NAFTA: Economic Effects on the United States After Eight Years
Report for Congress
NAFTA: Economic Effects on the United States
After Eight Years
Updated August 20, 2002
Consultant in International Trade and Finance
Foreign Affairs, Defense, and Trade Division
Congressional Research Service ˜ The Library of Congress
NAFTA: Economic Effects on the United States After
The North American Free Trade Agreement (NAFTA) among the United States,
Canada and Mexico went into effect on January 1, 1994. It is the first trade
agreement the United States has entered into with a geographically-close developing
country and has raised concerns about its economic effect, particularly on U.S.
communities and workers.
Since the mid-1980s, when Mexico began reducing trade restrictions, the U.S.
and Mexican economies have become more highly integrated. This is evidenced by
the rapid growth in U.S. merchandise trade with Mexico, which is now 12% of all
U.S. trade (up from 8% in 1993 and 6% in 1988) and especially by the growth of
intra-industry and intra-firm trade. This growing integration has, by reducing costs,
made the U.S. economy (and particularly the Mexican economy) more productive
and globally competitive. Greater integration of the U.S. and Mexican economies
probably would have occurred without the NAFTA, but the NAFTA, by improving
business confidence in Mexico, may have accelerated the process.
The U.S. foreign direct investment position in Mexico grew more rapidly before
U.S. foreign direct investment in Mexico is still relatively small, accounting for 2.8%
of all U.S. foreign direct investment abroad in 2000. U.S. flows of direct investment
to Mexico of $3.5 billion in 2000 are also very small compared with U.S. domestic
investment in U.S. plant and equipment of $1,362 billion.
It is estimated that NAFTA caused job dislocation for about 415,000 workers
between January 1, 1994 and December 31, 2001, about 34% of whom were in the
textile and apparel industry and 5% in the automotive industry. The number of U.S.
workers dislocated by NAFTA over eight years is less than 1% of the 134.3 million
U.S. workers employed in 2001.
Nevertheless, a study published in August 2001 by the U.S. General Accounting
Office of six communities who were hard-hit by job dislocation from NAFTA
illustrates the need for improving the skills and job opportunities of dislocated
workers. Generally, dislocated workers had less education than the U.S. workforce
as a whole and many had limited English language skills. These case studies suggest
that, in the past, trade adjustment assistance programs did not meet the needs of the
workers in these communities. The Trade Act of 2002 includes provisions that may
improve the effectiveness of trade adjustment assistance.
Overall, the NAFTA has had a relatively small effect on the U.S. economy.
Future free trade agreements between the United States and other countries, which
are not major U.S. trading partners and are geographically further away, will likely
have an even smaller effect on the U.S. economy.
This report will not be updated.
In troduction ......................................................1
Analysis of Merchandise Trade Trends.................................6
Trade Creation, Trade Diversion and The Dynamic Effects of Trade......7
Integration of the U.S. and Mexican Economies......................8
U.S. Merchandise Trade Balances with Canada and Mexico...........10
Trade in Services.................................................11
Foreign Direct Investment..........................................12
Job Dislocation Relative to Total U.S. Employment..................13
Job Dislocation in Selected U.S. Communities......................14
Effect on Job Security.........................................16
List of Figures
Figure 1. Gross Domestic Product 2001................................4
Figure 2. U.S. Merchandise Exports as a Percent of U.S. GDP, 2001.........5
Figure 3. U.S. Merchandise Trade with Canada and Mexico................6
Figure 4. U.S. Merchandise Trade Balances with Canada and Mexico.......10
Figure 5. U.S. Private Services Trade with Canada and Mexico............11
Figure 6. U.S. Direct Investment Abroad..............................12
This report was prepared under the general supervision of Raymond J. Ahearn,
Specialist in Trade Relations, CRS.
NAFTA: Economic Effects on the United
States After Eight Years
The North American Free Trade Agreement (NAFTA) among the United States,
Canada and Mexico went into effect on January 1, 1994 after a well-publicized
debate by the public and the Congress. The primary fear of NAFTA’s opponents was
that the United States would lose jobs as corporations shifted production to Mexico
where wage rates are substantially less and labor standards are not always enforced.
Ross Perot coined the term “the giant sucking sound” of U.S. jobs going to Mexico,
a phrase that resonated with the American public. At the same time, the Clinton
Administration focused on the large number of jobs that might be created by NAFTA
which, it was claimed, would expand U.S. exports to Mexico.
There are two basic reasons for analyzing the economic effects of NAFTA at
this time. First, many people are still apprehensive and concerned about NAFTA.
Two recent polls tell the story. One question in a 1999 poll was “Do you think the
NAFTA has been good or bad for the United States?”1 The responses were: 44%
good, 30% bad, 7% neither, and 19% don’t know. In a poll taken in 2001, one
question was “Do you think America should continue the NAFTA agreement, pull2
out of NAFTA, or should the NAFTA be continued with changes?” The responses
were: 28% continue NAFTA, 14% pull out of NAFTA, 37% change NAFTA, and
Second, since proposals to expand and deepen NAFTA are being discussed, the
lessons learned from NAFTA may be helpful. Negotiations are underway towards
a Free Trade Area of the Americas (FTAA) and a U.S. trade agreement with Chile.
The Bush Administration announced its intention to explore a U.S.-Central American
Free Trade Agreement among the United States and Costa Rica, El Salvador,
Guatemala, Honduras and Nicaragua in January 2002. Reportedly, the United States3
and the five Central American countries may begin negotiations by the end of 2002.
Mexico’s President Vincente Fox has proposed a “deepening” of NAFTA to
become a common market, which could include a regional fund to stimulate
1Gallop/CNN/USA Today Poll conducted in November 1999. Website:
2Poll conducted for the Association of Women in International Trade in October 2001.
3United States, Central American Nations May Launch Trade Talks by End of 2002.
International Trade Reporter. June 20, 2002, p. 1099.
Mexican economic growth.4 Others have suggested measures such as a common
external tariff, coordination of regulatory systems, and freer movement of citizens
among the three NAFTA partners. One Canadian analyst suggested that deeper
integration is necessary to prevent border closings for security reasons (such as that
in the immediate aftermath of September 11, 2001) which imposed an economic
hardship on Canada.5
The purpose of this report is to examine the available data to determine
NAFTA’s economic effect on the United States. The two questions addressed are:
How has NAFTA affected the U.S. economy? What is the magnitude of the effect?
The data studied include trade (both merchandise and services) flows, foreign direct
investment, and the number of U.S. workers certified as having lost (or been
threatened with the loss of) their jobs as a result of NAFTA (a proxy for job
dislocation). Selected case studies of U.S. communities adversely affected by plant
relocations to Mexico are summarized.
This report focuses more on the U.S. trade and investment relationship with
Mexico than with Canada. The NAFTA is the first major trade agreement the United
States has had with a geographically-close developing country where per capita
income is much lower than in the United States. In 2000, Mexico’s per capita GDP
was about $6,000, compared to $35,000 in the United States. Issues of concern in
U.S. trade with Mexico, such as lower Mexican wages, are not relevant to U.S. trade
with Canada, where the two economies are quite similar.
The main findings of this report are:
Since the late 1980s, when Mexico began reducing trade restrictions, the U.S.
and Mexican economies have become much more highly integrated. This is
evidenced by the rapid growth in U.S. trade with Mexico and especially by the
growth of intra-industry and intra-firm trade. This growing integration has, by
reducing costs, made the U.S. economy (and particularly the Mexican
economy) more productive and globally competitive. Greater integration of the
U.S. and Mexican economies in fact was happening before the NAFTA, but the
NAFTA, by improving business confidence in Mexico, may have accelerated
Between January 1, 1994 and December 31, 2001, about 415,000 U.S. workers
have suffered job dislocation as a result of U.S.-Mexican economic integration.
Although this number is very small relative to the size of the U.S. workforce,
it may have imposed a significant economic hardship on those workers and their
families who have lost jobs.
4Pastor, Robert A. Toward a North American Community: Lessons from the Old World for
the New. Institute for International Economics, 2001, p. 190.
5Dobson, Wendy. Shaping the Future of the North American Space: A Framework for
Action. C.D. Howe Institute Commentary, No. 162, April 2002, p. 1.
Both the benefits of the NAFTA (increased trade and investment and enhanced
global competitiveness) and the costs (job dislocations) are very small relative
to the size of the U.S. economy.
Economic Benefits and Costs of Free TradeAn implicit goal of the
NAFTA was to increase
Generally, when trade restrictions such as tariffseconomic integration among the
and quotas are reduced in a free trade agreement, thethree partner countries (by
economies of all participating countries benefit overincreasing trilateral trade and
time. Freer trade allows each country to specialize ininvestment) which, in turn, was
producing those goods in which it has a comparative
advantage. Each economy may also benefit from lowerexpected to raise each
production costs per unit as it produces for a largercountry’s productivity and
market or by trade’s stimulation of technologicalglobal competitiveness. In
developments. Consumers can choose from a wideranalyzing the NAFTA’s effect,
variety of products, and greater competition leads tohowever, it is important to
higher quality goods at lower prices.
recognize that economic
The costs of liberalizing trade generally fall to theintegration between the United
workers and industries whose products are no longerStates and its NAFTA partners
competitive. They are forced to adjust, sometimes veryhad been underway for some
painfully, as domestic plants close or relocate to other
countries. time before the NAFTA was
The benefits of free trade generally exceed theintegration of production
costs. The benefits, however, are widely distributed overresulted from market forces in
the entire population, and, as a result, each person’s gainwhich businesses, to be
is quite small. In contrast, the costs are borne by
relatively few people, but may be huge for those workerscompetitive, had to devise cost-
and their families who have lost jobs.saving strategies (such as
shifting part of their production
abroad), and by government
policies and bilateral agreements that reduced trade restrictions.
The economies of the United States and Canada, in particular, have been highly
integrated for a long time. The Automotive Agreement of 1965 between Canada and
the United States eliminated tariffs on shipments of autos and auto parts between the
two countries, which allowed cost-savings from cross-border production of
automobiles. The U.S.-Canada Free Trade Agreement (FTA), which went into effect
on January 1, 1989, phased out bilateral tariffs over a 10-year period. Consequently,
when the NAFTA superceded the FTA in 1994, most U.S.-Canadian tariffs were
already eliminated or very low.
In the mid-1980s, Mexico dramatically reversed its trade policies by unilaterally
liberalizing many tariff and nontariff barriers. The extent of the liberalization was
actually greater than the reductions in trade barriers in the NAFTA.6 At that time,
6Hinojosa-Ojeda et. al. The U.S. Employment Impacts of North American Integration After
NAFTA: A Partial Equilibrium Approach. School of Public Policy and Social Research.
Mexico also liberalized investment restrictions. Mexico’s accession to the General
Agreement on Tariffs and Trade (GATT, the forerunner of the World Trade
Organization, or WTO) in 1986 locked in the reductions in trade and investment
barriers. Cross-border production took place through Mexico’s maquiladora
program (established in the 1960s), in which foreign-owned businesses could import
intermediate goods tariff-free as long as the final product was exported. (The
original maquiladora program ceased to exist in 2001 when NAFTA rules applied to
Figure 1. Gross Domestic ProductThe NAFTA is a wide-
the trade reforms of the previous
decade in Mexico.7 By making it
less likely that the Mexican
government can reverse its
previous trade and investment
liberalizations, many argue that
business confidence in Mexico
There are several reasons why
NAFTA’s effect on the U.S.
economy would be expected to be
small. First, from an economic
perspective, the relationship
among the three countries is very
asymmetric. As shown in figure 1, Mexico’s GDP in 2001 of $618 billion was only
6% that of the United States of $10,208 billion. Even Canada’s 2001 GDP of $682
billion was only 7% of U.S. GDP.
UCLA. January 2000. p. 38.
7The NAFTA includes chapters on investment, services, intellectual property protection,
government procurement, agriculture, energy, and dispute settlement. In addition, side
agreements on the environment and labor standards facilitate cooperation on these issues by
the United States, Canada and Mexico.
Figure 2. U.S. Merchandise ExportsU.S. trade with Mexico
as a Percent of U.S. GDP, 2001and Canada provides another
perspective. Even though
Mexico is the second largest
U.S. export market, U.S.
exports to Mexico were only
about 1% of U.S. GDP in 2001
(see figure 2). U.S. exports to
Canada (the largest U.S. trading
partner) were 1.6% of U.S.
GDP. U.S. exports to Mexico
and Canada have increased
substantially since 1994 (see p.
6), but they are still relatively
small in relation to U.S. GDP.
Second, tariffs were
already low or nonexistent between the three countries when the NAFTA went into
effect and have continued to decline. Under the Canada-U.S. Free Trade Agreement,
which was superceded by the NAFTA, some tariffs were eliminated on January 1,
1989, and the rest phased out over 5 or 10 years. By January 1, 1999, virtually all
U.S.-Canadian tariffs had been eliminated.
When the NAFTA went into force in 1994, half of U.S. exports to Mexico
became duty-free, while the remaining half were to be phased out over 5, 10, or 15
years. By January 1, 2001, the average Mexican tariff on U.S. products had declined
from 10% to 1.27%, while the average U.S. tariff on Mexican products had declined
from 4% to .35%.8
8Office of the United States Trade Representative. 2001 Trade Policy Agenda and 2000
Annual Report of the President of the United States on the Trade Agreements Program.
Washington, 2001, p. 115.
Analysis of Merchandise Trade Trends
Figure 3 shows the trends in U.S. exports to and imports from Mexico and
Canada since 1988. Generally, U.S. trade with Mexico has been increasing faster
than U.S. trade with Canada since 1988, but the rate of increase accelerated from
1994, the year NAFTA
Figure 3. U.S. Merchandise Trade withwent into effect, to 2001.
Canada and MexicoMore specifically, from
1988 to 1993, U.S. trade
increased by 86% with
Mexico, 38% with
Canada and 37% with the
world. Even in the pre-
NAFTA period, then,
U.S. trade with Mexico
was increasing faster than
with Canada or with the
world. From 1993 to
2001, U.S. trade
increased 185% with
Mexico, 79% with
Canada and 79% with the
world. Again, the
increase in trade with
Mexico far exceeded that
with Canada and with the world.
As a result, U.S. trade with Mexico accounted for 12% of all U.S. trade in 2001,
up from 8% in 1993 and 6% in 1988. By contrast, U.S. trade with Canada remained
at about 20% of all U.S. trade since 1988.
It should be emphasized that the rapid growth in U.S. merchandise trade with
Mexico occurred since NAFTA, not necessarily because of NAFTA. Other factors,
especially the high economic growth period of the 1990s followed by the global
economic downturn in 1991, the peso-dollar exchange rate, and the growing amount
of trade in intermediate goods (discussed more fully later), contributed to changing
trade flows. Although difficult to quantify, several recent studies have used
econometric methods in an attempt to isolate the effects of NAFTA on U.S.-Mexican
trade. Most studies found that NAFTA had a small, but positive effect on trade9
growth. Generally, in such studies, changes in GDP had a much greater effect on
U.S. exports to and imports from Mexico than did NAFTA.
The 1995 peso crisis in Mexico had a large effect on trade in that year. The peso
depreciated 47% against the U.S. dollar in 1995, the year after NAFTA went into
effect, making U.S. imports from Mexico much cheaper and U.S. exports to Mexico
9Ferrantino, Michael J., Evidence of Trade, Income, and Employment Effects of NAFTA.
International Trade Commission. Industry Trade and Technology Review. December 2001,
more expensive (to Mexicans). At the same time, the peso crisis caused Mexico’s
real (inflation-adjusted) GDP to decline by 6% in 1995, further reducing Mexican
imports from the United States. Mexico’s economy rebounded quickly, growing 5%
in 1996 and almost 7% in 1997. Strong economic activity in all three countries
contributed to the substantial growth of trade in the late 1990s.
Trade Creation, Trade Diversion and The Dynamic Effects of
The growing share of Mexico in U.S. trade raises the question of whether
NAFTA has led to trade creation or trade diversion. Trade creation occurs when the
removal of trade restrictions in a regional trade agreement increases the flow of trade
between the trading partners because it becomes more efficient to trade some products
than produce them domestically. Economists believe that trade creation improves
economic welfare of both the trading partners and the world. Trade diversion, on the
other hand, occurs when a country imports from the trade agreement partner because
the tariff removal has made the product cheaper, even though a firm in a non-trade-
agreement country can produce the product more efficiently. In other words, by
reducing tariffs among a few trading partners and not others, trade could be diverted
from the non-participating country, which may be the more efficient producer, to the
trade agreement partner. Economists argue that trade diversion worsens economic
It is not possible to measure the magnitude of trade creation or trade diversion
with any degree of precision. Several studies, however, indicate that trade creation10
exceeds trade diversion among the NAFTA partners. The majority of U.S industries
whose exports to Mexico have grown most rapidly are those in which the United
States has a comparative advantage, such as capital goods. Similarly, U.S. import
growth has been fastest in Mexican labor intensive industries where Mexico has a
Preliminary data from one study suggests that trade diversion is occurring with
respect to U.S. imports from Mexico versus the rest of the world.11 A significant
amount of trade diversion is most likely occurring in the textiles and apparel sector,
partly because the rules of origin established in the NAFTA for this sector tend to12
favor trade within the NAFTA countries.
Fears of trade diversion may have encouraged some non-NAFTA countries to
negotiate trade agreements with Mexico. For example, the European Union
10Mutti, John. NAFTA: the Economic Consequences for Mexico and the United States.
Economic Strategy Institute, February 2001, Website: [econstrat.org/naftacons.htm], p. 2.
11Hinojosa-Ojeda, Op.Cit., p. 68.
12Rules of origin in a free trade agreement specify the proportion of the content of the
product that must originate within the partner countries in order to qualify for the benefits
of the agreement.
negotiated a free trade agreement with Mexico after its share of Mexican imports fell
from 15% in 1990 to 12% in 1993 and 9% in 1999. 13
In addition to the short-term static effects of trade creation and trade diversion,
the NAFTA may also generate positive long-term dynamic effects. It is likely that the
long-term dynamic benefits are greater in magnitude than the trade-creating or trade-
diverting effects of NAFTA. Achieving long-term dynamic effects was probably the
most important economic consideration when the United States, Canada and Mexico
agreed to negotiate the NAFTA.
There are three main long-term dynamic effects. First, as producers face
increased competition, they are likely to become more efficient. In fact, Mexico’s
goal of making its economy more efficient was one of the incentives for negotiating
a NAFTA. Second, being able to sell to a larger market enables firms in NAFTA
countries to take advantage of economies of scale, in which the per unit cost of
production falls as output increases. Economies of scale might arise from one
country’s being able to concentrate on a particular size or model of a product, and
were important considerations in Canada’s negotiating its free trade agreement with
the United States in 1988. Finally, the NAFTA (and anticipation of the NAFTA)
attracted foreign investment from non-NAFTA countries whose firms wanted to
produce within the NAFTA area to take advantage of the area’s lower trade barriers.
Foreign direct investment from non-NAFTA countries to Mexico increased during the
years immediately preceding NAFTA.
Integration of the U.S. and Mexican Economies
Even before NAFTA was negotiated, U.S. companies were in the process of
internationalizing production to become more competitive globally. In many cases,
this meant locating part of their production processes in Mexico to take advantage of
lower wages, which were only partly offset by lower productivity. As discussed
earlier, such shifts were facilitated by Mexico’s unilateral reduction in trade barriers
and the NAFTA, which made it less likely that Mexico could revert to higher tariffs
and nontariff barriers.
One way of estimating the extent to which production has become more
integrated is to examine changes in intra-industry trade between the United States and
Mexico. As cross-border production increased, it would be expected that exports and
imports of similar products (intra-industry trade) would also increase. According to
an OECD study, intra-industry manufacturing trade in 1996-2000 accounted for 73%
of manufacturing trade in Mexico, up from 62% in 1988-1991.14 The OECD study
credits the NAFTA, which strengthened Mexico’s trading links with the United States,
as a contributor to Mexico’s growing intra-industry trade.
13Mutti, Op. Cit., p. 2.
14Organization for Economic Cooperation and Development. Intra-Industry and Intra-Firm
Trade and the Internationalisation of Production. OECD Economic Outlook 71. June 2002,
Another finding of the OECD study was that U.S. intra-firm trade with Mexico
accounted for 66% of all U.S. goods traded with Mexico in 1999, up 3% since 1992.15
U.S. intra-firm trade is particularly concentrated in transportation equipment,
computer and electronic products, and machinery and chemicals. U.S. Department of
Commerce data show that, from 1989 to 1999, U.S. majority-owned affiliates’ exports
to Mexico (including exports to affiliates and non-affiliates) nearly doubled to 9% of
all U.S. exports by majority-owned affiliates to the world.16 In contrast, Canada’s
share, at 21% of all U.S. exports, remained about the same. Mexico’s increased
share, according to the Department of Commerce, reflected the growing presence of
foreign affiliates of U.S. parent firms in Mexico, as well as the higher Mexican tariffs
prior to NAFTA.
Since a high proportion of U.S-Mexican trade is intra-industry or intra-firm trade,
total U.S. exports to Mexico can be expected to grow at about the same rate as U.S.
imports from Mexico. Looking again at figure 3, it can be seen that U.S. exports to
and imports from Mexico are fairly well correlated, as are U.S. exports to and imports
from Canada, where the economies have been integrated for quite a while. One of the
implications, then, of increased integration of production is that when the U.S.
economy slows, not only will U.S. imports from Mexico and Canada decline, but U.S.
exports of components to Mexico and Canada will also decrease. This may account
for some of the decline in both U.S. exports to and imports from Mexico and Canada
in 2001 shown in figure 3. The decline in growth of U.S. GDP to 1.1% and the
negative growth of 0.6% for Mexican GDP in 2001 also contributed to the decline in
U.S. exports and imports.
The five U.S. industry sectors with the greatest volume of trade with Mexico and
Canada are automotive, chemicals and allied products, computer equipment, textiles
and apparel, and microelectronics.17 Not surprisingly, these sectors are also heavily
involved in intra-industry trade. For Mexico alone, between 1993 and 2001, U.S.
trade (exports plus imports) in all five sectors grew faster than U.S. trade with the
world.18 Bilateral automotive trade, by far the largest of the five sectors in absolute
terms, grew 254%, compared with 72% for the world. In computer equipment, the
growth was especially large; U.S-Mexico trade grew 553% while U.S. trade with the
world grew 80%. In contrast, U.S. trade with Canada grew slower than U.S. trade
with the world in all of the sectors except textiles and apparel.
15Ibid., p. 165.
16U.S. Department of Commerce. Operations of U.S. Multinational Companies. Survey of
Current Business. March 2002, p. 37.
17For a more detailed analysis, see CRS Report RL31386, Industry Trade Effects Related to
NAFTA, by M. Angeles Villarreal.
18Computed by CRS based on data in Ibid., p. 7 and 9.
U.S. Merchandise Trade Balances with Canada and Mexico
Figure 4. U.S. Merchandise TradeAs figure 4 indicates,
Balances with Canada and Mexicothe United States had a
merchandise trade deficit
with both Mexico and
Canada in almost every
year since 1988. The
trade deficit grew
substantially in 1995,
probably in response to
the Mexican peso crisis.
Since 1999, the trade
deficit has been especially
large, and, surprisingly,
the deficit with Canada
has been larger than that
with Mexico. In 2001,
the U.S. trade deficit was
$53 billion with Canada
and $30 billion with
Mexico. The 14% appreciation of the U.S. dollar against the Canadian dollar since
1996 likely contributed to the worsening U.S. trade deficit with Canada. As the dollar
appreciates, U.S. imports increase because the price in dollar declines, and U.S.
exports increase as the price in foreign currencies increases.
For perspective, it is interesting to compare the U.S. trade deficit with Canada
and Mexico with the total U.S trade deficit. In 2001, the U.S. trade deficit with
Canada and Mexico was $83 billion, about 20% of the U.S. trade deficit of $411
billion with the world. Moreover, the U.S. trade deficit with the world is growing
faster than the U.S. trade deficit with Canada and Mexico. Since 1996, the U.S. trade
deficit with Canada and Mexico has increased 112 %, while the U.S. trade deficit with
the world has increased 142%. This suggests that factors other than NAFTA,
especially a strong U.S. economy, have been the driving forces behind the growing
U.S. trade deficit with Canada and Mexico.
Trade in Services
U.S. private services trade with Mexico and Canada is small by comparison with19
merchandise trade. For example, in 2000 (the latest year for which services data are
available), U.S private
Figure 5. U.S. Private Services Trade withservices exports to
Canada and MexicoMexico were $14 billion,
compared with U.S.
merchandise exports to20
Mexico of $101 billion.
U.S. private services
imports from Mexico
were $11 billion, while
U.S. merchandise imports
from Mexico were $136
Private services trade
fluctuated quite a bit over
the 1988-2000 period.
There was a noticeable
drop in U.S. services
exports to Mexico in
1995, probably reflecting
the peso crises of that year. For example, the depreciation of the peso and the decline
in Mexican GDP in 1995 likely reduced the number of Mexicans traveling to the
border to purchase U.S. goods.
Generally, U.S. services trade (exports plus imports) with Mexico increased more
in the pre-NAFTA period (from 1988 to 1993) than since 1993. U.S. services trade
with Mexico increased 79% from 1988 to 1993, and 40% from 1993-2000. The
pattern for U.S. services trade with Canada and the world was just the opposite – the
growth rate was greater for 1993 to 2000 than for 1988 to 1993.
Unlike merchandise trade, the United States had a surplus in services trade with
Mexico and Canada in almost all years. In 2000, the U.S. services trade surplus was
$3 billion with Mexico and $7 billion with Canada. The $10 billion surplus is small,
however, compared with the $77 billion merchandise trade deficit with Canada and
Mexico in 2000.
19Private services trade include exports and imports of travel (purchases of goods and
services by U.S. persons traveling abroad and foreigners traveling in the United States),
passenger fares, other transportation (related to transport of goods), royalties and fees and
other (education; financial services; insurance; telecommunications; business, professional,
and technical services; and other).
20The services data in this section were taken from (or calculated by CRS based on data in)
U.S. Department of Commerce. U.S. International Services: Cross-Border Trade in 2000
and Sales Through Affiliates in 1999). Survey of Current Business. November 2001.
Not surprisingly, given their geographic proximity and the significant amount of
cross-border tourism and business travel, the largest category of services trade with
the NAFTA partners is travel, passenger fares and other transportation. In 2000, this
category represented close to 50% of U.S. services exports to Mexico and Canada.
For U.S. services imports, the amounts are much higher – 66% for Canada and 81%
for Mexico. The next highest category was “other services.” U.S. exports of other
services exceeded U.S. imports of other services by a considerable amount to both
Mexico and Canada in 2000.
U.S. services are also provided in Mexico through affiliates of U.S. multinational
firms. Although the amounts are relatively small, such sales have been growing
rapidly. For example, in 1999 (the latest data available), sales to U.S. affiliates in
Mexico were $4.8 billion, up from $1.3 billion in 1993, a growth of 269%. This
growth rate was much greater than for U.S. services sales through affiliates in Canada
(84%) and in the world (137%).
Foreign Direct Investment
The U.S. foreign direct investment (FDI) position (the cumulative amount, or the
stock) in Mexico grew rapidly in21
Figure 6. U.S. Direct Investmentthe pre-NAFTA period. From
Abroad1988 to 1993, the U.S. FDI
position in Mexico grew 166%,
compared with 12% in Canada22
and 68% in the world. Growth
continued at a somewhat lower,
but still high, rate after NAFTA
went into effect. From 1993 to
the U.S. FDI position in Mexico
grew 133%. In contrast, from
1993 to 2000, U.S. direct
investment in Canada and the
world grew 81% and 121%
respectively, greater than during
the period preceding NAFTA.
Despite fairly rapid growth rates, especially in the pre-NAFTA period, the
amount of U.S. direct investment in Mexico is currently small relative to total U.S.
direct investment abroad. In 2000, the U.S. direct investment position in Mexico was
$35.4 billion, only 2.8% of U.S. FDI in all countries of $1,244.7 billion, up from
21Direct investment is defined as the ownership or control, directly or indirectly, by one
person (individual, partnership, etc.) of 10% or more of the voting securities of an
incorporated business enterprise or an equivalent interest in an unincorporated business
enterprise. Data in this section are for FDI at historical cost.
22Computed by CRS from data in U.S. Department of Commerce. Survey of Current
Business. Various Issues.
1.7% in 1988 and 2.7% in 1993. Even the U.S. direct investment position in Canada
in 2000 was only 10.2% of the total U.S. FDI, down from 18.7% in 1988.
About 58% of all U.S. FDI in Mexico was in the manufacturing sector in 2000.
Food products, transportation equipment, and chemicals account for almost 70% of
U.S. manufacturing investment in Mexico.
Examining U.S. direct investment flows to Mexico shows that they are very small
relative to U.S. domestic investment. The amount of new U.S. FDI in Mexico was
$4.6 billion in 1998, $5.1 billion in 1999, and $3.5 billion in 2000. By comparison,
U.S. domestic investment in U.S. plant and equipment was $1,362.2 billion in 2000.23
The argument that NAFTA destroys U.S. jobs resonates strongly with the public.
It is indeed true that NAFTA has resulted in the elimination of some U.S. jobs,
especially in the textile and apparel and food products sectors. At the same time,
NAFTA has created new jobs as exports to Mexico in the capital goods and “new
economy” industries increased. On balance, it is likely that NAFTA’s net effect on
U.S. employment is close to zero. However, NAFTA has probably affected the
composition of jobs in the U.S. economy, as the demand for unskilled workers
declined while the demand for skilled workers increased.
Unfortunately, there are no available data bases to measure or estimate the
number (and types) of jobs created by NAFTA. As will be discussed below, it is
possible to roughly estimate the number of workers dislocated as a result of NAFTA.
This section examines the job dislocation caused by NAFTA from two
perspectives. First, the available data show that the estimated number of jobs lost is
relatively small compared to total U.S. employment. Second, several case studies of
the effect of NAFTA-induced job loss on several communities are described. It is
clear that while job dislocation may be small relative to aggregate employment, the
hardships faced by those workers and communities that lose jobs is a serious concern.
This section concludes with a brief discussion of NAFTA’s effect on job security in
the United States.
Job Dislocation Relative to Total U.S. Employment
The implementing legislation for the NAFTA (P.L. 103-182) included a
Transitional Adjustment Assistance (TAA) program, which provides employment
services, training, income support, and job search and relocation allowances to eligible
workers. To receive benefits, workers must be certified by the U.S. Secretary of the
Labor as having lost their jobs (or threatened with job loss) by increased import
competition from Mexico or Canada or by production shifts to Mexico or Canada.
23U.S. Department of Commerce. Survey of Current Business. July 2001, p. D-3.
One by-product of the NAFTA-TAA program is that it provides data on the
number of workers certified as eligible to receive NAFTA-TAA benefits. Since these
data may include workers who were never laid off from their jobs, or who quickly
found new jobs, it may overstate the number of dislocated workers. On the other
hand, since some firms and workers may not know of the TAA program or that their
displacement results from NAFTA, the data may understate the number of dislocated
workers. Consequently, these data should not be considered a definitive measure of
job dislocation, but rather an indication of the magnitude of NAFTA-related job
losses. Although clearly imperfect, they are the best available data on job dislocation.
In the eight years ending December 31, 2001, 414,761 workers were covered by
certification, of which 34% were in the textiles and apparel industry, and 5% in the
automotive industry.24 By comparison, total U.S. employment in the fourth quarter
of 2001 was 134.3 million workers.25 Thus the 414,761 workers certified over eight
years is relatively small – about three tenths of a % of total U.S. employment of 134.3
million. The number of dislocated NAFTA workers is also very small relative to the
total number of unemployed workers in the United States. The 414,761 workers
covered by certification over eight years is about 6% of the 6.7 million unemployed
in 2001. This suggests that the effects of NAFTA are swamped by macroeconomic
trends in the U.S. economy.
In a recent study, it is estimated that NAFTA has eliminated 766,030 job
opportunities in the United States between 1994 and 2000 because of the growing
trade deficit with Mexico and Canada.26 Most economists disagree with the
methodology used in making this estimate.27 Nevertheless, assuming the estimate is
realistic, the number of job opportunities lost is still relatively small – about six tenths
of a percent of total U.S. employment.
Job Dislocation in Selected U.S. Communities
According to the NAFTA-TAA program, from January 1, 1994 to December 31,
2001, 3,331 U.S. plants were closed as a result of import competition from NAFTA
or production shifts to Mexico. Of these, 1,549 plants (involving 210,621 workers)
were certified as eligible because of production shifts to Mexico. This section
summarizes the experience of a few of the communities in which plants closed or
production was shifted to Mexico as a result of NAFTA. It provides a perspective on
24U.S. Department of Labor. Office of Trade Adjustment Assistance. Data based sorted by
26Scott, Robert E. NAFTA’s Hidden Costs: Trade Agreement Results in Job Losses,
Growing Inequality, and Wage Suppression for the United States. In NAFTA At Seven: Its
Impact on Workers in All Three Nations. Economic Policy Institute Briefing Paper. 2001,
27The methodology is based on the U.S. trade deficit with its NAFTA partners. As discussed
earlier, the trade deficit primarily results from changes in macroeconomic factors such as
changes in exchange rates and economic activity in the three NAFTA countries.
the problems facing dislocated workers and the problems inherent in existing attempts
to aid such workers.
A recent case study by the U.S. General Accounting Office (GAO) included six
U.S. communities that were especially hard-hit by trade-related layoffs since the mid-
Owosso, Michigan; Washington and Chocowinity, North Carolina; El Paso, Texas;
and Martinsville and Henry County, Virginia. The purpose of the study was to
examine the impact of trade-related layoffs on the communities and their experience
with trade adjustment assistance. Most of the layoffs were in the apparel, electronics,
furniture and food processing industries. Layoffs attributed to NAFTA were not
separately identified, but it can be assumed (since data from the NAFTA-TAA
program were used) that many of the layoffs were due to import competition or plant
relocation to Mexico.
Of the six communities, El Paso, Texas had 17,069 workers certified as eligible
by NAFTA-TAA since 1994. Of 6,000 workers laid off in Martinsville and Henry
County, Virginia since 1993, 3,500 were certified by NAFTA-TAA. Washington and
Chocowinity, North Carolina lost 1,500 jobs in the 1997-99 period. All the other
communities lost fewer than 1,000 jobs since 1994.
Workers affected by the layoffs in these six communities had less education than
the U.S. workforce as a whole. Eighty percent of the trade adjustment assistance
participants in the six communities had a high school education or less, compared to
The majority of workers were in their 40s, and women accounted for two-thirds of the
unemployed. Generally, the workforces were not suited to the “new economy” jobs,
and therefore local government officials could not easily attract such industries to the
The layoffs caused a dramatic rise in the unemployment rate in the six
communities. Local government tax revenues declined, affecting the services that
could be provided. Retail sales dropped as workers had less money to spend.
Business suppliers or contractors to the plants that closed were also adversely affected.
Many of the workers found new jobs in the area, but often at a lower wage.
An important lesson from the GAO study is that current trade adjustment
assistance is not meeting the needs of the workers in these communities. The
requirement that separate TAA (the trade adjustment assistance program that is not
directly related to NAFTA) and NAFTA-TAA accounts be maintained is
administratively inefficient and confusing.29 Since dislocated workers receive income
support for l8 months and training benefits for 24 months, they often drop out of
training after 18 months.
28U.S. General Accounting Office. Trade Adjustment Assistance: Experiences of Six Trade-
Impacted Communities. GAO-01-838. August 2001.
29The NAFTA-TAA and TAA were consolidated into a single program with one set of
procedures and eligibility criteria in Title I of the Trade Act of 2002 (P.L. 107-210, 116 Stat.
News accounts tell similar stories of plant closings in different communities. For
example, Chilhowie, Virginia became a thriving manufacturing town in the 1970s,
producing clothing, and furniture as well as other products.30 Since NAFTA went into
effect in 1994, four apparel factories Buster Brown (300 jobs), Tultex (200 jobs),
Natalie Knitting Mills (350 jobs), and Spring Ford (450 jobs) have shut their doors.
Recently, American of Martinsville, a furniture manufacturer (450 jobs) announced
it would close its plant, citing competition from imports from China and the effect of
September 11 on the lodging industry.
Another example is the closing of three apparel plants in rural Louisiana.31
About 240 jobs will be eliminated in Olla, Many and Ville Platte, Louisiana as the
plants relocate to Mexico, the Middle East and other foreign countries. Steve Graf,
the company’s Louisiana director of manufacturing, was quoted as saying “It was
strictly a business decision. These people are hard-working, and they helped build
Holloway (the Ohio-based manufacturer), but unfortunately, due to NAFTA, we can’t
Effect on Job Security
While the number of plant relocations to Mexico appears to be relatively small,
some argue that many employers use the threat of moving production to Mexico at the
bargaining table, in organizing union drives and in wage negotiations with individual
workers. In this respect, then, NAFTA may be increasing the economic insecurity
among U.S. workers. Kate Bronfenbrenner, Director of Labor Education Research
at Cornell University’s School of Industrial and Labor Relations, researched this issue.
She found that in National Labor Relations Board (NLRB) certification elections,
more than 50% of employers made threats to close all or part of the plant during the
organizing drive.32 By comparison, in the late 1980s, only 29% of employers made
plant closing threats during NLRB campaigns. Also, it is possible that threats to
close plants prevented union organizing drives from getting underway.
U.S. merchandise trade with Mexico, especially intra-industry and intra-firm
trade, increased more rapidly since 1994 than in the pre-NAFTA period and now
accounts for 12% of U.S. merchandise trade. This suggests that the process of
economic integration of the two economies, which began prior to 1994, strengthened
after 1994. The NAFTA, by improving business expectations, probably played a
significant role in the growth of trade and investment since 1988. Given Mexico’s
trade liberalization in the 1980s, however, economic integration probably would have
occurred without NAFTA, although more slowly.
30A Town Out of Work; Globalization Takes Toll on Industries. Seattle Times. May 11,
31More Job Loss from NAFTA. Advocate (Baton Rouge, La.), May 11, 2002, p. 8-B.
32Kate Bronfenbrenner. Testimony, U.S. Trade Deficit Review Commission. Hearings, Oct.
U.S. trade and investment with Mexico will probably continue to grow, but in the
near future its effect on the large U.S economy will remain relatively small. At some
point, the process of U.S. economic integration with Mexico will likely slow down,
as it did with Canada. U.S. merchandise trade with Canada has remained at 20% of
total U.S. trade for some time. The experience of the NAFTA suggests that future free
trade agreements between the United States and other countries will have a small
effect on the U.S. economy. If integration with close geographical partners like
Mexico and Canada has a small effect on the United States, it is likely that trade
agreements with more distant countries will have an even smaller effect on the U.S
Growing economic integration between the United States and Mexico has
increased productivity and efficiency for the U.S. economy. These benefits are small,
however, especially when distributed over the entire population. The costs of
economic integration are disproportionately borne by the workers who lose jobs and
the communities in which they live. The low level of education and skills of the
dislocated workers makes it difficult for them to be retrained in a short period of time.
From a policy perspective, a central issue is that, in the past, trade adjustment
assistance programs have not been very effective in helping workers gain new skills
or find new employment. Congress has attempted to remedy this deficiency in Title
1 of the Trade Act of 2002 which includes changes to the trade adjustment assistance