NAFTA In Ohio

Originally Posted: February 25th, 2008

The debate surrounding NAFTA has always provided interesting questions for economists. The discourse leading up to the Ohio primary however, provided me with a particularly succinct and interesting question: Was there a way to measure the jobs lost to NAFTA in Ohio? While not an absolutely precise measure, the answer was yes. Whether the loss of jobs was worth the overall economic gains of NAFTA is another question entirely, though in terms of settling the argument between Senators Obama and Clinton, that part was fairly irrelevant.

So it would seem that strong words are flying once more in the Democratic race for the Presidential nomination, and this time it's around NAFTA (and all manner of other things, though this seems to me to be the only worthy point of contention this week). It seems that once again, the conventional wisdom is prevailing - but don't buy it for a second.

NAFTA was largely a success. That is, it did what it was supposed to. The majority of economists agree on this, and it really isn't debatable. As paraphrased from Moss' article, "Economic Developments during NAFTA's First Decade", the main provisions of the North American Free Trade Agreement were to eventually eliminate both tariff and non-tariff barriers to goods and services trade, to phase in free trade in agriculture, textiles, and automobiles, and to liberalize cross-border investment, all between Canada, the USA, and Mexico. Since going into effect on January 1st 1994, the vast majority of economists agree that the agreement had a small, though net positive effect for the United States, and a larger net positive effect for Mexico. (I neglect Canada here due to their prior free trade agreement with the US.)

All in all, returns to the U.S. from foreign direct investment in Mexico grew, and quantities of imports to and exports from the United States increased, as did tertiary sector jobs in the U.S., and secondary sector jobs in Mexico. Such benefits were the direct causal result of NAFTA.

What about unemployment? The problem for economists here is that isolating the other individual effects of NAFTA on the economy is a particularly tricky puzzle. Where unemployment is concerned, the typical rate in the U.S. is around 6%. Sure, unemployment fluctuated in both directions during the period of time that has elapsed since NAFTA was ratified - but who is to say how many of those jobs were lost or created due to NAFTA? Especially when one considers the range of other economic factors in play. The regular business cycle. The Mexican Peso Crisis. Supply shocks. The boom of the early to mid 1990's. I could go on.

Luckily, for those of us that are actually wondering, policy makers knew that NAFTA would bring significant change to the structure of the American economy. Many of NAFTA's proponents were right - jobs would be created as a result of increased capital for investment due to lower tariffs, etc. Many of NAFTA's detractors however, were also right. Lower skilled jobs, particularly in manufacturing, would be destroyed as capital holders took advantage of cheaper labor in Mexico and moved factories there. As a result, they created the NAFTA-Transitional Adjustment Assistance program (hereafter referred to as the NAFTA-TAA program).

The NAFTA-TAA program sought to provide assistance to workers "who lose their jobs or whose hours of work and wages are reduced as a result of trade with, or a shift in production to, Canada or Mexico", as well as both their upstream and downstream counterparts in industries with shared linkages. As a result, there is actually data that points to the number of jobs lost due to NAFTA. This data is not perfect - it is likely to understate the number of jobs lost, since fewer workers were aware of the programs existence. It is also likely to overstate the number of jobs lost, since not all workers will have lost their jobs. Perhaps these balance out - either way, NAFTA-TAA's certification process is the best available count of the number of workers displaced due to increased NAFTA exports.

So what do these numbers indicate? Approximately 3/4 of a million certifications were issued between NAFTA's ratification in 1994 and 2002. That's around 90,000 jobs lost a year, which, for the entire United States, is actually a fairly small number when compared to national unemployment. From this perspective, the Clinton's argument is fairly sound - NAFTA did little damage when it came to unemployment nationally, and in fact, the jobs created in the service industry are pointed to by many as part of the reason for the boom of the 90's.

Regionally however, the picture is very different. The majority of the manufacturing industry in the United States is densely concentrated by region, especially when one considers the concentration via migration of manufacturing infrastructure to the Midwest through the 80's, which persists even to this day. According to the American Manufacturing Association, 235,000 jobs have been lost in the manufacturing industry in the state of Ohio since 2000. Since January 2001, the TAA program has certified 56,414 workers in Ohio alone. That's around 24% of the manufacturing jobs lost in Ohio that can be directly tied to NAFTA, and it's probably a low estimate. From this perspective, Obama's argument is solid. Ohio voters can certainly blame a significant portion of the unemployment in their state's manufacturing industry on NAFTA, and it's something Hilary Clinton should own up to and provide one of her much vaunted "solutions" to, if she's to be believed, rather than trying to call her opponent a liar.

So there's the beef. There are winners and losers in free trade. Get used to it. Sadly, Ohioans* got the shitty end of the stick on this one.

*More specifically, Ohio manufacturers.