1. Introduction

Over the next decade, the great challenge for the world economy will be learning how to manage the process of integrating the large number of economies in transition into the global system. Throughout the developing world, countries are transforming their economies by implementing far-reaching, outward-oriented policies. The world market is therefore the fulcrum by which these countries will elevate themselves out of poverty. A friendly global market will be essential to their prospects.

However, this transformation process, which is unprecedented in scope and depth, is advancing in an unfavorable environment. While strong signs of cyclical recovery are emerging, especially in the United States and increasingly in Europe, the long-term prospects for economic growth in developed economies are poor. In the United States, the enduring weakness in productivity growth, the decline in the growth of the labor force, the modest increase in real wages over the last 20 years, and the widening of relative wage dispersion all reveal that the economy needs to find a new driving force. European economies, which suffer from high rates of structural unemployment, are also expected to witness a slowing of the growth rate in the medium to long term because of demographic and productivity changes. Even in Japan, an economy that has sustained rapid growth for nearly 40 years, long-term prospects are now in doubt.

The success of the transition process in developing countries is therefore necessary not only for those economies that are moving toward a market-oriented economic system but also for industrial countries, since the transforming economies represent a new source of economic growth.

However, in light of the structural adjustments required in both developing and industrial countries, managing the process will not be frictionless. The debate that took place in the United States on the North American Free Trade Agreement (NAFTA) is an excellent example of the difficulties associated with the transition process. In this paper, after briefly describing the main features of NAFTA, I will stress six lessons that can be drawn from this experience and that are relevant to similar arrangements.

2. The North American Free Trade Agreement

NAFTA is an agreement to achieve, over 15 years, complete free trade between the United States and Mexico, including trade in extremely sensitive sectors such as textiles and agriculture. This last sector is excluded in the agreement between Mexico and Canada.

It is worth bearing in mind that at the time NAFTA was negotiated, the average tariff rate protecting the Mexican economy was about 10 percent. U.S. tariffs on Mexican products averaged 4 percent and were based on the most favored nation (MFN) status accorded Mexico. But this average declines to around 2 percent when the special treatment under the Generalized System of Preferences (GSP) granted to some goods is taken into account. Therefore, the change in U.S. tariffs envisaged by NAFTA is very small—one reason why all the studies that tried to calculate the economic impact of this agreement obtained rather poor results. However, the real significance of NAFTA goes beyond the mere dismantling of tariff barriers among the countries involved, because NAFTA is about far more than trade. In fact, the main goal of the agreement is to create an environment conducive to investment flows, and meeting this goal would require deep and pervasive structural changes in the Mexican economy.

If NAFTA is compared with the process of integration in Europe, several clear differences emerge. First, NAFTA does not establish any common, supranational institution. Second, it does not require any harmonization of national regulations. It allows instead for mechanisms through which a member country can seek to have the rules enforced. Third, NAFTA does not envisage any mechanism for redistributing wealth among member countries and establishes no social funds. Free access to the U.S. market is the real compensation for the less-developed Mexican economy. Finally, the agreement does not establish the principle of full labor mobility. There is already an immense amount of labor migration between Mexico and the United States, but labor mobility was not the main driving force behind the agreement.

3. What Can Be Learned from NAFTA?

As I have already mentioned, the debate that took place in the United States on NAFTA is a remarkable example of the difficulties associated with the process of integrating an economy in transition—in this case, Mexico—into the global system. On the one side, there was Mexico, which had made remarkable progress in the outward-oriented reform program (led by the private sector) started in the mid-1980s. On the other side, there was the U.S. economy, which had been affected by a prolonged period of slow growth and an increasing polarization of income distribution. Therefore, the considerable debate that emerged about the need for the United States to enter into a free trade agreement with Mexico was not completely unexpected, despite the quite trivial economic impact the agreement might entail, at least in the short run. In fact, the Mexican economy is around one twentieth of the size of the U.S. economy, and Mexican exports to the United States amount to about 1 percent of U.S. GNP.

As for the agreement’s effect on employment, it was equally possible to demonstrate that NAFTA would create around 200,000 jobs and that it would destroy about that number. The impact on employment would therefore be very limited. NAFTA thus provided the U.S. economy with an opportunity to reorient its productive structure with rather small adjustment costs in the short run. In other words, it was an investment with the potential to pay off extremely well in the long run. NAFTA also represented an excellent opportunity for Mexico to deepen its reforms and make them irreversible. If NAFTA was economically important to Mexico, for the United States it was relatively unimportant economically but highly sensitive politically.

What lessons can be learned from the NAFTA debate that can be useful in other processes of integration? The first lesson is the agreement itself. Before this agreement, conventional wisdom held that full regional integration was possible only among countries at similar levels of economic development. NAFTA is proof that it is possible to negotiate a complete free trade agreement for goods, services, and capital among industrial and developing economies. In addition, a striking feature of this agreement is that it is based on the principle of reciprocity and offers the developing country no special or differential treatment. Under NAFTA, Mexico must adhere to exactly the same obligations as the other two partners.

The second lesson concerns the political relevance of the debate that took place in the United States over NAFTA. It is important not to underestimate the role of political forces. Such underestimation has been costly in Europe, particularly with respect to the Maastricht Treaty, suggesting that the process of economic integration requires considerable political groundwork. Even though the economic consequences of NAFTA were potentially rather small for the U.S. economy, there was nothing small about the debate, which was a highly relevant political discussion. The NAFTA debate created a pretext—certainly a context—for different groups in America to promote their particular critiques of what is wrong with the U.S. society and the U.S. economy. Something very similar can in fact be observed in the European discussion. When people consider the issue of economic integration, they do not look at the basic facts impartially; they look into their own affairs, as people did in both the United States and Europe. It is worth noting that while the U.S.-Mexican agreement served as a focal point for political pressure from various “interest groups” in American society, the U.S.-Canadian agreement took place almost unnoticed. This agreement was an important issue in Canada but was ignored in the United States, even though, given the economic importance of the countries involved, its impact on the U.S. economy was likely to be much greater than that of the U.S.-Mexican agreement.

The NAFTA debate brought out a variety of forces. First, there was Ross Perot, who held up (eventually unsuccessfully) the specter of the “giant sucking sound.” With this unforgettable phrase, he wanted to emphasize that the primary consequence of the agreement would be the displacement of jobs from the United States as a result of the lower wages in Mexico. Perot’s argument was simple but very effective. Since in today’s world technology and capital are mobile, and since Mexican assembly plant workers can do the same job as their counterparts in the United States, it would be rational behavior to move almost every economic activity to the south. However, in the Mexican regions where foreign multinationals have set up assembly operations, output per worker is around one seventh of that of the United States, and the wage per worker is also roughly one seventh that prevailing in U.S. manufacturing. So this situation looks much more like the kind of equilibrium found in the traditional Ricardian model than the kind of equilibrium Perot described. This is not to say that it is impossible to achieve in a single plant the productivity levels prevailing in the U.S. economy; certainly this scenario is realistic in some industries. But other factors—such as infrastructure, communications, and so on—are very important in determining productivity levels.

Second, organized labor in the United States made fighting NAFTA the top priority on its agenda. However, the only outcome of this dispute was to reveal how weak U.S. unions actually are, especially compared with those in Europe. Third, there were “right-wingers” like Pat Buchanan, who were particularly fearful about the erosion of U.S. sovereignty and so made it a major issue. There were also “left-wingers,” the so-called consumer advocates such as Ralph Nader, who mobilized around an essentially anticorporate ideology: what is good for a multinational must be bad for ordinary people. Finally, there were the environmentalists, who were divided in their loyalties. Some groups worked together with the administration, playing a crucial role in developing the side agreement on environmental standards. (In fact, on balance, NAFTA is an agreement that will improve the environment.) But there were also other more populist environmental groups that found NAFTA a useful issue for gathering support. Eventually, it turned out that these political forces were too weak to defeat the agreement. President Clinton deserves much credit for having fought hard for NAFTA in the face of very difficult political circumstances. But it was a close battle.

The third lesson of NAFTA is that, as mentioned earlier, the opportunities for transformation associated with the agreement go far beyond the simple removal of trade barriers. If all the attention is focused only on that aspect of the agreement, NAFTA’s significance is overlooked. To better understand what NAFTA is about, the distinction between shallow and deep integration should be borne in mind. Shallow integration is the old paradigm: borders are removed but national sovereignty is retained. To look at NAFTA from the traditional free trade perspective is to enter into an area of ambivalence. The establishment of a free trade area always gives rise to both trade creation and trade diversion. Therefore, regional agreements may or may not improve welfare. (It is not a matter of coincidence that the theory of the second best was put forward in relation to the study of free trade agreements.) But it seems clear that the paradigm of shallow integration is inadequate to describe NAFTA, because this paradigm fails to capture what the agreement was really about: deeper integration, which is something Europeans are very familiar with. Deeper integration involves changes in domestic policies and institutions. As protectionist barriers are removed, the real issues, such as financial regulation, intellectual property rules, labor and environmental standards, are revealed.

When the negotiations on NAFTA started, Mexico was already engaged in the process of reforming its own economy. However, NAFTA gave Mexico the opportunity to deepen these structural changes and make them irreversible. Mexico suddenly found its entire society subject to scrutiny, from its system of government to its respect for human rights; from its labor standards to its banking and financial system; from its environmental regulations to its industrial, agricultural, and transportation policies. For example, before NAFTA there were no intellectual property rules in Mexico, or at the most very weak ones. Today, Mexico is characterized by what the chairman of an important American chemical company called a “world class” intellectual property system. However, it is worth noting that Mexico was willing to change its economic and social system deeply, while other developing countries might not be so eager to change. The process of deeper integration has to be evaluated on a case-by-case basis, because the gains (in the form of improved access) may not be worth the costs (in the form of constraints on its institutions) to all countries.

NAFTA’s fourth lesson is that free trade agreements are not stationary contracts but require revision and deepening. Nations give up sovereignty with great reluctance. In fact, in a typical agreement on economic integration, half the provisions are devoted to dealing with the conflict between national sovereignty and supranational supervision. Europe has overcome this conflict by adopting the principle of mutual recognition. The agreement among the United States, Canada, and Mexico allows each country to retain its own regulations but establishes a system for jointly overseeing the administration of these rules.

Since Canadian and Mexican entrepreneurs were primarily interested in being allowed to sell freely in the American market, they were concerned not only about U.S. tariff barriers but also about U.S. trade rules, in particular antidumping and countervailing duty rules. When the U.S.-Canadian agreement was negotiated, the Canadians sought unsuccessfully to obtain special treatment regarding American antidumping rules and subsidy issues. There were two main reasons why Canada was unsuccessful in these efforts. First, Canada itself has a subsidy policy that is actually more pervasive than the U.S. policy. Second, the U.S. Administration was very reluctant to remove the antidumping provisions without replacing them with some other instrument of control. In the end it was agreed that the United States will continue to apply its antidumping rules but that Canada will have the right to appeal to a panel of representatives of both countries if an antidumping procedure is undertaken against a Canadian firm. A number of such panel appeals have taken place already.

While this system of joint oversight has given member countries a degree of increased control, considerable dissatisfaction remains. NAFTA does not actually put Canadian, Mexican, and U.S. firms on an equal footing, and, if the goal is equal access, similar rules will eventually have to be enforced for all three countries. The approach followed in Europe was to abolish the antidumping rules internally and to establish a common policy with respect to competition. A similar solution was adopted in the agreement between Australia and New Zealand. The United States, Canada, and Mexico will move inevitably toward an analogous configuration; the quid pro quo is replacing the antidumping rules with some kind of antitrust regime.

The fifth lesson concerns the macroeconomic effects of free trade agreements, which are about investment as much as they are about trade. But appropriate attention is usually not devoted to the implications an agreement such as NAFTA has for investment. From the standpoint of the developing country participating in such an arrangement, capital inflows may have both propitious and harmful consequences. On the positive side, the recovery of foreign capital inflows may make the developing country’s external position more viable. In the last few years, for example, Mexico has experienced a dramatic upsurge in capital inflows driven not only by the reform process already in place but also by expectations that the NAFTA negotiations would be successfully completed. This influx of foreign capital has helped Mexico shift from a situation in which the burden of servicing its external debt forced the country to run trade surpluses to a more sustainable situation that involves absorbing capital.

On the negative side, significant inflows of capital inevitably put upward pressure on the real exchange rate of a developing country, as happened not only in Mexico but also in Spain and Portugal. This appreciation of the real exchange rate exacerbates the impact of adjustment, since firms have to adjust not only to the removal of tariff barriers but also to a stronger currency. However, a stronger currency may introduce an important element of discipline into the economic system, thus fostering the anti-inflation process. In any event, these transformation processes entail a high degree of uncertainty about the level at which the exchange rate is going to settle.

The macroeconomic consequences of a free trade agreement for a developed country—a capital exporter—may well be that the adjustment is actually much easier than has traditionally been predicted. The capital outflows will in fact be associated with a real depreciation of the currency of the developed country, a development that in turn mitigates some of the adjustment difficulties. In the case of the United States, the dollar has depreciated vis-à-vis the Mexican peso, promoting U.S. exports to that market. In addition, the shift of the Mexican trade balance from a net surplus to a deficit suggests that, on balance, the employment impact has been more positive in the United States than in Mexico.

The final lesson to be drawn from NAFTA is that devil is in the details. The trade agreement between the United States and Mexico was heavily driven by political considerations. It is not surprising, then, that the agreement is a very thick document of about 1,600 pages, when only one page would have been sufficient to declare a free trade area among these three countries. NAFTA reflected the influence of powerful corporate lobbies that played an important role in framing the rules, some of which were crafted in an obnoxious way. The most deleterious are the rules of origin, of which those governing textiles and automobiles are the most striking. For example, in the case of textiles, a so-called triple transformation rule—sometimes even a quadruple transformation rule—applies. This rule states that exports of a particular garment between, say, Mexico and the United States are not subject to tariffs if three conditions are met, namely, the fiber is made in North America; the fabric is woven in North America; and the garment is sewn entirely in North America. This type of rule can have serious diversionary effects. Canadian, Mexican, and U.S. textile firms will no longer find it convenient to import fibers from Europe because products made from imported fibers cannot be exported within the NAFTA area free of charge. Therefore, while there is complete trade liberalization inside the NAFTA region, in fact there is diversion taking place with respect to the outside world. The rules for the automobile sector are not as draconian as those for the textile sector, but in any case they were carefully crafted to strengthen the competitive positions of the big three automakers in North America. Despite the enthusiastic rhetoric surrounding NAFTA, such rules may have a powerful impact on trade and could wind up creating a permanent source of friction with the rest of the world.

4. Conclusions

This article has described the six major lessons that can be drawn from the NAFTA experience. First, it is politically possible to reach an agreement on free trade between developed and developing countries. Second, the major obstacle to the conclusion and implementation of these agreements is not their economic consequences, which are not as great as people tend to believe, but rather the political reverberations. Third, because of the political pressures, free trade agreements are not simply about free trade. As a matter of fact, the Clinton Administration correctly assumed that the U.S. Congress would not approve NAFTA unless some side agreements on very sensitive issues, such as labor and environmental standards, were established. Such side agreements require some major changes in the domestic institutions of developing countries. For better or worse, the impact on these societies will be considerable. The fourth lesson is that free trade agreements are not stationary contracts but will inevitably require revision and deepening. Fifth, their macroeconomic consequences cannot be ignored. And finally, careful attention has to be devoted to the details. This last point suggests an important role for the GATT in policing the details of agreements such as NAFTA. In the case of the United States and NAFTA, the most troublesome of the details are the rules of origin. In the case of custom unions—in Europe, for example—the most troublesome are the antidumping rules and the way they are administered.