International Trade: Mexico's Maquiladora Decline Affects	 
U.S.-Mexico Border Communities and Trade; Recovery Depends in	 
Part on Mexico's Actions (25-JUL-03, GAO-03-891).		 
                                                                 
Mexico's maquiladoras have evolved into the largest component of 
U.S.-Mexico trade. Maquiladoras import raw materials and	 
components for processing or assembly by Mexican labor and	 
reexport the resulting products, primarily to the United States. 
Most maquiladoras are U.S. owned, and maquiladoras import most of
their components from U.S. suppliers. Maquiladoras have also been
an engine of growth for the U.S.-Mexico border. However, the	 
recent decline of maquiladora operations has raised concerns	 
about the impact on U.S. suppliers and on the economy of border  
communities. Because of these concerns, GAO was asked to analyze 
(1) changes in maquiladora employment and production, (2) factors
related to the maquiladoras' decline, and (3) implications of	 
recent developments for maquiladoras' viability.		 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-03-891 					        
    ACCNO:   A07724						        
  TITLE:     International Trade: Mexico's Maquiladora Decline Affects
U.S.-Mexico Border Communities and Trade; Recovery Depends in	 
Part on Mexico's Actions					 
     DATE:   07/25/2003 
  SUBJECT:   Foreign trade agreements				 
	     International cooperation				 
	     International relations				 
	     International trade				 
	     Labor statistics					 
	     Foreign trade policies				 
	     Mexico						 
	     North American Free Trade Agreement		 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-03-891

                                       A

Letter

July 25, 2003 The Honorable Max Baucus Ranking Minority Member Committee
on Finance United States Senate The Honorable Jeff Bingaman The Honorable
Kay Bailey Hutchison United States Senate Mexico*s Maquiladora program,
which was established in 1965 to attract investment and create jobs along
the U. S.- Mexico border, has evolved into the largest and most dynamic
component of U. S.- Mexico trade. U. S. companies own the vast majority of
maquiladora plants, and maquiladoras

import about 80 percent of their components from U. S. suppliers. With
double- digit growth rates in output and employment until its peak in the
last quarter of 2000, maquiladora- based production has been one of the
engines of regional employment and income growth for the U. S.* Mexico
border. However, over the past 2 years, the level of employment and

production in maquiladora operations has declined sharply. This decline
has led a number of observers to express concern about the future
viability of maquiladoras. Some observers in the United States are also
concerned that the decline of Mexico*s maquiladoras could adversely affect
U. S. companies that supply these plants and could hurt the economy of
border

communities. In response to your concern about the significance of these
developments, this report analyzes (1) the ways in which the communities
along the U. S. Mexico border are integrated and how maquiladoras have
contributed to U. S.- Mexico interdependence; (2) recent changes in
maquiladora production, employment, and cross- border trade; (3) factors
that have affected employment and production in the maquiladora sector;
and (4)

factors that could affect the maquiladora industry*s future viability. To
address these objectives, we met with U. S. and Mexican government
officials in Washington, D. C., and Mexico City. We contacted business and
nonprofit sector representatives, academicians, and other experts on the
maquiladora industry in the United States and Mexico, and we reviewed
extensive documentation and academic research provided by these sources.
We obtained and analyzed official data on employment and trade trends from
both U. S. and Mexican government agencies. We also

conducted a series of semistructured interviews with 29 representatives of
business associations consisting of maquiladora- specific and principal
industry sector organizations at the local and national levels. We relied
on business associations because, as representatives of the maquiladoras,
they could comment on issues facing their members, such as increased

competition, and could explain the reasons for plant closures, changes in
employment levels, and other changes within the industry. We conducted
site visits in three major border *twin cities*: San Diego, California*
Tijuana, Baja California; El Paso, Texas* Juarez, Chihuahua; and McAllen,
Texas*

Reynosa, Tamaulipas. Results in Brief A variety of social and economic
factors link U. S. and Mexican border

communities, and maquiladoras play a significant role in this
interdependence. Trade figures indicate that the four U. S. border states
account for about 62 percent of U. S. exports to Mexico, while 70 percent
of

these exports were destined for Mexican border states. Border communities
are also drawn together socially by family and educational ties and
economically by twin- plant production and retail commerce. Residents in
the twin cities cross the border about one million times every day to
work, shop, attend classes, visit family, and participate in other
activities. The maquiladora sector, which relies heavily on imports from
the United States and represents the principal industrial activity on the
Mexican side of the border, drives cross- border economic integration as
well as the increasing U. S.- Mexico interdependence. However, the border
is a diverse region, and the extent of interdependence between communities
along the border varies widely.

After growing rapidly during the 1990s, Mexican maquiladoras experienced a
sharp decline in production and employment after October 2000. In early
2002, employment in the maquiladora sector had contracted about 20
percent, losing nearly 290,000 jobs, and production had contracted about
30 percent. The decline was particularly severe in certain industries and
cities. For example, maquiladora employment in the Mexican electronics
industry declined 31 percent between 2000 and 2002, and Tijuana, a city
with significant maquiladora electronics manufacturing, experienced a 30
percent decline in maquiladora employment. In addition, overall
manufacturing production in the Mexican border region began declining in
2000. The downturn was felt on the U. S. side as well. For example in
2001, the value of U. S. exports through U. S.- Mexico land border ports
fell by 10

percent. Similarly, employment on the U. S. side of the border declined in
manufacturing and certain other trade- related sectors. Despite these

contractions, overall U. S. border employment grew in most U. S. border
metropolitan statistical areas.

According to government researchers, academicians, and industry
representatives, both cyclical and structural factors have contributed to
the decline in Mexico*s maquiladora employment and production since 2000.
Representatives of industry groups emphasized, and our economic analysis
confirms, that the cyclical downturn in the U. S. economy has been a
primary factor in the decline of the maquiladoras. However, industry
sources and other experts noted that Mexico*s maquiladoras also face
increased global competition in the U. S. market, particularly from China,

Central America and the Caribbean. The real appreciation of the peso
relative to the dollar and key competitors* currencies has heightened such
pressure. Additionally, industry representatives indicated that Mexican
government policies such as changing the tax regime applied to
maquiladoras have created a climate of uncertainty for investors.
Meanwhile, owing to commitments undertaken under the North America Free
Trade Agreement (NAFTA), Mexico has now phased out some benefits to the
maquiladora sector.

Factors affecting the recovery of Mexico*s maquiladoras include recent
industry and government actions and the prospect of future Mexican
reforms. The recent decline of the maquiladoras has added impetus to the
ongoing evolution in the industry toward more sophisticated manufacturing
and prompted the Mexican government to take several

steps in support of the maquiladoras. For example, the Mexican government
has greatly expanded the number of components that can be imported by
maquiladoras and other firms with little or no duty assessments. However,
government, industry, and other experts agree that additional fundamental
reforms by Mexico, in areas such as energy and labor practices, are still
necessary to restore the country*s attractiveness as a business and
investment location. Though difficult, tackling such reforms is made more
urgent by U. S. trade and homeland security policies that are likely to
present further challenges for maquiladora operations.

Background Mexico*s Maquiladora program has been a central feature of the
U. S. Mexico border. The U. S.- Mexico border stretches nearly 2,000
miles, from

the Pacific Ocean in California to the Gulf of Mexico in Texas. Four U. S.
states (Arizona, California, New Mexico, and Texas) and six Mexican states
(Baja California, Chihuahua, Coahuila, Nuevo Leon, Sonora, and Tamaulipas)
make up the border. Texas contains the longest section of the

U. S. border with Mexico, with several large and numerous small border
crossings across the Rio Grande. Compared with Texas, California*s border
with Mexico is relatively short, but it includes San Diego* Tijuana, the
single busiest U. S.- Mexico border crossing. Arizona*s principal border
crossing with the Mexican state of Sonora at Nogales plays a significant
role in agricultural trade. The relatively small border crossings between
New Mexico and Mexico reflect the sparsely populated areas in that region
of the border. Figure 1 shows the U. S.- Mexico border, including all U.
S. and

Mexican border states, some Mexican border cities with varying
concentrations of maquiladora plants, and some ports of entry on the U. S.
side of the border.

Figure 1: Map of U. S.- Mexico Border Twin Cities

During the 1990s, the population along the border experienced significant
growth. On the U. S. side, the population increased by 21 percent,
considerably more than the overall U. S. population, which grew by 13.2
percent. Some cities on the U. S. border experienced significant increases
in population, such as Yuma, Arizona, and McAllen, Texas* respectively,
the

third and fourth fastest growing metropolitan areas in the United States.
Population on the Mexican side of the border increased even more rapidly,
growing by 32 percent between 1990 and 2000. The majority of the border*s
residents live in communities along the border that are composed of twin
cities* a city on each side of the border* such as San Diego* Tijuana and

El Paso* Juarez. The San Diego* Tijuana area alone has a combined
population of about 4 million, and the El Paso* Juarez area has a
population of 1.9 million.

The Maquiladora 1 program was first established by the government of
Mexico in 1965 as part of the Border Industrialization Program (BIP) and
maquiladoras have been a driving force in the development of the U. S.
Mexico border region. Under the BIP, Mexico encouraged foreign
corporations to establish operations along the northern border to provide
employment opportunities for Mexican workers displaced after the
termination of a temporary cross- border work arrangement known as the
Bracero Program. 2 Also known as *in- bond* plants, 3 maquiladoras were
allowed to import temporarily, on a duty* free basis, raw materials and
components for processing or assembly by Mexican labor and to re- export
the resulting products, primarily to the United States.

The maquiladoras have undergone a dynamic evolution over the last four
decades. In the mid- 1960s, maquiladoras consisted primarily of basic
assembly operations taking advantage of Mexico*s low labor costs. By the
1980s, U. S. multinationals representing various industrial sectors
established maquiladora plants along the U. S.- Mexico border. Japanese
and European companies also established maquiladora plants in Mexico to
compete in the U. S. market. Since the 1980s, some firms moved from
lowskilled assembly work to more advanced manufacturing operations.
Researchers from Mexico*s Colegio de la Frontera and San Diego State

1 Maquiladora is a term derived from the Spanish word maquilar, which is
the service a miller provides when he grinds wheat into flour. Similarly,
a maquiladora provides assembly services without necessarily taking
ownership of the goods being assembled.

2 The Bracero Program allowed Mexican citizens to work on a temporary
basis in the United States between 1942 and 1964. It was initially
designed to address labor shortages in the U. S. agricultural and railroad
industries during World War II. 3 When the BIP was established, companies
with assembly plants in Mexico would deposit a

bond with the Mexican Department of Commerce and Industry (Secretaria de
Comercio y Fomento Industrial) for the value of the duty on imported
components and machinery. The bond would be returned when the finished
products assembled using the imported components were exported.

University note that the number of *technical workers* employed by
maquiladoras increased significantly from the early 1980s to the 1990s.
Some maquiladoras now employ workers in development and design as well as
manufacturing. For example, Delphi Automotive in Juarez, the largest
private employer among maquiladoras in Mexico, now has a sophisticated
research and development center that employs hundreds of highly skilled
workers and engineers.

Over the years, as maquiladoras evolved and expanded, the term

maquiladora has come to be used loosely to refer to almost any subsidiary
plant of a foreign company involved in export from Mexico, particularly
those located along the U. S. border. However, the Maquiladora program
continues to be quite distinct from other efforts initiated by the Mexican
government to encourage exports. 4 Firms must register with the government
of Mexico to be considered maquiladoras and, once registered, are eligible
for several key benefits, such as preferential tariffs on inputs and
machinery, and simplified Mexican customs procedures. In this report, we
define maquiladoras as those firms officially participating in Mexico*s
Maquiladora program. In addition to the Maquiladora program, the U. S.-
Mexico trade relationship

has also been influenced by other important developments such as NAFTA.
NAFTA was concluded between the United States, Mexico and Canada in 1992
and entered into force on January 1, 1994. This agreement provided, among
its other provisions, for the elimination of tariffs and other barriers to
U. S.- Mexico bilateral trade by 2008. It also required Mexico to change
certain provisions of the Maquiladora program, such as elimination of
duty4

Among the most important of these programs is PITEX (Program for Temporary
Importation to Manufactured Exports), established in 1990 as another
government of Mexico program that allows companies to import components
duty- free. PITEX requires companies to export a minimum of 30 percent of
their total annual sales. Companies operating under PITEX are more
commonly located in the interior of Mexico and typically use more Mexican
components than do maquiladoras, which are primarily located in Mexico*s
northern border region. The type of employment and production data the
Mexican government collects on maquiladoras is not available for PITEX
firms. The U. S.

International Trade Commission (ITC) issues an annual report on
production- sharing trade, including between the United States and Mexico,
as part of its Industry, Trade, and Technology Review series. In that
report, the ITC considers trade under the Maquiladora program and the
PITEX program as production- sharing trade. For comparison of Mexico*s
imports and exports under the Maquiladora Program and PITEX by Harmonized
Tariff

chapters, see Ralph Watkins, U. S. International Trade Commission,
*Production- Sharing Update: Developments in 2001,* Industry Trade and
Technology Review, USITC, pub. 3534, July 2002, appendix C.

free benefits for imports of components from non- NAFTA countries. U. S.
Mexico trade has expanded sharply since NAFTA*s inception. Much of this
trade involves *production sharing,* whereby final goods are produced

with parts, labor, and manufacturing facilities from the United States and
Mexico. Because it enables firms to increase specialization, take
advantage of low labor costs in Mexico, and attain other efficiencies,
production sharing is a key benefit to U. S. companies under the
Maquiladora program.

Maquiladoras A variety of social and economic factors create strong
linkages between Contribute to

communities on both sides of the U. S.- Mexico border, and maquiladoras
play a critical part in this interdependence. Residents in the twin cities
Integration along the

cross the border about one million times every day to work, shop, attend
Diverse U. S.- Mexico

classes, visit family, and participate in other activities. Maquiladoras
have Border

increased trade between the United States and Mexico and have helped to
develop the economies of several border regions. While communities along
the U. S.- Mexico border share certain traits, each region is distinct.
Multiple Social and

A wide range of social ties* educational, political, cultural, and
familial* Economic Ties Fuel

contribute to integration along the U. S.- Mexico border. For example,
Integration at the Border

certain U. S. universities in border cities offer combined degrees or
exchange programs with their counterparts on the Mexican side. In some
schools, such as the University of Texas at El Paso and the University of
Texas* Pan American, Mexican nationals cross the border regularly to
attend classes. Political interaction and cooperation between local
authorities of twin cities enhance integration. Cultural and family ties
also contribute significantly to integration at the border. The U. S.
counties with

the highest concentration of Hispanics are located along the southwest
border, and by far most of the Hispanics in southern border states are of
Mexican descent.

Trade and retail sales contribute to economic interdependence at the
border. Approximately $200 billion in trade went through the U. S.- Mexico
border in 2002. Much of U. S.- Mexico trade occurs between border states.
For example, 62 percent of U. S. exports to Mexico originated in Texas,
California, Arizona, and New Mexico; of this, 70 percent was destined for

Mexican border states. Research by the Federal Reserve Bank of Dallas
indicates that trade between the United States and Mexico has positive
effects on border communities, because U. S. border cities typically
provide

a variety of services such as transportation and customs brokerage. 5
Retail sales to Mexican nationals also contribute significantly to the
economies of cities on the U. S. side of the border. According to one
estimate, retailers in Texas annually make an estimated $15 billion in
sales to Mexican shoppers. In McAllen, Texas, 35 percent, or about $700
million worth, of retail sales are made to Mexican nationals. Residents
from Tijuana make 1.5 million trips per month into the San Diego area,
mainly to shop. In El Paso, Juarez residents account for more than 20
percent of retail sales. On the other hand, because of the high cost of
pharmaceuticals in the United States, a growing number of U. S. residents
regularly cross the border into Mexico to purchase prescription drugs.

Maquiladoras Drive U. S.* Maquiladoras import most inputs from the United
States and export most

Mexico Trade and Border of what they produce back to the United States.
Growth in U. S.* Mexico

Integration trade and economic interdependence at the border during the
last decade

can be explained to a great degree by the participation of maquiladoras in
supplying a strong U. S. market during the 1990s. Mexican exports
increased by about 340 percent between 1993 and 2001, in large part
because maquiladora- related exports increased by over 400 percent during
this time, according to a report by the Mexican Commission on Northern

Border Affairs. 6 By 2001, maquiladoras accounted for 41 percent of total
Mexican trade with all countries * 34 percent of Mexico*s imports and 48
percent of its exports (see fig. 2). Trade with the United States makes up
a significant share of maquiladora trade. In 2001, 79 percent of
maquiladora imports of components and parts for production were from the
United States and 98 percent of their exported products were destined for
the U. S. market. Maquiladora trade between the United States and Mexico
totaled

about $121 billion in 2001, with maquiladora exports ($ 75 billion)
accounting for more than half of Mexico*s total exports to the United
States. 7 Border cities are typically seen as the primary beneficiaries of

growing U. S.- Mexico trade. However, states such as Florida, Tennessee, 5
Lucinda Vargas, Federal Reserve Bank of Dallas, *The Binational Importance
of the Maquiladora Industry*, Southwest Economy, Issue 6, November/
December, 1999. 6 Northern Border Regional Development Program, 2001- 2006
(Programa de Desarrollo Regional, Frontera Norte 2001- 2006), Commission
on Northern Border Affairs (Comision para Asuntos de la Frontera Norte).

7 Exports from PITEX assembly plants ($ 46 million), accounted for another
one- third of Mexican exports to the United States.

and Ohio, which doubled their exports to Mexico during the second half of
1990s, have also benefited from growing U. S.- Mexico trade.

Figure 2: Maquiladoras* Share of Mexico*s Trade 1990- 2002 Furthermore,
maquiladoras are directly connected to U. S. companies through ownership
and production ties. The list of Mexico*s top 100 maquiladora employers
includes such U. S. firms as Delphi, RCA, Ford Motor Company, Tyco,
General Electric, General Instruments, Johnson &

Johnson, and ITT. All told, 79 percent of the top 100 maquiladora
employers are from the United States. Maquiladoras are important to the
United States because they are a strategic means by which U. S. companies
stay

competitive in the global marketplace. By offering lower production costs,
maquiladoras enable U. S. companies to produce goods more cheaply in
Mexico than in the United States. In essence, maquiladoras and U. S.

companies are part of a greater production- sharing model, which is an
important part of overall North American production. Moreover, more than

26,000 U. S.- based companies, located mainly in the Midwest, supply
maquiladoras with raw materials and components.

The Mexican border region has benefited in terms of job creation from the
dominant presence of maquiladoras on the Mexican side of the border.
Overall, 77 percent of all maquiladora establishments are located in the
six

Mexican border states shown in figure 3. Also, about 83 percent of
maquiladora employment was located in border states. During most of the
1990s, maquiladoras represented more than half of the industrial activity
in the states of Chihuahua and Tamaulipas. During the same time period,
the

maquiladoras represented nearly three quarters of industrial production in
the state of Baja California, which contained almost one third of Mexico*s
maquiladora firms. Cities on the U. S. side of the border have benefited
from the large flow of trade created by maquiladoras. Between 1990 and
2002, more than half a million jobs were added to the U. S. border region,
including jobs in services, retail trade, finance, and transportation, and
after 1995, employment growth in the U. S. border region exceeded the U.
S. national average (see app. I for details). The employment gains are
particularly notable, because the border region historically has had high
rates of unemployment. Some studies have outlined the effect of overall
border

economic trends on local border communities. For example, researchers
estimated that in one Texas border community, in 2001, services and
supplies purchased by maquiladoras amounted to $136 million and a total of
32,577 jobs were sustained by maquiladoras and related manufacturing
activity.

Figure 3: Map of Mexico Showing Share of Maquiladora Establishments, by
State

The same researchers estimated that 15 percent of maquiladora workers*
salaries was spent in the region on goods and services. In one Arizona
border community, researchers surveyed maquiladora workers and found that
workers who crossed the border to shop made an average of 5.5 trips a
month and spent about $35 on each trip. Almost one third of retail sales
in

the same Arizona community are attributed to Mexican nationals, according
to local sources.

Despite their role in generating employment in Mexico, the maquiladoras*
benefit to the country remains a subject of some debate. Some express
reservations about the maquiladoras* ability to generate economic
development for Mexico, since these plants have generally been unable to
establish a network of domestic inputs providers or create significant

linkages to the internal Mexican economy. In April 2002, for example, the
former Mexican Foreign Minister noted that without proactive Mexican
government policy to set up domestic suppliers, the benefits of the
maquiladora industry would never extend beyond the border. In addition,
critics in the environmental and labor movements on both sides of the
border also assail these plants. 8 Some environmental groups claim that
maquiladoras are responsible for the growing pollution problem in the
border region. Similarly, some labor organizations criticize the
maquiladoras for the low wages paid to workers and for allegedly poor
working conditions.

Border Has Distinct Regions Although communities along the U. S.- Mexico
border share certain traits,

with Varying Degrees of they are also quite distinct. The level of
integration between cross- border

Integration twin cities depends on location, population, economic profile,
and crossborder

political cooperation. We observed some of these differences during
fieldwork in three border areas: McAllen* Reynosa, El Paso* Ciudad Juarez,
and San Diego* Tijuana.

McAllen* Reynosa. McAllen and Reynosa are economically interdependent.
Both are medium- size cities, McAllen with a population of about 569,000,
and Reynosa with about 420,000, and there are no other sizeable urban
areas nearby on either side of the border. Officials at the McAllen
Economic Development Corporation (MEDC) capitalized on the interdependence
between McAllen and Reynosa and incorporated it into their economic
strategy for the region starting in 1988. At that time, McAllen had high
unemployment and Reynosa*s economy was based on

8 We have prepared several reports on environmental and labor issues
related to maquiladoras. See U. S.- Mexico Trade: Assessment of Mexico*s
Environmental Controls for New Companies, GAO/ GGD- 92- 113 (Washington,
D. C.: Aug. 3, 1992) and U. S.- Mexico Trade: The Work Environment at
Eight U. S.- Owned Maquiladora Auto Parts Plants, GAO/ GGD94- 22
(Washington, D. C.: Nov. 1, 1993).

subsistence farming. Working with political leaders in McAllen and
Reynosa, MEDC developed a strategy based on promoting industrial
development in Reynosa, recognizing that if companies opened maquiladoras
there, McAllen would benefit by providing inputs and offering management,
engineering, warehousing, trucking, legal, and accounting services. In the
14 years since its establishment, MEDC has recruited 178 companies to the
area, in diverse manufacturing sectors, such as electronics, auto parts,
and telecommunications.

El Paso* Ciudad Juarez. Although El Paso and Ciudad Juarez are also
closely integrated, such ties have developed differently than in McAllen*
Reynosa. Ciudad Juarez is a larger metropolitan area, with a population of
1.2 million, and it is home to more maquiladora employees than any other
Mexican city. On the U. S. side, El Paso and the neighboring communities
in southern New Mexico are much smaller. El Paso*s economy has been shaped
by economic activity in Ciudad Juarez, especially that of providing
services to maquiladoras. In addition, Juarez residents contribute to El
Paso*s economy by purchasing items ranging from cars to clothing and
services such as financial and health services. In 2001, there were
approximately 46 million northbound crossings via the three international

bridges that connect the two cities. However, business leaders and other
observers whom we met in the El Paso* Juarez area frequently noted that
integration and economic dependence between El Paso and Juarez occurred
spontaneously, rather than by design. The development of maquiladoras on
the Mexican side occurred much earlier than in Reynosa and was largely
attributed to individual efforts by entrepreneurs in Juarez

and El Paso that began in the 1960s, rather than to a collective vision.
However, in recent years, in Santa Teresa, a nearby border community in
southern New Mexico, developers created a strategic plan to build an

industrial park as a supplier base, with warehouse and distribution
facilities, to service maquiladoras. The Santa Teresa port of entry opened
11 years ago. Developers envisioned that this border crossing would serve
as an alternate point of entry to El Paso for cross border trade.

San Diego* Tijuana. The dynamics of integration between San Diego and
Tijuana are notably different from other cross- border twin cities. In
this area, economic dependence is more one- sided. Unlike El Paso or
McAllen, San Diego is a large metropolitan area in its own right, with a
population of close to 3 million. Many of the major economic activities in
San Diego, including defense and space manufacturing, biosciences, and
tourism, are not directly connected to Tijuana. While Hispanics account
for at least 50 percent of the population in most U. S. counties along the
Southwest

border, they account for only about 27 percent of the population in San
Diego County, suggesting lower levels of family ties or connections to
Mexico. In contrast, Tijuana, with a population of about 1.2 million, is
heavily dependent on maquiladoras, and the city is closely tied to the U.
S. market. In addition to U. S. companies, Tijuana has also been the
preferred location for Japanese and Korean maquiladora investments, which
have made this area the world*s leading producer of color televisions.
More than 600 maquiladora plants, employing approximately 150,000 people
are

located in Tijuana. Moreover people in Tijuana are more likely to cross
the border to shop or do business in San Diego than vice versa; in fact,
it is estimated that two out of three residents of San Diego have never
been to Tijuana. In contrast, Tijuana residents spend between 3 to 5
billion dollars

in purchases in the San Diego region, mostly in the communities adjacent
to the border. In addition, 7 percent of economically active people in
Tijuana work in San Diego, earning an estimated $650 million a year in

wages and salary income. After Rapid Growth,

Maquiladora production and employment grew rapidly throughout the
Maquiladoras and

1990s but declined sharply after October 2000. Within the diverse
maquiladora sector, the decline was particularly steep in certain
industries Border Region

and in some border cities. Overall, Mexican manufacturing production in
Experienced Declines

the border region also declined and cross- border trade flows fell. At the
Beginning in 2000

same time, U. S. border employment in manufacturing and certain other
trade- related sectors contracted. Nevertheless, the U. S. border region
continued to experience stronger employment growth than did the United
States as a whole.

Maquiladoras Grew Rapidly During the 1990s, maquiladoras proved to be one
of the more dynamic

in the 1990s, with Growth components of Mexican manufacturing. Maquiladora
production increased

Varied by Region and by 197 percent from January 1993 until its peak in
October 2000, while

Industry overall manufacturing production in Mexico increased by only 58
percent

in the same time period (see fig. 4). During that time period, maquiladora
employment tripled, adding more than 900,000 jobs to the Mexican economy.
In 2000, maquiladoras accounted for about 4 percent of total

employment and about 20 percent of manufacturing employment in Mexico.

Figure 4: Growth in Maquiladora and Total Mexican Manufacturing
Production, 1993* 2002

Note: Certain statistics on Mexican industrial production were not
available for years prior to 1993.

With respect to employment, most major Mexican border cities and
industrial sectors experienced growth in maquiladora employment over the
decade, although some grew faster than others. For example, Tijuana and
Mexicali tripled their maquiladora employment, and the electronics
industry more than doubled its maquiladora employment in the border
region. The electronics industry, which was already the largest
maquiladora employer, added more than 200,000 jobs in the border region
during the 1990s. For the Mexican border region as a whole, maquiladora
employment rose 145 percent* from 342,555 in January 1990 to 839, 200 in
October 2000 (see app. V, table 8, for more information). While
maquiladoras have

typically been concentrated in the border region, maquiladora employment
growth throughout the rest of Mexico was actually higher than in the

border region during the 1990s. Growth in the nonborder region was
particularly strong in the textile and apparel sector, in which employment
rose in the nonborder region from about 22,000 in 1990 to about 224,000
jobs in 2001 (fig. 5). As a result of the stronger growth in the nonborder
region, the share of textile and apparel maquiladora employment in the
border region fell from 49 percent in 1990 to 17 percent in 2001. 9 Much
of the investment in the apparel sector occurred in anticipation of duty-
free treatment for most U. S. imports of apparel from Mexico under NAFTA
in 1999. 10

9 Data on textiles and apparel in the border region were available
annually only through 2001. The National Institute of Statistics,
Geography and Information Technology (Instituto Nacional de Estadistica,
Geografia e Informatica) provided us data broken down by industry and
region through October 2002.

10 For information about investment in Mexico*s textile and apparel sector
in 1999, see Larry Brookhart and Ralph Watkins, U. S. International Trade
Commission, *Production- Sharing Update: Developments in 1999,* Industry
Trade and Technology Review, USITC, pub. 3335, July 2000, p. 15.

Figure 5: Maquiladora Textile and Apparel Employment, Nonborder and Border
Regions, 1990* 2001

Maquiladora Decline Started After growing since the program*s inception
over 35 years ago, particularly

in 2000, Unevenly Affecting in the 1990s, Mexican maquiladora production
and employment began to

Industries and Border Cities decline sharply in late 2000. Maquiladora
production declined about 30

percent from late 2000 to early 2002. At the same time, maquiladora
employment contracted about 20 percent, losing nearly 290,000 jobs
nationally, about 174,000 of which were located in the border region. 11
Similarly, the number of maquiladora establishments (factories) in
operation began to decline as well (see fig. 6). Nevertheless, even with
the 11 Mexico*s National Institute of Statistics, Geography and
Information Technology (Instituto

Nacional de Estadistica, Geografia e Informatica) defines the border
region as the 41 municipalities located along the U. S.- Mexico border.

pronounced declines, the overall numbers of maquiladora employees remain
at levels similar to those in 1998* 1999.

Figure 6: Maquiladora Employment and Establishments, 1990- 2002

While the Mexican maquiladora downturn was evident both nationally and in
the border region, certain industries experienced larger declines (see
fig. 7). For instance, in the border region, the electronics industry
experienced one of the steepest and largest maquiladora employment
declines of any industrial sector, contracting by 31 percent and losing
more than 112,000 jobs in the 2- year period between October 2000 and
October 2002. 12 In contrast, the automobile and auto parts industry
experienced a less severe

12 Nationally in Mexico, the electronics and electrical components sector
declined by 32 percent from September 2000 through April 2002.

maquiladora employment decline of 13 percent (about 24,000 jobs) in less
than a year, before resuming some growth in November 2001. Textiles and
apparel also experienced a steep employment decline, falling by 26 percent
and losing more than 12,000 jobs. Nationally, the textile and apparel
industry lost more than 70,000 jobs. In all other border industrial
sectors combined, maquiladora employment declined by about 16 percent over
a little more than a year but has grown by about 4 percent since January
2002.

Figure 7: Mexican Maquiladora Employment in the Border Region by
Industrial Sector, January 1997* October 2002

Note: Data broken down by industrial sector in the border region were
available only through October 2002.

As figure 8 illustrates, the decline in maquiladora employment also
affected cities in the Mexican border region differently. The two largest
border

cities, Juarez and Tijuana, both experienced significant declines in
maquiladora employment, accounting for over half of the total jobs lost in
the border region. After peaking in October 2000, by December 2002,
maquiladora employment had fallen 27 percent in Juarez and 30 percent in
Tijuana. The smaller city of Nogales, Sonora, experienced one of the

sharpest percentage changes in maquiladora employment in the border
region, declining by 44 percent. In contrast, the city of Reynosa
experienced a decline of only about 5 percent between September and
December 2000, and its maquiladora employment has since rebounded, with 7
percent growth since January 2001. Reynosa*s decline in electronics and
auto parts employment was much less severe than other cities. Figure 8:
Mexican Maquiladora Employment, by Border City, January 1990 - December
2002

Mexican Border The decline in Mexico*s maquiladora production contributed
to a decline in

Experienced Overall overall manufacturing production in Mexico*s border
region. 13 Figure 9

Decline in Manufacturing shows the growth of manufacturing production for
three Mexican border

and Cross- Border Trade states: Baja California, Coahuila, and Sonora.
Baja California, the state with

the largest share of maquiladoras, grew more rapidly than the other border
states but also experienced the largest decline in overall manufacturing
production after October 2000. Similarly, manufacturing production in
Coahuila, Nuevo Leon (not shown), and Sonora also experienced downturns
beginning in late 2000 and early 2001. 14

13 Overall manufacturing production consists of manufacturing by both
maquiladora and nonmaquiladora manufacturing companies. The growth in both
maquiladora production and total manufacturing production in Mexico is
shown in figure 3.

14 Data on manufacturing production were not available for the Mexican
border states of Chihuahua and Tamaulipas.

Figure 9: Growth of Manufacturing Production in Mexican Border States,
1993* 2002

During the maquiladora decline, exports, imports, and overall trade
through U. S.- Mexico land border ports also dropped. The value of
crossborder trade dropped 5 percent in 2001 and remained flat in 2002,
owing in large part to the 10 percent decline in U. S. exports to Mexico
through these ports. Although each of the four major land border ports
experienced some decline, Nogales experienced the greatest decline, losing
about 20 percent of its value between 2000 and 2002 (see app. IV, table 7,
for levels of U. S. trade with Mexico through the four main land border
points). Maquiladoras, which accounted for 40 percent of U. S. exports to
Mexico and 54 percent of Mexican exports to the United States in 2001,
contributed

to this decline.

U. S. Border Employment in The decline in Mexico*s maquiladoras was also
felt on the U. S. side, as

Manufacturing and manufacturing employment in border municipalities
declined by 6 percent

overall from 2000 through 2002. 15 Other U. S. sectors related to trade
also Transportation Services

experienced declines in employment at the border. U. S. border Fell, but
Overall

employment in transportation and public utilities, which includes trucking
Employment Growth

and warehousing, was down 4 percent, and employment in wholesale trade
Continued

was down 3 percent overall. Similar to the maquiladora employment declines
in Mexico, employment declines on the U. S. side of the border also varied
by region. For example, manufacturing employment declined by 18 percent
overall in Texas* border cities, and employment declines in wholesale
trade and transportation and in public utilities were more pronounced in
Arizona. (App. I provides a detailed analysis of employment trends in the
U. S. border region.)

Despite the contractions in manufacturing and certain other trade- related
sectors, other sectors in the U. S. border region grew. As a result, total
nonagriculture- related employment in the border area grew by 4 percent
even after the U. S. economic slowdown began in 2000 and national
employment contracted 1 percent through 2002. Some border metropolitan
areas maintained even stronger employment growth. For example, the McAllen
area grew by 9 percent between 2000 and 2002, while Laredo grew by 6
percent, and San Diego and Las Cruces grew by 5 percent each over the same
period. On the other hand, El Paso*s overall nonfarm employment

fell, primarily because its mix of industries is weighted towards sectors
that have been shrinking (see app. I for details).

Cyclical and Structural The decline in maquiladora production and
employment since the last

Factors Cited in quarter of 2000 is attributable to both cyclical and
structural factors. Government researchers, academicians, economic
studies, and industry

Maquiladora Decline representatives agree that the cyclical downturn in
the U. S. economy has

been a primary factor in the decline. However, industry sources and other
experts emphasized that the maquiladoras have also been adversely affected
by structural factors, such as increased competition in the U. S. market,
particularly from China, Central America and the Caribbean, and by the
strength of the Mexican peso, which has further eroded the

15 The U. S. border municipalities for which appropriate employment data
through 2002 were available are San Diego, California; Yuma, Arizona; Las
Cruces, New Mexico; and Brownsville, El Paso, Laredo, and McAllen, Texas.

maquiladoras* competitiveness. Changing Mexican tax policies have also
contributed to the maquiladora decline by creating a climate of
uncertainty for foreign investors. Meanwhile, owing to commitments
undertaken under NAFTA, Mexico has phased out some of the key benefits of
the Maquiladora program.

It is clear from our research that all of these factors were at work
before and during the recent maquiladora downturn, and that each was
changing in a direction adverse for maquiladora production and employment.
However, the sheer number of simultaneous changes over a relatively brief

period makes it difficult to isolate or quantify the impact of individual
factors. Although many government, academic, and industry sources
generally refer to the cyclical downturn in the U. S. economy as a
principal factor in the decrease in maquiladora employment and production
since the last quarter of 2000, there is no such agreement on the relative
importance of other factors associated with the decline of the
maquiladoras. Therefore, the order in which we present these other factors

is generally based on the results of our semistructured interviews with
industry associations (see app. VI).

U. S. Economic Slowdown In explaining the decline in maquiladora
production and employment

Adversely Affects beginning in the last quarter of 2000, government,
academic, and industry

Maquiladoras sources generally emphasized the role of the downturn in the
U. S. economy. Of the 23 industry association representatives we
interviewed

whose membership had experienced a decline in production or employment,
about three- quarters cited the recent downturn in the U. S. economy as a
major factor. As noted earlier in this report, maquiladora

production is often linked to U. S. manufacturing through
productionsharing arrangements. In fact, about 98 percent of maquiladora
production is destined for the U. S. market. Thus, it is not surprising
that the maquiladoras are very sensitive to fluctuations in U. S.
manufacturing and demand. Our analysis of economic data supports the
conclusion of experts and interviewees, demonstrating that historically
maquiladora employment typically grows when the overall U. S. economy
expands and is negatively influenced when the U. S. economy slows down
(see app. II for a discussion of the effect of the economic downturn in
the United States on employment

for various maquiladora industrial sectors). Moreover, maquiladora
employment has been even more sensitive to changes in U. S. manufacturing
production, particularly in sectors such as textiles and autos, and a
sharp drop in U. S. manufacturing has characterized the

present U. S. economic slowdown.

As figure 10 illustrates, maquiladora employment shows a correlation with
U. S. economic performance over the past two decades. On average,
maquiladoras added almost 118,000 employees annually from 1995 to 2000. 16
During this period, U. S. annual economic growth averaged 3.6

percent. However, in 2001, as U. S. economic growth slowed to 1.4 percent,
the maquiladoras lost nearly 229,000 jobs.

Figure 10: Annual Growth Rates of U. S. Gross Domestic Product and
Maquiladora Employment, 1980- 2002

16 Maquiladoras reached their peak employment level of 1.3 million
employees in October 2000.

Moreover, although the Mexican economy as a whole is very closely linked
to that of the United States, the maquiladoras appear to have been
affected by the U. S. economic slowdown more severely than the Mexican
economy

overall. While Mexico*s economy contracted .2 percent in 2001, it resumed
growth at .7 percent in 2002. However, the maquiladora sector declined
both years 9.2 percent and 8.3 percent, respectively. 17 Of the industry
associations indicating that their membership had experienced a decline in
employment or production, about half reported that the maquiladoras had
been more negatively affected by the U. S. economic downturn than had
other businesses in Mexico.

Mexico Faces Increased Among the 23 industry associations that indicated a
decline in their

Global Competition in the memberships* employment or production, mounting
foreign competition in

U. S. Market the U. S. market was the most frequently offered explanation
for the decline

of the maquiladoras over the past 2 years. Over one- half of the
representatives of industry associations referred specifically to the role
of China in the maquiladoras* decline. One maquiladora spokesman, for
example, suggested that China*s entrance into the World Trade Organization
(WTO) 18 has made that country a more attractive choice for foreign direct
investment, while foreign investment in Mexico*s maquiladoras has
decreased.

Among the major suppliers of imports to the United States, Mexico ranked
second and China third in 2002. As figure 11 illustrates, both Mexico and
China experienced significant growth in exports to the United States from

1995 to 2002. However, between 2000 and 2002, U. S. imports from Mexico
grew at a slower pace than those from China. As a result, the gap between
Mexico and China narrowed in China*s favor. 17 The figure for 2002 is
preliminary, based on data available through September.

18 China formally joined the WTO in December 2001.

Figure 11: Value of U. S. Imports from Mexico and China, 1995* 2002

As appendix III details, Mexico recently lost market share in 47 out of
152 major U. S. import categories. At the same time, China gained U. S.
market share in 35 of those 47 import categories, including toys,
furniture, electrical household appliances, television and video equipment
and parts, and apparel and textiles. Some of these industries represent
significant

sectors of maquiladora production. Recent International Trade Commission
(ITC) staff research suggests that while Mexico does face increased
competition from China in the U. S. market, some sectors are more
threatened than others. 19 According to the 19 Ralph Watkins, U. S.
International Trade Commission, *Mexico Versus China: Factors

Affecting Export and Investment Competition,* Industry Trade and
Technology Review,

USITC, pub. 3534, July 2002, p. 11ff.

ITC staff research, a growing share of some textiles and apparel products
sold in the United States are being produced in China rather than Mexico.
In contrast, this staff research notes that within the machinery sector,
the

data did not indicate a shift in competitiveness away from Mexico towards
China. Mixed results are apparent in the electronic products sector.
Mexico lost U. S. market share to China in the telephone and telegraph
equipment segment in both 2001 and 2002, and Mexico*s gain in the computer
hardware segment in 2001 was more than offset by a sharp loss to China in

2002. The ITC staff research noted above concludes that China has
competitive advantages over Mexico in terms of labor costs, electricity
costs, and diversity of component suppliers. In this context, it is worth
noting that wages along the U. S.- Mexico border, where the maquiladoras
are concentrated, tend to be higher than in other areas of Mexico. 20 More
recent ITC staff research indicates that the cost of water for industrial
uses

(important in the textiles industry) and corporate income tax rates are
lower in China. On the other hand, the ITC staff research suggests that
Mexico*s comparative advantages include lower transportation costs,
shorter transit time, and lower international communication costs. Mexico
also provides greater protection for intellectual property, more
transparency in regulation and administration, and a network of free-
trade agreements with third countries.

Several industry representatives noted that Mexico also faces increased
competition from countries in Central America or the Caribbean. One
industry spokesperson noted that the U. S. decision in May 2000 to grant
NAFTA- parity access to Caribbean Basin Initiative (CBI) countries had
eroded Mexico*s ability to compete in the U. S. apparel market,
particularly because a number of Central American and Caribbean countries
have

lower labor costs than Mexico. According to a Mexican economic research
group, manufacturing wages in Mexico are almost 67 percent higher than in
the Dominican Republic and about 92 percent higher than in Honduras. 21 20
Average wages in Mexico*s six border states are higher than in other
regions of Mexico except the central area around Mexico City.

21 Actual figures: Mexico: $2. 5/ hour; Dominican Republic: $1.5/ hour;
Honduras: $1.3/ hour. Source: *Perspectives of the Maquiladora Industry in
the Mexican Economy* (Perspectivas de la Industria Maquiladora en La
Economia Mexicana), Center for Analysis and Economic Projections of Mexico
(Centro de Analisis y Proyecciones Economicas de Mexico), December 9,
2002.

The heightened global competition from China and CBI countries is part of
a larger phenomenon in which the benefits enjoyed by maquiladoras and
other Mexican producers have eroded as U. S. trade preferences or
liberalization accorded to other countries have expanded. The recent
experience of the Mexican textiles and apparel industry, one of the major
maquiladora sectors, illustrates this point. In 1994, NAFTA gave Mexico
preferential access for its textiles and apparel. Other countries* exports
to the United States and Canada generally did not receive similar
advantages. 22 U. S. imports of Mexican textile and apparel products grew
rapidly, with Mexico*s share of total U. S. imports in this sector
doubling from 7 percent in 1994 to 14 percent in 2000 (see fig. 12).
Mexico surpassed both China and the Caribbean Basin countries to become
the largest supplier to the U. S. market. However, under the Trade and
Development Act of 2000, the United States allowed textile and apparel
products from Caribbean Basin countries that met certain requirements to
receive

preferential access to the U. S. market. This legislation also stipulated
that to benefit from the special treatment, CBI- based apparel operations
must use U. S.- made inputs, and, according to a Mexican textile industry
association, operations have shifted from using Mexican textiles. In
addition, under the WTO Agreement on Textiles and Clothing, all quotas on
textile and apparel products are being phased out by 2005. For some
products quotas have already been removed. Despite the recent U. S.
recession and a decline of total U. S. imports of textiles and apparel by
5

percent between 2000 and 2002, U. S. imports of textiles and apparel from
China rose 12 percent, making China again the largest foreign supplier to
the U. S. market. Figure 12 shows these changing patterns of U. S. imports
in textiles and apparel from Mexico, China, and CBI countries.

22 U. S. nonreciprocal trade programs, such as the Generalized System of
Preferences for developing countries and CBI for Caribbean and Central
American trade partners, generally excluded textile and apparel from
special preferences. Canada is an exception, also receiving benefits
through NAFTA.

Figure 12: U. S. Imports of Textiles and Apparel from Mexico, China, and
Caribbean Basin Countries, 1990* 2002

Note: Caribbean Basin Countries are those eligible for the Caribbean Basin
Trade Partnership Act preferential access. These countries are Antigua and
Barbuda, Aruba, Bahamas, Barbados, Belize, British Virgin Islands, Costa
Rica, Dominica, Dominican Republic, El Salvador, Grenada, Guatemala,
Guyana, Haiti, Honduras, Jamaica, Montserrat, Netherlands Antilles,
Nicaragua, Panama, St. Kitts and

Nevis, Saint Lucia, Saint Vincent and the Grenadines, and Trinidad and
Tobago.

Industry Representatives Many industry representatives whom we contacted
also called attention to

Blame Strong Peso for Loss the role of the strengthening Mexican currency
in eroding the

of Competitiveness maquiladoras* competitiveness. Historically, growth
periods in the

maquiladoras have been associated with devaluations of the peso. For
example, after the peso was devalued in 1984, there was a 3- year surge in
U. S. automotive industry investments in maquiladora plants. Similarly,
according to a study by the El Paso Branch of the Federal Bank of Dallas,

the peso devaluation in December 1994 played a key role in spurring the
expansion of Mexico*s maquiladoras during the second half of the past

decade. 23 However, beginning in the last quarter of 1998, the Mexican
peso consistently appreciated against the dollar in real terms, a trend
that continued while the maquiladoras experienced their greatest
employment decline, from the end of 2000 to the beginning of 2002 (see
app. II for the

relative dependence of maquiladora employment on the real peso exchange
rate). As the peso appreciated in real terms, maquiladora operating
expenses increased. Moreover, this real appreciation of the Mexican peso
took place as the currencies of some of Mexico*s East Asian competitors
were depreciating against the dollar. For example, figure 13 compares the
performance of the Chinese yuan to the Mexican peso, in real terms,
between 1995 and 2002. Unlike the peso, the yuan has actually depreciated
since early 1998.

23 William C. Gruben, Federal Reserve Bank of Dallas, *Was NAFTA Behind
Mexico*s High Maquiladora Growth?* Economic and Financial Review, Third
Quarter 2001.

Figure 13: Real Dollar Exchange Rate of Mexican Peso and Chinese Yuan,
1995* 2002

Changes in Mexican Tax Among industry groups whose members had experienced
losses in

Policies Raise Investor employment or production, about two- thirds of
those we interviewed

Uncertainty indicated that uncertainty resulting from Mexican government
tax policies

was a major factor in the maquiladoras* decline. These groups noted that
such uncertainty had caused some firms to withdraw from, or downsize,
their operations in Mexico and had also discouraged new foreign direct
investment in Mexico. In particular, industry representatives said that
frequent changes to the fiscal regime had increased the tax burden and
administrative costs to maquiladoras. They were also concerned that the
frequent changes reduced the maquiladoras* ability to develop long- term
investment plans.

In addition to the duty- free treatment on import of parts, components,
and other inputs, maquiladora plants enjoyed, at least until the mid-
1990s, a virtual freedom from taxation. Though legally subject to income
taxes, in practice, the companies paid only a small assets tax, a flat
minimum of 2 percent of the value of the maquiladora*s assets. Moreover,
the maquiladoras were permitted to use the cost of wages to offset their
tax on assets. This virtually eliminated taxes for some maquiladoras.
According to experts, the twin benefits of duty- free import and minimal
taxation were primary incentives for foreign firms to establish
manufacturing operations

in Mexico. The tax regime applicable to maquiladoras remained constant for
almost 30 years but began to evolve rapidly in the 1990s. The most
significant of these tax changes, the treatment of what are known as
*permanent

establishments* is frequently noted by industry groups and others as a
cause of investor uncertainty about the industry. A permanent
establishment typically is a branch of a company from one country that is
doing business in another *host* country, and which may be

taxed in that host country. According to U. S. Treasury officials,
permanent establishment is a concept found in virtually all double
taxation treaties. Mexico adopted the permanent establishment concept as
part of its income tax law in 1981. According to U. S. Treasury officials
and Mexican tax experts GAO consulted, Mexico essentially exempted
maquiladoras from

the tax that could be imposed on permanent establishments until 1998.
However, starting in 1998, Mexico began seeking to treat the foreign
parent companies of maquildoras as having permanent establishments in
Mexico for tax purposes. By treating the maquiladoras as permanent
establishments, the Mexican government could subject the foreign parent
companies to taxation, potentially allowing Mexico to increase the
revenues it collects from maquiladora operations.

The right of the Mexican government to tax maquiladoras as permanent
establishments was affirmed in the U. S.- Mexico tax treaty of 1992.
However, U. S. companies with maquiladora operations in Mexico were
concerned that Mexico*s application of permanent establishment to their
maquiladora operations would subject them to double taxation. This could
occur if Mexico imposed a broad definition of how permanent establishments
could be taxed that the U. S. Treasury would not accept, because it would
prevent the U. S. parent company from getting a full credit

in the United States from the taxes actually paid in Mexico. Resolution of
potential problems, such as double taxation, associated with the treatment

of maquiladoras as permanent establishments has necessitated a series of
additional bilateral agreements between the United States and Mexico. It
took several years and several different iterations to finally resolve
such practical problems, and this caused a prolonged period of uncertainty
for maquiladoras. Other changes in Mexico*s tax regime have contributed to
the climate of investor uncertainty. In 2002, for example, Mexico limited
the ability of businesses, including maquiladoras, to take a tax credit on
salaries. According to industry representatives, this provision could have
significantly increased the tax burden on some maquiladoras. However,
according to Mexican officials, this tax provision has subsequently been

ruled unconstitutional. NAFTA Phased Out Some The phasing out of
maquiladora benefits as part of NAFTA was also cited by Maquiladora
Benefits

industry associations as a major factor in the decrease in maquiladora
production and employment. When NAFTA was signed in 1993, it envisioned
fundamental changes to the maquiladora model. The most significant of
these changes was embodied in Article 303 of NAFTA, which eliminated duty
drawback (or refunds of duties) 24 for inputs of non- NAFTA origin as of
January 1, 2001, if the final products incorporating these inputs

are to be subsequently exported to another NAFTA country. For various
reasons, notwithstanding the 7- year grace period provided, the
maquiladoras did not develop a network of domestic suppliers in Mexico. As
a result, implementation of Article 303 has adversely affected the
competitiveness of maquiladoras that rely on non- NAFTA suppliers for
inputs and resulted in closure of some maquiladora firms.

According to officials with the Office of the U. S. Trade Representative,
some aspects of the Maquiladora program were not consistent with NAFTA*s
trade objectives. For example, the duty drawback provisions of the
Maquiladora program were in conflict with NAFTA*s rules of origin
requirements. Under NAFTA*s rules of origin, goods traded among NAFTA
partners are allowed duty- free status only when the goods comprise a
minimum percentage of North American content. However, the Maquiladora
program provided duty drawbacks for inputs imported to Mexico from any
source, including non- NAFTA countries, undermining the duty- free
benefits that North American products were to receive in Mexico as a
result of NAFTA. Second, such drawback programs represented an

24 *Duty drawback* refers broadly to the refund, waiver, or reduction of
customs duties owed on imported goods, on condition that the goods are
subsequently exported.

advantage for exporters versus firms involved in production for the
domestic market, since the latter would not receive an equivalent duty
drawback. In negotiating NAFTA, the United States hoped to reverse this
advantage, which led to the development in Mexico of an economic system
with separate production tracks for exports and for goods destined for
domestic consumption. In fact, U. S. officials explained that they
envisioned the gradual phasing- out of maquiladoras with the
implementation of NAFTA, as duty- free treatment would apply to all trade
among NAFTA member countries.

The rationale behind Article 303 was to encourage firms to develop North
American suppliers for critical inputs by providing an incentive for
maquiladoras to shift sourcing of components or inputs to North America,
including Mexico. The development of a network of North American

suppliers would mean that more value would be added during the production
process in Mexico, the United States, and Canada. The elimination of duty
drawback would necessitate significant changes in the sourcing of
maquiladora inputs, particularly for maquiladora operations of some
Japanese and other Asian companies that were heavily dependent on

certain inputs from the Far East. The implementation of Article 303 was
therefore scheduled for January 1, 2001, 7 years after NAFTA*s entry into
force, to allow the maquiladoras to relocate their supply chain to North
America. However, a network of Mexican domestic suppliers for the
maquiladoras largely failed to materialize during this period. Maquiladora
observers have suggested several explanations, principally the scarcity of
credit in Mexico to support entrepreneurial activity and the lack of an

entrepreneurial culture among Mexican businesses. Under NAFTA, Mexico
could have chosen to counter the loss of duty drawback following
implementation of Article 303 by reducing or eliminating its most favored
nation duties on key inputs. U. S. officials note that Canada eliminated
hundreds of its most favored nation duties before Article 303 took effect.
Instead, in order to cushion the impact of NAFTA Article 303, Mexico
instituted a measure known as the sectoral promotion program with targeted
and reversible tariff reductions. 25

Since Article 303 was implemented, maquiladoras that depend on inputs from
outside North America have seen their competitiveness erode. Some
maquiladoras have reported production cost increases of up to 20 percent

25 See discussion of the sectoral promotion program (PROSEC) below.

due to the implementation of Article 303. Japanese, Korean, and Taiwanese
companies involved in maquiladora production have been particularly
affected by the implementation of Article 303 and have led the way in
relocating from Mexico to other countries. Industry associations we
contacted, representing maquiladoras in the Tijuana area, where Asianowned
maquiladoras are concentrated, as well as an association representing
Japanese business in Mexico, attributed the departure of maquiladora firms
from Mexico, at least in part, to the implementation of Article 303.

Maquiladora Downturn Significant challenges continue to confront Mexico*s
maquiladoras,

Spurs Some Positive although recent industry and government action and the
prospect of future

Mexican reforms may bolster prospects for maquiladoras* recovery. The
Changes, but

downturn during the past 2 years has accelerated ongoing industry
Fundamental

evolution and has been a catalyst for several industry and government
Challenges to Future

changes to improve the competitiveness of the sector. However,
maquiladoras still face fundamental challenges. For the most part, meeting
Viability Remain

these challenges depends on further action by the government of Mexico,
but some of the challenges are related to U. S. policies that are likely
to put additional pressure on maquiladoras.

Maquiladoras Face Serious The factors described in the previous section as
having a role in the

Challenges and Questions maquiladora*s recent decline still confront the
industry. As a result, some

about Viability Mexican government officials have stressed the need to
move beyond the

current *maquiladora model* to attract a new generation of more
technologically advanced operations that would allow Mexico to remain
competitive. Given the continuous evolution of maquiladora operations,
Mexico*s maquiladora industry is now a complex sector with substantial
diversity. One academic expert concludes that as firms become involved in
more

sophisticated, capital- intensive operations, they are less likely to
close and move their plants because of cyclical downturns such as the one
maquiladoras faced after 2000. Some Mexican maquiladoras are now
recognized as having sophisticated production and management methods.
According to industry experts, such maquiladoras are better positioned to
weather the maquiladora downturn and deal with continuing challenges.

Nevertheless, researchers point out that the transition to more advanced

production practices is quite uneven. 26 Many maquiladoras remain oriented
toward lower- skill activities that involve few Mexican inputs besides
labor. The downturn of the last several years has resulted in a shake- out
involving

some losers, notably among this type of operations. Downturn Catalyst for
One positive aspect of the recent maquiladora downturn is that it has
Industry and Government

spurred some actions by industry and the government of Mexico to restore
Efforts to Improve

Mexican competitiveness. In the face of increased global competition,
Competitiveness

maquiladoras are seeking to capitalize on Mexico*s unique competitive
advantages, particularly those associated with that country*s proximity to
the United States and its growing network of free trade agreements. For

example, noting the recent establishment of plants in Juarez by several
computer manufacturing firms, one industry analyst explained that Mexico*s
quick time- to- market location is essential for the success of both new
products as well as repairs in the computer value chain. Similarly, a
senior industry expert noted that the growth of automotive maquiladoras,
in northern and central Mexico, underscores the competitive advantages
resulting from the efficient combination of U. S. and Mexican inputs.
According to this source, notwithstanding the arrival of new competitors,
the Mexican automotive industry is poised to take advantage of the full
opening of the regional North American automotive market that will occur
in 2004. Mexico also stands to benefit as a direct and indirect automotive
sector exporter to the United States and other countries with which Mexico
has signed trade agreements.

Some industry sources reported unexpected benefits associated with the
recent losses experienced by the maquiladoras. According to industry
representatives in Juarez, the rapid pace of maquiladora growth had put
intense pressure on local infrastructure during the late 1990s. Local
authorities simply could not keep up with the demand for health,
education, and other services associated with the dramatic increases in
population growth that accompanied the expansion of maquiladoras. They

viewed the slowdown of the past 2 years as a welcome respite. In addition,
a number of industry representatives noted that the downturn has resulted
in significant drops in employee turnover and in the associated hiring and
training costs. Prior to the downturn, they said, maquiladoras in 26 James
Gerber and Jorge Carrillo, *Are Tijuana*s and Mexicali*s Maquiladora
Plants Competitive?* San Diego Dialogue, July 2002.

some border cities reported very high employee turnover rates because the
rapid growth in maquiladora establishments allowed workers to continuously
find new jobs in other plants. One expert suggested that such turnover in
some border cities had reached 80 percent at the height of the

maquiladora boom. Consequently, employers had significant hiring and
training costs and were forced to keep some positions overstaffed to
compensate for the turnover. This could have more fundamental implications
for the ability of some maquiladoras to build a highly skilled workforce,
since it is not feasible to invest in significant training for workers
whose expected tenure with a firm is only a few months. Industry sources
told us that the turnover rates had dropped sharply since the downturn,
and some maquiladoras report that this has had a positive effect on
administrative costs as well as the cost of training new employees.

Finally, industry sources stressed the importance of Mexican government
action for the development of a favorable business environment that can
respond quickly to changing market forces faced by maquiladoras. In
response to industry pressure, the Mexican government recently undertook

several measures in support of the maquiladoras, primarily aimed at easing
irritants.

 On May 12, 2003, the Mexican government issued a decree modifying
certain aspects of the Maquiladora program. 27 The reforms are aimed
primarily at simplifying regulations that apply to companies that provide
support and logistic services to maquiladoras, and to enhance legal
certainty for Mexican exporters, including maquiladoras. An important
provision of the new decree will be streamlined customs requirements for
companies with several subsidiaries operating under the Maquiladora
program. This would allow such companies greater flexibility in the
transfer of finished or semi- finished products from one subsidiary to
another. The decree also contains provisions that would reduce
administrative costs and procedures. For example, a maquiladoras will only
have to submit a single report on an annual basis, which can be submitted
electronically. Based on initial industry

reaction, it is unclear whether the new decree will satisfy critics
seeking greater legal certainty and improved incentives for maquiladoras.

27 Decree for the Development and Operation of the Maquiladora Export
Industry (Decreto que reforma el diverso para el fomento y operacion de la
industria maquildora de exportacion), Official Daily Gazette (Diario
Oficial), May12, 2003, Section 1, page 107.

 In response to the recent crisis in the maquiladora industry, Mexico has
greatly expanded its sectoral promotion program (PROSEC). First launched
in November 2000, PROSEC was intended to reduce the impact of NAFTA
Article 303 (which became effective January 1, 2001) by providing that
duty rates on imported inputs from non- NAFTA suppliers of either 0 or 5
percent. Initially, maquiladora industry

representatives complained that PROSEC was too restrictive because it
applied to very few imported inputs. However, throughout 2001 and 2002,
the list of products eligible for tariff reduction under PROSEC was
progressively expanded to include more than 16,000 products from 22
industry sectors, including electronics and textiles and apparel.

 In September 2002, the Mexican government provided additional support to
the maquiladoras through a program called the Information Technology
Agreement (ITA) Plus. 28 ITA Plus immediately removed tariffs from inputs,
parts, and components used in the electronic and high- technology sectors,
regardless of the country of origin. It also provided for the gradual
removal of tariffs from semifinished and finished products in those
sectors. According to Mexican officials, in addition to lowering tariffs
on electronic and high technology inputs,

ITA Plus may help to reduce the administrative burden on the maquiladoras.

 The Presidential Council for Competitiveness was created in July 2002 to
promote investment, increase employment, and accelerate Mexico*s economic
growth. A cooperative effort between government and

business that is chaired by the Minister of Economy, the council*s
activities include the creation of fiscal stimulus packages for export
factories in twelve different sectors of the economy, including
maquiladoras as part of the in- bond industry. One objective of the
council is the development of manufacturing clusters, which will deepen
the supply chain in Mexico. In support of the work of the council, the
Secretariat of the Economy has agreed to fund, through the National
Council of the Maquiladora Export Industry (CNIME), a comprehensive study
on the maquiladora industry.

28 The Information Technology Agreement (ITA) was negotiated under the
auspices of the WTO and concluded in 1996, eliminating tariffs on
information technology products by the participating members. Mexico has
not joined the WTO ITA, but chose to unilaterally eliminate tariffs on
certain information technology products in its ITA Plus initiative.
However, because Mexico did not commit to these tariff cuts in the WTO, it
may change them at any time.

 Recent agreements between the United States and Mexico have largely
resolved the threat of double taxation of U. S. firms that was raised by
Mexico*s efforts to define maquiladora parent companies as permanent

establishments, discussed above. As a result of a Second Additional
Protocol to the U. S.- Mexico tax treaty, signed in 2002, the United
States will be able to provide a foreign tax credit to U. S. firms that
have paid income taxes to Mexico with respect to their maquiladora
operations. Mexico has also independently announced that it will make no
changes

to existing agreements on permanent establishment until 2007. These steps
by the Mexican government seem to reflect wider recognition by officials
in Mexico City of the maquiladoras* importance to Mexico. Industry
representatives complained that the Mexican government was slow to respond
to the challenges faced by the maquiladoras. According to these
representatives, the Mexican government initially took *a wait and

see* approach to the maquiladoras decline, in the belief that labor-
intensive maquiladora operations leaving Mexico would be readily replaced
by better paid, more profitable industries. As job losses continued in the
first three months of 2002, maquiladora representatives pressured the
government to implement remedial measures.

Future Challenges Remain, Notwithstanding the initiatives discussed above,
government, industry and

Involve Difficult Reforms academic sources suggest that meeting remaining
challenges to the future

success of the maquiladoras will, in some cases, require fundamental
Mexican reforms in several areas, including energy, infrastructure and
labor. However, the initiatives Mexico is pursuing in these areas may be
difficult to bring about.

Some Consider Energy Reform Government officials and industry
representatives stated that there is an

Vital to Mexico*s urgent need for energy reform in Mexico. Energy sector
reform is Competitiveness

important to the maquiladora industries because they require reliable and
competitive energy prices to compete with suppliers in other nations. The
ITC, for example, has noted that electricity and industrial water costs
are two areas in which Mexico is less competitive than China. The Fox
administration maintains that without energy reform, Mexico may experience
a power crisis as early as 2004, and it introduced an energy reform bill
in August 2002. The legislation stalled in the Mexican Congress, however,
because some legislators opposed aspects of reform dealing with

privatization that would entail amending the Mexican constitution.

Upgrading and Modernizing Maquiladora and other Mexican industry
associations cite improving

Mexican Infrastructure Is Mexico*s infrastructure as critical to advancing
Mexico*s competitiveness.

Critical According to a report by the Mexican Government Commission for
Border

Affairs, the six Mexican states that border the United States share the
advantage of an adequate basic infrastructure, with a road network
variously described as good, fluid, or satisfactory. However, even in this
region, about 32 percent of the Mexican federal highways are in poor
condition. Another study found that critical problems persist in Mexico*s
road infrastructure, notably, limited public or private investment in

highways in recent years. Some maquiladora representatives we spoke with
cited infrastructure shortcomings as a disincentive for potential
investors in maquiladoras. Need for Mexican Labor Reform

According to Mexican labor officials, as part of its platform to modernize
Acknowledged

Mexico and improve its international competitiveness, the government has
sought to reform the labor code. Maquiladora representatives stated that
improvements in labor productivity depend on reform of labor regulations
to provide increased flexibility to employers. The Fox administration has
responded to this need for labor reform by developing a labor reform
package that represents a compromise between labor groups, business, and
government. Key elements of the reform package include the use of secret
ballots in union elections, the allowance of more than one union to
represent worker interests, expanded employer flexibility to hire workers

on a trial basis, and a strengthened binding arbitration process. This
reform package was not passed by the Mexican Congress before congressional
elections were held in July 2003, in part because it lacked consensus
support within the Mexican labor movement.

Worker Skills in Mexico Must Be A consultant for the maquiladora industry
cites worsening shortages of

Improved trained labor in most cities where maquiladoras are concentrated
as among

the challenges confronting the industry that the government must address.
One academic study 29 of the maquiladoras* viability found that to develop
more technology- intensive operations, Mexico needs a large number of

highly educated workers. However, according to the Commission on Border
Affairs, the data indicate a low level of educational attainment in the
economically active population along the border, with over one- third of
adults having completed only primary education or less. The search for

29 John Sargent and Linda Matthews, The University of Texas- Pan American,
Center of Economic Studies, Boom or Bust: Is it the End of Mexico*s
Maquiladoras? Working Paper #2002- 6, August 2002.

better educated workers has led a number of companies to establish
assembly plants in cities further from the border, with better reputations
for good public secondary education and trade schools. 30

U. S. Policies May Exert Action by Mexico is key to the maquiladoras*
future viability, particularly

Additional Pressure on since U. S. approaches to trade liberalization and
homeland security may

Maquiladoras put additional pressure on maquiladora operations. Industry

representatives noted that present U. S. policies in these areas could
undermine current benefits and reduce future competitiveness.

U. S. Trade Liberalization Could Regarding U. S. trade policy, the future
development of the maquiladora

Affect Maquiladora Development industry in Mexico may also be affected by
further changes in competitors* access to the U. S. market. The United
States is engaged in trade

negotiations in several venues, including the Doha Round among the 146
members of the WTO, the Free Trade Agreement of the Americas (FTAA)
involving 34 nations of the Western Hemisphere, and the U. S.- Central
America Free Trade Agreement. These negotiations may reduce barriers to
non- NAFTA countries* products to levels similar to those enjoyed by NAFTA
participants, Mexico, and Canada. For example, in the WTO, the United
States has proposed to eliminate all industrial tariffs by 2015, and in
the FTAA, the United States has proposed to phase out textile and apparel
tariffs within 5 years after the agreement is implemented, if its
hemispheric partners reciprocate. As we concluded in a 2001 report, 31
expansion of trade benefits to wider numbers of competitors, while
benefiting U. S. consumers and other trade partners, dilutes the benefits
of prior trade preferences. Some business association representatives that
we interviewed expressed concern that future U. S. trade agreements would

erode benefits provided to Mexican suppliers in the U. S. market under
NAFTA. Representatives for one industry association expressed hope that
the United States would use negotiations such as the FTAA to strengthen
regional competitiveness relative to global competitors such as China.

30 For a discussion of the trend for new maquiladora investments to be
located in the interior of Mexico see Ruben Mata, U. S. International
Trade Commission, *Recent Developments in Mexico*s Assembly Industry,*
Production Sharing: Use of U. S. Components and Materials in Foreign
Assembly Operations, 1994- 1997, USITC, pub. 3146, Dec. 1998, pp. 2- 10ff.
31 U. S. General Accounting Office, International Trade: Comparison of U.
S. and European Union Preference Programs, GAO- 01- 647 (Washington, D.
C.: June, 2001).

U. S. Homeland Security Maquiladora industry experts also expressed
concern that U. S. security

Measures Could Slow CrossBorder measures instituted at ports of entry
after September 11, 2001, could erode

Movement of Goods and the Mexican maquiladora industry*s advantage of
proximity to U. S. Personnel

markets. Of particular concern are U. S. government measures that require
advance notice for transborder shipments of goods and additional
information on the entry into and departure from the United States of
every foreign citizen. 32 Companies that use just- in- time operations, an
important

element in some maquiladora operations, could be especially hurt by
requirements related to advance notice for shipments, because they could
not ship goods immediately on receiving an order. 33 Firms that rely on
regular and efficient movement of workers and service operations across
the border could be especially affected by the information requirements
for Mexican workers who cross the border frequently. For example, at one
major border crossing in downtown El Paso, less than a mile from
Interstate 10, significant congestion would result if U. S. authorities
had to screen traffic bound for Mexico to obtain information from every
departing alien. Successful implementation of these new requirements will
require

close coordination of U. S. and Mexican national and local officials as
well as adaptation of the private industry to the new requirements.
Conclusion Both the United States and Mexico have an interest in the
future of maquiladoras given their central role in U. S.- Mexico trade and
the border

economy. Partly driven by maquiladoras, Mexico has assumed a more
prominent place among U. S. trade partners in recent years, becoming the
United States* second leading trading partner, after Canada. Moreover,
production and employment linkages have developed between 32 Two measures
regarding advance notification of cargo shipments are cause for industry

concern: (1) an informal U. S. Customs proposal that would require trucks
to declare the contents of their cargo 4 hours before they enter the
United States and 24 hours before they enter Mexico, which falls under the
Advance Electronic Information provision of the Trade Act of 2002 and has
not yet gone into effect and (2) a U. S. customs measure known as the 24-
hour rule, effective since December 2, 2002, which requires ships
traveling to U. S. seaports to declare the contents of their cargo 24
hours before they depart from a foreign port * 19 CFR 4.7( b)( 2).
Regarding entry of foreign citizens, at issue is an Immigration and
Naturalization Service mandate that is part of Section 110 of the Illegal
Immigration Reform and Immigrant Responsibility Act of 1996.

33 Customs officials noted that they are keenly aware of the importance of
*Just in Time* delivery and have taken that into account in any programs
proposed for cargo clearance. Customs also intends to offer a program
known as Free and Secure Trade to speed the clearance of known shippers,
importers and carriers, and assist in moving traffic borderwide.

maquiladoras and producers throughout the United States and are based on
the high volume of U. S.- generated components used in maquiladora
operations. Businesses in communities on the U. S. side of the border
provide services to the maquiladoras, such as customs brokerage and
commercial transportation. Retail sales to Mexican citizens in U. S.
border communities contribute substantially to U. S. business and tax
receipts. The decline in Mexico*s maquiladora production and employment
has already taken its toll on cross- border trade and trade- related
employment in certain U. S. border communities. Maquiladoras have become
an even more important element of the Mexican economy, particularly over
the decade of

the 1990s, when maquiladora growth propelled Mexico into the ranks of the
world*s leading exporters and generated 900,000 new jobs. Employment
created by maquiladoras on the Mexican side of the border has become a
mainstay of economic activity in that country. The decline over the past 2
years has served as a catalyst for further transformation of the industry,
as well as Mexican industry and government efforts to restore
competitiveness. The challenges still confronting maquiladoras and the
pressure from U. S. trade and homeland security policies lend urgency to
Mexican efforts to create an environment where cross- border links between
U. S. and Mexican firms and communities can continue to prosper.

Agency Comments and We provided a draft of this report for comment to five
U. S. government

Our Response agencies: Department of State, the Office of the U. S. Trade
Representative,

U. S. Customs and Border Protection (formerly U. S. Customs), Department
of the Treasury, and the U. S. International Trade Commission. We also
asked for comments from three Mexican government agencies: the Ministry of
the Economy (Secretaria de Economia) the Ministry of the Treasury
(Secretaria de Hacienda), and the National Institute of Statistics,
Geography and Information Technology (Instituto Nacional de Estadistica,
Geografia e Informatica). We received informal written comments from all
of these U. S. and Mexican government agencies, except Mexico*s Ministry
of Economy. In addition, the Department of State provided formal written
comments, which are reprinted in appendix VII.

In general, all of the agency comments were technical or editorial in
nature, which we incorporated as appropriate in the text of our report. In
addition, U. S. ITC staff had more extensive comments related to our
decision to

exclude firms operation under the so- called PITEX program from the
general scope our work, noting that PITEX firms are important in certain
sectors, such as autos, and account for a substantial share of Mexico*s
total exports to the United States. While we recognize that firms
operating under

PITEX are an important element in U. S- Mexico production- sharing
operations, as are maquiladoras, we limited our report to the Maquiladora
program for several reasons. First, our requesters specifically expressed
an interest in the maquiladora industry and the effects of the recent
decline of the maquiladoras along the U. S.- Mexico border. Unlike
maquiladoras, which are still concentrated along the border, firms
operating under the PITEX program are spread throughout Mexico. Secondly,
the data the government of Mexico collects on maquiladoras are
significantly more extensive and are not altogether comparable to the data
collected on PITEX firms. Thus, there would have been problems in
comparing the two

types of operations. Finally, the data available on PITEX firms suggest
that they have experienced trends in recent years not unlike those
observed among maquiladoras. Including data on PITEX firms would not have
significantly altered our message.

We are sending copies of this report to other interested members of
Congress, the Secretary of State, the Secretary of the Treasury, the U. S.
Trade Representative, the Secretary of the Department of Homeland
Security, the Commissioner of Customs, and the Chairman of the U. S.
International Trade Commission. We will also make copies available to
others upon request. In addition, the report will be available at no
charge on the GAO Web site at http:// www. gao. gov.

If you, or your staff, have any questions about this report, please
contact me on (202) 512- 4347. Other GAO contacts and staff
acknowledgments are listed in appendix VIII.

Loren Yager Director, International Affairs and Trade

Appendi xes Structure of Employment Growth in the U. S. Mexico

Appendi x I

Border Area This appendix examines U. S. employment changes along the U.
S.- Mexico border and explores whether employment in the border areas of
the United States has been disproportionately affected by the recent
slowdown in U. S. economic activity and the associated decline in cross-
border trade between the United States and Mexico. For the purpose of this
analysis, the U. S. border with Mexico is defined as the metropolitan
statistical areas (MSA)

closest to the U. S.- Mexico border, comprising the MSAs for San Diego,
California; Tucson, Arizona; Las Cruces, New Mexico; and El Paso,
Brownsville, Laredo, and McAllen, Texas. 1

U. S. employment in the border area increased by approximately 591,000
jobs between 1990 and 2002, largely owing to the overall national trend in
employment growth. For example, according to our analysis, 60 percent of
the jobs gained were due to the growth of the national economy. However,
230,000 of those jobs could be linked to local factors, that is, factors

associated with the area*s attractiveness for employment creation. Most of
the new jobs were added from 1995 to 2002. However, the ways in which each
border subregion benefited from the employment growth vary considerably.

U. S. Border U. S. employment in the U. S.- Mexico border area grew by 35
percent

Employment Outpaced between 1990 and 2002, gaining 591,000 jobs. The
services sector was the

largest employer and accounted for approximately 48 percent of the job
Nation*s since 1995

growth (282,000 jobs) during this period. Other sectors with notable
employment growth were retail trade (93,000 jobs); finance, insurance, and
real estate (20, 000 jobs); transportation and public utilities (31,000
jobs); and government (128,000 jobs). As figure 14 shows, total nonfarm

employment growth rates in the border region were generally similar to
those observed for the United States from 1993 to 1995. However,
employment growth in the border MSAs exceeded employment growth at

1 The U. S. border with Mexico is defined by the states of California,
Arizona, New Mexico, and Texas. However, a meaningful description of the
border would require the exclusion of large portions of each of these
states. Many analysts define the border in terms of the contiguous
counties that have direct geographical links with Mexico. According to
this definition, the U. S.- Mexico border consists of the counties of San
Diego and Imperial in California; the counties of Yuma, Pima, Santa Cruz,
and Cochise in Arizona; the counties of Hidalgo, Luna, and Dona Ana in New
Mexico; and the counties of El Paso, Hudspeth,

Culberson, Jeff Davis, Presidio, Brewster, Terrell, Val Verde, Kinney,
Maverick, Dimmitt, Webb, Zapata, Starr, Hidalgo, and Cameron in Texas.

the national level after 1995. 2 Furthermore, growth of nonfarm employment
in the border area continued even after the U. S. economic slowdown began
in 2001. Laredo and McAllen grew fastest, followed by Brownsville, Tucson,
Las Cruces, and San Diego. Figure 14: Nonfarm Annual Employment Growth in
the United States and in U. S. Metropolitan Statistical Areas at the U. S.
Mexico

Border, 1991- 2002

Some border industries experienced a decline in employment in 2001 and
2002, particularly manufacturing (down 6 percent), transportation and
public utilities (down 4 percent), and wholesale trade (down 3 percent)
(see table1). As table 1 shows, declines in manufacturing were relatively
more severe in Texas (down an average of 18 percent), while declines in
wholesale trade and transportation and public utilities were more

2 Given that NAFTA was implemented in 1994, the graph suggests that NAFTA
had an employment- stimulating effect in the border counties.

pronounced in Arizona (down 9 and 11 percent, respectively). A closer look
at Texas further shows that the manufacturing, transportation, and public
utilities sectors declined after 2000 in all four Texas border MSAs.

Table 1: Employment Growth of the United States and U. S. Metropolitan
Statistical Areas at the U. S. Mexico Border by Industry: 1990* 2002 and
2000* 2002

Finance, Construction Transportation

insurance, Total & Manufac & public Wholesale

Retail & real

nonfarm mining turing

utilities trade trade estate Services Government

United 2000- States 2002 -1% -1% -9% -4% -4% 0% 2% 2% 3%

1990- 2002 20% 22% -12% 17% 8% 19% 16% 47% 16%

US- Mexico 2000- Border 2002 4% 5% -6% -4% -3% 7% 6% 5% 8%

1990- 2002 35% 48% -7% 43% 13% 27% 21% 65% 37%

San Diego, 2000- CA 2002 5% 9% -1% -1% -1% 8% 4% 6% 8%

1990- 2002 30% 46% -5% 40% 14% 22% 14% 59% 26%

Tucson, 2000- AZ 2002 2% -1% -4% -11% -9% 3% 7% 0% 9%

1990- 2002 42% 43% 21% 14% 32% 24% 35% 60% 49%

Las Cruces, 2000- NM 2002 5% 0% 3% -5% 8% 6% 0% 8% 6%

1990- 2002 37% 50% -8% 31% 56% 35% 25% 111% 14%

Brownsville, 2000- TX 2002 2% 7% -19% -4% 7% 6% 0% 5% 6%

1990- 2002 47% 96% -14% 66% 22% 38% 5% 93% 53%

El Paso, TX 2000- 2002 -1% -2% -17% -10% -9% 4% 12% 1% 8%

1990- 2002 22% 46% -24% 31% -4% 26% 35% 43% 38%

Laredo, TX 2000- 2002 6% -6% -22% -5% -10% 14% 11% 10% 13%

1990- 2002 62% 3% -22% 82% 12% 45% 63% 106% 81%

(Continued From Previous Page)

Finance, Construc Transportation

insurance, Total tion & Manufac & public Wholesale

Retail & real

nonfarm mining turing

utilities trade trade estate Services Government

McAllen, TX 2000- 2002 9% 4% -18% 0% 0% 9% 11% 19% 12%

1990- 2002 69% 92% -25% 120% 11% 55% 56% 165% 71% Source: GAO calculations
using U. S. Department of Labor, Bureau of Labor Statistics data.

Analysis of Employment To analyze the factors at the national and local
levels that contributed to

Trends the employment trends described above, we employed a methodology

known as shift- share analysis that decomposes employment growth (or
decline) in a region over a given time period into three components: the
national growth effect, the industry- mix effect, and the local
(competitive) effect.

1. National growth effect. The national growth effect is that part of a
regional change in total employment ascribed to the national growth rate
of total employment. It assumes that the region*s employment growth
matches the overall national rate. The national growth component is the
change that would be expected given that the local area is part of a
changing national economy. 3 Our analysis shows that from 1990 through
2002, the border counties

gained 339,100 jobs due to economic trends at the national level (see
table 2). However, the actual gain occurred prior to the year 2000 as an
estimated 15,800 jobs were lost due to the national trend in 2001 and

2002. The border area*s biggest employer, the service sector, had the
highest national growth component (97,300 jobs), followed by the
government (71, 200 jobs), and retail trade sectors (65,900 jobs). Our
analysis incorporating possible differences among the border subregions
shows that from 1990 through 2002, nonfarm employment growth in San Diego
accounted for nearly 50 percent of the increase in employment due to
employment expansion at the national level.

3 For example, during the time period 1990* 2000, nonfarm employment in
the United States grew by 20 percent (i. e., from 109.4 million to 130.8
million). Therefore, the national growth component of any region within
the United States during this period would be 20 percent of the the
region*s 1990 employment.

2. Industry- mix effect. An industry- mix effect is the amount of change
that a region would have experienced had each of its industries grown at
their industry national rates, less the national growth effect. This
component identifies the share of local job growth that can be attributed
to the region*s mix of industries and seeks to address

whether employment growth in an area outpaced the nation owing to a
concentration of faster growing industries. For the period 1990 to 2002,
the border area gained 21, 200 jobs owing to a concentration of faster
growing sectors there than in the nation as a whole. This gain in total
employment was achieved primarily with employment gains in the services
(114,400 jobs) and construction and mining (4,200 jobs), and it occurred
despite employment losses totaling

95,200 jobs from other sectors, notably manufacturing (69,800 jobs),
government (10, 500 jobs), and wholesale trade (8,500 jobs). Moreover, 47
percent of the employment growth due to the industry- mix effect

occurred between 2001 and 2002. In subregions, the industrial mix
component for all sectors decreased total nonfarm employment during 1990*
2002 only in El Paso, Texas. 3. Local (competitive) effect. A local
(competitive) effect seeks to isolate the extent to which factors unique
to the local area have caused growth

or decline in regional employment. The effect is defined as the employment
change that remains after the national and industrial mix components have
been accounted for, and it is therefore the purely regional aspect of the
region*s employment growth. If a region*s competitive share is positive,
the region is considered to have local advantage in promoting employment
growth. This advantage could result from such factors as local businesses
having superior technology, management, location, market access or the
local labor force*s having higher productivity, lower wages, or both. A
negative competitive share component could be caused by local shortcomings
in any or all of these

aspects. Local conditions appear to have been a significant factor in the
increase in U. S. border employment, particularly since 1995. Across all
sectors, the competitive share component* employment growth attributable
to local conditions* totals to a net addition of 230,000 jobs. This
indicates that the border area was competitive in securing additional
employment from 1990 through 2002. As figure 15 shows, nearly all of these
employment gains were realized in the years since 1995. Furthermore, 43
percent of border area employment gains owing to

local factors were achieved between 2001 and 2002. The top three sectors
in competitive share gains in employment from 1990 through 2002 were
services (70,600 jobs), government (67,200 jobs), and manufacturing (37,
500 jobs). However, for the 2000* 2002 period, the transportation and
public utilities sector showed a reduction in jobs (approximately 300
jobs) owing to local factors. In addition, factors

unique to the local area caused employment declines in certain subregions
and sectors during 1990* 2002, notably, in Laredo, Texas, in construction
and mining; El Paso, Texas, in wholesale trade and services; Brownsville,
Texas, in finance, insurance, and real estate; Tucson, Arizona, in
transportation and public utilities; and Las Cruces, New Mexico, in
government employment. Furthermore, subregions in Texas generally lost
their local edge in securing manufacturing

employment from 1990 through 2002 and this loss was more pronounced in
2001 and 2002. Similarly, owing to local factors from 2001 to 2002,
Tucson, Las Cruces, and El Paso lost jobs in transportation and public
utilities; Tucson, El Paso, and Laredo lost employment in wholesale trade;
and El Paso lost service employment; and Brownsville and Las Cruces lost
employment in the finance, real estate, and insurance sector.

Figure 15: Employment Gains (Losses) in Nonfarm Employment in Metropolitan
Statistical Areas at the U. S. Mexico Border Due to National, Industry-
mix, and Local Effects

Table 2: Components of Employment Changes by Sectors in U. S. Metropolitan
Statistical Areas at the U. S. Mexico Border, 1990* 2002

Employment in thousands

Transpor Finance, Construc tation and

insurance, Total tion & Manufac public Wholesale Retail

and real nonfarm mining turing

utilities Trade trade trade estate Services Government Local effect

US- Mexico 2001Border 2002 98. 1 7.5 10. 8 (0.3) 31. 1 0.9 29. 4 3.6 23. 3
25. 1 1990

2002 230. 5 21. 6 37. 5 17.1 35. 4 4.3 29. 0 5.3 70. 6 67. 4 San Diego,
2001CA 2002 67. 0 6.9 10. 7 1.6 19. 5 1.3 17. 8 1.4 16. 1 10. 8 1990

2002 91. 2 14. 0 10. 4 8.2 12. 6 3.0 8.5 (0.5) 29. 8 16. 6 Tucson, AZ
20012002 6.3 0.1 2.0 (0.9) 1.8 (0.6) 2.2 0.7 (2. 2) 4.8 1990 2002 42. 3
2.2 9.7 (0.8) 4.6 1.6 2.4 2.1 7.6 16. 9 Las Cruces,

2001NM 2002 2.7 0.0 0.4 (0.0) 0.7 0.2 0.6 (0.0) 1.0 0.6

1990 2002 6.8 0.4 0.1 0.2 1.7 0.4 1.3 0.1 4.7 (0.4)

Brownsville, 2001TX 2002 2.6 0.4 (1.2) 0.0 2.0 0.5 1.4 (0.1) 0.9 0.7

1990 2002 18. 6 1.5 (0.2) 1.4 3.7 0.5 3.0 (0.3) 6.7 5.9

El Paso, TX 20012002 0.5 (0. 2) (2.9) (1.0) 1.5 (0.6) 2.1 1.0 (0. 7) 2.9

1990 2002 7.6 1.4 (5.1) 1.3 1.2 (1.5) 2.6 1.5 (1. 6) 9.0

Laredo, TX 20012002 4.3 (0. 2) (0.2) (0.2) 1.9 (0.2) 2.0 0.2 1.1 1.6

1990 2002 16. 5 (1. 0) (0.2) 3.9 2.9 0.1 2.8 0.9 4.1 5.9

McAllen, TX 20012002 14. 7 0.5 (1.1) 0.2 3.6 0.3 3.2 0.5 7.1 3.8

1990 2002 47. 5 3.2 (1.7) 2.9 8.8 0.2 8.4 1.5 19. 3 13. 6

(Continued From Previous Page)

Employment in thousands

Transpor Finance, Construc

tation and insurance, Total tion &

Manufac public Wholesale Retail

and real nonfarm mining turing

utilities Trade trade trade estate Services Government

Industrial- mix effect

US- Mexico 20012002 Border 10. 0 (0. 6) (20.6) (3.1) (1.5) (3.0) 2.4 3.4
17. 2 15. 2 19902002

21. 2 4.2 (69.8) (1.8) (12.6) (8.5) (1.8) (2.8) 114. 4 (10.5) San Diego,

20012002 CA 5.1 (0. 3) (11.7) (1.5) (0.8) (1.7) 1.3 2.2 10. 1 7.2

19902002 16. 5 1.4 (38.9) (0.9) (6.8) (4.8) (1.1) (1.8) 68. 4 (4.9)

Tucson, AZ 20012002 2.3 (0. 1) (3.0) (0.3) (0.2) (0.4) 0.4 0.5 2.9 2.6

19902002 7.5 1.1 (8.9) (0.2) (1.8) (1.0) (0.3) (0.4) 19. 7 (1.9)

Las Cruces, 20012002 NM 0.7 (0. 0) (0.3) (0.1) (0.0) (0.0) 0.1 0.1 0.4 0.6

19902002 0.6 0.2 (0.9) (0.0) (0.3) (0.1) (0.0) (0.1) 2.3 (0.6)

Brownsville, 20012002 TX 0.5 (0. 0) (1.1) (0.2) (0.1) (0.1) 0.1 0.1 0.8
0.9

19902002 0.0 0.2 (3.8) (0.1) (0.6) (0.4) (0.1) (0.1) 5.1 (0.6)

El Paso, TX 20012002 (0.1) (0. 0) (3.3) (0.4) (0.2) (0.4) 0.3 0.3 1.6 2.0

19902002 (4.6) 0.6 (12.8) (0.3) (1.6) (1.3) (0.2) (0.2) 11. 0 (1.4)

Laredo, TX 20012002 0.4 (0. 0) (0.2) (0.4) (0.1) (0.1) 0.1 0.1 0.4 0.5

19902002 1.0 0.3 (0.5) (0.2) (0.4) (0.3) (0.0) (0.1) 2.2 (0.3)

McAllen, TX 20012002 1.2 (0. 0) (1.1) (0.2) (0.1) (0.2) 0.2 0.2 1.0 1.4

19902002 0.2 0.4 (3.9) (0.1) (1.0) (0.7) (0.1) (0.1) 5.7 (0.9)

National effect

US- Mexico 20012002 Border (15.8) (0. 9) (1.6) (0.7) (3.6) (0.6) (2.9)
(0.8) (4. 9) (3.2)

19902002 339. 1 17. 5 40. 0 15.4 80. 5 14. 7 65. 9 17.1 97. 3 71. 2

(Continued From Previous Page)

Employment in thousands

Transpor Finance, Construc

tation and insurance, Total tion &

Manufac public Wholesale Retail

and real nonfarm mining turing

utilities Trade trade trade estate Services Government

San Diego, 20012002 CA (8.6) (0. 5) (0.9) (0.4) (1.9) (0.4) (1.6) (0.5)
(2. 9) (1.5)

19902002 182. 8 8.7 21. 7 7.2 42. 4 7.9 34. 5 11.0 58. 0 33. 8 Tucson, AZ
20012002

(2.5) (0. 2) (0.2) (0.1) (0.5) (0.1) (0.4) (0.1) (0. 8) (0.5) 19902002 54.
5 3.8 4.8 2.3 12. 2 1.8 10. 4 2.2 16. 9 12. 3 Las Cruces,

20012002 NM (0.4) (0. 0) (0.0) (0.0) (0.1) (0.0) (0.1) (0.0) (0. 1) (0.1)

19902002 8.9 0.5 0.5 0.3 1.9 0.2 1.7 0.4 1.9 3.4

Brownsville, 20012002 TX (0.8) (0. 0) (0.1) (0.0) (0.2) (0.0) (0.2) (0.0)
(0. 2) (0.2)

19902002 16. 8 0.5 2.3 0.8 4.3 0.7 3.6 0.6 4.4 3.9

El Paso, TX 20012002 (1.8) (0. 1) (0.2) (0.1) (0.4) (0.1) (0.3) (0.1) (0.
4) (0.4)

19902002 42. 8 1.9 8.1 2.3 10. 2 2.3 8.0 1.6 9.4 9.2

Laredo, TX 20012002 (0.5) (0. 0) (0.0) (0.1) (0.1) (0.0) (0.1) (0.0) (0.
1) (0.1)

19902002 10. 5 0.8 0.3 1.7 3.0 0.5 2.4 0.4 1.9 2.4

McAllen, TX 20012002 (1.1) (0. 1) (0.1) (0.0) (0.3) (0.0) (0.3) (0.0) (0.
3) (0.3)

19902002 22. 7 1.3 2.3 0.8 6.5 1.2 5.3 0.8 4.8 6.2

Total employment change

US- Mexico 20012002 Border 92. 3 6.0 (11.4) (4.1) 26. 0 (2.7) 28. 8 6.2
35. 6 37. 1 19902002

590. 8 43. 4 7.7 30.7 103.4 10. 4 93. 1 19.6 282. 3 128. 1 San Diego,
20012002 CA 63. 4 6.0 (1.9) (0.3) 16. 8 (0.7) 17. 5 3.1 23. 3 16. 4
19902002 290. 5 24. 1 (6.8) 14.5 48. 2 6.2 42. 0 8.7 156. 2 45. 6

(Continued From Previous Page)

Employment in thousands

Transpor Finance, Construc

tation and insurance, Total tion &

Manufac public Wholesale Retail

and real nonfarm mining turing

utilities Trade trade trade estate Services Government

Tucson, AZ 2001- 2002 6.1 (0. 2) (1.3) (1.3) 1.1 (1.0) 2.1 1.0 (0. 1) 6.9

1990- 2002 104. 4 7.1 5.6 1.3 14. 9 2.4 12. 5 4.0 44. 2 27. 3 Las Cruces,

2001- NM 2002 3.0 0.0 0.1 (0.1) 0.6 0.1 0.6 (0.0) 1.3 1.1

1990- 2002 16. 3 1.1 (0.3) 0.5 3.3 0.5 2.9 0.4 8.9 2.4

Brownsville, 2001- TX 2002 2.3 0.3 (2.4) (0.2) 1.7 0.3 1.4 0.0 1.5 1.4

1990- 2002 35. 5 2.2 (1.7) 2.1 7.3 0.8 6.5 0.2 16. 2 9.2

El Paso, TX 2001- 2002 (1.4) (0. 3) (6.4) (1.6) 0.9 (1.1) 2.0 1.2 0.4 4.4

1990- 2002 45. 8 3.9 (9.8) 3.3 9.9 (0.5) 10. 4 2.9 18. 8 16. 8 Laredo, TX
2001-

2002 4.2 (0. 2) (0.4) (0.6) 1.7 (0.3) 2.0 0.3 1.4 2.0 1990- 2002 27. 9 0.1
(0.4) 5.4 5.5 0.3 5.2 1.2 8.2 7.9

McAllen, TX 2001- 2002 14. 7 0.4 (2.2) (0.0) 3.2 (0.0) 3.2 0.6 7.8 4.9

1990- 2002 70. 4 4.9 (3.3) . 14. 3 0.7 13. 6 2.2 29. 8 18. 9 Source: GAO
analysis using U. S. Department of Labor, Bureau of Labor Statistics data

Note: Numbers in parentheses indicate employment losses.

Effect of U. S. Economic Conditions on

Appendi x II

Employment in Mexican Maquiladoras Our statistical analysis shows that the
key factors cited in our semistructured interviews as responsible for the
maquiladora downturn* namely, the U. S. general economic slowdown,
particularly in U. S. manufacturing, and the real peso- dollar exchange
rate* are significant

determinants of maquiladora employment. We found a strong relationship
between maquiladora employment and U. S. economic conditions. This
relationship is stronger than that between maquiladora employment and the
real peso- dollar exchange rate, but considerably weaker than that between
maquiladora employment and changes in U. S. manufacturing shipments. We
also found that maquiladora sectors are more sensitive to changes in U. S.
manufacturing shipments than to broader U. S. economic conditions.

A major reason for the rapid growth of the maquiladora industry has been
its direct tie to the U. S. economy, particularly to U. S. manufacturing.
As a result, the maquiladoras are partly independent of Mexico*s internal
economic trends. This independence from the Mexican economy has made the
industry a stabilizing force when the Mexican economy heads into
recession. 1 However, the direct tie to U. S. manufacturing also makes the
industry predisposed to U. S. business cycles. As mentioned previously in
the main body of this report, the number of maquiladoras and the
employment they generate has declined from a peak reached in 2000. 2 This
decline has been attributed to several factors. The most important of
these factors is the downturn in the U. S. economy. An additional factor
that has been alleged to contribute to the apparent decline has been cost
increases due to increases in the inflation adjusted value of the peso
relative to the dollar, i. e., the real exchange rate of the peso. 3 This
appendix investigates the relationship

1 For example, in 1995, when Mexico's GDP fell by 6 percent, employment in
the maquiladora industry grew by more than 9 percent. During 1998, when
export earnings by the oil industry were off significantly, the
maquiladora industry became the largest source of foreign

revenue. See, Gerber, J, "Whither the Maquiladora? A Look at the Growth
Prospects for the Industry After 2001," Comercio Exterior, Bancomext. 9:
3, 1999.

2 In comparison to 2000 figures, latest statistics (November 2002) show
452 fewer maquiladora companies consisting of 310 garment maquiladoras, 56
electronic and electric accessory maquiladoras, 69 furniture assembly
maquiladoras, and approximately 17 companies in the rest of manufacturing.

3 The real exchange rate reflects the relative price of goods. It is the
nominal exchange rate adjusted for differences in inflation rates between
trading partners.

between maquiladora employment in Mexico and U. S. economic performance
and the real peso exchange rate. 4

To determine the link between maquiladora employment and U. S. economic
conditions, we assembled data on maquiladora employment in total and by
main sectors as well as data on U. S. GDP on a quarterly basis from
January 1980 to December 2002. We then converted all of these data to
their natural logarithms and performed a regression of maquiladora
employment on the real peso- dollar exchange rate and the real U. S. GDP.
5 The results of the regression are presented in table 3.

4 Maquiladoras are particularly sensitive to movements of real exchange
rates since they generate their revenues in dollars by exporting their
output to the United States while incurring production costs (labor and
other local inputs) in pesos. An appreciation of the real exchange rate of
the peso makes goods made in the United States cheaper relative to their
cost in Mexico. For example, in 2002, Mexico had an average annual rate of
inflation in

consumer prices of 5.7- percent, while the United States had an average
inflation rate of 2.4- percent. Consequently, the peso lost purchasing
power inside Mexico at a rate of 5.7 percent, but if its exchange rate
were not allowed to adjust, its loss of purchasing power in the United
States would be 2.4 percent. As a result, the goods purchased in the
United States would be 3.3 (5.7 percent minus 2.4 percent) percent cheaper
relative to their cost in

Mexico. The peso- dollar exchange rate has, in effect, appreciated in real
terms even if the nominal exchange rate does not change. An appreciation
of real exchange rate of the peso, therefore, makes maquiladoras less
competitive in the U. S. market.

5 The regression equations we estimated are represented by the following
relationship:

lnX j = + lnY + ln +

Where X j is Maquiladora employment, Y is U. S. Gross Domestic Product, is
the exchange rate of the dollar relative to the peso, and , , are positive
constants to be estimated. J represents the maquiladora sectors and ln
indicates natural logarithms.

Table 3: Summary of Regression of Maquiladora Employment and U. S. GDP and
Real Peso Exchange Rates

U. S. GDP in billions of Real

chained pesos per Constant 1996 dollars

dollar R- square

Total maquiladora employment -19.76 3.68 0.17 0.99

0.46 0.05 0.05

Textile products -29.50 4.64 -0.31 0.97

0.90 0.09 0.10

Footwear & leather products -12.98 2.32 0.70 0.80

1.15 0.12 0.12

Furniture products -33.03 4.77 0.47 0.94

1.28 0.13 0.14

Transportation equipment -33.77 4.94 0.87 0.93

1.40 0.15 0.15

Electronics -12.15 2.73 0.11 0.98

0.41 .04 .04

Electrical & electronic machinery -4.71 1.74 0.22 0.96

0.37 0.04 0.04

Electrical & electronic materials & accessories -17.04 3.24 0.06 0.98

0.52 0.06 0.06

Other manufacturing -18.82 3.56 0.15 0.99

0.40 0.04 0.04

Source: GAO analysis of Bank of Mexico (Banco De Mexico) data on
maquiladora employment, U. S. Department of Commerce, Bureau of Economics
data on GDP, and U. S. Department of Agriculture, Economic Research
Service data on exchange rates Note: Numbers in italics are standard
errors. All estimated coefficients were significant at 99 percent of
confidence.

As table 3 shows, maquiladora employment is very sensitive to U. S.
economic growth and the exchange rate. Our results show that a 1 percent
rise (or fall) in U. S. GDP increases (decreases) total maquiladora
employment by 3.68 percent, while a 1 percent rise in the real peso
exchange rate decreases maquiladora employment by 0.17 percent. 6
Maquiladora employment is consequently more responsive to changes in 6 It
should be noted that other factors not explicitly captured in our
estimates may also affect

maquiladora employment.

the U. S. economy than to changes in the real exchange rate of the peso. 7
In addition, maquiladora employment in the automotive sector is most
responsive to change in U. S. GDP, while maquiladora employment in
electrical apparatus and machinery is least responsive. The automotive
sector is also the most responsive to real exchange rate variations, while
the electrical materials sector is least responsive.

To investigate the stability of our estimates, we divided our sample into
three separate time periods: 1980 to1985, 1986 to1993, and 1994 to 2002.
Respectively, these three periods correspond roughly to the periods before
and after the implementation of Mexican economic policy reform and after
the implementation of NAFTA. Our analysis of the effect of U. S. GDP and
the real peso- dollar exchange rate on total maquiladora employment during
these three periods is shown in table 4. As the table shows, the

responsiveness of maquiladora employment to U. S. economic conditions and
the real peso exchange rate is fairly consistent with our results in table
3. However, the strongest maquiladora employment responsiveness to U. S.
GDP growth occurred in the pre- NAFTA reform period (1986 to 1993). The
post- NAFTA period (1994 to 2002) has a lower response coefficient for GDP

and a higher response coefficient for exchange rates. It should be noted
that the peso depreciated considerably in December 1994, after the onset
of the *peso crisis.*

7 It should be noted that the peso- dollar exchange rate has been more
volatile then U. S. GDP during this period.

Table 4: Summary of Regression of Maquiladora Employment and U. S. GDP and
Real Peso Exchange Rates for Three Subperiods

U. S. GDP in billions of chained 1996 Real pesos Constant

dollars per dollar R- square

1980- 85 -18.34 3.51 0.14 0.97

1.32 0.16 0.05

1986- 1993 -25.58 4.38 0.02 a 0.97

4.01 0.43 0.12

1994- 2002 -16.75 3.31 0.34 0.92

1.95 0.20 0.11

Source: GAO analysis. a All coefficients were significant at least at 95
percent of confidence, except the coefficient for pesodollar

exchange rate (1986- 1993). Note: Numbers in italics are standard errors.

We also looked into whether the U. S. manufacturing sector has a unique
effect that cannot be captured by overall U. S. GDP. To do so, we obtained
data on U. S. manufacturing shipments and performed a set of regressions
similar to those we performed using GDP. The results of our analysis
appear in table 5.

Tabl e 5: Summary of Regression of Maquiladora Employment and U. S.
Manufacturing Shipments and Real Peso Exchange Rates U. S. manufacturing
shipments in

Real pesos Constant

1996 dollars per dollar R- square

Total maquiladora employment -72.36 6.73 0.48 0.78

5.17 0.40 0.21

Textile products -103.44 9.07 0.22 a 0.89

4.73 0.37 0.19

Footwear & leather products -42.45 3.97 0.73 0.52

5.42 0.42 0.22

Furniture products -95.34 8.27 0.78 0.65

8.67 0.68 0.35

Transportation equipment -94.21 8.23 1.21 0.60

9.47 0.74 0.38

Electronics -54.51 5.20 0.35 0.85

3.19 0.25 0.13

Electrical & electronic machinery -31.81 3.35 0.43 0.86

1.95 0.15 0.08

Electrical & electronic materials & accessories -66.96 6.21 0.32 0.83

4.04 0.32 0.16

Other manufacturing -70.33 6.55 0.47 0.79

4.80 0.38 0.19

Source: GAO analysis. a Coefficient was not significant at 95 percent. All
other coefficient estimates are significant with at least

a 95 percent level of confidence. Note: Numbers in italics are standard
errors.

As can be seen from table 5, a 1 percent change in U. S. manufacturing
shipments induces an employment growth in maquiladoras of approximately
6.7 percent. Overall, the table shows, the maquiladora

employment*s response to changes in U. S. manufacturing shipments is
larger than its response to changes in U. S. GDP. We also found that
certain maquiladora sectors, such as textile products, furniture and
transportation equipment, are particularly sensitive to changes in U. S.
manufacturing shipments.

Mexico- China Competition in the U. S. Market

Appendi x III

for Imports Although U. S. imports from Mexico ($ 130.8 billion) 1
exceeded those from China ($ 109.2 billion) in 2001, these figures
represented a decline of 3.2 percent for Mexico and an increase of 1.9
percent for China. In 2002, both countries experienced growth, but U. S.
imports from China grew faster than U. S. imports from Mexico. This
development, coming at a time of decreased maquiladora employment and
increased plant closings, has led to speculation that Mexico is losing
ground because of China*s production cost advantages.

To highlight the competition between Mexico and China, we selected U. S.
imports items from Mexico in 1995 and 2002, with a value of more than $100
million in 2002. We also obtained information on U. S. imports from China
that matched the categories of the imports from Mexico. We then selected
U. S. imports for which the share from Mexico had declined between 1995
and 2002 and matched them with imports for which the share from China had
increased between 1995 and 2002.

In 2002, the United States imported from Mexico 152 categories of items
valued at more than $100 million each. The total value of these items was
$123.1 billion, while the total value of the same categories of items from
China was $88.2 billion. From 1995 to 2002, the share of U. S. imports
from Mexico decreased for 47 of the 152 categories. For these 47
categories, in 2002, the total value of imports from Mexico was $25.5
billion and the value of imports from China was $23.4 billion. China*s
share of U. S. imports increased for 35 of the 47 categories. The total
value of these 35 items was $20 billion for Mexico and $23 billion for
China.

Table 6 shows the top 25 U. S. import categories in which imports from
China increased, while imports from Mexico declined between 1995 and 2002.
As the table shows, Mexico and China appear to be in direct competition
for several import categories. Although a direct causal link is difficult
to establish, China seems to have gained U. S. market shares at the same
time that Mexico has lost them in some import categories, such as toys,
furniture, electrical household appliances, television and video

equipment and parts, and apparel and textiles. Maquiladoras are 1 The
figures used in the analysis presented in this appendix are from the U. S.
International Trade Commission. They differ somewhat from figures cited
earlier in this report for U. S. imports from Mexico, which were provided
by the Mexican National Institute of Statistics,

Geography and Information Technology (Instituto Nacional de Estadistica,
Geografia e Informatica).

concentrated in the categories where China appears to have gained U. S.
market shares.

Table 6: Top 25 U. S. Imports from China for Which China's Share of U. S.
Imports Grew, while Mexico's Share Declined between 1995 and 2002

Dollars in million

Mexico China Value of Share in

Share in Value of Share in

Share in U. S. U. S. total U. S. total

U. S. U. S. total U. S. total imports in imports

imports imports in imports

imports Item

2002 1995 2002

2002 1995

2002

Toys, puzzles, scale models $180 4.5% 2.2% $6, 927 73.3% 84. 3% Furniture
and parts thereof 595 6.9% 4.7% 4,932 12.2% 38. 7% Articles and equipment
for general physical exercise, etc. 176 5.4% 4.8% 2,011 28.7% 55. 3%

Electrical transformers, power supplies for adp machines or units 1,625
25.4% 24. 8% 1,553 10.8% 23. 7%

Electromechanical domestic appliances 408 24.8% 20. 6% 1,062 35.8% 53. 5%
Television receivers, including video monitors and video projectors 4,797
65.6% 47. 5% 860 2.6% 8.5%

Made- up articles of textile materials 340 29.6% 19. 1% 839 33.9% 47. 1%
Articles of jewelry and parts thereof 158 2.9% 2.6% 524 1.0% 8.6% Parts
for television, radio, and radar apparatus 1,021 34.7% 24. 3% 523 3.6% 12.
5% Stoves, ranges, grates, cookers, barbecues, braziers, and similar
nonelectric domestic

appliances 325 42.5% 26. 8% 467 7.1% 38. 5% Exports of articles imported
for repairs etc.; imports of articles exported and returned 3,870 11.6%
11. 5% 371 1.0% 1.1%

Articles of stationary, of paper or paperboard 116 20.6% 16. 5% 349 33.1%
49. 6% Electric filament or discharge lamps and parts thereof 244 18.0%
16. 0% 341 7.0% 22. 4%

Unrecorded media 266 12.5% 9.5% 333 10.3% 11. 9% Brassieres, girdles,
corsets, braces, suspenders, garters, and similar articles 201 18.8% 11.
8% 309 5.2% 18. 2%

Garments of textile fabrics, made- up of fabrics of felt or nonwovens 233
46.9% 28. 1% 252 18.7% 30. 4%

Trailers and semitrailers; other vehicles, not mechanically propelled; and
parts thereof 113 40.7% 14. 7% 165 8.4% 21. 4%

Articles of aluminum 180 22.3% 22. 3% 162 7.0% 20. 1%

(Continued From Previous Page)

Dollars in million

Mexico China Value of Share in

Share in Value of Share in

Share in U. S. U. S. total U. S. total

U. S. U. S. total U. S. total

imports in imports imports

imports in imports imports

Item 2002 1995

2002 2002

1995 2002

Women's or girls' slips, petticoats, briefs, panties, nightdresses,
pajamas, negligees, bathrobes, and similar articles 196 9.0% 9.0% 149 4.4%
6.8%

Centrifuges; filtering or purifying machinery and apparatus 449 19.7% 15.
5% 136 1.7% 4.7% Petroleum oils and oils from bituminous minerals (other
than crude) 569 2.4% 2.1% 96 0.0% 0.4%

Safety glass, consisting of toughened (tempered) or laminated glass 279
41.6% 41. 0% 84 0.6% 12. 3%

Orthopedic appliances; splints etc.; artificial parts of the body; hearing
aids and other appliances 143 13.0% 5.2% 82 2.0% 3.0%

Sanitary fixture including ceramic sinks, washbasins and pedestals, baths,
bidets, water closet bowls and flush tanks, urinals 210 51.4% 44. 4% 71
0.0% 15. 0% Measuring or checking instruments, parts and accessories
thereof 141 8.7% 7.8% 39 0.3% 2.2% Source: GAO calculation using U. S.
International Trade Commission data.

Appendi x IV

U. S.* Mexico Trade, by U. S. Port, 1999* 2002 Trade with Mexico through
U. S.- Mexico border crossings dropped in 2001 and remained flat in 2002.
Whereas, total trade through the 4 major land border ports fell by 5
percent in 2001, U. S. exports to Mexico through these ports fell by 10
percent. The port of Nogales, Arizona, experienced the sharpest decrease
in trade, with total trade declining by 9 percent in 2001 and 13 percent
in 2002. Table 7 provides information on U. S. imports,

exports, and total trade with Mexico by border crossing. The four border
crossings examined* Laredo, El Paso, San Diego, and Nogales* are Customs
districts that represent 33 individual ports of entry along the U. S.
Mexico border.

Tabl e 7: Trade Flows through Major U. S.- Mexico Land Border Crossings U.
S. dollars in millions Percent change 1999 2000 2001 2002 1999- 2000 2000-
2001 2001- 2002 Exports Total All Ports 692,821 780,419 731, 026 693,257
11% -7% -5%

Laredo, TX 45, 351 57, 659 52, 081 48, 937 21 -11 -6 El Paso, TX 13, 171
18, 045 16, 299 16, 476 27 -11 1 San Diego, CA 10, 760 12, 662 12, 342 12,
873 15 -3 4 Nogales, AZ 5,631 7,325 6,217 5,366 23 -18 -16

Border subtotal 74, 913 95, 691 86, 939 83, 652 22 -10 -4 Imports Total
all ports 1,024,766 1, 216,888 1,141, 959 1, 163,549 16% -7% 2%

Laredo, TX 51, 611 63, 298 62, 877 65, 351 18 -1 4 El Paso, TX 21, 007 24,
333 24, 151 24, 938 14 -1 3 San Diego, CA 19, 077 22, 263 21, 303 23, 013
14 -5 7 Nogales, AZ 10, 516 13, 050 12, 477 11, 112 19 -5 -12

Border subtotal 102,211 122,944 120, 808 124,414 17 -2 3

Trade (exports+ imports)

Total all ports 1,717,587 1, 997,307 1,872, 985 1, 856,806 14% -7% -1%

Laredo, TX 96, 962 120,957 114, 958 114,288 20 -5 -1 El Paso, TX 34, 178
42, 378 40, 450 41, 414 19 -5 2 San Diego, CA 29, 837 34, 925 33, 645 35,
886 15 -4 6 Nogales, AZ 16, 147 20, 375 18, 694 16, 478 21 -9 -13

Border subtotal 177,124 218,635 207, 747 208,066 19 -5 0

Source: U. S. Department of Commerce, Bureau of the Census, official trade
statistics. Note: Values for imports are general imports at customs value,
and values for exports are total exports at Free- Alongside- Ship (FAS)
value.

Appendi x V

Maquiladora Employment Statistics After growing rapidly throughout the
1990s, Mexican national maquiladora employment peaked in October 2000 and
declined sharply through March 2002. However, the rise and decline in
maquiladora employment varied by state and city. As table 8 shows, the
city of Tijuana experienced both the greatest percentage increase in
maquiladora employment (233 percent from 1990 through October 2000) and
the greatest decline (30 percent through December 2002). For each state or
city, table 8 shows the number of jobs in 1990, followed by the number of
jobs at the peak of employment (usually around October 2000) and at the
lowest point, or trough, following the peak. The table also includes the
changes in employment in absolute and percentage terms. The rise and
decline of maquiladora employment also varied by industry. Table 9 shows
employment changes for three key industries* electronics, autos and parts,
and textiles and apparel* along

with details on the rise, peak, and trough for the top five border region
cities in terms of maquiladora employment.

Table 8: Maquiladora Employment by State and City, 1990* 2002 Percent
Percent

Percent Change change Change change change Date in jobs:

in jobs: Duration:

in jobs: in jobs:

in jobs: Industry/ Jobs1990 Jobs at

of 1990 to

1990 to Jobs at

Dateof peak to peak to

peak to Jobs in trough area peak peak

peak peak trough trough

trough trough

trough 12/ 02 to 12/ 02 All Industries

National 1 year, 6 446, 436 1,347,803 Oct- 00 901,367 202% 1,060, 173 Mar-
02 months -287, 630 -21% 1, 084,911 2%

Border 1 year, 5 states 402, 432 1,045,410 Oct- 00 642,978 160 829, 954
Feb- 02 months -215, 456 -21 834,216 1

Baja 1 year, 7 California 87, 657 293,248 Oct- 00 205,591 235 215, 837
Apr- 02 months -77,411 -26 218,887 1

Coahuila 1 year, 3 30, 952 116,428 Oct- 00 85,477 276 102, 683 Dec- 01
months -13,745 -12 116,258 13

Chihuahua 2 years, 1 163, 953 336,995 Oct- 00 173,042 106 253, 722 Oct- 02
month -83,273 -25 262,558 3

Sonora 2 years, 2 38, 924 111,591 Nov- 00 72,667 187 70, 525 Dec- 02
months -41,066 -37 70, 525 0

Tamaulipas 1 year, 4 80, 947 187,581 Oct- 00 106,634 132 161, 139 Jan- 02
months -26,442 -14 165,988 3

Nuevo 1 year, 7 Leon 13, 868 72, 878 Sep- 00 59,010 426 50, 423 Mar- 02
months -22,455 -31 52, 181 3

Border 1 year, 11 cities 342, 555 839,200 Oct- 00 496,645 145 665, 637
Aug- 02 months -173, 563 -21 667,046 0

Juarez 2 years, 3 122, 231 264,241 Oct- 00 142,010 116 192, 485 Dec- 02
months -71,756 -27 192,485 0

Tijuana 2 years, 3 59, 870 199,428 Oct- 00 139,558 233 140, 069 Dec- 02
months -59,359 -30 140,069 0

Matamoros 1 year, 4 38, 360 69, 989 Oct- 00 31,629 82 52, 396 Jan- 02
months -17,593 -25 55, 183 5

Reynosa 23, 541 68, 199 Sep- 00 44,658 190 64, 877 Dec- 00 4 months -3,
322 -5 69, 389 7 Mexicali

1 year, 7 20, 729 65, 494 Oct- 00 44,765 216 49, 056 Apr- 02 months
-16,438 -25 55, 191 13

Share of border states in national 90% 78% 77%

Share of border cities in national 77% 62% 61%

(Continued From Previous Page)

Percent Percent Percent Change change Change change change Date

in jobs: in jobs:

Duration: in jobs:

in jobs: in jobs:

Industry/ Jobs1990 Jobs at

of 1990 to

1990 to Jobs at

Dateof peak to peak to

peak to Jobs in trough area peak

peak peak

peak trough trough trough

trough trough 12/ 02

to 12/ 02

Share of 5 border cities (above) in the overall

border 77% 80% 77% Source: National Institute of Statistics, Geography and
Information Technology (Instituto Nacional de Estadistica, Geografia e
Informatica).

Note: The *share of border states in national* is the percentage share
that maquiladora employment in border states makes up of national
maquiladora employment in Mexico. The *share of border cities in national*
is the percentage share that maquiladora employment in the 41 Mexican
border municipalities makes up of national maquiladora employment. The
*share of 5 border cities in the overall border* is the percentage share
that the five border cities listed in the table make up of total
maquiladora employment in the 41 border municipalities that comprise the
border region.

Table 9: Maquiladora Employment by City and Industry, 1990* 2002 Percent

Percent Percent Change change Change change change Jobs in jobs: in jobs:

Duration: in jobs:

in jobs: in Jobs: Industry/ Jobs at

Date of 1990 to

1990 to Jobs at Date of

peak to peak to

peak to Jobs in trough area 1990

peak peak peak

peak trough trough trough

trough trough 10/ 02

to 10/ 02 Electronics

National 166,501 468,254 Sep- 00 301,753 181% 320,282 Apr- 02 1 year, 8
-147, 972 -32% 326, 229 2% months Border

142,330 360,857 Oct- 00 218,527 154 248,163 Oct- 02 2 years, -112, 694 -31
248, 163 0 cities 1 month Juarez 48, 647 111,023 Oct- 00 62,376 128 68,908
Aug- 02 1 year,

-42,115 -38 71, 100 3 11 months

Tijuana 27, 598 97, 551 Oct- 00 69,953 253 61,971 Apr- 02 1 year, 7
-35,580 -36 64, 566 4 months Matamoros 17, 806 32, 052 Oct- 00 14,246 80
18,486 Feb- 02 1 year, 5

-13,566 -42% 19, 177 4 months Reynosa 13, 809 31, 073 Sep- 00 17,264 125
28,299 Jan- 01 5 months -2,774 -9% 29, 049 3

Mexicali 8,257 32, 963 Sep- 00 24,706 299 21,656 Jun- 02 1 year, -11,307
-34% 23, 270 7 10 months

Share of border cities in national 85% 77% 76%

(Continued From Previous Page)

Percent Percent Percent Change change Change change change Jobs in jobs:

in jobs: Duration:

in jobs: in jobs:

in Jobs: Industry/ Jobs1990 at

Date of 1990 to

1990 to Jobs at Date of

peak to peak to

peak to Jobs in trough

area peak peak peak

peak trough trough trough

trough trough 10/ 02

to 10/ 02

Share of 5 cities (above) in border 82% 84% 83%

Autoparts and other transportation National 104,487 250,635 Oct- 00
146,148 140 218,289 Oct- 01 1 year, 1 -32,346 -13 233, 747 7 month Border

77, 200 188,572 Jan- 01 111,372 144 164,599 Oct- 01 10 -23,973 -13 171,
289 4 cities months Juarez 39, 272 98, 821 Oct- 00 59,549 152 82,798 Oct-
02 2 years, -16,023 -16 82, 798 0

1 month Tijuana 1,021 6,701 Jan- 01 5,680 556 3,647 Dec- 01 1 year -3,054
-46 4,826 32 Matamoros 14, 086 21, 376 Apr- 01 7,290 52 15,543 Oct- 01 7
months -5,833 -27 21, 073 36 Reynosa* 4,753 14, 624 Sep- 00 9,871 208
13,470 Dec- 01 1 year, 4

-1154 -8 14, 450 7 months Mexicali 3,010 7,247 Dec- 00 4,237 141 6,063
May- 01 6 months -1,184 -16 7,353 21

Share of border cities in national 74% 75% 73%

Share of 5 cities (above) in border 80% 79% 76%

Textiles and Apparel

National 42, 464 294,855 Jul- 00 252,391 594 224,230 Mar- 02 1 year, 9
-70,625 -24 240, 315 7 months Border

20, 891 47, 493 May- 01 26,602 127 34,908 Apr- 02 1 year -12,585 -26 35,
217 1 Cities Juarez 8,634 10, 649 Jul- 97 2,015 23 4,592 Apr- 02 4 years,
-6,057 -57 4,857 6

9 months Tijuana 2,483 9,875 May- 01 7,392 298 6,782 Apr- 02 1 year -3,093
-31 7,394 9 Matamoros 368 2,537 Sep- 99 2,169 589 1,662 Jan- 02 2 years,
-875 -34 1,675 1

5 months Reynosa 925 3,141 Oct- 99 2,216 240 2,089 Jul- 02 2 years, -1,052
-33 2,267 9 10 months

Mexicali 2,454 2,758 Nov- 97 304 12 1,429 Mar- 02 4 years, -1,329 -48
1,562 9 4 months

(Continued From Previous Page)

Percent Percent Percent Change change Change change change Jobs in jobs:

in jobs: Duration:

in jobs: in jobs:

in Jobs: Industry/ Jobs at

Date of 1990 to

1990 to Jobs at Date of

peak to peak to

peak to Jobs in trough

area 1990 peak peak

peak peak trough trough

trough trough

trough 10/ 02 to 10/ 02

Share of 49% 16% 15% border cities in national

Share of 5 71% 61% 50% cities (above) in border

Source: National Institute of Statistics, Geography and Information
Technology (Instituto Nacional de Estadistica, Geografia e Informatica).
Notes: Data on the border region broken down by industrial sector were
only available through October 2002.

Data listed for textile and apparel jobs in Matamoros in 1990 are for
1994. Data were not available for prior years for that city and industry.
The *share of border cities in national* is the percentage share that
maquiladora employment in the 41

Mexican border municipalities makes up of national maquiladora employment.
The *share of 5 border cities in the overall border* is the percentage
share that the 5 border cities listed in the table make up of total
maquiladora employment in the 41 border municipalities that comprise the
border region.

Appendi x VI

Objectives, Scope, and Methodology Our work focused on employment and
production trends on the U. S. Mexico border and recent trends in the
maquiladora industry. We also analyzed data on overall U. S.- Mexico trade
and compared trends along the border with developments in the broader U.
S. and Mexican economies. To complete our objectives, we conducted
interviews with government

officials in the U. S. and Mexico, as well as semistructured interviews
with 29 industry associations. Between November 2002 and February 2003, we
conducted site visits in three areas of the border with a considerable
maquiladora presence: McAllen, Texas* Reynosa, Tamaulipas; El Paso,

Texas* Juarez, Chihuahua; and San Diego, California* Tijuana, Baja
California. Our selection criteria consisted of two characteristics
integral to the maquiladora industry: (1) the number of maquiladora
employees and (2) the number of maquiladora plants.

In addition to conducting site visits in selected border areas, we met
with U. S. officials and traveled to Mexico City to meet with Mexican
government officials. In the United States, we met with officials from the
Department of State, Office of the U. S. Trade Representative,
International Trade Commission, Environmental Protection Agency,
Immigration and Naturalization Service, Department of Labor, Department of
Transportation, Department of the Treasury, and U. S. Customs. In Mexico,
we met with officials from the Ministry of Economy; Ministry of Labor;
National Institute of Statistics, Geography and Information Technology;
Ministry of Treasury; Ministry of Government; and Ministry of

Environment. We obtained, reviewed, and analyzed data from maquiladora
industry experts, nongovernmental organizations, and Mexican and U. S.
government agencies. We also met with academics at educational
institutions in Mexico and the

United States, including San Diego State University; the University of
California, Los Angeles; University of California, San Diego; University
of Texas at El Paso; Colegio de la Frontera, Tijuana; Universidad Nacional
Autonoma de Mexico; and Universidad Autonoma Metropolitana de Xochimilco.
In addition, we met with numerous representatives of industry and
nongovernmental organizations as well as other maquiladora experts.

To understand how communities along the U. S.- Mexico border are
integrated and the role that maquiladoras play in U. S.- Mexico
interdependence (objective 1), we interviewed experts on the maquiladora
industry, academics, and representatives of nongovernmental organizations.
We reviewed extensive documentation and academic research provided by
these sources, analyzing economic, social, and

political linkages between border communities and the influence of the
maquiladora industry in the border region. We identified similarities and
differences between border communities with regard to social and economic
integration.

To review the status and trends in trade, employment, and output
(objective 2), we obtained original official data on employment and trade
from both U. S. and Mexican government agencies. We analyzed the data to
identify trends in employment and production in the U. S.- Mexico border

area. We compared our analysis of trends along the border with
developments in the broader U. S. and Mexican economies. For example, for
the United States, we conducted a shift- share analysis that decomposes
employment growth (or decline) in a region over a given time period into
three components: the national growth effect, the industry mix effect, and
the local (competitive) effect. To assess the quality and reliability of
the data, we conducted in- person meetings with government officials of
the National Institute of Statistics, Geography and Information Technology
in Mexico City to discuss the methodology for collecting the data and any
known limitations or biases. For instance, statistics on maquiladora

employment and production are affected when companies leave the program.
Although establishments and employees are no longer considered part of the
maquiladora sector and statistics correctly show a decline in maquiladora
employment, the firms and employees may still remain in operation outside
of the program. We also analyzed the data sources for internal
consistency, as well as external consistency with other sources of
information, such as our structured interviews. Although both U. S. and
Mexican statistics have some limitations, we consider the data
sufficiently

reliable to present general trends and magnitudes in production,
employment and trade. To identify the factors that have affected
employment and production in the maquiladora industry (objective 3), we
analyzed economic data and conducted semistructured interviews.
Specifically, to determine the link between maquiladora employment and U.
S. economic conditions, we assembled data on maquiladora employment in
total and by main sectors as well as quarterly U. S. GDP data from January
1980 to December 2002. We then converted all of these data to their
natural logarithms and performed a

regression of maquiladora employment on U. S. real GDP, U. S.
manufacturing shipments and the real peso- dollar exchange rate. The
semistructured interviews were conducted in person and by telephone with
29 representatives of business associations, consisting of organizations

representing principal industrial sectors involved in maquiladora
operations, and maquiladora associations at the local and national level.

Of these 29 organizations, 23 reported their members experienced a decline
in employment and/ or production. We asked these 23 organizations to
discuss the major reasons for the maquiladoras* recent decline. We relied
on business associations to identify the factors affecting employment and
production in the maquiladora industry, because of the direct experiences
of their membership with plant closures, changes in employment levels, and
other company changes. We also relied on associations to comment generally
on issues facing the industry, such as increased competition, and for
information on overall industry trends. In selecting potential interview
participants from maquiladora and other business associations, to ensure
representation throughout the industry, we considered three criteria:
geographic location, industry sector, and country of origin or region of
representation.

Of the 29 associations interviewed, 17 were maquiladora associations and
12 were industry- specific associations. The maquiladora associations were
primarily identified through the membership list for Mexico's National
Council of the Maquiladora Export Industry (Consejo Nacional de la
Industria Maquiladora de Exportacion -- CNIME) that has a membership
including 22 maquiladora associations located across Mexico. We contacted
all 22 members and the national association, and we completed

interviews with the national association and 14 local member associations.
We completed additional interviews with two maquiladora associations that
were not members of CNIME, but were included to broaden representation of
country of origin/ region of representation (i. e., Japan and

the United States). Of the 12 industry- specific associations, we sought
interviews with associations representing major industrial sectors,
specifically targeting the electronics, automotive, and apparel sectors. 1
Of the 29 associations, Mexico, the United States, and Japan were the
country of origins/ regions of representation included.

1 We identified major industry sectors with information presented in
MEXICONOW on the 100 largest maquiladora firms. Although the service
industry was one of the largest, it represents a very diverse group of
operations with too little in common to allow the identification of
factors of the recent industry downturn that would be applicable to all of
the industry*s members. Therefore, we sought to interview participants
that represented the electronics, automotive, and apparel sectors.

We developed 14 questions for the semistructured interview guide, based on
previous research. Six questions were closed ended, and eight were open
ended. Participants* responses to the open- ended items were
contentanalyzed by two trained coders, and intercoder reliability values
were computed. Reliability values ranged from 58 percent to 100 percent.
The coding category scheme was modified until 100 percent agreement was
reached between the two coders. The results will not be generalizable
outside our sample; however, we believe we have included associations in a
way that is as balanced and inclusive as possible within the number of

interviews we were able to conduct. To identify the implications of recent
developments in the maquiladora industry for the border region and U. S.-
Mexico trade (objective 4), we analyzed documents and interviews citing
factors that might influence the recovery of maquiladora production. We
also analyzed the debate about the viability of the industry and some
initiatives to identify and address its recovery. The information on
foreign laws in this report does not reflect our independent legal
analysis, but is based on interviews and secondary sources.

We performed our work from July 2002 through July 2003 in accordance with
generally accepted government auditing standards.

Appendi x VII Comments from the Department of State

Appendi x VIII

GAO Contacts and Staff Acknowledgments GAO Contacts Kim Frankena (202)
512- 8124 Juan Gobel (213) 830- 1031 Acknowledgments In addition to those
listed above, Joel Aldape, Bronwyn Bruton, Gezahegne

Bekele, Francisco Enriquez, Reid Lowe, Alison Martin, and Timothy Wedding
made key contributions to this report.

(320147)

Report to Congressional Requesters

July 2003 INTERNATIONAL TRADE Mexico*s Maquiladora Decline Affects U. S.-
Mexico Border Communities and Trade; Recovery Depends in Part on Mexico*s
Actions

03- 891

Contents Letter 1

Results in Brief 2 Background 3 Maquiladoras Contribute to Integration
along the Diverse

U. S.- Mexico Border 7 After Rapid Growth, Maquiladoras and Border Region
Experienced Declines Beginning in 2000 14

Cyclical and Structural Factors Cited in Maquiladora Decline 23
Maquiladora Downturn Spurs Some Positive Changes, but

Fundamental Challenges to Future Viability Remain 36 Agency Comments and
Our Response 44

Appendixes

Appendix I: Structure of Employment Growth in the U. S.- Mexico Border
Area 46 U. S. Border Employment Outpaced Nation*s since 1995 46

Appendix II: Effect of U. S. Economic Conditions on Employment in Mexican
Maquiladoras 57

Appendix III: Mexico- China Competition in the U. S. Market for Imports 63

Appendix IV: U. S.* Mexico Trade, by U. S. Port, 1999* 2002 66

Appendix V: Maquiladora Employment Statistics 67

Appendix VI: Objectives, Scope, and Methodology 72

Appendix VII: Comments from the Department of State 76

Appendix VIII: GAO Contacts and Staff Acknowledgments 78 GAO Contacts 78
Acknowledgments 78

Tables Table 1: Employment Growth of the United States and U. S.
Metropolitan Statistical Areas at the U. S. Mexico Border by

Industry: 1990* 2002 and 2000* 2002 48 Table 2: Components of Employment
Changes by Sectors in U. S.

Metropolitan Statistical Areas at the U. S. Mexico Border, 1990* 2002 53
Table 3: Summary of Regression of Maquiladora Employment and U. S. GDP and
Real Peso Exchange Rates 59

Table 4: Summary of Regression of Maquiladora Employment and U. S. GDP and
Real Peso Exchange Rates for Three Subperiods 61 Table 5: Summary of
Regression of Maquiladora Employment and U. S. Manufacturing Shipments and
Real Peso Exchange

Rates 62 Table 6: Top 25 U. S. Imports from China for Which China's Share
of

U. S. Imports Grew, while Mexico's Share Declined between 1995 and 2002 64
Table 7: Trade Flows through Major U. S.- Mexico Land Border

Crossings 66 Table 8: Maquiladora Employment by State and City, 1990* 2002
68 Table 9: Maquiladora Employment by City and Industry, 1990* 2002 69

Figures Figure 1: Map of U. S.- Mexico Border Twin Cities 4 Figure 2:
Maquiladoras* Share of Mexico*s Trade 1990- 2002 9

Figure 3: Map of Mexico Showing Share of Maquiladora Establishments, by
State 11 Figure 4: Growth in Maquiladora and Total Mexican Manufacturing
Production, 1993* 2002 15

Figure 5: Maquiladora Textile and Apparel Employment, Nonborder and Border
Regions, 1990* 2001 17 Figure 6: Maquiladora Employment and
Establishments, 1990- 2002 18

Figure 7: Mexican Maquiladora Employment in the Border Region by
Industrial Sector, January 1997* October 2002 19 Figure 8: Mexican
Maquiladora Employment, by Border City,

January 1990 - December 2002 20 Figure 9: Growth of Manufacturing
Production in Mexican Border

States, 1993* 2002 22 Figure 10: Annual Growth Rates of U. S. Gross
Domestic Product

and Maquiladora Employment, 1980- 2002 25 Figure 11: Value of U. S.
Imports from Mexico and China, 1995* 2002 27

Figure 12: U. S. Imports of Textiles and Apparel from Mexico, China, and
Caribbean Basin Countries, 1990* 2002 30 Figure 13: Real Dollar Exchange
Rate of Mexican Peso and Chinese

Yuan, 1995* 2002 32 Figure 14: Nonfarm Annual Employment Growth in the
United

States and in U. S. Metropolitan Statistical Areas at the U. S.- Mexico
Border, 1991- 2002 47

Figure 15: Employment Gains (Losses) in Nonfarm Employment in Metropolitan
Statistical Areas at the U. S. Mexico Border Due to National, Industry-
mix, and Local Effects 52

Abbreviations

BIP Border Industrialization Program CBI Caribbean Basin Initiative CNIME
Mexico's National Council of the Maquiladora Export Industry

(Consejo Nacional de la Industria Maquiladora de Exportacion) FAS Free-
Alongside- Ship GDP Gross Domestic Product ITA Information Technology
Agreement ITC International Trade Commission MSA Metropolitan Statistical
Area MEDC McAllen Economic Development Corporation NAFTA North America
Free Trade Agreement PROSEC Sectoral Promotion Program (Programa de
Promocion

Sectoral) USTR Office of the U. S. Trade Representative WTO World Trade
Organization

This is a work of the U. S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

a

GAO United States General Accounting Office

After growing rapidly during the 1990s, Mexican maquiladoras experienced a
sharp decline after October 2000. By early 2002, employment in the
maquiladora sector had contracted by 21 percent and production had
contracted by about 30 percent. The decline was particularly severe for
certain industries, such as electronics, and certain Mexican cities, such
as Tijuana. The downturn was felt on the U. S. side of the border as well,
as U. S. exports through U. S.- Mexico land border ports fell and U. S.
employment in manufacturing and certain other trade- related sectors
declined.

The cyclical downturn in the U. S. economy has been a principal factor in
the decrease in maquiladora production and employment since 2000. Other
factors include increased global competition, particularly from China,
Central America, and the Caribbean; appreciation of the peso; changes in
Mexico*s tax regime for maquiladoras; and the loss of certain tariff
benefits as a result of the North American Free Trade Agreement.

Maquiladoras face a challenging business environment, and recent
difficulties have raised questions about their future viability.
Maquiladoras involved in modern, complex manufacturing appear poised to
meet the industry*s challenges. Still, experts agree that additional
fundamental reforms by Mexico are necessary to restore maquiladoras*
competitiveness. U. S. trade and homeland security policies present
further challenges for maquiladoras.

Maquiladora Component of U. S.- Mexico Trade, 2001

Mexico*s maquiladoras have evolved into the largest component of U. S.*
Mexico trade. Maquiladoras import raw materials and components for
processing or assembly by Mexican labor and reexport the resulting
products, primarily to the United States. Most maquiladoras are U. S.
owned, and

maquiladoras import most of their components from U. S. suppliers.
Maquiladoras have also been an engine of growth for the U. S.* Mexico
border. However, the recent decline of maquiladora operations has raised
concerns about the impact on U. S. suppliers

and on the economy of border communities. Because of these concerns, GAO
was asked to analyze (1) changes in maquiladora employment and production,
(2) factors related to the maquiladoras* decline, and (3) implications of
recent developments for maquiladoras* viability. www. gao. gov/ cgi- bin/
getrpt? GAO- 03- 891. To view the full product, including the scope

and methodology, click on the link above. For more information, contact
Loren Yager at (202) 512- 4347. Highlights of GAO- 03- 891, a report to

congressional requesters

July 2003

INTERNATIONAL TRADE

Mexico*s Maquiladora Decline Affects U. S.- Mexico Border Communities and
Trade; Recovery Depends in Part on Mexico*s Actions

Page i GAO- 03- 891 International Trade

Contents

Page ii GAO- 03- 891 International Trade

Contents

Page iii GAO- 03- 891 International Trade

Page 1 GAO- 03- 891 International Trade United States General Accounting
Office Washington, D. C. 20548

Page 1 GAO- 03- 891 International Trade

A

Page 2 GAO- 03- 891 International Trade

Page 3 GAO- 03- 891 International Trade

Page 4 GAO- 03- 891 International Trade

Page 5 GAO- 03- 891 International Trade

Page 6 GAO- 03- 891 International Trade

Page 7 GAO- 03- 891 International Trade

Page 8 GAO- 03- 891 International Trade

Page 9 GAO- 03- 891 International Trade

Page 10 GAO- 03- 891 International Trade

Page 11 GAO- 03- 891 International Trade

Page 12 GAO- 03- 891 International Trade

Page 13 GAO- 03- 891 International Trade

Page 14 GAO- 03- 891 International Trade

Page 15 GAO- 03- 891 International Trade

Page 16 GAO- 03- 891 International Trade

Page 17 GAO- 03- 891 International Trade

Page 18 GAO- 03- 891 International Trade

Page 19 GAO- 03- 891 International Trade

Page 20 GAO- 03- 891 International Trade

Page 21 GAO- 03- 891 International Trade

Page 22 GAO- 03- 891 International Trade

Page 23 GAO- 03- 891 International Trade

Page 24 GAO- 03- 891 International Trade

Page 25 GAO- 03- 891 International Trade

Page 26 GAO- 03- 891 International Trade

Page 27 GAO- 03- 891 International Trade

Page 28 GAO- 03- 891 International Trade

Page 29 GAO- 03- 891 International Trade

Page 30 GAO- 03- 891 International Trade

Page 31 GAO- 03- 891 International Trade

Page 32 GAO- 03- 891 International Trade

Page 33 GAO- 03- 891 International Trade

Page 34 GAO- 03- 891 International Trade

Page 35 GAO- 03- 891 International Trade

Page 36 GAO- 03- 891 International Trade

Page 37 GAO- 03- 891 International Trade

Page 38 GAO- 03- 891 International Trade

Page 39 GAO- 03- 891 International Trade

Page 40 GAO- 03- 891 International Trade

Page 41 GAO- 03- 891 International Trade

Page 42 GAO- 03- 891 International Trade

Page 43 GAO- 03- 891 International Trade

Page 44 GAO- 03- 891 International Trade

Page 45 GAO- 03- 891 International Trade

Page 46 GAO- 03- 891 International Trade

Appendix I

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 47 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 48 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 49 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 50 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 51 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 52 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 53 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 54 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 55 GAO- 03- 891 International Trade

Appendix I Structure of Employment Growth in the U. S. Mexico Border Area

Page 56 GAO- 03- 891 International Trade

Page 57 GAO- 03- 891 International Trade

Appendix II

Appendix II Effect of U. S. Economic Conditions on Employment in Mexican
Maquiladoras

Page 58 GAO- 03- 891 International Trade

Appendix II Effect of U. S. Economic Conditions on Employment in Mexican
Maquiladoras

Page 59 GAO- 03- 891 International Trade

Appendix II Effect of U. S. Economic Conditions on Employment in Mexican
Maquiladoras

Page 60 GAO- 03- 891 International Trade

Appendix II Effect of U. S. Economic Conditions on Employment in Mexican
Maquiladoras

Page 61 GAO- 03- 891 International Trade

Appendix II Effect of U. S. Economic Conditions on Employment in Mexican
Maquiladoras

Page 62 GAO- 03- 891 International Trade

Page 63 GAO- 03- 891 International Trade

Appendix III

Appendix III Mexico- China Competition in the U. S. Market for Imports

Page 64 GAO- 03- 891 International Trade

Appendix III Mexico- China Competition in the U. S. Market for Imports

Page 65 GAO- 03- 891 International Trade

Page 66 GAO- 03- 891 International Trade

Appendix IV

Page 67 GAO- 03- 891 International Trade

Appendix V

Appendix V Maquiladora Employment Statistics

Page 68 GAO- 03- 891 International Trade

Appendix V Maquiladora Employment Statistics

Page 69 GAO- 03- 891 International Trade

Appendix V Maquiladora Employment Statistics

Page 70 GAO- 03- 891 International Trade

Appendix V Maquiladora Employment Statistics

Page 71 GAO- 03- 891 International Trade

Page 72 GAO- 03- 891 International Trade

Appendix VI

Appendix VI Objectives, Scope, and Methodology

Page 73 GAO- 03- 891 International Trade

Appendix VI Objectives, Scope, and Methodology

Page 74 GAO- 03- 891 International Trade

Appendix VI Objectives, Scope, and Methodology

Page 75 GAO- 03- 891 International Trade

Page 76 GAO- 03- 891 International Trade

Appendix VII

Appendix VII Comments from the Department of State Page 77 GAO- 03- 891
International Trade

Page 78 GAO- 03- 891 International Trade

Appendix VIII

GAO*s Mission The General Accounting Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities

and to help improve the performance and accountability of the federal
government for the American people. GAO examines the use of public funds;
evaluates federal programs and policies; and provides analyses,
recommendations, and other assistance to help Congress make informed
oversight, policy, and funding decisions. GAO*s commitment to good
government is reflected in its core values of accountability, integrity,
and reliability.

Obtaining Copies of GAO Reports and Testimony

The fastest and easiest way to obtain copies of GAO documents at no cost
is through the Internet. GAO*s Web site (www. gao. gov) contains abstracts
and fulltext files of current reports and testimony and an expanding
archive of older products. The Web site features a search engine to help
you locate documents using key words and phrases. You can print these
documents in their entirety, including charts and other graphics.

Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as *Today*s Reports,* on its
Web site daily. The list contains links to the full- text document files.
To have GAO e- mail this

list to you every afternoon, go to www. gao. gov and select *Subscribe to
e- mail alerts* under the *Order GAO Products* heading. Order by Mail or
Phone The first copy of each printed report is free. Additional copies are
$2 each. A check

or money order should be made out to the Superintendent of Documents. GAO
also accepts VISA and Mastercard. Orders for 100 or more copies mailed to
a single address are discounted 25 percent. Orders should be sent to:

U. S. General Accounting Office 441 G Street NW, Room LM Washington, D. C.
20548 To order by Phone: Voice: (202) 512- 6000 TDD: (202) 512- 2537 Fax:
(202) 512- 6061

To Report Fraud, Waste, and Abuse in Federal Programs

Contact: Web site: www. gao. gov/ fraudnet/ fraudnet. htm E- mail:
fraudnet@ gao. gov Automated answering system: (800) 424- 5454 or (202)
512- 7470

Public Affairs Jeff Nelligan, Managing Director, NelliganJ@ gao. gov (202)
512- 4800 U. S. General Accounting Office, 441 G Street NW, Room 7149
Washington, D. C. 20548

United States General Accounting Office Washington, D. C. 20548- 0001
Official Business Penalty for Private Use $300 Address Service Requested

Presorted Standard Postage & Fees Paid

GAO Permit No. GI00
*** End of document. ***