Speeches and Papers

Reviving the Doha Round

by Jeffrey J. Schott, Peterson Institute for International Economics

May 2004

© Peterson Institute for International Economics




In February 2003, I concluded a lecture at Stanford University on prospects for trade negotiations in the World Trade Organization (WTO) with the following somber prognosis:

“If WTO negotiators cannot agree on the modalities for agricultural negotiations, then officials will begin to doubt whether the Doha Round can meet the ambitious objectives set by ministers in November 2001 across the whole range of issues. Such uncertainty could put the Cancun Ministerial in September 2003 at risk. Developing countries could well respond to the cloudy signals on farm reform by withholding support for starting negotiations on the Singapore issues (investment, competition policy, trade facilitation, and transparency in government procurement). Trade officials would then have to work hard to prevent a further unraveling of the Doha agenda. At best, the pace of talks would decelerate, and they could possibly seize up.”

I often make mistakes, and I wish I had misjudged developments in the Doha Round more than a year ago. But, unfortunately, events played out as I predicted. The meeting of trade ministers in Cancún in September 2003 failed. Instead of accelerating the pace of the Doha Round, the Mexican standoff impeded the global trade talks and made their conclusion increasingly difficult and uncertain. As a result, concluding the talks as planned by yearend 2004 is no longer feasible. Finishing the talks at all will require significant effort.
In this short paper, I will examine why the WTO negotiations stalled at the Cancún ministerial and what needs to be done to revive the Doha Round. But such analysis first requires a discussion of why it has become so difficult to negotiate agreements in the WTO.


Why the New WTO does not work like the Old GATT

When I was a US trade negotiator during the Tokyo Round, multilateral negotiations took place among a handful of countries. In essence, the United States and the European Community, joined by Japan, Canada, and a few other delegations (mostly developed countries), framed the prospective deals and brought them forward to the broader GATT membership. Most developing countries weren’t involved and weren’t obligated to accept new GATT obligations. They received concessions under the GATT’s most-favored nation (MFN) obligation but did not have to reciprocate by opening their own markets to foreign competition.

Developing countries essentially were free riders on the GATT system—at least until the Uruguay Round. Their nonparticipation came at a cost, however, because the negotiated results often excluded improved access to industrial country markets for the competitive agricultural and manufactured exports of developing countries. In short, developing countries protected their own markets, but in turn had to accept the maintenance of high foreign trade barriers against their most competitive exports.

Though such policies never yielded big economic rewards, they were politically convenient. Many developing countries relied on protected home markets and commodity exports to support modest growth; some followed a strategy of export-led growth and became platforms for the assembly and export of light manufactures. Their success in turn provoked a wave of new protectionism in developed markets via so-called voluntary export restraints, antidumping and countervailing duties, and special protection regimes like the Multi-Fiber Arrangement.

This strategy of limited engagement in the world trading system was never productive and became increasingly untenable over the past two decades with the globalization of economy activity and growing competition from China. Today, more countries have a stake in world trade; more countries have undertaken substantive international trade obligations under the Uruguay Round agreements and the new WTO. The implications of the new trade organization perhaps were not well understood when the WTO entered into force in 1995—but it is now clear that the new WTO does not work like the old GATT.

First, conducting and concluding WTO accords are much more complex than in prior GATT rounds. The WTO now has 147 member countries; membership could increase to 170 by the end of this decade if many of the current applicants complete their accession negotiations. Most of the members are developing countries; unlike the GATT era, many of them now have an important stake in international trade and therefore an important stake in getting something out of the trade agreements. Each country needs to be able to bring home a trophy to justify the concessions that they make to their trading partners, so WTO talks have to produce a big package of agreements that accommodates the diverse interests of its large membership. However, since the WTO still operates by consensus, the task of crafting a set of agreements that meets the demands of the large and increasingly disparate membership has become much more difficult (as evidenced in Cancún).1

Second, achieving a negotiated “balance of concessions” is further complicated by the fact that the United States and the European Union have very little left to give at the negotiating table in terms of market access, except things that are very difficult to give—i.e., the protection in agriculture and textiles that has survived eight previous rounds of multilateral trade negotiations and that is of major export interest of developing countries. To get the United States, Europe, and Japan to commit to significant reforms in long-standing protection in agriculture and in some manufacturing sectors, other WTO members—including middle income developing countries—need to offer concrete reductions in their protection as well. But the developing countries object to lowering their own generally much higher trade barriers without increased and more secure access to industrial markets—and the United States, European Union, and Japan did not send a clear signal in preparations for Cancún that they were willing to change their current policies. Indeed, the US-EU compromise proposal on agriculture in August 2003 seemed to be structured to maintain existing farm programs on each side of the Atlantic, and deserved the harsh critique of the newly formed G-20 (more on this point below).

Third, WTO deliberations are now the subject of intense coalitional politics that was unknown in the GATT era when the United States and European Community were the main hegemonic powers. There are now too many active players to conduct trade talks among a select few delegations, as we did during the Tokyo Round. Members now need to build alliances by issue (the Cairns Group on agriculture) or region (the Caribbean Regional Negotiating Machinery) or groups of developing countries (G-20 that formed before Cancún); the alliance building also takes place through regional trade initiatives that strengthen trade relations among the partner countries and help clarify their common objectives in WTO talks.

Fourth, there is a sizeable gap in the ability of countries to participate actively in WTO deliberations. Resource constraints are real, and many developing countries allocate their representation to regional bodies that offer more immediate pay-offs in terms of unilateral trade preferences. Unfortunately, this leads to foot-dragging in the WTO for two reasons.

Trade preferences give developing countries a brief head start on competing for investment and export markets. Those that meld the trade preferences with domestic reforms improve their competitiveness and reap long-term benefits. Those that don’t find their trade gains transitory and their investment footloose.

These differences suggest that a lot needs to be done not only to address the challenges of the mandate for the Doha Round, but also to improve the functioning of the WTO itself. Much of this latter task should be done apart from negotiating rounds and over the medium term—as the EU Commission finally realized after its knee-jerk reactions to Cancún were succeeded by more sober reflection on WTO institutional reform.


Why did the Cancún ministerial fail?

Many people have theories of cause and effect to explain why the WTO ministerial failed last September. But the basic reason is simple: the Geneva process did not narrow differences between countries on the parameters of the key areas of negotiation (especially agriculture) in the Doha Round. This problem has several major components:

First and foremost was inadequate progress on agriculture. The major trading nations—developed and developing--needed to make real commitments to reform of current practices, not just revised rulemaking obligations that still allowed national programs to continue as before. The United States put forward radical proposals in the summer of 2002 that would require it to rewrite its farm bill; the European Union was unable to do anything until it agreed to new farm policies in the summer of 2003. By that point, Geneva negotiators had failed to meet their assigned deadline to agree on “modalities” for farm trade talks; indeed, by the Montreal mini-ministerial in late July 2003 the talks were at an impasse. Countries recognized that the draft ministerial declaration developed by the WTO Council chairman was untenable and would cause the Cancún meeting to be still-born.

As a result, ministers in Montreal turned to the big powers for a new initiative to save Cancún. From the US perspective, it was a no-win situation: working with the European Union probably would produce a bad deal. The two sides had only about 10 days to craft a proposal, and the EU negotiators had little flexibility to deviate from their freshly minted negotiating mandate. Thus any compromise would reflect in large measure the EU position. But not working with the European Union would doom Cancún. So the US strategy, as I see it, was to deal with the Europeans and then hope that other WTO members would push the United States back toward its original proposals on agricultural reform.

Unfortunately, US officials, among others, underestimated the backlash to the US-EU compromise proposal tabled in late August 2003, which created mistrust and provoked the formation of the G-20.2 This hastily formed alliance of developing countries (led by Brazil, China, India, and South Africa) were united on their demands but couldn’t agree to a common position for their own contributions, thus provoking the ire of US officials and others. Interestingly, this did not inhibit Brazil from pushing for substantive discussions in Cancún, but the curtain closed on the show shortly after intermission and the negotiations were not engaged.

The G-20 was founded—and still exists in large measure—to push the United States, Europe, and Japan to liberalize their barriers to agricultural trade. That position is not antithetical to US interests and objectives in the Doha Round…provided that the large and middle-income developing countries in the group also reduce barriers to trade in their markets.3

Second, negotiators were uncertain about what the European Union needed to get from other countries to enable it to follow through on hoped-for farm reforms that would go beyond their limited new mandate. Initially, the Singapore issues seemed to be EU priorities, though it was hard to identify political constituencies that actively promoted these objectives. Japan also strongly supported negotiations on the Singapore issues (particularly investment) for both substantive and tactical reasons. At Cancún, however, Japanese negotiators seemed to be following defensive strategies to avoid decision points on agriculture more than substantive interests in an investment accord.

While the European Union (plus Japan and Korea) wanted comprehensive negotiations on the Singapore issues, almost no one else did. The expected deal would probably have involved negotiations on a narrowly circumscribed set of transparency obligations—but developing countries were worried about opening the door to new obligations on investment policies in subsequent WTO rounds—that is, in talks that might take place 10-15-or even 20 years from now! In essence, developing countries opted to either delay or forego benefits from market access reforms this decade in order to reduce their anxiety about talks that might take place in a decade or two.

Third, many developing countries misconstrued what could be done in a “Development Round” in the WTO. Some officials recognized that political resistance in the major industrial countries to needed reform commitments in agriculture meant that the WTO talks would necessarily be extended well past the stated deadline of January 2005, and thus adopted positions that deepened divisions instead of narrowing gaps on key market access and rulemaking initiatives. Others were less well informed—which is another problem in the WTO—and demanded advance payments on trade reforms and monetary compensation for the Doha Round to proceed—demands that stretched well beyond the jurisdiction of the WTO and exceeded the competence of the ministers involved in the WTO. The fact that this latter group of developing countries are hurt the most by setbacks to the multilateral trading system seemed beyond their comprehension.


How do we get the Doha Round back on track?

Soon after Cancún, I was invited by The Economist to write an article suggesting what needed to be done to get the multilateral negotiations back on track.4 I argued that “there was—and still is—a good economic case for new trade “concessions” by both developed and developing countries (though recognizing that only the middle-income developing countries, and not the least developed countries, were likely to be required to reciprocate).

Many countries have since taken up the challenge. I am most pleased that my own government has resumed a leadership role. USTR Robert Zoellick sent a letter in January 2004 to his ministerial colleagues clarifying US interests and objectives in the Doha Round and committing to substantial reforms in US policies in the context of a substantive package of WTO accords. Since then, intensive efforts have been made by Zoellick and his European counterpart, Pascal Lamy, to deepen the involvement of ministers in the crafting of the “framework agreements” which are needed to provide political guidance to Geneva negotiators in areas like agriculture, nonagricultural market access, the Singapore issues, and development issues.

To avoid continued drift in the WTO talks, it would be highly desirable to reach a consensus on these negotiating modalities before the summer holidays. Hopefully, WTO negotiators can achieve what should have been done in Cancún and get the blessing of ministers to begin work on putting flesh (and hard numbers) on these skeletal accords.

Agriculture, of course, holds the key. Developed countries need to demonstrate their willingness to reduce both the absolute value of subsidies provided their farmers and the tariffs and other nontariff barriers that protect agriculture. Their commitments need not be as radical as the reforms proposed in the initial US proposal of July 2002 but must go beyond the US-EU joint proposal of August 2003. The G-20 formed primarily to attack that paper, arguing correctly it would not have changed US and EU policies very much. A framework is doable, if countries understand that concrete commitments will not be on offer until the talks enter their final stages—probably in 2006. The framework should include the following elements:

Such a framework would mean ending most export subsidies soon after the completion of the Doha Round; real reductions in domestic subsidy disbursements by the United States and the European Union on cotton and other major products; deep cuts in high farm tariffs and larger import quotas for Japanese and Korean rice; and meaningful, albeit less comprehensive, reforms by middle-income developing countries of their farm trade restrictions.

On nonagricultural market access, the United States has put forward the best approach on industrial tariffs—that is, get rid of them by 2015 (with some flexibility for implementation by developing countries and technical assistance for those countries that currently are heavily dependent on trade taxes for government revenue). Some entire product sectors should be able to accelerate this process and should be encouraged to do so.

Decisions on all of the Singapore issues remain to be taken, though a consensus seems to be developing to move forward exclusively on the topic of trade facilitation. Too bad, because developing countries had a lot to gain from greater transparency of government regulations and policies on all four issues under review. More transparency would yield important dividends in terms of combating corruption, reducing uncertainty about rules for accessing and competing in national markets, and encouraging investment. But most developing countries have resisted moving forward in these areas.

To be fair, developing countries did face substantial challenges in administering and enforcing the Uruguay Round accords, and had legitimate concerns that violations of those rules could provoke retaliation blocking access to foreign markets for their goods. The European Union seems to have demurred and WTO members will likely relegate discussion of investment and competition policy to ongoing working groups and accept negotiations only on trade facilitation issues.

Finally, WTO negotiators need to agree on how to handle the numerous proposals to provide “special and differential treatment” for developing countries. WTO members explicitly committed in the Doha Declaration to liberalize restrictions that adversely affect the trade of developing countries. Paragraph 16 of the Doha Declaration commits “to reduce or as appropriate eliminate tariffs, including the reduction or elimination of tariff peaks, high tariffs, and tariff escalation, as well as nontariff barriers, in particular on products of export interest to developing countries.” Moreover, WTO agreements should afford the opportunity for developing countries to undertake “less than full reciprocity in reduction commitments.”

In my view, this clause merely reiterates standard practice in the WTO, namely that developing countries commit to market access reforms to the extent practicable given their level of development and are accorded a longer transition period to implement those reforms. However, it has been incorrectly interpreted by some delegations to mean that developing countries can have a “free ride” in the market access talks. Such a view neither comports with political realities or the economic development interests of the developing countries themselves. Development goals should be supported by flexible implementation schedules and narrowly focused exceptions from WTO reforms—plus generous technical and developmental aid from national and international financial institutions.

The Doha Round can succeed, if the priority interests of both developed and developing countries are accommodated in the final package of agreements. The leading trading powers need to take the lead—and they now seem to be doing so. But to get the United States, Europe, and Japan to commit to significant reforms in long-standing protection in agriculture and in some manufacturing sectors, other WTO members—including middle-income developing countries—need to offer concrete reductions in their protection as well.



1. The consensus rule is still preferable to a voting scheme, but reform of the WTO’s consensus-building process could make WTO decision-making more efficacious and equitable (see Jeffrey Schott and Jayashree Watal, “Decision-making in the WTO,” Policy Brief 00-02, Washington: Institute for International Economics, 2000).

2. Developing countries were skeptical that either the US-EU proposal, or the Cancún ministerial text, would have changed much beyond how each country’s subsidies were classified under the green, blue, and amber boxes of the WTO Agreement on Agriculture.

3. Indeed, the United States and Brazil share key objectives regarding agriculture in the WTO and should be working together to advance their common interests.

4. Jeffrey Schott, “Unlocking the benefits of world trade,” The Economist, 1 November 2003, 65–67.

© 2014 Peter G. Peterson Institute for International Economics. 1750 Massachusetts Avenue, NW.
Washington, DC 20036. Tel: 202-328-9000 Fax: 202-659-3225 / 202-328-5432
Site development and hosting by Digital Division