Our friends at the Economic Policy Institute are at it again, issuing another study this week that shows some particular trade agreement has cost X thousands of jobs over a certain number of years.

The latest target of EPI’s flawed model is the North American Free Trade Agreement. Enacted in 1994, NAFTA has created a free trade zone comprising the United States, Canada, and Mexico. According to the EPI report,

U.S. trade deficits with Mexico as of 2010 displaced production that could have supported 682,900 U.S. jobs; given the pre-NAFTA trade surplus, all of those jobs have been lost or displaced since NAFTA. This estimate of 682,900 net jobs displaced takes into account the additional jobs created by exports to Mexico.

The report’s author, Robert Scott, claims it foreshadows job losses if Congress passes pending trade agreements with South Korea, Colombia, and Panama.

The EPI model has little relevance to the real American job market. As I’ve pointed out before (here and here), its model is based on an overly narrow view of trade’s impact on the job market. Yes, some people do lose their jobs because of import competition, no news there, but trade also creates jobs through increased exports. And even if we run a trade deficit with a country such as China or Mexico, jobs are also being created by the net inflow of foreign capital, which spurs domestic job creation through lower interest rates and direct investment. The money we save from lower-priced imports also liberates consumer dollars to fuel growth elsewhere in our economy, and cuts costs for import-consuming businesses, boosting their sales and employment.

Next, consider the EPI numbers on their face. Those alleged 682,900 net jobs lost came over a 16-year period. That’s a bit more than 40,000 jobs lost per year. That is a drop in the bucket in a dynamic economy like ours that creates and eliminates about 15 million jobs each year. Even when unemployment is low, 300,000 or more Americans file for unemployment insurance in a typical week. So even if true, the EPI job loss numbers amount to less than one day’s worth of job displacement for the whole year.

When we look at the actual job market performance since NAFTA was enacted, the irrelevance of the EPI model becomes plain. In the first five years after NAFTA’s passage, 1994-98, when we could have expected it to have the most impact, the U.S. economy ADDED a net 15 million new jobs, including 700,000 manufacturing jobs. In the 16 years since its passage, despite two recessions, our economy still employs 20 million more workers than it did the year before NAFTA passed. (Check out the employment tables in the latest Economic Report of the President.)

In my own April 2011 study of trade and the economy, “The Trade-Balance Creed,” I found that civilian employment in the past 30 years has actually grown quite a bit faster during periods of rising trade deficits compared to periods of declining deficits, just the opposite of what EPI’s distorted model would predict.