back to Corp Glbzation Impacts | co-globalize | rat haus | Index | Search | tree

( PDF | ASCII text formats )

This document is mirrored from its source at:

April 2001 | EPI Briefing Paper

For a printer-friendly version of this report, download an Acrobat PDF version of this paper.
To view the press release for this report, click here.

Its impact on workers in all three nations

Jump to a specific report:

NAFTA’s Hidden Costs
Trade agreement results in job losses, growing inequality, and wage suppression for the United States

by Robert E. Scott, Economic Policy Institute

The North American Free Trade Agreement (NAFTA) eliminated 766,030 actual and potential U.S. jobs between 1994 and 2000 because of the rapid growth in the net U.S. export deficit with Mexico and Canada. The loss of these real and potential jobs[1] is just the most visible tip of NAFTA’s impact on the U.S. economy. In fact, NAFTA has also contributed to rising income inequality, suppressed real wages for production workers, weakened collective bargaining powers and ability to organize unions, and reduced fringe benefits.

NAFTA’s impact in the U.S., however, often has been obscured by the boom and bust cycle that has driven domestic consumption, investment, and speculation in the mid- and late 1990s. Between 1994 (when NAFTA was implemented) and 2000, total employment rose rapidly in the U.S., causing overall unemployment to fall to record low levels. Unemployment, however, began to rise early in 2001, and, if job growth dries up in the near future, the underlying problems caused by U.S. trade patterns will become much more apparent, especially in the manufacturing sector. The U.S. manufacturing sector has already lost 759,000 jobs since April 1998 (Bernstein 2001). If, as expected, U.S. trade deficits continue to rise with Mexico and Canada while job creation slows, then the job losses suffered by U.S. workers will be much larger and more apparent than if U.S. NAFTA trade were balanced or in surplus.

Growing trade deficits and job losses
NAFTA supporters have frequently touted the benefits of exports while remaining silent on the impacts of rapid import growth (Scott 2000). But any evaluation of the impact of trade on the domestic economy must include both imports and exports. If the United States exports 1,000 cars to Mexico, many American workers are employed in their production. If, however, the U.S. imports 1,000 foreign-made cars rather than building them domestically, then a similar number of Americans who would have otherwise been employed in the auto industry will have to find other work. Ignoring imports and counting only exports is like trying to balance a checkbook by counting only deposits but not withdrawals.

The U.S. has experienced steadily growing global trade deficits for nearly three decades, and these deficits have accelerated rapidly since NAFTA took effect on January 1, 1994. Although gross U.S. exports to its NAFTA partners have increased dramatically—with real growth of 147% to Mexico and 66% to Canada—these increases have been overshadowed by the larger growth in imports, which have gone up by 248% from Mexico and 79% from Canada, as shown in Table 1-1. As a result, the $16.6 billion U.S. net export deficit with these countries in 1993 increased by 378% to $62.8 billion by 2000 (all figures in inflation-adjusted 1992 dollars). As a result, NAFTA has led to job losses in all 50 states and the District of Columbia, as shown in Figure 1-A.

Table 1-1

Figure 1-A

The growing U.S. trade deficit has been facilitated by substantial currency devaluations in Mexico and Canada, which have made both countries’ exports to the United States cheaper while making imports from the United States more expensive in those markets. These devalued currencies have also encouraged investors in Canada and Mexico to build new and expanded production capacity to export even more goods to the U.S. market.

The Mexican peso was highly overvalued in 1994 when NAFTA took effect (Blecker 1997). The peso lost about 31% of its real, inflation-adjusted value between 1994 and 1995, after the Mexican financial crisis. The peso has gained real value (appreciated) recently because inflation in Mexico has remained well above levels in the U.S. As prices in Mexico rose, its exports become less competitive with goods produced in the U.S. and other countries because the peso’s market exchange rate was unchanged between 1998 and 2000. High inflation in Mexico also made imports cheaper, relative to goods purchased in the U.S.

By 2000 the peso’s real value had risen to roughly the pre-crisis levels of 1994.[2] Thus, the peso was as overvalued in 2000 as it was when NAFTA took effect. As a result, Mexico’s trade and current account balances worsened substantially in 1998-2000, as imports from other countries surged, despite the fact that Mexico’s trade surplus with the U.S. continued to improve through 2000. Given Mexico’s large overall trade deficits, and the rising value of the peso, pressures are building for another peso crisis in the near future.

The Canadian dollar has depreciated over the past few years. The Canada-U.S. Free Trade Agreement—a precursor to NAFTA—took effect in 1989. Initially, the Canadian dollar rose 4.1% in real terms between 1989 and 1991, as Canada’s Central Bank tightened interest rates. During this period, Canada maintained short-term interest rates that averaged 2.25 percentage points above those in the U.S. (1989 to 1994), which caused the initial appreciation in its currency. Canada then began to reduce real interest rates in the mid-1990s. Between 1995 and 2000, short-term interest rates in Canada were 0.75 percentage points below U.S. rates, a net swing of 3.0 percentage points. The Canadian dollar began to depreciate in the mid-1990s, as interest rates were reduced, relative to the U.S. Overall, between 1989 and 2000, the Canadian dollar lost 27% of its real value against the U.S. dollar.[3]

NAFTA and the devaluation of currencies in Mexico and Canada resulted in a surge of foreign direct investment (FDI) in these countries, as shown in Figure 1-B. Between 1993 and 1999 (the most recent period for which data have been published), FDI in Mexico increased by 169%. It grew rapidly between 1993 and 1997, following the peso crisis, and then declined slightly afterwards, because of the steady appreciation of Mexico’s real exchange rate between 1995 and 2000.

Figure 1-B

FDI in Canada more than quadrupled between 1993 and 1999, an increase of 429%, largely as a result of the falling value of the Canadian dollar in this period. Inflows of FDI, along with bank loans and other types of foreign financing, have funded the construction of thousands of Mexican and Canadian factories that produce goods for export to the United States. Canada and Mexico have absorbed more than $151 billion in FDI from all sources since 1993. One result is that the U.S. absorbed an astounding 96% of Mexico’s total exports in 1999.[4] The growth of imports to the U.S. from these factories has contributed substantially to the growing U.S. trade deficit and the related job losses. The growth of foreign production capacity has played a major role in the rapid growth in exports to the U.S.

NAFTA costs jobs in every state
All 50 states and the District of Columbia have experienced a net loss of jobs under NAFTA (Table 1-2). Exports from every state have been offset by faster-rising imports. Net job loss figures range from a low of 395 in Alaska to a high of 82,354 in California. Other hard-hit states include Michigan, New York, Texas, Ohio, Illinois, Pennsylvania, North Carolina, Indiana, Florida, Tennessee, and Georgia, each with more than 20,000 jobs lost. These states all have high concentrations of industries (such as motor vehicles, textiles and apparel, computers, and electrical appliances) where a large number of plants have moved to Mexico.

Table 1-2

While job losses in most states are modest relative to the size of the economy, it is important to remember that the promise of new jobs was the principal justification for NAFTA. According to its promoters, the new jobs would compensate for the increased environmental degradation, economic instability, and public health dangers that NAFTA brings (Lee 1995, 10-11). If NAFTA does not deliver net new jobs, it can’t provide enough benefits to offset the costs it imposes on the American public.

Long-term stagnation and growing inequality
NAFTA has also contributed to growing income inequality and to the declining wages of U.S. production workers, who make up about 70% of the workforce. NAFTA, however, is but one contributor to a larger globalization process that has led to growing structural trade deficits and has shaped the U.S. economy and society over the last few decades.[5] Rapid growth in U.S. trade and foreign investment, as a share of U.S. gross domestic product, has played a large role in the growth of inequality in income distribution in the last 20 years. NAFTA has continued and accelerated international economic integration, and thus contributed to the growing tradeoffs this integration requires.

The growth in U.S. trade and trade deficits has put downward pressure on the wages of “unskilled” (i.e., non-college-educated) workers in the U.S., especially those with no more than a high school degree. This group represents 72.7% of the total U.S. workforce and includes most middle- and low-wage workers. These U.S. workers bear the brunt of the costs and pressures of globalization (Mishel et al. 2001, 157, 172-79).

A large and growing body of research has demonstrated that expanding trade has reduced the price of import-competing products and thus reduced the real wages of workers engaged in producing those goods. Trade, however, is also expected to increase the wages of the workers producing exports, but growing trade deficits have meant that the number of workers hurt by imports has exceeded the number who have benefited through increased exports. Because the United States tends to import goods that make intensive use of less-skilled and less-educated workers in production, it is not surprising to find that the increasing openness of the U.S. economy to trade has reduced the wages of less-skilled workers relative to other workers in the United States.[6]

Globalization has reduced the wages of “unskilled” workers for at least three reasons. First, the steady growth in U.S. trade deficits over the past two decades has eliminated millions of manufacturing jobs and job opportunities in this country. Most displaced workers find jobs in other sectors where wages are much lower, which in turn leads to lower average wages for all U.S. workers. Recent surveys have shown that, even when displaced workers are able to find new jobs in the U.S., they face a reduction in wages, with earnings declining by an average of over 13% (Mishel et al. 2001, 24). These displaced workers’ new jobs are likely to be in the service industry, the source of 99% of net new jobs created in the United States since 1989, and a sector in which average compensation is only 77% of the manufacturing sector’s average (Mishel et al. 2001, 169). This competition also extends to export sectors, where pressures to cut product prices are often intense.

Second, the effects of growing U.S. trade and trade deficits on wages go beyond just those workers exposed directly to foreign competition. As the trade deficit limits jobs in the manufacturing sector, the new supply of workers to the service sector (displaced workers and new labor market entrants not able to find manufacturing jobs) depresses the wages of those already holding service jobs.

Finally, the increased import competition and capital mobility resulting from globalization has increased the “threat effects” in bargaining between employers and workers, further contributing to stagnant and falling wages in the U.S. (Bronfenbrenner 1997a). Employers’ credible threats to relocate plants, to outsource portions of their operations, and to purchase intermediate goods and services directly from foreign producers can have a substantial impact on workers’ bargaining positions. The use of these kinds of threats is widespread. A Wall Street Journal survey in 1992 reported that one-fourth of almost 500 American corporate executives polled admitted that they were “very likely” or “somewhat likely” to use NAFTA as a bargaining chip to hold down wages (Tonelson 2000, 47). A unique study of union organizing drives in 1993-95 found that over 50% of all employers made threats to close all or part of their plants during organizing drives (Bronfenbrenner 1997b). This study also found that strike threats in National Labor Relations Board union-certification elections nearly doubled following the implementation of the NAFTA agreement, and that threat rates were substantially higher in mobile industries in which employers can credibly threaten to shut down or move their operations in response to union activity.

Bronfenbrenner updated her earlier study with a new survey of threat effects in 1998-99, five years after NAFTA took effect (Bronfenbrenner 2000). The updated study found that most employers continue to threaten to close all or part of their operations during organizing drives, despite the fact that, in the last five years, unions have shifted their organizing activity away from industries most affected by trade deficits and capital flight (e.g., apparel and textile, electronics components, food processing, and metal fabrication). According to the updated study, the threat rate increased from 62% to 68% in mobile industries such as manufacturing, communications, and wholesale distribution. Meanwhile, in 18% of campaigns with threats, the employer directly threatened to move to another country, usually Mexico, if the union succeeded in winning the election.

The new study also found that these threats were simply one more extremely effective tactic in employers’ diverse arsenal for thwarting worker efforts to unionize. At 38%, the election win rate associated with organizing campaigns in which employers made threats was significantly lower than the 51% win rate where there were no threats. Win rates were lowest—32% on average—when threats were made during organizing campaigns involving more mobile industries, such as manufacturing, communications, and wholesale distribution. Among this last group, companies targeted for organizing are much likelier than they were in 1993-95 to be subsidiaries of large multinational parent companies with foreign operations, customers, and suppliers. The 30% win rate for organizing campaigns with these global multinational companies suggests that the existence of other sites in Latin America, Asia, or Africa serves as an unspoken threat of plant closing for many U.S. workers.

Bronfenbrenner (2000) described the impact of these threats in testimony to the U.S. Trade Deficit Review Commission:

Under the cover of NAFTA and other trade agreements, employers use the threat of plant closure and capital flight at the bargaining table, in organizing drives, and in wage negotiations with individual workers. What they say to workers, either directly or indirectly, is if you ask for too much or don’t give concessions or try to organize, strike, or fight for good jobs with good benefits, we’ll close, we’ll move across the border just like other plants have done before.[7]

In the context of ongoing U.S. trade deficits and rising levels of trade liberalization, the pervasiveness of employer threats to close or relocate plants may conceivably have a greater impact on real wage growth for production workers than does actual import competition. There are no empirical studies of the effects of such threats on U.S. wages, so such costs simply have been ignored by other studies of NAFTA.

NAFTA, globalization, and the U.S. economy
The U.S. economy created 20.7 million jobs between 1992 and 1999. All of those gains are explained by growth in domestic consumption, investment, and government spending. The growth in the overall U.S. trade deficit eliminated 3.2 million jobs in the same period (Scott 2000). Thus, NAFTA and other sources of growing trade deficits were responsible for a change in the composition of employment, shifting workers from manufacturing to other sectors and, frequently, from good jobs to low-quality, low-pay work.

Trade-displaced workers will not be so lucky during the next economic downturn. If unemployment begins to rise in the U.S., then those who lose their jobs due to globalization and growing trade deficits could face longer unemployment spells, and they will find it much more difficult to get new jobs.

When trying to tease apart the various contributing causes behind trends like the disappearance of manufacturing jobs, the rise in income inequality, and the decline in wages in the U.S., NAFTA and growing trade deficits provide only part of the answer. Other major causes include deregulation and privatization, declining rates of unionization, sustained high levels of unemployment, and technological change. While each of these factors has played some role, a large body of economic research has concluded that trade is responsible for at least 15-25% of the growth in wage inequality in the U.S. (U.S. Trade Deficit Review Commission 2000, 110-18). In addition, trade has also had an indirect effect by contributing to many of these other causes. For example, the decline of the manufacturing sector attributable to increased globalization has resulted in a reduction in unionization rates, since unions represent a larger share of the workforce in this sector than in other sectors of the economy.

So, although NAFTA is not solely responsible for all of the labor market problems discussed in this report, it has made a significant contribution to these problems, both directly and indirectly. Without major changes in the current NAFTA agreement, continued integration of North American markets will threaten the prosperity of a growing share of the U.S. workforce while producing no compensatory benefits to non-U.S. workers. Expansion of a NAFTA-style agreement—such as the proposed Free Trade Area of the Americas—will only worsen these problems. If the U.S. economy enters into a downturn or recession under these conditions, prospects for American workers will be further diminished.

April 2001

Jana Shannon provided research assistance and Jung Wook Lee provided administrative assistance. Eileen Appelbaum and Jeff Faux offered helpful comments.

EPI gratefully acknowledges the support of the Ford Foundation for the Workers and the Global Economy project.

Methodology used for job-loss estimates
This study uses the model developed in Rothstein and Scott (1997a and 1997b). This approach solves four problems that are prevalent in previous research on the employment impacts of trade. Some studies look only at the effects of exports and ignore imports. Some studies include foreign exports (transshipments) — goods produced outside North America and shipped through the United States to Mexico or Canada — as U.S. exports. Trade data used in many studies are usually not adjusted for inflation. Finally, a single employment multiplier is applied to all industries, despite differences in labor productivity and utilization.[8]

The model used here is based on the Bureau of Labor Statistics’ 192-sector employment requirements table, which was derived from the 1992 U.S. input-output table and adjusted to 1998 price and productivity levels (BLS 2001a). This model is used to estimate the direct and indirect effects of changes in goods trade flows in each of these 192 industries. This study updates the 1987 input employment requirements table used in earlier reports in this series (Rothstein and Scott 1997a and 1997b; Scott 1996).

We use three-digit, SIC-based industry trade data (Bureau of the Census 1994 and 2001), deflated with industry-specific, chain-weighted price indices (BLS 2001b). These data are concorded from HS to SIC (1987) classifications using conversion tables on the Census CDs. The SIC data are then concorded into the BLS sectors using sector-plans from the BLS (2001a). State-level employment effects are calculated by allocating imports and exports to the states on the basis of their share of three-digit, industry-level employment (BLS 1997).


  1. Potential jobs, or job opportunities, are positions that would have been created if the trade deficit with Mexico and Canada had remained constant, in real terms (and holding everything else in the economy constant). The total number of jobs and job opportunities is a measure of what employment in trade-related industries would have been if the U.S. NAFTA trade balance remained constant between 1993 and 2000, holding everything else constant.

  2. EPI calculations and International Monetary Fund (IMF) (2001).

  3. IMF (2001) and EPI calculations. This analysis compares overnight money market rates in Canada (annual averages) with the comparable federal funds rate for the U.S.

  4. Bureau of the Census (2000) and EPI calculations.

  5. Globalization includes rapid growth in imports, exports, and the share of trade in the world economy, and even more rapid growth in the international flows of foreign investment around the world. The term is also used to refer to the international convergence of rules, regulations, and even the social structure and role of government in many countries. This process is often viewed as a “race-to-the-bottom” in global environmental standards, wages, and working conditions.

  6. See U. S. Trade Deficit Review Commission (2000, 110-18) for more extensive reviews of theoretical models and empirical evidence regarding the impacts of globalization on income inequality in the U.S.

  7. Bronfenbrenner (1999).

  8. Other studies—see California State World Trade Commission (1997), which finds 47,600 jobs created in California from increased trade with Canada alone—have allocated all employment effects to the state of the exporting company. This is problematic, because the production—along with any attendant job effects—need not have taken place in the exporter’s state. If a California dealer buys cars from Chrysler and sells them to Mexico, these studies will find job creation in California. However, the cars are not made in California; so the employment effects should instead be attributed to Michigan and other states with high levels of auto industry production. Likewise, if the same firm buys auto parts from Mexico, the loss of employment will occur in auto-industry states, not in California.

Bernstein, Jared. 2001. Jobs Picture. Washington, D.C.: Economic Policy Institute. (March 9)

Blecker, Robert A. 1997. NAFTA and the Peso Collapse—Not Just a Coincidence. Briefing Paper. Washington, D.C.: Economic Policy Institute.

Bronfenbrenner, Kate. 1997a. “The effects of plant closings and the threat of plant closings on worker rights to organize.” Supplement to Plant Closings and Workers’ Rights: A Report to the Council of Ministers by the Secretariat of the Commission for Labor Cooperation. Lanham, Md.: Bernan Press.

Bronfenbrenner, Kate. 1997b. “We’ll close! Plant closings, plant-closing threats, union organizing, and NAFTA.” Multinational Monitor. Vol. 18, No. 3, pp. 8-13.

Bronfenbrenner, Kate. 1999. “Trade in traditional manufacturing.” Testimony before the U.S. Trade Deficit Review Commission. October 29. <>

Bronfenbrenner, Kate. 2000. “Uneasy terrain: The impact of capital mobility on workers, wages, and union organizing.” Commissioned research paper for the U.S. Trade Deficit Review Commission. <>

Bureau of the Census. 1994. U.S. Exports of Merchandise on CD-ROM (CDEX, or EX-145) and U.S. Imports of Merchandise on CD-ROM (CDIM, or IM-145). Preliminary data for December 1993 (month and year to date). Washington, D.C: U.S. Department of Commerce, Bureau of the Census.

Bureau of the Census. 2001. U.S. Exports of Merchandise on CD-ROM (CDEX, or EX-145) and U.S. Imports of Merchandise on CD-ROM (CDIM, or IM-145). Preliminary data for December 2000 (month and year to date). Washington, D.C.: U.S. Department of Commerce, Bureau of the Census.

Bureau of Labor Statistics. 1997. ES202 Establishment Census. Washington, D.C.: U.S. Department of Labor.

Bureau of Labor Statistics, Office of Employment Projections. 2001a. Employment Outlook: 1994-2005 Macroeconomic Data, Demand Time Series and Input Output Tables. Washington, D.C.: U.S. Department of Labor. <>

Bureau of Labor Statistics, Office of Employment Projections. 2001b. Private email communication with Mr. James Franklin about 2000 price deflator estimates.

California State World Trade Commision. 1996. A Preliminary Assessment of the Agreement’s Impact on California. Sacramento, Calif.: California State World Trade Commission.

International Monetary Fund. 2001. International Financial Statistics. Database and browser, March.

Lee, Thea. 1995. False Prophets: The Selling of NAFTA. Briefing Paper. Washington, D.C.: Economic Policy Institute.

Mishel, Lawrence, Jared Bernstein, and John Schmitt. 2001. State of Working America: 2000-01. An Economic Policy Institute book. Ithaca, N.Y.: ILR Press, an imprint of Cornell University Press.

Rothstein, Jesse and Robert E. Scott. 1997a. NAFTA’s Casualties: Employment Effects on Men, Women, and Minorities. Issue Brief. Washington, D.C.: Economic Policy Institute.

Rothstein, Jesse, and Robert E. Scott. 1997b. NAFTA and the States: Job Destruction Is Widespread. Issue Brief. Washington, D.C.: Economic Policy Institute.

Scott, Robert E. 1996. North American Trade After NAFTA: Rising Deficits, Disappearing Jobs. Briefing Paper. Washington, D.C.: Economic Policy Institute.

Scott, Robert E. 2000. The Facts About Trade and Job Creation. Issue Brief. Washington, D.C.: Economic Policy Institute.

Tonelson, Alan. 2000. Race to the Bottom. New York, N.Y.: Westview Press.

Trade Deficit Review Commission. 2000. The U.S. Trade Deficit: Causes, Consequences, and Recommendations for Action. Washington, D.C.: U.S. Trade Deficit Review Commission.

[back to top]

For a printer-friendly version of this report, download an Acrobat PDF version of this paper.


EPI home

back to co-globalize | rat haus | Index | Search | tree