Seeing Through Nafta's New Clothes

To oppose the North American Free Trade Agreement (NAFTA) is to be labeled protectionist, jingoist, apologist for declining U.S. industries, as well as callously indifferent to Mexico's poverty. I am none of these. Rather, my case against NAFTA is Keynesian.

Keynesian economics holds that total purchasing power (aggregate demand) needs to roughly balance the economy's capacity to produce; otherwise, supply exceeds demand and productive potential goes unfulfilled. To have a fiscal and monetary policy, not to mention a labor policy, you need a government. But North America is neither a country nor a government. The U.S., Canada, and Mexico have radically different laws, living standards, and notions of what is minimally decent. By pretending we are one country, we risk the country with the lowest wages and the fewest labor rights and environmental protections getting the jobs.

The good free-trader replies that by embracing open trade, we stimulate efficiency and thereby improve those very conditions. But the recent golden age of growth was the post-World War II boom--when trade was encumbered by relatively high tariffs and regulatory barriers that sheltered national economic-development strategies. There was growing global commerce, but it was far from free trade. The advanced nations all had wages that rose with productivity, completing the Keynesian virtuous circle. Japan and Korea, the growth leaders, were among the most highly protected. Mexico, with state-owned and heavily regulated industries, enjoyed annual growth rates in excess of 5%, despite--or perhaps because of--economic nationalism.

FORCED CONVERSION. When Mexico abandoned its economic nationalism in the early 1980s, it was not because the policy had failed or because Mexican leaders had suddenly seen the light. It was because excessive foreign borrowing based on mistaken projections of oil prices--and crushingly high interest costs imposed by Paul A. Volcker's Federal Reserve Board--suddenly gave the U.S. leverage to demand that Mexico's leaders become converts to free-market policies.

In the 1980s, Mexico's real wages fell by over 30%. As interest rates have come down, Mexico has begun to recover, but real income is still well below its 1980 level. Against this history, NAFTA was devised as a reward for Mexico's forced conversion to the economic theories of the Reagan-Bush era. As repentant free-marketers, the Mexicans would enjoy preferential access to the U.S. market.

I offer this revisionist history not to commend protectionism but to suggest that the case for free trade is exaggerated. Extreme protectionism is surely bad. When every nation protects, as in the 1930s, the world economy contracts. But far more important than perfectly free trade is whether nations and the world system are pursuing high-growth, full-employment policies.

SHORT-LIVED BOOM. Defenders of NAFTA also claim that the gains of freer trade will be roughly symmetrical. As a poor, low-skill country, Mexico will attract low-skill jobs, leaving better ones to materialize here. But as University of California at Berkeley researcher Harley Shaiken has shown, there is a huge divergence between Mexico's rising skills and lagging wages. It is precisely this disparity that makes relocation there so attractive. As skilled jobs in the auto and electronics industries move south, there is no pressure to raise Mexican wages because of its massive unemployment. And as long as Mexico's wages lag behind its productivity, the purchasing power necessary to import goods from the U.S.--and hence to provide offsetting U.S. jobs--will lag, too. The current boom in exports of U.S. capital goods to Mexico is likely to be short-lived as Mexico diversifies its suppliers.

Henry Ford's insight was Keynesian: It's smart to pay employees enough to enable them to buy the products they make. But Mexico's auto workers, though nearly as productive as their U.S. counterparts, are paid under $2 an hour and cannot afford to buy the cars they build. As wages lag behind output, supply mutstrips demand. And as Mexico becomes an adjunct of the U.S. economy, the low-wage drag on Mexico's prosperity becomes a drag on our own.

To date, there is one useful byproduct of the NAFTA debate. Last month, when the proposed side agreement on labor standards was shown to House Majority Leader and NAFTA critic Richard A. Gephardt (D-Mo.), he dismissed it as window dressing. After hasty consultations, Mexican President Carlos Salinas de Gortari offered a new concession: Mexican wages would begin rising in proportion to Mexican productivity. This unenforceable promise introduces a Keynesian test into the trade debate.

If we truly wish to improve living standards in Mexico, it is not smart to throw away our own. Rather, we might gradually liberalize U.S.-Mexico trade if Mexican wages and conditions rise with productivity. Please note that this Keynesian case against NAFTA is rather different from that of Ross Perot, who is no Keynesian. Presumably this strange bedfellowship embarrasses Perot as much as it embarrasses me.

Before it's here, it's on the Bloomberg Terminal.