Friday, January 3, 2014

NAFTA, twenty years out

On December 8th, 1993 President Bill Clinton signed NAFTA--the North American Free Trade Agreement--into law. Earlier in that same year, NAFTA legislation was passed in both Mexico and Canada. Following its passage in the United States, NAFTA went into effect on January 1st, 1994, just about twenty years ago. The initial pact between the three North American nations was made in 1992, between President George H.W. Bush of the United States, Prime Minister Brian Mulroney of Canada, and President Carlos Salinas of Mexico.

Bush had very much wanted to achieve passage of the agreement during his first term, but despite his "fast track" authority (granted him by the Trade and Tariff Act of 1984) to make such a deal, he was unable to make it happen, thus it fell to Clinton--who was very much behind the idea--to shepherd it through Congress. This he did, with Republican support. NAFTA was--from the beggining--something of a political hot potato in the United States (and in Canada). Ross Perot's economic platform for his unsuccessful Presidential campaign in 1992 was largely built around his opposition to NAFTA, a key point that distinguished him from Bill Clinton and George H.W. Bush during the campaign and likely the chief reason for his notable success as a third-party candidate. His characterization of the widely predicted job flight to Mexico (if NAFTA passed) as a "giant sucking sound" is well-remembered to this day.

Even though there was significant support for NAFTA in both the Republican and Democratic parties, there was also significant opposition. It is difficult to define these divisions, from an ideological standpoint. There was, at the time, a great deal of criticism from NAFTA opponents in Congress of their brethren who supported the agreement. More often than not, the accusation leveled at the pro-NAFTA crowd was that they had essentially sold their voted for political capital or for simple pork, or that they were beholden to large corporations. Those supporting NAFTA portrayed the opposition as out of touch with reality, as stuck in a past world of protectionism and isolationism.

But now, with two decades of economic development, growth (and contraction), and change to look at, we can perhaps begin to assess the consequences of NAFTA.

This is no easy task, to be sure. There are a number of pieces out there already attempting an assessment, but by and large most are far from objective--from what I have seen--and attribute things directly to NAFTA in ways that cannot be fully justified (if at all). Consider this piece by Jeff Faux of the Economic Policy Institute (a liberal/progressive think tank which I've discussed previously). It opens with the following:
The Agreement created a common market for goods, services and investment capital with Canada and Mexico. And it opened the door through which American workers were shoved, unprepared, into a brutal global competition for jobs that has cut their living standards and is destroying their future.
That certainly sound pretty bad: NAFTA has lowered living standards for American workers and it is destroying their future. But what are these conclusions based on? Not a whole heck of a lot, judging from the article. Mostly, it seems Faux is simply looking at the current state of affairs and arbitrarily blaming NAFTA for the problems he sees. Faux claims that NAFTA has "directly cost the U.S. a net loss of 700,000 jobs," a number that probably comes from a 2011 EPI paper by Robert Scott, which estimates that since the passage of NAFTA, some 682,900 jobs have been "displaced." Here is the reasoning behind this figure:
The employment impacts of trade deficits are assessed using an input-output model that estimates the direct and indirect labor requirements of producing output in a given domestic industry. The model includes 202 U.S. industries, 84 of which are in the manufacturing sector. The model estimates the amount of labor (number of jobs) required to produce a given volume of exports and the labor displaced when a given volume of imports is substituted for domestic output. The net of these two numbers is the estimated number of jobs displaced by changes in the trade balance, holding all else equal.

U.S. exports to Mexico in 2010 supported 791,900 jobs, but U.S. imports displaced production that would have supported 1,474,800 jobs, as shown in the bottom half of Table 1. Therefore, the $97.2 billion U.S. trade deficit with Mexico in 2010 displaced 682,900 jobs. Since the United States had a small trade surplus in 1993 (not shown), all of those jobs were displaced between 1993 and 2010. On average, 40,200 jobs have been lost or displaced per year since NAFTA took effect.
Scott is basically imagining these numbers, using an economic model that assumes a static world, apart from the net changes to trade surpluses and deficits between the United States and Mexico. It's an impressive analysis, mathematically speaking, but is ultimately a waste of time, as the rest of the world is not static at all. Thus, Faux's conclusion--that NAFTA directly cost the U.S. these jobs--is nonsense.

This isn't to say that NAFTA has had no consequences in this regard, that it hasn't led to changes in available jobs between the U.S., Mexico, and Canada. It most certainly has, but attempts to absolutely quantify these changes are pointless, because it just can't be done. Still, it is valid--to an extent--to look at trade balances as a means of assessing NAFTA's impact. And most certainly, Mexico has benefited here, with increased industrial activity and exports for its economy as a whole.

People who accept Faux's flawed lines of thinking would have us believe, however, that all Mexican (and Canadian) gains must come at the expense of the United States. This is simply not true, as other factors that are independent of NAFTA have impacted all three economies. For instance, the much-talked about shrinking manufacturing base in the United States is not mostly--much less wholly--dependent on NAFTA at all. This piece provides some needed balance. More recently, this article from Time indicates that manufacturing in the U.S. is on the rise.

What does all of this have to do with NAFTA? Simple, at it's root NAFTA is about eliminating trade barriers--tariffs, quotas, and investment dollars--between the three nations that are a party to it. Necessarily, this means movement and reallocation of capital and other resources (including labor): this is Adam Smith's "invisible hand" at work. But rather than having such movement limited by national borders, the specifics of NAFTA allow more freedom in this regard.

The question is, does such freedom work to the benefit or detriment of the U.S. economy and U.S. workers (ditto for Mexico and Canada)? Rather than using complicated models that are fundamentally flawed to answer this question, let's first take a look at the big picture items: GDP and the unemployment rate. First GDP (in current dollars):

Next, the unemployment rate:

Now, I've limited the data sets to the years 1984 through 2007. Why? Well, the recession of 2007/2008 had a serious impact on both GDP and unemployment and there is really no way to argue--with a straight face--that the recession was related to NAFTA in any meaningful way. I extended the sets back to 1984 to better see the direction of trends. But first, one more data set, that of GDP per capita (in current dollars):

NAFTA, again, went into effect at the beginning of 1994. Looking at the above three charts, there is little evidence for claiming that NAFTA was highly detrimental to the economies of the three signatories. Indeed, exactly the opposite. True enough, the tech sector in the late 80's and early 90's--in the U.S.--was gearing up, about to explode into the "dotcom bubble," but the rest of the economy was humming along pretty well, too. There were some hiccups, no doubt (one which cost George H.W. Bush the 1992 Presidential Election) but GDP was increasing by most every measure and the unemployment rate dropped below 6% by the end of 1994, staying there until after the events of 9-11 in 2001 (and even after those events, it stayed lower than it was before NAFTA).

The GDP in Canada and Mexico trended upwards throughout this NAFTA period, as well (Mexico experienced a serious currency crisis at the end of 1994, but as the above charts indicate, things turned around in short order). Unemployment in these other two nations? Again, the charts speak for themselves. If NAFTA led to substantial and permanent job losses, one would think there'd be a little evidence in that regard.

One might argue that the supposed job losses cited by Faux are not enough to significantly move the needle on the unemployment rate. But see, according to his line of thinking, those are just the direct losses. The loss of all that buying power should have had something of a snowball effect. Yet there is no evidence of this, at least not until after 2008. Ironically, Faux claims that those who champion NAFTA and similar policies are engaged in demagoguery, even as he uses the consequences of the recent financial crisis to stoke up outrage over NAFTA. It's high comedy.

Having said all of this, I am under no illusions here. I recognize the limitations of both GDP and unemployment as metrics for this kind of analysis. But my point is not that NAFTA was responsible for either the rising GDP of the above period nor the dropping unemployment rate. Rather, I'm suggesting that there is no evidence that NAFTA has done the opposite: decreased GDP or increased the unemployment rate.

And that's because NAFTA--and other "free trade" style agreements--are not actual drivers in this regard. They don't make things happen, they merely remove impediments for the market as a whole. The consequences after that are dependent on a host of other things, all of which are themselves a product of human agency. Such changes can result in volatile periods, to be sure. But even then, over the long term the market tends to sort itself out, because the upshot of such agreements is simply the creation of larger markets.

So what can we say about the consequences of NAFTA on the U.S., Mexican, and Canadian economies twenty years after its passage? Mostly, we can say that the doom and gloom prognostications were all wrong. NAFTA has not been a disaster. The changes that have ensued have been absorbed and used, just as we--well, those of us with a clue, at any rate--thought would be the case. Regardless of the economic conditions in the moment--any moment--these changes created opportunity. And that's really the name of the game.

Cheers, all. And happy twentieth birthday, NAFTA.


  1. Being in the manufacturing industry myself, I've actually seen first-hand the job displacement that Perot predicted. Large corporations have used the agreement to shift significant numbers of manufacturing jobs to a lower cost structure in Mexico (and when they concentrate on profits and shareholders, that's ultimately what they're paid to do). Right or wrong is in the eye of the beholder, of course, but to say that there is no evidence NAFTA has impacted unemployment is a bit naive, in my opinion.

    Granted, there's no way to know whether the high-cost jobs in America that were displaced would have driven the price of goods to a point where they could not compete, so......

  2. Again though, such job losses are reflective of the shift created by an expanded market in goods, capital, and labor. The losses are not permanent in the least. The markets adjust, new opportunities are created (more than are eliminated, imo). And they didn't have a snowball kind of effect, at all. Thus, one can't--in my opinion--claim NAFTA had severe negative consequences for the economy as whole.

    Anecdotally, any economic policy can be made news for some, no doubt. But such policy shouldn't--imo--revolve around maintaining a status quo at all costs. Otherwise, we wouldn't even have free trade within the United States, itself.