NAFTA's Lessons: From Economic Mythology to Current Realities¹

by James M. Cypher

I.

Introduction

The selling of the North American Free Trade Agreement (NAFTA) between the United States and Mexico constituted one of the largest and most sustained bipartisan efforts on the part of the U.S. government, the largest U.S. corporations, and the U.S. economics profession ill living memory. The Mexican government also brought its impressive public relations skills to the political process with a single-minded intensity that surprised many long-time observers of Mexican political affairs.² The myths surrounding the effects of NAFTA were many and varied — all under the rubric of “win-win-win” for the United States, Canada, and Mexico.

The momentum to pass NAFTA was propelled by four primary myths: (1) NAFTA would be a trade agreement not a project to shift production to Mexico. (2) The impact of NAFTA would he felt only after passage: NAFTA was led by a binational search for efficiency. It was not a government/corporate-led strategy to regain lost United States economic dominance which arose in the 1980s. (3) NAFTA would reduce consumer prices in the United States. (4) NAFTA would quickly create 170,000–200,000 jobs in the United States (Hufbauer and Schott, 1993). Jobs would he created in all nations as a result of balanced trade. This article is primarily concerned with this fourth myth regarding employment effects. However, to introduce the argument presented here in a broader context, in this section the first three myths are briefly discussed.

The policy process which culminated in NAFTA was under way by 1986 with the private-sector-funded study by “the Bilateral Commission on the Future of United States-Mexican Relations” favoring further consolidation of the growing economic ties between the United States and Mexico. By November 1987 President Reagan had signed a “Bilateral Framework Agreement on Trade and Investment” with Mexico (Erb and Greenwald, 1989: 135). Mexico then responded with a new foreign investment law in 1989 which greatly increased the scope and latitude of foreign investors (Lustig, 1992: 158–160). (In the 1980s Mexico’s tariffs were not a major concern of U.S. corporate capital, but the Mexican foreign investment laws certainly were of central concern to many prospective investors.) In March 1990 Mexican President Salinas announced his support for a “Free Trade Agreement” with the United States. Now the word “investment” had been purged from the title, even though the elimination of all barriers to the free movement of U.S. corporate capita into (and out of) Mexico was and is at the core of NAFTA. Thus, the first myth of NAFTA is that the agreement exists to facilitate free trade.

In the late 1980s and early 1990s the U.S. auto and autoparts indus try proceeded to make a heavy investment cominitinent in Mexico, con fidently premised on passage of NAFTA (Ramírez de la O, 1998: 59–63 Mortimore, 2000: 1617). This illustrates the second myth of NAFTA — the final passage of the agreement at the end of 1993 constitutes the starting point for an analysis of NAFTA. While it is impossible to set re precise “start” date, the impact on Mexico of the NAFTA project wa, overwhelmingly large years before final passage.

A third myth of NAFTA is that U.S. “consumers” are the beneficiaries of enhanced trade (and investment) with Mexico. Consider that roughly 85 percent of all imports from Mexico are manufactured goods and that roughly 25 percent of these goods are auto-related products. In the U.S. direct labor cost (as a share of total cost) in auto production range from 20 to 30 percent, while in Mexico direct labor cost drops to roughly 5 percent (Dicken, 1998; Juarez Nunez, 1999: 177–178). Seeking this cost advantage, in 1994 General Motors moved 50 percent of its GM “Suburban” production to Mexico, cutting wages from $19 per hour to $1.54. In 1994, before Mexican output came on line, the Suburbans sold for $21,000–$24,500 each. By 1996, with Mexico generating one-half the output, their average price had risen by 20 percent — while the general auto price index (four all vehicles sold in the United States) increased by only 5.4 percent (Anderson, Cavanagh, and Lee, 1999: 51).³ Naive economists often predict great “efficiency,” and thus consumer benefits, from “free” trade. But the auto industry — and many other industries — does not function as a purely competitive industry. As an oligopoly industry of few sellers, it will not pass its cost savings on to “consumers.” This, of course, is exactly why large corporations put so much effort into the creation of NAFTA.

This article will focus on the fourth and primary myth of NAFTA — the employment effects: Proponents asserted that the agreement would facilitate a structural shift in both nations toward greater efficiency, expanding trade, thereby creating employment in both nations. A linked effect would be a tightening in the labor markets of both nations and therefore a mutual increase in wages.

The research presented in the following sections rejects the above view, finding that the employment effects on the United States have been consistently negative, with NAFTA’s structure responsible for employment losses of approximately 315,939 in 1999 due to both trade effects and investment shift effects which have adversely impacted U.S. workers. Further, NAFTA works to adversely impact labor through the threat effect to move operations to Mexico. NAFTA has deepened the competitiveness effect which is realized through off-shore production sites where sub-minimum wages are paid. U.S. corporations have become more competitive while U.S. labor has been adversely affected. Far from producing a mutual increase in wages, Mexican wages have stagnated throughout the industrial sector. Current wages are approximately 25–30 percent below manufacturing wages realized in the early 1980s. Yet, the search for a competitive production base has lowered the wages of U.S. production workers. Finally, NAFTA has deepened the maquiladorization effect whereby Mexico, far from developing its industrial base, has become an assembly site for (primarily) U.S. corporations which thrive on a workforce whose worklife averages ten years. NAFTA has created roughly 753,000 new maquila jobs from 1993–2000, at an average wage-per-hour level of roughly $1.

In conclusion, it is argued that NAFTA’s lessons demand a major reorientation of strategic effort on the part of not only organized labor, but also socially concerned policymakers, labor educators, human-rights advocates, and committed students in the United States, Mexico, and Canada in pursuit of a renegotiated NAFTA. A new agreement must incorporate heavy sanctions for labor rights violations. It must ensure, through sanctions and other measures, Core Labor Standards and the freedom of cross-border cooperation and negotiations for all organized labor. With these conditions and standards in place both U.S. and Mexican labor will be in a position to halt the “race to the bottom” which NAFTA has both enhanced and perpetuated.

Investment and Trade Effects of NAFTA

In the 1991–1993 period there were several “models” of NAFTA that purported to show how economic science could be used to predict the effects of the agreement (USITC, 1992a; USITC, 1992b). All assumed future “balanced” trade (U.S. exports to Mexico would equal U.S. imports), moderate growth of trade, and (with one exception) that the agreement would not induce a significant shift of U.S. capital outlays in plant and equipment (i.e., direct foreign investment) to Mexico. Table I records the actual trade statistics, along with an estimate of the jobs created or foregone in any given year.

Table 1 demonstrates that the large U.S. trade deficit with Mexico is nest merely the result of the peso collapse of 1994–95. In fact, the deficit for 1999, $22.662 billion, was substantially above the $16.5 billion deficit 111 1995 (column 1 , U.S. Exports to Mexico, minus column 2, U.S. Imports from Mexico). The 1995 deficit occurred when the peso was weak; in 1999 the peso was strong. The peso’s Value is but one determinant of Mexico’s exhort capability. More important is the strategy followed by transnational corporations which have made a commitment to Mexico rind use their plants for intra-firm trace in the context of long-term strategic planning for competitiveness.⁴

Table 1
U.S.-Mexico Trade and the Employment Impact of NAFTA

	1	2	3	4
	US Exports ($ Billions)	U.S. Imports ($ Billions)	Jobs per $1 Billion	U.S. Jobs Foregone
1993	$41,635	$40,745	15,123	+13,475
1994	50,840	50,356	14,361	+6,951
1995	46,311	62,756	13,774	-226,513
1996	56,791	74,111	13,258	-226,513
1997	71,378	87,167	12,755	-201,389
1998	78,772	94,269	12,181*	-193,154
1999	87,044	109,706	11,633*	-264,092

Sources: USITC, lit http://dataweb.usitc.gov; U.S. Bureau Of the Census, Foreign Trade Division; Banamex, Review of the Economic Situation of Mexico (October 1999); Bolle, 1998.

*Estimated

Column 3 shows a U.S. government estimate for U.S. jobs (fulltime for one year) created per billion dollars of exports (Bone. 1998: 15). This estimate includes both direct and induced (or secondary) employment created across the U.S. economy when exports increase by $1 billion. The numbers drop rather rapidly from 1993 to 1999 because they are adjusted each year for the rate of productivity growth (which reduces the number of workers needed) and for inflation (which reduces the number of workers which $1 billion of spending will employ). These are average, economywide estimates, with only a modest portion of the $1 billion actually received by workers, the rest taken in management salaries, profits, dividends, interest, rent, and depreciation.

Column 4 registers the U.S. jobs foregone due to both the restructuring of the Mexican economy and the offshore sourcing strategies pursued by U.S. transnational corporations from the 1980s onward. Thus for 1999 it is estimated that 264,092 jobs were not created in the United States because goods consumed in the United States (Mexico’s import) were not balanced by an equivalent level of goods produced in the United States and exported to Mexico.⁵ Notice that the estimate is derived by multiplying the jobs-per-billion number (1 1,633) by the deficit ($22.662 billion). To produce this estimate, the jobs-per-billion number was used for both exports and imports.⁶

Jobs Foregone Due to Investment Shifts

Free trade economists tend to become eloquent when discussing the imagined benefits of trade, but silent regarding the impact of capital mobility. In the case of Mexico, direct foreign investment from the United States has been at the annual level of $6.5 billion for the past six years. Using the jobs-per-billion-dollars calculation of Table 1, this means that roughly 75,615 jobs in the production of investment goods were foregone in 1999 — with the largest impact falling can the U.S. construction industry, including the building trades unions.

This estimate likely underemphasizes the job-loss effect from investment diversion to Mexico: Further negative multiplier effects could well be felt from infrastructure investment by government which was foregone because capital shifted out of the United States. (There seems toy be no current reliable research on this effect — but it is far from negligible.)

Combining the estimated job loss through investment diversion with that of the previous section, it is possible that in 1999 the total job loss-jobs foregone effect (trade plus investment effects) of NAFTA was nearly as high as 315,939.⁷ Measured against the entire U.S. labor force this may not seen 1 like a startlingly high number. But measured against the 18 million manufacturing workers in the United States, the number
takes on greater significance. Measured against union workers in manufacturing  and construction, these numbers begin to convey the anxiety organized labor has continually expressed regarding NAFTA, fast track, and the Enterprise for the Americas Initiative.⁸

NAFTA and the Threat Effect

Even if the trade balance indicated that no direct evidence of jobs foregone could be produced-the situation in 1993 and 1994-the largest single effect of NAFTA on labor would remain. NAFTA’s enabling structure has made the threat of moving offshore very real and very tangible.

Kate Bronfenbrenner’s study “The Effects of Plant Closing” — a comprehensive analysis of private sector organizing efforts in the 1993-95 period — produced the following conclusion: The “most dramatic finding” was that in one of every eight organizing victories employers shut down all or soiree of the plant — to move offshore in most instances. Further, the plant closing rate has more than doubled from the early 1990s, suggesting “that NAFTA has both increased the credibility and effectiveness of the plant closing threat for employers and emboldened increasing numbers of employers toy act upon that threat” (Bronfrenbrenner, 1996: 26–27). Bronfrenbrenner found that during organizing campaigns the majority of employers threatened to shut the plant and that “many workers appear to take even the most veiled employer plant closing threats very seriously” (Bronfrenbrenner, 1996: 31).

Perhaps the best recent example of the threat effect involves the General Electric Coinpany. In September 1999, GE eliminated 1,400 pro(lucticil jobs in Bloomington, Indiana — cutting 44 percent of its refrigerator print workforce, while shifting this production line to Mexico (New York Times, 1999: C4). In Mexico, production will be provided via a joint venture (JV).⁹

GE also cut 7,500 workers from its Louisville Appliance Park plant — moving most of the work to Mexico (Bernstein, 1999: 78). To prevent further shifts to joint ventures in Mexico the International Union of Electrical, Radio, and Machine Workers (IUE) was forced to make major pay concessions, including acceptance of new techniques to raise productivity. GE has annual sales of $10.3 billion, current profits of $1.7 billion, with 163,000 U.S. employees and production workers represented by 14 unions — the largest of which is the IUE with 47,000 workers (Bernstein, 1999). GE also has 30,000 employees in Mexico plus an unknown number involved in GE-controlled subcontracting, strategic alliance agreements, and JVs. GE’s current campaign is to take several major lines of business into Mexico by forcing subcontractors to set up new operations in Mexico for suppliers of aircraft engines, power systems, industrial systems, specialty metals, and aircraft instruments. It appears that GE’s campaign is affecting hundreds of suppliers, some of whom will shift production to a new aerospace industrial park in Monterrey, Mexico. GE has cut its U.S. labor force by 87,000 since 1988, adding 80,000 foreign workers pit the same time.¹⁰ The threat effect impacts labor even when it is not directly subjected to displacement either by imports or offshore shifts in investment and outsourcing. When wages move down in industries directly impacted by NAFTA, downward pressure spreads on all closely innerlinked labor markets, regionally and nationally.¹¹ NAFTA, in making the threat credible, contributes to U.S. wage dispersion either through forcing production workers wages to (1) fall, (2) stagnate, or (3) rise well below increases in productivity.

The U.S. Competitiveness Effect

When (in the late 1980s) the United States became seriously committed to a new investment and trade regime with Mexico, it did so because of growing concern on the part of state managers and policymakers and U.S. business groups that the United States was losing out in the “competitiveness” struggle with Japan — and to a much lesser degree with Europe. One of the primary policy drivers in this “competitiveness crisis” was the U.S. auto industry, which was falling behind in the more competitive global auto industry of the 1980s.

A decade later, with Mexico as the United States’s second-largest trading nation, the auto industry has been rejuvenated. The changes in the auto industry point to a broader conclusion: United States-Mexico trade is being driven not by “arms length” market transactions nor by Mexican-owned firms increasingly exporting products to the United States, as trade economists anticipated. Rather, we are now witnessing, as Raúl Urtega of Mexico’s Secretariat of Commerce stated, the full realization not of a “trade deal” but of a co-production scheme of exceedingly large proportion:

The objective of NAFTA was to solidify what the maquiladora industry started: co-production. NAFTA is consolidating the vertical integration of entire sectors. The big three [auto producers] in the United States have assembly plants and facilities in the three countries. That shows that the auto) industry is perhaps the first fully integrated industry in North America (Adelson, 1999a: 10)

For the auto industry the decision was to make Mexico the center for production of small car and engine plants. By 1994 more than two-thirds of auto sector exports entailed vehicles, whereas ten years before the auto sector had emphasized engines and parts. International competitiveness arid advanced technologies had become underlying features of the U.S. Alto industry’s Mexico operations by the early 1990s. Michael Mortimore portrays Mexico as an example of “immiserizing international competitiveness” whereby production rises, exports soar, foreign corporations invest, and yet the majority of the nation languishes in deep poverty and rising inequality (Mortimore, 1998: 426). He spells out the cruel dilemma of NAFTA as part of the enabling neoliberal “transnational corporation-centric” framework which U.S. transnational corporations sought as their means of restructuring in the 1980s and early 1990s: ...the Mexican transnational corporation-centric model has enabled many Transnational Corporations, especially U.S. ones, to defend their market shares in their home markets. However, national Mexican companies for the most part are not major participants in the most dynamic industries of international trace, rather they operate in generally undynainic sectors such as cement, glass, etc....The Mexican transnational corporation-centric version could be interpreted even to have produced an initial contradiction between national value-added and international competitiveness. (Mortimore, 1998: 426) That is, Mexico succeeds as a competitive site for transnational corporations so long as it fails to receive significant value-added income in the co-production scheme. Sub-minimum wages are Mexico’s contribution to the “model” of Mexico’s integration into the new regional bloc created by the United States.

From 1989 to 1997, to cite recent data, Mexico’s exports grew at the rate of 14 percent per year. At the same time, the role of the transnational corporations increased, Such that from 1993 to 1998 the share of Mexico’s exports attributable to the transnational corporations rose from 56.5 percent to 64.2 percent (CEPAL, 2000: 105 ). Leading this phenomenal growth has been the auto sector: in 1985 passenger cars accounted for 1 percent of Mexico’s exports, by 1996 the figure stood at 11 percent, and by 1999 it rose to 16 percent. In 1985 exports for the entire auto sector (parts, motors, etc.) accounted for 10.5 percent of exports (oil for 31.4 percent) while in 1996 the entire auto sector accounted for 21.6 percent (oil for only 8.5 percent) (CEPAL, 2000: 106). (The export boom in manufactured products had also made Mexico the number one foreign supplier to the United States of electronics industry products and apparel in 1998. )

Meanwhile, Japan’s auto exports to the United States have stagnated since 1994. By 1998 Mexico’s output of cars arid trucks delivered to the U.S. market was 51 percent of that of Japan — up from 21 percent of
Japan’s in 1994 (CEPAL, 2000: 107). And the U.S. auto industry is increasing its emphasis on Mexican production.¹² The boon in autos strongly coincides .with NAFTA-induced policy changes which drastically reduced Mexico’s national content requirements (specifying a minimum value of national produced parts, materials, and other inputs). By 2004, required national content value added will drop from 32 percent to zero. At this point (2004) analysts now predict that Mexico Could have more auto sector workers than the United States.

For Mexico, the boom in auto production has been accompanied by a significant increase in employment even as new production techniques have raised productivity. In 1990 the industry (auto and parts) employed 295,000, while by mid-2000 employment had soared to 611,000. Productivity rose by 10.3 percent from 1994 to 1999, but wages in the auto sector fell by 20 percent (Smith, et al., 2000: 78; Lippert, 2000: 8). Employment in the Mexican auto industry points to a much larger dilemma: Mexico’s “boom” is dependent upon direct foreign investment (DFI), principally U.S. DFI. Yet, as recent research indicates, any wage increase in Mexico relative to the United States will reduce DFI (Love and Lage-Hildalgo, 1999: 91).¹³ Thus the vicious circle: When Mexico receives DFI or subcontracting or JVs or strategic alliances it increases the threat effect on U.S. wage rates. And U.S. wage growth (other things remaining the same) stagnates or declines while in Mexico wages must remain low to maintain the export-led growth of the Mexican economy. As capital shift from North to South the relative demand for less skilled labor in the United States goes down, thereby increasing the inequality of wages. At the same time, in Mexico the demand for skilled labor goes up because tile processes transferred to Mexico are — by Mexico’s standards — relatively skilled. Thus DFI, outsourcing, JVs, and strategic alliances also create greater wage dispersion in Mexico — while falling to address Mexico’s deed to employ more than 1.2 million new workers per year (Hanson and Harrison, 1999: 284; Feenstra. 1998: 42).

The Maquiladorization Effect

Over the past ten years maquiladora plants, primarily located along the U. S.-Mexico border, have grown at a blistering pace. By mid-2000 the maquilas lead roughly 1.3 million employees working in 3,500 plants. (In 1990 the maquilas employed only 448,000.) Only 15 percent of these plants account for 85 percent of the output. In 1998, wages hovered at $120 to $180 per month (plus benefits) (Philpott, 1999: 3). From 1994 to mid-1999 Mexico created only 1.54 million permanent jobs — 35 percent of these coming from the maquila sector.

Labor turnover rates pit the plants average between 15 percent and 25 percent of the labor force per month, or 3.5 million job switches per year (González, 1999: 15). The average worklife for a maquila worker is only ten years, due to injuries, health problems, and the firing of women workers who become pregnant. (Maquila workers are essentially ‘throwaways’ who leave no union protection and who must descend into the ‘informal’ sector when their brief cycle of formal employment ends. In the informal sector they work as street vendors or as domestic servants at a fraction of the incoine received in the maquila sector.)

The maquilas do virtually nothing to develop Mexico, aside from a short cycle of employment for more than a million workers. Aside from labor, domestic value added from materials ranges from 2 percent to 4 percent. U.S. and joint U. S.-Mexican-owned firms account for 51 percent of all the plants — but in terms of the value of output the influence of the United States is much higher (McCosh, 1997: 12).

The maquilas are, nevertheless, crucial: by 2000 they accounted for 47 percent of all exports and 54 percent of all manufactured exports. The maquilas trade surplus was approximately one-third greater than gross oil exports in 1998 (Banamex, November 1998: 414). Reviewing this situation, former Finance Minister David Ibarra stated, “We are fundamentally a maquiladora exporter, or rather, a cheap labor exporter” (Adelson, 1999b: 3).

In spite of the phenomenal growth in employment and output, maquila production workers (low-skilled workers who make up 82 percent of the workforce) experienced a real wage decline of 6 percent from 1993 to 1998 (Philpott, 1999: 3). Meanwhile, demonstrating the heightened role of the maquilas, the number of non-maquila manufacturing workers actually declined somewhat from 1993 to 1998. Further validating the shift to maquiladorization is the rapid increase in maquila DFI as a share of total Mexican Foreign Investment: in 1994 maquilas were absorbing only 6 percent of all DFI; in 1996 maquilas drew 14 percent of foreign investment; and in 1999, 26 percent (Banamex, December 1999: 473).

Direct Conclusions

Since passage of the NAFTA agreement the United States has suffered a growing trade deficit with Mexico. While trade economists confidently anticipated balanced trade and strong employment effects for the United States, in 1999 trade with Mexico reduced employment opportunities in the United States by an estimated 264,092 jobs-Further, NAFTA has encouraged a significant shifting of capital investment to Mexico. Combining the trade and investment effects of NAFTA, roughly 315,939 job opportunities were foregone in 1999 due to the NAFTA accord with Mexico. While impossible to quantify, NAFTA has constituted a threat effect as corporations threaten to, and do, shift production sites to Mexico, driving down wages and working conditions in the United States. Far from the prediction of trade economists that enhanced trade with Mexico would increase wages in both nations, U.S. wages have stagnated or fallen or lagged behind productivity increases in industries such as autos, which face direct competition with Mexico. NAFTA has increased some employment opportunities in manufacturing/assembly in Mexico, but not real wages. (And NAFTA has created millions of new unemployed in the agricultural sector due to the surge of U.S. agricultural imports.) NAFTA has also greatly increased the role of maquiladora assembly plants with one consequence being that a greater percentage of Mexico’s manufacturing workers are without union protection and must subsist with a worklife expectancy of only ten years. In major areas of trade between the United States and Mexico, such as autos, there is no evidence that U.S. consumers have been the primary or major beneficiary of low-wage production.

NAFTA was designed to reduce labor’s share of production and income in the United States, and it has clearly done so. U.S. transnationals understood that new direct investment in Mexico would not drive Lip the Mexican wage rate to any appreciable degree both because of the pervasiveness of “protection” labor contracts at U.S. maquiladora plants and because of the huge and growing “surplus” population of young workers. Shifting capital to Mexico would either drive down the U.S. wage rate car cause it to stagnate. At the same time, shifting capital to Mexico would ensure the viability of the threat creating the climate of fear which has ensured a weak response by labor in bargaining situations.

Some Broader Conclusions

Labor is now forced to retrieve the losses of NAFTA. Cross-border organization is one response — but this tactic involves projecting U.S labor into a dramatically different social structure, a profoundly different Culture, and a strikingly disparate mode of production in Mexico. For U.S. labor, cress-border organization will demand the mastery of Mexico’s complex institutions and cultural practices. It will demand the ability to overcome centuries of deep distrust regarding the motives of all U.S, citizens in Mexico. This cannot be either quick or easy. Organized labor has always struggled, and often won, in situations which offer no quick or easy solutions. Confronting the challenge of NAFTA demands clarity on the effects of this agreement which has disenfranchised workers on both sides of the borders. NAFTA’s lessons bring into concrete focus the vague and false promises of “globalization”: the actual impacts of NAFTA, as detailed here, should allow labor educators, union members, and the concerned Public a definite context in which to advocate for major policy changes, specifically the renegotiation of NAFTA.

For U.S. labor, deep linkages with Mexico must now be achieved not merely in Mexican unions, but throughout various levels of society. The current realities of NAFTA should force a reconceptualization of industrial relations: the expansion of international trade union cooperation, cross-border organization and contract negotiations, the struggle for international labor standards; and the ratification and enforcement of basic International Labor Organization (ILO) conventions must now became prioritized strategic goals of U.S. labor.

The “Lessons of NAFTA,” as outlined here, should present a major argument for passage and meticulous enforcement of Core Labor Standards (and for the inclusion of the rights of cross-border contracts in a renegotiated NAFTA). Some Mexican trade union leaders suggest that tile majority of contracts are actually “protection contracts” allowing employers to “guarantee” compliance with the labor law yet operate without :any oversight by an independent union (Fonte Z., et al., 1999: 548). “Freedom of Association” and “Effective Recognition of the Right to Collective Bargaining” are two of the five components of Core Labor Standards — with these two in place, Mexico’s labor force would be in position to stop the race to the bottom and effectively engage in cross-border labor accords and organizing drives.

Notes

1 I am grateful for the comments received from participants of, the UCLEA conference on “Unions in the Global Economy” in April 2000 and to the three anonymous reviewers for this journal. return to text

2 This was written before publication of John MacArthur’s The Selling of “Free Trade”: NAFTA, Washington and the Subversion of American Democracy (New York: Hill and Wang, 2000) which documents the determination of the Clinton Administration and the USA*NAFTA lobbying effort to assure passage of NAFTA. With its focus on labor issues, MacArthur’s book is a welcome addition to the small number of publications which have examined the effects of NAFTA. return to text

3 In 1999 the Suburban was GM’s most profitable vehicle with a net income of $5,000 per unit. Labor productivity in the Mexican plant was 9 percent higher than the U.S. plant. The Mexican plant registered the best record ()f all GM truck plants for quality control (Lippert, 2000: 3–4). return to text

4 Contrary to mainstream trade theory, which hinges export performance on the dollar/peso exchange rate, the transnational corporations — most particularly the auto corporations and their linked autoparts companies, which account for nearly 25 percent of Mexico’s manufactured exports-seem all but impervious to the swings of the peso. Intra-industry or intra-firm trade accounted for 48% of Mexico’s exports in 1998-up from 44% in 1995 (Banamex, April 1999: 120). return to text

5 Some economists strenuously object to the use of these numbers to calculate the jobs-per-billion of imports (Hinojosa, et al., 2000). The issues raised in this dispute are complex and lengthy. This dispute was detailed in the original conference version of this paper, but it has been eliminated here due to space limitations. return to text

6 Several economists have gone to considerable effort in attempting to address the issue of net jobs created/lost through NAFTA (Bolle, 1998; Hinojosa, et al., 2000; Scott, no date). Their approach, if applied to Table 1 would be to add the job effect numbers in column 5 to estimate the total employment effect of NAFTA. The reason for this method is unclear: Each year a given number of workers is employed to generate the total value of output. Adding the jobs from year to year (jobs created or foregone) does not seem to address the issue. return to text

7 Several subtle adjustments would be required to make this a firm estimate. Accounting for goods shipped through the United States would reduce the trade numbers by about 9 percent. Thus, here we have taken the figure from the previous section — 264,092 — reducing it by 9 percent following Robert Scott’s research estimate of the effect of transhipped goods (Scott, no date). Other downward adjustments, largely beyond the estimating abilities of economists, would have to be made relating to such issues as the fact that when imported goods are marketed in the United States some jobs created in wholesale, retail, finance, and transport. return to text

8 Obviously, not all of the jobs foregone would be organized. Yet consider the fact that nearly 25 percent of the manufacturing exports of Mexico are in the auto sector. For the United States, unionization rates are 55% in autos and auto parts. return to text

9 Notice that investment here is displaced onto the joint venture partner, which might not be a U.S. firm. The data presented in the previous section on investment diversion would not record GE’s actions. Likewise, when U.S. firms engage in outsourcing or strategic alliances the effect cannot be traced through U.S. outbound DFI. Use of a JV is increasingly commonplace, making it difficult to actually trace the impact of NAFTA. Outsourcing, JVs, and strategic alliances are not registered through any international statistics, but observers know they are now very large — probably as much as 50 or 100 percent as large as DFI (Dicken, 1998; Dunning, 1999). return to text

10 Another recent manifestation of the threat effect involves the America Axle & Manufacturing Holdings Corporation, which opened a new, state of-the-art gear cutting plant in Guanajuato, Mexico, in March 2000. American Axle hired 328 workers at $2.60 per hour — these workers have established a perfect record for quality control thus far. Using the leverage of the Mexico plant, American Axle then announced that it would build a identical production site in Three Rivers, Michigan — but only after the UAW agreed to drop the normal machinists’ wage rate by 23 percent, from $22 per hour to $17 (Lippert, 2000: 7). return to text

11 Even though the threat effect makes perfect economic sense, mainstreai trade economists have long challenged the idea that increasing U.S. trade rind foreign investment has had much to do with the growing inequality c income distribution. This approach has served to diminish attention o external trade and investment, even among critical economists such as jams Galbraith, who recently stated: “It is...hard to see how North-South trade, which remains comparatively minor, can have such pervasive effects on the structure” (Galbraith, 1998: 276). It is therefore important to note that the research of some trade economists is now moving in the direction of positions long held by critics of NAFTA. Of particular note is the work of Robert Feenstra, who has demonstrated that by incorporating the impact of outsourcing on labor, rather than limiting the analytical focus strictly to trade in final goods, it is possible to conclude that the role of transnational production is much more important than previously thought in terms of explaining the growing income gap between skilled and unskilled workers (Feenstra, 1998). Thus while some researchers claimed that only one-half of 1 percent of the decline in demand for less skilled workers was due to increasing U.S. trade, Feenstra has shown that, when outsourcing is included in the analysis, about 20 percent of the wage dispersion between skilled and less skilled workers can be explained through “globalization” (Feenstra and Hanson, 1997). return to text

12 Thus in 1998–99 the auto industry invested $5.5 billion in both new plant and equipment’ and in the expansion of existing production facilities (CEPAL, 2000: 114). In addition, in April 1999 Ford announced a massive investment plan for 2000–2001, including $2.5 billion of new capital outlays for its export operation which will involve 120 Mexican suppliers, and another $2.7 billion to expand the export capacity for vehicles (Senzek, 1999: 11). In May 2000 DailmerChrysler stated it would expand again in Mexico with $2 billion spread over the next five years. return to text

13 At the top of the scale, average auto workers’ pay was $15.80 a day in August 1997 at an organized plant employing 10,450 workers. These “high” wages have driven at least one major producer to build an In dustrial park adjacent to the plant where outsourced suppliers paid $ 8.78 per day (Juarez, Nuñez, 1999: 199, 205 ). return to text

References

Adelson, Naomi. 1999a. “Co-production: NAFTA Continues What the Maquiladoras Started,” El Financiero International (November 22): 10.

_____ 1999b. “About Face: The Move to Export-Led Growth,” El Financiero International (June 21): 3.

Anderson, Sara, John Cavanagh, and Thea Lee. 1999. Field Guide to the Global Economy. New York: New Press.

Banamex. Review of the Economic Situation of Mexico (various issues).

Bernstein, Aron. 1999. “Welch’s March to the South,” Business Week (December 6): 74, 78.

Bolle, Mary Jane. 1998. “NAFTA: Estimated U.S. Job Gains and Losses by State,” CRS Report to Congress (September 16 ): 1-18.

Bronfenbrenner, Kate. 1996. “The Effects of Plant Closing on the Right of Workers to Organize,” New York School of Industrial and Labor Relations, Cornell University (September): 1–56.

CEPAL. 2000. La Inversion Extranjera en America Latina y el Caribe, 1999. Santiago: CEPAL.

Dicken, Peter. 1998. Global Shift. New York: Guilford Press.

Dunning, John. 1999. Alliance Capitalism. London: Routledge.

Erb, Guy, and Joseph Greenwald. 1989. “U.S.-Mexico Trade Issues,” in The Economics of Interdependence: Mexico and the United States, William Glade and Cassio Luiselli (eds.), 107–148. San Diego: University of California, San Diego.

Feenstra, Robert. 1998. “Integration of Trade and Disintegration of Production in the Global Economy,” Journal of Economic Perspectives 12, no. 4: 31–50.

Feenstra, Robert, and Gordon Hanson. 1997. “Productivity Measurement and the Impact of Trade and Technology on Wages,” NBER Working Paper (Cambridge, Mass.), No. 6052 (June).

Fonte Z., Luis, Evaristo Alvarez A. y Guillermo Rivera R. 1999. “La Autentica Contratacion Colectiva y Los Sindicatos de Proteccion,” in Confronting Change, Humberto Juarez Nunez and Steve Babson (eds.), 545-550. Puebla, Mexico: BUAP

Galbraith, James. 1998. Created Unequal. New York: Free Press.

González, Maria. 1999. “Maquiladoras in Crisis,” El Financiero International (July 19): 15.

Hanson, Gordon, and Ann Harrison. 1999. “Trade Liberalization and Wage Inequality in Mexico,” Industrial and Labor Relations Review 52, No. 2: 271–288.

Hinojosa, Raul, et al. 2000. “The U.S. Employment Impacts of North American Integration After NAFTA.” School of Public Policy and Social Research, UCLA. http://www.naid.sppsr.ucla.edu (cited January 15).

Hufbauer, Gary, and Jeffery Schott. 1993. NAFTA, An Assessment. Washington, D.C.: Institute for International Economics.

Juarez Nuñez, Humberto. 1999. “La Productividad y et Trabajo en et Contexto de la Produccion Esbelta,” in Confronting Change, Huberto Juarez Nunez, and Steve Babson (eds.), 173–206. Puebla, Mexico: BUAP

Lippert, John. 2000. “”Mexico Becomes Motown South,” Bloomberg Markets. (July) pp. 1–9. http://www.bloomberg.com/marketsmagazine/

Love, James, and Francisco Lage-Hidalgo. 1999. “The Ownership Advantage in Latin American FDI,” Journal of Development Studies 35, No. 5 (June): 76–95.

Lustig, Nora. 1992. “Mexico’s Integration Strategy with North America,” in Strategic Options for Latin America in the 1990s, Colin Bradford (ed.), 155–181. Paris: OECD.

McCosh, Dan. 1997. “Maquila Growth is Back on Track,” El Financiero International (September 20): 12.

Mortimore, Michael. 1998. “Mexico’s Transnational Corporation-centric Industrialization Process,” in Transnational Corporations and the Global Economy, Richard Kozul-Wright and Robert Rowthorn (eds.), 401–430. New York: St. Martin’s.

_____ 2000. “Corporate Strategies for FDI in the Context of Latin America’s New Economic Model,” World Development 28, No. 9: 1611–1626.

New York Times. 1999. “General Electric Is Cutting 1,400 Jobs.” (September 10): C4.

Philpott, Tom. 1999. “Boom and Gloom: Manufacturing Rises as Wages Fall.” El Financiero International (June 14) : 3 .

Ramírez de la O, Rogelio. 1998. “The Impact of NAFTA on the Auto Industry in Mexico,” in The North American Auto Industry Under NAFTA, Sidney Weintraub and Christopher Sands (eds.), 48–91. Washington, D.C.: Center for Strategic and International Studies.

Scott, Robert. No date. “North America Free Trade After NAFTA,” Economic Policy Institute Briefing Paper, 1–19.

Senzek, Alva. 1999. “Economic Growth Holds in 1st Quarter,” El Financiero International (April 26): 11.

Smith, Geri, et al., 2000. “Car Power: Latin American Auto Industry,” Business Week (October 23 ): 73–82.

USITC (United States International Trade Commission). 1992a. Economy-wide Modeling of the Economic Implications of a FTA with Mexico, Publication 2508. Washington, D.C.

_____ 1992b. Economy-wide Modeling of the Economic Impact of a FTA with Mexico, Publication 2516. Washington, D.C.

June 23, 2003