W__5 W   qro
POLICY RESEARCH WORKING PAPER   1950
Free Trade and Deep                                                 Arguments that common or
similar antitrust rules are
Integration                                                         essential for eliminating
antidumping do not have a
Antidumpingr and  Antitruststrong economic foundation
IkflLitU1dum          and  Antitrust                                 and are only weakly
in  Regional Agreements                                              supported by existing
regional agreements.
Bernard Hoekmnan
The World Bank
Development Research Group
International Trade                                                          H
July 1998



POLICY RI sI:.AxuCFi WOyRKL.(, PAi-ER 1950u
Summary findings
Preferential trading agreements (PTAs) are increasingly         In practice, it may be that the demnise of antiduiping
including elements of "deep" integration -  efforts to        in PTAs is constrained because governments are
agree on common regulatory regimes.                           concerned about the potential for their partners to
Hoekman explores what the PTA experience suggests           engage in beggar-thy-neighbor industrial policies. They
about the relationship between shallow integration -          may consider antidumping a useful defensive instrument
attaining unconditional intra-area free trade (including      in this connection, as it can substitute for instruments
the abolition of antidumping) -  and deeper integration,      such as countervailing duties, which have a much higher
especially agreement about common antitrust rules.            foreign policy content and may be more difficult to
He argues that common antitrust disciplines in PTAs         pursue.
tend to be driven by a broader agenda -  which revolves          If so, antidumping is a particularly ineffective and
around attaining economic integration (for example, by        costly instrument. Eliminating it in PTAs woould help
creating a single market), not by a need to abolish           focus attention on the real source of trade problems
antidumping. Many PTAs continue to apply antidumping          (industrial policies and government intervention) rather
to internal trade flows.                                      than on the symptoms (allegations of unfair dumping).
Earlier versions of this paper - a product of International Trade, Development Research - were presented at the Brookings
Conference on Private Practices and Trade Policy and at a CEPR/Institut d'Analisi Economica (CSIC) workshop on competition
and trade policy (Barcelona, November 1997). Copies of this paper are available free from the World Bank, 1818 H Street NW,
Washington, DC 20433. Please contact Lili Tabada, room MC3-333, telephone 202-473-6896, fax 202-522-1159, Internet
address Itabada@worldbank.org. The author may be contacted at bhoekman@worldbank.org. July 1998 (42 pages).
The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about
development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than folly polished. The
papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this
paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive IDirectors, or the
countries they represent.
Produced by the Policy Research Dissemination Center



Free Trade and Deep Integration:
Antidumping and Antitrust in Regional Agreements'
Bernard Hoekman
World Bank and CEPR
JEL classification: F 13
Keywords: trade policy, preferential trade agreements, economic integration
Correspondence: World Bank, 1818 H St. NW, Washington DC 20433; tel: 202 473 1185; fax:
202 676 9810; bhoekman@worldbank.org
* Earlier versions of this paper were presented at the Brookings conference on Private Practices
and Trade Policy and a CEPR/Institut d'Analisi Economica (CSIC) workshop on competition
and trade policy held in Barcelona in November 1997. I am grateful to Geza Feketekuty, Robert
Lawrence, Petros Mavroidis, Patrick Messerlin, Russell Pittman, Andre Sapir, Maurice Schiff,
Alasdair Smith, Alan Winters and participants in both meetings for helpful comments and
discussions, and to Chuck Stark for providing materials.



Summary
Arguments have been made that the WTO should be extended to include multilateral disciplines
on competition (antitrust) regimes. Rationales for this vary, but include a perception that private
anticompetitive practices may restrict market access opportunities and effectively nullify or
impair a country's trade liberalization commitments; assertions that national antitrust regimes
may impose negative spillovers on other countries; that effective antitrust enforcement against
finns with global market power requires a certain degree of harmonization of rules and
cooperation between national enforcement agencies; and that a WTO agreement in the area of
antitrust might prove to be an effective mechanism to constrain the use of contingent protection
(antidumping).
Contingent protection, in particular antidumping, has become a major bone of contention in
intemational trade relations. Perspectives on this issue vary widely, ranging from the perception
that antidumping is an appropriate response to an activity that is condemned by the GATT to the
view that antidumping is straightforward protectionism without any economic justification.
Somewhere in between are those who argue in favor of the use of antidumping on second best
grounds. For example, the need for antidumping may be motivated on the basis that foreign firms
are able to leverage entry restrictions in their home markets into an "unfair" competitive
advantage in export markets, and that absent common rules on antitrust disciplines, antidumping
is required to "level the playing field."
A working group has been established in the WTO with the mandate to explore the relationship
between trade and competition policies. This group can draw on a significant amount of
experience that has already been generated in this connection in the context of preferential trade
agreements (PTAs). These illustrate that governments can agree to abolish antidumping on intra-
area trade flows. They also reveal that governments have been able and willing to agree to
common competition disciplines. This paper explores what the PTA experience suggests
regarding the relationship between the attainment of intra-area free trade and the pursuit of
"deep" integration--actions by member states to effectively harmonize national regulatory
regimes. More specifically, the following two questions are asked: (i) does free trade require
deep integration?; and (ii) if so, is agreement on antitrust a necessary element of such deep
integration?
Many economists are of the view that insofar as dumping is potentially an economic problem for
an importing nation, it can and should be dealt with through enforcement of national competition
law. This can be done unilaterally--there is no need for international agreement or harmonization
of competition regimes. The implication is that if governments desire to form a free trade area
there is no need to negotiate common antitrust rules as a precondition for the abolition of
antidumping. Others argue that antidumping is a strategic tool that firms/governments use to
counteract the extraterritorial impact of differences in domestic policies and that consequently
that common competition disciplines are required in PTAs in order to abolish antidumping.



This paper argues that the economic and empirical basis for such "linkage" is questionable. The
regional experience illustrates that shallow integration--unconditional free trade--may be
associated with the pursuit of a deeper integration agenda. However, the linkages that are
sometimes argued to exist in practice between antidumping elimination and enforcement of
common antitrust disciplines tend to be driven by a broader agenda that revolves around the
attainment of economic integration (e.g., the creation of a "single" market). Although agreement
on common antitrust disciplines is not necessary to eliminate antidumping, common disciplines
in other regulatory areas affecting competition between firms in each jurisdiction may be needed
to completely remove the reach of contingent protection.
In practice, the demise of antidumping in PTAs may be constrained because governments are
concerned about the potential for their partners to engage in beggar-thy-neighbor industrial
policies. They may consider antidumping a useful defensive instrument in this connection,
because it can be used as a substitute for instruments like countervailing duties which have a
much higher "foreign policy content" and may be more difficult to pursue. If so, it must be
recognized that antidumping is a particularly ineffective and costly instrument. Indeed, a case can
be made that this is another reason to eliminate it in the PTA context as this will help focus
attention on the real source of problems (industrial policies; government intervention), rather
than on the symptoms (allegations of "unfair dumping").



I.   Introduction
Contingent protection, in particular antidumping, has become a major bone of contention in
international trade relations. Perspectives on this issue vary widely, ranging from the perception
that antidumping is an appropriate response to an activity that is condemned by the GATT to the
view that antidumping is straightforward protectionism without any economic justification.
Somewhere in between are those who argue in favor of the use of antidumping on second best
grounds. For example, the need for antidumping may be motivated on the basis that foreign firms
are able to leverage entry restrictions in their home markets into an "unfair" competitive
advantage in export markets, and that absent common rules on antitrust disciplines, antidumping
is required to "level the playing field."
A tentative start has been made in the WTO context to initiate discussions on the subject.
A working group has been established with the mandate to explore the relationship between trade
and competition policies. The work program for this group has been heavily contested, as was its
terms of initial reference. Strong opposition exists on the part of import competing industries in a
number of countries against any effort to discipline the reach of antidumping through the
introduction of linkages to antitrust principles.' Conversely, some competition authorities are
opposed to the introduction of antitrust into WTO because of a worry that antitrust enforcement
might become affected by trade policy considerations.
1/ This was illustrated in the Singapore Ministerial meeting of the WTO in December 1996
where the United States and the EU jointly agreed on a statement "clarifying" that the Ministerial
declaration establishing a working group on trade and competition policy "is specifically directed
at a work plan addressing antitrust issues and will not affect domestic antidumping standards and
provisions." Statement on Competition Policy, USTR press release 96-95, 13 December 1996.
Or, to cite the business view expressed in a 1996 ACTPN report on competition policy: "As long
as exporters may engage in dumping, there will be a need for national antidumping laws"
(http://www.ustr.gov/reports/actpn/policy.html). See Hoekman (1997) for a discussion and
references to the rapidly expanding literature on this topic.
1



This paper explores what can be learned from preferential trade agreements (PTAs)
regarding the conditions that are perceived as necessary by governments in order to move to a
free trade stance, including the removal of antidumping and other instruments of contingent
protection from their trade policy arsenal. A number of PTAs have illustrated that governments
can agree to abolish antidumping on intra-area trade flows. They also reveal that governments
have been able and willing to agree to common competition disciplines. My interest is to explore
what the PTA experience suggests regarding the relationship between the attainment of intra-area
free trade and the pursuit of "deep" integration--actions by member states to effectively
harmonize national regulatory regimes. More specifically, the following two questions are asked:
(i) does free trade require deep integration?; and (ii) if so, is agreement on antitrust a necessary
element of such deep integration?
It is well known that many PTAs have not eliminated the reach of trade policy--
governments often exempt certain sectors and retain the right to impose antidumping,
countervailing duty and "emergency" protection. The focus of attention in this paper is limited to
PTAs that have managed to abolish the scope to pursue such actions and thos_ where serious
attempts have been (are being) made to do so by one or more members. In reviewing the
approaches that have been taken in PTAs concerning the elimination of trade policy on intra-area
trade, particular attention will be given to one dimension of deeper integration: the extent of any
linkage to the adoption of antitrust disciplines. Many economists are of the view that insofar as
dumping is potentially an economic problem for an importing nation, it can and should be dealt
with through enforcement of national competition law (Finger, 1993; Hoekman and Mavroidis,
1996). This can be done unilaterally--there is no need for international agreement or
harmonization of competition regimes. The implication is that if governments desire to form a
free trade area there is no need to negotiate common antitrust rules as a precondition for the
2



abolition of antidumping. Others argue that antidumping is a strategic tool that
firms/governments use to counteract the extraterritorial impact of differences in domestic
policies and that consequently the "capacity of firms and governments to reach directly restrictive
business practices wherever they take place within the regional area is a pre-condition sine qua
non, before antidumping measures can be effectively phased out" (Marceau, 1995, pp. 35-36).
That is, it is claimed that common competition disciplines are required in PTAs in order to
abolish antidumping.
This paper argues that the economic and empirical basis for this line of reasoning is
questionable. Section II defines terms and discusses the economics of trade policy-antitrust
linkages. Section III reviews the evidence offered by PTAs that have been established since the
second World War involving countries that have tended to use contingent protection relatively
intensively--Australia, Canada, Western Europe, and the US--as well as several "new entrants"
such as Argentina, Brazil and Mexico. Section IV summarizes the regional experience and draws
some tentative lessons. Section V turns briefly to some implications of the regional experience
for the reform of multilateral disciplines on PTAs. Section VI concludes.
II.   Definitions and Conceptual Issues
As average tariffs of industrialized countries have fallen (to less than five percent currently) and
quantitative restrictions been abolished, international negotiations have increasingly centered on
domestic regulatory policies that are alleged to impede the ability of foreign firms and products
to contest a market. Examples of the latter include technical regulations that aim to safeguard
public health or the environment, and licensing or certification requirements for service
providers. Dealing with such "behind the border" policies has been termed "deeper integration"
3



in the literature,2 in contrast with the "shallow" integration that has been the staple of the General
Agreement on Tariffs and Trade (GATT), which revolves around reducing the prevalence of
measures applied at the border (tariffs, quotas). Rather than use the border-nonborder
distinction, in this paper shallow integration is defined to comprise actions to eliminate
discrimination between foreign and domestic firms-i.e., to apply the principle of national
treatment. This implies not only zero tariffs and quotas, but also the abolition of contingent
protection. Deep integration consists of explicit actions by governments to reduce the market
segmenting effect of differences in national regulatory policies that pertain to products,
production processes, producers and natural persons. In practice this will require decisions: (i)
that a partner's policies are equivalent (mutual recognition); or (ii) to adopt a common
regulatory stance in specific areas (harmonization). The latter approach may be complemented by
decisions to cede enforcement authority to a supra-national entity.
Many PTAs are only about a shallow type of shallow integration: all that is sought is to
eliminate measures that discriminate against foreigners by granting national treatment to
products originating in the preferred partner country. But shallow integration can also extend to
factors ofproduction such as labor and capital--in this case nondiscrimination is applied to
foreign labor and/or foreign investors. Allowing foreigners to operate on domestic markets does
not constitute deep integration according to the definition used in this paper because there is no
harmonization/recognition of the regulatory regimes that are applied to labor markets or to
investment in each of the participating countries. Governments remain sovereign and
unconstrained in this regard. Clearly the economic impact of shallow integration may far exceed
that of deep integration. Shallow does not mean that economic effects are limited.
2/ The term is due to Robert Lawrence, see e.g., Lawrence (1996).
4



Competition disciplines are one of the policy areas that may figure on a deep integration
agenda. It is important in this connection to distinguish competition policies in general from
antitrust or competition law. Antitrust involves instruments that control or regulate the
permissible behavior of private firms or natural persons. They are a subset of competitionpolicy.
Generally antitrust laws prohibit anticompetitive practices such as price fixing, collusion between
firms to restrict output, or the abuse of a dominant position. Competition policy spans the
broader set of measures and instruments that may be pursued by governments to enhance the
contestability of markets. This goes beyond antitrust to include efforts to privatize state-owned
enterprises, deregulate, reduce firm-specific subsidy programs, liberalize trade and investment,
etc. Elements of competition policy broadly defined have been on the agenda of the GATT for
many decades.
There are two possible motivations for deeper integration efforts, and more specifically,
why antitrust might figure on the agenda of a trade agreement. The first is that there are gains
from cooperation in this particular area and an international trade agreement allows these to be
attained. The second is that common disciplines on competition policy are necessary to achieve
the objective of regional free trade (shallow integration). In the first case, a distinction must be
made between situations where national competition policies result in outcomes that are sub-
optimal (inefficient) from a joint welfare perspective and those where they do not. In
Prisoners Dilemma-type situations there is a prima facie case for cooperation, but it is not
clear why a preferential trade agreement is necessary as a vehicle to capture the gains from
cooperation on competition policy. If such gains exist, in principle governments can pursue
issue-specific cooperation that results in outcomes that raise welfare in both countries. In
practice, of course, that is what has often been done, as reflected in bilateral cooperation
5



agreements between antitrust authorities.3
The case for inclusion of competition policy in a PTA becomes more compelling if the
situation ex ante is akin to a zero sum game: efficient national enforcement of competition
policies reduces the welfare of a trading partner or there are costs involved for a country to
move its competition regime to conform more closely to that of a partner.4 Here the issue
becomes a distributional one, as moving away from the status quo ante may impose a loss on a
country. Alternatively, there may be scope for increasing joint welfare through coordination
of antitrust, but interest groups that would lose from a move to a more efficient (joint)
outcome may be able to block this (insist on compensation). In such cases, side-payments are
required to achieve agreement to adopt common competition regimes. Such side-payments
may be easier to realize in the context of a broader trade agreement that has a large number of
issues on the negotiating agenda. Indeed, they may be a necessary condition for deeper
integration.
A second possible rationale for the introduction of antitrust on the agenda of a PTA is
that governments regard common competition disciplines as a necessary condition for
achieving shallow integration (the elimination of trade policy). Most PTAs have tended to be
imperfect in the sense that many holes and loopholes are maintained that imply that intra-
regional free trade is not achieved for a long period of time, if ever. Whole sectors may be
3/ Although this is not to say that existing arrangements are adequate. Many observers have
concluded that such bilateral cooperation between competition authorities could benefit from
substantial strengthening and expansion. See e.g., Commission of the European Communities
(1995).
4/ This is not unlikely to be a frequent occurrence in practice, given the significant differences in
the substantive provisions and interpretations of competition law across even OECD countries
and he prevalence of exemptions for export cartels. See, e.g., Boner and Krueger (1991) for a
review of national rules and provisions.
6



excluded from the ambit of liberalization and a variety of instruments of contingent protection
may continue to be applicable to intra-area trade. Full achievement of shallow integration
therefore has been more the exception than the rule. This raises the question whether shallow
integration requires introduction of deeper integration elements into the process, and, in
particular, common disciplines in the area of antitrust.'
Logically, there at least four "types" of PTAs: (i) those that achieve unconditional internal
free trade (shallow integration) without any deep integration; (ii) shallow integration is
accompanied with deep integration; (iii) shallow integration is not achieved (or sought) but the
PTA embodies elements of deeper integration; and (iv) neither shallow nor deep integration is
attained or pursued. Observations of the first type of PTA are sufficient to reject the view that
agreement on antitrust is necessary for the abolition of antidumping. More difficult is to
determine what motivates deeper integration in instances where this is observed in PTAs that
realize shallow integration. There are three possibilities. First, deep integration may be
considered necessary by governments in order to give up trade policy instruments. Second, it
may be the case that no link is between trade and antitrust is perceived to be necessary and that
agreement on common antitrust disciplines simply reflects the existence of gains from
cooperation in this area. This might just as well have been pursued independently of the PTA
5/ Strong statements can be found in the legal literature implying that the answer is positive. For
example: "..antidumping measures can be phased out between two countries only if and when:
standards for transnational restrictive business practices (RBPs) exist; private firms have legal
standing before the courts of the territory where the RBP takes place to enforce these standards;
and/or when this issue is dealt with before domestic courts (other than that of the territory where
the RBP takes place), the market of reference is transnational and the domestic courts are entitled
to address transnational issues" (Marceau, 1995, p. 53).
7



and in the absence of the PTA might have been.6 Finally, it may be that shallow integration
does not require deeper integration, but cooperation on antitrust requires side payments. These
may be achieved by linking cooperation on antitrust to the broader agenda of the PTA.
Distinguishing between these three situations in cases where PTAs embody both shallow
and deep integration is important in understanding the role of antitrust disciplines in trade
agreements. The more one observes instances of the first situation the greater the likelihood that
agreement on antitrust will be necessary in the WTO if the multilateral trading system is to move
further down the path towards free trade. Unfortunately, characterizing real world PTAs on the
basis of this three-fold typology may not be straightforward. Recourse will have to be made to
statements by the governments involved in such agreements, as well as the specific language and
drafting/negotiating history of each agreement.
Why Link Abolition ofAntidumping to Common Antitrust Rules?
Before reviewing the regional experience it is helpful to discuss briefly the most interesting type
of PTA: those where deep integration is held to be a necessary condition for shallow integration.
Specifically, why might governments be unwilling to abolish antidumping unless an agreement
on common antitrust standards can be negotiated? Possible rationales for antidumping are to
protect domestic agents from predatory (monopolizing) dumping (Viner, 1923) or as an
instrument to counteract foreign strategic trade policy. A linkage argument must then
demonstrate that absent common antitrust rules, a nation becomes open to predation or foreign
rent-shifting policies.
6/ This is the case among NAFTA members, for example, where common antitrust rules were
not adopted but where national authorities have concluded a variety of bilateral cooperation
agreements. These are pursued independently of the FTA (see below).
8



Predation is the standard economic potential rationale for antidumping. Predatory
dumping arises if a foreign firm (or cartel) seeks to force domestic competitors to exit the market
by pricing below cost (with the intention of raising prices once the competition has exited). It has
been pointed out frequently that if there is predation, national antitrust should in principle be able
to address the problem and that there is no need for common antitrust disciplines. More
important, predation is unlikely to be a strategy that is often used by firms. In practice, research
suggests it is very much the exception, not the rule in antidumping cases.7 Indeed, in real world
antidumping actions an antitrust authority would not have intervened on competition, let alone
predation, grounds (Messerlin, 1996; Sch6ne, 1996). The consensus among most observers is
that antidumping as it is practiced today in the major jurisdictions has nothing to do with
predation.
This leaves the second possible rationale for antidumping: a strategic policy instrument
that aims to affect a country's terms of trade. Assuming a PTA aims to achieving free trade
(eliminate trade policy), governments may be concerned to constrain the ability of other
members to use alternative instruments (such as antitrust or subsidies) to circumvent the free
trade goal. Absent such agreements, contingent protection may be seen an insurance mechanism
to offset the strategic use by other member governments of policies that could reduce domestic
welfare. This line of argument puts foreign government policy in the center of attention. The
issue then is to determine if targeting individual firms (as antidumping does) is an efficient way
of dealing with such foreign government policies. This depends in part on the source of the
"problem". One possibility is that foreign firms benefit from high trade barriers that protect their
7/ There are many reasons for this, including the difficulty of restricting entry in the future and
the likelihood that the importing country government would regulate the resulting monopoly (or
cartel).
9



home market, and that antidumping is seen as an instrument with which to put pressure on the
foreign government to reduce such barriers (Stewart, 1996). However, given that in the PTA
context presumably the objective is to eliminate trade barriers, this line of reasoning implies that
antidumping should be removed once "classic" trade barriers such as tariffs and quotas have been
eliminated. No link with antitrust is required.8
Another possible "problem policy" is foreign antitrust. This may have a negative effect
on domestic welfare. For example, a merger between foreign finns that is allowed by the foreign
antitrust authorities may result in an increase in export prices. The larger the share of the foreign
entity in domestic consumption the more likely the resulting decline in consumer surplus will
exceed the increase in producer surplus of import-competing firms (Ordover and Willig, 1984;
Dixit, 1984, 1998; Barros and Cabral, 1994). This is a variant of the export cartel issue-an
antitrust authority of an industry that mostly exports (or where behavior on the home market can
be regulated) has an incentive to allow the formation of export cartels as it improves the terms of
trade. The relevant question in this context is what antidumping can do to counteract the effect of
a "nationalistic" foreign antitrust policy. The answer is: very little. Whether trade policy in
general (tariffs) may offset (some of) the welfare reducing effect of the increase in foreign market
power depends on a number of factors, including market structure and demand. Collie (1997)
concludes that in this type of situation tariffs may have a positive or negative effect and are
inferior to domestic production subsidies. In any event, antidumping is not very relevant in this
context as the effect of greater concentration (market power) is to increase export prices, thus
reducing the probability of dumping. One can also question whether (the threat of) an
8/ As noted below, in a number of PTAs where the elimination of antidumping has been
achieved or is an objective, abolition was (is) linked to the transition period for phasing out
tariffs (e.g., the EC and free trade agreements between Canada and Chile, and between Australia
and New Zealand).
10



antidumping duty can put pressure on affected firms in the export market to lobby their
government to pursue a more vigorous antitrust stance-or to consider amending its practices to
conform more closely with those demanded by its trading partner. This is because current
antidumping enforcement takes no account of whether price discrimination or selling below cost
is a function of a "strategic" antitrust stance by the home government of an exporter.
It is sometimes argued that as countries move to eliminate trade policies, they may
perceive that possible negative spillovers due to differences in national antitrust regimes will
increase after the achievement of free trade. That is, the incentive to use antitrust strategically
may increase: once free trade is achieved, government's may shift to alternative instruments to
shift foreign profits or surplus to domestic agents. However, analysts that have explored this
issue question whether the abolition of trade policy will give rise to greater incentives to use
antitrust strategically, even in areas where it would seem most easily applied--e.g., merger
policies and export cartels (Horn and Levinsohn, 1996; Motta and Onida, 1996).
A final potential "strategic" rationale for antidumping is as an instrument to counteract
foreign industrial policies such as investrnent restrictions, subsidies and state involvement in
industry. Such government support may give rise to significant negative repercussions on
domestic firms. Consequently, the less PTAs do to control industrial policies of governments, the
greater the resistance may be to giving up "commercial defense" instruments. There are valid
theoretical economic justifications for the use of countervailing duties. Indeed, partial
countervailing of a foreign export subsidy can be the optimal response by an importing country
(Dixit, 1988; Collie, 1991). However, in general contingent protection is an inefficient
instrument to deal with the effects of foreign subsidies or industrial policies as it imposes
additional deadweight costs on domestic consumers without greatly increasing the incentives of
the foreign government to change its ways (Deardorff and Stem, 1987). The appropriate policy is
11



to draft rules that restrict the ability of a PTA member to use industrial policies in ways that are
detrimental to the welfare of other member countries. In practice, this may be difficult to achieve,
as is illustrated in many PTAs. If so, governments may seek to continue to pursue subsidy and/or
countervail policies in part motivated by strategic considerations. But antidumping should not
have a role to play in this connection as it does not focus on the existence of detrimental foreign
export subsidies as a precondition for action. Of course, it may be a substitute for countervailing
duties for exactly that reason: there is no need to confront foreign governments with allegations
that their subsidy policies are detrimental to domestic industries. Indeed, it may be defended as
allowing non-subsidy-based industrial policies to be targeted as well. But that is no virtue.
Instead, this illustrates one of the major problems with antidumping from an economic policy
perspective-it allows proponents to hide behind the cover of vague allegations of "unfair" trade
practices without specifying what these are.
There has been a lot of research on the incentive effects of antidumping.9 This has
concluded that antidumping can have a pro-competitive impact in situations of oligopolistic
rivalry. However, once rent-seeking incentives are recognized, account is taken of the fact that
antidumping as actually applied does not consider the variables that matter from a strategic point
of view, and that the possible "spillover" effects of antidumping on downstream industries can be
significant (Hoekman and Leidy, 1992), in practice it is much more likely that antidumping has a
detrimental effect on domestic welfare. Moreover, given the technical and "automatic' nature of
antidumping and the fact that it is driven by private petitions, it is difficult to conceive of
antidumping being used as an effective strategic instrument by governments.
Arguments that a necessary condition for the elimination of antidumping in the PTA-
2/ See, e.g., Dixit (1988), Leidy and Hoekman (1990), Messerlin (1990), Reitzes (1993), Staiger
and Wolak (1989; 1992), Veugelers and Vandenbussche (1997).
12



context (or in the WTO context for that matter) is the adoption of common competition rules are
therefore not compelling. Insofar as the fear is predation, this can be achieved by the application
of domestic antitrust. If the fear is foreign "strategic" policy, be it antitrust-related or more
general industrial policies, antidumping can do little to address the problem. Antidumping as
currently practiced is predominantly simple protectionism (Finger, 1993). There is therefore
nothing special about antidumping that requires linkages to other policy areas-as just another
form of trade policy it should be eliminated in conjunction with tariffs and quotas more generally
as a PTA is implemented.
III.  Regional Trade Agreements
Many preferential trade agreements have been negotiated over the years. Competition policy in
general and antitrust in particular has played an important role in only few of these agreements.
However, a number of the more recent agreements have pursued common antitrust disciplines.
What follows looks at a number of PTAs that have gone far towards achieving either full free
trade (shallow integration) or have sought to incorporate disciplines on antitrust. As noted
earlier, the main focus of interest is to determine whether cooperation extends to the area of
competition law and if so, what can be said about whether this appears to be a necessary
condition for achieving free trade. Given the voluminous literature that exists on most of the
PTAs considered here, the discussion will be succinct.'0
European Economic Community (EEC)
The EEC (now the European Union) is unique in that it involves the complete liberalization of
JQ/ This section draws in part on Hoekman and Mavroidis (1996) and on Hoekman (1998).
13



trade in goods, services, labor and capital. It is also unique in the extent and reach of the
constraints that are imposed on member states regarding regulatory policies that may have an
impact on trade. Thus, common disciplines are imposed on Member states with respect to state
aids (subsidies), monopolies, government procurement practices, and competition policy, and
these are complemented by detailed directives and regulations aimed at the achievement of the
Internal Market. These disciplines are enforced by supranational bodies (the Commission and
the European Court of Justice) as well as by national institutions.
Articles 85 and 86 of the European Community Treaty (ECT) prohibit practices that
restrict or distort competition and abuses of dominant positions insofar as they affect intra-
European trade flows. Art.85 deals with agreements and concerted practices and Article 86
prohibits abuse of a dominant position.'" Public undertakings and entities granted special or
exclusive rights are subject to the competition rules as long as this does not impede the
realization of their assigned tasks (Article 90).12 State-aids affecting trade flows are prohibited
(Art. 92), although generally available subsidies are permitted in principle, as is aid targeted at
disadvantaged regions (Article 92.3a).
In short, common disciplines are imposed on Member states with respect to state aids
(subsidies), monopolies, government procurement practices, and antitrust. These are
complemented by detailed European legislation relating to the achievement of the Internal
Market. These disciplines have direct effect (supercede national law) and are enforced by
11/ Dominance is determined on the basis of the relevant product and geographic markets; abuse
may involve unfair trading, price discrimination, tie-ins, bundling, or restricting output or access
to markets.
12/ State monopolies of a commercial character must also ensure nondiscrimination regarding
the conditions under which goods are procured and marketed between EC nationals (Article 37
ECT).
14



supranational bodies (the Commission and the European Court of Justice) as well as by national
courts. The various competition provisions were considered necessary in order to achieve the
objective of creating an integrated European market. As has been emphasized many times by the
European institutions and numerous commentators, the goal is to ensure a "level playing field" or
"equal competitive opportunities" on European markets for suppliers originating in member
states.'3 This "trade effects" focus has implied that in the enforcement of EC antitrust rules (Arts.
85-86) the Commission and Court may pursue a more "interventionist" policy stance than a
national antitrust authority. The latter will generally be concerned with national welfare, and will
balance the various economic effects of potential anticompetitive measures. In the EC context,
the application of Arts. 85 and 86 is concerned as well with whether specific actions by entities
are consistent with the objective of market integration (i.e., do not restrain intra-EC trade)."4
A noteworthy feature of the EC regime is that no efforts were made to harmonize national
antitrust regimes, which differed very substantially across countries. Indeed, some members
states did not even have antitrust legislation; Italy adopted a comprehensive antitrust statute only
in 1990 (Siragusa and Scassellati-Sforzolini, 1992). Particularly for smaller states this implied
that EC rules had little impact, as the affected markets could easily be too small to satisfy the
"trade effect test". In such cases effective enforcement requires that national authorities take the
lead. If for whatever reason this does not occur, EC competition rules may be largely
13/ For more comprehensive discussions of the EC, see e.g., Bourgeois (1992), Brittan (1990)
and Ehlermann (1992).
14/ This may lead to findings by the EC institutions that are inappropriate from a "pure"
economic efficiency viewpoint. Treatment of vertical agreements provides an example.
Exclusive distribution agreements between a producer in one country and a distributor in another
may fall foul of EC law even if vigorous interbrand competition (including from imports) exists
on all EC markets because the agreement implies market partitioning. See Weatherill and
Beaumont (1993, Chapter 22) for an extensive discussion.
15



irrelevant."5 Over time greater attention has come to be centered on the adoption of more
effective national antitrust legislation in member states. For example, in 1983 Portugal drafted
and adopted its antitrust legislation in part with a view to its future accession to the EC (Barros
and Mata, 1996), and a number of member states have amended their statutes over time to
conform more closely to the letter and spirit of EC rules. To some extent this has been driven by
the EC Merger Regulation, which requires that national bodies exist to review mergers or
acquisitions if these are referred to them by the European Commission, and by a more general
concern to implement the principle of subsidiarity (in the process reducing somewhat the burden
of enforcement that is carried by the Commission and the Court)."6
As mentioned, the primary focus of the European Commission and the Court of Justice
has been on contesting measures (public and private) that segment national markets. Facilitation
of arbitrage has therefore been an important objective. This is perhaps best illustrated in the
extensive case law pertaining to Article 30 (prohibiting the use of quantitative restrictions and
measures with equivalent effect) and the Cassis de Dijon decision which led to adoption of the
principle of mutual recognition for standards (Weatherill and Beaumont, 1993, Chapter 15). But
it is also reflected in the vigorous enforcement of competition provisions on the behavior of
private agents and States that prevent arbitrage across markets. An example is the ban on
restrictions on parallel imports-preventing distributors of a product from selling to entities that
have the intention of re-selling the products in other member states.
The Treaty of Rome prohibits antidumping on intra-area trade once the transition period
for full implementation of the treaty provisions has expired. However, during the transition
15/ For an example, see Fingleton (1996), who argues that this was the case in Ireland until
passage of new legislation in 1991.
16/ See Neven, Nuttall and Seabright (1993) for an analysis of the EC's Merger Regulation.
16



period, members were permitted to petition the European Commission for authorization to
impose antidumping (Art. 91).'7 No formal linkage was established between the application of
the common competition rules (or national antitrust) and the elimination of antidumping; instead
the presumption appears to have been that once tariffs and quotas had been removed,
antidumping should go as well. It might be argued that there is an informal linkage, as Art. 91 is
in the competition-related section of the ECT, preceded by provisions regarding anticompetitive
practices by undertakings and followed by a provision dealing with state aids. However, this
cannot be seen as evidence of a strong antidumping-antitrust link as the criteria of Articles 85-86
are very different from those that are applied in antidumping. Consequently, the argument that
the EC is a prime example of a PTA where antidumnping was removed because of a perception
that the common antitrust disciplines could deal with the "problem" of dumping is not very
compelling. Instead, the view was taken that antidumping could have no place in a common
market.
The European Economic Area (EEA)
The EEA was negotiated between the European Free Trade Association (EFTA) states (minus
Switzerland) and the EU in 1992. It involves the former adopting much of the EC's acquis
communautaire, including not only the EC's competition rules for firms and governments but
also some 1,600 pieces of EC legislation (Norberg, 1992).18 The EEA built upon free trade
17/ During the transition period, Article 91:2 required that products originating in one Member
state and exported to another be free of duties, quotas and measures with similar effect if they are
re-imported. Thus, during the transitional period governments were required to ensure that
arbitrage between markets was not impeded by maximizing the incentive for consumers of the
exporting country to re-import dumped goods. Transition periods varied across new members.
18/ The main exception is that the EU's Common Policies (such as agriculture and external
trade) are not adopted by the EFTA states.
17



agreements negotiated between the EC and the EFTA states in the early 1 970s. These contained
disciplines on the application of competition policies, notwithstanding which the EC continues to
extend the reach of anti-dumping to EFTA-based exporters. The abolition of antidumping duties
in the EEA was one of the negotiating goals of the former EFTA countries, as their exporters
sought to eliminate the threat of such actions completely (Hindley and Messerlin, 1993). This
was achieved: Article 26 EEA states: "Anti-dumping measures, countervailing duties and
measures against illicit commercial practices attributable to third countries shall not be applied in
relations between the Contracting Parties, unless otherwise specified in this agreement."19 Given
adoption of the acquis and the effective extension of the EU Single Market to EEA members,
clearly there was no option for the EU to retain antidumping. As in the EC proper, the driving
force behind abolition of antidumping was the creation of a single market. Deep integration
played an important part in this process, as reflected in the numerous directives and regulations
that were adopted. But antitrust regimes per se did not.
Enforcement of the EC competition disciplines is similar as in the EC proper, through a
mix of direct effect (allowing the rules to be invoked by firms before the courts) and supra-
national enforcement. However, the latter task is shared between two institutions: the European
Commission and the EFTA Surveillance Body (the EFTA analogue of the Commission). It was
agreed that the latter has jurisdiction if at least one third of the turnover of the firm(s) subject to
complaint is in EFTA states.
The EEA is a free trade area, not a customs union, suggesting that whatever other
necessary conditions must be satisfied for elimination of antidumping, a common trade policy
19/ The exceptions are listed in a protocol which states that it only applies to the areas covered
by the provisions of the EEA in which the Community acquis is fully integrated (thus excluding
agriculture); and that it is without prejudice to any measures which may be introduced to avoid
circumvention of anti-dumping or similar measures by third countries.
18



vis-a-vis non-members is not required. Indeed, the EU illustrates that antidumping can co-exist
with a customs union. Thus, in the customs union with Turkey, European firms may continue to
petition for antidumping actions to be imposed on Turkish exports (see Togan, 1997).
EUAgreements with Eastern European and Mediterranean Countries
Starting in the early 1 990s, the EU initiated a process of negotiating Association agreements with
neighboring countries.20 These commit signatories to achieve free trade on a reciprocal basis,
usually over a transition period ranging from 10 to 12 years. As in the EEA, in the so-called
Europe Agreements (EAs), Central and East European (CEE) countries committed themselves to
adopt the EU's rules relating to agreements between firms restricting competition, abuse of
dominant position, the behavior of public undertakings (state-owned firms) and competition-
distorting state aids that have an effect on trade (Articles 85, 86, 90 and 92 ECT).2'
Despite the agreement to adopt EU-compatible competition disciplines there is no
analogue to Art. 91 ECT in the Europe Agreements specifying that antidumping will be phased
out or eliminated. EU firms will therefore be permitted to engage in price discrimination or sell
below cost on the EU market, whereas CEE firms will be constrained in pursuing such a strategy
by the existence of EU antidumping procedures. On EU markets, price discrimination by CEE
firms in the sense of selling products at prices below those charged at home may lead to
antidumping petitions if this injures EU firms (Hoekman and Mavroidis, 1995).
20/ Formally, the EEA is also an Association Agreement under Art. 238 ECT.
21/ State-aid, compatible with EU rules for disadvantaged regions (Article 92.3a ECT), may be
applied to the entire territories of the associated states during the first five years, and transition
periods of 3 and 5 years were granted for abiding with Arts. 90 and 37 disciplines. Cooperation
between the EU and CEE antitrust authorities is based on the principle of positive comity (this
grants a state whose interests are affected the legal option of requesting another state to initiate
appropriate enforcement proceedings in its market).
19



This situation has been justified by the Commission on the basis that antidumping and
similar instruments must remain applicable to trade flows as long as it is still unclear to what
extent partner countries have completed the transition to a market economy. Subsequent to the
negotiation of the Europe Agreements the EU clarified its stance on this issue. At the December
1994 Essen Summit, the European Council declared that the Union "should be ready to consider
refraining from using commercial defense instruments for industrial products" conditional upon
the "satisfactory implementation of competition policy and control of state aids ... together with
the wider application of other parts Community law linked to the internal market, providing a
guarantee against unfair competition comparable to that existing inside the internal market."
Clearly there is no firm commitment here to eliminate antidumping. Note that the EU insists that
application of competition laws and principles are not enough; what is necessary (but not
sufficient!) is that all of the Single Market directives are applied as well.22
The situation is similar in the Euro-Mediterranean Partnership agreements that have been
(are being) concluded by the EU. Under these agreements Mediterranean countries commit to
abolish barriers to EU imports over a 12 year period and to abide by the principles of Articles 85,
86, and 92 of the ECT. For the first five years (as opposed to three for the CEE countries),
Mediterranean nations will be regarded as disadvantaged regions for the purposes of the rules on
subsidies (Art. 92). Rules to enforce the provisions of Arts. 85-86 and 92 are to be adopted by
the Association Council after an initial five year period. The agreement does not require
nondiscrimination in government procurement, and does not impose any requirements with
respect to the liberalization of trade and investment in services. The Euro-Med agreements are
22/ Indeed, as noted by Holmes (1996, p.5), in practice the EU is requiring its partners (future
members) to adapt national competition rules to the EU's standards in a much more rigorous
fashion than has been done by existing member states. A recent survey by Pittman (1997)
documents that the CEE countries have already gone far down the harmonization road.
20



therefore less far-reaching than the Europe Agreements, and it should come as no surprise that
instruments of contingent protection remain applicable to intra-area trade flows.23
The Australia-New Zealand Closer Economic Relations Trade Agreement (ANZCERTA)
ANZCERTA is a free trade agreement between Australia and New Zealand established in 1983.
Article 12(1)(a) of the ANZCERTA requires the two countries to: "examine the scope for taking
action to harmonize requirements relating to ...restrictive trade practices." Competition regimes
in the two countries differed significantly in 1983. Australian antitrust laws followed the US
model, whereas New Zealand's legislation was much more modeled on that of the UK. In 1986,
New Zealand's Parliament enacted new competition legislation which was much more similar to
the Australian system.24 Although this was largely motivated by a desire of the government to
enhance competition in the economy, it also facilitated discussions between the two countries to
eliminate antidumping on bilateral trade flows.
In a review of ANZCERTA in 1988 a Protocol on Acceleration of Free Trade on Goods
was appended to the Agreement. The Preamble to this Protocol states that: "...the maintenance of
anti-dumping provisions in respect of goods originating in other Member States ceases to be
appropriate as the Member States move towards the achievement of full trade in goods between
them." Article 4 of the ANZCERTA was modified to read: "The member States agree that anti-
dumping measures in respect of goods originating in the territory of the other Member States are
not appropriate from time of achievement of both free trade in goods between the Member States
on 1 July 1990, and the application of their competition laws to relevant anti-competitive conduct
23/ See Galal and Hoekman (1997) for assessments of the content and economic effects of the
Euro-Med agreements.
24/ Ahdar (1991, p. 332).
21



affecting trans-Tasman trade in goods"... and "Each Member State shall take such actions as are
appropriate to achieve the application of its competition law by 1 July 1990 to [anticompetitive
conduct affecting trans-Tasman trade] in a manner consistent with the principles and objectives
of the Agreement." With this amendment the CER achieved unconditional free trade. Since the
amended antitrust legislation came into effect, no cases have been brought alleging abuse of a
dominant position that deters intra-Tasman competition.
Ahdar (1991) argues that in the ANZCERTA context full trade liberalization was judged
necessary but not sufficient to eliminate the need for antidumping. Such elimination required
active enforcement of similar competition laws and agreement that the jurisdiction of
competition agencies extend to matters affecting trade between New Zealand and Australia. In
this connection it was agreed that nationals of one state could be made the subject of an enquiry
by the competition authorities of the other state and be required to respond to requests for
information. Australian (New Zealand) antitrust legislation was amended to extend its scope to
the behavior Australian and/or New Zealand firms with market power on either one of the
national markets or the combined Australia/New Zealand market; Courts were empowered to sit
in the other country; orders may be served in the other country; and judgements of Courts or
authorities of one country are enforceable in the other country. In 1994 the competition
authorities of the two countries concluded a bilateral Cooperation and Coordination Agreement
to reduce the possibility for inconsistencies in the application of legislation in instances where
this is not required by statutory provisions (WTO 1996).
As in the EU, elimination of antidumping was linked to the transition path for the
realization of free trade (July 1990). There was no effort to gradually increase the "competition-
22



consistency" of antidumping.25 The elimination was accompanied by action by one member of
the PTA to move towards the adoption of antitrust legislation that conformed relatively closely
with that enforced by the partner country. The application of antitrust remedies remains strictly
national. Ahdar (1 991) argues that competition policy harmonization played a role in the decision
to abolish antidumping. However, it is unclear why or how the link was made. A reading of the
CER suggests rather that the focus was on the realization of free trade (a single market). The
agreement includes various deeper integration elements that aim at achieving this goal. Thus,
disciplines on subsidies are included. The 1988 Protocol banned industry-specific subsidies
("bounties and subsidies providing long term support can no longer be regarded as a viable
instument of industry policy");26 export subsidies are also prohibited (these were eliminated by
1987). Although investment (capital flows) are not covered by CER, Australia and New Zealand
also maintain relatively liberal investment regimes, and significant cross-investment flows have
occurred.27 ANZCERTA also unifies the labor market of the two countries (there is free mobility
of labor), and contains relatively far-reaching commitments to liberalize trade in services. It is
therefore similar to the EU in that the objective is to move towards the establishment of a single
"trans-Tasman" market. As an instrument of protection, antidumping was inconsistent with the
25/ A facilitating factor could be that antidumping had become increasingly difficult to obtain.
Between 1983 and 1988, New Zealand initiated 39 dumping investigations against Australia. Of
these only two led to an affirmative finding. Similarly, Australia imposed duties in only 3 of 34
cases during this period (Ahdar, 1991).
26/ Lloyd, 1991, p. 24
27/ As of the mid 1990s, Australian FDI stocks in New Zealand equaled US $8 billion,
equivalent to 25 percent of the stock of inward FDI. The comparable figure for New Zealand FDI
in Australia was US $5 billion, or 5 percent ofthe total stock. Kahler (1996, p.110) notes that a
factor constraining extension of CER to cover investment is the Treaty of Nara between Australia
and Japan, which would require that Australia extend any benefits granted New Zealand in this
area to Japan.
23



realization of this objective.
The North American Free Trade Agreement (NAFTA)
Both the Canada-US Free Trade Agreement and its successor NAFTA substantially liberalize
trade and investment flows between signatories. However, NAFTA does little to discipline the
use of subsidies-it simply restates the existing multilateral (WTO) disciplines and remedies in
this area-or to apply common competition disciplines to intra-NAFTA trade. Antidumping
remains applicable to internal trade flows. Despite the Canadian government's interest in
disciplining the use of antidumping, negotiators were unable to move far down the path towards
elimination. Although ostensibly antidumping and antitrust were not linked in the negotiations,
in part this may have reflected differences in the competition rules and their enforcement in the
member countries, (e.g., triple damages in the US; rights of private action for breach of
competition rules). But the major factor underlying the failure of the Canadian government to
remove the threat of antidumping appears to have been the strength of the US lobby that
supported the continued existence of antidumping.
As a result, the approach taken in NAFTA towards antidumping was limited to
introducing mechanisms to ensure that domestic legislation is applied correctly. Independent
binational panels may be called upon to review final antidumping determinations under the
domestic law of each party. Moves to agree to link the abolition of antidumping to the
implementation of common disciplines on antitrust, subsidies or exclusive rights are not on the
agenda. This is not to say there is no recognition of potential gains from cooperation in these
areas. For example, as far as antitrust is concerned, parties have agreed to apply the positive
comity principle and are engaged in a process of discussing the trade-competition policy
interface. Article 1504 of the NAFTA established a working group on trade and competition.
24



This group is required to report to the NAFTA Commission on issues concerning the relationship
between antitrust law and policy and intra-NAFTA trade before the end of 1998. This working
group has mostly focused on a comparison of competition legislation in the three member states
with a view to determine whether differences have any implications for intra-NAFTA trade. A
perusal of the working group's 1997 interim report on its activities suggests that antidumping has
not figured prominently in the discussions.28 Instead the focus has been on the application and
possible trade effects of competition law enforcement.
MERCOSUR
The Southern Cone Market (Mercado Commun del Sur-MERCOSUR) was established in
1991 29 It is a common market and therefore has a common external tariff (implemented in
1995). The MERCOSUR treaty was amended in December 1994 to include a variety of
institutional provisions, including on dispute settlement. In contrast to the EU-also a common
market-there are no provisions for supranational enforcement of the agreement; disputes tend to
be settled through negotiation.
While GATT Article XIX type safeguard actions ("emergency protection") on intra-
MERCOSUR trade were prohibited as of the beginning of 1995, antidumping actions remain
possible on internal trade. However, this possibility is intended to be temporary, and it is
envisaged that antidumping will be eliminated once agreements has been reached on a set of
28/ "Interim Report of the NAFTA 1504 Working Group to the NAFTA Commission" February
1997.
29/ The agreement was notified to the WTO under the Tokyo Round Enabling Clause, which
exempts developing countries from establishing full conformity of the agreement with the WTO
rules on economic integration (Art. XXIV GATT and Art. V GATS) (Laird, 1997, p. 3). Laird
(1997) provides a detailed description of the provisions and implementation of MERCOSUR.
25



"Common Rules on the Defense of Internal Competition." The question whether antidumping is
to be maintained is to be reviewed by December 2000. MERCOSUR members have initiated a
process to harmonize national competition laws and create a mechanism for coordinated action to
prevent anticompetitive practices (Rowat, Lubrano and Porrata 1997).3° In December 1996 a
Protocol for the Defense of Competition within MERCOSUR was adopted. This establishes a
prohibition on concerted practices that restrict or distort competition and affect trade between
member states, and gives MERCOSUR institutions the power to enforce these rules, although
implementation remains the responsibility of national competition agencies.3" Progress is
therefore being made which may allow antidumping to be eliminated at some point in the coming
years. Presently, however, antidumping continues to be applied. For example, Argentina
initiated 33 antidumping cases on imports from Brazil during the 1992-96 period, making Brazil
the number one target for such actions (Tavares and Tineo, 1997).
In sum, it appears that Mercosur is following an approach that resembles ANZCERTA,
with elements of the EU regime as well. The reach of competition policy disciplines are
conditional upon a trade effects test; there will be a common external trade policy at some point;
but competition policy will be enforced by national bodies. There is no explicit requirement or
commitment that antidumping on intra-area trade will be eliminated, or any formal statement to
the effect that in principle antidumping has no place in a common market. However, this is on
the agenda: members are committed to review the situation by the end of 2000.
30/ Decision 21/94, "Basic Elements of the Defense of Competition in MERCOSUR," December
14, 1996.
31/ As MERCOSUR institutions operate by consensus, a national competition authority will
have the option of refusing to implement a decision. The Protocol must be ratified by
parliaments of members states before it becomes effective. Some countries have yet to create the
necessary institutions (e.g., Uruguay and Paraguay).
26



Canada-Chile
In early 1996 Canada and Chile initiated negotiations to establish a bilateral FTA. It was
envisaged that such an FTA would be an "interim" agreement to provide a bridge to full NAFTA
accession for Chile.32 An agreement was concluded within one year. It is an interesting
agreement because it goes beyond NAFTA by eliminating antidumping.33 As in the EU,
elimination is conditional upon the abolition of tariffs and NTBs: antidumping on specific
products ceases to be applicable on intra-FTA trade on the date that tariffs on that product are
eliminated (defined at the 8-digit level). In no case is this period to extend beyond January 1,
2003. The abolition of antidumping does not extend to countervailing duties (CVDs). A
Committee on Antidumping and Countervailing Duties has the mandate to consult with a view to
defining subsidy disciplines further and eliminate the need for CVDs. Not much was done in the
way of harmonizing antitrust. Article J-01 (Competition Policy) simply requires each party to
adopt or maintain measures to proscribe anticompetitive business conditions and to take
appropriate action where necessary. It also requires competition authorities to consult and
cooperate, including via the exchange of information. Parties are prohibited from invocation of
dispute settlement procedures on the basis of the competition policy article. There is no
reference to the principle of positive comity, nor are there restrictions on the right of parties to
maintain monopolies or state enterprises.
This is an example of a PTA that aims to eliminate the reach of antidumping illustrates
that this is feasible without harmonization of antitrust regimes. The explicit exception that is
made to maintain CVDs suggests that subsidy/industrial policies were considered to be a greater
32/ Government of Canada, Department of Foreign Affairs and International Trade, Press
Release 240, December 29, 1995.
33/ The full text of the agreement can be found at http://www.sice.oas.org.
27



cause for concern. Given the strategic dimension that possibly underlies the bilateral elimination
of antidumping in Canada-Chile context, it is too soon to claim that progress can be made on the
antidumping front in the absence of "deeper" integration.34 Nonetheless, the agreement has set a
precedent by revealing that antidumping can be "delinked" from the problem of disciplining
industrial policies.
Prospective Agreements: the FTAA and APEC
Given their prominence it is perhaps useful to discuss briefly the two major regional initiatives
that are under construction in the Asia-Pacific region and the Americas: the Asia Pacific
Economic Cooperation (APEC) initiative and the Free Trade Area of the Americas (FTAA).
Under APEC, high income (developing) member countries have agreed to achieve free trade and
investment within the region by 2010 (2020). A large number of bodies have been created by
APEC that are responsible for exploring possible "collective actions" in issue areas that are
relevant in helping to attain and maintain the free trade goal. Competition policy is one of these
topics, although contingent protection is not. APEC's objectives in this area are to introduce and
maintain "effective or adequate competition policy and/or laws and associated enforcement
policies, ensuring the transparency of the above, and promoting cooperation among APEC
economies, thereby maximizing, inter alia, the efficient operation of markets, competition
among producers and traders, and consumer benefits" (PECC, 1997, p.2). The main focus of
work in this area has been on information gathering and sharing and on the provision of technical
assistance. The end result may be the development of non-binding principles on competition law
34/ One of the mandates of the Canada-Chile Committee on Antidumping and Countervailing
Duties is illuminating in this connection: it is to "consult on opportunities for working together
with other like-minded countries with a view to expanding agreement on the elimination of the
application of antidumping measures within free trade areas" (Article M-05).
28



and policy in APEC.
There appears to have been little effort to link the antidumping and antitrust issue areas.
The implications of a "competition framework" for various regulatory areas is one of the
dimensions of the competition policy work program, and trade remedies are part of this agenda.
However, given the US antipathy to address this issue, the limited attention that seems to have
been given to antidumping as an "anticompetitive" practice is not surprising. Given that any
principles that may be agreed upon in the competition area will be non-binding, it appears likely
that antidumping will continue to figure in APEC trade relations for some time to come.
Similar to the APEC process, in the FTAA context efforts are also being devoted in a
working group on competition policy to exploring possible modalities for cooperation. The
terms of reference of the relevant working group are to compile a database/inventory of national
legislation and practices, exchange views on the application/operation of competition policy
regimes and their relationship to trade in a free trade area, and make specific recommendations
on how to proceed in the construction of the FTAA in this area (OAS, 1997). Between May
1996 and July 1997 the working group met four times. A working group on subsidies,
antidumping and countervailing duties was also created. This group has mostly engaged in a
process of information collection and exchange, with a particular emphasis placed on eliminating
remaining agricultural export subsidies and other trade distorting measures affecting agriculture.
The terms of reference of the group do not make any mention of developing recommendations
for disciplining the use of contingent protection.
IV.  Free Trade and Deep Integration in Regional Agreements
The regional experience to date illustrates that PTAs vary on many dimensions. Of primary
interest here is whether PTAs achieve full shallow integration (free trade) (or at least have this as
29



an objective) and whether shallow integration was conditional on the pursuit of deeper
integration, particularly the adoption of common antitrust rules.
If shallow integration is defined as unconditional free trade in goods (including the
elimination of antidumping), only four of the agreements qualify: the EC, EEA, CER and
Canada-Chile (Table 1). This will increase to five if MERCOSUR members decide to abolish
antidumping in 2000. If the definition of shallow integration includes freedom of movement of
labor and capital, only the EC, the EEA and CER qualify.35 Four of these PTAs have adopted
common rules in the area of antitrust: the EC, EEA, CER and MERCOSUR. In all cases this is
part of a broader competition policy agenda, including disciplines on industrial policy.
Table I here
There are significant differences in approaches. ANZCERTA involves significant
harmonization of antitrust legislation and agreements allowing enforcement of national law in the
other nation's jurisdiction. In contrast, the EC was not aimed at harmonization of national legal
regimes. Instead, the objective of market integration is pursued through adoption of common
competition disciplines for practices that restrain trade, enforcement by a supranational body and
application of the principle of direct effect of EC law. The focus on trade effects was driven by
the concern that private practices not be permitted to segment national markets. Both CER and
MERCOSUR are similar to the EC in applying a trade effects criterion.
The close overlap between the agreements that have achieved shallow integration (or are
seeking to do so) and those that have common antitrust rules does not indicate there is a
relationship between the two. In the EU and EEA context abolition of antidumping was
accompanied with not only complete liberalization of trade in goods, services and factors of
35/ Formally, as noted above, CER does not incorporate a right of establishment, but in practice
cross-investment is relatively unconstrained.
30



production (shallow integration), but also coordination and substantial harmonization of
competition policies, including disciplines on government procurement, state aids (subsidies) and
entities granted exclusive rights. Indeed, it now appears that as far as the EU is concerned, a
country must adopt all of its internal market rules before the EU will remove the reach of its
"commercial defense" instruments. But there is not direct link between abolition of antidumping
and adoption of common antitrust rules. The focus of the EU (and the EEA) is on the creation of
a single market, and that has been the main factor underlying the elimination of antidumping.36
Although market integration is somewhat less-reaching in the Australia-New Zealand CER
agreement, it also has a "Single Market" dimension, illustrated by a common labor market,
disciplines on the use of subsidies, and mutual recognition agreements for product standards.
Here again the basic premise was that there was no scope to maintain antidumping if the
objective was to attain a single trans-Tasman market. The concern underlying cooperation and
convergence in antitrust was to ensure that similar rules would apply to anticompetitive practices
with a detrimental impact on intra-area trade. Dumping did not figure prominently in this
context.
Table 2 here
As illustrated in Table 2, all possible permutations of shallow and deep integration can be
found in PTAs. Agreement to coordinate or even agree on common antitrust rules is neither
necessary or sufficient to allow antidumping to be eliminated. Even far-reaching regulatory
commitments (deep integration) that extend beyond antitrust per se may not be sufficient to
36/ The weakness of the link is also illustrated by the fact that the conditions for elimination of
antidumping are not spelled out clearly in the EU's free trade agreements with East European and
Mediterranean countries. There is no language linking the elimination of antidumping to the
implementation of competition policy disciplines broadly defined. A decision to eliminate the
reach of antidumping is left firmly in the political domain.
31



make elimination of antidumping feasible (e.g., the Europe Agreements).37 Examples can also be
found where intra-regional free trade was attained without any move towards incorporation of
rules on antitrust regimes (e.g., Canada-Chile). Finally, antitrust authorities have also pursued
formal bilateral agreements independent of the existence of PTAs.
Although it is difficult to say anything definitive concerning the necessary or sufficient
conditions for the abolition of antidumping, a number of conclusions can be drawn. First, explicit
linkages between abolition of antidumping and adoption of common antitrust rules are not
prevalent. Second, antidumping may be used as a tool to extract concessions from potential
partner countries (e.g., the Europe Agreements; the Euro-Med Agreements). Third, if members
of a PTA aim at the creation of a single market, there can be no role for antidumping. Although
common antitrust disciplines may be adopted as part of a deeper integration effort, this is driven
by the integration objective, not by a perception that such disciplines are required to abolish
antidumping. Fourth, much may depend on the extent to which a PTA imposes disciplines on
the ability of governments to directly intervene in or support domestic industries. This seems to
be one of the major concerns underlying the EU's unwillingness to give up its right to impose
antidumping duties on imports from Central European and Mediterranean countries. The fact that
NAFTA did little to impose disciplines on subsidies may be an important element in the US
unwillingness to give up antidumping.38 Canada-Chile is also noteworthy in this connection in
that countervailing duties remain applicable.
37/ In the case of the Europe Agreements, deep integration is part of the "price" to be paid by
Central European nations to obtain duty-free access to the EC market and increase the likelihood
that at some point in the future unconditional free trade will be attained (ultimately through
accession).
38/ Of course, in principle countervailing duties are the legal instrument that should be used in
this context, but in practice it is well known that antidumping and countervail are substitutes,
with firms frequently expressing a preference for antidumping (Leidy, 1997).
32



Finally, it can be noted that PTAs have done little in the way of adopting approaches that
are aimed at the gradual elimination of the instrument. Instead, abolition is either a dichotomous
event-here one day, gone the next-or there is a transition period, usually linked to the
elimination of tariffs more generally, after which it may not be used. No use has been made of
mechanisms of the type that have been proposed in the trade policy literature to introduce
competition criteria into the antidumping process. Such options include introduction of an
"injury to competition" standards or a requirement that the importing country's competition
authorities is given the mandate to block the imposition of antidumping measures if this would
endanger competition on the market.39 This also suggests the absence of strong link between
elimination of antidumping and antitrust disciplines.
V.   Implications for Multilateral Disciplines on Regional Agreements
GATT Article XXIV allows for the formation of PTAs between WTO members as long as these
meet a number of conditions. One of these is that "duties and other restrictive regulations of
commerce ... are eliminated with respect to substantially all trade between the constituent
territories (Art. XXIV:8a). The only exceptions concern quantitative restrictions that satisfy
GATT provisions (e.g., agriculture, balance of payments, and health or safety considerations).
As no mention is made in Art. XXIV of antidumping as a valid exception, strictly speaking PTAs
maintaining this instrument are "illegal" (being a form of restrictive regulation of commerce)
(Mavroidis, 1996). How members of a PTA eliminate existing restrictions on commerce is not
specified by Article XXIV. Thus, they are free to make cooperation in the area of antitrust a pre-
condition for the elimination of antidumping if they desire to do so, but are certainly not required
3.2/ For more detailed discussion of these and other options, see, e.g., Wood (1989), Morris
(1993), Messerlin (1994, 1996), Hoekman and Mavroidis (1996) and Sch6ne (1996).
33



to. It could be argued that if deep integration is required to achieve unconditional free trade, Art.
XXIV:8 cannot be held to have been violated as the WTO does not extend to deep integration.
But many analysts have argued that regional agreements should only be tolerated on systemic
grounds if they go beyond the rules and disciplines of the WTO (see, e.g., Bhagwati and
Panagariya, 1996). Insofar deep integration is a (political) requirement to eliminate the reach of
antidumping, this would imply that the PTA does go beyond the WTO. The same conclusion
applies if deep integration is not required and one accepts the argument that antidumping is a
perfectly legitimate exception to Art. XXIV:8. In that case, a prohibition on antidumping would
also go beyond the WTO, thus satisfying the criterion.
Most of the PTAs notified to the GATT/WTO Secretariat over the years have not resulted
in the elimination of contingent protection (if this existed) and have not included a transition
period during which this was to be achieved.40 Despite this, there has never been an instance
where a GATT working party disapproved of a PTA because of the continued existence of
antidumping. Of course, this no doubt reflects the more general weakness of the WTO process.
In any event, abstracting from the political constraints affecting scrutiny of PTAs in the WTO
context, the most straightforward solution to the problem of antidumping in the PTA context
would be to require that to be compatible with the WTO any PTA should abolish the reach of
contingent protection on intra-area trade
VI.  Concluding Remarks
The regional experience illustrates that shallow integration--unconditional free trade--may be
associated with the pursuit of a deeper integration agenda. However, arguments that common or
40/ Marceau (1994, pp. 147ff).
34



similar antitrust rules are a necessary condition for the elimination of antidumping do not have a
strong economic foundation and are only weakly supported by the regional evidence. The
linkages that are sometimes argued to exist in practice between antidumping elimination and
enforcement of common antitrust disciplines tend to be driven by a broader agenda that revolves
around the attainment of greater economic integration (e.g., the "single" market goal of the EC,
EEA and CER).
All the PTAs that have led to the abolition of antidumping, or are moving in that
direction, go beyond the WTO in at least some important respects. This is also the case for the
Canada-Chile agreement, which is NAFTA-like in the extent of its disciplines in the investrnent
area. One can conclude from the available evidence that although agreement on common
antitrust disciplines is not necessary to eliminate antidumping, common disciplines in other
regulatory areas affecting competition between firms in each jurisdiction may be needed to
completely remove the reach of contingent protection. The Canada-Chile agreement is again
suggestive in this connection, as there are no strong disciplines on the use of subsidies and
consequently countervailing duties continue to be applicable. In practice therefore, it may be that
the demise of antidumping in PTAs is constrained because governments are concerned about the
potential for their partners to engage in beggar thy neighbor industrial policies. They may
consider antidumping a useful defensive instrument in this connection, because it can be used as
a substitute for instruments like countervailing duties which have a much higher "foreign policy
content" and may be more difficult to pursue. If so, it must be recognized that antidumping is a
particularly ineffective and costly instrument. Indeed, a case can be made that this is another
reason to eliminate it in the PTA context as this will help focus attention on the real source of
problems (industrial policies; government intervention), rather than on the symptoms (allegations
of "unfair dumping").
35



Experience has revealed that antidumping can be eliminated in the regional context. The
burden of proof should therefore be put on those that argue for the maintenance of this
instrument in a PTA. Existing WTO rules in this connection appear unambiguous in requiring
that all duties and other restrictive regulations of commerce must be eliminated in a PTA. Given
the pervasive spread of PTAs, perhaps a start could be made to reduce the reach of antidumping
by beginning to take the existing provisions of the WTO more seriously.
36



References
Ahdar, Rex. 1991. "The Role of Antitrust Policy in the Development of Australian - New
Zealand Free Trade," Northwestern Journal of International Law & Business, 12:317-334.
Barros, Pedro and Luis Cabral. 1994. "Merger Policy in Open Economies," European Economic
Review, 38:1041-55.
Barros, Pedro and Jose Mata. 1996. "Competition Policy in Portugal," presented at the
CEPRIEuropean University Institute workshop on Recent Developments in the Design and
Implementation of Competition Policy, Florence, 29-30 November.
Bhagwati, Jagdish and Arvind Panagariya. 1996. The Economics of Preferential Trade
Agreements. Washington DC: American Enterprise Institute.
Boner, Roger and Reinald Krueger. 1991. The Basics ofAntitrust Policy: A Review of Ten
Nations and the European Communities. Washington D.C.: World Bank Technical Working
Paper No. 160.
Bourgeois, Jacques. 1993. "Competition Policy and Commercial Policy," in M. Maresceau
(ed.), The European Community's Commercial Policy after 1992: The Legal Dimension. Leiden:
Kluwer.
Collie, David. 1991. "Export Subsidies and Countervailing Tariffs," Journal of International
Economics, 31:309-24.
Collie, David. 1997. "Mergers and Trade Policy Under Oligopoly," presented at a CEPR/Institut
d'Analisi Economica (CSIC) workshop on competition and trade policy, Barcelona, November.
Commission of the European Communities. 1995. "Competition Policy in the New Trade
Order: Strengthening International Cooperation and Rules," Report by the Group of Experts
(DG-IV, July)
Deardorff, Alan and Robert Stem. 1987. "Current Issues in Trade Policy: An Overview," in
Robert Stem (Ed.), US. Trade Policies in a Changing World Economy. Cambridge: MIT Press.
Dixit, Avinash. 1984. "International Trade Policies for Oligopolistic Industries," Economic
Journal, 94:1-16.
Dixit, Avinash. 1988. "Antidumping and Countervailing Duties under Oligopoly," European
Economic Review, 32:55-68.
Ehlermann, Claus-Dieter. 1992. "The Contribution of EC Competition Policy to the Single
37



Market, Common Market Law Review, 29, 257-82.
Fingleton, John. 1996. "Competition Policy in Ireland," presented at the CEPR/European
University Institute workshop on Recent Developments in the Design and Implementation of
Competition Policy, Florence, 29-30 November.
Galal, Ahmed and Bernard Hoekman (eds.) 1997. Regional Partners in Global Markets: Limits
and Possibilities of the Euro-Med Agreements. London: CEPR.
Hindley, Brian and Patrick Messerlin. 1993. "Guarantees of Market Access and Regionalism,"
in Kym Anderson and Richard Blackhurst (eds.). Regional Integration and the Global Trading
System. London: Harvester-Wheatsheaf.
Hoekman, Bernard. 1997. "Competition Policy and the Global Trading System," The World
Economy, 20:383-406.
Hoekman, Bernard. 1998. "Competition Policy and Preferential Trade Agreements," presented
at a Brookings Institution conference on Private Practices and Trade Policy.
Hoekman, Bernard and Michael Leidy. 1992. "Cascading Contingent Protection," European
Economic Review, 36:883-92.
Hoekman, Bernard and Petros C. Mavroidis. 1995. "Linking Trade and Competition Policies in
Central and Eastern Europe," in L. Alan Winters (ed.), Foundations of an Open Economy: Trade
Laws and Institutions for Eastern Europe. London: Center for Economic Policy Research.
Hoekman, Bernard and Petros C. Mavroidis. 1996. "Dumping, Antidumping and Antitrust,"
Journal of World Trade, 30:27-52.
Holmes, Peter. 1996. "Competition Policy and Integration: Levelling or Tilting the Playing
Field," Global Economic Institutions Working Paper Series No. 21.
Horn, Henrik and Jim Levinsohn. 1996. "Merger Policy and Trade Liberalization," NBER
Working Paper 6077.
Kahler, Miles. 1995. International Institutions and the Political Economy of Integration.
Washington DC: Brookings Institution.
Laird, Sam. 1997. "Mercosur: Objectives and Achievements," WTO, mimeo.
Leidy, Michael. 1997. "Macroeconomic Conditions and Pressures for Protection under
Antidumping and Countervailing Duty Laws: Empirical Evidence from the United States," IMF
StaffPapers, 44:132-44.
38



Leidy, Michael and Bernard Hoekman. 1990. "Production Effects of Price- and Cost-based,
Antidumping Laws," Canadian Journal of Economics, 23:873-95.
Lloyd, Peter. 1991. "The Future of CER: A Single Market for Australia and New Zealand,"
Committee for Economic Development of Australia. Monograph No. 96. Wellington: Victoria
University Press.
Marceau, Gabrielle. 1994. Antidumping and Antitrust in Free Trade Agreements. Oxford:
Oxford University Press.
Marceau, Gabrielle. 1995. "The Full Potential of the Europe Agreements: Trade and
Competition Issues," World Competition, 19(2):35-69.
Mavroidis, Petros C. 1996. "The Treatment of Dumping, Subsidies and Anti-Competitive
Practices in Regional Trade Agreements: Comment," University of Neuchatel, mimeo.
Messerlin, Patrick. 1990. "Antidumping Regulations or Procartel Law? The EC Chemical
Cases," The World Economy, 13, 465-92.
Messerlin, Patrick. 1994. "Should Antidumping Rules be replaced by National or International
Competition Rules?," Aussenwirtschaft, 49:351-74.
Messerlin, Patrick. 1996. "Competition Policy and Antidumping Reform," presented at the
conference The World Trading System: Challenges Ahead, Institute for International Economics,
Washington D.C., June 24-25.
Morris, John. 1993. "International Trade and Antitrust," University of Cincinnati Law Review,
61:945-53.
Motta, Massimo,and Fabrizio Onida. No date. "Trade and Competition Policy," mimeo.
Neven, David, Robin Nuttall and Paul Seabright. 1993. Merger in Daylight: The Economics
and Politics of European Merger Control. London: CEPR.
Norberg, Sven. 1992. "The Agreement on the European Economic Area," Common Market Law
Review, 29:1171-98.
Ordover, Janusz and Robert Willig. 1986. "Perspectives on Mergers and World Competition,"
in R. Grieson (ed.), Antitrust and Regulation, Lexington MA: D.C. Heath.
Organization of American States (OAS). 1997. "FTAA Working Groups: Competition Policy,"
http://www.alca-ftaa.oas.org.
39



Pacific Economic Cooperation Council (PECC). 1997. "Competition Principles Project,"
September 25.
Pittman, Russell. 1997. "Competition Law in Central and Eastern Europe: Five Years Later,"
US Department of Justice, mimeo.
Reitzes, J. 1993. "Antidumping Policy," International Economic Review, 34:745-63.
Rowat, Malcolm, Michele Lubrano and Rafael Porrata Jr. 1997. "Competition Policy and
Mercosur," World Bank, mimeo.
Sch6ne, Rainer. 1996. Alternatives to Antidumpingfrom an Antitrust Perspective. University of
St. Gallen doctoral dissertation.
Staiger, Robert and Frank Wolak. 1989. "Strategic Use of Antidumping To Enforce Tacit
International Collusion," NBER Working Paper 3016.
Staiger, Robert and Frank Wolak. 1992. "The Effect of Domestic Antidumping Law in the
Presence of Foreign Monopoly," Journal of International Economics, 32:265-87.
Stewart, Terrence. 1996. "Why Antidumping Laws Need Not Be Cloned After Competition
Laws Nor Replaced By Such Laws," presented at a conference Antidumping and Competition
Policy: Complements or Substitutes?, Center for Applied Studies in International Negotiations,
Geneva, July 11-12.
Tavares de Araujo, Jose and Luis Tineo. 1997. "The Harmonization of Competition Policies
Among Mercosur Countries," Organization of American States, mimeo.
Togan, Subidey. 1997. "Opening Up the Turkish Economy in the Context of the Customs Union
with EU," Journal of Economic Integration, 12:157-179.
Veugelers, Reinhilde and Hylke Vandenbussche. 1997. "European Antidumping Policy:
Profitability of National and International Collusion," European Economic Review, forthcoming.
Viner, Jacob. 1923. Dumping: A Problem in International Trade.
Weatherill, Stephen and Paul Beaumont. 1993. EC Law. London: Penguin.
Wood, Diane. 1989. "'Unfair' Trade Injury: A Competition-based Approach," Stanford Law
Review, 41:1153.
40



Table 1. Summary Comparison of PTAs
EC    EEA    EA   EMA   NAFTA    CER   MERCO   CAN-
SUR       CHILE
Free labor          Yes       Yes       No       No        No         Yes        No         No
mobility
Free capital        Yes       Yes       Yes      No        Yes        No         No         Yes
flows
Free services       Yes       Yes       Yes      No    Significant   Signi-      No        In part
trade                                                                ficant
Competition         Yes       Yes       Yes      No        No         Yes        No         No
policy rules*
Harmnonization    Partly,    Partly,   Partly,    No       No        Signi-   In part       No
of national       ex post   ex post  ex ante                         ficant
antitrust
Area-wide           Yes       Yes       Yes      Yes       n.a.       Yes       Yes         n.a.
antitrust rules
conditional on
"trade effects"
test
Formal              n.a.      Yes       Yes      No        Yes        Yes      t.b.d.       Yes
cooperation
agreements
between antitrust
authorities
Supranational      Yes        Yes       No       No        n.a.       No         No         n.a.
enforcement of
antitrust
Binding dispute     No        No        No       No        n.a.       No        t.b.d.      n.a.
settlement on
antitrust
Elimination of      Yes       Yes       No       No        No         Yes      t.b.d.       Yes
contingent
protection
* Defined as significant disciplines on industrial policies (e.g., subsidies)
n.a.: not applicable; t.b.d.: to be determined.
Source: adapted from Hoekman (1998).
41



Table 2. Shallow/Deep Integration Permutations
Full shallow integration  Incomplete shallow integration
Deep integration   EC, EEA, ANZCERTA          Euro-Med Agreements, Europe
Agreements, MERCOSUR
NMo `eep           Canada-Chile               NAFTA
n ;egration
Note: Shallow integration is defined as free trade in manufactures and abolition of antidumping.
Deep integration comprises common competition policies, including antitrust.
42



Policy Research Working Paper Series
Contact
Title                            Author                  Date              for paper
WPS1925 Half a Century of Development      Jean Waelbroeck         May 1998           J. Sweeney
Economics: A Review Based on                                                31021
the Handbook of Development
Economics
WPS1926 Do Budgets Really Matter?          Emmanuel Ablo           June 1998          K. Rivera
Evidence from Public Spending    Ritva Reinikka                             34141
on Education and Health in Uganda
WPS1927 Revenue-productive Income Tax      Fareed M. A. Hassan     June 1998          A. Panton
Structures and Tax Reforms in                                               85433
Emerging Market Economies:
Evidence from Bulgaria
WPS1928 Combining Census and Survey Data Jesko Hentschel           June 1998          P. Lanjouw
to Study Spatial Dimensions      Jean Olson Lanjouw                         34i529
of Poverty                       Peter Lanjouw
Javier Poggi
WPS1929 A Database of World Infrastructure   David Canning         June 1998          A. Abuzid
Stocks, 1950-95                                                             33348
WPS1930 The Main Determinants of Inflation in llker Domac          June 1998          F. Lewis
Albania                          Carlos Elbrit                              82979
WPS1931 The Cost and Performance of Paid   Ariel Dinar             June 1998          F. Toppin
Agricultural Extenion Services: The   Gabriel Keynan                        30450
Case of Agricultural Technology
Transfer in Nicaragua
WPS1932 Air Pollution and Health Effects:  Bart D. Ostro           June 1998          C Bernardo
A Study of Respiratory Illness   Gunnar S. Eskeland                         31148
Among Children in Santiago, Chile   Tarhan Feyzioglu
Jose Miguel Sanchez
WPS1933 The 1997 Pension Reform in Mexico  Gloria Grandolini       June 1998          C. Zappala
Luis Cerda                                 87945
WPS1934 WTO Accession for Countries        Constantine Michalopoulos June 1998        L. Tabada
in Transition                                                               36896
WPS1935 Explaining the Increase in Inequality  Branko Milanovic   June 1998           G. Fvans
during the Transition                                                       85734
WPS1936 Determinants of Transient and      Jyotsna Jalan           June 1998          P. Sader
Chronic Poverty: Evidence from   Martin Ravallion                           33902
Rural China
WPS1937 Aid, the Incentive Regime, and     Craig Burnside          June 1998          E. Khine
Poverty Reduction                David Dollar                               37471



Policy Research Working Paper Series
Contact
Title                           Author                   Date              for paper
WPS1938 What Explains the Success         David Dollar            June 1998          E. Khine
or Failure of Structural Adjustment   Jakob Svensson                       37471
Programs?
WPS1939 Second Thoughts on Second         Arturo J. Galindo       June 1998          M. Cervantes
Moments: Panel Evidence on      William F. Maloney                         37794
Asset-Based Models of Currency
Crises
WPS1 940 The Structure of Labor Markets in    William F. Maloney  June 1998          M. Cervantes
Developing Countries: Time Series                                          37794
Evidence on Competing Views
WPS1941 Are Labor Markets in Developing   William F. Maloney      June 1998          M. Cervantes
Countries Dualistic?                                                       37794
WPS1942 Poverty Correlates and Indicator-    Christiaan Grootaert  July 1998         G. Ochieng
Based Targeting in Eastern Europe  Jeanine Braithwaite                     31123
and the Former Soviet Union
WPS1943 The Implications of Hyperbolic    Maureen Cropper         July 1998          A. Maranon
Discounting for Project Evaluation    David Laibson                        39074
WPS1944 Detecting Price Links in the World   John Baffes          July 1998          J. Baffes
Cotton Market                                                              81880
WPS1945 Evaluating a Targeted Social      Martin Ravallion        July 1998          P. Sader
Program When Placement Is       Quentin Wodon                              33902
Decentralized
WPS1946 Estonia: The Challenge of Financial  Carlos Cavalcanti    July 1998          L. Osborne
Integration                     Daniel Oks                                 38482
WPS1947 Patterns of Economic Growth: Hills,  Lant Pritchett       July 1998          S. Fallon
Plateaus, Mountains, and Plains                                            38009
WPS1948 Comparative Advantage and the     Aart Kraay              July 1998          A. Kraay
Cross-Section of Business Cycles   Jaume Ventura                           35756
WPS1 949 Education and Earnings Inequality   Ulrich Lachler       July 1998          C. Lazcano
in Mexico                                                                  37776