I would like to express my sincere thanks to Maria Helena for inviting me to participate, and accepting me as a last minute addition to the conference.

Even though I don't have a paper, for which I'm apologetic to our discussants today, I do have a presentation. Luckily for you, unlike the very excellent talks we've heard so far, I'm not going to talk about Brazilian politics. Luckily because I know very little about Brazilian politics! So, in that sense, my paper is somewhat unique and it's also a little different in the context of the papers we've heard so far. It's very much an empirical study and I have much less theoretical to say, although I think that one might be able to relate it with the ideas and things we've talked about so far.

I was very interested in Fabiano's paper this morning, and although we have somewhat different objectives in mind, I do think that I can get a lot from learning more about the ratification process on trade policy in Brazil. And perhaps Fabiano can draw something from my discussion today about the US process.

I am going to examine three major trade policy debates in the US during the past decade: the NAFTA agreement, the WTO or Uruguay Round Agreement, and the failure of the fast track proposal that President Clinton tried to get through Congress in 1997 and 1998.

In looking at these cases, I have two objectives. First of all, I'd like to assess how well some of the common theories and ideas found in the literature about the politics of trade help us understand these particular cases and their outcomes. I'm going to look at three theories, each particularly focused on predictions about societal cleavages over trade. What kind of coalitions should we expect to form around trade issues? The second objective is to switch gears a little bit toward the end of the talk and examine the change in the domestic climate over trade policy in the US during the 1990s.

The first two cases that I am going to talk about are NAFTA and the WTO. Although especially NAFTA was very contentious, nevertheless free trade proposals passed in both those cases. In the case of fast track, however, the critics of trade liberalization prevailed. I think, for reasons I'll suggest later, it's an interesting puzzle to consider why protectionist forces may have grown stronger during this decade.

Just a little background on cases with which most of you are probably quite familiar. As you know, NAFTA ­ North American Free Trade Agreement ­ represented an expansion of a preexisting US/Canadian free trade pact that was negotiated successfully in the late 1980s to include Mexico. The NAFTA Agreement generated enormous debate in the United States, but it did pass in 1993 in the House of Representatives by a majority of 234 to 200 ­ so, a fairly narrow passage - and by a somewhat more comfortable margin in the Senate: 61 to 38 votes.

Note however that NAFTA was not considered a Treaty. In the United States, a formal treaty requires two-thirds majority in the Senate. The House does not vote on treaties. NAFTA ­ that's why it's called Free Trade Agreement, and not Free Trade Treaty ­ was not regarded as a treaty, and therefore the requirement for passage was a simple majority in both Houses of the Congress. I'll come back to that a little bit later.

The second case is the agreement that came out of the Uruguay Round of Global Trade Negotiations, and the creation of the World Trade Organization, the WTO. This treaty came to a vote in 1994 and passed. Since this was a treaty, it was only voted on in the Senate. It met the  two-thirds requirement with a vote of 76 in favor and 23 opposed.

The third case is the proposed fast track legislation. Through this legislation, the Congress authorizes the president to negotiate trade agreements under special rules of legislative consideration. Fabiano was talking about delegation, and this is the same idea: the Congress would delegate authority to the president. Basically, the Congress would say: "Okay, for the next three years you can go out and negotiate trade agreements, and when you bring those agreements back, we agree to vote on any agreement that you bring back without amendments. There will be no changes, revisions, and additions to the legislation. We'll just vote up or down". Also, a very important part of this deal was that the Congress agrees to vote on the trade agreement within 90 days of its submission, so it won't just delay forever. It has to vote within 90 days.

This kind of arrangement was first extended by Congress to the President in 1974. Every few years it would lapse and then Congress would always renew fast track authority. So, the President continuously enjoyed fast track authority from the Congress between 1974 and 1994, when finally this authority lapsed and, as we'll see later, Clinton decided to wait a while before seeking to renew that authority.

The basic rationale behind this is that Congress does not trust itself. It knows that there are great temptations for the members of Congress to tack on changes to trade agreements to satisfy particular interests. The fear is that if an agreement comes before Congress and we don't have a fast track arrangement like this, Congress will make so many changes unacceptable to our trade negotiating partners - to other countries that have already agreed to a certain set of arrangements ­ that the president will be forced to go back and renegotiate. So, it's believed that fast track rules make it possible for Congress to deal with trade issues without overwhelming and overburdening the process with all kinds of parochial considerations.

One thing that I should note is that when Clinton first wanted to bring up the fast track proposal for a vote in November, 1997, he discovered that he didn't have the votes to pass it. The Democratic leadership in the Congress ended up coming out publicly against fast track, and the President calculated that he was going to lose. Therefore the proposal was withdrawn without a vote in 1997. But it was clearly a failure for Clinton and fast track.

In 1998, the proposal was brought up again. Interestingly, Republicans called the vote this time. The president didn't want a vote in 1998. Even though he did want fast track authority, he decided that the timing of the vote was bad, because it was a congressional election year, and the fast track issue ­ in fact, trade in general ­ is a very divisive issue for the democrats, the president's own party. So, he was very reluctant in the months preceding the elections to raise the issue, or force the democrats to vote on an issue that would reveal divisions within the party, and between the party and the White House.

The Republicans however felt that a vote on fast track was a good idea, because they knew that they could embarrass the president by compelling the Democrats to vote on this measure. So, they forced a vote on fast track, and it was defeated by a vote of 180 in favor to 243 against. Only 29 Democrats voted in favor of fast track, but we should realize that the president basically told congressional democrats: "Eventually I want your votes on fast track, but this year, for this vote, you are off the hook. I am not going to even ask you to vote in favor. Vote whatever you think will be most helpful for you in getting you reelected". And in that case, only 29 actually voted in favor.

One question that arises here is: Why did Clinton delay? In 1994 fast track authority lapsed. Instead of automatically seeking renewal as other presidents had done, Clinton decided to wait three years to bring it back up. A couple of explanations seem likely. There had just been two very difficult trade debates ­ NAFTA and WTO - right in a row and, as I said, this is a very divisive and difficult issue for the Democrats. Clinton did not want to have yet another trade debate right on the heels of these other two very difficult and divisive ones. Another reason is that he had just wrapped up two big agreements, and there were no new negotiations on the horizon that promised to bring near term agreements. So, he didn't need fast track authority for a while anyway. He wasn't planning to bring a new agreement to Congress for some time.

Since the 1998 defeat in the Congress, there has not been another vote on fast track, and it's very unlikely that fast track will be voted on again until some time after the 2000 presidential elections. So, at the moment, the president does not have this kind of authority from Congress. It doesn't mean he can't go out and negotiate trade agreements, but he cannot bring them to consideration in the Congress under fast track rules.

Now, how debilitating is this for the president? Well, it's often exaggerated just how debilitating, because in fact Clinton has, or his representatives in the Executive branch have, reached over 200 trade agreements since he took office. Only two of those have been brought before the Congress for a vote under fast track, and those were NAFTA and WTO.

So, many smaller trade agreements and small issues, often bilateral or just a few countries, are reached all the time. The US is still negotiating and reaching trade agreements, but these are considered executive agreements that don't require congressional authorization or approval.

How do you determine whether an agreement is so significant that it requires congressional approval? If it does require congressional approval, how do you know whether it should be submitted as a treaty, like WTO, or as an agreement? I think this is not so much a legal question as a political one. The White House could have brought NAFTA up as a treaty. There was nothing stopping them from doing that. So, I think the president makes the determination. Some kinds of agreements, though, it's clear that the political cost of trying to get away without congressional approval would be greater than going through the debate itself.

Now let us examine cleavages over trade policy in society using these cases, and see how existing theories fit. The first theory I am going to talk about is based upon an economic model called Stopler-Samuelson, and this economic model has been applied to the politics of trade by Ronald Rogowski (1990) in his book Commerce and Coalitions.

Basically, the Stopler-Samuelson theory says that the owners of abundant factors of production benefit from rising exposure to trade. The term "factors" refers to inputs to the production process. The classical factors of production are land, labor and capital. When a country's exposure to trade is growing, those who own or intensively use relatively abundant factors of production will benefit. If trade is rising, however, the owners of relatively scarce factors within that country will tend to be hurt ­ they are the losers. So, in other words, growing trade creates both winners and losers. Trade therefore has important distributive effects.

Now, in relative terms, the US has abundant land and capital, but relatively scarce labor. That suggests, following this theory, that farmers and business firms (i.e., capital) should favor freer trade while labor should oppose freer trade.

Is that true? Well, if we look at the trade debates of the 1990s, very generally it is true. All of the major business associations supported NAFTA, WTO and fast track. Business spent twenty-five million dollars in support of the pro-NAFTA lobbying campaign. However, even within the business community, there were divisions and exceptions, to which I will return in just a moment.

What about Labor? Labor opposed all three of these agreements. And yet, there is a twist here, which I think is important. Labor did not adopt the sort of blunt protectionist stance that Stopler-Samuelson and Rogowski would lead us to anticipate. The position of the labor unions was this: "We will support trade agreements only if they give the same kind of rights and protections to labor that, for instance, they often give to intellectual property rights". So, this was not an unconditional condemnation of trade agreements, but rather a demand that trade deals must take Labor's interests into consideration.

Most Democrats have embraced these kinds of demands. Republicans and business firms generally oppose any connection between labor standards and trade agreements.

Clinton responded to these sorts of pressures in the case of NAFTA. Clinton and President Carlos Salinas of Mexico negotiated two so-called side agreements that, while not formally part of the NAFTA agreement, committed the U.S. and Mexico to address labor and environmental issues connected with trade between the two countries. This was an attempt to increase the chances of passing NAFTA and to satisfy Labor and Democrats, who were worried about the agreement.

These labor standards were very weak, and none of the unions that opposed NAFTA were won over by this action. None of them said: "Okay, now we have a labor side agreement; we will support NAFTA". Some Democrats in the Congress were, however, satisfied by the side agreements and came around to vote for NAFTA.

By the way, the link between labor standards and trade becomes even more important once we look at fast track. So, in a very general way, there's some match between Stopler-Samuelson and these cases, but we can do much better if we also look at other theories. For instance, a second set of predictions draws upon the Ricardo-Viner model. This model has been popularized and applied to trade politics by Jeffry Frieden (1988; Frieden and Rogowski, 1996; also see Ferguson, 1984). Before returning to our cases, let us explore, for a moment, the differences between the Stopler-Samuelson and the Ricardo-Viner models of how trade affects various societal groups.

Stopler-Samuelson assumes what is called "perfect factor mobility". This means that when capital, labor and land are displaced from their present uses by imports, these factors of production will shift quickly and costlessly to other equally profitable uses in different industries. So, if you have a factory producing goods that are getting hammered by imports, you can move your assets easily to another industry with no losses, and make just as much profit in the new use.

The problem is that this is very often an unrealistic assumption, especially in the short run. A particular machine, for instance, is designed to do only a specific task, and to produce a certain kind of good. It can't be used to produce just anything. Workers have specific skills that allow them to do certain kinds of production and they're not very good in a different industry. Or land is good for a certain kind of agricultural crop, but due to climate and rain, it's not good for a different one. You can't grow orange trees in Iowa, where I am from; you grow corn and soybeans, that's about it!

This Ricardo Weiner model says: "Let's then assume factors specificity. Let's assume that factors are immobile. They cannot shift from one use to another. Any capital, labor or land that's in a particular use when trade policy changes or trade rises is stuck there". If one assumes factor specificity, then one derives a different set of trade preferences among owners of capital and among landowners as compared with Stopler-Samuelson.

Under the assumptions of the Ricardo-Viner model, export industries and those that use imports as inputs in their own production process will favor freer trade. However, those firms that are located in import-competing industries and cannot move out of these industries because of factor specificity - they're stuck there - are going to favor protection, because without protection they are going to lose a market share, profit, workers and jobs.

As a result, if you assume factor specificity, like Ricardo-Viner, business divides into internationalist and nationalist camps ­ internationalist free traders and nationalist protectionists.

How does this model apply to coalitions over trade in the nineties? Well, this model does help to explain some things that Stopler-Samuelson cannot. For instance, there was some division within the business community. Large businesses, the national business associations and probably the majority of businesses favored NAFTA. But many smaller firms and import-competing industries opposed NAFTA and lobbied against it. This camp included fruit and vegetable growers, producers of brooms, ceramics, flatware, glassware and luggage - all opposed NAFTA.

One study actually looks very closely at the probable sources of congressional votes (Kahane, 1996; also see Cox and Skidmore-Hess, 1995). Many congressional representatives who opposed NAFTA likely did so because Labor Unions were very well represented in manufacturing industries in their districts. Perhaps Labor Unions gave lots of money to their campaigns. So NAFTA's congressional opponents voted the way Labor wanted, right? This study says that something over half of the 200 House votes against NAFTA were probably inspired by a strong labor movement in the relevant congressional district. On the other hand, the study also finds that there were a large number of congressional representatives who had very little Union support and very little Union presence in their district, and yet they voted against NAFTA. Why? In most cases, they voted against NAFTA because import-competing industries had a strong presence in their districts. Nationalist firms who saw NAFTA as a threat to their profits influenced these congressional critics of NAFTA. This suggests that not all the responsibility for the anti-NAFTA vote can be laid at the doorstep of Labor. A significant business faction also inspired some of the congressional opposition to NAFTA.

The Ricardo-Viner model would suggest that those industries that opposed NAFTA did so due to factor immobility. These firms found it too costly to consider reallocating their capital and other resources to new uses as imports eroded traditional markets. I'm not aware of any studies that have tried to go back and confirm the proposition that anti-NAFTA firms or industrial sectors were characterized by unusually high levels of factor specificity.

There's another interesting division in the business community that is not anticipated in either of the models I've discussed so far. This is a distinction amongst supporters of NAFTA (Cox, 1996). Not all businesses supported NAFTA for the same reasons. One faction that supported NAFTA consisted of what we might call globalists. These were usually very large multinational firms that do business around the world. Such firms support global free trade. They saw NAFTA as a stepping stone to a global free trade agreement, particularly at a time when the WTO and the Uruguay Round of negotiations were stalled. The globalists hoped that the passage of NAFTA would prompt movement toward completion of the Uruguay Round and thus speed a global free trade agreement.

The other faction consisted of regionalists, represented particularly by the US auto and electronics industries. This faction of business was interested not in global free trade, but in creating a somewhat closed economic bloc that would support their competitive strategy vis a vis firms in Japan and Europe. Basically, US auto and electronic industries, facing growing competition in the eighties from Japan and Europe, wanted to move their assembly operations to Mexico and take advantage of the lower wages there to reestablish their price competitiveness and win back markets in the United States. Most of the vehicles assembled in Mexico would be sold back into the US market. Of course, that would be easier if the tariffs on imported cars from Mexico were eliminated under NAFTA.

But it was more than just that. The American auto industry also supported rules, such as the so-called rules of origin or domestic content rules, which said that products not possessing high North American content would not be given free trade treatment. Furthermore, they supported rules favoring firms that already had long established factories in Mexico at the time of the agreement. Say a Japanese firm wanted to come in two years after NAFTA's passage and claim that "Our car is a Mexican product. So, we're going to send it into the US free of trade barriers!". - even though 90% of the value of the production is done in Japan while only 10% is done in a Mexican factory. That would be a great deal for the Japanese. To counter this possibility, US auto firms argued for North American content rules included in NAFTA that would prevent Japanese firms from exporting tariff free into the US market unless a large proportion of the content of the vehicle was actually produced in Mexico.

In the end, NAFTA's terms made if very difficult for Japanese auto firms to set up assembly plants in Mexico in hope of selling cars in the US markets without tariffs. It was much easily for US firms to do so. Regionalist business firms therefore favored NAFTA because it established a discriminatory bloc favorable to their competitive strategy.

The last theory I will review has to do with domestic regionalism, and its effects on US foreign policy and trade policy. Peter Trubowitz (1998) argues that the US economy is regionally diverse. Factor endowments vary considerably from one region to another. Each region therefore specializes in different kinds of production and their exports go to different parts of the world. As a result of this regional specialization, there are important regional differences in trade policy preferences.

One way you can gauge this is to ask: "How vulnerable are industries in a particular region to imports?". The North has a very high level of vulnerability whereas the South and West have much lower levels. So, this allows us to predict that the North, particularly the Northeast, would be more protectionist than the South and the West, because they fear imports more. The analysis is more complex, but that gives you the basic idea.

An analyses of a number of votes taken in the House of Representatives in the 1980s, shows the percentage of representatives from each region who voted in favor of a protectionist policy ­ these were protectionist votes. I have circled every case where the vote for protection exceeded 50%. Notice there are a lot of circles for the Northeast region; there are many fewer circles for the South and the West. So, in the 1980s, the North was much more protectionist.

Now, let's look at our three cases. What do they show, NAFTA, WTO and fast track? NAFTA and fast track fit the pattern. The total vote in favor of NAFTA in the Senate was 62%, yet in the Northeast it's only 58, and in the West it's 68. So, the Northeast was the least supportive of NAFTA. It's not a dramatic difference, but it's noticeable. More dramatic is fast track. I should note that the figures I will report are not the actual vote, but instead a headcount that was done by a private organization based upon how people publicly said they were going to vote if fast track had come to a vote in 1997. So, this is our best estimate. If there had been a vote, this is probably close to what it would have looked like. This headcount shows that, overall, 43% would have voted in favor of fast track, 57% against fast track. In the Northeast, however, only 30% would have supported fast track, whereas in the West 57% would have supported fast track. So this is a fairly clear regional difference.

Now, where the model breaks down - where the pattern doesn't fit - is the case of the WTO vote. Here, the percentage who supported WTO was highest in the North - higher than in the South or in the West. What's the reason for this? My guess - and this is only a guess ­ is that the WTO involved agreements that would open up service industries to trade and the Northeast is very strong in the service sector, making the Northeast more favorable to the WTO treaty. This might have had an influence.

In sum, all three models have some utility. We find divisions along class (Stopler-Samuelson), sector (Ricardo-Viner) and region (Trubowitz), but no single model tells the whole story by itself.

Now, very quickly, let me just talk about the other issue. Why has support for free trade declined in the 1990s? Why do I think that's puzzling? Globalization? The revival of the US economy and the decline of unemployment? The revival of growth and yet we're becoming more protectionist?! The failure of fast track is not the only evidence of low domestic support for trade liberalization. Look at some public opinion data: 61% of Americans oppose fast track; 57% of Americans oppose new trade pacts with Latin America; 87% believe that trade agreements should protect the environment; 73% of Americans believe that labor and environmental issues should be negotiated as part of trade agreements; 56% believe that trade leads to a decrease in US jobs; only 26% believe the US has benefited from NAFTA; 51% believe that America's involvement in the global market just benefits big corporations.

Why this skepticism toward free trade? There are several important factors. Number one, and this is also a discussion about why fast track failed: the Republicans won the Congress in 1994. You might think, "Well, Republicans are free traders. If they won the Congress, wouldn't that improve the chances for free trade?" The surprising answer is no. The reason is that when the Democrats controlled Congress, business money was split between Republicans and Democrats ­ business gave to Republicans because they liked Republicans; business gave money to Democrats because they had a majority in Congress and controlled legislation. Once the Democrats lost control of both the House and Senate in 1994, big business lobbyists said: "Forget it, we're not going to give you money anymore. We'll give all our money to the Republicans because one, we like Republicans and two, now they control things". So, the Democrats now faced a big disadvantage in fund raising. Who then could they turn to for money? LABOR UNIONS!

So the Democrats now depend more on Labor Unions than ever! Before 1994 the democrats could say, "Well, I can afford to offend Labor because I'll pick up some money on the Business side". Now they can't do that. Plus, the Labor Unions after NAFTA explicitly withdrew campaign funding from Democrats who voted in favor of NAFTA ­ they did this very systematically. They knew these kinds of threats were serious. If Labor withdrew funding, a democratic candidate would be penniless in the next campaign.

Another factor working against fast track passage was that Clinton was a lame duck who was hated by the Republicans ­ they just hated Bill Clinton. They tried to impeach him. So some Republicans who might have favored fast track voted against it because they wanted to embarrass Clinton.
 
Second point: growing linkages between trade and social issues like the environment and human rights. In the case of NAFTA, the environmental movement was initially opposed to a free trade deal, but the environmental side agreement won over about half the groups. They said, "We'll support it now because we have this environmental side agreement". Once they saw how the environmental side agreement worked after NAFTA passed, these groups changed their minds because, in fact, the side agreement has been useless as a means of improving environmental performance in Mexico. Also, the money that was promised has not been spent to clean up the border. So, by the time fast track came around, the environmentalists were all against it due to their experience with NAFTA.

Even though the NAFTA side agreements on Labor and the environmental were very weak, the important thing is that they set a precedent. When fast track came around, the NAFTA precedent guaranteed that demands for linking trade to labor and the environment would grow. Moreover, the argument that trade should never be linked to any non-trade issues was undermined by the fact that during the 1990s the US went around imposing economic trade sanctions on almost every country in the world! Sixty countries are now subject to US trade sanctions for all kinds of reasons: some political, some strategic, some economic. So, it's hard to argue that we cannot destroy the purity of trade by attaching it to anything else, when we're attaching it to all these other things.

The third point is the overselling of NAFTA. The supporters of NAFTA said it would be "so great, you're going to love it!". For instance, in Mexico, NAFTA supporters promised economic growth, rising wages, cleaner environment, less illegal immigration to the US, less drug trade, greater political stability, no problems. Now, what really happened is that almost immediately after NAFTA was signed - in fact, the very day that NAFTA went into effect -  the Zapatista rebellion broke out in Chiapas. And then, very soon after, Mexico witnesses two major political assassinations. Also, Mexico's top drug official is arrested because he's playing around with the drug lords. Then the peso crisis forces the US to step in and offer billions of dollars to bail out Mexico! There is no improvement in immigration or in environment. None of the rosy promises came true. It was therefore no surprise that many people concluded that NAFTA was not such a good agreement after all.

A fourth factor is that the US has continually persisted in large and very quickly rising trade deficits. Our trade deficit is approaching a level of 300 billion dollars a year, and that imbalance creates an unfavorable climate for trade liberalization.

In conclusion, I would argue that further major moves toward trade liberalization are going to be difficult. In particular, the free trade in the Americas proposal is going to be very difficult. Consider that the US has witnessed growing protectionist sentiment at a time of 4% unemployment and a 3 to 4% economic growth rate ­ a good economy. Just think what is going to happen when the economy goes bad in the US and people begin to look for someone to blame. And it is always easy to blame foreign imports for bad economic circumstances.

Remember, also, that Americans are more skeptical about free trade agreements with developing countries than with developed countries. Many Americans fear that the low wages levels offer impossible competition. I would argue that fast track definitely won't be brought up again before the 2000 elections. Now, after the elections, it depends a lot on who wins the presidency and the Congress. My guess is that some president is eventually going to get fast track approved again, but the interesting question is: Will they have to attach serious labor and environmental side agreements to get it passed? I think that's a real possibility. If trade becomes permanently attached to labor and environmental issues, it is going to change the whole nature of the trade debate, not only in the US, but, because the US is so important, at the global level more generally.

Bibliography

In addition to citations referenced in the text, this bibliography includes key sources on the domestic politics of U.S. trade policy.

Cohen, Stephen, Joel Paul and Robert Blecker (eds.), Fundamentals of U.S. Trade Policy, Westview Press, 1996

Cox, Ronald W. and Daniel Skidmore-Hess, "The Politics of the 1993 NAFTA Vote," Current Politics And Economics Of The United States, vol. 1 #2/3 1995

Cox, Ronald W., "Explaining Business Support for Regional Trade Agreements: The Cases of NAFTA and CBI," in Ronald W. Cox (ed.), Business and the State in International Politics, Westview Press 1996

Destler, I.M., American Trade Politics, Institute for International Economics, 3rd ed., 1995

Ferguson, Thomas, "From Normalcy to New Deal: Industrial Structure, Party Competition and American Public Policy in the Great Depression," International Organization, Winter, 1984

Frieden, Jeffry, "Sectoral Conflict and U.S. Foreign Economic Policy, 1914-1940." International Organization, vol. 42, 1988

Frieden, Jeffry and Ronald Rogowski, "Internationalization, Institutions and Political Change," in Robert Keohane and Helen Milner (eds.), Internationalization and Domestic Politics, Cambridge University Press, 1996

Goldstein, Judith, Ideas, Interests and American Trade Policy, Cornell University Press, 1994

Gourevitch, Peter, Politics in Hard Times: Comparative Responses to International Economic Crises, Cornell University Press, 1986

Ikenberry, G. John, David Lake and Michael Mastanduno (eds.), The State and American Foreign Economic Policy, Cornell University Press, 1988

Kahane, Leo, "Congressional Voting Patterns on NAFTA: An Empirical Analysis," American Journal of Economics and Sociology, October, 1996, vol. 55, n. 4

Keohane, Robert and Helen Milner (eds.), Internationalization and Domestic Politics, Cambridge University Press, 1996

Milner, Helen, Resisting Protectionism: Global Industries and the Politics of International Trade, Cornell University Press, 1988

Rogowski, Ronald, Commerce and Coalitions: How Trade Affects Domestic Political Allignments, Princeton University Press, 1990

Trubowitz, Peter, Defining the National Interest: Conflict and Change in American Foreign Policy, University Chicago Press, 1998