Go to Brad De Long's Home
NAFTA and Jobs
Remember the "Giant Sucking Sound"?
Chris De Long, Brad De Long, and Sherman Robinson
In 1993 an odd debate took place over the North American Free Trade
Agreement ("NAFTA"). It was as if we were living in a magical-realistic
novel, a scenario constructed by Gabriel Garcia Marquez or Carlos Fuentes.
Bizarre and perverse arguments were made with straight faces, and accepted
as normal. The distinction between reality and fantasy was erased. The solid,
substantive arguments for (and against) NAFTA hit the press--and died. Few
seemed interested in the U.S.'s long-run national interest in a prosperous
Mexico, or in the rationale behind openness to world trade as a development
strategy, in NAFTA as a joint U.S.-Mexican commitment to economic liberalization,
or in the potential economic threat America's agribusiness poses to Mexico's
Those were boring.
More interestng was the giant sucking sound of American jobs heading south,
or the ability of American companies to use Mexican labor force as a mercantilist
weapon in a struggle for worldwide economic domination, or the threatened
abrogation of American sovereignty, appeared more interesting. Never mind
that they were fantasy.
But it is important--if we are to have economic policy debates on a higher
level in the future--to take a look back and see that these arguments were
fantasy. So take a look at what the passage of time has done to the most
common argument: the "giant sucking sound" of American jobs destroyed
that those like Ross Perot and Pat Choate on the right, and Ralph Nader
on the left, predicted with such confidence.
According to Perot and Choate, five millions of jobs--four percent of U.S.
employment--were "at risk" as a result of the NAFTA. Moreover,
everyone's calculations assumed a relatively prosperous Mexico in the aftermath
of NAFTA. But Mexico has been suffering a depression since the peso crisis
of December 1994. This recession has depressed Mexican wages and exchange
rates, making Mexico an even lower-wage competitor. And this recession has
greatly trimmed Mexican demand for U.S. made consumer goods. So--if there
was any reality in the giant sucking sound--one might well conclude that
many more than five million American jobs should have been lost to Mexico
in the two and a half years since the implementation of NAFTA.
The year before NAFTA the U.S. unemployment rate averaged close to seven
percent. Has the U.S. unemployment climbed since then--to eleven percent
or more? No, the U.S. unemployment rate is now at five and a half percent.
The fall in the unemployment rate has not been due to any rise in the number
of discouraged workers not counted as unemployed: more than nine million
net new jobs have been created since the start of the Clinton administration.
The truth of the matter is that the level of employment in the U.S. is much,
much more the result of decisions made by the Federal Reserve's Open Market
Committee in Washington, rather than changes in trade with Mexico. There
never was any reason to think that NAFTA would have any significant effect
on aggregate employment in the United States--and it looks those who claimed
that it would were much more knaves than fools.
Increased imports from Mexico have displaced some American workers. There
have been applications to the Labor Department for the transitional training
and job search funds that were established to assist U.S. workers adversely
affected by NAFTA. Through the end of 1995, about 35 such petitions were
being filed a month, and about 2,000 workers each month have been added
to those who have claimed aid under these programs.
In the U.S. economy today, nearly two million workers lose their jobs every
month. Thus, NAFTA-related churning of the employment market caused perhaps
one out of every thousand job separations in 1995. The U.S. would almost
surely be a better place if it had more stable employment, and if individual
employers thus saw their strong collective interest in training workers.
But NAFTA's contribution to the churning of the U.S. job market is orders
of magnitude less than the statistical error in the monthly employment reports.
Maybe under NAFTA Mexico has been stealing not net jobs but good jobs? Maybe
U.S. workers have been being fired from high-paid manufacturing jobs as
a result of Mexican imports, and have been re-hired for low-paid service
There is no sign of any such effect. The inflation-adjusted wages of America's
workers have (except near the top of the income distribution) stagnated
for more than two decades, but not because the U.S. has been losing "good
jobs." The sources of net job creation in the U.S. economy during the
Clinton business cycle recovery have been in relatively high-paid occupations
and in relatively high-paid industries: median wages have stagnated because
the inflation-adjusted pay rates for already-existing jobs have fallen,
not because the new jobs created have been low-paying jobs.
Of course, the benefits of NAFTA to the U.S. have also been small. Mexico's
economy is comparable in size to that of metropolitan Los Angeles. Perhaps
production in capital goods-making industries is marginally higher, and
perhaps a few tens or hundreds of thousands of additional U.S. workers will
be employed making capital goods for export to Mexico than if NAFTA had
not been implemented; but if so then domestic interest rates are also marginally
higher, and fewer U.S. citizens are making capital goods for domestic use.
On net it is probably a small plus, but a plus smaller than the data errors
and revisions in even one quarter's worth of GDP.
NAFTA could never have had a major effect on the U.S. economy. So why did
so many people make wild assertions that it would? We think that they did
so to advance other political goals: a Ross Perot appears to hope to benefit
politically by telling Americans that the "establishment" is out
to get them; a Pat Choate appears to want to boost the power of his faction
of the trade warrior caste by frightening the public with the Mexican bogeyman
and then using that fear to power a "hawkish" trade stance toward
Japan. A Ralph Nader who has seen the good arguments for a more liberal
turn in American domestic policy crash or fizzle over the past fifteen years
is eager to embrace the bad arguments.
During the debate over the ratification of NAFTA there was a whiff of Europe's
politics in the 1920s and the 1930s--when right and left joined together
with glee to assault the center, both sure that they would benefit when
the center collapsed. In the short run the fascists benefitted. In the long
run no one benefitted from the collapse of the center, for interwar Europe
dissolved into chaos and World War II.
This piece is adapted from an article appearing in the May-June 1996
issue of Foreign Affairs.
Go to Brad De Long's Home
Associate Professor of Economics Brad De
Long, 601 Evans
University of California at Berkeley; Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax