The "giant sucking sound" Ross Perot used to talk about is back, only this time it is not Mexico sucking away American jobs. It is China sucking away Mexico's jobs. And jobs from Taiwan and South Korea, Singapore and Thailand, Central and South America, and even from Japan. Globalization is entering a fateful new stage, in which the competitive perils intensify for the low-wage developing countries much like the continuing pressures on high-wage manufacturing workers in the United States and other advanced economies. In the "race to the bottom," China is defining the new bottom.

This turn of events is difficult to see against the gathering threat of global recession, but in the long run it will be more meaningful. As one economy after another sinks into contraction, output subsides nearly everywhere–more layoffs and closed factories, more unsold goods. So the migration of production to China will not become fully apparent until after the recovery, when some of the closed factories never reopen. While it is impossible to know the full dimensions at this point, the downdraft on wages and competing economies induced by China's ascendancy may produce a terrible reckoning. For many poor nations that thought they had gained a foothold on the ladder, the reversal will be quite ugly.

This is the "treadmill" that ensnares developing countries–writ large. If they attempt to boost wages or allow workers to organize unions or begin to deal with social concerns like health or the environment, the system punishes them. The factories move to some other country where those costs of production do not exist.

In Mexico, the manufacturing wage level rose a bit in the last couple of years and is now around $1.50 an hour. In China, it is 20-25 cents an hour. After NAFTA, Mexico's manufacturing base expanded robustly year after year–except that most new factories are located in the maquiladora export zones along the US border and in the interior, essentially separate from the Mexican economy and largely producing components for US multinationals. Yet Mexico may already have peaked as an emerging player in global manufacturing. Its manufacturing base is now shrinking, due first to the US recession but also because the factories are leaving. American companies that were cheerleaders for NAFTA back in 1993 are shutting down and moving to greener–that is, cheaper–pastures. An American source in multinational business explained the trend: "When you consider the wage difference, moving the factory, which was usually leased anyway, or moving the more value-added product lines is a veritable no-brainer, if you want to increase profits. I expect this to intensify over the next few years, leaving considerable excess capacity and unemployment, particularly in northern Mexico but also in Central and South America, and the Caribbean."

During the past year, employment in the maquiladora industries fell by 12 percent, more than 170,000 jobs from the peak. The number seems modest by American standards, but those jobs have been the core of positive growth. The maquiladora sector produces about one-third of the nation's hard-currency income from abroad–the dollars that support its foreign borrowing–and so its loss could contribute to yet another currency crisis. The export-zone wages may seem pitiful to Americans (and to many Mexicans), but they are virtually the only bright spot in job development.

Poignant evidence of Mexico's dilemma is the fact that the government has slapped antidumping duties of 189 percent on electronics imports from China. Electronics was supposed to be one of Mexico's bright spots, remember, but Mexico now claims China is exporting its surplus output at a price below what it costs in China. Guadalajara, a production center for dozens of US technology companies,was down 16 percent in exports and lost 15,000 jobs in the first half of the year, according to Business Week. When US steel companies pursue anti-dumping remedies, the free-trade orthodoxy disparages them as backward protectionists, blocking the future for poorer countries. But Mexico is still very poor itself and feeling the same squeeze. Mexico's top three steel producers, incidentally, are all in grave financial trouble, like the American companies, and for the same reason. Worldwide overcapacity drives down steel prices and rewards the lowest-wage producer–China.

In Mexico, other shrinking sectors include shoes, tires, apparel and auto parts. The Big Three are moving auto components to China from both Mexico and the United States. General Electric, which over the years has moved a lot of US jobs to Mexico, is now moving production of mini-bar refrigerators from there to China. An executive of SCI Systems, which employs 10,000 in Guadalajara, told Business Week: "I'm an absolute believer in this country, but anything that is really price-sensitive is considering moving lock, stock and barrel to Asia."

Mexico provides a most ironic example because its supposed success after NAFTA has been so fictionalized by the American establishment. The general propaganda line–Mexico's modernization–is still the story told by US political leaders and the media. The case of Mexico matters especially because it is intimately integrated with the United States, and also because it is dramatic refutation of the upward-and-onward mythology promoted by globalization's advocates and apologists.

They would say this ongoing movement of production is the natural course of events: Nations start their development with the humblest industries–shirts, shoes and toys–then move up the ladder to higher-value production–cars, steel and semiconductors–that brings higher incomes and eventually middle-class prosperity. Japan led the way, South Korea and Taiwan followed. Dozens of others are trying to emulate Japan's export-led model. Now China is embarked on the same path. That familiar dogma will be severely tested in the next few years by China's enormous scale–1.2 billion people, relatively well educated and extremely cheap to employ. The sucking sound is not just jobs. China also takes scarce foreign capital from other developing countries, as investors seek the most promising returns. China, one should add, is not doing anything to gain its advantage that other nations didn't do before it or that the global system considers illegitimate.

But the killer question asked by critics, myself included, is whether China can fulfill its vast ambitions without smashing the dreams of other striving nations, not to mention manufacturing in the high-wage countries. Too many producers, too few consumers in a global system where too many workers cannot afford to buy the things they make–that's the central contradiction. The destructive qualities and repeated crises are sure to continue, critics would argue, so long as the system advances by this roving exploitation of labor and prevents developing countries from pursuing more balanced, albeit more gradual, strategies.

The first essential thing to understand about Mexico's dilemma is that its post-NAFTA miracle was utterly mythical for most Mexicans–confined to a few sectors and the wealthy. The success depended almost entirely on the investment in export production by foreign multinationals, most of them American but some from Europe and Japan. Since 1993 investment in new maquiladora factories has grown from $895 million to $2.8 billion a year, and manufacturing exports have expanded by 16 percent, due almost exclusively to the maquiladoras. This growth was supported almost entirely by US consumers. While Mexico tripled its trade surplus with the United States, it imported even more from other countries and thus runs a large and growing trade deficit overall.

Mexico's spectacular growth as an export platform was divorced from the larger reality of the country's economy, where things mostly got worse. Total foreign investment in Mexico actually fell, despite increased flows of financial speculation. Official unemployment declined, but this was grossly misleading because the government counts the rapid growth of part-time and even nonpaying jobs in the informal economy–street vendors, unpaid work in small family enterprises, for example. During the NAFTA years, the real incomes of Mexican wage earners lost 20 percent of their purchasing power as the labor market deteriorated, pushing many more workers down the job ladder. The national minimum wage, an important bargaining floor set by the government, was allowed to fall by nearly 50 percent in value during the decade.

Mexico City economist Carlos Salas reported, in an Economic Policy Institute survey, this startling fact about Mexico's supposed progress: "a virtual halt in the process of urbanization." In developing economies, people typically migrate from the impoverished countryside to the city, drawn by the growing availability of wage incomes. In Mexico, this long-running pattern stopped during the 1990s–a telltale sign that modernization actually stalled. "The entire process of development has been halted and, in some cases, it even may have been reversed," Salas wrote. In fact, notwithstanding NAFTA's propaganda, most Mexicans did not get any closer to an American standard of living, but fell farther away. In 1975 Mexican production workers earned 23 percent of US wages. They now earn 11 percent. Mexicans took the other option–heading north–in swelling numbers, and that migration is sure to grow as jobs in the maquiladoras disappear.

The corporate types, as they withdraw outposts from Juarez or Monterrey, are no longer claiming miracles for Mexico. Instead, they voice the usual complaints made about struggling poor countries: inadequate education, terrible infrastructure, lousy public services, too much red tape. That's true, of course, but it sounds like a nasty afterthought. The marriage Mexico's leaders made with US multinationals more or less guaranteed these outcomes, since NAFTA's narrow terms compelled the government to keep taxes down, cut public spending, suppress wages and do nothing to disturb the profitability of the new factories. That deal appears to be ending for Mexico. The companies are making a new one with China.

The coming economic crunch will likely be horrendous for millions of Mexicans and ought to inspire a progressive mobilization in the United States. This is an opportunity to change the politics in both countries: for the United States to forge a more sensible relationship with Mexico and for Mexico to devise an alternative format for development, far more equitable and forward-looking than the present system allows. Mexico, with honest help from the United States and Canada, could become the model for how human-scale globalization ought to work.

Achieving more meaningful economic and social integration obviously involves huge, complex issues, from Mexican immigration and US development aid to the relationships between currencies, legal systems and environmental standards. But the most difficult issue, one that cannot be evaded, is wages. No one should pretend that US-Mexican wage tensions can be entirely reconciled–of course not–but what is required is a wage-floor trade agreement that, as labor likes to say, "brings the bottom up, instead of pulling the top down." Mexico could only accept this arrangement if it had a genuine preferential status with the United States and Canada–including both significant trade privileges and investor guarantees of long-term commitments as well as serious aid for education, health and infrastructure. Think of this "North American union" as a first step toward someday imposing an international "living wage" standard on the production of traded goods, enforced by penalty tariffs on countries and companies that decline to participate. Producers would have a choice: Pay decent wages to their workers or pay penalty tariffs on their exports, the money to be recycled into development aid.

Obviously, the world is not ready for this (neither are Mexican and American politics), but the road to global reform has to start with a few like-minded nations willing to experiment with new terms because they see mutual self-interest in the bargain. A healthier, self-sustaining Mexico would be a lot better for the United States than a cheap-labor export zone that makes a few people very rich but survives on the backs of desperate immigrants and drug smugglers. US consumers might have to pay marginally higher prices on some items, but US commerce would gain a far more promising market for its exports, and that would help to reduce US trade deficits. Mexico would regain a measure of self-determination, the ability to chart its own course free of the neoliberal straitjacket.

The relationship would borrow a lot from the European Union's economic integration of rich and poor nations ranging from wealthy Germany to low-wage Portugal and Spain. The European Union delivers substantial aid conditioned on democratic standards and labor rights, implicitly encouraging rising wages in the poorer countries. The poorer countries enjoy the considerable trade advantages extended exclusively to EU members. A North American union, in addition to North/South development aid, would require concrete legal obligations: If US taxpayers are asked to invest in Mexico's future, US commerce cannot be allowed to enjoy NAFTA benefits, then pick up and leave whenever it sees fit.

The wage-enforcement system might not be as difficult to implement as it sounds. As things stand now, every component and input into goods produced in the three North American countries must be carefully measured to determine whether the goods qualify for NAFTA's trading preferences. Wage standards could simply be grafted onto those calculations. If the three countries formed a customs union similar to Europe's and adopted a common trade policy, this mechanism could also protect against free riders from outside–including China. None of this halts the "race to the bottom," not immediately at least, but it creates a reverse dynamic in global trade, a model that demonstrates there is a more promising path.

The usual multinational actors will naturally be opposed–their profits, after all, ride on roving exploitation–but developing countries, once they see the possibilities, might begin to change their views. Most of them well understand the high risks in export-led growth dominated by foreign capital, but it's the only game available. They understandably fear labor and environmental rights as threatening to their futures, and securing those values in international agreements is extremely unlikely until poorer nations can see that practical alternatives exist–a way to get off the treadmill. In the meantime, those countries are observing, once again, that what corporate globalization gives to the poor, it may also take away.