The WTO/World Bank Conference on Developing Countries' in a Millennium Round
WTO Secretariat, Centre William Rappard,
Geneva
20-21 September 1999

Enforcing Multilateral Commitments: Dispute Settlement and Developing Countries *

Bernard M. Hoekman
World Bank and CEPR

and

Petros C. Mavroidis
University of Neuchatel

September 14, 1999

It is frequently argued that one of the major results of the Uruguay Round--especially for developing countries--was the strengthening of GATT dispute settlement procedures (DSP) (Schott and Buurman, 1994; Croome, 1999). A unified dispute settlement process was created covering all multilateral agreements. The right to request a panel became (virtually) automatic, as did adoption of panel reports. Stricter procedural disciplines were adopted, including time limits and deadlines for each stage of the process, and a standing Appellate Body was established to which panel decisions could be appealed.¹ As a result, many observers believed "...that smaller countries (including developing countries) will now be able to raise disputes more effectively and have them settled in their favor" (Whalley, 1996, p. 422).

Developing country participation in dispute settlement in the first five years after the creation of the WTO would appear to support the optimistic expectations. A significant expansion in the number of dispute cases was observed relative to that under the GATT. Over 160 requests for consultations were brought to WTO in its first five years of operation; three times more on a per annum basis than under the GATT.² Developing countries are more often plaintiffs than in the past--about 25 percent of all cases were brought by developing countries. They were respondents in another 25 percent of all cases (Table 1). Disputes between developing countries have begun to emerge--the first case brought under the WTO was by Singapore against Malaysia. This development, in conjunction with the success some developing countries have had in contesting actions by large players (e.g., Costa Rica--US restrictions on cotton textiles; Venezuela and Brazil--US gasoline regulations) is frequently cited as proof that the system has become of greater relevance to developing countries.³

Greater use of the system does not necessarily reflect greater confidence in the effectiveness of DSP as an instrument to enforce market access rights. Hudec (1998) argues that the increase in cases is largely due to the expansion in legal obligations, one effect of which has been "Lawyers, Lawyers, Everywhere." There are costs associated with the expanded use of DSP, which are disproportionately heavy for developing countries. Least developed countries are not involved: there have been no cases involving sub-Saharan countries (Table 1), and intra-developing country cases remain limited. Many developing countries believe existing mechanisms can be improved to allow them to be used more effectively to defend their WTO rights, and that actions are required to "level the playing field" by enhancing their ability to both prosecute and defend cases. One issue that has attracted a lot of attention is the need for legal assistance for developing countries; a related concern is to make procedural disciplines more sensitive to the institutional and resource constraints that affect developing countries. More generally, the dispute between the US and the EU on bananas suggests that dispute settlement procedures can be improved.

This paper discusses the issue of enforcement of multilateral commitments from the perspective of developing countries. The approach taken is largely normative.⁴ We argue that the dispute settlement mechanisms embodied in the WTO are only one--albeit important--dimension of enforcement. To use them effectively, governments need to be able to bring cases to the WTO. To do that, there must be domestic mechanisms through which export interests can channel information to the government. For export interests to have an incentive to do this, they need to be aware of what the WTO rules are and have confidence that the expected return to bringing cases to their government exceeds the costs of doing so. This in turn will depend in part on the probability that the government will pursue the case, and conditional on that, the expected payoff associated with winning the case. In short, the effectiveness of mechanisms to enforce multilateral commitments depends importantly on the efficiency of "upstream" input activities--which comprise a complex interlocking chain. The efficiency of the "downstream" WTO panel process is only a part of the equation that must be considered.

The domestic "upstream" dimension of defending WTO rights--identification of potential cases, transmission of information to the government, and getting the government to act--is in turn just one determinant of the relevance of the WTO for domestic stakeholders. As, if not more, important is the feasibility and effectiveness of using national legal mechanisms to enforce commitments made by governments. Inspired by the public choice and transactions cost literature, it is sometimes argued that a major potential benefit of GATT (now WTO) membership is that the multilateral commitments made by a government have a "constitutional" function. They safeguard the interests of a diffuse majority by increasing the cost of adopting policies--certain types of import barriers--that benefit only small but powerful interest groups (e.g., Tumlir, 1985).⁵ For this to be the case, it is necessary that multilateral commitments can be the basis of claims by individuals before national tribunals or courts. The greater the feasibility and effectiveness of enforcement at the national level, the greater the incentive for traders to defend "their" WTO rights before national courts.

We argue in this paper that in practice there is excessive reliance on countries with an export interest to invoke WTO DSP in cases where a government is alleged to violate a WTO commitment, and that this is particularly detrimental to developing countries. Foreign entities with export interests can only be relied on to pursue this path in the case of large markets. This is the main reason why Quad countries are usually involved in WTO cases on one side or another. Foreign exporters will not have strong incentive to contest a WTO violation by a small poor country: the market is simply too small. Moreover, import interests in developing countries may not be able to contest actions by their government that violate WTO commitments. A consequence of this situation is that developing countries may have de facto "exempt" status in terms of implementation of their WTO obligations. At the same time, the absence of mechanisms at the domestic level to overcome problems of asymmetric information imply developing countries find it more difficult to use either the WTO or foreign national court systems to enforce WTO rights.

We conclude that efforts to assist developing countries use WTO DSP (e.g., an advisory centre in Geneva, greater technical assistance by the WTO secretariat), must be supplemented by: (i) strengthening the upstream, domestic, parts of the WTO enforcement chain; (ii) allowing individuals to use national courts or specialized tribunals to contest actions by their government that violate WTO obligations (tariff bindings; valuation; etc.); and (iii) changing the Dispute Settlement Understanding (DSU) to make it easier and more efficient to bring cases where the absolute value of the trade flows concerned is small, or the relative importance of the affected trade for the plaintiff is large.

Pursuing the first two tracks is vital if one believes (as we do) that a major role of the WTO is to be a credibility instrument, an anchor, for domestic policy. Strengthening domestic enforcement mechanisms can help make the WTO a more relevant instrument from an economic development perspective by increasing the "ownership" of negotiated commitments. Absent domestic enforcement mechanisms, the primary burden of ensuring compliance is put on external entities. This may have negative political repercussions, augment ownership problems--already an important issue (Finger and Schuler, 1999)--and often will be ineffective in any event.

The paper is organized as follows. A number of domestic and multilateral dimensions of enforcement of WTO commitments are discussed in Section 1. In Section 2 we briefly summarize some of the major problems at the WTO level that have been identified by developing countries as part of the recent review of the Dispute Settlement Understanding (DSU). Section 3 discusses domestic mechanisms to enhance information flow and transparency with regard to potential violations of WTO obligations. Section 4 discusses options for reforming WTO dispute settlement mechanisms. Section 5 concludes.

Without effective enforcement mechanisms, agreements may have little value.⁶ The standard device used within a country--a central authority: the apparatus of the State--is not available internationally. Given the absence of a central enforcement body, WTO agreements (as is the case for most international agreements) must be self-enforcing if commitments are to be credible. Various mechanisms are used by WTO members to support compliance with negotiated agreements, including regular communication and interaction in committee meetings (review of legislation; notification requirements); publication requirements for laws and regulations; and periodic multilateral surveillance (the Trade Policy Review Mechanism--TPRM). Dispute settlement complements these information and transparency mechanisms.⁷

The dispute settlement mechanisms embodied in the WTO aim at assuring (preserving) negotiated commitments. The role of the dispute settlement process is three-fold:

1. to provide WTO members with an agreed procedure through which to raise issues and exchange information (consultations);
2. in cases where consultations fail to resolve a matter, delegate to a panel of neutral experts the task of determining whether a WTO rule has been violated or a market access concession nullified or impaired; and
3. to put pressure on a member found to have violated its commitments to bring its policies back into conformity with WTO disciplines.

Four aspects of WTO DSP are particularly important: the available remedies, the incentives for losers to comply with rulings, the reliance on export interests to drive the process, and the fact that only governments may bring cases to the WTO.

For enforcement threats to be credible, there must be effective remedies. The WTO panel process takes a substantial amount of time to come to a conclusion--up to 30 months. Rapid resolution of a problem is not in the cards if consultations do not lead to a settlement, reducing the present discounted value of winning a case that goes to a panel. More important, generally the remedy proposed by a panel will not involve any compensation for the loss that has been incurred by a firm or industry. The government found to be in violation is simply told to bring its measures into compliance or conformity with WTO rules.⁸ The most panels can do is to make specific suggestions regarding the way a losing party can bring its measures into conformity. Even if compensation or restitution would be a suggested remedy (quite unlikely given US and EU opposition--see Palmeter and Mavroidis, 1999), this would accrue to the government winning the case. Whether this would be transferred to the firms that petitioned for the case is an open question.

It has been argued that current WTO dispute settlement practice creates a bias in favor of "innocent" recommendations that WTO members bring their measures into conformity with their obligations, rather than potentially "provocative" suggestions that specific actions be undertaken to achieve compliance (Matsushita, Mavroidis and Schoenbaum, 2000). Panels, in contrast to the standing Appellate Body, are formed on an ad hoc basis. In most cases panelists are selected by the parties to a dispute, and usually comprise (ex) government officials who are sensitive to diplomatic and pragmatic considerations, and are generally not inclined to suggest remedies that might "rock the boat." The recent Bananas case brought to the fore some of the problems that exist with the remedy dimension of WTO dispute settlement. In particular, it has become evident that the DSU has a potentially serious weakness in dealing with cases where it is alleged that the losing party has not implemented the panel’s recommendation. Such cases can easily arise when panels limit themselves to standard recommendations to bring measures into compliance, or, alternatively, if a panel’s suggestions are not interpreted as binding.

The DSU (Art. 21.3c) provides that the "reasonable period of time to implement panel or Appellate Body reports should not exceed 15 months from the date of the adoption of a panel or Appellate Body report." If there is disagreement between the parties as to the adequacy of implementing measures after this period has expired, the complaining party can request a panel (if possible, the original one) to pronounce on this issue (Art. 21.5 DSU). This can give rise to a recurring series of panels dealing with essentially the same issue, the Respondent country effectively obstructing recourse to retaliation by the complaining country by doing something which is claimed to meet the panel’s recommendations. The Complainant cannot obtain authorization to adopt countermeasures, because it is not the same dispute anymore, in the sense that a different measure than the one addressed by the original panel is now subject of scrutiny. Countermeasures can be used only when the Respondent remains passive.

The Bananas dispute illustrates that this scenario is not just a theoretical possibility. Currently, a WTO Member can effectively avoid implementation if it keeps on doing something but clearly not enough to bring its measures into compliance. This can easily erode to the legitimacy of the WTO legal apparatus. Preventing a repetition of what happened in Bananas is therefore important, especially for the small players in the trading system.⁹ Within the current system of rules, the best way to prevent strategies of "implementation avoidance" is for plaintiffs to request specific remedies (suggestions) every time they bring forward a complaint (Hoekman and Mavroidis, 1996).

It has often been pointed out by economists that there are asymmetric incentives for countries to deviate from the WTO, with small nations having few, if any, instruments to force large ones to abide by the rules of the game (e.g., Bhagwati, 1990). This is because the ultimate threat that can be made against a violator is retaliation, something small countries cannot credibly threaten, as it will have little impact on the target market while being costly in welfare terms because the instrument of retaliation involves raising trade barriers. Thus, pressure to comply with panel rulings is largely moral in nature. In practice the system has worked rather well, in that recourse to retaliation has rarely been required to enforce multilateral dispute settlement decisions (Hudec, 1993). This is largely a reflection of the repeated nature of the WTO "game" and the resulting value that governments attach to maintaining a (reasonably) good reputation. The importance of information provision and reputational effects is often underestimated by economists analyzing WTO dispute settlement.¹⁰

It remains true, however, that in the past, GATT contracting parties have tolerated or accepted policies that violated previously negotiated rules (e.g., the Multifibre Arrangement or agricultural protection), indicating that the large players can dominate the weaker ones when they desire to. The classic recommendation by economists to address asymmetry in power is to share the cost of retaliation by use of collective enforcement. This involves extending the nondiscrimination principle to retaliation: non-implementation of panel recommendations would be punished by withdrawal of market access commitments by all WTO members, not just those who brought a case to the WTO. This would greatly enhance the credibility of the threat of retaliation. To date, this suggestion has always been resisted--developing country proposals to increase the cost of violating the GATT for high-income countries in the 1960s and subsequently have never gotten off the ground (Hudec, 1987).

The DSU does not permit the use of counter measures by third parties. However, there is no requirement that a country bringing a case have a direct trade interest--all that is required is a potential trade interest and its interest in a determination of its WTO rights (Palmeter and Mavroidis, 1999, p. 24). Thus, there is nothing to prevent multiple countries from initiating a joint action; indeed, in principle, all developing (and other) countries can join together to prosecute a case. This in turn suggests that the threat of joint retaliation can already be made within the existing system--although there are clearly costs associated with overcoming potential free rider problems, and not all WTO members will be willing to join. But given that the costs of participating in a case are low (assuming the primary interested party (parties) carries the cost of prosecuting the case), in principle a large number of developing countries could bring a joint case. A possible reason this is not done is that a defendant can refuse to hold joint consultations with more than one potential plaintiff (Art. 4.11 DSU) and can manipulate the calendar to oblige the DSB to establish more than one panel on essentially the same issue. Moreover, it cannot be guaranteed that the same panelists will be selected to adjudicate the various (essentially identical) cases. If this is the case, there is no way one will end up with one panel that is competent to adjudicate multiple complaints on the same subject.

Another problem that arises related to joint action concerns the magnitude of the threat that can be exercised. If a country has no trade interest, it is generally difficult to determine the level of the appropriate counter measure that can be taken in the event of non-compliance with a panel recommendation. Indeed, if a WTO member is found to be discriminating against a particular country or set of countries, this may benefit other countries. Thus, pursuing the joint action path requires agreement that the basis for retaliation be changed.¹¹ This is a problem that also affects the approach often suggested by economists, i.e., continued reliance on a system where the main affected parties pursue the dispute, but changing the rules to extend the MFN principle to retaliation, if this becomes necessary (Hoekman, 1993).¹² A mandatory requirement that all WTO members participate in the retaliation will avoid free rider problems, but continues to confront the problem of determining the magnitude of retaliation.

A basic problem with retaliation is that it involves raising barriers to trade, which is generally detrimental to the interests of the country that does so, and to world welfare more generally. Moreover, experience with economic sanctions has demonstrated that the efficacy of such instruments is limited. We are therefore of the view that strengthening the retaliation provisions of the WTO may be of limited usefulness. Much better from an economic perspective would be to create stronger disciplines requiring compensation to be offered to affected WTO members for damages incurred as the result of an illegal action. Financial compensation is unlikely to be feasible (politically acceptable) although it is certainly a legally thinkable remedy.

One way of pursuing this path to a somewhat greater extent than is done currently would be to encourage greater use of the provisions in the WTO for re-negotiating concessions. This would ensure that the net impact of dispute resolution would lead towards more liberal trade, rather than create mechanisms through which trade barriers are raised. This option is of course built into the WTO. Arts. XXVIII GATT and XXI GATS allow for re-negotiation, which involves a member compensating members affected by the withdrawal of a concession through a reduction in other trade barriers. Thus, a losing party that finds it difficult to comply with a panel’s suggestions could offer compensation by lowering its tariffs on sectors of interest to the winning party.¹³ More generally, the problems associated with strengthening retaliation mechanisms in the WTO suggest that the focus of enforcement efforts may be better placed on domestic institutions (see below).

The WTO dispute settlement process is a decentralized one that relies on export industries to petition their governments to bring cases to the WTO--first bilaterally through consultations, then to a panel.¹⁴ In practice this means that small markets (poor countries) attract less scrutiny than large (rich) ones. This has a downside, as it can reduce the value of WTO membership for developing countries by creating greater uncertainty that a government will maintain a WTO-consistent policy stance. Other mechanisms embodied in the WTO aim to attenuate this problem--the TPRM being the foremost example--but are of limited effectiveness. The TPRM process is arguably too infrequent to be useful for enforcement (most countries are reviewed only once every six years or more).¹⁵

Information is clearly a crucial input into the enforcement process. Such information must be compiled at the national level. Information on foreign barriers can be compiled and transmitted to government officials through a variety of channels. Some countries have implemented formal mechanisms that foster communication and interaction between industry and government (the US is the leading example). The major players produce annual reports listing complaints expressed by their national firms regarding discriminatory policies.¹⁶ Although the information that is contained in these compilations is ad hoc, with no attempt being made to rank order the issues that are identified by firms on the basis of a national welfare criterion, they do reflect a pro-active stance on the part of both industry and the government. Without mechanisms to generate and transmit information, enforcement is doomed to be weak. Governments need to be informed by the private sector what is going on in export markets, and the private sector needs to know the WTO rules so it can identify violations. On both fronts, developing countries lag behind OECD countries.

Only governments have standing to bring cases to the WTO DSP. This means that export interests must operate through a government filter. A factor determining the credibility of multilateral commitments is therefore whether the private sector has an incentive to bring cases forward. If there is substantial uncertainty whether the government will be willing to bring the case to the WTO, this may not happen. This issue is important, but is often ignored. Governments may engage in "tacit collusion" by refraining from contesting measures through fear of stimulating counter claims (the "glass house" syndrome). More generally, governments have multidimensional relationships with other countries. It may well be that a government is unwilling to "rock the boat" by bringing a case if they fear this will have detrimental consequences in other dimensions (e.g., continued aid flows; defense cooperation; migration quotas, etc.). As such, there is nothing wrong with this--it is the role of government to determine priorities and to make the tradeoffs it deems most beneficial for the nation as a whole. At the same time, a major function of the WTO is to de-politicize trade, and reduce concerns of governments of possible repercussions of bringing trade cases. Arguments that bringing a case would "disturb" a country’s relationship with a major OECD country to some extent nullify the raison d’etre of the WTO--the establishment of a rule-based as opposed to a power-based system of international trade relations.

One option to avoid the problem of "politicization" of trade disputes for exporters is to given them direct access to the WTO, i.e., "privatize" DSP. This is a complex issue that has not attracted much research. In an interesting paper, Levy and Srinivasan (1996) develop a simple model that explores the implied tradeoffs. They conclude that as long as a government maximizes national welfare it may have good reasons not to pursue a trade case because the expected national return is negative (due to issue linkage by the partner). They also argue that removing a government’s discretion to decide whether or not to prosecute a case can make it more difficult to participate in trade negotiations (to make commitments), and conclude that there is a good case to be made for permitting only governments to bring cases to the WTO. Their argument is compelling, in that it recognizes the real politik of inter-governmental relations, although it is depressing for those who believe that the main role of the WTO is move the trade regime towards a rule-based system.¹⁷

The Levy and Srinivasan analysis has several implications for enforcing multilateral commitments. First, and foremost, it suggests that enforcement mechanisms at the national level are important. If trade violations can be contested through channels that do not allow threats to be exercised in non-trade areas, the probability of compliance increases. Thus, if WTO obligations can be invoked before national courts or tribunals, exporters may be better off bringing a case to a domestic court in the importing country (or inducing their importers to do so). The problem of course is that national law should permit this--more on this below. Second, it again illustrates the importance of information. This is required to allow governments to determine accurately what the cost to economy is of not taking action. Third, the argument assumes that governments defend the national interest--domestic consultation and "transparency" mechanisms may need to be strengthened to ensure that the "right" calculation or tradeoffs are indeed being made.¹⁸ Finally, it illustrates the importance of multilateral mechanisms such as the TPRM that reduce the burden on individual countries of identifying and contesting a WTO violation.

The foregoing has identified a number of systemic aspects of the WTO dispute settlement mechanism, all of which are important for developing countries. Asymmetry in power and resources matters--both in terms of defining the threat points that apply in any given dispute, and in determining the incentives poor countries have to bring cases to the WTO. Not surprisingly, many developing countries perceive the dispute settlement playing field to be slanted against them. Criticism is directed in particular at the cost of the system, the weakness of available remedies, and the limited implementation of provisions on special and differential treatment in disputes involving developing countries.¹⁹

One issue that is frequently raised is that rich countries such as the US, the EU or Canada--the major players in terms of dispute settlement activity (Table 1)--have extensive resources on which they can draw. They are well equipped with legal talent, are well briefed by export interests, and have a worldwide network of commercial and diplomatic representation. Developing countries, in contrast, have limited national expertise available and find it difficult to collect the type of information that is required to bring or defend cases. Instead, they must rely on (expensive) third-party expertise.²⁰ The cost of bringing cases to the WTO is an important factor restraining invocation of dispute settlement. These costs, which are not only financial but also related to scarcity of administrative resources, prevent many developing countries from using the system to their full advantage (ACWL, 1999; Michalopoulos, 1999).

Article 27:2 DSU provides for technical assistance to be given to developing countries by the WTO. The Secretariat has only a very limited ability to advise developing countries--legal technical assistance services are provided by two academic experts on a part-time basis. The adequacy of the assistance on offer is further reduced by the DSU requirement that such assistance can only be provided after a Member has decided to submit a dispute to the WTO. Thus, assistance in evaluating whether practices are inconsistent and determining what might be "winning" cases cannot be given. This requires expertise at the level of the national administrations, something many developing countries do not have. Technical assistance therefore is mostly used when developing countries are respondents. The consensus among developing (and many OECD) countries is that the available assistance is inadequate.²¹

There is also concern regarding the length of time it can take for the DSP to run its course--up to 2½ years--which is a long time for exporters to be subject to a measure that may be a violation of the WTO, especially for countries that do not have a diversified export base. Other concerns relate to the fact that there are no mechanisms to compensate for losses incurred during the adjudication period; the weakness of the threat of retaliation; the fact that language regarding special and differential treatment of developing countries as regards dispute settlement has generally been ignored by panels (see South Centre, 1999 for a summary of existing provisions); facilitating participation in DSP as third parties; reducing the scope for sequential ("copy cat") cases; and defining more clearly the task of the Appellate Body (some countries perceive this Body as having exceeded its authority in several cases by engaging in de facto fact finding and taking decisions that should be left to WTO members--see South Centre, 1999).

Available statistics suggest that while many of these concerns are valid, it is important to note that the implementation record to date is quite good. Abstracting from high profile and politically sensitive cases such as Bananas, in most cases the country found to be at fault in a DSP has undertaken to comply with the DSB’s recommendations within the agreed time frame. Unfortunately it is impossible to know what the counterfactual would be under an "ideal" DSP as far as developing country participation is concerned. Undoubtedly resource constraints play a major inhibiting role. But the pattern of dispute resolution to date also suggests possible cause for concern. Over two-thirds of all cases that are notified to the WTO are settled subsequent to bilateral consultations. While reducing the burden of DSP for the WTO, this is potentially worrisome in that the outcome of cases is generally unknown.

Enforcement requires that agents concerned know about public commitments, are able to identify when commitments appear to be violated, and can take action to contest the violation. Knowledge of the prevailing rules of the game is clearly a prerequisite for commitments to be enforced. The private sector has to know what governments have committed themselves to. And government must have the in-house capacity to interact with and support private sector efforts to enforce commitments. In many developing countries there is a great need for education and training on the WTO and on national commitments made under WTO auspices that is tailored to the background of the audience. CEOs need to be convinced there is value to devoting resources to training and cooperating in the delivery of training--e.g., through chambers of commerce or industry associations. Governments must acquire not just trained legal expertise to pursue/defend cases, but more generally educate staff in all potentially affected Ministries how their work relates to the WTO. This is an area where much is already being done, often supported by donor agencies such as the European Commission, USAID, and the WTO itself. But it is clear that more is needed.

What follows focuses on two key dimensions of enforcement at the domestic level: (i) generating the information required to defend WTO rights; and (ii) creating institutions that reduce the need to rely on WTO-level DSP. The latter is crucial in order to hold governments accountable to domestic constituencies for international trade obligations and to increase the ownership of negotiated commitments.

The prevailing type of firm-government-WTO interaction is sketched out in Figure 1. A firm has two options in contesting foreign government policies that restrict its ability to contest a market: (i) petition the foreign government, either through direct lobbying and/or through its legal system if a WTO obligation has been violated (more on this below); and (ii) lobby its own government to take up the issue with the foreign government in question. From the firm's perspective going the WTO route often will not be attractive. First, it must convince its government that the case is worth bringing. This requires considerable resources--a collective action problem must be overcome to bring other firms in the industry on board. If the case is pursued, as discussed above, the outcome of dispute settlement may not do much to address the concerns of the firm. Thus, many violations are unlikely to be addressed through the WTO.

This is not necessarily bad. High "entry" barriers or thresholds may be beneficial in ensuring that only major cases are brought, ones that cannot be resolved through alternative, private mechanisms. It is difficult to make a judgement, however, because the available data on the prevalence and effect of discriminatory policies (whether or not these violate the WTO) are quite limited. Violations of the WTO often are not contested because firms and industry associations do not know what a government committed to or lack the incentive to do so. Company employees involved in sales, forwarding or intra-company trade that are most likely to know what government agencies are doing on the ground may not have an interest in documenting and reporting trade restrictions or discriminatory policies. Indeed, they will often find it difficult to determine whether a given measure is a WTO violation, given the difficulty in obtaining information on commitments.

There is therefore a collective action problem (Olson, 1965) that may lead to an inefficient outcome, both for individual firms and for the trading system as a whole. The problem is similar to that of the production of a public good, where without cooperation under-provision is likely to result. Solutions to the problem require increasing the benefits to firms of collecting data on potentially WTO illegal policies; and reducing the costs of doing so. The former can be pursued by establishing mechanisms to facilitate private sector cooperation, within and across countries to compile information; the latter can be pursued by devising mechanisms to facilitate identification of potential violations and reducing the need to involve governments in the enforcement of WTO commitments and increase reliance on domestic enforcement mechanisms.

One option to deal with the information problem is for the private sector to cooperate and to create mechanisms through which data on trade (and investment) barriers is collected and analyzed. This could be realized through periodic surveys of a representative sample of companies that is undertaken by an independent entity on a fee basis. The types of firms covered by the survey should encompass both multinationals that have plants in multiple countries and engage in substantial intra-firm trade, national firms that produce for export, trade associations, and consumer organizations. Different entities will have different interests, so it is important that a broad cross-section of industry and consumer interests from a wide range of countries are represented. Figure 2 provides a stylized illustration of a possible mechanism. Firms (and other interested groups in civil society) are periodically canvassed and the resulting information is compiled and organized by the surveying agency. The data can be used to assess the status quo on export markets, and help identify potential cases (more on this below). Data can also be transmitted to the government and to the TPR Department in the WTO.

As indicated in Figure 2, firms and other stakeholders have an interest in cooperating across countries as well. Ideally, an independent "transparency" body can be envisaged that would coordinate national efforts and assist in the identification of potential WTO violations, lobby for joint actions to be brought to the WTO, and generate publicity. This would help complement the TPRM mechanism by bringing in the private sector more directly into the enforcement process. It would not involve direct access to the WTO, however--national governments would continue to have sole access to WTO DSP. As mentioned earlier, incentives to collect information and pursue DSP depend importantly on the perceived payoff. More and better information can not only help identify potential WTO violations but also assist in overcoming unwillingness on the part of governments to pursue cases motivated by foreign policy reasons or by concerns about possible repercussions in other areas of cooperation (e.g., development aid flows). If firms from a large number of nationalities are involved in the information collection and analysis effort and a number of governments are petitioned to take (joint) action, an individual government’s incentive not to take up a complaint may be reduced.²²

Although at the margin better information may induce governments to be more pro-active, political realities--asymmetric distribution of power; threats of cross-issue linkages--will always be a powerful force constraining governments to assist firms to pursue their rights. This suggests that dispute settlement should be a collective endeavor not only for the reasons emphasized by economists (to increase the credibility of the threat of retaliation), but also to ensure that individual firms or governments do not end up bearing the brunt of an "offending" government’s displeasure. Clearly, governments cannot be obliged to cooperate, but business and civil society more generally can play a role in lobbying for joint actions to be brought. A global, private transparency body can help provide a focal point for such efforts.

Many fora already exist through which business cooperates and coordinates policy/lobbying positions vis-a-vis governments. The most prominent international body is the International Chamber of Commerce (ICC) located in Paris. The ICC has consultative status in the United Nations system, and is the most visible avenue for international business to express its views on international policy matters. Mention can also be made of the Business and Industry Advisory Committee, which provides business input into the deliberations of OECD bodies, and the Alliance for Global Business. In many countries there are similar institutions, as well as numerous industry-specific bodies. They are primarily focused on lobbying for changes in regulations or policies. Although business associations have become more aware of the potential payoff of investing resources to influence international negotiations to reduce barriers to trade and investment, they have done relatively little to compile the type of information and undertake the analysis that would help policymakers identify the key constraints to competition and provide an input into better enforcement of WTO agreements. Participation in the transparency and information collection effort should not be restricted to export-oriented industry groupings, but should extend to importer associations, consumer groups and other interest groups that have a stake in the functioning of the global trading system.

Analysis of the information that is compiled must be undertaken if it is to be useful for enforcement purposes. A problem associated with DSP is the need to assess the legality of an observed practice or situation. Information collection efforts must be complemented by creation of capacity to advise the private sector of the legality of observed trade practices. Advisory centers to do this could be public-private partnerships, and have a regional dimension--e.g., build on the institutions that are created in the context of creating and implementing regional integration agreements (Weston and Delich, 1999). Cooperation across countries can allow economies of scale to be captured and reduce the "unit costs" of analysis and information processing. Advisory centers that are geared towards the private sector could also help to give stakeholders a greater incentive to take an interest in the functioning of the multilateral trading system.

WTO rules and commitments are valuable in part because they provide assurances to investors in--and citizens of--a country that a given trade policy stance will be maintained. For the potential of instruments such as the TPRM and other transparency devices such as reporting and publication requirements, as well as the DSP, to be fully realized, it is necessary that domestic stakeholders can invoke WTO law and obligations in domestic courts or specialized fora. In many countries that is not the case. At most, they can contest actions of government entities through internal administrative appeals mechanisms.

When states enter into international treaty regimes what is guaranteed ipso facto is "external" accountability: i.e., they are accountable vis-à-vis the other signatories of the agreement (Art. 27 Vienna Convention on the Law of Treaties, pacta sunt servanda). The question of "internal" or national accountability vis-à-vis domestic citizens is a question of domestic constitutional law. Thus, there is a separation between the international and the national legal orders. The challenge is to break down this dichotomy, which greatly reduces the relevance of international treaty obligations to national constituencies. In principle, there are three options through which this can be done:

1. acknowledgement of the direct effect of international agreements;²³
2. drafting implementing legislation that gives private parties standing to pursue suits against their governments for non-compliance with international obligations; and
3. creation of "challenge" mechanisms that allow private parties to challenge WTO inconsistent behavior by government entities before domestic courts or specialized tribunals.

Traditionally, GATT rules have not had direct effect in national legal orders, including the US and the EC.²⁴ This has denied the real "users" of the multilateral system (firms, consumers) the opportunity to contribute to its development. Antitrust law, for example, has developed to a large extent because of invocation by private parties. As a result, law in this area has developed to reflect changing conditions. The process in the WTO has been much less direct. As argued by Tumlir (1985), this is a crucial weakness in the multilateral trading system, not just because of the weakness of available multilateral remedies, but more importantly, in terms of building a strong constituency that supports the development of multilateral rules guaranteeing nondiscriminatory access to markets.

Direct effect is a constitutional matter with serious implications for the legal system of a country generally, and is often linked to non-legal considerations (reciprocity). The second option may also be difficult for countries to adopt, as it may be difficult to restrict this to WTO agreements only. Also both of the first two options will require cases to be brought to domestic courts, which may not be efficient in providing for timely redress of disputes. The "challenge" option has the advantage of having more teeth than mechanisms which are limited to a transparency role, while being less far-reaching than the other two options. In some sense it is a halfway house to full direct effect (Davey, 1999). A first step in this direction was made at the WTO level in the (plurilateral) Agreement on Government Procurement (GPA). Without saying anything on the issue of direct effect, the agreement requires signatories to give private parties access to mechanisms that allow them to contest actions by procuring entities that violate the GPA. In effect, the "challenge" mechanisms required by the GPA imply that private parties may invoke its provisions before domestic courts (see Hoekman and Mavroidis, 1997).

A potential downside of generalizing the use of domestic challenge mechanisms to enforce WTO commitments is that it could give rise to diverging interpretations of WTO law, with domestic tribunals rulings being inconsistent with those of WTO panels. This can be avoided through the use of advisory opinions by the Appellate Body on the correct interpretation of particular terms. Another option would be to oblige national courts to defer to the WTO under conditions similar to those prescribed in Art. 177 Treaty of Rome. While formal agreement that WTO members will adopt challenge mechanisms would be beneficial, this is not required. Any government can decide to do this unilaterally.

A less far-reaching alternative (or complement) to enhancing the legal enforceability of WTO agreements in domestic fora is to create national institutions that have the mandate to monitor and contest government actions. Examples are a national ombudsman or a competition authority-cum-transparency agency. While such agencies may be prohibited from taking direct action against the government, they can facilitate debate on the magnitude and distribution of the benefits and costs of government policies, and help transmit information to affected parties, which can then be used in the political process.²⁵

WTO dispute settlement procedures are substantially stronger (and more "legalistic") than those in the GATT. Most observers believe that while certainly not perfect, the experience to date suggests the system is working quite well. Abstracting from the crucial "upstream" dimensions of enforcement (discussed above), the major issues from a developing country perspective concern the costs of DSP (both financial and in terms of human resources) and the relative weaknesses of the available remedies.

Financial and human resource constraints reduce the ability of a developing country government to participate in the multilateral dispute settlement process. Least developed countries in particular are at a disadvantage in bringing cases and defending their rights because of the absence of representation in Geneva and a severe scarcity of both financial resources and relevant expertise (Croome, 1998; South Centre, 1999). A noteworthy development in this connection has been the formation of a group of countries that has agreed to seek to establish an Advisory Centre on WTO Law with resources to finance assistance for dispute settlement cases on a cost-sharing basis (ACWL, 1999). Analogous to the role a "public defender" can play in domestic legal systems to ensure that all citizens are able to defend themselves, an international mechanism to subsidize the ability of poor nations to bring and defend cases can help to "level the playing field" to some extent. But, as discussed above, there is also an urgent need to develop homegrown resources, which requires resources for training and education. Many governments do not have the trade lawyers they need in order to maximize the benefits of WTO membership. Outside lawyers can help when there is case to be fought, but before this can be done there must be a capacity to determine if a case is worth bringing.

Many cases that go to the WTO involve relatively small trade volumes. One option to consider is to introduce ‘light’ dispute settlement procedures for such cases (e.g., where less than $1 million of exports is involved). In such cases, a single panelist should be appointed and the whole judicial review process be required to be completed within three months. This would be beneficial to developing countries as the bulk of litigation involving developing countries are "small" cases and the cost of litigation would be substantially lower (indeed, no lawyers may be needed). This would also be beneficial to the WTO system by reducing the burden that is put on panels. A related possibility is to create mechanisms that provide for rapid review of cases that are large in relative terms for the country or countries bringing the case (e.g., account for more than 3 percent of total exports). For developing countries a small absolute trade flow in dollar terms may represent a large proportion of its total exports.

A variety of remedies-related issues have been discussed in this paper. Clearly one aspect of this that deserves attention is to reduce the scope for "implementation avoidance" strategies of the type that have been alleged to have occurred in Bananas. Another concerns the type of remedies that panels are permitted (encouraged) to suggest. These issues are related in that the more specific a panel’s suggested remedies, the less scope there will be for disputes regarding whether a WTO member has complied with the DSB’s conclusions. Encouraging panels to make specific suggestions can do much to help reduce the uncertainty of the process.

A more important remedy-related issue concerns compensation and damages. Developing countries in particular can make a strong case that violations of the WTO are disproportionately burdensome for them given the fragility of many of their export industries and the fact that their export base is generally much less diversified than in high income countries. The absence of compensation for damages incurred is therefore more detrimental for developing countries. Financial compensation for damages incurred is unlikely to be politically acceptable to WTO members, although this is the preferable route to follow in principle. As noted earlier, in our view greater reliance by WTO members on re-negotiation of concessions would be beneficial in cases where a member finds it difficult to comply with a panel ruling. This would ensure that compensation is provided to affected parties and to all WTO members via the MFN principle. Of course, this is not useful for the negatively affected export interests that brought a complaint--who clearly are best served by direct compensation for damages incurred--but at least it would be superior to retaliation (counter-measures).

The fact that the majority of cases are settled informally can also be regarded as potentially problematical for developing countries. Settlements are essentially non-transparent. No rules dictate how they should occur, no control over the merits of settlement exists. The only obligation of WTO Members is to notify the outcome of the settlement under Art. 3.6 DSU and the ensuing obligation that all settlements must be "MFN-friendly" in accordance with 3.5 DSU. However, the notification record of 3.6 DSU settlements is very poor and does not allow a determination of whether 3.5 DSU has been complied with. Settlements occur behind closed doors with the WTO rarely finding out about them. In the absence of a proper discovery process, the WTO is helpless to deal with this non-compliance. Since enforcement of WTO obligations is decentralized--only Members can take action--affected developing countries are left with the burden of finding out what actually happened behind the closed doors. Lack of administrative resources generally guarantees that this will rarely occur. Moreover, the "closed door" nature of the process may facilitate the negotiation of settlements that are not in the best interest of the developing countries that are involved.

Non-violation disputes are a potentially important avenue for countries to contest WTO legal measures that have serious adverse consequences for their economies. Given the asymmetry in bargaining power between developing and high-income countries, WTO negotiations can result in outcomes where very high trade barriers or other forms of intervention (e.g., subsidies) are maintained that are detrimental to their export interests. Non-violation in principle allows such cases to be brought up, and can act as a valuable "discovery" device. The recent Kodak-Fuji litigation has had the effect of substantially reducing the scope for raising non-violation complaints, which may be relatively more detrimental to developing countries. However, the key problem in this area concerns available remedies--there essentially are none. This suggests that the cost of recent developments in this area are probably limited.

One of the innovations of the Uruguay Round was the creation of a two-tier dispute settlement mechanism. A standing Appellate Body was created to review the legal reasoning of panels at the request of parties involved in a dispute.²⁶ The creation of Appellate Body has had some negative implications: the cost of litigation has been substantially raised; the time required to resolve a dispute has been extended; The first is clearly not in the interest of developing countries. The second could work both ways. But since settlements do occur and most of the time are between parties of unequal power, developing countries will often find themselves not in a position to profit from the extended time at their disposal to implement their WTO obligations.

It can be questioned whether this two-tier process is in the best interests of developing countries. Hudec (1998) argues the Appellate Body should be transformed into "the" WTO court, and that the WTO Legal Office should take on the function of an Avocat General (of the type found in the EC and many national legal orders). This is likely to be beneficial for developing countries in terms of "leveling the playing field" by ensuring greater coherence and consistency, and reducing the diplomatic element of the panel process. Giving the WTO Secretariat the right to identify possible violations of WTO commitments in Trade Policy Reviews (TPRs) and the power to initiate "inquiries" into the legality of observed measures are other possibilities (Wolf, 1984). Giving such ex officio power to the Secretariat is likely to be resisted by many WTO members who are concerned that the Secretariat remain impartial. One way around this constraint--which is a valid one in our view--is to consider the creation of an independent "Special Prosecutor" that is given the mandate to identify WTO violations on behalf of developing countries. While this is unlikely to be an option that will be adopted in the near future, seeking to move in this direction deserves serious consideration by developing countries.

A final issue that is relevant for developing countries is the suggestion to increase transparency and strengthen the role of non-governmental organizations (NGOs) in DSP by giving them direct access to the WTO. Responding to criticism from NGOs, President Clinton has proposed ²⁷ that "hearings by the WTO be open to the public, and all briefs by the parties be made publicly available" and that "the WTO provide the opportunity for stakeholders to convey their views ... to help inform the panels in their deliberations" (Croome, 1998). Greater transparency of WTO processes would be beneficial--although it must be noted that great strides have been made in this regard in the last five years: the situation under the WTO is an order of magnitude more transparent than was the case under the GATT.²⁸ However, giving NGOs direct access to the WTO DSP is unlikely to improve the DSP and can be expected to meet vigorous opposition on the part of developing countries. While stakeholders should have access to mechanisms that allow them to have a voice in the development of national positions in WTO negotiations and allow them to invoke WTO law at the national level, a good case can be made that this should not extend to direct access to the WTO. In addition to the considerations identified by Levy and Srinivasan (1996) (discussed above), doing so could lead to a situation where NGOs have rights that are stronger than those accorded to WTO members (South Centre, 1999). More fundamentally, it will be very difficult to determine who speaks for (is representative of) global civil society as a whole. Wolf (1999) argues that organizations (NGOs) "can only represent themselves. If NGOs were indeed representative of the wishes and desires of the electorate, those who embrace their ideas would be in power. Self-evidently, they are not." Problems of representativeness are obviously compounded at the global (WTO) level.

The incentive for governments to negotiate and abide by international trade agreements depends in part on the effectiveness of enforcement provisions. Enforcement is particularly important for developing countries, as these nations will rarely be able to exert credible threats against large trading entities that do not abide by the negotiated rules of the game. In this paper we have discussed many aspects of the enforcement problem. Our major conclusions can be summarized as follows.

• Domestic enforcement is a vital dimension in enhancing the relevance of multilateral commitments to domestic stakeholders (importers, exporters, consumer groups). In most countries, including high-income nations, domestic interests are restricted in their ability to contest actions by national government agencies that violate WTO commitments. Civil society has a strong interest in seeking to maximize the extent to which individuals can invoke international treaty obligations in national legal systems. This will remove a number of layers of uncertainty and complexity associated with bringing cases to the WTO. Strengthening national enforcement mechanisms can help make the WTO a more relevant instrument from an economic development perspective by increasing the "ownership" of negotiated commitments. It also relaxes the constraint of having to convince one’s government to bring a case to the WTO and will reduce the burden of DSP at the WTO level. The easiest way of making WTO commitments enforceable nationally is to expand on the "challenge" mechanisms that have been introduced in the GPA.
• The private sector must play a much greater role in enforcement. In part this can be achieved by designing domestic legal mechanisms that increase the incentive for them to collect, compile and transmit information on the measures that are being applied by governments, both their own and foreign. Thus, there is a close link to our first point--giving private interests standing in domestic fora (via a general "challenge" procedure) can be expected to be a great motivator for greater involvement in both the development and enforcement of multilateral disciplines. But greater private sector participation is also vital in order to ensure that developing countries can defend their rights through DSP at the WTO level. This "upstream" dimension of DSP at the WTO is as important as the efficacy of the "downstream" panel and Appellate Body process. Greater efforts to disseminate information and data can also be made by WTO members, starting with TPR reports and the various databases that are maintained by the Secretariat (Francois, 1999).
• Although greater attention is required to improve and augment domestic dimensions of enforcement, efforts can also usefully be made to make the WTO "easier" to use by developing countries. A key problem here that has become widely recognized is that the process is costly and human-resource intensive. Developing countries are at a disadvantage, given their limited resource base. Giving consideration to changing the DSU to make it easier and more efficient to bring cases where the absolute value of the trade flows concerned is small, or the relative importance of the affected trade for the plaintiff is large could help developing countries increase their use of DSP. Similarly, providing legal assistance along the lines that has been proposed by a number of WTO members (ACWL, 1999) would clearly be beneficial to developing countries. Finally, consideration could be given to the creation of an independent "Special Prosecutor" that is given the mandate to identify WTO violations that are detrimental to developing countries.

Advisory Centre on WTO Law. 1999. Final Proposal Document (www.itd.org).

Benson, Bruce. 1989. "The Spontaneous Evolution of Commercial Law," Southern Economic Journal, 55:644-61.

Bhagwati, Jadgish. 1990. "Aggressive Unilateralism: An Overview," in J. Bhagwati and H. Patrick (eds.), Aggressive Unilateralism: America’s Trade policy and the World Trading System. New York: Harvester-Wheatsheaf.

Bronckers, Marco. 1996. "Private Participation in the Enforcement of WTO Law: The New EC Trade Barriers Regulation," Common Market Law Review, 33:299-318.

Croome, John. 1998. "Trade Negotiations in the WTO: The Present Outlook," World Bank Policy Research Working Paper (www.worldbank.org/trade).

Croome, John. 1999. Reshaping the World Trading System: A History of the Uruguay Round. The Hague: Kluwer Law International.

Davey, William. 1999. "Roundtable on the role of adjudication in the WTO," Journal of World Arbitration, April.

European Communities. 1999. "Developing Countries and the WTO Dispute Settlement Mechanism," WT/GC/W/148, February 24.

Finger, J. Michael. 1982. "Incorporating the Gains from Trade into Policy," The World Economy, 5:367-77.

Finger, J. Michael and Philip Schuler. 1999. "Implementation of Uruguay Round Commitments: The Development Challenge," World Bank, mimeo (www.worldbank.org/trade).

Francois, Joseph. 1999. "Maximizing the Benefits of the Trade Policy Review Mechanism for Developing Countries," mimeo (www.worldbank.org/trade).

Francois, Joseph and Will Martin. 1999. "Commercial Policy Uncertainty, Bindings, and Market Access," Tinbergen Institute, mimeo (www.worldbank.org/trade).

Gilligan, Michael. 1997. Empowering Exporters: Reciprocity, Delegation and Collective Action in American Trade Policy. Ann Arbor: University of Michigan Press.

Greif, Avner. 1993. "Contract Enforceability and Economic Institutions in Early Trade: The Maghribi Traders Coalition," American Economic Review, 83:857-82.

Greif, Avner, Paul Milgrom and Barry Weingast. 1994. "Coordination, Commitment and Enforcement: The Case of the Merchant Guild," Journal of Political Economy, 101:745-76.

Hilf, Meinhard and Ulrich Petersmann (eds.). 1993. National Constitutions and International Economic Law. Deventer: Kluwer.

Hoekman, Bernard. 1993. "Multilateral Trade Negotiations and Coordination of Commercial Policies," in R. Stern (ed.), The Multilateral Trading System: Analysis and Options for Change. Ann Arbor: University of Michigan Press.

Hoekman, Bernard. 1997. "International Business and the World Trading System: Strengthening Private Enforcement of Public Commitments," presented at the conference The Manager in the Trading System, University of Neuchatel, May 20.

Hoekman, Bernard and Petros C. Mavroidis. 1996. "Policy Externalities and High-Tech Rivalry: Competition and Multilateral Cooperation Beyond the WTO," Leiden Journal of International Law, 9:273-318.

Hoekman, Bernard and Petros C. Mavroidis (eds.). 1997. Law and Policy in Public Purchasing: The WTO Agreement on Government Procurement. Ann Arbor: University of Michigan Press.

Horn, Henrik and Petros C. Mavroidis. 1999. "Remedies in the WTO Dispute Settlement System and Developing Country Interests," World Bank Institute (www.worldbank.org/trade).

Horn, Henrik, Petros C. Mavroidis and Håkan Nordström. 1999. "Equity in the WTO Dispute Settlement System: Participation," mimeo.

Hudec, Robert. 1987. Developing Countries in the GATT Legal System. Aldershot: Gower Press for the Trade Policy Research Centre.

Hudec, Robert. 1993. Enforcing International Trade Law. Butterworth: New Hampshire.

Hudec, Robert. 1998. "The Role of the GATT Secretariat in the Evolution of the WTO Dispute Settlement Procedure," in Jagdish Bhagwati and Mathias Hirsch (eds.), The Uruguay Round and Beyond: Essays in Honor of Arthur Dunkel. Ann Arbor: University of Michigan Press.

Hungerford, Thomas. 1991. "GATT: A Cooperative Equilibrium in a Non-Cooperative Trading Regime?" Journal of International Economics, 31:357-369.

Jackson, John H. 1997. The World Trading System. Cambridge: MIT Press.

Jackson, John H., William Davey and Alan Sykes. 1995. Legal Problems of International Economic Relations: Cases, Materials and Text. West Publishing Company.

Kovenock, Dan and Marie Thursby. 1992. "GATT, Dispute Settlement and Cooperation," Economics and Politics, 4:151-170.

Levy, Philip and T.N. Srinivasan. 1996. "Regionalism and the (Dis)advantage of Dispute Settlement Access," American Economic Review (Papers and Proceedings), May.

Matsushita, Mitsuo, Petros C. Mavroidis, and Thomas Schoenbaum. 2000. International Economic Law. Oxford: Oxford University Press, forthcoming.

Mavroidis, Petros C. and Werner Zdouc. 1998. "Legal Means to Protect Private Parties’ Interests in the WTO: The Case of the EC’s New Trade Barriers Regulation," presented at the conference The Manager in the Trading System, University of Neuchatel, May 20.

Michalopoulos, Costas. 1999. "The Developing Countries in the WTO," World Economy, January.

Milgrom, Paul Douglass North and Barry Weingast. 1990. "The Role of Institutions in the Revival of Trade: The Law Merchant, Private Judges, and the Champagne Fairs," Economics and Politics, 2:1-23.

Olson, Mancur. 1965. The Logic of Collective Action. Cambridge: Harvard University Press.

Palmeter, David and Petros C. Mavroidis. 1999. Dispute Settlement in the World Trade Organization: Practice and Procedure. The Hague: Kluwer Law International.

Petersmann, Uli. 1998. "How to Constitutionalize International Law and Foreign Policy for the Benefit of Civil Society," Michigan Journal of International Law, 20-1-30.

Roessler, Frieder. 1985. "The Scope, Limits and Function of the GATT Legal System," The World Economy, 8:287-98.

Schott, Jeffrey and J. Buurman. 1994. The Uruguay Round. Washington DC: Institute of International Economics.

South Centre. 1999. "Issues Regarding the Review of the WTO Dispute Settlement Mechanism," Working Paper No. 1 (February).

Tumlir, Jan. 1985. Protectionism: Trade Policy in Democratic Societies. Washington D.C.: American Enterprise Institute.

Vermulst, Edwin, Petros C. Mavroidis and Paul Waer. 1999. "The Functioning of the Appellate Body After Four Years--Towards Rule Integrity," Journal of World Trade, 33(2):1-50.

Weston, Ann and Valentina Delich. 1999. "Settling Trade Disputes After the Uruguay Round: Options for the Western Hemisphere," Latin American Trade Network Discussion Paper (June) (downloadable from "other WTO 2000 links" on www.worldbank.org/trade).

Whalley, John, 1996. "Developing Countries and System Strengthening in the Uruguay Round," in W. Martin and L. Alan Winters (eds.), The Uruguay Round and the Developing Countries. Cambridge: Cambridge University Press.

Wolf, Martin. 1984. "Two-Edged Sword: Demands of Developing Countries and the Trading System," in J. Bhagwati and G. Ruggie (eds.), Power, Passions and Purpose: Prospects for North South Negotiations. Cambridge: MIT Press.

Wolf, Martin. 1999. "Uncivil Society" Financial Times, September 1, p. 12.

Note: Excludes third parties.
Source: Horn and Mavroidis (1999).

* We are grateful to Mike Finger and David Palmeter for helpful comments and discussions. The views expressed are personal and should not be attributed to the World Bank.

1 The practice and procedures followed by the WTO in this area are discussed in detail in Palmeter and Mavroidis (1999). For a brief introduction into the rules and the mechanics of the process, see the WTO home page: www.wto.org.

2 Precise numbers in this area are difficult to obtain. Hudec (1993) analyzes 207 formal GATT cases, whereas a database cited by Jackson (1997) lists some 400 disputes. Using the higher number gives an average of 10 disputes per year under GATT, compared to 30 under the WTO. As of September 1, 1999, 179 requests for consultations had been notified to the WTO--averaging out to some 3 every month (WTO home page).

3 Other cases between developing countries were between Brazil and the Philippines (desiccated coconut); Guatemala and Mexico (antidumping actions on cement), and India and Turkey (textiles). Information on disputes can be obtained from the WTO web site (www.wto.org).

4 Readers interested in a comprehensive positive analysis of the determinants of use of WTO dispute settlement to date are referred to Horn, Mavroidis and Nordstrom (1999).

5 See also Roessler (1985), Petersmann (1998), and the contributions in Hilf and Petersmann (1993).

6 Throughout this paper we assume that WTO rules are good ones, i.e., we assume that WTO agreements are Pareto improving deals and that countries therefore have an incentive to enforce them. Clearly one can question this in the case of some specific agreements, and one must consider the cost of implementation in all cases. Some of these issues are discussed in Finger and Schuler (1999).

7 And is another mechanism itself for generating relevant information. For formal analyses of the importance of the information role of DSP, see Hungerford (1991) and Kovenock and Thursby (1992).

8 Reimbursement of duties paid--i.e., financial compensation--has only been recommended in the case of a few antidumping related cases prior to the creation of the WTO. None of these panel decisions were ever adopted by the GATT. See Palmeter and Mavroidis (1999).

9 The problem of "implementation avoidance" was effectively circumvented in the Bananas dispute, where a legal oddity occurred: the same panel was simultaneously examining a request by the United States to impose countermeasures in case the EC action was found not to be in compliance with the WTO and another request by the EC to the effect that its action was in compliance. Logically and legally only if the latter has been established can the former be examined. This is the unequivocal role that has been assigned to Art. 21.5 DSU. The panel report effectively reads 21.5 DSU out. The panel report examining the US request came out on April 9, 1999 whereas the report examining the adequacy of the EC action came out three days later. It is probably safe to speculate that this will remain a one-off incident in the annals of the WTO, mandated by reasons of political expediency rather than the letter of law. However, in order to assure that this remains a one-off event, legislative amendment is necessary.

10 An exception is Kovenock and Thursby (1992). An interesting literature has emerged in the last decade exploring the emergence and maintenance of legal norms in the absence of central authority. A conclusion that emerges from these studies is that the threat of ostracizing a "member of a club" who is reliant on repeated interaction with other members can have a powerful impact as an enforcement device. See, e.g., Benson (1989), Milgrom, North and Weingast (1990), Greif (1993), and Greif, Milgrom and Weingast (1994).

11 To some extent a basis for this may already exist. Art. 21.8 DSU states that in cases brought by developing countries, the DSB "shall take into account not only the trade coverage of measures complained of, but also their impact on the economy of developing country Members concerned."

12 For a recent proposal along these lines on behalf of developing countries, see South Centre (1999).

13 Compensation is an option in the DSU, but only as an interim measure if panel recommendations are not implemented within a reasonable period of time (Art. 22.1 DSU).

14 Of course, it also relies heavily on export interests in negotiating trade agreements (e.g., Gilligan, 1997). Large traders such as the EC and the US have developed legal instruments through which the private sector can petition the government to take action against foreign government regulations that restrict market access and/or violate multilateral commitments. Section 301 in the US is the most famous example. The EC has a similar instrument, the Trade Barriers Regulation. See Bronckers (1996) and Mavroidis and Zdouc (1998) for discussions on the EC; Bhagwati (1990) on the US.

15 The larger traders are reviewed every 2 or 4 years depending on size. It can be argued that the frequency of surveillance is the inverse of what is required, given that export interests have the greatest incentives to monitor the policies that are applied by the largest trading countries. The counter argument is that high-income nations obtain relevant information from their commercial attaches around the world and their private sector, whereas developing country networks of commercial attaches are much more limited, and their private sector export interests will tend to be concentrated in fewer markets. Thus, the payoff to multilateral surveillance of major markets may be more important for them.

16 E.g., the National Trade Barriers Report (USTR) or the annual Report on U.S. Barriers to Trade and Investment prepared by the European Commission (DG-I).

17 The experience with privatization of trade policy in the area of antidumping--where governments cannot on political or national interest grounds decide not to intervene even if the technical criteria (injurious dumping) have been found to be satisfied by investigating authorities--also suggests that governments should keep control of their trade policy.

18 This is also important in light of efforts by special interests with non-trade concerns to use trade agreements as a vehicle to pursue their aims. The foregoing argument regarding the need for government to keep control of the dispute settlement process applies equally to efforts of single issue NGOs to introduce new rules on non-trade issues in the WTO.

19 What follows draws on submissions by developing countries as part of the review of the DSU, as well as Croome (1998) and South Centre (1999).

20 Until Bananas III countries also were impeded from bringing non-government, private legal counsel before the panel, but an Appellate Body decision to allow representation by private lawyers removed this constraint as far as the Appellate Body was concerned. A subsequent panel then decided there were no provisions in the WTO or the DSU that prevented a WTO member from determining the composition of its delegation to panel meetings (Palmeter and Mavroidis, 1999).

21 The EU has proposed to enhance the Secretariat’s ability to provide to provide case-specific technical assistance by establishing an independent unit to provide advice on the likely strength of a potential case. The unit would be prohibited from representing WTO members in panels, however, in order to maintain impartiality (EC, 1999).

22 See Hoekman (1997) for more detailed discussion on the design aspects of a possible survey instrument. There are a variety of free rider problems that must be overcome, and in practice what is required is the formation of a "privileged group" of large firms and trade/industry associations that is willing the bear the costs of cooperation and tolerate free riding by smaller firms.

23 Direct effect allows private litigants to raise relevant points of WTO (international public) law before national courts. This is a matter that is determined by domestic constitutional law. Whether direct effect applies is not dependent on whether or not implementing legislation is needed, or whether international treaties enter into force automatically. In both cases, domestic courts retain the discretion to decide whether a particular norm has direct effect.

24 The status of GATT law in both sides of Atlantic has raised questions regarding the seriousness of both the US and EC when it comes to implementing their international obligations. In the Suramerica litigation (Jackson, Davey, and Sykes, 1995) a US court of appeals essentially accepted that a subsequent domestic statute prevails over GATT law. In the Bananas litigation, the European Court of Justice (ECJ), responding to a complaint by Germany, refused to examine the compatibility of the EC bananas regime with its WTO obligations--in effect denying that Germany has standing. Some hope was restored in the EC with the recent Hermes case, which on an optimistic reading appears to be a promise by the ECJ to start to interpret EC trade policy in the light of WTO obligations. Direct effect is a far cry for both transatlantic partners: the immediate priority is to ensure that the WTO is taken seriously through faithful implementation at the domestic level.

25 For an early proposal along these lines, see Finger (1982).

26 Space constraints preclude a detailed analysis of the functioning of the Appellate Body. See Vermulst, Mavroidis and Waer (1999) for a discussion.

27 Speech by President Clinton at WTO Ministerial Conference, 18 May 1998.

28 The best illustration of this is the WTO home page, which provides access to most documentation that is prepared by and submitted to the WTO. But much more can be done. As noted by Francois (1999), much of the data generated by the WTO is not freely available, significantly reducing the transparency role of the WTO, and thus the potential value of the commitments made by members as an uncertainty reducing device (on the latter, see Francois and Martin, 1999).