The WTO's Doha Round, which began in late 2001, has been effectively deadlocked on most of its issues for years. The long stalemate has caused many to think our prominent international institution is gradually losing its relevance as a venue for writing new rules for world trade, in competition with preferential trade agreements and bilateral investment treaties. At their December 2013 conference in Bali, WTO ministers finally reached an agreement on at least a small package of partial steps, which gave multilateralists some encouragement, although the decision by India to block adoption of the Protocol to embody the new Trade Facilitation Agreement into the WTO makes it unclear at the time of writing whether even this limited achievement will be implemented. Even if the Bali decisions are all implemented, they will have only a small impact on world GDP and will not resolve most of the Round's issues.
Many of us would prefer to see the WTO's relevance revived and re-energized. One possible means would be to launch another new comprehensive round of negotiations after the Doha Round ends. This article identifies pitfalls to avoid and strategies for success in such an endeavor. How should Members set the agenda next time, if there is a next time? The central argument is that they would have to conduct another complex agenda negotiation like those that culminated in Punta del Este, Seattle, and Doha, and I suggest steps that might increase the odds of success.
Two caveats: First, this paper is not designed to forecast what will happen in the DDA or defend particular ways to end it. Under one scenario, governments eventually reach a comprehensive agreement that ends the round. This looks unlikely at the time of writing. In a second scenario, the parties simply agree to terminate these talks under the 2001 mandate. Consensus would be difficult to reach, however, without something to replace it. For one thing, the many Members demanding changes to bring farm trade more in line with the general rules would be difficult to satisfy with a proposal to simply stop trying.
In a third scenario, the Members decide to end the DDA and the 2001 mandate by replacing both with a new round and an agenda that incorporates unfinished Doha issues along with new issues that have arisen since 2001.Footnote 1 Major Doha issues include agriculture and food security, non-agricultural market access, services, electronic commerce, and special and differential treatment. A list of new issues has been proposed for updating the WTO: multilateralizing regionalism; plurilateral agreements (whether to authorize specified new agreements conditional on a new multilateral framework to safeguard interests of non-signatories and the WTO); accommodating green energy and climate policies in the trading system; currency misalignment; WTO institutional reforms such as authorizing the Director General to make proposals in the common interest; and others.Footnote 2 Rather than attempting to evaluate these alternative scenarios, this article assumes Doha ends in some way, in order to concentrate instead on what could come after.
This paper also does not defend the claim that attempting to launch another comprehensive round would be the best path forward. Wolfe (Reference Wolfe2015) documents the distributional struggles among the big players that continue to stall the WTO. Given the talks underway about mega-regional preferential deals – a possible Trans-Pacific Partnership, a Transatlantic Trade and Investment Partnership, an EU–Japan FTA, and others – now underway among some of the largest economies, WTO negotiators may prefer to wait to learn what if anything these large players do before launching another WTO round. Meanwhile, other paths besides a new GATT-style round are conceivable. Members could cooperate on a variety of issues through the WTO's bodies. They can be sites for deliberation as well as legislation (Evenett and Jara, Reference Evenett and Jara2013). Examples are joint studies of new negotiating techniques and rules being developed in PTAs (Messerlin, Reference Messerlin2012), and joint study of regulatory differences across countries, asking whether the WTO can add value toward regulatory convergence (Hoekman, Reference Hoekman and Mavroidis2012b). Deep PTAs are writing new rules on such issues in response to business demand.
But suppose some advocates are optimistic and feel it would be a good idea to attempt to launch a new WTO round. How should they go about it? The next section introduces selected insights from scholarship about international negotiation and cooperation, highlighting key factors that will narrow the range of likely outcomes in an agenda negotiation and on which it would be useful to collect evidence and make judgments. Subsequent sections make several recommendations for the advocate of a new round.
Conceptual building blocks
A global public good and a changed distribution of power and ideology
The WTO is a global public good to a significant degree, which means that keeping the institution up to date with changing conditions requires that some Members must pay the institutional costs despite an incentive to take a free ride. Part of the value of this public good arrives in the tangible national gains in economic welfare that the Member state experiences from trade agreements negotiated under its umbrella. Another part takes the form of less tangible contributions. Economists have argued that investors gain certainty from decisions by many governments to bind themselves not to increase their protection above ceiling levels. Governments respect these obligations much of the time because they believe doing so strengthens their economies, they know many other states are respecting theirs to the benefit of all, and because they can face an adverse legal ruling in Geneva otherwise. The WTO's dispute settlement institution also helps Members shift their trade disputes out of the arena of brute force politics into an arena where politics still operates but under constraints of law. The large majority of small- and medium-sized trading states would be in much weaker positions in a world without this relatively strong multilateral institution built on the norm of non-discrimination, despite its flaws. Putting reliable numbers on intangible benefits is difficult, but trade officials and authors from many countries have affirmed repeatedly that they are real.
One of the costs of maintaining this global public good is willingness during negotiations to make the concessions needed to seal deals that will keep the WTO from losing its relevance. In the unusual early decades after World War II, the capitalist world had a K group, a small set of players, dominated by the USA and the European Community, that were large enough economically to believe they could capture much of the benefit of the GATT, and so believed it was in their self-interest to create it and pay the costs, disregarding free riders.Footnote 3 It is possible that another reason for their leadership was their leaders' belief that it would strengthen their military alliance against the Soviet Union. Reflecting on their origins, therefore, GATT/WTO rules embody the preferences of the ‘stronger’ much better than those of the ‘weaker’.
Today the WTO no longer has a K group. More than two dozen additional states have joined since 1994, including China and Russia. China and India expanded to become two of the five largest Member economies (counting the EU as one Member), and today Russia and Brazil are immediately behind them in the top ten Members by GDP. And while there was always some ideological diversity in the Quad countries (US, EC, Japan, and Canada), this power dispersion has elevated players that are less dedicated to economic liberalism than the USA, at least. In addition, after 1995 many middle-income countries greatly improved their technical preparation for WTO negotiations, and even the weakest Members now organize and operate as coalitions to a much greater degree (Odell, Reference Odell2010). Today's different set of big players has been unable during the Doha round to overcome concerns about the distribution of benefits and costs among themselves to pay this cost for their common institution.
At the same time, international negotiation research has documented many cases in which the power structure leaves much variance unexplained. Political outcomes have varied within the same power structure, depending in part on how parties negotiated (Odell, Reference Odell, Carlsnaes, Risse and Simmons2012 and references there). Most fundamentally, this is because power alone does not tell us the particular goals that a state will choose to seek with its power. States are often divided over the path to follow in foreign policy, especially on trade issues. How governments perceive their alternatives is also subject to domestic and international persuasion to some degree, and change sometimes takes years.
The huge information problem facing a proponent
Another general point is that the complexity of launching a multilateral trade round can be immense, which generates serious information problems for any planner in even the most developed countries. Before launching an attempt, the rational advocate will first ask: is there any chance of getting an agreement on a new agenda, and which agenda? Proponents would benefit from having economic forecasts showing baseline trends in GDP (as one indicator of alternatives to agreement). Regarding a given issue, they would benefit from analyses estimating how potential agreements would affect each major national economy or group. Regarding potential sets of issues, which sets offer the prospect of trade-offs to attract sufficient support or acquiescence from all major groups of countries? Also desirable would be political information such as how domestic politics will affect delegations and eventual ratification decisions, delegations' likely true minima on a variety of issues at least at the outset, how delegations will react to proposals and to one another, and whether likely coalitions of states will hold together or fragment in the end.
Today we have plenty of information about governments' and coalitions' positions on Doha issues, which will be relevant to the extent that a new round could attempt to build on that work. We have much less information on proposed new issues or, especially, analysis of how governments would react to new alternative issue packages.
The perceived alternative to agreement
An especially crucial type of information for shaping negotiator behavior will be how the parties evaluate their alternatives to a new WTO agenda agreement. In order to accept the proposal, Members would need to view any WTO agenda proposal as having the potential to deliver gains relative to their alternative courses of action. Here the state's or group's reservation value in the talks might be thought of as the country's GDP growth rate and trade growth rate absent a new round, as a first approximation. More central for behavior is how leaders and negotiators perceive their alternatives. The Member state's resistance point or ‘bottom line’ is its reservation value plus or minus a subjective component, and the resistance point may change during and because of the negotiation. Some negotiators settle above or below their material reservation values (White and Neale, Reference White and Neale1991). For instance, a developing country excluded from the green room in 1999 views the WTO process in Geneva and Seattle as unfair, sets its resistance point below its material reservation value, and chooses impasse until its demand for fair process is satisfied. Or a delegation may factor in the intangible value the country places on the WTO and what the failure of a round would mean for the institution, and thus accept commercial costs somewhat above what it would prefer on narrow material terms. Or an unexpected proposal could rearrange the issue space and require rethinking the national interest. The existence of this subjective level opens up possibilities for persuasion and learning to move states toward or away from agreement. It is also another locus of uncertainty, which can allow miscalculation. In 2000, EU leaders evidently hoped they could persuade developing countries that a new behind-the-border agreement on investment policy would be a gain for them, but persuasion attempts failed in many countries and the EU dropped this proposal in 2004.
While the set of the parties' resistance points is a theoretical idea that can be useful for directing research and anticipating what will happen, these points are extremely difficult to measure in practice from outside delegations. It will be difficult enough to project, with any precision, the material effects of possible agreements on trade, welfare, and the environment to compare with a baseline, and still harder to estimate the changeable subjective elements on a uniform basis in advance. But some approximations are discussed below.
Multiple issues but not deal breakers
A familiar formula for generating a joint-gain agreement is to include multiple issues that can be linked into complementary trade-offs. If the EU and Japan are going to have to give largely one-sided concessions in agriculture, they need to gain concessions on other issues to go home with a deal that exceeds their resistance point and can be ratified, and likewise with other Members.
Hoekman (Reference Hoekman2014) proposes a new way to increase the productivity of issue linkages. The new round could add a parallel process for examining barriers to supply chain trade. An increasing share of world trade is taking place inside such production networks, each spanning several countries; think autos and electronics. Some developing countries follow strategies of attracting foreign investment and inserting their economies into the global economy through these channels. More small and medium enterprises and low-income countries might be candidates to join this game in the future. But supply chains face costs due to not only border measures but also domestic monopolies and regulations in services, intellectual property, local content requirements, environmental conservation, and other areas. WTO negotiators are accustomed to discussing each of these policy issues separately, and they may not be aware of how the issues in different policy silos jointly affect a supply chain. The proposal is not to negotiate ‘supply chain agreements’ or to include this topic as a formal agenda item, but to create several bodies for parallel joint study of the problems of a few representative supply chains. Business participation in each forum could improve transparency and understanding about which policy changes would generate how much gain and loss for whom. This improved information might enable negotiators to find more valuable joint-gain exchanges of concessions across issues.
If any issue seems likely to stimulate a significant blocking coalition, adding it to the agenda may of course prove counter-productive. In the early 2000s, the proposed new WTO commitments concerning behind-the-border polices on investment, competition, and labor rights were deal breakers for many developing countries, and they said so often and intensely. That the EU insisted so long on these agenda items, despite this vigorous opposition, undermined confidence that the process truly would be sensitive to the wishes of developing countries. Similarly, in 2001 binding rules assuring developing countries of concrete special and differential treatment (SDT) probably were a deal breaker for developed countries. They did agree to put SDT on the Doha agenda, in order to gain developing country support for launching the round. But their analysis told them these rules have not promoted economic development, and developed countries were likely to block more than minor changes in this area right from the beginning.
In practice, including unpopular issues on a future agenda could nevertheless prove necessary to achieve agreement to launch a round. States and non-state actors often want to raise issues that seem less-than-ripe in the hope of changing minds, or failing that to raise consciousness for the future after the round. And if such a divisive issue jeopardized the rest of the agenda again, states could agree on a shallow pact obliging them only to continue joint study of the divisive issue, or drop it entirely from the talks.
A neutral negotiation analysis
One practical preliminary step would be to address the information problem by asking a set of universities and think tanks to produce a neutral political-economic analysis to help delegations think through this complex problem of choosing and negotiating a set of agenda items. During the 1970s, a team at the Massachusetts Institute of Technology (MIT) estimated the costs and benefits of a future deep sea mining enterprise, something that did not exist at the time. The chair of the Law of the Sea negotiating group on financing informed delegations about this model. Although it had no formal status in the talks, the MIT model reduced uncertainty and some states changed their negotiating positions because of what they learned from it (Antrim and Sebenius, Reference Antrim, Sebenius, Bercovitch and Rubin1992). In the past, trade economists have projected the global benefits of possible GATT and WTO agreements.
This research team could begin by assembling a list of candidate agenda items nominated by others – from Doha, recent statements by governments, interest groups, NGOs, and other analysts – plus ideas of their own. The team could study say the largest ten to 20 WTO Members by share of world GDP or trade. For Member A, the team would project likely gains and losses from each item on the agenda list except where it is impossible to model the effects or find appropriate data. The team could also collect evidence on the attitudes of the government and key constituent groups toward each agenda item, using public statements and confidential interviews in the capital. It would be especially important to consult business groups and others likely to take an active part in the domestic politics surrounding the possible round. What issues do constituency groups most want addressed (and not addressed) today?
The team would make rough judgments whether Member A will see itself as probably gaining, losing, or both on issue 1 by itself, issue 2 by itself, and so on. The relative salience of the issues for A should also be ranked roughly (high, medium, low) even though these can change, since exchanges of concessions on issues of differential salience are a primary way to create joint gains. The team would estimate A's perception of its alternatives to another WTO round and how eager or reluctant it is to participate relative to its alternatives.
In addition, it would be desirable to identify groups of mid-sized and small Members and estimate analogous judgments for each group. A practical approach could select sets that revealed their preferences by organizing bargaining coalitions during the Doha Round and negotiating jointly on a sustained basis, such as the African Group, the Group of 33 on agriculture, the Least Developed Countries, and the Small and Vulnerable Economies. For new issues, different sets might be needed.
The team should certainly consider the Members' collective interests as well. Projections of gains or losses in world welfare, studies of how issues could affect the natural environment, and analysis of how individual items and sets of items could affect WTO law, dispute settlement, and relations with other international organizations should be included.
Finally, the team would consider packages of agenda items, beginning with the smallest potential packages. It would ask whether any package seems promising relative to the Members' views of their alternatives, and hence might be politically feasible. This would entail checking whether each country's or group's perceived losses on some issues are sufficiently offset by perceived gains on other issues. If more than one package seems feasible, the possible trade-offs of expanding the agenda should be illuminated, and it would be best to rank the feasible packages according to their potential for (net) joint gain. Naturally, some assumptions would need to be made to reach these conclusions. It would be more valuable to have a comprehensive analysis that combines the best possible quantitative and qualitative analysis and judgment with appropriate caveats, than to reduce the study to only conclusions that can be demonstrated quantitatively.
Bearing in mind the WTO's Member-driven norm, this independent analysis, like the MIT model, would have no formal status. The report should be financed in a way that reduces complaints that the financing biased the conclusions.
GATT and WTO states have never commissioned such a large negotiation analysis in advance. But this global organization and the task of managing its negotiations have become even more complex than before. An innovation like this might help the Members avoid problems that have troubled them in past agenda negotiations. The information and insight might save far more than they would cost. Of course, Members would be free to conduct their own analyses as well.
If today's pessimists are right, such research will show that no agenda for a comprehensive new round is sufficiently promising, and if governments then decide not to proceed down this path, they would avoid the costs of another lengthy effort yielding a small or no payoff. If today's optimists are confirmed, the preliminary analysis will provide careful estimates to guide Members in designing a round with good prospects.
Procedural rules
The procedural rules in effect will also influence whether they reach an agenda agreement.
The consensus norm
During the WTO's first decade, the Members, large and small, strongly reaffirmed their commitment to the traditional GATT norm that decisions will be made by consensus of the entire membership, defined as the absence of formal objection by any Member. While many have been frustrated by the inefficiency of this norm and some respected experts have recommended consideration of formally weighted voting, a decision to make that change – such that a Member (e.g. USA or India) that dissented on a question could then be outvoted by a majority – could itself be adopted only by consensus. These ideas are not close to achieving consensus among the Members. Today there is no practical alternative to operating by consensus, for setting the agenda and approving the round's final act. Better to look for supplementary rules that reduce incentives for states to block consensus.
Variable geometry
To reduce the risk of re-enacting Doha frustrations, consideration should be given to agreeing that not all Members would be required to sign every component pact.Footnote 4 Some or all pacts can be critical-mass agreements (CMAs) with unconditional MFN treatment or plurilateral agreements (PAs) with conditional MFN treatment. Every WTO Member would be invited to participate in each club negotiation, and each could sign or not sign each resulting pact. This type of club agreement could not be used to change provisions of a WTO agreement previously adopted by the whole Membership by consensus (Warwick, 2007; Gallagher and Stoler, Reference Gallagher and Stoler2009; Hoekman, Reference Hoekman, Narlikar, Daunton and Stern2012a; Rodriquez Mendoza, Reference Rodriguez Mendoza, Ortiz, Bellmann and Rodriguez Mendoza2012).
A critical mass agreement could provide all its benefits to non-signatories as well as signatories on an unconditional MFN basis, greatly reducing a country's reasons to block. A spoiler country or coalition might be tempted to threaten to block the world's leading traders in order to coerce them to give up a side payment. But the credibility of this threat would be low, since all would know the spoiler can escape material losses by not signing any pacts, and since carrying out such a threat would jeopardize the prospect of support from its major partners in future WTO issues and damage the WTO's credibility and value for that country.
The critical-mass approach has generated some objections, and one is that it would permit free riding. While this is true, its competitive significance would be small if the free riders amounted to only a small share of the world economy. Free rider countries could, however, expand relative to the membership in the future. If this is a concern, the rule could be amended to provide that if a non-signatory Member's GDP or covered trade should expand to some objective level (such as some share of world GDP or covered trade or the share of the smallest signatory), the signatories would have the right, as a group or individually, to deny the pact's benefits to the growing non-signatory three years later unless it made commitments that satisfied the signatories.
Another objection is that requiring unconditional MFN treatment could make variable geometry a dead letter since if any large players chose to stay out of the club, which seems quite plausible, others would have no interest. A well-known remedy is the plurilateral agreement applied with conditional MFN. All WTO Members would be welcome to join the negotiation. The benefits would be enjoyed by all who make the commitments needed to join the club and not shared with those not ready to make these commitments. This permitted discrimination would create an incentive to join, perhaps later enlarging the club and its achievements.
Members could authorize additional plurilaterals under WTO Agreement Article II.3. Members could first devise a multilateral framework to address PA disadvantages and protect interests of non-signatories. For instance, this framework could require that new PAs must (a) be open to every WTO Member at the outset; (b) cover a substantial share of the relevant trade or GDP (as the Government Procurement Agreement does); (c) report specified information to the WTO periodically for transparency; (d) provide that when disputes arise under its provisions, signatories will use the common WTO Dispute Settlement Understanding (in contrast to the Tokyo round codes); and (e) build in a mechanism for non-signatories to accede later, allowing the option of multilateralizing the plurilateral eventually.
Some might object that allowing variable geometry will marginalize the smaller and weaker Members and solidify control over the WTO by the stronger Members. Nothing in this rule, however, will prevent any Member from taking an active part in these negotiations, including by joining a coalition of weaker players to magnify their influence. Unfortunately, it is true that many least developed Members are marginalized in a commercial sense by their low level of development. With their tiny shares of world GDP, they have little with which to bargain in market access bargaining. These countries need many things: better education systems, health care, agricultural productivity, infrastructure, climate resilience, governance, sustainable economic growth, diversification, capacity building, and effective development assistance toward these ends. Trade policy is only one tool, and other international organizations focus on development in a more comprehensive way. If some weaker Members are unable to make new WTO commitments now, they will nevertheless receive the benefits granted by signatories (in the case of unconditional MFN) and will be able to join the top tier later when they are ready (with either type of club). Another disadvantage is that by this means the WTO would relax the counterweight it could otherwise provide to non-signatory governments, which can offset domestic special interests that have disproportionate influence over their politics and economic policies.
But the choice is always between available alternatives, each with advantages and disadvantages. The WTO has tried the universal single undertaking rule. Some governments fear that giving a green light to PAs would sacrifice the bargaining leverage they have in the WTO. But how much has that leverage gained them during the Doha Round? And, meanwhile, how much has the stalemate cost the WTO in credibility? Variable geometry will make future negotiations more efficient. Further, if larger players agree to greater liberalization in a club and this accelerates their growth, their expanded spending may spill partly into greater imports from non-signatories without their having to commit.
And critically, achieving such pacts could strengthen the WTO as an institution by dissuading businesses and governments from turning further to regional PTAs. PTAs also discriminate against non-signatories, and arguably they are under less effective WTO discipline than future plurilaterals would be under such a framework. Article XXIV is not enforced effectively; some PTAs' transparency could be greater; their members are not required to use WTO dispute settlement for disputes under PTA rules; they are not required to accept new members (Hoekman and Mavroidis, this issue).
Discrimination by a PA against non-signatories would be a disadvantage relative to the ideal MFN norm. But blocking this option in the WTO has only led to discrimination by other means.
A restricted single undertaking (SU) rule
If during this round, three sets of Members negotiated toward agreement on three new CMAs or PAs, they might achieve only minimal joint gains or none at all. Pact 1 might well create gains distributed more favorably to some Members than others. A disfavored state A could then try to gain enough to justify accepting this pact by linking it to another pact where it would gain more. A could demand that B make commitments on pact 2 or else A would reject 1. This threat would have greater force, however, if the set of pacts, negotiated simultaneously, were linked as a single undertaking, binding now only on the WTO Members that chose to join the club. (In the Doha round, the single undertaking rule applies to all WTO Members). Adding a restricted single undertaking would magnify the influence of a threat to withdraw, encouraging club Members to work hard for the greatest ambition they can accept. Without the single undertaking, it might be difficult to assemble a package of pacts that achieved the degree of balance needed for any of them to be adopted.
Two types of agreement could be made exempt from the restricted single undertaking and allowed to move ahead separately. If gains from a single pact, such as trade facilitation, are relatively balanced, delinking the pact will not forego much ‘linkage value.’ And if a pact between certain states had little impact on other countries either way, and the others did not care much about it, such as duty-free–quota-free access for least developed countries, it too would have little value as a chip in the wider bargaining process (Hoekman, Reference Hoekman and Mavroidis2012b).
On the other hand, this restricted single undertaking rule could discourage a state from signing one or two acceptable pacts when it cannot accept others. If the rule generated stalemate even with reduced numbers, the Members could drop it (and perhaps drop the idea of a round) and work on individual critical-mass or plurilateral pacts simultaneously or sequentially.
Consultative board
Private discussions in smaller groups are essential for discovering joint-gain deals and breaking stalemates in all multilateral negotiations. Several nations and respected authors have proposed creating a formal standing representative WTO board for this purpose, on which a small number of Members would have seats at a given time. Other international organizations have such small bodies, and without one, the WTO process of finding consensus is less efficient. But many WTO states have rejected the idea of a formal body that excludes some Members, even for consultation only, and consensus support at present seems beyond reach here as well.
Negotiators and chairs did, however, evolve an analogous informal practice under present rules between 2000 and 2005. After the Seattle debacle, facing the serious mistrust that darkened the Organization in 2000, Kåre Bryn, Norway's Ambassador and chair of the General Council, began calling small off-the-record meetings only after announcing each meeting in advance to all Members, on the understanding that any Member could make its views known. It was also understood that small meetings would never make decisions for the whole; any results must be reported to the plenary for decision. After six months of Members following these informal understandings without any objections, Bryn issued a written description of the practice, and no one objected to his statement.Footnote 5 This process understanding was followed again in 2001 in reaching agreement to launch the Doha Round. In 2002, the full Trade Negotiations Committee agreed to work on the basis of its best practices of the past and cited the 2000 Bryn statement.
During the early 2000s, Members including small developing countries also increasingly joined bargaining coalitions on different issues and met with their coalition partners prior to meetings of formal bodies. Chairs calling small informal meetings began inviting an ambassador from each coalition active on the issue. Eventually, chairs tended to invite whichever delegation the coalition itself had chosen as its leader.Footnote 6 Leaders such as Indonesia for the Group of 33 discussed the issues with their fellow coalition members before and after the global steering meetings. This new representative practice marked a significant departure from the less transparent GATT tradition. Complaints from delegations about a lack of internal transparency declined sharply after 1999. Today it might be wiser to encourage continuation of this flexible informal representative practice than to attempt to formalize it against opposition.
Lessons from experience for WTO chairs
Whether efforts to agree on an agenda and launch a new round succeed or fail will also depend in part on how WTO chairs play their roles. Members regularly turn to the Director General, ministers, and Geneva ambassadors to chair their conferences and committees, and part of their task is mediation, assisting divided Members to find consensus. Little research has illuminated what chairs do or their effects, but some is beginning to appear (Odell, Reference Odell2005; Tallberg, Reference Tallberg2010; Kanitz, Reference Kanitz2011) and some preliminary observations from experience can be summarized briefly.
Chairs have limited but significant capacity to influence the efficiency and legitimacy of WTO negotiations. Members delegate functions to chairs because Members face uncertainty about parties' true resistance points on the several issues and about what multilateral agreements are feasible. The mediator's job is subtle and entails risks. He or she serves the Members and has no authority to make any decisions for them. The chair too lacks complete political information, and thus even an experienced trade diplomat can bet on a mediation strategy that proves ineffective politically or even reduces the odds of agreement.
Chairs' tactics have ranged from the more passive to the more interventionist. The most passive and least risky form of mediation is to gather information privately, communicate back to Members how much support and opposition different ideas are attracting, and develop his or her own diagnosis of the key bargaining problems and opportunities. One obvious pitfall would be for the mediator to act in a way that undermines delegates' trust, causing them to withhold private information that could be used against them and that is needed for accurate diagnosis and intervention. Observers of the 1999 Seattle ministerial complained that the chair, US Trade Representative Charlene Barshefsky, fell into this problem by giving the impression in public that she was driven only by US national interests, and spending little time speaking privately with ministers to build consensus.Footnote 7 Most chairs avoid this pitfall, but another common challenge will be when a delegate privately states a position as his or her bottom line, and the chair then needs to judge whether this is a true minimum or not. Misjudgment could lead to an agreement being reached too early when there was still opportunity for further negotiation and therefore opportunities for greater gain. The opposite error, to regard a true minimum as a bluff, can be more dangerous, as we will see in a moment.
WTO chairs regularly go beyond these passive steps to moderately interventionist formulation tactics. In a round, first they create a complex formal structure of multiple chairs and mechanisms to coordinate among themselves. Chairs then call and lead informal small meetings of leading rival Members to explore possibilities for agreement that might break deadlocks at the level of ministers as well as ambassadors.
An important formulation tactic, after delegations have defended their positions several times and agreement remains elusive, is to issue an informal text in the chair's name. The chair's text is based on ideas circulated by delegations but no agreement as such has been reached by any delegation at that stage. The tactical goal is to simplify the options for discussion and create a new intermediate starting point that the parties may use as a vehicle for further negotiations. A cautious variant presents multiple alternative options, in square brackets, for the same item. Even this step can omit some proposals that are circulating. With a bolder variant, the informal single negotiating text, the mediator proposes a single option for each issue and presents a package intended to seem like a balanced compromise. A chair's announcement of the intention to issue a single text introduces a subtle change in negotiators' incentives. Before then, as each delegation conceals how much flexibility it has, the parties find it difficult to coordinate their expectations about how one another would behave in response to a compromise proposal. But once they know the chair is going to craft a compromise, they have a new incentive to reach agreement among themselves before the chair decides for them (Buzan, Reference Buzan1981).
In 1999, in Geneva, after delegations had made many conflicting proposals, the chair of the General Council assembled a cautious chair's text to set an agenda for Seattle, and listed conflicting proposed options for several issues in many square brackets. This case illustrated a risk of caution. Because many delegations now saw their preferred language in an official text, moving toward agreement meant that many delegates would be required to give up something they felt they had achieved. This cautious tactic might have inadvertently encouraged delegates to dig in their heels – though this was not the only reason for the Seattle debacle.
In 2001, Stuart Harbinson, the Permanent Representative of Hong Kong and chair of the General Council preparing for Doha, again waited until delegations had failed to reach to agreement, and then in September issued a bolder single negotiating text with very few square brackets. Ambassadors naturally criticized it where it departed from their positions, but few rejected it as a platform for further talks. In Doha, after considerable further haggling, the ministers settled on an agenda for the round that was nearly identical to the Harbinson draft. Several participants cite this Geneva mediator's work as a major reason why they reached agreement in 2001.Footnote 8
This chair was making a bet; he did not have certain knowledge whether all Members would accept the positions he published for them. The obvious risk of a bold single text is that the bet will turn out to be too optimistic. The Derbez text laid on the table at the 2003 Cancún ministerial illustrates that less happy outcome. Mexico's Minister Luis Ernesto Derbez, chair of that ministerial conference, issued a text including two Singapore issues as proposed by the EU and Japan, going in their direction further than the earlier Geneva chair's text had gone, evidently gambling that developing countries that had sworn they would not accept them were bluffing, and would do so if satisfied on other issues. Derbez also included a pro-US text for the cotton section in place of stronger commitments proposed by the African group. The African group, India and others condemned this Derbez text with hostility and suspicion. Many participants report that the mood among negotiators that night was unusually ugly, indicating a collapse of mutual trust. Minister Derbez and Director General Supachai Panitchpakdi decided to end the ministerial the next day, when many key Members were still ready to continue working, which was another controversial decision.Footnote 9 The decision to introduce this text might have unintentionally made the task of settling in Cancún more difficult.Footnote 10
When selecting a minister or ambassador to take on this important, delicate assignment, the top priorities should go to personal command of the technical economic and legal issues under negotiation in the WTO, experience in diagnosing and resolving negotiation impasses, and widespread trust in the individual's impartiality. The Secretariat is available to support a chair, but ambassadors and ministers decide how much they want to listen to the Secretariat. Should a chair be less aware of the details than the delegates, experienced WTO delegates will use their technical knowledge to their advantage as much as possible. The WTO need not select the host country to chair a ministerial conference held on its territory. Chairs and other ministers chosen to ‘facilitate’ particular issues should be chosen for their expertise in mediating these issues. Ministers and ambassadors currently representing the largest trading Members, however, should never be chosen as chairs or facilitators. The giants have trade interests at stake in many issues on the table and well-organized constituency groups pressing for their own advantage. Even the most skilled and respected diplomats from these Members will have a built-in disadvantage in generating the trust needed for effective mediation.
The term of office for Geneva ambassadorial chairs has been limited to one or two years in most cases, and for the ministerial chairs to only one week. Such short terms deprive the organization of the benefits of experience in the art of chairing in this specialized environment. The organization attempts to compensate for the short terms by assigning promising ambassadors first to chair less central bodies in Geneva where they can gain experience, before choosing a few of those ambassadors to chair the most important bodies. But the organization would benefit if it dropped automatic time limits on chairs' terms.
The WTO should also consider adding a regular mechanism for drawing lessons from experience and transmitting them to new chairs. For instance, the Director General could host a private retreat once a year to help prepare incoming chairs of Geneva bodies. Experienced former chairs from diverse countries could be invited to participate, and all could share questions and lessons about how to deal with chairs' dilemmas that recur, independent of the issues of the day (Odell, Reference Odell2005).
The DG and the chair of a coming ministerial conference could also hold a private meeting, a month prior to the conference, of the team of ministers asked to help mediate. In the past some ‘facilitators' have arrived at the conference with little preparation for the delicate jobs they are given. At this advance meeting, they could hear briefings on lessons drawn from previous ministerials, and briefings on the state of play on each current issue. The latter could be given by the ambassadors who currently chair the Geneva negotiating bodies and have been consulting intensively with delegations about possible settlements. The ministers would discuss problems expected to arise, ways to deal with them, the conference schedule, and how they will coordinate. They would establish or deepen personal working relationships.
Summary
If WTO Members wish to launch a new round to follow Doha, setting the agenda will require complex negotiation as in the past, however Doha ends. To reduce the serious information problems they face and prepare the way, advocates should commission an independent research team to produce a comprehensive negotiation analysis before they decide to move further. The team should gather evidence on how the Members, including their major constituent sectors, judge the possible alternatives, including not to launch a new round.It should report whether any set of agenda items is preferred by most Members over their alternatives. And it should attempt to rank alternative issue packages roughly as to their potential for net joint gains.
Reaching an agreement on an agenda will depend on the procedural rules that apply in the agenda negotiation and the subsequent round. They should consider four procedural rules that seem legitimate today and most likely to help Members find a joint-gain agenda. (1) Decisions on the agenda and the final act will be made by consensus. (2) They should not repeat the universal single undertaking rule. Resulting pacts can be critical-mass agreements with unconditional MFN, or failing that, plurilateral agreements with conditional MFN, governed by a new framework to safeguard non-signatories' interests. (3) Club agreements resulting from the round could be linked as a limited single undertaking that would bind only signatories of these new pacts. (4) The WTO should continue to use its current informal representative practice for small meetings that is analogous to a formal consultative board, rather than seeking to formalize it.
Reaching an agreement on an agenda could also depend in part on decisions by WTO chairs during this negotiation. Experience illustrates the potentials and possible pitfalls to avoid. The Organization and its Members would benefit from relaxing its current norm that automatically ends all chairs' terms after a very short time and sacrifices valuable experience. It should also discuss adding an institutionalized mechanism for transmitting lessons of experience from former chairs to new chairs.
