Happily Ever NAFTA?

The North American Free Trade Agreement (NAFTA) has sparked fierce academic and political disputes -- not to mention an armed rebellion or two. Looking back on its nearly nine years of existence, has NAFTA delivered or disappointed? The answer will go a long way toward determining the future of regional trade pacts. U.S. critics clash with Mexico's original NAFTA architects on whether free trade in North America is a blessing or a curse.

A Bad Idea That Failed
By John Cavanagh and Sarah Anderson

A Bad Idea That Failed
By John Cavanagh and Sarah Anderson

NAFTA offers a rocky road map for the Americas. More than eight years of monitoring reveal that, yes, the accord has boosted investment and trade, just as the negotiators promised. And yes, increased international competition may have helped fuel the dramatic rise in labor productivity rates during the 1990s, particularly in Mexico and the United States. But workers, communities, and the environment in all three countries have suffered from the agreement’s flaws.

In Mexico, for example, 50 percent productivity growth didn’t prevent an 11 percent slide in real manufacturing wages between 1994 and 2001, according to a Global Policy Network study by Mexican labor economist Carlos Salas. The U.S. government reports that even in nominal dollar-value terms, Mexican manufacturing wages were no higher in 2000 than in NAFTA’s first year and considerably lower than in 1981, prior to Mexico’s sweeping free market reforms.

Making ends meet in Mexico’s rural areas is even tougher. NAFTA opened the floodgates to cheap U.S. corn imports, leading to an 18-fold increase between 1993 and 2000. The devastating impact on Mexican small farmers is reflected in the rising rural poverty rate, which climbed from 79 percent in 1994 to 82 percent in 1998, according to the World Bank.

Why have increased trade and investment failed to reduce poverty or raise wages? Part of the answer is that in a globalized marketplace, highly mobile employers have even more power to suppress workers who fight for their fair share of the benefits. And these firms often find allies among governments desperate for foreign investment. Just ask the mostly female workers at Duro Bag Manufacturing in Río Bravo, Mexico, a U.S.-owned plant that makes decorative bags for Hallmark. When these workers demonstrated peacefully in June 2000 for their right to form an independent union, local police swept in and beat them, reportedly sending one pregnant woman to the hospital. Later the Fox administration reneged on promises to allow a secret ballot in the union election, forcing terrified Duro workers to vote in front of management, with armed thugs allegedly hovering nearby.

In the United States, workers face globalization pressures of their own. Cornell University Prof. Kate Bronfenbrenner has documented how U.S. employers increasingly threaten to move their factories to Mexico and other low-wage countries in order to fight unions and restrain wages. Such "whipsaw bargaining" was a major factor in the meager level of U.S. real wage growth in the late 1990s, despite near-record low unemployment.

Unfortunately, the agency set up under the NAFTA labor side agreement has proved incapable of holding governments or corporations accountable for worker rights violations. More than 20 complaints have been filed regarding alleged violations in all three NAFTA countries, but in not a single case has the process yielded more than a bit of public exposure to the problem.

Residents on both sides of the U.S.-Mexico border also face rising environmental hazards related to the NAFTA-induced industrial development that has far outstripped investment in environmental infrastructure. Although NAFTA promoters theorized that trade-related economic growth would produce greater environmental spending, a forthcoming Tufts University study reveals that such expectations were pure fantasy. Despite steady growth in gross domestic product, Mexican government investment in environmental protection has declined in real terms by about 45 percent since 1994. Environmental funding from a trinational commission established under NAFTA has amounted to a paltry $3 million per year. Meanwhile, air pollution from Mexican manufacturing has nearly doubled.

When officials in any of the NAFTA countries attempt to tackle environmental problems through regulation, they face the threat of expensive lawsuits, thanks to NAFTA rules allowing foreign investors to sue governments directly over any act that might diminish the value of their investment. Following one such suit, the Mexican government was ordered to pay nearly $17 million to a California firm that was denied a permit from a municipality to operate a hazardous waste treatment facility in an environmentally sensitive location. Similar suits in Canada and the United States have stirred up rancor among state and local governments that have historically supported free trade agreements.

Of course, not everyone is worse off under NAFTA. According to a January 2002 International Monetary Fund working paper by Ana Corbacho and Gerd Schwartz, increased incomes at the top explain why inequality in Mexico was higher in 2000 than in any year since the mid-1980s. In Canada and the United States, the wealth gap has also widened.

Ten years ago, we cautioned NAFTA negotiators to heed the lessons of the European Union, where a "social protocol" combined with the channeling of resources into the poorer nations has helped level the playing field as economic integration advanced. Instead, the negotiators argued that free trade alone would lift all boats. We argued that strong controls were needed to ensure that trade and investment supported social goals, rather than the narrow interests of large corporations. Again, we were ignored. Today, trade officials continue to ignore these concerns, despite mounting opposition to NAFTA-style free trade across the hemisphere.

The Proof Is in the Paycheck
By Jaime Serra and J. Enrique Espinosa

NAFTA’s fundamental objectives as a free trade and investment pact have been achieved. John Cavanagh and Sarah Anderson even acknowledge that "yes, the accord has boosted investment and trade, just as the negotiators promised. And yes, increased international competition may have helped fuel the dramatic rise in labor productivity rates during the 1990s, particularly in Mexico and the United States." For instance, compare the evolution of Mexican exports to the United States with the rest of Latin America’s exports to the United States. Ten years ago, exports from Mexico and from the rest of the region were about equal. Today, Mexico’s exports are nearly double those of the rest of the Latin American countries put together. When looking at foreign direct investment (FDI), the results for Mexico have been equally dramatic. During the eight years before NAFTA, the average annual flow of fdi into Mexico was approximately $3.47 billion. But since 1994, this average has exceeded $13 billion. Such growth confirms that the preferential market access granted by NAFTA has indeed produced increased trade and investment flows.

Cavanagh and Anderson argue that NAFTA has had important adverse effects on wages, agricultural activities, the environment, and income distribution. But their evidence is weak and casuistic. Consider each of their claims:

First, their assertion that real wages have declined in Mexico is based on a biased selection of dates. The base years they use (1981 and 1994) happen to be years in which the Mexican peso was historically overvalued, a condition soon corrected by massive devaluations. Thus, wages measured in dollars for those years are grossly distorted. In fact, a careful analysis shows opposite results: There is a clear positive relationship between the growth of Mexico’s dollar-denominated manufacturing wages and the growth of exports. This evidence shows that one cannot acknowledge NAFTA’s positive impact on trade without recognizing NAFTA’s favorable effects on Mexican wages. This result should provide some comfort to those, like Cavanagh and Anderson, who are concerned with the migration of U.S. jobs to lower-wage countries. Wages and labor conditions in these countries are often better in export-oriented firms than in traditional nonexporting companies. From 1994 to 1996, Mexican firms that exported more than 80 percent of their total sales paid between 58 and 67 percent higher wages than the average wage rate. Over time, access to markets of industrialized countries, such as that obtained by Mexico under NAFTA, will help narrow the wage gap.

Regarding agriculture, Cavanagh and Anderson claim that "NAFTA opened the floodgates to cheap U.S. corn imports… [with a] devastating impact on Mexican small farmers." But they misrepresent the facts. First, since NAFTA entered into force, the average duty for U.S. corn imports into Mexico has been either 177.4 percent or $0.1695 per kilogram (whichever is larger), and the sector will not even be fully liberalized until 2008. Second, under NAFTA’s tariff-rate quota system, a certain quantity of a product can enter duty-free; all imports exceeding that quota are subject to tariffs. Since 1994, total U.S. corn imported into Mexico has exceeded the tariff-rate quota by 162 percent. This excess in imports has resulted from internal supply shortages and cannot be considered an effect of NAFTA. In this sense, tariff quotas have not been binding. The poverty of Mexican farmers is a legitimate concern, but it is hardly attributable to NAFTA.

Cavanagh and Anderson also point to the declining investment by the Mexican government in environmental protection. But they fail to recognize that overall public investment has been declining due to budgetary constraints — a problem that has little to do with NAFTA. We have yet to meet a Mexican citizen who does not desire a better environment. But the issue involves availability of funds rather than good wishes. Trade liberalization has increased per capita income in Mexico and has thus created better opportunities to tackle its serious environmental problems. The true risk is that protectionist measures might be adopted to punish polluters. Cavanagh and Anderson worry about the "expensive lawsuits" that foreign investors may file when individual NAFTA countries attempt to regulate environmental conditions. However, the authors ignore the July 2001 clarification letter signed by the three NAFTA parties, regarding fair and equitable treatment in the dispute settlement mechanisms established under Chapter 11 of NAFTA.

Finally, the serious inequality of Mexico’s income distribution remains a secular problem, fundamentally explained by deep educational and cultural factors rather than by specific trade regimes. But if anything, one could argue that trade protection boosts profit margins, reduces the scale of operations of firms, and maintains high unemployment and underemployment rates, thus keeping wages low. Mexico was a heavily protected economy for over six decades and has only been liberalizing its trade for the last 15 years. And NAFTA has only reached its eighth anniversary. Hard data show that trade liberalization tends to improve income distribution: Blue-collar wages and jobs respond more positively to exports than do white-collar wages and jobs. We agree with Cavanagh and Anderson that this positive trend could be accelerated under a social program funded by the wealthier NAFTA parties, thereby closing the social gap between our countries.

Nice Theories, Sad Realities
John Cavanagh and Sarah Anderson respond.

Predictably, Serra and Espinosa focus on trade and investment flows to gauge NAFTA’s success. But NAFTA was not sold on these terms. All three governments boasted that the deal would support broader social goals, from creating good jobs to cleaning up the environment. Former Mexican President Carlos Salinas de Gortari even promised that NAFTA would give such a boost to Mexican living standards that illegal immigration to the United States would drop. On these indicators, NAFTA is a failure.

On wages, Serra and Espinosa object to our use of U.S. Labor Department data denominated in dollars. However, they offer no evidence to dispute our peso-denominated figures on real wages. Although Mexican factory workers saw some improvement in 2000 and 2001, their real wages remain lower than in the year NAFTA began. We also question the claim that export-oriented firms pay higher wages. More recent data from the U.S. Labor Department show that in 1999, hourly wages in the maquiladora export plants were considerably lower than in Mexican manufacturing as a whole ($1.74 versus $2.12). Perhaps most damning, the share of Mexicans living in poverty rose from 51 percent in NAFTA’s first year to more than 58 percent four years later, according to World Bank data.

The authors ignore our argument that disappointing wage levels in all NAFTA countries are partly due to the agreement’s failure to protect the right to organize independent unions. This omission was not surprising since, according to congressional testimony by AFL-CIO official Thea Lee, Serra once told a group of Mexican business people not to worry about NAFTA’s side agreement on labor, reassuring them that it was too full of loopholes to pose any threat.

Moreover, Mexico’s bargain wages have attracted foreign investors, but the strategy is short-lived at best, since countries such as China can offer even lower labor costs. Nearly 500 maquiladoras closed their doors between January 2001 and March 2002, victims of either the U.S. downturn or relocation to Asia.

We remain unconvinced that Mexico’s inequality is rooted not in economic policy but in "deep educational and cultural factors." According to Corbacho and Schwartz, "following several decades in which the country moved toward a more even distribution of income, Mexico’s income disparities have generally widened since the 1980s." We’re not aware of changes in Mexico’s education or culture during that time that would explain the growing gap. We do know, however, that this period marked the era of Mexico’s economic liberalization.

For free traders, Serra and Espinosa seem awfully proud of the remaining protections on U.S. corn imports. However, by citing average tariffs, they mask the fact that NAFTA requires a phaseout of such barriers. The Mexican government has accelerated the process by declining to collect even the full amount allowed. As protections fall to zero by 2008, small farmers are being devastated by competition from cheap subsidized U.S. corn. Tens of thousands of protesters already have marched on Mexico City to register their opposition.

On the environment, Serra and Espinosa recite the orthodox mantra that "trade liberalization has increased per capita income in Mexico and has thus created better opportunities to tackle its serious environmental problems." They then concede that no actual improvements have occurred. Once again, reality conflicts with textbook theory. The drop in environmental spending is blamed on overall budgetary constraints, but in fact Mexico’s real government expenditures were about level throughout the 1990s. The most puzzling statement of all was that "the true risk is that protectionist measures might be adopted to punish polluters." Do Serra and Espinosa mean to suggest that it is more important to protect polluters than communities? What music to the ears of the global firms that poison the border region’s water and air!

Finally, Serra and Espinosa attempt to calm fears about NAFTA’s excessive investor protections by referring to a 2001 official "clarification." Unfortunately, this document ignored the most objectionable aspects of NAFTA’s Chapter 11 rules. It did nothing to address the overly broad definition of expropriation that allows corporations to sue over any government act that may diminish the value of a foreign investment. Nor did it respond to criticism of the unaccountable and secretive arbitration panels that rule on Chapter 11 claims. By allowing profit-hungry corporations to undermine public-interest regulations, these protections are an assault on democracy.

We believe NAFTA is the wrong model for hemispheric integration. But we do not oppose new rules to govern relations among the nations of the Americas. We have engaged in collaborative processes to develop positive, alternative platforms under the auspices of the International Forum on Globalization, a global alliance of researchers and activists, and the Hemispheric Social Alliance, a coalition of unions and civil society networks representing about 50 million people in the Western Hemisphere. Rather than allow a race to the bottom, these alternatives would create protections for small farmers as well as for water and other common resources. They would place checks on the global financial casino, use debt cancellation and other measures to attack inequality, and guarantee that investment does not undermine workers’ rights or the environment. These measures are not the sort of "protectionism" vilified by Serra and Espinosa but rather sound rules and incentives that harness trade and investment to support sustainable societies everywhere.

More Accuracy, Less Activism
Jaime Serra and J. Enrique Espinosa respond.

Cavanagh and Anderson cannot help but acknowledge that, in terms of its explicit trade and investment objectives, NAFTA has been an unqualified success. However, they invoke the neoprotectionst cliché that the agreement "was not sold on these terms." They correctly note that NAFTA was also intended to support broad social goals, including job creation and environmental protection, but they conclude on the basis of superficial and biased analysis that NAFTA has failed. Their arguments provide an excellent example of the misleading results obtained when political activism taints serious research.

Cavanagh and Anderson point out that "although Mexican factory workers saw some improvement in 2000 and 2001, their real wages remain lower than the year NAFTA began." However, they fail to acknowledge that average real hourly wages declined in Mexican manufacturing only in 1995 and 1996 and that such decline is explained by the severe rebound in inflation in those years, as internal prices adjusted to the massive devaluation of the peso in late 1994 and 1995. This devaluation can hardly be attributed to NAFTA, since it occurred when the agreement’s gradual liberalization was only beginning. Instead, it stemmed from a mismatch between Mexico’s payment obligations to international creditors and the foreign exchange available to meet them. This crisis was similar, in many respects, to those experienced more recently by several Asian and Latin American countries — none of which is party to a NAFTA-like agreement with the United States or with any other industrialized nation.

According to data published by the National Institute of Statistics, Geography and Informatics (Mexico’s leading source of economic statistics), the average hourly real wage for factory workers in Mexican manufacturing fell 20 percent between 1993 and 1996 but has increased every year since, as the NAFTA phaseout of trade barriers took hold. By 2001, this wage had recovered 95 percent of its pre-NAFTA value. More important, the comparable average hourly wage in the maquiladora export plants fell by less than 14 percent from 1993 to 1996 and by 2001 showed a 8.4 percent gain over its pre-NAFTA level.

Hourly wages in maquiladora export plants are lower than in Mexico’s manufacturing industry, as the former tend to employ workers with lower skill. However, this gap narrowed by nearly 25 percent between 1993 and 2001, even as the number of workers employed by maquiladoras more than doubled during this period. Clearly, low-skilled Mexican workers have benefited substantially from NAFTA. In fact, they seem to have benefited more than the higher skilled workers in Mexico’s manufacturing industry, where the number of blue-collar jobs fell by about 12 percent between 1993 and 1995 but more than recovered by 2000. These figures suggest that NAFTA has helped moderate Mexico’s income inequality.

Cavanagh and Anderson’s lack of objectivity is also obvious when they declare that NAFTA’s alleged harm to wages and employment results partly from "the agreement’s failure to protect the right to organize independent unions." All they show in support of this claim is the hearsay testimony of an AFL-CIO official. Moreover, they ignore that NAFTA’s side agreement on labor was always intended to protect workers’ rights while preventing the undue use of labor-related claims as protectionist tools. Any serious analyst of Mexican labor unions would agree that their political independence has greatly increased since the onset of NAFTA. Such independence was a factor in the 2000 election of an opposition candidate as president of Mexico.

Convinced as we are of the benefits of free trade, we have always realized that liberalizing trade in agricultural goods ranks among the most difficult topics in multilateral and regional trade negotiations. But unlike Cavanagh and Anderson, we are also aware that duty-free U.S. corn imports benefit the poorest among Mexico’s poor — rural workers with no land of their own who grow no corn but need to buy tortillas. Such imports also benefit farmers who use corn as an input (such as poultry and livestock producers). Any unbiased analysis of trade liberalization, even in agriculture, must balance the benefits accrued to consumers against the costs incurred by producers. The need to attain this delicate balance was recognized at all times when NAFTA’s agricultural provisions were negotiated.

Cavanagh and Anderson dismiss the notion that, by helping raise per capita income, liberalized trade improves the ability of countries to protect the environment. In so doing, our debate counterparts dismiss the findings of many serious studies linking economic development and environmental protection as "orthodox mantra." On this topic, we encourage them to read the World Bank’s "World Development Report 1992," devoted entirely to exploring these links.

As part of the team that crafted NAFTA, we are convinced that the agreement has supported broad social goods, including job creation and environmental protection. However, we have always acknowledged that the agreement would not provide an instant and universal remedy to Mexico’s problems. In a 1992 speech presenting the NAFTA text to the Mexican Senate for approval, one of us warned that "the agreement will not be a panacea. Now that the negotiations have concluded, I want to reiterate this idea so that no false expectations are created."

More from Foreign Policy

Residents evacuated from Shebekino and other Russian towns near the border with Ukraine are seen in a temporary shelter in Belgorod, Russia, on June 2.
Residents evacuated from Shebekino and other Russian towns near the border with Ukraine are seen in a temporary shelter in Belgorod, Russia, on June 2.

Russians Are Unraveling Before Our Eyes

A wave of fresh humiliations has the Kremlin struggling to control the narrative.

Chinese President Xi Jinping (R) and Brazilian President Luiz Inácio Lula da Silva shake hands in Beijing.
Chinese President Xi Jinping (R) and Brazilian President Luiz Inácio Lula da Silva shake hands in Beijing.

A BRICS Currency Could Shake the Dollar’s Dominance

De-dollarization’s moment might finally be here.

Keri Russell as Kate Wyler in an episode of The Diplomat
Keri Russell as Kate Wyler in an episode of The Diplomat

Is Netflix’s ‘The Diplomat’ Factual or Farcical?

A former U.S. ambassador, an Iran expert, a Libya expert, and a former U.K. Conservative Party advisor weigh in.

An illustration shows the faces of Chinese President Xi Jinping and Russian President Vladimir Putin interrupted by wavy lines of a fragmented map of Europe and Asia.
An illustration shows the faces of Chinese President Xi Jinping and Russian President Vladimir Putin interrupted by wavy lines of a fragmented map of Europe and Asia.

The Battle for Eurasia

China, Russia, and their autocratic friends are leading another epic clash over the world’s largest landmass.