Democratizing the World Trade Organization

Sunday, October 1, 2000

I’ve repeatedly said I thought the WTO process was too closed if I can persuade more of my colleagues that, if they don’t want people like the protestors outside of every trade meeting ’til the end of time, they’re going to have to open the process so that the voices of labor, the environment and the developing countries can be heard, and so that the decisions are transparent, the records are open, and the consequences are clear, [or] we’re going to continue to have problems.

—President Clinton. December 1, 1999.1

Last year, the four days of World Trade Organization (WTO) talks in Seattle were disrupted by violent public demonstrations. Nongovernmental organizations (NGOs), such as People’s Global Action, organized the protests, drawing together groups worried about the effects of globalization on labor standards, human rights, and the environment. One issue that united this diverse coalition was democratizing global governance. Marchers protested that they had no political voice within the WTO, denouncing it as an undemocratic institution, secretive and uncaring. Placards accused the WTO of representing big business and trampling on the rights and interests of workers and environmentalists. The European Union’s trade commissioner, Pascal Lamy, sympathized:

The theory that the WTO is a black box in the hands of unknown and mysterious multinationals has played well among NGO’s for years and there’s some truth to that.2

There might be some truth to that, but there’s not much. If the WTO is seen as a black box, it’s because of ignorance about its organizational structure. The WTO’s dispute mechanism and bargaining rules are largely transparent. The WTO is not about global governance, it’s about the right to trade; as such it’s simply a set of rules about multinational negotiations and dispute settlement. If the WTO is seen as a tool of “unknown and mysterious multinationals,” it’s because of ignorance about the structure of representation within the WTO. International trade negotiators are bureaucrats, but their actions are accountable to politicians (WTO rules and trade agreements are ratified by Congress). In turn these politicians are accountable to their domestic constituents, not some international interest. U.S. politicians tend to represent parochial special interests rather than the interests of multinational corporations. Bill Clinton, Al Gore, Pat Buchanan, even Bill Bradley, claimed they were sympathetic to the Seattle protesters’ claims.3 The future of world trade and prosperity runs a poor second to winning an extra handful of votes in the upcoming presidential race.

It’s certainly true that the WTO trade rounds are closed negotiations and that not all WTO transcripts or records are open to the public. Yet this is a feature shared by all representative governments. Negotiations over the formation of coalition governments or the deliberation within cabinet meetings are as much behind closed doors as deliberations within the WTO. The democraticness of the WTO hinges on howrepresentativeit is of both the winners and the losers from free trade. Opening the WTO process to greater participation won’t necessarily make the WTO more democratic. There are more winners than losers from free trade, but because the costs and benefits are distributed unevenly, only the losers face strong incentives to organize and lobby the WTO. Most trade lobbyists seek protection, not open markets.

This essay is about which countries determine the course of policy proposals in the WTO and to what extent the winners or losers from free trade are championed by national trade negotiators. To explain the future direction of the United States within the WTO, we need to understand the mechanism of representation at work here. The key question is, which U.S. special interests get represented at closed WTO negotiations? The answer to this question depends on one, how much representation the United States gets within the WTO (in other words, how much influence does the United States have over bargaining outcomes and rule making)? and two, which domestic groups the U.S. trade representative (USTR) represents: Is it multinational corporations or the NGOs? Within the WTO, the USTR has a disproportionate influ-ence on making the rules and in determining policy outcomes. The U.S. trade representative is increasingly constrained by Congress, which pushes the interests of U.S. domestic industry and trade unions. It is the interests of U.S. multinational corporations that are increasingly trampled on and ignored. Opening the process won’t make the WTO more democratic. Allowing NGOs to participate could potentially pull outcomes further from the interests of the majority and toward the interests of unelected special interests, but most probably it would result in the stagnation of further talks to liberalize world trade.

Where the WTO is blatantly undemocratic is in underrepresenting the interests of its poorer countries (more than 100 of the 136 members). Of course, if developing countries have more power within the WTO, it is less likely that the United States will be able to link trade to environmental issues or labor rights. If, as they claim, American trade unions, environmentalists, and human rights activists want to use the WTO to influence the course of globalization, they’d be better off using e-mail to contact their representative in Washington, D.C., than battling in the streets of Seattle to democratize the WTO.


Based in Geneva, the WTO oversees the rules of international trade.4 It is an arena for organizing multilateral trade negotiations and settling trade disputes between governments. The WTO was created in 1995, replacing the General Agreement on Tariffs and Trade (GATT). The GATT was organized after World War Two, at the instigation of the United States. It evolved from a set of guidelines for an institutional forum where trade negotiations took place and trade disputes were adjudicated. Since 1948, a series of eight multilateral trade rounds has led to wide-ranging trade concessions, largely tariff cuts on manufacturing goods but recently also nontariff barriers in farming, intellectual property rights, and government appropriations. The most recent round, in Uruguay, had mixed success, but the bigger picture is that GATT multilateral negotiations have dramatically lowered both tariff and non-tariff barriers around the world, creating freer trade and contributing to postwar world growth. One of the main differences between the WTO and the GATT is the WTO’s tougher dispute settlement mechanism (individual countries can longer veto the dispute panel’s ruling).

As did the GATT, the WTO works on the principle that free trade will allow all of its members (136 and 30 waiting to join)5 to achieve higher levels of growth. The important norms within the WTO are nondiscrimination and reciprocity. Nondiscrimination means that members cannot levy different duties on the same good coming from different countries. Reciprocity implies a give-and-take approach to creating mutually advantageous agreements. It makes trade agreements more politically palatable. Both reciprocity and nondiscrimination are useful tools for driving down trade barriers among a large and diverse group of countries. But the first step is getting the 136 members to agree to an agenda for a future trade round. Human barricades aside, this is turning out to be a difficult task.


Each member of the WTO has an equal vote. However, within the WTO, the United States has a disproportionately large influence on outcomes. In part, this is because of the rules organizing trade negotiations. Putting the WTO’s 136 members into a room and letting them hash it out would be too slow and unwieldy. Instead, the WTO uses the most favored nation (MFN) rule6 in combination with principal supplier negotiations to organize talks. The biggest principal supplier (or potential principal supplier) of a product negotiates with the principal purchaser. Whatever agreement they come to is broadened to include other countries through the MFN principle (principle of nondiscrimination)7 described below:

Any advantage, favor, privilege or immunity granted by any contracting party to any product originating in or destined for any other country shall be accorded immediately and unconditionally to the like product originating in or destined for the countries of the other contracting parties. (GATT Article I.)

If the goal is to open trade, it makes sense to structure negotiations around the biggest countries, as this has the biggest echo effect through the MFN provision. Fewer actors make it quicker and easier to reach decisions, and because the actors involved have large markets, it is easier to negotiate agreements under which they can trade concessions on different products. Reciprocity seals the deal. In practice what these rules mean, however, is that the same groups of countries dominate all trade negotiations. The United States and European Union (EU) together account for approximately two-fifths of world trade. Not surprisingly, almost all principal supplier and principal purchaser negotiations include the EU on one side and the United States on the other.

The countries that are consistently excluded from these negotiations are the developing ones. Rarely are third world countries principal suppliers or principal purchasers of a particular good except in agricultural issues, which are typically excluded from trade negotiations. The WTO does make allowances for its poorer members. All countries are bound by the outcome of principal supplier/purchaser negotiations, but poor countries are often allowed to opt out (in part because a sharp shift in trade patterns can cause huge economic dislocation for them). But not all poor countries want to be sheltered from tariff changes; many want more aggressive tariff reductions. India is very unhappy about the course of textile liberalization; although it wasn’t shut out of U.S.-EU talks, most of the liberalization is back loaded until 2004. More than 100 of the 136 WTO members are developing nations. Most of the time, few are represented in trade negotiations. Each has a vote, and all have to vote to agree on the final set of agreements. Of course, they vote yes because to vote against an agreement opening trade in the industrialized North is against their interests. That does not mean they are signing an agreement that comes close to meeting their policy goals.

In general then, the market size of the United States gives it a disproportionately large influence in negotiating trade agreements. WTO trade rules are also highly reflective of U.S. interests, although it does not always appear that way. Take the WTO ruling on the Kodak/ Fuji dispute (1997). The WTO’s disputes are settled through binding arbitration. If countries agree to enter into WTO dispute resolution, they agree to abide by the panel of independent experts’ decision. If the panel judges that WTO rules have been breached, and there is not widespread opposition to the panel’s findings in the council of representatives, the wronged country can withdraw trade concessions worth a similar amount (again, that principle of reciprocity). Very publicly, the United States lost the complaint of unfair trade that it filed against Fuji film. To date, however, the United States has won far more WTO trade disputes than it has lost (twenty-two of the twenty-four that it has filed). In two recent cases, the WTO panel ruled in favor of the United States, finding that the EU unfairly discriminated against Latin American/U.S. bananas and U.S. beef.

The WTO has restrained the United States’ right to impose unilateral trade restrictions. However, it has also overlooked unauthorized retaliatory action that the United States has taken against the EU and Japan. The United States has one of the largest, richest markets in the world system. If the United States doesn’t honor WTO rules (as it didn’t in the recent banana/beef hormone cases with the EU), it seriously undermines the credibility of the WTO. The WTO may be full of faceless bureaucrats and independent experts, but all are aware of the importance of U.S. participation in the future of the WTO. This affects how WTO rules are interpreted; they never stray too far from either the U.S. or the EU’s interests. In both the WTO’s multilateral trade negotiations and its dispute resolution mechanism, the United States is a powerful player.

The United States has strong representation within the WTO, but which domestic interests does the U.S. trade representative represent? In other words, what are the U.S. trade representative’s preferences?


The Seattle protestors believe that the WTO is dominated by U.S. multinational corporations. To determine whether or not this is true, we need to ascertain who the USTR is accountable to. Officially the USTR works for the president. She or he is nominated by the president for a fixed term. Presidential appointees, however, need the approval of the Senate. As such, the choice of USTR reflects both the president and the Senate’s tastes (in recent years the Senate has pushed for nominees with a background in business).

Most USTRs advocate freer trade; however, the USTR has not always acted in the interests of freer trade. In part, this is because the USTR’s actions are constrained by Congress. Congress can refuse to ratify international trade agreements. Since NAFTA, Congress has retained the right to amend international trade agreements. Any settlement reached by the president is structured so as not to alienate possible opposition coalitions in Congress.

So the USTR negotiates, but through their powers to appoint and veto, both Congress and the president play important roles in the formation of trade policy. Congress and the president represent different constituencies—for example, the president is more receptive to multinational corporations (MNCs) than Congress. So who gets represented—unions, MNCs, environmentalists, tradables, nontradables, and so forth—depends on whether the president or Congress dominates the formation and implementation of trade policy. It is worth explaining the history of this relationship because the power to make trade policy has shifted back and forth between Congress and the president over time.

In the 1920s Congress began delegating its tariff-making powers to the president by authorizing the president the power to levy antidump-ing duties. By 1934 it had given the president the power to negotiate extensive reciprocal tariff cuts with other countries (the Reciprocal Trade Agreements Act). Congress delegated trade policy to the president for a number of reasons: Congress lost interest in legislating tariff rates when they became a less important form of financing for government spending. Members of Congress also complained that tariff bills took up too much of their time. In 1934, it appears that members of Congress had come to believe that free trade was beneficial and that delegating the authority to reduce tariffs to the president was the only way to open international trade. Only a few years before, Congress had passed the highly protectionist Smoot-Hawley tariff. Many members of Congress felt that this act was responsible for the severity of the Great Depression. Delegating to the president was the solution to creating a more open trade policy. The president is less vulnerable to protectionist pressures and so tends to be more free trade oriented than Congress. The USTR negotiated on the president’s behalf, and Congress’s role was to ratify international trade agreements. This veto power was formalized in legislation known as fast track. Under these rules Congress committed to vote either for or against trade agreements negotiated by the president. By signing fast-track authority (which it periodically renewed), Congress limited its ability to attach protectionist amendments to international trade agreements.

In the past, Congress has periodically renewed fast-track authority (approximately every three years). However, as trade has become a more salient issue and domestic pressures for protectionism have increased, Congress has started to take back some of its influence over trade policy.

Since the 1970s Congress has used the threat of nonrenewal of fast track to restrict the duty-reducing authority of the president. For example, in the 1974 trade bill Congress named industries to be excluded from international negotiations. These were not simply suggestions but were restrictions on the USTR’s actions. If the USTR fails to follow congressional guidelines, Congress can refuse to ratify the treaty. Congress never has, but the threat of veto appears to have influenced the industry concessions negotiated by the USTR.8 Any settlement reached in the WTO is structured so as not to alienate possible protectionist coalitions in Congress.

President Clinton has not held fast-track authority since the passage of NAFTA. In 1999 Congress, for the first time, refused to renew fast-track presidential authority. This has made it difficult for the USTR to strike international trade deals. Congress is able to tack amendments to any agreement negotiated by the USTR. Trade concessions offered by the USTR during negotiations are less credible because Congress can amend any agreement. That said, the amended agreement would have to be renegotiated in the WTO, and this would constrain Congress’s decision to amend it in the first place.9 Nonetheless—because of increased uncertainty—it is more difficult for the USTR to make international trade agreements without fast-track authority. Following this next election, a new president may well be able to persuade Congress to authorize fast track. But this episode demonstrates that what Congress gives, Congress can take back. And increasingly, Congress is taking back its power to determine U.S. trade policy.


Free trade increases the nation’s aggregate wealth, but there are also domestic winners and losers. For example, consumers get lower prices and more choice (although national governments have less control over consumer safety and health standards), whereas uncompetitive industries are driven out of the market. The costs and benefits of opening trade are also distributed unevenly across winners and losers. The ben-efits of opening trade are spread thinly across consumers, whereas the costs are concentrated in uncompetitive industries. The incentives to organize and lobby are greater for losers than for winners. Free trade might benefit the aggregate, but most interest groups in the United States lobby against free trade.

Because Congress is elected from small geographic jurisdictions, it is particularly vulnerable to protectionist pressures. This is because the costs of trade protection are spread out over all congressional districts (consumers paying higher prices) and the benefits are concentrated in those electoral districts and states that contain the protected industries. Protecting local industry hurts the nation but benefits voters of the member of Congress’s district or state. Reelectionincentives make members of Congress receptive to local demands for protection.

Presidents have tended to be much more free trade oriented than Congress. Magee, Brock, and Young (1989) argue that this is because the nation is the president’s constituency and he increases his reelection chances by maximizing the national welfare, while each member of Congress cares first and foremost about increasing the welfare of her or his constituency.11 The one caveat here is that the president is not elected by direct popular vote; he is elected from an electoral college of state delegates. Thus, the president is very concerned about winning votes in marginal states with large number of electoral college votes. So although the president is more likely to care about the welfare of the nation than the average member of Congress, he still cares about parochial interests in large states, such as California and Florida.

So which domestic interests get representation? Both Congress and the president have geographic jurisdictions. The size, spread, and location of winners and losers affect their political clout with the House, the Senate, and the president. In the case of Congress, a majority coalition is needed to veto international trade agreements. A coalition could form between several districts containing geographically concentrated industries. It is easier, however, to form a coalition from electoral districts that support the same industry. If an industry is in several districts, this acts as a focal point for coalition formation in the House of Representatives. Likewise, if an industry is located in several different states, it can help create a majority in the Senate. The size and location of a domestic group influences who—the president, the House, or the Senate—represents it.

Large and dispersed industries, such as textiles and apparel, receive widespread support in both the House and the Senate. The textile industry is concentrated in North Carolina and South Carolina (60 percent). However, it also employs a large number of workers in Penn-sylvania and Alabama. The apparel industry is spread throughout the United States but heavily concentrated in the mid-Atlantic states, such as New York and New Jersey. In part because of its widespread congressional support, numerous exclusions for U.S. textiles were written into both the NAFTA and Uruguay GATT agreements. The United States does not have to liberalize textiles until 2005.

Another large and dispersed industry that wields a disproportionate amount of political influence in Congress is agriculture. In the initial formation of the GATT, the United States demanded that farm trade be excluded from tariff reductions. Now the USTR is fighting hard to bringing agriculture back onto the WTO agenda. U.S. agriculture is highly competitive; however, U.S. farmers claim that unfair subsidization in Europe undermines their competitiveness on the world market.12 There is huge support in the Senate for agriculture—in part because rural states are overrepresented. Beef ranchers, fruit growers, and wheat farmers are powerful actors across many rural states. When the USTR pushes to put agriculture back on the WTOs agenda, he or she is pushing on behalf of farmers’ representatives in the House and Senate.

Although it is politically helpful to be dispersed, interests that are too dispersed tend to be underrepresented in Congress. In part, this is because they have difficulty organizing effectively. In part as well, it is because highly dispersed industries are unlikely to be a strong voting force in any single district. The member of Congress is unlikely to go out of her or his way to offer these industries support. Both consumer and environmental groups face this problem. Environmental groups made a lot of noise in Seattle, but they tend to have dispersed support across the country. In every congressional district there are some environmentalists, but they are not a strong voting force in any single district. Environmentalists, however, do have political clout with the president (in the aggregate, they add up to a lot of votes). Hence the statements by Gore, and other presidential hopefuls, sympathizing with environmentalists.

Another underrepresented group in U.S. politics are interests that are highly concentrated geographically, in other words, interests that are located in a single district or state. They are strongly supported by their representatives and senators; nonetheless, they are promoted by only a few representatives and senators. Print and photographic equipment is an excellent example of this type of industry. Kodak is a highly concentrated industry in upstate New York but does not wield a lot of influence in Congress. In fact, some of the lowest U.S. tariffs are on print and photographic equipment. Nor are there quotas, such as voluntary export restrictions, on print and photographic equipment. The president has negotiated such quotas for other industries but not for Kodak. Upstate New York is not a politically competitive race in presidential elections. Lacking the political clout to influence the president or Congress, Kodak has been forced to seek import relief from an international body, the WTO.

A few highly concentrated industries do have political clout. It helps to be located in a district (state) whose representative (senator) has important agenda-setting powers. Footwear is a small and highly concentrated industry (until recently, in Maine and Missouri). However, over the years, the footwear industry has garnered disproportionate industry protection because of the important institutional positions held by its representatives. Especially important to the footwear industry was the strong institutional positions of Senator Danforth (R-MO) and, later, Senator George Mitchell (D-ME). Senator Danforth was the chairman of the Finance Subcommittee on Trade from 1981 to 1986, and Senator Mitchell was a member of the Finance Committee and subsequently became the Senate majority leader from 1989 to 1994. With the retirement of these members in 1994, the footwear industry lost its key supporters and is increasingly likely to find trade policy moving against its interests. According to one Clinton administration aide, “there are so few places with any footwear industry in them now that no one cares.”13 As mentioned earlier, the president is elected nationally, and we would expect him to be interested in free trade because it increases the aggregate wealth of the nation. However, the president is elected by an electoral college. He cares about winning states with competitive races and with large numbers of electoral college votes. The president represents the interests of industries located in states that are politically important to his reelection (or that of his vice president). Although few American workers are unionized, AFL-CIO workers are concentrated in four states—Illinois, Michigan, Ohio and Pennsylvania—these large states are critical to Al Gore’s chances in the primary race. Hence, Gore’s movement from free trader to labor rights activist.

In a similar way, the steel industry has used its geography to its benefit. The major source of the power for the steel industry comes from its main union, the United Steel Workers, which represents workers in steel, iron ore, copper, aluminum, chemical facilities, and rubber. According to one steel union lobbyist:

Steel making is concentrated in the middle of the country but we are in every congressional district. We have about 650,000 members nationwide The geographic spread of our union . . . gave more support to the steel movement than from just the geographic concentration of just steel workers. We can get our members of other industries [in our union] to write their members in support of our efforts.14

In response to demands from steel during the recent Asian financial crisis, the president and Congress organized a quota agreement (illegal under the WTO) restricting steel imports into the country.

What about the “mysterious” influence of MNCs? MNCs have political clout but largely with the president. MNCs tend to support free trade because it helps open foreign markets. They donate capital to presidential campaigns and can threaten to exit the country if the president does not push for the agreements that they want. Indeed, in 1999 the president tried to negotiate a multilateral agreement on investment favored by MNCs. That said, the president ultimately cares about winning large, marginal states. So the needs of MNCs often come second to those of domestic constituencies (i.e., Clinton’s support of a labor rights working group at the WTO). In part, the talks at Seattle failed because MNCs did not lobby very hard. They realized the president was absorbed with winning the electoral support of domestic labor groups.15 MNCs have relatively little political influence in Congress. Precisely because they are multinationals, many of their employees are located outside congressional electoral districts. This is problematic because Congress is taking an increasingly active role in trade policy. Congress’s refusal to grant the president fast-track authority has undermined the tariff-reducing authority of the president. In part, the talks at Seattle failed because, without presidential fast-track authority, there were few incentives for MNCs to lobby for a new set of trade rounds.16 In summary, U.S. industry, trade unions, and environmenalists are not shut out of U.S. trade policy. Some interests have more political clout than others, and environmenalists are one of the weaker interests. But, increasingly, so are the MNCs. The important point is that the losers from globalization have three places to block change: the president, the Senate, and the House!


The World Bank has opened its process to special-interest groups, coopting their representatives into the decision-making process. Clinton wants to democratize the WTO by opening the process to special-interest groups. However, it is important to understand how doing so will change the structure of representation within the WTO. The potential winners from free trade greatly outnumber the losers. However, contrary to their claims, the losers from globalization do have representation in the WTO. The have three places to block the USTR’s actions: the president, the Senate, and the House. Any settlement reached by the USTR is structured so as not to alienate possible opposition coalitions in Congress. As Congress increases its involvement in trade policy, the interests of MNCs and consumers are increasingly ignored. Opening the WTO by including NGOs in the negotiation process could potentially pull outcomes even further from the interests of the majority and toward the interests of unelected special interests. This is because the losers from globalization have the biggest incentives to organize and participate.

Congress originally delegated trade negotiating authority to the president in 1934 precisely to isolate decisions from special interests. It recognized then, as it does now, that if the process is more open, it is more difficult to reach free trade agreements. Opening the WTO to special interests would almost certainly result in is the stagnation of further talks to liberalize world trade. As Marge Simpson said of participatory democracy, “I guess one person can make a difference, but perhaps they shouldn’t be able to.”17

Michael Paulson interview with President Clinton, December 2, 1999. Seattle P
David Evans, December 6, 1999, “EU’s Lamy Sees U.S. Election Stalling WTO Deal,” eu 1. htm.
The Economist, “Storm over Globalization,” December 3, 1999.
For more background, see Hoekman and Kostecki, The Political Economy of the World Trading System: From GATT to WTO (Oxford, Eng.: Oxford University Press, 1995).
As of April 2000
This terminology has changed recently; see the WTO web site: http://
For a fuller description of the MFN rule, see “Trading into the Future” by the WTO (1999).
R. Baldwin, The Political Economy of U.S. Import Policy (Cambridge, Mass.: MIT Press, 1985).
Thanks to an anonymous reviewer for making this point
This section of the paper draws on work from McGillivray (1997), “Party Discipline as a Determinant of the Endogenous Formation of Tariffs,” American Political Science Review, and McGillivray and Schiller (1998), “The Political Geography of Lobbying: Forming Coalitions to Minimize Influence in Trade Politics,” Working Paper.
S. Magee, W. Brock, and L. Young, Black Hole Tariffs and Endogenous Policy Theory (Cambridge: Cambridge University Press, 1989).
Note that the U.S. government also aids its farmers but through direct payments. However, through the WTO’s “green box” policy, these are labeled as nontrade distorting payments because—unlike subsidies—they are decoupled from production.
For a fuller description of the footwear and other cases, see McGillivray and Schiller, “The Political Geography of Lobbying.”
Thanks to an anonymous reviewer for making this point.
See discussion in The Economist, “The Storm over Globalization,” No-vember 27, 1999.
The Simpsons, FOX Network.