The North American
Free Trade Agreement
as Negotiated: A U.S.
Perspective

Sidney Weintraub

Introduction

THE NORTH AMERICAN FREE TRADE AGREEMENT (NAFTA) as negotiated is a massive document of thousands of pages. Much of the verbiage deals with exceptions to free trade. In some sectors, such as textiles and apparel and agriculture, complex and lengthy legal contortions were needed to move from protection on a global scale to less onerous restrictions applicable to trade in North America.

Consequently, the agreement is not an easy read. NAFTA is sure to keep many lawyers, economists, and public servants gainfully occupied for years making interpretations that will have modest social welfare benefits. Fortunately, the complexities will diminish over time. North America will have something approaching free trade in goods, including agricultural commodities, after a transition period of 10 years, but up to 15 years in some sectors and, in a few cases, even longer. The agreement could have been "clean" and relatively brief-certainly shorter than it is-if it contained fewer exceptions to free trade and freedom of cross-border investment. Yet, despite the derogations, the agreement does move toward greater freedom in economic relations among the three countries.

The discussion that follows looks at the agreement from a U.S. perspective. First, the structure of the agreement is examined. This is followed by an analysis of what is omitted and what is most contentious. The two final sections will contain this observer's analysis of the likely consequences of the agreement on the U.S. economy and the nature of the debate that is unfolding over approval of the agreement by the U.S. Congress. There is a brief concluding comment.

Structure of the Agreement

The agreement has a preamble and 22 chapters. [These are objectives, general definitions, national treatment and market access, rules of origin, customs procedures, energy, agriculture, emergency action, standards-related measures, government procurement, investment, cross-border trade in services, telecommunications, financial services, competition policy monopolies and state enterprises, temporary entry for business people, intellectual property, publication notification and administration of laws, review and dispute settlement in antidumping and countervailing duty matters, institutional arrangements and dispute settlement procedures, exceptions, and final provisions.] The phasing of tariff reductions is set forth in a separate, lengthy document.

The chapters have their own annexes to deal with particularly contentious themes or where the text needs clarification or elaboration. Thus, there are detailed annexes on the automotive sector and textiles and apparel in chapter 3 (national treatment and market access); separate annexes in chapter 10 (government procurement) clarify the extent of obligations undertaken, such as listing the government entities for which procurement obligations are liberalized; and in chapter 12 (cross-border trade in services), annexes set forth the extent to which professionals and providers of land transportation can operate in countries other than their own. Chapter 7 on agriculture has two subchapters, on market access and on sanitary and phytosanitary measures.

In addition to annexes to the specific chapters, there is an extensive section of reservations by each country to the obligations stated in the text of the agreement for chapters 11 (investment), 12 (cross-border trade in services), and 14 (financial services). These are primarily what are called "transitional exceptions." For example, Mexico's gradual opening of its financial services market (including banking, insurance, and investment services) to foreign investment is set forth in this part of the agreement. By contrast, the nonapplicability to Mexico of provisions on foreign investment in oil and natural gas and for sharing supplies of these products with the other two countries during periods of shortage is written into the agreement itself (in chapter 6, energy). The nonapplicability in this case is intended to be durable. Canada's exclusion of cultural industries from the provisions of NAFTA is written right into chapter 21 (exceptions) as an annex. This, too, is not intended to be a transitional exception.

The use of separate annexes for transitional exceptions should clarify the negotiating context for countries which later seek accession to NAFTA. If transitional issues can be dealt with outside the core text of the agreement, this is intended to make clear that accession negotiations should deal only with the transitional provisions. Nevertheless, Mexico was able to negotiate permanent derogations from the text of the Canada-U.S. free trade agreement (FTA), which was the model used in the NAFTA negotiations. The oil exclusion noted above is a particularly significant example of this. There is no way to guarantee that other acceding countries will be unable to do the same in the future.

Mexico entered more exceptions and reservations, both in sheer numbers and in significance, than the other two countries combined. The most important exception is in the energy chapter. Indeed, the first sentence of this chapter reads as follows: "The Parties confirm their full respect for their Constitutions." (Article 601(1)). This makes clear that the Mexican state reserves for itself exploration and exploitation of crude oil and natural gas, and various activities that flow from this, including refining and foreign trade in these products. The Mexican reservations that are intended to be transitory, as in automotive trade and investment and in the provision of financial and other services, are best seen as Mexico's way of gradually opening itself to free trade. [Reading through these Mexican reservations, one gets a sense of how Mexican nationalism developed. For example, in the reservations under chapter 12 (cross-border trade in services), one learns that only Mexicans "by birth" may serve as ship captains, pilots, ship masters, machinists, mechanics, etc.] Mexico's tariff reduction schedule is slower than for the other two countries. These differences in trade and investment opening reflect Mexico's less-developed status coupled with its more closed economy as the transition gets under way. For the most part, all three countries will be at equal levels in trade liberalization at the end of the transition period, or generally after 10 years.

The text of the agreement incorporates provisions of the General Agreement on Tariffs and Trade (GATT) throughout. The three countries, all of which are contracting parties to the GATT, resolve in the preamble to build on their respective rights and obligations in the GATT. This is done again in chapter 1 (objectives), in chapter 3 (trade in goods) under which various GATT articles are incorporated into NAFTA, chapter 6 (energy and basic petrochemicals), and elsewhere. The intent is to emphasize that NAFTA is a supplement, not a replacement, to the rights and obligations of the three countries in the GATT. However, article 103 states that in the event of an inconsistency between NAFTA and other agreements, the provisions of NAFTA shall prevail, unless otherwise specified. Various international environmental agreements are also incorporated into the NAFTA and, in these cases, NAFTA is subordinated in the event of an inconsistency.

As long and as detailed as it is, the text of the agreement is not the final word on the structure of the arrangement. There are provisions in almost every chapter for the formation of committees or working groups for further work on aspects of the agreement. There is thus a committee for consultation and problem solving for trade in goods, on worn clothing, trade in agriculture, sanitary and phytosanitary measures, standards, small business, financial services, and private commercial disputes. Working groups are to be created on rules of origin, agricultural subsidies, and for temporary entry for business persons. These committees and working groups are merely those listed in chapter 20 (institutional arrangements and dispute settlement procedures). Other groups, less formal in structure, are called for throughout the agreement.

An agreement as comprehensive in subject matter as NAFTA requires continual consultation to work out the problems that will inevitably arise. Some of these problems will deal with policy; a free trade commission made up of cabinet level representatives or their designees is established for this purpose. Others will deal with implementation matters. Rosters of experts will have to be put together under the dispute-settlement provisions of the agreement. Like the Canada-U.S. FTA, NAFTA has a variety of dispute-settlement provisions. Chapter 19 authorizes the replacement of judicial review of antidumping and countervailing duty determinations by binational panels, similar to what now exists in chapter 19 of the Canada-U.S. FTA. Chapter 20 of NAFTA authorizes the establishment of arbitration panels to make recommendations on disputes not resolved through the good offices, conciliation, or mediation of the commission. The agreement calls on the commission to establish an advisory committee on private commercial disputes. Chapter 11 establishes a procedure for settlement of cross-border investment disputes. Disputes arising under the chapter on financial services will have their own settlement procedures.

The purpose of the agreement is to facilitate increased interaction among traders and investors in the three countries. In addition, the framework of the agreement requires much consultation among government officials, in many cases assisted by private business people and technicians, to clarify ambiguities that will arise. If the agreement succeeds in its primary purpose, to facilitate trade and production in North America, this will inevitably lead to the deepening of interchange among nationals in all three countries.

The agreement, if approved by the three legislatures, will enter into force on January 1, 1994. Any party can withdraw six months after providing written notice of its intention to do so. The accession clause leaves open membership to any other country, regardless of location, subject to a prior negotiation and the necessary approval procedures in the existing member countries.

The agreement does not cover all the themes that have arisen in the debate on free trade. The Mexican authorities, when they took their initiative for free trade with the United States, wished to include provisions on migration. Other than for chapter 16 (temporary entry for business people), this theme is not included.

U.S. and Canadian labour unions and legislators sympathetic to their position wanted inclusion of workplace standards. This theme is not included in the agreement, although two side arrangements were concluded between Mexico and the United States to deal with this issue. The first is a memorandum of understanding on labour cooperation, essentially a procedure for the exchange of information, and the second an agreement which establishes a consultative commission on such issues as labour laws, workplace conditions, and labour standards applicable to migrant workers. [The text of this agreement is included in a "Report of the Administration on the North American Free Trade Agreement and Actions Taken in Fulfilment of the May 1, 1991 Commitments," dated September 18, 1992. This document accompanied the submission of the text of the NAFTA to the Congress by the Bush administration and was designed to demonstrate that the commitments made by President Bush at the time fast track was extended were met.]

Environmentalists in all three countries made a successful effort to include provisions on this theme in the agreement as well as in parallel understandings. Other than passing references to environmental issues in article XX of the GATT, this is the first major trade agreement that deals consciously and comprehensively with this theme. [Article XX(b) of the GATT permits contracting parties to take measures "necessary to protect human, animal or plant life or health"; article XX(g) permits such measures "relating to conservation of exhaustible natural resources if such measures are made effective in conjunction with restrictions on domestic production or consumption."] As such, NAFTA may well be a precursor for inclusion of such provisions in other agreements, including the GATT itself in future rounds of trade negotiations.

Trade agreements in the past generally dealt exclusively with trade issues. The Treaty of Rome establishing the European Economic Community was much more comprehensive, but this was seen from the outset as setting in motion a political process in Western Europe. The Canada-U.S. FTA, by contrast, was seen as a trade agreement without the deep political content of the Treaty of Rome. The free trade rather than the customs-union option was chosen deliberately for this reason. Nevertheless, the Canada-U.S. FTA dealt not only with trade, but also with related issues. The NAFTA is even more comprehensive in its inclusion of environmental and intellectual property issues. The NAFTA covers agricultural trade more fully than did the Canada-U.S. FTA. One can quarrel with omissions from the NAFTA, but it is not possible to argue that this is a narrow trade agreement.

Omissions from the Agreement

Labour unions in the United States argue that the biggest omission is the failure to deal with Mexico's enforcement of its safety and sanitary standards. This issue will be discussed later in the section on the debate in the United States on whether and with what conditions to approve the agreement.

Energy

Mexico's exceptions in chapter 6 (energy and basic petrochemicals) are a major omission. Since the expropriation in 1938 of foreign oil properties and the formation of Petróleos Mexicanos (Pemex), the Mexican state has had exclusive control over what in the agreement are called "strategic activities and investment" in the oil and natural gas sector. [In annex 602.3, this reservation of activities to the Mexican State lists the following: "(a) exploration and exploitation of crude oil and natural gas; refining or processing of crude oil and natural gas; and production of artificial gas, basic petrochemicals and their feedstocks; and pipelines; and (b) foreign trade; transportation, storage and distribution, up to and including first hand sales of the following goods: crude oil; natural and artificial gas; goods covered by this Chapter obtained from the refining or processing of crude oil and natural gas; and basic petrochemicals."] Those familiar with Mexican history are aware of the sensitivity of this issue. The expropriation of the foreign oil properties is seen even today as a second declaration of independence. Mexico, in the last few years, has been willing to alter other parts of its constitution, for example to modify the ejido system of land tenure, itself a highly emotional issue, and to permit restoration to private buyers of the commercial banks nationalized in 1982; but state control of the oil industry was considered sacrosanct. Or, at least it was by the government and the Mexican negotiators. In the end, this position was accepted. The inability to alter the Mexican position on petroleum risk contracts was a major disappointment to U.S. negotiators.

The issue is not trivial. U.S. imports of crude oil from Mexico were US$4.3 billion in 1991, about 14 percent of all imports from Mexico. [U.S. Department of Commerce, International Trade Administration, U.S. Foreign Trade Highlights 1991, (Washington, D.C., 1992) p. 129.] The level of trade is less germane than the security of supply from a neighbour with whom a free trade agreement is being contemplated. Little investment has been made in recent years in oil and natural gas exploration and exploitation in Mexico because of the lack of resources during the difficult economic stretch in the 1980s. The main request of the U.S. negotiators, therefore, was to find some way to permit foreign investment in exploration and exploitation. This is possible in other countries in which the state controls this resource. Mexico was prepared to offer service contracts, which it has always been willing to do. It was not prepared to offer contracts under which the investor would take the risk but also share in the benefits that might accrue. Mexico argued that a risk contract implies some share of ownership of the profits and would thus violate the constitution. The most the Mexican negotiators would concede was a provision that reads: "The Parties shall allow state enterprises to negotiate performance clauses in their service contracts." It remains to be seen what this will mean in practice.

In times of supply restriction, Article 605 stipulates that any party to the agreement cannot reduce the proportion of total exports of energy or basic petrochemical products to the other parties. This is based on the proportion during the previous 36-month period for which data are available. Any restriction must not disrupt normal channels of supply of importing countries. The same article prohibits raising export prices for these products higher than domestic prices. Mexico exempted itself from this article. It applies, therefore, only in oil and natural gas trade between Canada and the United States.

Investment

One of the main reasons why Mexico took the initiative for free trade with the United States was to attract foreign investment. Indeed, much investment has been flowing into Mexico in recent years. Foreign investment, direct and portfolio, increased from US$3.5 billion in 1989, to US$4.6 billion in 1990, and then to US$12.3 billion in 1991. Of this, more than a third, or US$4.8 billion, was direct investment in 1991. [Banco de México, Indicadores Económicos. This is a publication for which new, updated loose leaf sheets are published monthly. The data above come from the May 1992 sheets, p. IV-2.] Mexico was attractive to foreign investors because of a combination of its own, open economic policy, plus the attraction of the large North American market for its industrial output.

Under these circumstances, one would have expected Mexico to eliminate its restrictions on foreign direct investment. This was mostly done, but not fully. Mexico retains a screening process for acquisitions of more than 49 percent of ownership of an enterprise controlled by Mexican nationals, with a modest threshold of US$25 million at first but rising to US$150 million after 10 years. After that, the threshold will be adjusted for growth in Mexican gross domestic product "but in no case will the threshold to be applied exceed that of Canada. [NAFTA agreement, Reservations, Annex I, Schedule of Mexico, pp. I-M-6 and 7.]" By agreeing to a screening and threshold for U.S. foreign direct investment in the Canada-U.S. FTA, the United States set itself up for similar asymmetrical treatment by Mexico. Giving up a clean agreement on foreign investment was a concession to the Mexicans that U.S. negotiators would have preferred not to make.

In the investment chapter, the three parties agree to eliminate performance requirements. These refer to mandatory export levels, achievement of a given level of domestic content, forcing investors to purchase local goods and services, relating the level of permitted imports to the value of exports achieved, or relating domestic sales that are permitted to export accomplishments. However, Mexico will eliminate these performance requirements only gradually. It will take 10 years for full elimination of performance requirements in the automotive sector. Once again, Mexico took its cue from the example in the Canada-U.S. auto agreement. These are not crucial omissions or limitations because they disappear over a reasonable transition period.

Agriculture

The provisions on trade in agriculture set up, in effect, two separate bilateral agreements for the United States, one with Mexico and the other with Canada. The rules affecting tariffs and nontariff barriers in U.S.-Canada trade in agricultural products are generally those contained in the existing FTA between the two countries. There are trilateral provisions on domestic support and export subsidy issues; a working group on agricultural subsidies will be established. A single agreement on agricultural trade in North America would have been preferable.

However, the agricultural understandings between Mexico and the United States are quite impressive, especially in light of the protectionism that has prevailed for decades in this sector. The main impediment to coming to closure in the Uruguay round in the GATT has been precisely the unwillingness of many countries-the European Community (France) and Japan in particular-to fully address issues of agricultural protection. Mexico and the United States accomplished much more in freeing bilateral agricultural trade than almost all observers had expected.

Under the agricultural chapter (chapter 7), Mexico and the United States agree to convert all nontariff barriers in their agricultural trade to tariff-rate quotas as opposed to ordinary quotas. These tariff quotas call for zero tariffs within the quota limits with over-quota amounts subject to regular tariffs. These over-quota tariffs will be set at a rate deemed to be equivalent to the protection offered by the current quota limitations and then decline to zero, generally over a 10-year period. The phase-out period for tariffs will be 15 years for some sensitive products-corn and dry beans for Mexico, orange juice and sugar for the United States. The 15-year phaseout for these four products represents significant concessions on both sides. These products have long protective histories and the concessions as negotiated will face stiff internal opposition in the two countries.

While this discussion examines the agreement from the vantage of the United States, a word is useful on the commitment undertaken by Mexico. Corn and beans are the main staples in the Mexican diet, as sugar and frozen orange juice are not for the United States. Mexico's agricultural policy has long been based on supporting the production of these two products, subsidizing urban consumers of these products, and limiting imports to the extent possible. Mexico is still about one-third rural and by far the largest cohort of growers and labourers for any product are engaged in corn production in rainfed areas of Mexico. The owners of minifundio and the daily labourers involved in corn production number from three to four million people. Mexico is thus taking a gamble on agricultural policy far more audacious than anything being undertaken by Canada or the United States, and certainly more bold than anything contemplated in other industrial countries.

Mexico's domestic program during the 15-year transition period to free trade in corn calls for assistance to establish rural-based industries and to facilitate a gradual migration out of the villages where the bulk of this population lives. Some concern has been raised that this liberalization will lead to mass emigration not just to Mexican cities, but to the United States. Hinojosa and Robinson estimate that the elimination of protection for corn will lead to the emigration of 800,000 Mexicans to the United States. [Raul Hinojosa-Ojeda and Sherman Robinson, "Alternative Scenarios of U.S.-Mexican Integration: A Computable General Equilibrium Approach," working paper 609, Department of Agricultural and Resource Economics, University of California, April 1991.] Even if they are correct, their analysis omits the fact that the population involved is probably the poorest in Mexico: people with low incomes, inadequate diets, little health care, and few educational opportunities. The Mexican authorities concluded that a restructuring of corn policy was needed-free trade or no free trade-and NAFTA offered the opportunity to accomplish this.

One final comment on agriculture is warranted, this time dealing with U.S. sensitivities. U.S. sugar policy has long been highly protectionist. Sugar growers and their supporters have been able to maintain U.S. prices well above world prices and in the process distribute the cost to foreign countries, such as the Dominican Republic, the Philippines, and Australia-all U.S. allies. The agreement also liberalizes the U.S. sugar import program. The formula is complex; however, at the end of a 15-year transition period, restrictions on sugar trade between Mexico and the United States will be eliminated. [There is a modest exception to this statement in that imports of Mexican sugar for U.S. sugar re-export programs will remain subject to U.S. most-favoured-nation tariff rates.] This, of course, will provide no benefit to other countries that export sugar to the United States.

Principal U.S. Sensitivities

The most contentious aspect of NAFTA for the United States is not what is in the agreement itself, but the very fact of having free trade with a developing country whose wages are so much lower-about 14 percent of the U.S. level for production workers in manufacturing in 1991. [The percentage figure is from the U.S. Bureau of Labour Statistics. It is calculated on the basis of the market exchange relationship between the U.S. dollar and the Mexican peso.] This competitive concern is at the heart of the debate in the United States on approval of the agreement.

It is precisely this fear of low-wage competition that makes three aspects of the agreement highly contentious. These are the provisions on trade in the automotive industry, textiles and apparel, and fresh fruits and vegetables. The first two of these, automotive and textiles and apparel, are covered primarily in separate annexes to chapter 3 (national treatment and market access). Automotive trade between Canada and the United States will continue to be guided mostly by the terms of the 1965 auto pact, as subsequently amended.

The U.S. auto industry has been severely buffeted in recent years. Profitability, market share, and employment have all gone down. It is only in recent years that U.S. producers have begun to correct their inadequacies, but these changes will require further lowering of employment levels. This was precisely the basis for a strike by a local of the United Auto Workers (UAW) at a General Motors tool and die plant in Lordstown, Ohio, in late August and early September 1992. GM had announced plans to close this plant, eliminating 240 jobs. The strike was intended to be a protest not only about closure of the Lordstown operation, but against GM's announced intention to trim a further 74,000 jobs from its payroll. The nine-day strike was settled after causing much disruption to GM production at other plants that depended on the output from Lordstown, but the deeper issue remains. [New York Times, September 6, 1992, p. l, has a discussion of the strike.

Problems in the auto industry are related much more to competition with Japanese and European producers-many of whom have established U.S. production facilities-than with Mexico. But growth in the Mexican industry has not been negligible. Complete motor vehicles and parts are now the most important U.S. imports from Mexico. Taken together, their value was higher than crude oil imports in 1991. Motor vehicle parts are the most important imports from maquiladora plants in Mexico. [These imports are based on U.S inputs sent outside the United States and, under U.S. harmonized tariff schedule subheadings 9802.00.60 and 9802.00.80, pay the duty on the value added outside the United States when they re-enter the country.] The U.S. labour unions believe that the maquiladora plants, because they are based in large part on cheap Mexican labour, deprive U.S. workers of jobs. A common critical metaphor for NAFTA used by U.S. labour unions is that it will lead to the maquiladorization of Mexico-that is, make Mexico a haven for low-wage jobs and require the Mexican authorities to keep wages low in order to attract this kind of foreign investment. There is no convincing evidence that the Mexican authorities are deliberately seeking to keep wages down. Indeed, real wages have been rising in the past two to three years after declining for most of the 1980s along with the general economic decline in Mexico, but the charge has wide currency nevertheless.

The automotive provisions in the agreement as they relate to Mexico have four interconnected elements: reduction in import barriers; liberalization of investment laws and regulations; the elimination of performance requirements by Mexico; and rules of origin. The first three can be summarized rapidly.

U.S. tariffs on passenger cars will be eliminated immediately upon entry into force of the agreement (these duties are now only 2.5 percent ad valorem); the tariff on light trucks will be reduced immediately to 10 percent (the most-favoured-nation, or MFN, rate is 25 percent), and then phased out over five years; U.S. tariffs on other vehicles will be phased out over 10 years. Mexican automotive tariffs are higher than those in the United States. Mexico will reduce its tariff on passenger automobiles by 50 percent immediately (the current MFN rate is 15 percent) and phase out the remaining tariff over 10 years; the tariff on light trucks will be reduced by 50 percent immediately and be phased out over five years; and the tariffs on all other vehicles will be phased out over 10 years. All three countries will eliminate duties on parts in three stages: some immediately; others over five years; and a small proportion over 10 years. The agreement calls for Mexico to gradually phase out its restrictions on the import of used vehicles. The full elimination of prohibitions for used cars of all ages will not occur until January 1, 2019.

The investment provisions in this sector are governed generally by the investment provisions in chapter 11 of the agreement. There are two important exceptions. Foreign investors are now limited to a maximum of 40 percent equity in auto parts production, although exceptions are possible. Under the agreement, this will rise to 100 percent immediately for what are called "national suppliers" in Mexico's auto decree (that is, those firms that meet particular levels of national value added in their production) and to 49 percent for other auto parts producers. The latter will rise to 100 percent after five years.

Mexico's performance requirements will be reduced gradually, as noted earlier. The automotive provisions of the agreement call for Mexico's national value-added requirements for auto manufacturers to decline gradually to 29 percent by the year 2003, and trade balance requirements will be eased over the same period. The agreement calls for a review of the status of the North American automotive sector no later than December 31, 2003.

Finally, perhaps the most contentious issue in the automotive negotiations was setting the rules of origin-how to calculate them and what percentage should apply. A free trade agreement permits each member country to maintain its own tariffs against nonmember countries. Consequently, to avoid the problem of transshipment of imported items from the low tariff to the high tariff member of an FTA, a rule of origin is needed to determine what constitutes a national or regional product eligible for duty-free treatment. However, this technical matter can be transformed into a protectionist one by increasing the level of regional content required for an item to receive preferential treatment. This protectionist aspect took its most severe form in the NAFTA in two sectors, automotive and textiles and apparel.

The rules of origin for automotive goods are complex. (They are set forth in chapter 4, rules of origin.) Simplifying for the purpose of this discussion, the regional value content that makes a vehicle eligible for free trade under NAFTA will rise in two steps to 62.5 percent after the year 2002 for passenger automobiles and light trucks and engines and transmissions for such vehicles, and to 60 percent for other vehicles and parts. This is under the net-cost formula for calculating regional value. [The agreement provides two methods for calculation of regional value, a net cost method and transaction value.] To avoid disagreements of the type that arose in calculating the regional value content of Hondas shipped from Canada to the United States, the elements that go into the calculation are spelled out more precisely than was the case in the Canada-U.S. FTA. In addition, the three countries have said they will trace the value of automotive parts imports from outside the NAFTA region to improve on the accuracy of the calculation.

The purpose of the high regional value content is to prevent automotive producers other than those now producing in the three countries from taking advantage of preferential provisions of the NAFTA by merely establishing assembly plants in one of the three countries-particularly Mexico-and then enjoying duty-free treatment in the entire North American market. The rules of origin make it necessary to carry out much of the production of parts in the three countries to enjoy duty-free treatment for their vehicles.

The rule of origin in the Canada-U.S. FTA is 50 percent. However, the nature of the calculation is different from NAFTA and there was no deep tracing provision. Thus, a precise comparison cannot be made between the Canada-U.S. FTA 50 percent rule and the NAFTA 62.5 percent provision. It is by no means clear, either, how well deep tracing will work or how much it will cost. The big three U.S. auto producers had fought hard for a content provision of 65 percent for passenger cars. The Canadian negotiators forced the compromise to 62.5 percent.

It is by no means evident, either, how NAFTA will affect the automotive industries in the three countries. Mexico, over the transition period, will have to discard many of the elements that helped it develop its industry-high protection, forcing foreign producers to balance auto sector imports with exports, and insisting on minimum levels of Mexican content. However, Mexico will still retain many competitive advantages, such as lower wages combined in some plants with levels of productivity comparable to plants in Canada and the United States. The Canadian weakness is its parts sector. The U.S. disadvantage is the incompleteness of its restructuring in this industry, especially for General Motors, coupled with higher wages. The auto producers believe that they need the Mexican production-or if not Mexico, some other low-wage location-for those operations with a high labour content if they are to remain competitive with Japanese and European production. The UAW, by contrast, sees this as a low-wage strategy that will inevitably deprive its members of jobs.

Some analysts of the automotive industry believe that North American production and sales integration will be a net gain for the U.S. and Canadian industries. [James P. Womack, Daniel T. Jones, and Daniel Roos, The Machine that Changed the World: The Story of Lean Production (New York: Harper Perennial paperback, 1991, from original Macmillan edition, 1990), pp. 264-267.] The thrust of the argument is that the Mexican market will itself grow and that a division of production will occur under which cheaper, entry-level cars and trucks will be produced in Mexico using parts from the maquiladoras in northern Mexico, while larger cars and trucks will be produced in Canada and the United States. This scenario requires changing the U.S. corporate fuel economy (CAFE) regulations to count small autos produced in Mexico by U.S companies as domestic U.S. products. This change is included in the text of the agreement.

The second sector of great contention in the United States is textiles and apparel. This is really two separate industries, the mill sector producing textiles and the apparel sector. These have long been protected industries in the United States and other industrial countries, as witness the durability of the import quota structure of the multifiber arrangement (MFA) and its predecessors.

The sensitivity of opening the U.S. market for textiles and apparel rests on several foundations. The two industries together have 29,000 plants spread throughout every state in the union, ship about $64 billion each in value (1990 data), and together employ almost 1.4 million people (1990) [Gary Clyde Hufbauer and Jeffrey J. Schott, North American Free Trade: Issues and Recommendations (Washington, D.C.: Institute for International Economics, 1992), pp. 264-266.]. Yet this is a case in which the Mexican industries are as deeply concerned, perhaps more so, than their U.S. counterparts. The cost of mill production in Mexico is considerably higher than in the United States because of inferior inputs, outdated technology, and capacity underutilization. [Ovidio Botella C., Enrique García C., and José Giral B., "Textiles: Mexican Perspective," in Sidney Weintraub, Luis Rubio F., and Alan D. Jones, eds., U.S.-Mexican Industrial Integration: The Road to Free Trade (Boulder, Colorado: Westview Press, 1991), pp. 193-220.] Because of its labour-intensive nature, Mexico has a cost advantage in apparel production, but the industry is nevertheless in a precarious condition. This is an industry in which workers in both countries share similar concerns.

The provisions of NAFTA in this combined textile and apparel sector take precedence over those of the MFA. The NAFTA language is extremely detailed and complex for textiles and apparel, in part because the agreement must deal with liberalization, item by item, using the starting point of the MFA. Complexity of protection has given way to complexity of liberalization, mainly because it is staged and deals with a variety of tariff and nontariff barriers on hundreds of individual items. In addition, the rules of origin are, if anything, more burdensome (more protectionist) than in the automotive sector. Three aspects of the agreement will be noted here: tariff and quota reduction; safeguards, because these are important in an industry with so many workers; and rules of origin. What follows is a simplified description of the agreements in these three areas.

Tariffs are to be phased out by all three countries over a maximum period of 10 years for products that meet the rules of origin. The United States will immediately remove import quotas on those textile and apparel products produced in Mexico that meet the rules of origin and will gradually phase out quotas, generally over 10 years, on the products that do not meet those rules.

The safeguard provisions have their own complex rules. For goods meeting the rules of origin, safeguard measures must take tariff form. The language as to when a safeguard can be invoked is the usual artful language for these emergency actions. [If the product "is being imported into the territory of another Party in such increased quantities, in absolute terms or relative to the domestic market for that good, and under such conditions as to cause serious damage, or actual threat thereof, to a domestic industry producing a like or directly competitive good, the importing party may, to the extent necessary to remedy the damage or actual threat thereof...", which is then followed by the allowable steps. Annex 300-B, Textiles and Apparel Goods, Section 4(1).] Under these circumstances, for goods meeting the origin criteria, emergency action can consist of suspending further tariff reduction or increasing the rate of duty to the MFN rate in existence at the time or to the rate when NAFTA went into effect, whichever is lower. This action normally can last for no more than three years and may be taken only once during the transition period. When the rule of origin is not met, then emergency action can take quota form, again normally to last for up to three years. There are other specific limits on the quota action that may be taken.

Finally, the rule of origin is based on the principle of yarn forward. A product must be made of NAFTA cloth, and of textiles made of NAFTA yarn. There are exceptions for certain fabrics in short supply in North America, such as silk and linen. Other exceptions permit tariff rate quotas (that is, preferential treatment up to a given quota level) for yarns, fabrics, and apparel that do not meet the rules of origin. The tariff quotas for Canadian products in the Canada-U.S. FTA have been increased. The whole structure of textile and apparel rules of origin will be reviewed no later than January 1, 1998.

This is one sector (or two, if the two industries are treated separately) in which significant trade effects may occur. If the MFA is retained, one should expect diversion of production, and hence exports to the United States and Canada, from other suppliers to Mexico. Hufbauer and Schott anticipate that U.S. exports will increase in unfinished goods and that Mexico will increase its exports of finished products. This, they believe, will come from the diversion of investment to Mexico to exploit its advantage if the MFA remains in place. They speculate that Mexican exports of textiles and apparel to the United States may reach US$3 to US$5 billion in 10 years, an increase of US$2 to US$4 billion. [Hufbauer and Schott, p. 278.] The U.S. International Trade Commission also expected Mexican exports of textiles and apparel to the United States to increase. While the ITC did not provide precise estimates, its tone is less pessimistic from the U.S. trade balance view than that of Hufbauer and Schott. [U.S. International Trade Commission, The Likely Impact on the United States of a Free Trade Agreement with Mexico , USITC publication 2353 (Washington, D.C.: USITC, 1991), pp. 4-38 to 4-41.]

Import liberalization in the agricultural sector was most sensitive for Mexico for its staple crops, corn and beans. Canada's special safeguard measures as set forth in chapter 7 (agriculture) are for fresh cut flowers, tomatoes, onions, cucumbers, broccoli, cauliflower, and frozen strawberries. U.S. sensitivities in this sector are similar to Canada's. Special tariff classifications were established in the U.S. harmonized schedule for tomatoes, onions, eggplants, chili peppers, squash, and fresh watermelons to differentiate between those periods when these products come to market from U.S. production and when imports from Mexico are more welcome.

The fresh fruit and vegetable growers in the United States, mostly in California, Florida, and Texas, stand to lose from NAFTA. Thus, a tradeoff exists within the U.S. agricultural sector. The gains should go primarily to grain and soybean producers and the losses are more likely for fresh fruit and vegetable growers. How this gets resolved must still play itself out.

While this discussion has concentrated on a few areas of particular sensitivity, trade liberalization is always contentious in the United States, as it is elsewhere. It pits the general public, which should be a force for import liberalization but is a dispersed group, against special interests which are more focused in their desire for protection. This is the scenario as the U.S. Congress opens deliberations on approval or rejection of NAFTA. The U.S. debate, as it unfolds, will deal with many more commodities than have been listed here as being particularly sensitive to import competition.

Impact on the U.S. Economy

There is no single answer to the question of how NAFTA will affect the U.S. economy. If there is a consensus among economists, particularly those who have subjected it to quantitative analysis, it is that the short-term effect will be small but positive, and that the long-term impact will depend much on the dynamic effect in Mexico. [The quantitative analyses of NAFTA were done before the proposed text was released. However, there is little in the text that would affect the projected outcome of these quantitative studies from what was anticipated earlier.] This conclusion is reflected in a recent book reporting on a conference at the Brookings Institution. The direct effects are expected to be small because U.S. import duties are already low and nontariff barriers will be phased out gradually. Because the United States supplies some 70 percent of Mexico's imports, sustained economic growth there should lead to a bilateral U.S. trade surplus, heavily concentrated in capital goods, extending over several decades. [Barry P. Bosworth, Robert Z. Lawrence, and Nora Lustig, "Introduction," in Lustig, Bosworth, and Lawrence, eds., North American Free Trade: Assessing the Impact, (Washington, D.C.: Brookings Institution, 1992), pp. 2-3.] This scenario of greater growth in U.S. exports than imports, is, as the participants at the Brookings conference note, ". . . strikingly at odds with much of the public debate that foresees large-scale relocations of production facilities. [Ibid, p. 2.] " Even if there is a modest increase in U.S. economic growth and gross employment from NAFTA, there will almost surely be intersectoral shifts and thus dislocations in particular industries and communities. The studies of the effects of NAFTA on the U.S. economy have taken a variety of forms. Those that have the greatest credibility, at least to economists, are computable general equilibrium (CGE) models, mostly static, but a few incorporate dynamic aspects. The CGE models are mostly the work of academic economists. [Twelve CGE models are contained in U.S. International Trade Commission, Economy-Wide Modeling of the Economic Implications of a FTA with Mexico and a NAFTA with Canada and Mexico, USITC publication 2508 (Washington, D.C.: USITC, 1992).] The CGE models cannot be used as predictions because the conclusions vary with the nature of the assumptions-whether the projections are based on constant or increasing returns to scale, the nature of product differentiation incorporated in the modeling, and the extent of investment flows that are posited. But they do give an indication of the direction of changes that can be expected from NAFTA.

Drusilla Brown has provided an overview of a number of the leading CGE models, including an evaluation of the role of different modeling assumptions. [Drusilla K. Brown, "The Impact of a North American Free Trade Area: Applied General Equilibrium Models," in Lustig, Bosworth, and Lawrence, eds., North American Free Trade, pp. 26-68.] Her conclusion is that the models clearly indicate that NAFTA will increase welfare in North America, including in the United States, but most particularly in Mexico. The gains are modest, less than 1 percent of GNP, when differentiated products and constant returns to scale are assumed. The welfare gains are higher, about 2 to 4 percent of GNP for Mexico, when products are assumed to be homogeneous across producers, incorporate increasing returns to scale, or both. Adding capital flows increases the welfare gains to Mexico to 4 to 7 percent, and then even in the range of 10 percent when productivity growth is endogenized.

Weintraub examined three types of models: overall CGE models, sectoral models, and those that use other quantitative techniques or are essentially descriptive. [Sidney Weintraub, "Modeling the Industrial Effects of NAFTA," in Lustig, Bosworth, and Lawrence, pp. 109-133.] The static CGE model of Brown, Deardorff, and Stern showed positive wage and growth effects in all three countries and a need for only modest intersectoral factor reallocations. [Drusilla K. Brown, Alan V. Deardorff, and Robert M. Stern, "A North American Free Trade Agreement: Analytical Issues and a Computational Assessment," paper presented at a conference on North American free trade sponsored by the Fraser Institute, the Center for Strategic and International Studies, the University of Toronto Centre for International Studies, and the Stanford University Americas Program, Washington, D.C., June 27-28, 1991.] The dynamic models showed larger welfare gains for all three countries. [The two that were examined were Timothy J. Kehoe, "Modeling the Dynamic Impact of North American Free Trade," working paper 491, Federal Reserve Bank of Minneapolis, March 1992; and Leslie Young and José Romero, "A Dynamic Dual Model of the Free Trade Agreement," University of Texas at Austin and El Colegio de México, 1991.] Perhaps the most widely cited U.S. study of the effects of NAFTA is that of Hufbauer and Schott, presumably because it is presented in language comprehensible to noneconomists. This is not a CGE model, but bases its projections on historical experiences of other countries that have taken the kind of trade liberalizing measures in which Mexico is engaged. Their model shows higher growth in mutual exports between Mexico and the United States and in employment in the two countries than do the static CGE models.

A paper by Faux and Lee comes to a conclusion diametrically different from the CGE and the Hufbauer-Schott models discussed above. [Jeff Faux and Thea Lee, "The Effect of George Bush's NAFTA on American Workers: Ladder Up or Ladder Down," Washington, D.C.: Economic Policy Institute, 1992.] This paper anticipates substantial job losses-the authors cite studies that project losses of between 290,000 and 490,000 U.S. jobs over 10 years-as a result of diversion of investment from the United States to Mexico. In addition, the paper concludes that wages for unskilled U.S. workers are apt to decline. This paper is labelled a "briefing paper." It draws on other studies and newspaper articles for its major conclusions. [The most important of these other studies are Edward E. Leamer, "Wage Effects of a U.S.-Mexican Free Trade Agreement," NBER working paper 3991, Cambridge, Mass.: National Bureau of Economic Research, February 1992; and Timothy Koechlin, Mehrene Larudee, Samuel Bowles, and Gerald Epstein, "Effect of the North American Free Trade Agreement on Investment, Employment and Wages in Mexico and the U.S.," mimeograph, February 1992.] Its conclusions, in my view, are not well substantiated, but its figures on U.S. employment losses are widely cited by opponents of NAFTA.

The Office of Technology Assessment (OTA) in a recent study came to what can best be described as ambiguous conclusions-that open trade could increase living standards in both Mexico and the United States, but only if there are significant other changes in U.S. and Mexican policies. [U.S. Congress, Office of Technology Assessment, U.S.-Mexico Trade? Pulling Together or Pulling Apart? ITE-545 (Washington, D.C.: U.S. Government Printing Office, October 1992).] The main policy prescription concerns the need to improve worker skills in the United States. The study also makes recommendations for a comprehensive worker adjustment program, and suggests further negotiations to establish a number of North American commissions on the environment, and labour and social welfare. The central conclusion of the study is that ". . . whether a NAFTA works for or against either country will depend on how integration is managed. [Ibid, p. 4.]" What is then described in the study, however, is only marginally about how NAFTA is managed, but mostly about how the U.S. economy is managed. The recommendations are mostly consistent with those of the Democratic majority in the Congress.

My own view is similar to that of the majority of mainstream economists. This is that the United States will benefit economically from NAFTA, modestly at first and then more substantially over time as Mexico's economy grows, but that compensation is required for those people who will suffer from the adjustments that will take place in the U.S. economy. These adjustments are required, with or without NAFTA. There can be little long-term benefit for the U.S. economy if noncompetitive sectors are protected. A high-wage future requires activities which embody much research, innovation, and technological advance. These activities require a skilled workforce and not one that seeks to compete with Mexico on the basis of which country pays lower wages. Necessary requirements for a high-wage future are much more investment in education and training, more emphasis on research, and the willingness to adjust to the changing global situation. The countries with which the United States must compete if it wishes to remain an advanced economy are Germany, others in Western Europe, and Japan, and not developing countries like Mexico. All these points are made in the OTA study, although that study talks about management of NAFTA while my judgments deal with the management of the U.S. economy.

The argument in favour of NAFTA is that North America is a natural market for the United States. In 1991, 28 percent of U.S. exports went to Canada and Mexico. The two countries are already the first and third trading partners of the United States. For them, the dominant market is the United States; Canadian and Mexican business people are only now beginning to know each other. When the Mexican economy declined during the 1980s, so too did Mexican imports. In 1986, U.S. exports to Mexico were only US$12 billion. The Mexican economy started to recover that year and by 1991 U.S. exports to Mexico had climbed to US$33 billion. In 1992, they are running closer to US$40 billion; Mexico is beginning to compete with Japan as the second largest U.S. export market. The principal determinant of Mexico's import level is its income growth. And the principal source of Mexico's imports is the United States.

It is clear from scanning the themes dealt with in NAFTA that it is much more than a trade agreement. U.S. trade with the other two countries both follows and stimulates investment in them. Much of the trade that takes place is based on coproduction arrangements. Many of the engines produced in Mexico end up in cars assembled in the United States. The parts and accessories shipped from the United States to Mexico often end up in cars assembled in Mexico or are further manufactured in maquiladora for re-export back to the United States. [U.S. imports of internal combustion engines from Mexico were almost US$700 million in 1991. U.S. motor vehicle parts and accessories shipped to Mexico in 1991 were valued at US$3.2 billion.] More than half of Mexico's manufactured exports to the United States are now intrafirm. The same is true for Canada's manufactured exports to the United States. As trade in industrial inputs as opposed to finished products increases, the need for low trade barriers becomes imperative if North American producers are to remain competitive.

This, as much as absolute levels of trade, explains the support of multinational producers for free trade in North America. The ability to choose manufacturing location on efficiency grounds-on relative factor endowments-without regard to import barriers should enhance competitiveness of North American producers.

The Free Trade Debate in the United States

President Bush submitted the legal text of NAFTA to the Congress on September 18, 1992. The fast-track legislation gives Congress 90 calendar days to examine the agreement before the President can sign it, which he did in December. While congressional hearings began even before the text was formally submitted, they dealt largely with a few issues. These are the adequacy of the environmental provisions in NAFTA and concern over the lack of attention in the agreement to workplace health and safety standards in Mexico. A third labour-related issue is the degree to which the U.S. administration will provide financial support and retraining for workers who lose jobs as a consequence of increased imports from Mexico stemming from NAFTA. This is not an issue appropriate for inclusion in an international agreement, but rather one for the administration to work out with the Congress.

The implementing legislation, which constitutes the true vote on NAFTA, will not take place until the new U.S. Congress meets in 1993. It is thus relevant that Bill Clinton has stated that he supports NAFTA but will not sign implementing legislation "until we have reached additional agreements to protect America's vital interests. But I believe we can address these issues without renegotiating the basic agreement. [Remarks by Governor Bill Clinton, North Carolina State University, Raleigh, October 4, 1992.]" The main issues he and congressional Democrats have stated must be addressed are those listed above: the environment, work standards, and retraining and adjustment assistance. While these three issues are at the center of the public debate, there is another level of debate, less public but more informed, about the details of the NAFTA dealing with specific concessions. The two levels come together in that many congressional opponents of NAFTA are quite prepared to stress Mexico's lack of enforcement of workplace standards or environmental laws when their real concern is competition with an industry in their districts, whether automobiles and parts, textiles and apparel, fresh fruits and vegetables, or whatever. The micro-level debate will become more public when the implementing legislation is taken up and the special protectionist interests are out in full force.

Environment

The nature of the debate taking place can be sketched in here. [Some of the details of the environmental and labour standards debate as they affect NAFTA can be found in the collection of three essays in "The Social Charter Implications of the NAFTA," Canada-U.S. Outlook, vol. 3, no. 3 (August 1992). Canada-U.S. Outlook is a quarterly publication of the National Planning Association, Washington, D.C.] The environmental issue burst forth when President Bush informed the Congress of his intention to negotiate a free trade agreement with Mexico. In order for him to enter into these negotiations, he required extension of fast-track authority under which these discussions are conducted. One of the conditions the Congress imposed was to include environmental issues in the agreement. After some initial resistance on the grounds that this was a trade and not an environmental agreement, President Bush consented and promised that environmental issues would either be made part of the agreement or be dealt with on a parallel track. The U.S. administration did both.

In chapter 1 (objectives), NAFTA accords priority to three international environmental agreements to which the United States is a party: the convention on international trade in endangered species of wild fauna and flora, 1973; the Montreal protocol on substances that deplete the ozone layer, 1987 and amended 1990; and the Basel convention on the control of transboundary movements of hazardous wastes and their disposal, 1989, on its entry into force for the three countries. If this last does not enter into force, NAFTA accords priority to two bilateral agreements: that between Canada and the United States concerning the transboundary movement of hazardous waste, signed at Ottawa in 1986; and the agreement between Mexico and the United States on cooperation for the protection and improvement of the environment in the border area, signed at La Paz, Baja California Sur, 1983. In addition, Mexico and the United States agreed in the spring of 1992 to a cooperative plan for improving the environment at their border. This was part of the parallel track.

The concern expressed in the debate on the NAFTA has been less on Mexico's legal structure for protection of the environment, although it is weak and poorly funded, than on misgivings over Mexico's willingness and ability to enforce its laws and regulations. The opposite side of this argument is that NAFTA is the best assurance the environmentalists can now have about Mexico's commitment to improving its environment. Recent increases in Mexican budgets for environmental infrastructure and enforcement of laws and regulations would almost certainly not have come about if NAFTA had not provided the impetus.

NAFTA has other provisions designed to assure those in the United States concerned about the environment. Each country is empowered to choose the level of protection of human, animal, and plant life or environmental protection that it deems appropriate. Each country can adopt standards and sanitary and phytosanitary measures that it chooses, including those more stringent than international standards. Each country agrees not to lower its health, safety, or environmental standards to attract investment. This is not fully enforceable if there is bad faith on the part of any country; but it would be even less enforceable-the provision might not even exist-in the absence of NAFTA. The factual issues in disputes concerning environmental standards can be submitted for resolution in the NAFTA framework and the burden of proof is on the complaining country.

The Clinton speech supporting NAFTA proposed the establishment of an environmental protection commission with substantial powers and resources to, as he put it, prevent and clean up water pollution, encourage the enforcement of each country's own environmental laws, and provide a forum to hear complaints. This commission would presumably be trilateral. The OTA study had a similar recommendation for the establishment of a bilateral U.S.-Mexico environment protection commission.

The proposal has aroused Mexican concern. The official in charge of the environmental arm of Mexico's Secretariat of Social Development (Sedesol in its Spanish acronym) has stated that Mexico will not accept a supranational authority. [Sergio Reyes Lujan, as quoted in the U.S.-Mexico Free Trade Reporter, October 19, 1992.] Presumably, neither will Canada. And probably neither will the United States.

Workplace conditions

Workplace sanitary and safety issues raise a different set of problems, although U.S. labour unions and some environmental groups made common cause in opposing the grant of authority to the president to negotiate NAFTA. Basic worker rights as defined in U.S. legislation include the right of association, the right to organize and bargain collectively, prohibition of any form of forced or compulsory labour, a minimum age for employment of children, and acceptable conditions of employment with respect to minimum wages, hours of work, and occupational safety and health. These issues are not included in the text of NAFTA. Parallel negotiations are taking place on them.

The debate in the NAFTA centres, once again, not on Mexico's laws and regulations, but on their enforcement. Mexico does not have and cannot afford an army of inspectors to assure compliance with workplace safety and health conditions, as the United States does-or tries to do, not always successfully. The minimum age for employment in Mexico is 14, but young people aged 14 to 16 may not work more than six hours a day or in hazardous occupations. The maximum work week for persons other than minors is 48 hours, although many contracts call for fewer hours. Mexico does have a system of minimum wages, but they are low. Overtime work calls for double the hourly salary.

The most sensitive issue for U.S. workers in competing activities is the fear that low Mexican wages will give Mexican products a competitive edge and force down U.S. wages in these activities. This is hard to deal with head-on in an agreement. Organized labour therefore opposes NAFTA. But if there is an agreement, worker rights is an appropriate subject for discussion. By being on a parallel track and not part of NAFTA itself, violation of worker rights is not subject to the dispute settlement mechanism. This may prompt a drive to incorporate these rights into the agreement, either directly or by reference to the parallel track understanding.

Here again, Clinton-and the OTA-proposed establishment of a commission for worker standards and safety which would have extensive powers to educate, train, develop minimum standards, and settle disputes. A U.S.-Mexico consultative commission on labour matters has, in fact, been established pursuant to a side agreement completed at about the same time that the NAFTA negotiations were concluded. Will a consultative commission as opposed to one with extensive powers satisfy the U.S. Congress? Would Mexico or Canada be prepared to enter into a commission that has supranational power over labour standards? Would the United States? Stand by for the answer in 1993.

Retraining and adjustment assistance

President Bush stated when he obtained congressional approval to negotiate free trade with Mexico that worker adjustment assistance would be provided. Then, in August 1992, President Bush proposed a worker adjustment initiative to be funded at $10 billion over five years. Of the $2 billion each year, the president proposed that $335 million be reserved specifically for NAFTA-related dislocations, and an equal additional amount be available for this purpose if the Secretary of Labour so decided.

This, in the end, may be the most divisive issue of the congressional debate because it requires the appropriation of substantial funds from a tight budget. President Bush never said what would have to give way to make room in the budget for his proposal. The contentious aspects of this issue will be: retraining for what; for how long should financial benefits be provided to affected workers; how to fund the costs. A related theme that will need resolution is the criteria that will be used to determine if a worker is entitled to retraining and financial assistance. Labour union leaders would like this to be an entitlement, not a discretionary program, but this will be a hard sell at a time when other entitlement programs are mushrooming and consuming large parts of the U.S. budget.

This is a debate whose resolution must take place at home, in the United States, and not within the structure of NAFTA. Yet its resolution will be essential for congressional enactment of the implementing legislation. My judgment is that this will be the most difficult issue to resolve for NAFTA to come into existence.

Concluding Comment

Despite its comprehensive nature, U.S. negotiators did not obtain all they sought. This is most evident in the energy field with respect to foreign risk contracts in petroleum exploration, but is true also in other areas. The United States would have preferred symmetry in all three countries on dealing with foreign investment, but both Canada and Mexico reserve the right to screen acquisitions of domestic companies. The big three auto producers would have preferred a rule of origin of 65 percent for passenger cars rather than the 62.5 percent figure in the agreement. All these differences will come out in the debate on implementing legislation.

Beyond that, the United States made concessions that will be opposed by interests that now enjoy protection. Fresh fruit and vegetable growers in California, Florida, and Texas will certainly object to the concessions on these products. U.S. sugar producers will object to the eventual loss of protection should Mexico augment its sugar production. U.S. truckers and teamsters will undoubtedly express concern over lower-paid Mexican drivers having the right, after a transitional period, to make deliveries in the interior of the United States.

The combination of these micro-level objections with the broader issue of free trade with a low-wage, developing country whose past record on enforcement of its own laws on working conditions and the environment makes it impossible to say with certainty whether NAFTA will be approved in the United States. Economists generally favour low international trade barriers. It is thus no accident that most economists who have studied NAFTA have come out in its favour. Producers and workers are not necessarily impressed with the arguments of economists. Their interests are more immediate. Their positions are determined by how they believe NAFTA will affect their well-being. Legislators, who must ultimately say yea or nay on NAFTA, are clearly influenced by all these pressures.

The Canada-U.S. FTA was a relatively simple matter for most U.S. legislators. Canada is a more familiar place to them than Mexico. It is a developed country with roughly comparable levels of wages, working conditions, and social structure. Mexico's relatively low wage level has stimulated much protectionist concern in the United States. Beyond that, the debate in the United States has made it clear that there is much distrust that Mexico will enforce its own laws and regulations on working conditions and environmental issues and, a fortiori, live up to the letter and spirit of its undertakings in NAFTA.

Yet I find it unfortunate that the debate has taken this mistrustful tone. I hope that by the time the U.S. Congress must act, it will do so on the basis of what is agreed in the text and in parallel undertakings-which is substantial-and not on what their stereotypes about Mexico might be.

info@fraserinstitute.ca You can contact us at the above email address for any comments or information requests. Please report any dead links or technical problems.

If you know someone who would be interested in this web page, please enter their email address below, and we will forward this URL to them: Email Address:

Last Modified: Wednesday, October 20, 1999.