As American political footballs go, trade deals rank second probably only to the Affordable Care Act.
The Trans-Pacific Partnership was thrown for an enormous loss during the presidential campaign even though the share of Americans who understood the issues even partially probably reached the hundredths of a percent. But the grandfather of controversial deals remains NAFTA, the North American Free Trade Agreement, which went into effect in January 1994 and remains in President Trump's crosshairs. (He's already moved to withdraw the U.S. from the TPP.)
Kevin Drum at Mother Jones tried Sunday to mediate a debate over NAFTA between economists Brad DeLong of UC Berkeley and Dani Rodrik of the Harvard Kennedy School. He takes DeLong's point that the overall economic impact of NAFTA has been minuscule as a share of gross domestic product. "NAFTA's impact on the U.S. — whether good or bad — is inevitably tiny," Drum concludes, and he's right.
But there's more to the debate. In fact, NAFTA is a perfect example of the difference between economics and politics. Put simply, DeLong is right that NAFTA's aggregate effects, especially on U.S. manufacturing employment, have been small to the point of insignificance (and about some other aspects as well). Rodrik is right in observing that, even so, its impact on certain discrete communities has been "severe." But Rodrik only touches on the fact that those severe but limited impacts were ruthlessly exploited by Trump to distort the overall NAFTA story.
That's important, because Trump's approach to trade is all about those limited impacts. Those impacts are far beyond his ability to fix, even if he really wanted to, which is questionable.
Let's take a look at what's really happened.
DeLong reminds us that NAFTA instantly eliminated Mexican tariffs on half of the industrial goods imported from the U.S. and eliminated other Mexican protectionist rules; phased out most tariffs among Mexico, Canada and the U.S.; and gave Mexico tariff-free access to the immense U.S. consumer market, which spurred investment in Mexican factories.
Did that kill American manufacturing? Delong says no, "period." He observes that American manufacturing employment has been in a decline dating back to the 1950s for several reasons. The biggest factors, which brought the manufacturing share of nonfarm employment from 30% in 1950 to 12% by the turn of the century, were "rapid growth in manufacturing productivity and limited demand." Bad economic policies starting in the Reagan era pared another three percentage points, and the rise of China as a global competitor accounted for about another third of a percentage point. NAFTA took another tenth of a percentage point off the manufacturing share of employment.
"And yet," DeLong writes, "the American political system right now is blaming all, 100%, every piece of that decline from 30% to 8.6% and every problem that can be laid at its door on brown people from Mexico."
Rodrik accepts other economists' finding that NAFTA has been good for the U.S., but only in a tiny way. The deal contributed a gain of 0.08% to net economic welfare, according to Lorenzo Caliendo of Yale and Fernando Parro of the Federal Reserve. Mexico benefited the most, with a gain of 1.31% mostly from increased trade, and Canada lost a very marginal 0.06%, mostly from higher prices for imports.
But Rodrik points to the severe impacts on the pockets of blue-collar American workers whose employers had coasted on tariff protections that were now removed: makers of "apparel, textiles, footwear or structural clay products like brick and tile." High school dropouts in those industries "experienced reductions in wage growth by as much as 17 percentage points relative to wage growth in unaffected industries," Rodrik notes. Those workers, and residents of their communities who were indirectly dependent on their wages — waitresses, barbers, auto repair workers, etc. — felt the pain.
They were small in number — DeLong places the "high-end credible estimate" of manufacturing jobs lost directly to NAFTA at 200,000, or 0.14 percentage points of nonfarm employment and 1/112 of the total manufacturing job share loss since 1971. He argues this has been compensated for by higher real incomes in the U.S. derived from cheaper imported goods, along with the more inchoate benefits of having a more prosperous and therefore more stable Mexico on our southern border.
The pain of the disenfranchised workers is not only real, but exploitable by unscrupulous politicians with no apparent policies to help them. Republicans and Democrats alike deserve blame for leaving these communities behind. Government job-training initiatives long have been more talk than action, and the economic collapse that was fostered by years, even decades, of tax-cut and deficit-spending policies favoring the 1% hamstrung any efforts to assist those workers.
What can be done now? These manufacturing jobs are almost certainly not coming back, and abrogating NAFTA won't bring them back. The handful of factory relocations trumpeted by Trump is deceptive; they're not employers' responses to economic factors or Trump employment policies, whatever they are. Any effort by the White House to stem job losses by imposing tariffs on imports from Mexico or anywhere else will mostly cost U.S. consumers.
To the extent Trump administration trade policy is focused on trade deals such as NAFTA or TPP, it's likely to fail economically. The only question is its impact politically.
"Is Trump deluded on NAFTA's overall impact on manufacturing jobs?" Rodrik asks. "Absolutely, yes. Was he able to capitalize on the very real losses that this and other trade agreements produced in certain parts of the country in a way that Democrats were unable to? Again, yes."
DeLong is more withering. "Politicians are taking claims that have a very tenuous connection to economic reality — claims that feel true — and running with them, sometimes out of ignorance, sometimes because of cynical calculation," he writes. On trade as on so many other issues, American politicians have their eyes on the wrong ball.