box be Modified?*
T
he concept of agricultural subsidies that have no (or at most minimal) production and trade-distort-
ing effects was introduced in the Uruguay Round Agreement on Agriculture (URAA). It provided a
direct link between the parallel processes of domestic agricultural policy reform, and reform of the
multilateral trading system. The process of moving from price support to direct payments has been an essen-
tial part of domestic reforms, offering better targeting of support and reducing surplus production. Rules that
ensure that domestic agricultural programs are more consistent with open international markets will continue
to allow for the development of a more satisfactory multilateral trade system. The designation in the URAA of
certain types of agricultural payments that are free from reduction commitments (i.e., put in the Green Box)
is thus pivotal for the continuation of domestic policy reform, and for acceptance by other countries that the
policies adopted are desirable for the development of international trade.
Despite this, many think the current Green Box criteria permit policies that have significant trade-distorting
effects. Others consider that Green Box definitions do not allow for the development of a full range of legiti-
mate domestic programs, such as those relating to environmental issues, without payments being counted
against domestic support limits. Yet others claim that the Green Box does not take into account the difficul-
ties that developing countries have in providing direct payments, and underplay the desirability of payments
coupled to production and price in the context of rural development. Another concern is that widespread use
of Green Box measures by developed countries gives them a competitive edge, even if direct output effects
are difficult to determine.
As a result of these concerns, several countries have called for a review of Green Box payments. The main
users of the Green Box, the US and the EU, have argued against re-opening definitional issues. As a result,
the Framework Agreement of August 2004 called for negotiations to “review and clarify” the existing criteria
(WTO 2004a), a step short of actually changing the definitions themselves. Little discussion on the topic actu-
ally took place before the Doha negotiations were suspended in July 2006, and it is unlikely that the Green
Box will form a major part of the talks after the resumption of the negotiations.
The lack of focus on Green Box criteria does not, however, mean that the issue is unimportant. On the con-
trary, the Green Box is significant because its existence is tied to the reduction of Amber (and Blue Box) pay-
ments. As these more trade-distorting forms of subsidies are reduced, the conditions under which payments
can qualify for the Green Box become crucial. A credible WTO outcome to the Doha Round that includes
deep cuts in trade-distorting support requires agreement on criteria for policies considered non-trade-distort-
ing. Thus, even if Green Box issues will not form a major part of the Doha talks if they resume, WTO members
may need to agree on a “built-in” agenda to tackle these issues upon conclusion of the Round.
The Green Box will likely remain contentious, even if the Doha Round is abandoned. Political discussions on
the criteria have been divisive, indicating that there are likely to be future challenges to notifications. And, fol-
lowing the ruling by the WTO cotton panel, the legal status of some policies currently notified as Green Box
is unclear (WTO 2004b, 2005a).
*David Blandford is Professor in the Department of Agricultural Economics and Rural Sociology, Pennsylva-
nia State University, and Timothy Josling is a Senior Fellow at the Freeman Spogli Institute for International
Studies, Stanford University. Thanks are extended to participants in the IPC plenary meeting in Washington,
October 2006, for helpful comments on an earlier draft.

In this paper, we discuss the current criteria and significance of the Green Box to the process of domestic and
trade policy reform. We focus on three related topics:
• Issues that need to be addressed in a Green Box modification, in particular (a) plugging loopholes to
prevent the use of the Green Box to avoid other disciplines; (b) preserving incentives to move away
from trade-distorting policies; (c) allowing developing countries to pursue legitimate development
policies for their agricultural sector; (d) finding a better way to monitor the use of the Green Box; and
(e) avoiding confusion between allowable Green Box policies under the URAA and actionable subsi-
dies under the Subsidies and Countervailing Measures Agreement (SCM);
• Changes already proposed in the Doha Round negotiations, and what additional modifications might
improve the transparency and effectiveness of constraints on policies that fall under the Green Box;
and
• The implications of changes in Green Box definitions for policies in the major developed countries,
particularly with reference to options currently under discussion for the U.S. 2007 Farm Bill, for
example an expansion in the use of conservation payments, and what reforms of the Common Agri-
cultural Policy (CAP) imply (and may imply in the future) for use of the Green Box by the European
Union.
Background
The Green Box was introduced in the Uruguay Round to encompass government payments that were con-
sidered trade-neutral, or at least minimally trade-distorting. As such, it serves the same purpose as the
class of permitted subsidies in the non-agricultural area, as defined by the SCM.1 In the Uruguay Round of
negotiations, WTO members realized that they would not be able to agree on disciplines for all agricultural
payments. Negotiators agreed however that rules governing border instruments would not be sufficient to
correct the trade-distorting impact of farm programs, and felt that it was essential to impose some discipline
on domestic farm policies. Hence, it was essential to adopt a classification of agricultural subsidies that distin-
guished between those that are trade-distorting and those that are relatively trade-neutral, in order to include
all domestic support instruments under WTO disciplines.
The fundamental criterion for classifying a subsidy as “trade-distorting” is linkage between the subsidy and
the incentive to produce. Subsidies that directly affect output or input prices, or vary with the quantity of out-
put, are most likely to provide an incentive to expand production. However, those that have no direct link to
prices or output (i.e., are “decoupled”) were viewed as less likely to expand production.2 Hence, the Green
Box is tied inextricably to the notion of decoupling, and Green Box subsidies were thought to have a minimal
impact on production.
Countries have taken very different positions on both the interpretation of current Green Box rules and
the need for changes in the way the Green Box is defined and disciplined. Those that view the Green Box
positively see the advantage of permitting the inclusion of both direct income payments and environmental
subsidies, and sheltering these from challenge, even if they are not completely production neutral. Propo-
nents view shifting from highly coupled support to largely decoupled support (one form of box-shifting) as a
constructive contribution to the reform of domestic agricultural policies. There is also the added advantage
that decoupled payments can be “re-coupled” to reward desirable ecological practices and to provide public
goods, such as rural amenities.

March 2007
In contrast, those who are skeptical about the Green Box are concerned that it simply provides a means for
delivering income support to farmers under another guise. From this viewpoint, new justifications for support
to farmers, such as the provision of public goods, are a political convenience without economic rationale.
Even in cases where public goods can be identified (e.g., a positive contribution to environmental quality), the
payment provided may exceed that necessary to elicit the desired supply of a public good. Moreover, many
other sectors of the economy provide public goods, but are not given the same treatment as agriculture.3
If the intent of Green Box policies is to redistribute income from taxpayers to farmers, rather than provide
public goods, one might question whether such an objection is relevant. If governments choose to redistribute
national income among various groups, that is essentially a domestic political issue and should not be dictat-
ed by trade rules. A more substantial claim is that Green Box payments, whatever their rationale, are likely to
distort trade (i.e., are unlikely to be fully decoupled) and so need to be disciplined. In that case, shifting from
Amber to Green Box support may simply keep resources in agriculture that would otherwise be used else-
where, negating the beneficial adjustment and trade effects of a reduction in Amber Box subsidies. Clearer
rules are needed to prevent countries from switching the method of payment (or how payments are labeled)
simply to avoid subsidy constraints. That type of “box–shifting” may do little to remove trade impacts.
Significance of the Green Box

The Green Box is politically important for two reasons. First, it offers countries the possibility of making pay-
ments to the agricultural sector in support of objectives other than income support. The traditional argument
for such support – chronically low income for farmers and farm families – is less relevant today in many devel-
oped countries than in the past, particularly when considering farm families’ combined income from farming
and off-farm activities. Other objectives are becoming more important, such as providing public goods, or im-
proving efficiency through investment in rural infrastructure and research and extension. Payments targeted
to such objectives should not be “counted” as trade-distorting subsidies by trading partners (even though they
may have an indirect impact on trade flows). Second, the Green Box provides countries a mechanism for
moving from price- and production-related policies towards “decoupled” measures that are less disruptive of
trade. Typical among such policies are direct payments based on past participation in price-related programs,
often given as compensation fo decreases in price-related programs.
For these arguments to be convincing, two further conditions must hold. First, payments must be commen-
surate with the cost of achieving policy objectives. In particular, the provision of public goods must not be a
conduit through which income support is channeled to agriculture to keep resources employed in the sector
that could otherwise be employed elsewhere. And second, subsidies must be provided in such a way that
they do not generate incentives to produce more agricultural output than market conditions would justify.
Even if total decoupling is an unrealistic goal, there must be agreement that new policies limit incentives to
increase production.4
The attractiveness of these arguments depends on the trade stance of countries and the extent of their
domestic programs for agriculture. The exporting countries of the Cairns Group, for instance, are wary that
a Green Box that is too accommodating will invite abuse. They are less convinced by the public goods ar-
gument, perhaps because of their lower population density and hence lower demand for rural amenities by
urban residents. Developed importing countries, such as the EU, start with both a higher level of protection
through domestic programs and a greater demand for rural amenities.

Green Box supporters in developed countries argue that it is not politically possible to reduce subsidies
without providing an option for moving to less trade-disruptive alternatives. Green Box payments provide a
means for making the transition out of trade-distorting measures. From this viewpoint, “box-shifting” is a sign
of success, not a loophole to be plugged; it should be made as easy as possible to facilitate the move away
from price- and production-linked subsidies. And if the public is receptive to the idea of “re-coupling” pay-
ments to environmental objectives, then such a move satisfies both domestic and trade priorities. Implicit in
this argument is the principle that “the best should not be the enemy of the good” – setting the bar too high
for environmental or other payments may play into the hands of those who would prefer to keep current farm
programs.
It may be impossible to design Green Box criteria to accommodate these differing views completely. Howev-
er, an acceptable compromise could be one in which payments under a range of policy objectives are allowed
under the Green Box, but with criteria that are sufficiently well-defined to prevent abuse. In order to achieve
this, improved monitoring of Green Box measures is needed. We examine these issues in the remainder of
this paper.
Current Green Box Criteria

The Agreement on Agriculture (Annex 2) identifies the set of payments that are exempt from reductions.
General Criteria
There are general criteria for payments that fall within the Green Box and specific criteria for individual pay-
ment types. All payments have to meet the general criteria, but they must also meet specific criteria for the
category under which they are notified.
The fundamental requirement is that Green Box payments should have “no, or at most minimal trade-distort-
ing effects on production” (paragraph 1). Two criteria are specified to ensure that this requirement is met:
1. support should not involve transfers from consumers; and
2. measures should not provide price support to producers.
The restriction that a payment be publicly funded reflects the need to exclude transfers generated through
the market by raising output prices or reducing input costs. This criterion is consistent with the move from
coupled to decoupled policies. Though the condition appears unambiguous, it could be difficult to apply in
situations where the consumer is the purchaser of a farm-related service. For example, suppose a payment
is provided to farmers for the provision of landscape amenities through the enforcement of property rights.
The government may permit controlled access to the countryside (e.g., through the creation of parks) and
consumers could be required to pay an amenity charge, which is then distributed to farmers. A strict interpre-
tation of the general condition would cast doubt on the Green Box conformity of such a scheme, even though
it results in the creation of a market for a previously un-priced service.5
Similarly, the term “price support” is ambiguous when payments are linked to environmental performance,
or to the method of production. For example, in order to receive payments, farmers may be required to keep
land in a particular agricultural use (e.g., as pastureland for grazing livestock), or to keep land in good agri-

March 2007
cultural condition. It is an open question as to whether such conditions are an implicit form of price support for
the production of agricultural commodities. Even more problematic may be cases where decoupled policies
are administered in a way that makes them more accessible to farmers enrolled in price- or production-linked
subsidy programs. If eligibility for or access to decoupled payments is not clearly separated from “Amber Box”
subsidies, they may be ineligible for the Green Box.
These exceptions aside, most subsidies are likely to fall clearly on one side or the other with respect to the
fundamental criteria.
Specific Criteria
The types of government programs specified under the Green Box (Annex 2 of the URAA) are grouped under
eleven headings:
1. the provision of general services for agriculture, such as research and extension, pest and disease
control, product inspection, and technical assistance and training for producers;
2. expenditures on public stockholding for domestic food security;
3. expenditures on domestic food aid;
4. payments to producers for decoupled income support;6
5. payments for income insurance and income safety net programs;
6. subsidies for disaster relief (including crop insurance);
7. payments designed to promote structural adjustment through producer retirement;
8. payments designed to promote structural adjustment through resource (land) retirement;
9. payments designed to promote structural adjustment through investment;
10. payments under environmental programs; and
11. payments to producers in disadvantaged regions under regional assistance programs.
The first of these headings is the broadest, including many of the programs that governments, in both devel-
oped and developing countries, have operated for decades to assist growth and development of the agricul-
tural sector. No government has suggested that expenditures on such programs be limited. The next heading
refers to stockholding and food aid, and also includes substantial programs that are not usually considered
“distorting” to international trade. Payments identified under the next three headings are more directly related
to farm income goals, including protecting against fluctuations, and are among the most contentious in the
Green Box. The three “structural” programs are of a longer run nature, benefiting not just individual farmers,
but improving the performance of the sector as a whole. Payments under environmental programs are con-
sidered more beneficial to society, and regional payments address situations in which local conditions create
permanent economic disadvantages.
The criteria for each category of payments are summarized in Table 1.7 They appear precise, but still leave
plenty of room for interpretation. A key aim underlying the definition of the specific criteria is to prevent pay-
ments that provide incentives to increase production. However, there is also recognition that payments may
be associated with the provision of public goods. The balance between allowing for such public good aspects
and avoiding trade distortions is difficult to determine conceptually, and even more difficult to achieve in
practice.

As the programs that fall under these headings are designed to respond to very different social needs, the
criteria are tailored accordingly. There are differences in the extent to which the criteria address potential con-
cerns of competitive producers in other countries. The following discussion points up some of the difficulties
of including so many different types of policies in the Green Box. The eleven categories distinguished in the
Green Box are grouped into five main types of policy objectives: general services, food security, structural
assistance, income support and environmental programs.

March 2007
Expenditure on General Services

Several sub-categories of expenditures (or revenue foregone) are included under this heading. These are:
1. research;
2. pest and disease control;
3. training services;
4. extension and advisory services;
5. inspection services;
6. marketing and promotion services; and
7. infrastructural services.
General services are included in the Green Box even though they can affect production. Indeed, they may be
more important than coupled subsidies in influencing agriculture’s productive capacity. As with non-specific
subsidies in the SCM, this class of domestic subsidies is considered to be permissible even though it may
have trade impacts. The underlying logic is that many of these expenditures have public good aspects. This
is most obvious for expenditures on inspection to ensure health and safety, but there are public good com-
ponents in the control of agricultural pests and diseases, and in agricultural research. The control of pests
and diseases is likely to lead to higher marketable production, but it would not make sense to object to such
expenditures on that basis. Society as a whole will value a pest and disease-free agriculture, regardless of
any resulting financial gains for individual farmers.
Expenditures on research that result in increased productivity or on infrastructure that lowers transactions
costs are likely to expand agricultural output. While there will be gains to producers from such expenditures,
much of the benefit will be passed onto consumers through lower food prices. It would be difficult to argue

from either an economic or a political perspective that expenditures of these types should be considered
“trade-distorting,” although they may affect the demand for imports or supply of exports.8
The public good argument for some of the other expenditures under the general services category is less
clear-cut. Public funding for training and for extension and advisory activities may contribute to a general
increase in productivity in the farm sector (similar to research) with consequent beneficial effects for consum-
ers. This would be the case, for example, for activities directed to the adoption of technologies leading to im-
proved productive efficiency. However, training and extension activities can be directed primarily at increas-
ing the income of individual farmers, as in the case for activities designed to improve business management
and marketing skills.9 To the extent that higher income (increased returns) helps to retain resources in agri-
culture, there could be a production and trade-distorting effect from such activities. Where the public sector
competes with the private sector in the provision of services, there would seem to be a strong argument that
public expenditure be considered a production and trade-distorting subsidy. Here again we see that the issue
is the balance between private and public benefits resulting from improvements in human capital.
Expenditures on marketing and promotion services also appear to generate private gains rather than solely
a public good. Despite the restriction that expenditures should not allow sellers to reduce their selling price
or to provide any other direct economic incentive for purchases of the product, this category of expenditures
could be viewed to be a specific subsidy under the criteria of the SCM.
Thus, many of the payments under the general services heading create general benefits to society beyond
any direct advantage to producers. The prospect of challenges by other countries to expenditures on re-
search and infrastructure, in particular, seem to be remote, and certainly not in the spirit of international trade
as an engine of development.
Financing for Public Stockholding for Domestic Food Security

and Domestic Food Aid
Expenditures on domestic food security and assistance are included in the Green Box. In addition to evident
humanitarian benefits, there would seem to be little objection if such schemes increase effective demand for
food to the potential benefit of both domestic and foreign suppliers. This is the case for schemes that increase
the real income of a segment of the population, such as the U.S. Food Stamp Program, which forms a large
part of the U.S. notification for the Green Box (see below). Under this program, there is no requirement that
food stamps (which are equivalent to a cash transfer) have to be used to purchase domestic products in
preference to imported ones. In contrast, food aid schemes that rely on the public procurement of domestic
products or restrict consumer purchases to products of domestic origin could be considered a specific sub-
sidy within the criteria of the SCM. Schemes that rely on the direct distribution of food to target groups, rather
than increasing their ability to purchase food, are particularly susceptible to such a judgment.
Public stockholding for food security purposes may provide a limited margin of preference for domestic
producers. The effect on competing suppliers is likely to be limited, provided that domestic markets are rea-
sonably open to imports.10 Taking all these factors into consideration, these components of the Green Box,
though possibly having some trade impact, are relatively uncontroversial.
Payments for Income Support, Income Insurance and

Disaster Relief, and for Producers in Disadvantaged Regions
The inclusion of this category of payments in the Green Box raises many issues. As noted above, some other
types of Green Box payments may have production and trade effects and could be considered specific sub-
sidies under the criteria applied in the SCM. This may be even more so for this category of payments.
10
March 2007
A key issue is the effectiveness of specified constraints on the linkage between payments and production.
These are as follows:
1. several types of payments (decoupled income support, income insurance and income safety-net
payments) are permitted only if these are not linked to the volume of production or factors of production em-
ployed in any year after the base period; and
2. payments made under regional assistance programs (limited to disadvantaged regions) cannot be
related to production, other than to reduce the volume of production and must be available to all producers
within a region. Payments may be linked to production factors but must be made at a degressive rate above
a threshold level of the factor concerned, and must be limited to the extra costs or loss of income involved in
undertaking agricultural production in the disadvantaged area.
The effectiveness of such criteria in limiting any production-enhancing effect of payments is at the heart of
the controversy over the Green Box. With the notable exception of decoupled income support, sub-criteria
are defined that attempt to limit the amount of payments. There are caps on the compensation that can be
provided through income insurance and safety-net programs and through disaster relief. This presumably
reflects the view that payments that vary inversely with annual farm income are likely to affect production
decisions, and there is a need to prevent such payments from being open-ended.
Trading partners may be relatively unconcerned about any production enhancing effects of payments to farm-
ers in disadvantaged areas since these areas often have low agricultural productivity. Any resulting produc-
tion and trade distortions might be modest. There is likely to be greater concern about the lack of constraints
on general decoupled income support, since this can be paid to highly productive farmers. The critical issue
then becomes the extent to which apparently decoupled income support might influence production. This is-
sue is discussed in more detail below.
Structural Adjustment Payments

Three categories of payments are identified under this heading:
1. Producer retirement;
2. Resource retirement; and
3. Investment aids.
These payments are designed to remove resources from agriculture or to promote farm restructuring. The
producer retirement provision is designed to ensure that payments are made for the permanent exit of farm-
ers from commercial production. The resource retirement provisions are designed to facilitate the retirement
of land for a minimum of three years and the liquidation of livestock herds. No limitation is placed on the du-
ration or size of payments in either case. The retirement of land from agriculture and liquidation of livestock
herds should be expected to reduce production. Payments to induce the retirement of farmers might have
a similar effect, although if the retiree transfers land and other resources to another operator who manages
these more efficiently or combines them with an existing farm operation to exploit scale or size economies,
output could increase.11 In general, however, trading partners will not likely object strongly to the use of mea-
sures that reduce the resources employed in agriculture.
Another issue arises in connection with the land retirement criteria – the requirement that retired land should
not be used for “marketable agricultural production.” This raises the potentially divisive issue of what products
11
should be included in this category. Switchgrass, a promising cellulosic feedstock for ethanol, might not be
considered as a marketable agricultural product, as there is no other market for food or feed (Howse, et. al.
2006, p. 21). However, payments for production of ethanol feedstocks that have an agricultural use on set-
aside land could perhaps be excluded from the Green Box. 12
The status of investment aids is less clear-cut than the resource retirement provisions. These payments are
designed “to assist the financial or physical restructuring of a producer’s operations in response to objec-
tively demonstrated structural disadvantages.” (Annex 2, paragraph 11a). If there are efficiently functioning
capital markets and if investment in restructuring generates a positive net return, one might question why
government would need to provide investment aid. The inference might be that there is actually no net com-
mercial return to restructuring, and that the aid is therefore a subsidy. To the extent that restructuring would
be expected to increase productivity and output, such aid might be considered a specific subsidy within the
criteria applied by the SCM.
Payments Under Environmental Programs

For payments under environmental programs, the criteria specify that these can be linked to production
methods and inputs, but that the amount of payment must be limited to the extra costs or loss of income
involved in complying with clearly defined environmental or conservation programs. As in the case of the
land retirement provisions discussed earlier, payments oriented to removing land from production appear to
be inconsistent with the basic Green Box requirement that subsidies should not be tied to output. However,
competing producers are unlikely to object to reductions in production, even though trade may strictly speak-
ing be “distorted” as a result. Thus programs that encourage farmers to withdraw land from crop production
(for example, in order to reduce erosion or to protect wildlife) are not considered a threat to other suppliers
and are unlikely to be challenged.
On the other hand, payments that reward producers for the provision of environmental services might be of
greater concern if these services are closely linked to production. The criteria seek to impose a constraint on
the use of such payments and their size by requiring that payments should not exceed the extra costs or loss
of income involved in complying with an environmental program.
These criteria appear to allow a wide range of programs to be included in the Green Box category, though
with the proviso that they would have to be crafted with the specific criteria to avoid challenge.13 Some of the
issues raised by current environmental programs in the EU and US are discussed in more detail later in the
paper.
Payments under Regional Adjustment Programs

The criteria for regional payments attempt to impose constraints in recognition of the fact that maintaining
agriculture in a region will inevitably keep production at a higher level than otherwise. The strongest limitation
appears to be that payments must be limited to the “extra costs or loss of income involved in undertaking
agricultural production in the prescribed area.” This criterion is of dubious effectiveness since the extra costs
or income losses of undertaking agricultural production will inevitably be related to the amount of agricultural
production in an area. Consequently, an effective constraint on the amount of payments used to support in-
comes to maintain agricultural production in disadvantaged areas appears lacking.
Issues and Problems

A number of critical issues are raised by current Green Box criteria. These issues and the possible need for
clarification of the criteria can be separated into those of a conceptual nature, those that relate to the legal
12
March 2007
definitions of the Green Box subsidies, and those that are more empirical. Each of these will be considered
briefly before addressing possible improvements.
Conceptual Issues
Constraints on domestic support introduced in the Uruguay Round Agreement were primarily designed to re-
strain the use of price supports, deficiency payments, and input subsidies. The focus was on curbing the use
of payments that stimulate domestic output through means other than restricting imports or encouraging ex-
ports. The Green Box as a safe haven for decoupled subsidies from current price or output was an important
step in the process of introducing disciplines on domestic farm programs. It moved countries along a path in
line with domestic notions of policy reform: the better targeting of payments and shifting the focus of support
away from commodities and toward farmers. It was not intended to deal with the range of rural subsidies that
have emerged in recent years for a variety of different purposes.
The argument in favor of allowing decoupled income support payments to be unconstrained is simple and
persuasive – up to a point. Subsidies that expand output cause distortions in trade patterns, displacing more
efficient producers in other countries. This was the basis for the inclusion of domestic subsidies under Ar-
ticle XVI of the GATT: trade rules were focused on reducing the externalities imposed on other countries by
subsidies “within the border” as a necessary complement to reducing border protection. This argument has
its strongest appeal when the objective of the subsidy is income support for farmers in rich countries. Such
was the case until the early 1980s when the focus of rural policy was primarily on the maintenance of farm
incomes in industrialized societies.14
Unfortunately, the argument is conceptually weaker when we consider other motivations for subsidies. There
are several types of subsidies to rural businesses, including but not limited to farmers, for which the concept
of decoupling from output has less relevance (Table 2). The provision of public goods, such as recreational
and scenic amenities, may be related to the type and method of production of private goods, such as ag-
ricultural crops. And the stimulus to economic activity in rural areas might be a legitimate aspect of rural
development in rich or poor countries. Some of these objectives can be met by the use of direct payments
and targeted subsidies, but others may have to be achieved by encouraging particular farming activities. If
there is a need to encourage the production of public goods provided by the farm sector, or if their production
plays a key role in the development of regional economies, one has to admit the possibility that decoupled
payments are not the most efficient method of providing public goods. Hence, there is an underlying tension
between the granting of subsidies for achieving domestic goals and the possible impact this can have on
competing producers in other countries.
At the heart of the issue is the question raised above – whether the private goods being subsidized are joint
products with the public goods that are underprovided by the market. If they are indeed joint products then it
is quite likely that a purely decoupled instrument will be inefficient in achieving policy aims. This would not,
of course, condone all farm subsidies, or even imply that there are many instances of public goods that can
be most efficiently provided by subsidizing a jointly-produced private (agricultural) good. But to assume that
no such cases exist is to open trade rules to the criticism that the importance of public goods is discounted
relative to private goods. The challenge is to find the appropriate rules that will allow “legitimate” objectives
to be followed with a minimum of disruption to other countries.15
Similar arguments apply to other aspects of “multifunctionality,” such as granting subsidies to compensate for
loss of competitiveness as a result of stricter animal welfare standards, the promotion of rural development,
and the gamut of programs under the heading of food security. In each case, the same question arises: is
13
there a necessary link between domestic farm output and the achievement of the goal? If so, the argument
for decoupling is weakened. If not, lump-sum payments given as compensation or to redistribute income are
the desired policy responses.16
In the light of the conceptual ambiguity surrounding the Green Box, and of decoupling more generally, the
best we can expect from trade rules is a political agreement that has a broad rationale, a mechanism to moni-
tor the application of the trade rules to provide full transparency, a way of ensuring that the rules themselves
do not hamper desirable development policies, and a process for handling changes in policy instrumenta-
tion.
Legal Issues
Much of the discussion of the compatibility of existing programs with the Green Box is by nature speculative.
There has been no WTO dispute panel ruling specifically on Green Box compatibility. No government has
challenged the Green Box notifications of others. In the one case where a notification was questioned by a
Panel (the U.S. cotton case, see below), no country has demanded that the US re-notify past domestic sup-
port (AMS) totals.17
Lack of successful challenges thus far does not mean that none is possible in the future. One such challenge
could be on the basis that a payment for environmental services that does not simply compensate produc-
ers for compliance costs or income foregone may have a non-trivial impact on output. Most programs that
involve payments for environmental services include an incentive component, i.e., an element that rewards
producers for adopting environmentally-friendly practices.18 As such, this may be actionable under the SCM.
The United States has argued for a renewal of the Peace Clause in part to avoid such challenges. But this
may only be effective if the policies are Green-Box compatible. The likelihood of establishing a Peace Clause
14
March 2007
that covers AMS payments is more remote. So, the possibility of such a challenge remains. Legislation needs
to be crafted with this in mind. In particular, payments should not be restricted to specific farms, though they
can be regionally based, and should involve a mechanism through which a reasonable level for the payment
can be determined.19
The more serious challenge is not that of “serious prejudice” under the SCM, but of non-compliance with the
Green Box definition in the Agreement on Agriculture. Payments for environmental services would clearly
meet the general conditions of not being tied to production or prices. The specific conditions are more prob-
lematic. One might argue that rewards for ecosystem services from the farm are payments above and beyond
the costs of complying with an existing government program. However, this may depend on the way in which
the program is formulated. If the farmer has a choice between joining an environmental stewardship pro-
gram or keeping eligibility for current farm income payments (direct payments, counter-cyclicals, etc.) then
income forgone includes the amount of those payments. And if the payments are calculated with the cost to
the farmer in mind, it would be difficult to argue that they are not equivalent to the “extra costs” involved in
complying with the program.20 So all, or most, of the stewardship payments might be notified as Green Box
payments.21
A WTO ruling on a dispute between Brazil and the United States over U.S. upland cotton programs recently
called into question the legality of certain Green Box payments (WTO 2004b, 2005a). The essence of the
case was that a wide range of U.S. domestic subsidies, as well as some export programs, had caused seri-
ous prejudice to Brazil by depressing world cotton prices and increasing the U.S. share of world exports. The
domestic programs challenged included some notified as Amber Box support (including marketing loans),
some Green Box payments (including direct payments), and some not yet notified but potentially includable in
an expanded Blue Box (counter-cyclical payments). The United States disputed Brazil’s allegations, arguing
that the Peace Clause sheltered many of these programs from challenge under the subsidy rules, and that in
any case serious prejudice had not been the result of U.S. programs. The WTO Panel found that the export
programs were in breach of U.S. limits for such subsidies and that many of the domestic programs did in fact
cause serious prejudice (WTO 2005a).
As the panel was considering the extent to which U.S. direct payments were correctly notified under the
Green Box, they concluded that some do not satisfy the fundamental requirement, because they are linked to
production. In the panel’s judgment, this linkage was created by limitations on planting flexibility for land upon
which the payments are based. Producers who wish to receive payments cannot plant fruits and vegetables
on eligible land. This was interpreted to mean that there is, in fact, a linkage between payments and produc-
tion and that, consequently, they do not qualify for the Green Box.22 This judgment is highly significant for the
design of programs that countries wish to be exempt from reduction commitments – since it appears to imply
that any direct linkage established between payments and production decisions would make their Green Box
status potentially open to challenge.
The distinction between the political decision in the Uruguay Round Agreement on Agriculture to classify sub-
sidies on the basis of whether they were considered trade distorting, and therefore had to be reduced, on the
one hand, and the political decision in the Subsidies and Countervailing Measures Agreement to distinguish
among prohibited and allowed subsidies on the other, is at the heart of the problem. Subsidies can be noti-
fied to the Agriculture Committee as Green Box by the country concerned and not be challenged. But a policy
could still be actionable even if allowable, and it could be found to cause serious prejudice.23 Moreover, the
treatment of agricultural subsidies in the URAA was premised on a firm distinction between domestic support
15
and export subsidies, but both were legal if within scheduled commitments. The Canada dairy case (WTO
1999) cast doubt on the validity of that distinction when applying WTO rules: domestic programs can have the
effect of providing an export subsidy. The EU sugar case (WTO 2005b) rested on the same premise, and the
result was similar. The SCM includes export subsidies as one of two types of prohibited subsidies. Thus, any
such subsidy is actionable (since the expiry of the Peace Clause) and the challenge does not have to show
impacts on other countries. So, a panel can declare a policy deemed to fall under the heading of domestic
support in the URAA to be an export subsidy when viewed through the lens of the SCM.
The SCM Agreement gives a legal definition of the term “subsidy that must have three basic elements:24
• it must be a financial contribution;
• it must be made by a government or any public body within the territory of a Member; and
• it must confer a benefit.
However, even if a measure is a subsidy under the definition of the SCM Agreement, it is not subject to the
disciplines of the SCM Agreement unless it is a specific subsidy.25 Specific subsidies fall into two categories:
those that are prohibited and those that are allowed. Two types of subsidies are prohibited: export incentive
subsidies that are contingent on export performance, and local content subsidies granted for use of domestic
inputs over imported goods. The WTO SCM Agreement provides a clear process through which to identify ac-
tionable subsidies. All specific subsidies are actionable under the SCM agreement. But the SCM agreement
deals differently with prohibited subsidies and other actionable subsidies, depending on the trade distorting
nature of specific subsidies. For non-prohibited actionable subsidies, a Member can initiate remedial mea-
sures only if it can prove there is a serious prejudice to its interests. Serious prejudice may arise in any case
where one or more of the following apply: displaced domestic imports; displaced exports to the third country;
significant price undercutting; and increase in world market share.26
The URAA, on the other hand, permits domestic subsidies and (existing) export subsidies. However, Mem-
bers undertake not to provide export subsidies except as specified in that Member’s schedule, and gradually
to reduce overall levels of domestic support, measured by the total Aggregate Measure of Support (AMS).
The classification of policy instruments into export subsidies and domestic support is crucial to the current
Round of negotiations. Export subsidies are expected to be phased out and domestic support payments are
to be further reduced. But countries do not currently know what subsidies are likely to pass the test if exam-
ined by a panel. This has implications for current negotiations on disciplines on the boxes. Why negotiate
separate restraints if these are not found relevant by WTO panels? These are the challenges that the cotton
and sugar cases have raised.
Empirical Issues
Empirical evidence complicates the question of whether to clarify or modify the Green Box criteria by estab-
lishing a link between various types of policy measures and their expected output and trade impact. The con-
cept of the Green Box, as containing subsidies that are trade-neutral, has come under increased scrutiny. On
a purely practical level, it is not easy to define “minimally trade-distorting” policies. Abler and Blandford (2005)
summarize the reasons why payments that are apparently decoupled may, in fact, influence production:
1. Income effects. The additional income generated by payments enhances the ability of producers to
cover fixed and variable production costs. Payments may serve to keep production higher in the short-run,
if that production would otherwise be unprofitable at prevailing market prices.27 For this to apply, producers
must choose to use the payment to cover production costs, rather than to increase their consumption or
16
March 2007
savings. Producers may also choose to use some of the increased income generated by payments to make
longer-term investments in their farm operations. They may do this because they have a preference for in-
vesting in agriculture rather than in other sectors of the economy or to take advantage of non-transferable
skills. Alternatively, some producers may be operating on the downward portion of their average total cost
curve, i.e., under increasing returns to scale. In such a case, it would be economically rational to use some
or all of the additional revenue from direct payments to expand output.28
2. Capital constraints. If producers face limitations in securing capital from traditional lenders due to
imperfections in the capital market, the additional income generated by payments may allow them to make
production-enhancing investments in their farming operations. In addition, lenders may be more willing to
lend to producers if the risk of non-repayment is considered lower through the additional income provided
through a known stream of future payments, rather than having to rely on the uncertain stream of producers’
market returns. To the extent that payments increase the value of fixed assets, particularly land, an increase
in farmers’ equity may also enhance their ability to secure loans.
3. Risk aversion. If producers are risk averse, the increase in wealth created by payments may make
producers less risk averse than otherwise, causing them to expand production by planting crops on land that
would otherwise considered too risky.29 This effect might be strengthened if payments vary inversely with
market prices (as with U.S. counter-cyclical payments), since this would reduce income variability and pro-
vide an insurance effect.
4. Expectations. Even if producers are not risk averse, expectations about the conditions attached to
future payments might influence production decisions. Producers might expect there to be a future updating
of the area upon which payments are based. They might be reluctant to reallocate land from program crops
to other crops or to idle marginal land when prices fall in order to protect their future eligibility for payments.
5. Resource retention. Payments may prompt some producers to remain in agriculture, rather than
exiting the industry. If exit would result in land abandonment or the conversion of land to other crops, the
provision of payments would unequivocally result in the production of supported crops being maintained at a
higher level than would otherwise be the case.30
In summary, there are a number of reasons to believe that apparently decoupled payments will actually have
some impact on production. The magnitude of the impact of a given payment scheme is difficult to determine
a priori since it depends partly on the design of the scheme and, most crucially, on the response of producers
to it.31 In terms of the Green Box debate, the issue is the extent to which the current criteria are sufficient to
ensure that any production and trade distortions associated with direct payments are minimal. Decoupling
may not be the best criterion for the desirability of domestic policy that does not have farm income as its
(sole) objective. But the empirical question is whether the distortions that remain in a world of decoupled sub-
sidies are serious enough to offset the benefits of a shift from policies that tie payments to price and output.
What gets in the way of such a rational calculation is the politics (or political economy) of domestic subsidies.
Such subsidies become embedded in attitudes about farm programs in both exporting and importing coun-
tries, and in the perception of equity among developed and developing countries. Competing exporters will
protest strongly that subsidies are merely an excuse for favoring domestic producers over international com-
petitors. They do not get the benefit of the public goods provided, and do not have to stand in elections in the
subsidizing country. Conversely, the representatives of producers who receive subsidies will stress the public
17
good argument and reject alternative policies as infeasible. Nobody should be surprised by the rhetoric that
surrounds the issue of “multifunctionality” on both sides. Developed countries are in a position to pay such
subsidies if they wish to protect their relatively small farm sector. Developing countries cannot afford such
transfers, even if they wished to increase the provision of public goods in rural areas. They are also skeptical
of the argument that such payments in rich countries do not hurt them.
Negotiating Proposals
Several countries have proposed changes to the Green Box criteria in the Doha Round in an attempt to im-
prove the effectiveness of constraints on policies that fall under the Green Box heading. They can also be
judged on whether they aid the reform of domestic policies.
The first set of proposals focused both on the Green Box criteria and on whether to cap payments. Table 3
summarizes the opinions of the US, EU and Cairns Group in the first two tiers of the agricultural talks.32 The
Cairns Group argued for significant modifications, essentially narrowing the Green Box by moving structural
aids and safety net programs to the Amber Box, along with certain direct payments. The Green Box would be
restricted to payments for the provision of public goods. Canada would cap even those payments. In contrast,
the EU and the US argued that such a deconstruction of the Green Box would be counterproductive because
it would hinder the movement towards decoupled support. The EU went further and requested clarification
that measures to promote animal welfare and environmental programs are included in the Green Box, as well
as those tied to rural development programs.
In the Draft Modalities paper of March 2003, the then Chairman of the Agricultural (Negotiating) Committee,
Stuart Harbinson, suggested no change in the Green Box criteria, except to add compensation payments
for animal welfare regulations to those currently allowed under environmental programs, and to expand the
categories applied to developing countries. WTO members did not accept the Harbinson draft as the basis for
further negotiations, but some of the specific language of that paper may reappear in the Modalities.
18
March 2007
A sharp change in the dynamics of the Round took place in the run-up to the Cancún Ministerial. In August
2003 the US and the EU submitted a “joint “paper designed to accelerate the move towards agreed modali-
ties. But a group of developing countries, led by Brazil and India with support from the developing Cairns
Group countries, China and South Africa, considered the joint paper to be self-serving. A particular point of
concern was that US and EU domestic agricultural policies were going to be “let off the hook” to the detriment
of developing countries. Though the Green Box issue was not one of the major impediments to agreement
at the Ministerial Meeting in Cancún, member countries recognized it as being closely linked to achieving
significant reductions in trade-distorting support.
In 2005 the G-20 produced a proposal that laid out their view on preventing the Green Box from becoming
a means for continuing to provide support under a different label. The G-20 “Review and Clarification” paper
(G-20 2005a) makes the case for changes to address many of the issues raised above. The paper shows
unease not so much with the concept of the Green Box as with its contents. It identifies as a major problem
that the Green Box includes both “direct payments” often introduced to compensate for the removal of price
supports, and “public good payments” that encourage farmers to provide benefits to society other than their
marketed output. The G-20 would presumably prefer that the former be moved to the Amber Box or otherwise
be constrained. They argue that these payments are not truly production-neutral. But the EU and the US
are not likely to accept such a change, as policy changes in these countries have been geared towards the
greater use of direct payments. So the G-20 focused on the modification of the Green Box criteria to minimize
the wealth and insurance effects inherent in current direct payments even when these are not tied directly to
price or output.33
Regarding subsidies designed to address non-trade concerns (or to promote or reward multifunctionality),
there is now less controversy than earlier in the negotiations. Those pushing for wider recognition of such
concerns have been obliged to settle for “Green Box” shelter rather than protection through tariffs.34 Public
goods provision is recognized by the G-20 as not contributing significantly to trade problems, though some
exporters still query the apparently open-ended nature of payments in this realm. 35 As mentioned above, the
EU has long made it clear that it is keen to find a place in the Green Box for several of its current payment
schemes aimed at helping farmers change their farming practices to be more animal or environmentally-
friendly. But as the Harbinson draft (WTO 2003) included provision for these items, and the G-20 has not sug-
gested language that would exclude them, it seems reasonable to assume that they could be incorporated
in the Green Box.
With respect to the specific issues that have arisen through experience with Green Box payments, the G-20
“Review and Clarification” paper suggests several changes of wording. These include restraining direct pay-
ments by eliminating the requirement that land be kept in agricultural use and by preventing the updating of
bases or addition of new products that may encourage farmers to anticipate increases in payments in the
future. The first change is primarily an extension of the existing provision that “no production be required in
order to receive [direct] payments.” The new provision would add the phrase that “land … shall not be re-
quired to be in agricultural use.” Though the logic behind this change is impeccable, its political acceptability
is in doubt. The prospect of the depopulation of agricultural regions through the abandonment of farming still
concerns governments, particularly in Europe. In that case one could argue that there is some “public good”
benefit to subsidies that keep agriculture viable. So the justification of a subsidy tied to the continuation of
some form of agricultural production could shift from income support to rural amenity provision, and the sub-
sidy could still be in the Green Box.
19
The G-20 review paper suggests that decoupled income support be targeted to farmers with low levels of
income, and based on production in a fixed and unchanging base period. The paper also seeks to limit direct
payments in cases where Amber Box and Blue Box payments exceed a certain percentage of the value of
production.36 The agreement to fix the base is entirely within the spirit of the current rules, and could play
a useful role in changing expectations on the part of producers. But any attempt to cap total Green Box
payments is much less likely to be successful and much more difficult to support as a wise strategy. Such
a change would run contrary to the concept of a Green Box as a desirable way of allowing governments to
respond to political realities and to develop rural policies that are minimally trade distorting. Any capping of
Green Box payments is likely to be seen as setting trade rules in contradiction to broadly desirable trends in
domestic agricultural policies over the past two decades.
The applicability of the Green Box to developing countries is an extension of the broader debate about a “de-
velopment box,” a safe haven for development policies that might otherwise seem to violate WTO rules. But
the G-20 paper restricts its suggestions to additions to the text of Annex 2 to make it clear that certain policy
instruments common in developing countries (and not markedly trade distorting) are included. As this would
not alter the reality of the current situation greatly, developed countries may agree to such changes.
In addition to suggestions by the G-20, other countries have come forward with specific proposals which focus
on the clarification and possible extension of the Green Box to encompass development programs, including
those related to land reform, to fit better with policies common in developing countries. This extension, often
referred to as a “development box,” could take account of the fact that developing country rural policy is usu-
ally designed to expand production. Rules that are premised on the notion that agricultural subsidies add to
surpluses and retain inefficient productive capacity may not be well tuned to the situation in all developing
countries. Such a shift in emphasis from developed to developing country policy circumstances may indeed
require some clarification, though most of what developing countries currently do in the way of general ser-
vice and structural programs is arguably already covered by some aspect of the Green Box criteria.37
Specific suggestions on the reform of the Green Box criteria fall under two headings. There are those that
seek to exempt developing countries from specific constraints on the eligibility of certain policy instruments.
One such exemption would allow the subsidy element in public stock purchases (the difference between the
acquisition price and the release price) to be excluded from the total AMS for developing countries. A similar
relaxation has been argued for domestic food aid. In effect, both of these changes would broaden the defini-
tion of the Green Box to include measures that are included in the Amber Box for developed countries.
Less drastic changes have been suggested for income insurance, to relax the criterion that only 70 percent
of losses be covered. Similarly, more flexibility could be given to developing countries in the area of disas-
ter relief, where current rules limit payments for crop losses. The criteria for regional assistance programs
could also be loosened for developing countries, in particular by relaxing the condition that regions be clearly
designated, contiguous areas with a definable economic and administrative identity. There may be little to
be gained by defining Green Box policies in such a way that developing countries have difficulty in meeting
criteria that were largely formulated with developed countries in mind. The July Framework (WTO 2004a) is
explicit on the issue of reviewing the criteria for policy payments that fall under the Green Box:
Green Box criteria will be reviewed and clarified with a view to ensuring that Green Box measures have no,
or at most minimal, trade-distorting effects or effects on production. Such a review and clarification will need
20
March 2007
to ensure that the basic concepts, principles and effectiveness of the Green Box remain and take due ac-
count of non-trade concerns. The improved obligations for monitoring and surveillance of all new disciplines
foreshadowed in paragraph 48 below will be particularly important with respect to the Green Box (para. 16).
The Hong Kong Ministerial confirmed the commitment to clarification and review and added that the Green
Box should adequately cover developing country policies that cause minimal trade distortions. Subsequently,
the Chairman of the Agricultural Committee restated the need to ensure that developing country policies not
causing significant trade distortions be included. Relatively little negotiating effort has been put into this as-
pect of the talks, but progress will be needed as a part of a broader package. The danger is that a hasty “last
minute” deal on Green Box definitions could undermine the clarity of the URAA agricultural subsidy classifica-
tion and lead to years of controversy on the consistency of the regulation of domestic support payments.
Monitoring Proposals and the Green Box
The URAA set up an Agriculture Committee (Article 17) to review the implementation of the commitments
(Article 18) “on the basis of notifications submitted by Members” as well as documentation prepared by the
Secretariat on request. In addition, “any new domestic support measure … for which exemption from reduc-
tion is claimed” has to be notified promptly. Countries can raise matters relating to the implementation of the
URAA and bring to the Committee’s attention any measure that they consider ought to have been notified
(Article 18:7). The Committee has met roughly four times a year since 1995 and discussed the notifications
submitted by WTO members.
In the area of domestic support, the significance of the Committee is that it provides a check against mislead-
ing reporting and surreptitious policy shifts to the detriment of other countries. The opportunity to question
the notifications (in other WTO Committees called “counter-notification”) has been used extensively, though
with little evidence of impact on policy choice. Most counter-notifications are requests for information and
pose little challenge to the ingenuity of the respondent in crafting a reply. Categorization has not led to ma-
jor disputes and no member has requested the establishment of a panel to resolve a “box” dispute.38 As a
mechanism for increasing transparency, the notification process was useful in the early years following the
conclusion of the URAA, but has been rendered much less useful by lags in reporting (which means that an
issue is mainly of historical interest by the time the Committee discusses it) and by the fact that as compre-
hensive negotiations became imminent, notifications became more sensitive.
The July Framework attempts to improve this situation. It includes the statement:
“Article 18 of the Agreement on Agriculture will be amended with a view to enhancing monitoring so as to ef-
fectively ensure full transparency, including through timely and complete notifications with respect to the com-
mitments in market access, domestic support and export competition. The particular concerns of developing
countries in this regard will be addressed.”
Few of the proposals in the current negotiations have addressed the issue of monitoring. An exception is the
G-20 (G-20 2005b). Their paper sets out concerns about the current system of notifications and suggests
improvements. Specifically, it suggests the creation of a sub-group of the Agriculture Committee, the Sub-
committee on Notification and Surveillance, which would significantly increase the timeliness and transpar-
ency of the notification process and would generate conclusions and recommendations on compliance with
the (modified) URAA. Though the G-20 paper emphasizes that monitoring should be improved in all areas,
including export competition and Special and Differential Treatment, the benefits of such improvements are
likely to be most apparent in the area of domestic support.
21
Improved timeliness and completeness of notification is a necessary, if not sufficient, step for improved
monitoring of domestic support. This is evidenced by the current confusion over the extent to which recent
reforms to the EU’s Common Agricultural Policy have changed the level of trade-distorting support, and
hence the impact of further limits on support levels. Article 18 of the URAA already stipulates that members
should notify promptly “any new domestic support measure, or modification to an existing measure, for which
exemption from reduction is claimed.” So merely adhering to current rules would resolve some of the prob-
lems. But adding some structure to the process could help. A subcommittee dedicated to the overview of the
monitoring process could be constructive, if that body were adequately supported by information collected
by the Secretariat. The process could be coordinated with that used for the Trade Policy Reviews. Advance
notification of significant policy changes should not be a problem, as these are typically the subject of lengthy
debate and legislative procedures.
The subcommittee could also conduct the evaluation of the compatibility of domestic measures with the
URAA. However, the limits to such evaluation should be more clear, since is not obvious that the Committee
on Agriculture, or any subcommittee thereof, could decide on the compatibility of any policy with other as-
pects of international law, such as the Agreement on Subsidies and Countervailing Measures (SCM). Legal
issues will continue to be the province of the dispute settlement process. But discussions in the Committee
on Agriculture could well influence the decision of countries on whether to pursue litigation.
In the remainder of the paper, we provide a review of developments in the use of the Green Box in selected
countries, and how recent or future changes in policies in the United State and the European Union relate to
Green Box issues discussed above.
22
March 2007
Green Box and U.S. Farm Policies

Green Box payments are an important component of the support provided to US agriculture (Figure 1). The
annual average of US$50 billion of support provided under this category was 75 percent of total support noti-
fied to the WTO for 1995-2001. Domestic food aid, in particular the Food Stamp program, accounted for 70
percent of total notified support under the Green Box.
The United States notified direct payments provided to producers under the 1996 Farm Act as Green Box.
These payments replaced the Blue Box deficiency payments of the former legislation, as reflected in the com-
position of support in Figure 1. However, the cotton case ruling discussed above could have a considerable
impact on the ability of the United States to adhere its commitments under the URAA, if its direct payments
as a whole are judged to be inconsistent with Green Box criteria.
The URAA limits the total AMS for the United States to a maximum of just over US$19.1 billion. If the direct
payments that have previously been included in the Green Box were required for inclusion in the calculation
of the total AMS, the United States could exceed this limit. The situation for 2000-06 is depicted in Figure
2, which assumes that direct payments are included in the non product-specific AMS category of support.
Following the procedure adopted by the United States in notifying Marketing Loss Assistance Payments pro-
vided prior to the 2002 Farm Act, the Counter-Cyclical Payments introduced under that Act are also included
in the non product-specific category. These assumptions result in that category of support exceeding the de
minimis threshold of 5 percent of the value of production in three of the seven years shown (2000, 2001 and
2006). These estimates are indicative of the importance of preserving Green Box eligibility for direct pay-
ments for the United States. That eligibility would become even more vital if the Doha Round of negotiations
ultimately results in a reduction of allowable trade distorting support (the sum of total AMS, de minimis and
Blue Box payments).
One obvious way in which the United States might overcome the problems posed by the cotton case ruling
for its direct payments would be to eliminate the restriction stipulating that fruit and vegetables cannot be
produced on land upon which the payments are based. Producers of fruit and vegetables in the United States
do not receive direct assistance though price supports and government payments, although some products
are protected from competing imports through tariffs. Producers of previously supported commodities could
decide to redirect land into fruit and vegetables if they could produce these profitably. As a result, there
could be some adjustments in prices and the location of production of certain crops.39 The elimination of the
fruit and vegetable restriction would not create a market distortion unless producers use direct payments to
cross-subsidize other agricultural activities. But in that case, the whole basis of the rationale that such direct
payments are minimally production and trade distorting would be in question!
The importance of the ability to provide payments under the Green Box would increase if there is a successful
outcome of the Doha Round. The United States, for example, has proposed a 60 percent reduction in its total
AMS limit (to US$7.64 billion). If that limit had been in place for the period shown in Figure 2 the United States
would have exceeded the limit in every year shown, with the possible exception of 2004 (depending on the
value of production criterion adopted for non product-specific de minimis). This indicates that some additional
changes will be required in future U.S. farm legislation if the United States is to stay within negotiated limits
for trade-distorting domestic support.
23
Perhaps in recognition of the potential challenges created by WTO issues, there is an increasing amount of
discussion about a reorientation of U.S. agricultural policies to permit a larger proportion of government sup-
port to be notified as Green Box. The four principal areas that are being discussed in this context are:
1. expenditures on services designed to increase competitiveness;

2. revenue insurance;
3. environmental programs; and
4. bio-energy programs.
Services to Increase Competitiveness

An increase in expenditures on basic and applied research leading to increased productivity is permissible
under the Green Box, even if research programs are oriented towards specific products. Investment in in-
frastructure is also permitted, provided that on-farm facilities are not subsidized. It would therefore be pos-
sible to subsidize a general upgrading of transportation facilities (e.g., locks on the Mississippi river) or other
types of infrastructure (e.g., broadband internet networks) provided that farmers’ marginal costs of using the
infrastructure were not subsidized through preferential user charges. As currently specified, expenditure on
extension and training activities could also be increased, even if these were primarily targeted to increasing
farm profitability, rather than productivity.
Income Insurance and Safety Net Programs

The term “safety net” can be used in a number of different contexts. Under the Green Box, the term is taken
to apply to programs that provide income insurance. The amount of compensation is limited and payments
cannot relate to the type and volume of production or to prices. The Green Box conditions relating to insur-
ance do not include the qualifier that this linkage should not solely apply to years after a base period (as is
the case for decoupled income support) – the lack of linkage appears to be an absolute requirement.
Under the current Federal program, approved insurance companies provide crop insurance to producers.
These policies provide coverage for a loss in yield or revenue for over 350 commodities (USDA 2006). In
some states, whole-farm gross revenue policies are also available. Current U.S. crop insurance programs
do not satisfy the conditions set out in Annex 2. Consequently, the United States has notified these as non
product-specific support and this has been included in total AMS calculations.
Proposals have been made to increase expenditures on revenue insurance as an alternative to current
price-based support (e.g., AFT 2006; NCGA 2006). These proposals include compensation for reductions in
revenue on an individual crop basis rather than a whole-farm basis. This is consistent with the way in which
other US payments (direct and counter-cyclical, for example) have been structured. However, revenue sta-
bilization schemes linked to individual crops appear to be inconsistent with the criteria set out in Annex 2.
Consequently payments under this approach would seem to be open to challenge if notified as Green Box.40
A switch from current price-based and other programs that fall in the Amber Box to alternative programs that
would also fall in the same category would seem feasible only if the new programs would lead to lower expen-
ditures. This would be particularly important if reductions in the total AMS binding are eventually agreed.
Environmental payments
US agricultural policy since the 1930s has featured environmental programs relating to soil conservation..
The 1985 Agriculture Act represented an extension of policy to broader environmental concerns, such as the
24
March 2007
protection of environmentally sensitive land and the preservation of wetlands. The trend was continued in
the 1996 Act by establishing several new programs, including the Environmental Quality Incentives Program
(EQIP). This program encourages farmers and ranchers to adopt farming practices that reduce environmen-
tal damage associated with production. The pressure to increase expenditure on conservation (agri-environ-
mental) programs is evident in the debate on the 2007 Farm Bill.
The Annex 2 criteria governing payments under environmental programs appear quite restrictive and pose
questions for the Green Box compatibility of several current U.S. agri-environmental programs. A list of the
major programs and a summary of the associated issues is given in Table 5.
In general, the likelihood of challenge to the Green Box compatibility of US agri-environmental programs
centers on the following issues:
1. The extent to which the requirement that payments be part of a clearly-defined environmental or
conservation program is met;
2. Whether or not payments include an incentive component above and beyond compensation for ad-
ditional costs incurred or income loss; and
3. The extent to which payments are linked to the production of marketable agricultural production.
If payments that are justified on environmental grounds are to increase significantly, the compatibility of pro-
grams with these conditions is likely to come under increasing scrutiny, with pressure to have these counted
as Amber Box.
25
Bio-energy Programs
Recent increases in the price of energy and concerns about dependence on imported energy supplies have
led to heightened interest in bio-energy programs. There may be pressure to increase payments to farmers
to produce crops for use as a feedstock for ethanol or to produce bio-diesel.
An incentive for the supply of ethanol is currently provided by a reduction of 51 cents per gallon in the Federal
excise tax compared to that imposed on gasoline. Although the exemption does not discriminate between
domestically produced and imported ethanol, there is also an import duty of 54 cents per gallon. As a result
of the two measures, the excise tax reduction is in essence a subsidy to domestically produced ethanol. As
noted earlier, a key issue in terms of the status of the domestic subsidy is whether ethanol is considered to be
an agricultural product, and hence whether the indirect subsidy provided to the agricultural feedstock used to
produce it is a subsidy for a marketable agricultural product. If this is indeed the case, then the subsidy might
arguably fall under the Amber Box.
But what if direct subsidies for the production of feedstock (corn or oilseeds) for bio-energy were introduced?
In that case, these subsidies would be almost certainly classified as Amber rather than Green. Subsidies that
are paid for the production of a marketable agricultural product are not allowable under the Green Box, re-
gardless of the ultimate use of that product.41 Other forms of aid, for example, investment aids for processing
plants that are specific to these agricultural products might also be subject to challenge if notified as Green
Box.
As in the case of environmental payments, an increase in the use of bio-energy subsidies that are closely
linked to production as part of future U.S. agricultural policies would not seem likely to qualify for inclusion in
the Green Box and would not contribute to a reduction in Amber Box support.
26
March 2007
Green Box and the CAP

Prior to recent changes in the Common Agricultural Policy (CAP) the Green Box was a less important com-
ponent of the support provided to EU agriculture than in the United States. The annual average of €20 billion
of support provided under this category represented 22 percent of the total support notified to the WTO for
1995-2001 (Figure 3). The three leading categories of support were general services, investment aids and
environmental payments—each of which averaged from €4.5 to €5.5 billion per year (Table 5). Regional as-
sistance was the other significant category, averaging roughly €2.5 billion annually.
Because of the lack of more recent notifications, the figures in Table 5 do not show the likely effects of the
significant changes in the CAP since 2003. The introduction of the single farm payment (SFP) scheme in
2005 would probably result in a substantial increase in the amount of decoupled income support that would
be notified to the WTO. It is difficult to estimate the exact impact, but a rough indication of an order of mag-
nitude can be derived from the notifications. Over the period 1995-2001, roughly 24 percent of the support
notified to the WTO was in the Blue Box. If roughly 80 percent of the support provided under the SFP would
qualify for the Green Box, then this would have roughly doubled this category of support over the notification
period to around 42 percent of the total.42
In the future, Green Box support may become even more significant for the EU as a result of the following:
1. Reform of the sugar regime in which a reduction in the support price for white sugar is being replaced
by a decoupled payment;
2. The potential adoption of direct payments for other commodities, particularly wine, fruit and veg-
etables; and
3. Possible replacement of remaining coupled components of direct payments made to commodities
covered by the 2003 reforms.43
The future status of payments that the EU may wish to modify as Green Box is open to question. The SFP
involves a number of restrictions on land use that could be interpreted as involving a linkage to current pro-
duction (European Communities 2003) and could open the SFP to a challenge on the basis of Green Box
compatibility. These are:
1. Land upon which a SFP is made may not be used for the production of fruit and vegetables (Article
51) – this parallels the condition applied to direct payments in the United States, which were ruled by the
WTO panel not to be in conformity with Green Box rules in the case of cotton.
2. Member states are required to ensure that areas under permanent pasture, upon which payments
are based, remain in that use (Article 5). To the extent that this increases the marketable production of live-
stock or livestock products it could be subject to challenge.
3. There is a requirement that land upon which payments are based should be kept in good agricultural
and environmental condition (Article 5). The general criteria (Annex IV) used to define these include several
elements relating to land management practices than could be interpreted as linked to current production.
The most obvious of these is a criterion for minimum stocking rates for livestock.
Also, producers receiving the SFP must respect a range of statutory management requirements (Article 3)
relating to environmental conditions; public, animal and plant health; and animal welfare. To the extent that
these are production linked, they are likely to constrain production, rather than increase it, and are therefore
27
unlikely to be challenged.
As in the United States, a lot of attention has been directed recently to increasing the supply of bio-energy
in the European Union. Under the 2003 changes to the CAP, €45 per hectare is offered to farmers who pro-
duce energy crops. A maximum of 1.5 million hectares of eligible area was originally specified, but there are
proposals to increase this to 2 million hectares. There is also a proposal to allow member states to provide
investment subsidies for up to 50 percent of the costs of the establishment of multi-annual energy crops.
Whether these would be included in the Amber Box would depend on whether the crops are considered ag-
ricultural or not.44
Although payments under bio-energy schemes are yet to be notified to the WTO, it seems likely these will
primarily fall under the Amber Box rather than the Green Box. As in the case of U.S. policies in this area, it
seems unlikely that production-linked aids for marketable agricultural crops used in the production of bio-en-
ergy will be eligible for inclusion in the Green Box.
In summary, it appears likely that the EU will continue to shift the structure of support towards decoupled
direct payments, and that these will be linked to some extent to production practices. The future Green Box
compatibility of these payments will be a major issue for the EU to satisfy more stringent constraints on Am-
ber Box payments.
Use of the Green Box by Other Countries

Though the EU and the US are the largest users of Green Box subsidies, they play an important role in the
farm policies of other countries. Among the other developed countries, domestic support in Japan is particu-
larly significant, both in magnitude and in its external visibility. Green Box payments form an important part of
total domestic support in Japan. From 1995-2002 these payments varied between 2.3 to 3.2 billion Yen (Table
7). The vast majority of these payments fall under the general services category. This has accounted for be-
tween 75-85 percent of total Green Box payments since 1995. The largest sub-category of expenditures is
on infrastructural services for the agricultural sector and rural areas, including the construction of irrigation/
drainage facilities and rural roads, and for land consolidation. This sub-category has typically accounted for
28
March 2007
at least half of total Green Box expenditures by Japan. Japanese government expenditures on infrastructure
are a large proportion of GDP (roughly 15 percent in recent years) and there is pressure to reduce these ex-
penditures due to persistent budget deficits. It is possible that this will be reflected in reductions in Green Box
expenditures, particularly for infrastructure projects for agriculture, in future notifications to the WTO.
The notification of domestic support in Japan changed significantly in 1998, when a modification to the rice
policy abolished the system of administered prices, though protection at the border was not reduced. As a
result, the notified total AMS fell sharply. Green Box expenditures accounted for roughly 45 percent of total
domestic support before the rice policy change. Since then, the Green Box percentage has risen to over 70
percent of the total (Figure 4).
The totals for Green Box payments in several other countries are given in Table 8.45 The table also shows
Green Box payments as a percentage of total domestic support (the sum of current total AMS, de minimis,
Blue Box, and Green Box payments) for the years for which data are available from the WTO notifications.
Green Box payments make up the bulk of domestic support in a few developed countries such as Australia
and New Zealand, and in many of the developing countries included in the table (such as Egypt and Indone-
sia, which did not report any non-Green support). This proportion acts as a measure of the degree to which
direct payments and income support have taken over from price support policies and input subsidies. Green
Box subsidies have increased in importance in several countries, most notably South Africa, India and Swit-
zerland, in recent years, due to changing policies.. But Norway did not significantly shift to Green Box policies
over this time period, and Brazil moved away from such policies in 1999.
An overall picture of the changing composition of support in WTO member countries from 1995-2001 is
shown in Figures 5 and 6. These charts depict aggregates based on all available WTO domestic support no-
tifications expressed in US dollars. Since not all WTO countries have notified their Green Box support for the
entire period, and the dollar tended to increase in value against many currencies over the period considered,
the data in Figure 5 may overstate an apparent reduction in total support. Despite this, Figure 6 suggests
that Green Box support has become increasingly important, rising from 44 percent of total support in 1995 to
51 percent by 2001. Although it is difficult to compute more recent aggregate figures due to the absence of
notifications in many countries, it is highly likely that the change in policy in the European Union alone as a
result of the 2003 reforms will have increased this proportion even further.
This tendency to switch to Green Box policies is clearly in the direction indicated by the incentive structure of
the URAA. Those that support the concept of the Green Box view it as an indicator of success in the reform
of domestic policies. However, it might also be considered a source of concern by those that view the Green
Box as simply a convenient means for sheltering subsidies from international disciplines.
Conclusion
The Green Box, as agreed to in the Uruguay Round Agreement on Agriculture, has both an economic and
a political rationale. As an economic concept, constraints on domestic support were designed to limit sup-
port provided through subsidies and deficiency payments on output, and subsidies on inputs. As a result, the
focus centered on creating a class of “decoupled” subsidies that did not directly stimulate output. In political
terms, the notion of the Green Box as a “safe haven” for subsidies decoupled from current price or output
was a necessary step in the process of introducing disciplines on domestic farm programs.46 It helped to
move countries along a path in line with domestic policy reform: better targeting of payments and a shift in
the focus of support from commodities to farmers. Though some countries are still skeptical of the extent and
consequences of this shift, most observers conclude that it is a move toward better policies.
29
The Green Box rests on the concept of decoupling; this is both its strength and its weakness. The notion
of decoupling has considerable merit in providing a framework for moving from price supports to direct
payments. The latter are likely to be less trade-distorting in that they are less output-enhancing and avoid
reducing consumption. But decoupled payments are not a perfect fit for all situations. As payments have be-
come less tied to output and price, competing demands for the redirection of subsidies have emerged. The
conceptual weakness of decoupling is that it does not address these alternative rationales. The granting of
30
March 2007
payments to compensate for previous price supports may require one set of criteria; payments that are tied
to environmental benefits may require another. In the latter case, the effectiveness of a subsidy may be more
important than its impact on production. What may be missing is the notion of “least trade-distorting” ways
to achieve particular objectives.47
As a consequence, the core of the debate over the Green Box should now shift from whether it has helped
with the reform of domestic policies (which it has) to another set of issues, including:
• Does the Green Box adequately address the provision of public goods? Though many such pay
ments are apparently covered under the current definition, there remains the problem of how to pay
for the production of public goods (environmental services) that may be associated with private
goods (farm products). Such subsidies are open to challenge as not conforming to Green Box crite
ria, even if they are the most effective way of achieving legitimate environmental objectives.
• Does the Green Box ignore the output effects of apparently decoupled farm income payments
through the impact on wealth, the removal of capital constraints, the expectation of future base acre
changes, and the removal of some price uncertainty? Should one restrict the method or size of pay
ments to farmers so as to avoid these output effects? Should one establish a separate category for
income payments and subject them to additional disciplines?
Both of these issues have relevance for future policy changes in the European Union and the United States. The
US is contemplating increases in environmental stewardship payments to give an incentive to farmers to follow
ecologically sound practices. Though most would consider such a move to be consistent with trade rules, cur-
rent Green Box criteria restrict payments to compensation for costs incurred or income foregone. From the point
of view of designing effective environmental programs, there is a serious weakness with such limitations. If the
maximum offered to producers is an amount equivalent to additional costs or income foregone, what incentive
(apart from altruism) is there for many producers to participate voluntarily in programs? Given the transaction
costs involved, it is unlikely that individuals who base their decisions solely on economic rationality would ever
choose to participate in a program if the additional returns from participation are negative.48 Some of the pay-
ments made under current environmental programs by the United States (and by other countries) or income sup-
port payments that involve environmental cross-compliance may not qualify for inclusion under the Green Box.
The EU is well on the way to delivering the bulk of its farm subsidies through a Single Farm Payment that
consolidates the benefits previously given by price supports. Once again, most would applaud a move away
from price supports. But it is not clear that the SFP is consistent with the Green Box criteria, as it still retains a
connection to production. Apart from the restriction on the use of land for fruits and vegetables, requirements
mandating that pastureland be kept in that activity, and that minimum stocking rates be maintained seem to
preserve the link between payments and production.
The Green Box needs to include direct and environmental payments to ensure a satisfactory alternative to
current programs. Environmental programs that are effective and minimally trade-distorting should be en-
couraged. Those payments that are introduced as part of a move away from price supports but are not made
for the provision of environmental or other services need to be monitored to ensure that the basic notion of
moving to less trade-distorting policies is not compromised.
31
The resolution of these Green Box issues will be extremely difficult to achieve for a swift conclusion of the
Doha Development Round. Steps should be taken in the short run to improve timely notification, monitoring
and surveillance of Green Box payments (in line with more timely notification of all elements of domestic sup-
port). Ad hoc modifications or extensions of the existing conditions made in haste, however, run the risk of
creating more problems than solutions. The only practical option is to postpone the full review and clarifica-
tion of Green Box criteria to a designated period following the conclusion of the Round.. Given that closure
is still needed on other critical issues, and that there is unlikely be sufficient time to deal appropriately with
Green Box issues, we urge for an agreed process of future review and clarification to address the interests
of both developed and developing countries.
32
March 2007
Figures
33
Figures
34
March 2007
Figures
35
References
Abler, D. and D. Blandford (2005). “A Review of Empirical Studies on the Acreage and/or Production Response to U.S.
Production Flexibility Contract Payments under the FAIR Act and Related Payments under Supplementary Legislation.”
OECD Papers Volume 5 (11). No. 421. Paris. Also available at http://www.oecd.org/dataoecd/15/15/34997377.pdf
American Farmland Trust (2006). “Integrated Farm Revenue Program: Enhancing the long-term viability and competitive-
ness of American agriculture.” Available at http://www.farmland.org/. Accessed November 2006.
Blandford, D., T. Josling and K. Arha (2006). “U.S. Environmental Programs and their Compatibility with Green Box Provi-
sions under the WTO Agreement on Agriculture” prepared for the National Forum on US Agricultural Policy and the 2007
Farm Bill: Conserving the Ecological Integrity and Economic Resilience of American Farmlands.
Brink, L. (2002). “Assessing the Proposals of the Cairns Group, Canada, and the United States for WTO Disciplines on
Domestic Support.” Paper presented at the IATRC Annual Meeting, Monterey, California, December 15-17.
Economic Research Service, USDA (2006). “Total support, United States, European Union, Japan, and all others, 1995-
2001.” http://www.ers.usda.gov/db/wto/ Accessed November 2006.
Chau, N.H. and H. de Gorter (2000), “Disentangling the Production and Export Consequences of Direct Farm Income
Payments.” Paper presented at the American Agricultural Economics Association Annual Meeting, Tampa, Florida.
European Communities (2003). Council Regulation (EC) No 1782/2003 establishing common rules for direct support
schemes under the common agricultural policy and establishing certain support schemes for farmers. Official Journal of
the European Communities, Brussels.
G-20 (2005a). Review and Clarification of Green Box Criteria, February 6. http://www.g-20.mre.gov.br/conteudo/19082005_
Breviario.pdf. Accessed November 2006.
G-20 (2005b) Improving Monitoring and Surveillance, October 19. http://www.g-20.mre.gov.br/conteudo/19082005_Bre-
viario.pdf. Accessed November 2006.
Hennessy, D.A. (1998). “The Production Effects of Agricultural Income Support Policies Under Uncertainty,” American
Journal of Agricultural Economics 80(1), pp 46-57.
Harris, D. and Rae A. (2006). “Agricultural Policy Reform and Industry Adjustment in Australia and New Zealand.” In D.
Blandford and B. Hill (eds). Policy Reform and Adjustment in the Agricultural Sectors of Developed Countries. CAB In-
ternational, Wallingford, Oxfordshire.
Howse, R., P. Van Bork, and C. Hebebrand,. (2006). “WTO Disciplines and Biofuels: Opportunities and Constraints in the
Creation of a Global Marketplace.” IPC Discussion Paper, Washington DC. Available at http://www.agritrade.org/Publica-
tions/wto_biofuels.html
Josling, T. (2000). “New Farm Programs in North America and their Treatment in the WTO” American Journal of Agri-
cultural Economics, 82(3), pp 775-777.
Josling, T., D. Roberts and D. Orden (2004). Food Regulation and Trade. Institute for International Economics, Washington
DC.
Josling, T. (2004) “Domestic Farm Policies and the WTO Negotiations on Domestic Support.” In G. Anania, M. E. Bohm-
an, C. A. Carter and A. F. McCalla (eds.) Agricultural Policy Reform and the WTO: Where are we Heading? Edward Elgar,
Cheltenham, UK and Northampton, USA.
National Corn Growers Association (2006). Testimony to the Agriculture Committee, U.S. House of Representatives, Sep-
tember 20. Available at http://www.ncga.com/. Accessed November 2006.
OECD (2000). “Decoupling: A Conceptual Overview,” OECD Agricultural and Trade Directorates, COM/AGR/APM/
TD/WP(2000)14/FINAL
OECD (2001). Multifunctionality: Toward an Analytical Framework. Paris.
Rude, J. (1999). “An Examination of Nearly Green Programs: Case Study for Canada” American Jounral of Agricultural
Economics, 82(3), pp 755-761.
36
March 2007
UNCTAD (2006). “Green Box Subsidies; A Theoretical and Empirical Assessment.” UNCTAD-India, September.
USDA (2006). “Risk Management.” 2007 Farm Bill Theme Papers. Available from http://www.usda.gov. Accessed Novem-
ber 2006.
WTO (1999). Canada – Measures Affecting the Importation of Milk and the Exportation of Dairy Products (Panel report,
as modified by the Appellate Body), WT/DS/103/AB/R.
WTO (2000). Member’s usage of domestic subsidy categories, export subsidies and export credits. G/AG/NG/S/12. 15
June.
WTO (2002). Domestic Support. TN/AG/S/4. 20 March.
WTO (2003). Negotiations on Agriculture: First Draft of Modalities for the Further Commitments (Revision) TN/AG/
W/1/Rev.1, 18 March.
WTO (2004a). Doha Work Programme: Decision Adopted by the General Council on 1 August 2004, WT/L/579, Geneva:
World Trade Organization (The July Framework Agreement).
WTO (2004b). United States – Subsidies on Upland Cotton, Report of the Panel, WT/DS267/R (Sept. 8).
WTO (2005a). United States—Subsidies on Upland Cotton, Reports of the Appellate Body, WT/DS265/AB/R, WT/
DS266/AB/R, & WT/DS267/AB/R (adopted March 21).
WTO (2005b). European Communities – Export Subsidies on Sugar, Report of the Appellate Body, WT/DS265/AB/R,
WT/DS266/AB/R & WT/DS283/AB/R.
Young, C.E. and P.C. Westcott (2000). “How Decoupled is U.S. Agricultural Support for Major Crops?” American Journal
of Agricultural Economics 82 (3), pp 762-767.
Figures
Source: WTO notifications.

Source: WTO notifications and personal communication from D. Sumner.
Source: WTO notifications.
Source: WTO notifications
Source: Economic Research Service, USDA (2006).
Economic Research Service, USDA (2006).
37
Endnotes
1 These include subsidies that are not considered to be specific (i.e., directed to an enterprise or industry, or to a group of
enterprises or industries) or are specific but relate to research and development, disadvantaged regions or environmental
requirements. In the latter case, the subsidies must also satisfy a set of specific criteria. The relevance of the SCM for the
Green Box is discussed in more detail later in the paper.
2 At its simplest, decoupling can be interpreted as the lack of a direct link between payments and current production deci-
sions. Thus a payment based on a historical volume of production is more decoupled from current production decisions
than one that varies with current output. Payments that appear to be decoupled may still have an impact on production as a
result of longer-term effects on the allocation of land, labor and capital (Abler and Blandford 2005).
3 One might, for instance, ask why, if farmers are paid to provide visual amenities through their stewardship of the land,
residents of attractive rural villages are not also eligible to receive such payments.
4 There can be a contradiction between the aim of removing the incentive to produce and the provision of environmental
public goods if there is jointness between agricultural production and the supply of such goods. The extent to which joint-
ness is likely to exist, particularly in the production of specific commodities, is the source of considerable debate (OECD
2001). Differences in opinion on the issue serve to strengthen the view that disciplines are needed on the form and amount
of environmental payments
if these are not to become an open-ended vehicle for potentially trade-distorting income support.
5 The adjustment payments made in Australia as part of the elimination of the support program for dairy products in 2001
could presumably be challenged on the same basis, since these were funded by imposing a levy on fluid milk consumption.
6 Annex 2 requires that any program not fitting in to any of the specific categories has to satisfy the criteria for decoupled
payments as well as the general criteria.
7 Annex 2 lists the categories of payments included in the Green Box and the specific criteria that apply to each of these.
The groupings used here are similar to those in Annex 2.
8 A recent study has attempted to quantify the output impact of Green Box measures considered as an aggregate (UNC-
TAD 2006). As might be expected, countries that spend more on research and infrastructure seem to have higher levels of
agricultural output. This appears to support the claim that such policies give the developed countries a competitive edge,
even if they are not specifically tied to production and price levels. However, it does not follow that research and infrastruc-
ture spending should be halte
d to prevent the expansion of output in developed countries: it could highlight the need for developing countries to focus
on the same areas. The study does not address the issue of box-shifting and, in particular, that previous policies in developed
countries could have been even more production-enhancing.
9 Advocates might argue that by contributing to the existence of economically viable farms, these types of activities make
a contribution to the supply of environmental and other public goods by agriculture. Also to the extent that the activities
are part of an agricultural adjustment program, e.g., are designed to facilitate the adaptation of labor to reduced domestic
support or freer trade, they might be considered to have a significant public good component.
10 If the domestic market is protected by high tariffs, any trade distortions relate to that instrument rather than to public
stockholding activities. With well-functioning markets and free trade, public stockholding can have only a limited effect on
domestic market prices.
11 This effect has been noted in connection with the restructuring of dairy operations following the reform of dairy policies
in Australia in 2001 (Harris and Rae 2006).
12 Ethanol is classified as a processed agricultural product, in the Harmonized System (HS) of tariff headings, though this
heading includes both fuel and non-fuel ethanol.
13 See Blandford, Josling and Arha (2006).
38
March 2007
14 It is worth emphasizing that the domestic support constraints apply almost entirely to the developed countries. Brink
(2002) reports that only 15 developing countries have registered total AMS commitments in the WTO ( two of these,
Mexico and Korea, are OECD members).
15 This is the formulation employed in the Technical Barriers to Trade Agreement, that explicitly allows the use of technical
standards and regulations in support of legitimate objectives (an open ended list is given) provided that a less trade-distort-
ing standard with the same effect is not available.
16 The case of animal welfare is more complex than suggested in the table. The appropriate policy also has to take into ac-
count the extent to which consumers in the importing country care about the treatment of animals in exporting countries.
If they do, a subsidy on domestic production to encourage particular farming practices may be inferior to an import restric-
tion, which in turn could be less desirable than a bilateral (or multilateral) agreement. But if enough consumers are willing
to pay for the attribu
tes of animal welfare then no subsidy will be necessary. See Josling, Roberts and Orden (2004) for a discussion of the regula-
tory dilemmas with respect to animal welfare in the presence of international trade.
17 This may be a result of the delay in notification, which makes retrospective challenges less plausible, or due to the fact
that countries have been engaged in negotiating a comprehensive trade agreement in the WTO.
18 In fact, in the absence of personal costs of benefits associated with an environmental issue or altruistic behavior it is
difficult to imagine that producers would participate in environmental programs without such an incentive component.
19 The use of an auction mechanism, similar to that employed in the Conservation Reserve Program of the United States,
can provide a way to establish the payment necessary to yield the desired supply of environmental services (i.e., their op-
portunity cost). The use of such mechanisms may make a payment less likely to be challenged.
20 If the payments were to be calculated as the amount needed to encourage adequate participation (e.g. by distributing
payment entitlements equal to the benefits from the system), could this still be interpreted as “the extra cost … involved in
complying with the government program”?
21 The precedent exists for a country to notify a part of the payments for a program as Green-Box compatible, leaving other
parts of the payment to be added to the Current Total AMS.
22 The argument of the panel was that cotton production could still be encouraged as a result of the restriction on possible
(more profitable) alternatives. Thus the subsidy was not fully decoupled from cotton production. Of course one could argue
that the restriction gave a benefit to fruit and vegetable growers, by reducing domestic competition. But as the restriction
acts to reduce the production of fruits and vegetables, it is unlikely that rival producers abroad would pursue such a chal-
lenge.
23 A panel could find that the policies in question did not fit under the definition of the Green Box (as happened in the
U.S. cotton case), and thus were incorrectly notified. But this would only be a signification violation if it caused a country to
exceed its scheduled subsidy commitments.
24 Agreement on Subsidies and Countervailing Measures, Article 1.
25 Id, Article 2.26
Id, Article 6:3.
27 It would appear to be economically irrational for producers to use a direct payment to cover current variable costs of
production, although some might do this to cope with short-run financial difficulties or on a longer-term basis if they view
farming to be a recreational activity. Chau and de Gorter (2000, p. 6) argue that a producer may decide to use a payment
to cover current production costs rather than exiting the industry, under the expectation that adjustments by others will
eventually cause market prices to rise. In any event, the use of payments to cover current production costs is likely to be
unsustainable over the longer term and will have only a short-run effect on production.
28 The prices of fixed assets in farming, particularly land, may change when farmers’ revenues change, as a result the rev-
enue-enhancing effect of payments may largely be reflected in changes in land rental rates and values, rather than changes
in production.
29 This is the “wealth” effect of support identified by Hennessy (1998).
39
30 The exit of existing farmers may result in the land being acquired by more efficient farmers. The amalgamation of land
parcels may permit economies of scale to be realized, leading to increased production efficiency and lower average produc-
tion costs. Larger scale farms may be in a better position to obtain financing from lenders to fund purchases of variable
inputs or to make output-enhancing investments, due to their higher income-earning capacity or greater equity to provide
security for loans. Any or all of these effects associated with the exit of farmers and resulting structural changes might lead
to an increase in production in the absence of payments.
31 See Young and Westcott (2000), Rude (2000) and Josling (2000) for further discussion of empirical evidence on the im-
pact of decoupling.
32 The agricultural negotiations themselves had started in the year 2000, but it was not until November 2001 that they be-
came part of a wider round, launched in Doha.
33As noted, economic analysis suggests that such payments can have output-increasing effects. However, the magnitude of
these effects is still debated and is likely to be specific to the circumstances of a particular program and crop.
34 The provisions for Sensitive Products and Special Products in the Market Access talks provide a way to address many of
the non-trade concerns in a less trade-friendly manner than through Green Box payments.
35 Ironically, the output effect of some subsidies designed to address non-trade concerns may be larger than those associ-
ated with wealth and other effects of decoupled policies. Policies that focus on public goods are likely to affect the produc-
tion of private goods.
36 There would also be a provision that precludes investment subsidies from being related to production in any year after
the fixed and unchanging base year.
37 Moreover, the chances of a challenge to the notification by a developing country of a subsidy as Green can be consid-
ered unlikely in most cases. Notifications by India, China and Brazil may be scrutinized on account of the size of potential
impacts from measures such as investment aids. If there is little slack in the AMS for such countries, a successful challenge
on Green Box status could require a change in policy.
38 This is in contrast to the constructive use of counter-notifications in the SPS and TBT Committees, where many issues
related to health and safety regulations, as well as technical standards, have been resolved . See Josling, Roberts and Orden
(2004) for a discussion of the work of these two Committees.
39 Green Box transitional assistance could be used to address any adjustment problems that arose from the change.
40 Other elements of the proposals may be inconsistent with Annex 2, for example, the definition of a revenue shortfall
that triggers payments and the amount of compensation provided.
41 Many agricultural crops, for example corn, are used in the production of a range of industrial products, but there has
been no attempt in the notifications to adjust estimates of agricultural subsidies for this.
42 This is probably a conservative estimate of the impact of CAP reform on the composition of EU support, since it will
also have affected the size of the AMS.
43 EU members were allowed to retain limited proportion of direct aids as coupled. There is no explicit requirement that
these be changed in the future. However, a further review of the CAP that will take place in 2008 may well result in pressures
to reduce the use of coupled payments.
44 As mentioned above, if the feedstock has no agricultural use, its subsidized production on retired land should not run
afoul of Green Box rules.
45 It should be noted that the numbers for developing countries do not include notifications of aid provided under the
Special and Differential Treatment provisions for development programs (Article 6.2 of the URAA).
46 One might expect that taxpayers would question the rationale for payments that do not require farmers to farm. But the
link to environmental stewardship has so far quieted this opposition to decoupling and the characterization of such pay-
ments as “welfare”.
47 Such a concept is central to the WTO Sanitary and Phytosanitary Measures Agreement and to that on Technical Barriers
to Trade.
40
March 2007
48 Transactions costs might be includable as a cost of compliance with environmental programs. However, a generous
interpretation of the components or magnitude of such costs would probably increase the likelihood that payments under
those environmental programs would be challenged.
41
42
March 2007
43
About IPC
The International Food & Agricultural Trade Policy Council (IPC) promotes a more open and
equitable global food system by pursuing pragmatic trade and development policies in food
and agriculture to meet the world’s growing needs. IPC convenes influential policymakers,
agribusiness executives, farm leaders, and academics from developed and developing coun-
tries to clarify complex issues, build consensus, and advocate policies to decision-makers.
www.agritrade.org
44

Green Box

Загружено:

Сведения о документе

Исходное описание:

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Green Box

Загружено:

Авторское право:

Доступные форматы

International

Food & Agricultural Trade

IPC Discussion Paper

Should the Green Box be Modified?

© 2007 International Food & Agricultural Trade Policy Council

Published by the International Food & Agricultural Trade Policy Council

Membership of the International Food & Agricultural Trade Policy Council

Piet Bukman (Chairman), The Netherlands Marcelo Regúnaga (Vice-Chairman), Argentina

Should the Green Box be Modified?*

Significance of the Green Box

Current Green Box Criteria

Expenditure on General Services

Financing for Public Stockholding for Domestic Food Security

Payments for Income Support, Income Insurance and

Structural Adjustment Payments

Payments Under Environmental Programs

Payments under Regional Adjustment Programs

Issues and Problems

Green Box and U.S. Farm Policies

1. expenditures on services designed to increase competitiveness;

Services to Increase Competitiveness

Income Insurance and Safety Net Programs

Green Box and the CAP

Use of the Green Box by Other Countries

Source: WTO notifications.

Вам также может понравиться