Deaths During Undocumented Migration - Trends and Policy Implications in the New Era of Homeland Security

1

Deaths during undocumented migration:

trends and policy implications in the new era of homeland security

Karl Eschbach

Jacqueline Hagan

Nestor Rodríguez

Presented at the

26th Annual National Legal Conference

on Immigration and Refugee Policy

Published in In Defense of the Alien, Vol. 26, pp. 37-52.

Washington, DC

April 2003

2

Border Death Project

The Center for Immigration Research at the University of Houston has been

conducting an ongoing study of deaths of undocumented migrants along the Southwest

border since 1995 through fieldwork and quantitative research (Eschbach, Hagan,

Rodriguez, Hernandez and Bailey, 1999; Eschbach, Hagan and Rodriguez 2001). The

fieldwork has involved interviews with Border Patrol agents, medical examiners, funeral

directors, local law enforcement agents, undocumented migrants and human rights

advocates. Through these interviews we have sought to understand both the number and

reasons for migrant deaths in each area along the border. The quantitative component

involves the systematic study of trends in undocumented deaths along the full border

using a standardized data source. We believe that together these two approaches provide

a comprehensive understanding of migrant deaths in recent decades. In this presentation

we discuss our findings in relation to the themes of the conference session, border

enforcement policy after September 11.

Failure of IRCA and continued undocumented immigration

To understand the emergence of concern about the deaths of undocumented

migrants in the middle 1990s, it is important to place the problem in the context of

immigration policy during this period. The dominant policy strategy during the 1990s

was a substantial increase in the effort expended by the United States to prevent illegal

entry through increased and targeted patrol of the Southwest border.

3

This border enforcement policy emerged in the wake of the failure of the

Immigration Reform and Control Act of 1986 (IRCA) to control undocumented

migration. Recall, IRCA attacked undocumented migration through a three-pronged

strategy: a legalization program, an employer sanction program, and a border

enforcement program. In the short run of a few years, IRCA did reduce the flow of

undocumented migrants by legalizing several million migrants engaged in circular flow

between the United States and Latin America, moving them into legal border crossing

channels. However, while IRCA temporarily changed the stock of undocumented

migrants through the legalization program, in the long run it did not change the basic

dynamic of undocumented migration to the United States. By the 1990s, migration flows

had returned to their pre-IRCA levels.

The Gatekeeper Complex

The response to the failure of IRCA has been a progressive increase in the

resources devoted to border control. Some of these increases were mandated under the

terms of IRCA, and began in the late 1980’s. Most were ushered in the 1990s. Resources

devoted to border enforcement were dramatically increased. The annual budget for

enforcement operations increased sevenfold between 1980 and 1995, and then tripled

between 1995 and 2001. It currently exceeds $2.5 billion (Reyes, Johnson, and

Swearingen, 2002), The number of Border Patrol agents along the southwest border more

than doubled between FY 1993 to FY 1999, from roughly 3,400 to 7,200 (Andreas 2001).

The new policy initiative also included a huge influx of new technological resources such

4

as night vision cameras and ground sensors, along with the construction of new physical

barriers along the border. In the San Diego sector alone, for example, the length of border

fencing has more than doubled since 1994 (Andreas, 2000). Indeed, the Congressional

Research Service reported that roughly $3.3 billion has been invested on the Border

Patrol since 1994 (cited in Suro, 1998:3).

The Border Patrol deployed these additional resources in the most popular urban

crossing locations in a series of highly publicized border control operations. Operation

Blockade (later renamed Hold-the-Line) was implemented in El Paso in 1993, followed

the next year by Operation Gatekeeper in San Diego. Subsequent years saw the extension

of intensive border control to other stretches of the border. We call the border control

efforts of this period the “Gatekeeper Complex,” named for the largest and most visible

of the border control operations.

The Gatekeeper Complex had the intended purpose of raising the cost of

migration by closing off the easy and well established crossing channels. Since El Paso

and San Diego were adjacent to the largest border cities in northern Mexico (Juarez and

Tijuana), transportation services to these locations within Mexico were plentiful. When

they got to the border, migrants were implicitly aided by the large number of other

persons trying to cross at the same places, which tended to overwhelm Border Patrol

resources (Espenshade, Baraka and Huber, 1997). The proximity to urban areas meant

that migrants could receive assistance from and blend to these cities to plan the next leg

of their northbound journey. Networks of alien smugglers were well established in these

cities. In the stretch from San Diego to Los Angeles, attempts by the Border Patrol to

5

control transit up Interstate 5 corridor from San Diego to Los Angeles had to contend

with the fact that this Interstate is among the most heavily traveled in the country.

By initially concentrating enforcement on the most used crossing points, the

Border Patrol intended to deny these relatively easy corridors to the migrants, redirecting

those who would still try to smaller secondary cities. The extension of increased

enforcement to secondary cities was intended to increasingly redirect undocumented

migration flows out of urban areas into the surrounding countryside. Because there were

relatively few roads leading away from the border in many of these rural areas, the

Border Patrol could in these areas control flow on the highways using a combination of

fixed checkpoints and mobile patrols. The Border Patrol could then deploy additional

agents and technological aids to “cut for sign” of northbound intruders. The length of

these cross country journeys--sometimes taking more than 50 miles on the United States

side of the border alone—gave the Border Patrol more time to apprehend migrants than in

urban settings.

It was the expressed hope of the INS officials who designed the new enforcement

policies of the 1990s that the intensified and targeted control would discourage many

would-be migrants from even attempting the journey because of the additional physical

difficulty, and the increased financial and psychological costs of the northbound journey.

New technologies and laws that increased the penalties faced by migrants who

were apprehended during an unauthorized entry accompanied these increases in resources

devoted to border control. The penalty that had long been paid by the person apprehended

while attempting an unauthorized entry had been voluntary return to the country of

6

origin. The implementation of new "IDENT" technology created and maintained a

database of biometric identifiers that allowed the Border Patrol to identify repeat entrants

and increase the penalty to prison time for some.

Increasing dangers to migrants?

Almost as soon as the plans for the Gatekeeper Complex became public, migrant

advocates expressed their fear that the collateral damage from this enforcement initiative

would be increased dangers and risk of death for the migrants. The rural terrain to which

migrants were being redirected exposed them to increased risk. As the probability of

apprehension and the legal penalties associated with undocumented migration went up,

the impulse to take dangerous risks through more dangerous concealment strategies to

avoid capture also went up. So too did the incentive to engage in either high-speed flight

or violence in response to an imminent apprehension.

For its part, the INS did not contest the idea that there were dangers associated

with the redirection of undocumented migration to rural areas. INS officials made several

responses to these allegations. First, they observed that the migrants themselves were

responsible for these increased risk since undocumented entry is a crime in the United

States. Second, they pointed to professional alien smugglers—the so-called coyotes-- as

the principal villains, by encouraging and sometimes misleading migrants to take

unnecessary risks. Third, they observed that the border was already dangerous before the

implementation of the concentrated enforcement campaign, especially in the out-of-

control San Diego crossing corridor.

7

When some of the predictions about the dangers to migrants in the restructured

border environment proved true, the Border Patrol implemented the Border Safety

Initiative Program in 1998, in cooperation with the Mexican government. This initiative

included increased deployment of emergency medical service (EMS) units, increased

EMS training for agents, increased supply if life-saving equipment in Border Patrol

vehicles, an emphasis on patrolling-for-rescue as well as patrolling for apprehension,

particularly during dangerously hot weather. Another part of this initiative was an

advertising campaign in Mexico and Central America warning potential migrants of

border dangers.

What did the Gatekeeper Complex accomplish?

There is little evidence that the Gatekeeper Complex has reduced undocumented

migration and the unauthorized labor supply (Hanson and Spilimbergo 1999; Reyes et al

2002; GAO 1997; 1999; 2001). Many critics of border enforcement policies in this period

suggest that the primary result of increased danger and expense of illegal border crossing

has been to decrease cross-border circulation by lengthening the time spent in the United

States by undocumented persons. (Kossoudji, 1992; Espenshade et al, 1997; Massey,

Durand, and Malone, 2002; Reyes et al, 2002). The 1990s saw large increases in the

stock of undocumented immigrants living in the United States. These increases are

estimated to be about half a million per year. The undocumented population in the United

States has been estimated at about 8.5 million in 2000 (Passel and Fix, 2001).

8

Although there is no evidence to show that the Gatekeeper Complex

accomplished its policy goals, there is no doubt it dramatically restructured the

geography of border crossing. In particular, the massing of enforcement resources and

construction of effective fencing in the extreme western part of San Diego County west

of Otay Mountain and close to the San Ysidro port of entry substantially reduced illegal

migration through this formerly popular migration corridor.

Figure 1 shows the magnitude of this restructuring by showing the changing

distribution of apprehensions to different Border Patrol operational sectors along the

Southwest border from 1985 to 2002. Apprehensions are an imperfect measure of the

location of migration flows, because the probability of apprehension (and hence the

relationship between apprehension totals and successful migration attempts) varies at

different places on the border. However, the patterns in apprehensions are so striking that

they give a clear indication of what happened. The percentage of apprehensions in the El

Paso and San Diego border sectors peaked at 70 in 1992. By 1998, this figure had

dropped to less than 25 percent. At the same time, the percentage of apprehensions

increased in the area from Imperial County California through Arizona.

Insert Figure 1 Here

By the summer of 1998, the western San Diego area had become so well

controlled that middle class housing was under development in areas on the United States

side of the border that had formerly been overrun by undocumented border crossing. In

the same year, however, Border Patrol officials in Calexico estimated that the ratio of

"got-aways" to apprehensions in that area may have been as high as ten-to-one, because

9

tens of thousands of undocumented migrants had shifted to this area, while the Border

Patrol did not markedly increase resources. By 2000, increased resources devoted to

border enforcement in the Calexico area had shifted flows further eastward yet, making

the Arizona desert the primary terrain where unauthorized entries are attempted

Within the border areas, unauthorized migration increasingly shifted to the open

countryside surrounding the relatively well-controlled border cities. In any case, border

cities in Arizona are relatively small. Even if an undocumented migrant successfully

entered an Arizona border city such as Nogales, Douglas, or Naco, he or she would still

need to traverse the parched landscape northward to Tucson or Phoenix. These open areas

have become the new playing field for the sometimes deadly game of cat-and-mouse

played by the undocumented migrants and United States border control personnel.

How did the Gatekeeper Complex change the deaths of undocumented migrants

along the border?

The Gatekeeper Complex has had limited success from the point of view of its

policy goals, but what has been the effect on death and danger to migrants? The question

is not as straightforward as at first appears. Government officials did not keep border

wide records of the deaths of undocumented migrants until 1998, when the Border Patrol

began to record deaths occurring in furtherance of an illegal entry in border areas in the

United States. The Mexican government also began to tabulate deaths of its own citizens

attempting undocumented entry in the 1990s. The Mexican series includes deaths in near-

border areas in northern Mexico, as well as deaths in the United States. The series is

10

aggregated from reports assembled from the network of Mexican consulates in the border

region. It is marred in its early years, by inconsistent coverage of different border areas.

Coverage of the San Diego border region is complete dating from the early 1990s, but

was incomplete through most of the decade in other areas, and especially in Texas.

The vital registration database for the United States is a useful alternative source

of data about longer term trends in deaths during undocumented migration. This source is

intended to be a complete registry of all deaths occurring within the United States. Death

records do not contain any information about the immigration status of decedents, or

activity at time of death. Thus these data cannot be used to generate a definitive count of

deaths during undocumented migration. Vital registration death data do contain

information about decedents’ place of residence and citizenship, as well as information

about underlying cause of deaths. Using this information we can construct a data series

for accidental deaths of foreign transients (foreign-born, non-U.S. residents) and for

unidentified bodies in near-border counties. Comparison of the Border Patrol counts of

undocumented migrant deaths with the death data supplied by the San Diego medical

examiner and Imperial County coroner show that the large majority of the accidental

deaths of foreign transients identified in the vital registration data for near border

counties do in fact occur to undocumented migrants attempting to enter without

inspection. Thus the shaded area of overlap in figure 2 is large, and the vital registration

data provide a useful data source to learn about longer term trends in the numbers and

causes of deaths occurring in the course of undocumented migration (Eschbach, Hagan,

Rodriguez 2001).

Insert Figure 2 Here

11

Trends in deaths

Figure 3 shows the trends in deaths that we compiled at the University of Houston

from vital registration sources (1985 to 2000). We extend this series from to 2002 using

Border Patrol counts. The number of deaths reported follows a u-shaped curve. In the late

1980s, the number of foreign transient deaths usually exceeded 300, and peaked in 1988

at 355. Thereafter, the number of deaths fell to 180 in 1993 and 1994. After 1994 the

number of deaths started to increase again, peaking in 2000 at 370. Border Patrol counts

for 2001 and 2002 show a small decrease in the number of deaths in those years

compared to 2000.

Insert Figure 3 Here

As Figure 4 shows, the leading causes of deaths of foreign transients in the border

region across the period were drowning, motor vehicle accidents, auto-pedestrian

accidents, deaths from exposure to environmental heat and cold, and deaths from

unknown causes. (Unknown cause deaths pertain primarily to bodies found as skeletal

remains in open areas.) In considering these distributions, it is important to keep in mind

that not all deaths occur during an undocumented entry. For example, deaths from

homicides likely include a large number of undocumented aliens, but a substantial

portion could pertain to persons who were engaged in drug trafficking or other illegal

activities (Eschbach, Hagan, Rodriguez 2001).

Insert Figure 4 Here

Figure 5 shows that there were several important changes in the causes of deaths

of foreign transients in the border region. Two causes, homicide deaths and auto-

12

pedestrian accidents, peaked in 1988 and declined thereafter. Since 1994, both causes

together have contributed about 50 deaths a year. Deaths from exposure to environmental

heat, cold and dehydration declined from the middle 1980s to the middle 1990s before

increasing sharply.

Insert Figure 5 Here

. The large majority of auto-pedestrian deaths occurred in the San Diego area. This

cause of death was substantially reduced before the implementation of Operation

Gatekeeper in 1994. Auto-pedestrian deaths had occurred primarily in the vicinity of the

Border Patrol’s checkpoint on Interstate 5 at San Clemente. These deaths declined after

the fencing of the Interstate median in the early 1990s made it difficult for northbound

migrants to cross the southbound lanes on foot south of the checkpoint.

Like auto-pedestrian deaths, homicide deaths occurred primarily in San Diego

County. The number of deaths began to drop rapidly in the early 1990s. Investigation of

these deaths in the archives of the San Diego Union-Tribune shows that a few of the

deaths involved the victimization of a migrant worker crossing the San Diego border.

Many more were related to disputes among drug-traffickers. The number of homicides

declined markedly and has remained low in the San Diego border areas after it was

brought under control by Operation Gatekeeper. The shifting of undocumented migration

to other border areas did not result in an increase in homicides to the areas were

undocumented migration was shifting.

The final noteworthy change in cause-specific death totals was a sharp increase

after 1993 in deaths from environmental heat, cold, exposure and dehydration. There

13

were 33 such deaths reported in 1985, declining to just 6 deaths border wide from these

causes in 1992 and 1993. This cause of death has skyrocketed since 1995, to 99 in 2000.

Data from the Border Patrol and the Mexican Consulates show that this cause of death

has remained high since 2000. This figure understates the number of deaths from

environmental causes, because many bodies may remain undiscovered in open areas

throughout the region.

Motor vehicle accidents and drowning are both leading causes of deaths. Both

causes of death fluctuate in number throughout this period, with no overall pattern related

to changes in enforcement and crossing patterns. Our studies have shown that that the

majority of drowning deaths occur in the Rio Grande, and that water levels along that

river are more strongly associated with the number of drowning deaths than is the

changing location of border crossing. The migrants have been fortunate that extended

drought conditions in the Rio Grande basin through most of the 1990s have kept water

levels in the river at historic lows (Eschbach, Hagan and Rodriguez 2001).

In summary, the data partially confirm the fears of the human rights advocates

that the redirection of flows from urban to rural crossing points is increasing the danger

to migrants. It is now unambiguous that the redirection of migrant flows to the Sonoran

deserts of Arizona and eastern California, and to the ranchlands of South Texas, takes the

lives of perhaps 100 persons per year that we can document through vital registration

data, and an uncounted number of additional persons whose bodies have yet to be - and

perhaps may never be – discovered.

14

We should not overlook, however, that undocumented border crossing was

dangerous before the Gatekeeper Complex. Returning to Figure 5, we can see that in the

middle 1980s, drowning (mostly in the Rio Grande) was the leading cause of death for

undocumented migrants. Deaths from environmental causes were reported even in this

period. Moreover urban areas have their dangers, as shown by the chaos near the border

and on the freeways in San Diego County in the late 1980s.

Had the increases in flow associated with the peso devaluation in 1995 all been

channeled through the San Diego and El Paso areas, we are not sure that the resulting

death totals would have been markedly lower than what was observed, though the cause

and location of death would have been different. While the Gatekeeper Complex

probably did result in an increase in the number of migrant deaths, undocumented

migration across the Southwestern border has always been dangerous. Resurrecting the

status quo in the days before Operation Gatekeeper is not a solution to the dangers of

undocumented border crossing.

Enforcement after September 11.

The terrorist attacks of September 11 had immediate ramifications for United

States border policy. The most immediate effect was to slam shut the door that had been

opened to the adjustment of the legal status of millions of undocumented Mexicans in the

United States, and perhaps to the creation of a new guest worker program in the United

States. To be sure, it was by no means clear what policy would have emerged from the

discussions about a more liberal immigration policy that Presidents Fox and Bush had

opened in the months before 9/11. It is clear that in the aftermath of the attacks, such a

15

liberalization was taken off of the table for the time being, and restrictionist voices gained

strength in public debates.

In the Gatekeeper era the mantra of the restrictionists was that we needed to

control the borders against “alien and drug smuggling.” The new threat of foreign

terrorists after September 11 added a third and very powerful justification for increased

vigilance at our borders, despite the fact that none of the September 11 terrorists are

known to have entered illegally across the Southwestern border. In a climate in which

legal migrants faced suspicions and restrictions in the name of anti-terrorist counter

measures, unauthorized migrants could expect little sympathy.

The ineffectiveness of attempts to halt undocumented migration across the

Southwest border stands in odd symbolic juxtaposition to the new claims that the United

States will seal its border against terrorist entry. Conventional wisdom is that

undocumented migrants always successfully run the gauntlet of border control or die

trying. This fact sits uneasily next to the somewhat different challenge of control of the

border against terrorism. In opening the new Department of Homeland Security,

Secretary Ridge promised to “seal our borders from terrorists and their cargo.” Yet if a

circular flow of more than a million undocumented migrants can make the journey each

year, how well sealed can the border be against a threat that is potentially devastating in

its consequences if just a single terrorist manages to enter?

On the ground at the border itself, September 11 has not changed the basic

dynamic of the cat-and-mouse game played out between undocumented migrants and

United States border enforcement personnel. The number of apprehensions did fall

16

sharply after September 2001, continuing a downward trend that had begun the previous

year. Increased border enforcement, immigrant fear of increased Border Control in the

context of 9/11, changing Border Patrol strategies, and slow-downs in the United States

economy likely all have contributed to the decline.

In the face of declining apprehensions, the number of deaths of undocumented

migrants along the border reported by the Border Patrol continues to exceed 300, which

is substantially higher than the death totals from the middle 1990s. In 2003, the Arizona

border continues to be the primary place where undocumented entry is attempted and

contested. Migrant death from heat and dehydration remains a tragic concomitant of the

border enforcement game in the arid deserts of the border region.

After September 11, national security concerns are increasingly intertwined with

border enforcement policy. It is unlikely that either Congress or the general public would

agree to reduce border enforcement to the level seen before the Gatekeeper Complex was

implemented. For this reason, calls to “Stop Operation Gatekeeper”—implicitly to return

to the enforcement policies of the 1980s—increasingly fall on deaf ears.

As mentioned above, simply rolling back enforcement policy to before

Gatekeeper would not in any case have eliminated the deaths of undocumented migrants.

Deaths will stop when undocumented flows are regulated through legal rather than illegal

channels. Evidence for this is provided by the lull in deaths that we observed in the late

1980s and the early 1990s. The most plausible explanation for this decline was the

temporary reduction in the circular flow of undocumented migrants because IRCA

legalized more than two million immigrants. This took the migrants out of the rivers and

17

deserts of the border region, and put them onto commercial buses, eliminating most of the

dangers associated with attempted unauthorized entry.

The policy alternative confronting the United States today is not between a lax

and a stringent effort to control the border, but rather between a policy that acknowledges

the dependency of the American economy on foreign labor and a policy that continues to

keep a large portion of this labor in undocumented status. It is likely that for the

foreseeable future the quantity of border enforcement effort expended will exceed

historical levels. But increased control of unauthorized entry is not inconsistent with

increased authorization of legalized entry. The combination of increased authorized entry

and increased border control may be the most effective policy for national security, to

accomodate the labor demands of the United States economy, and to reduce the toll of

migrant death at the Southwest border.

September 11 may have created a new willingness of the American public to

implement new technologies for the control of undocumented labor supply at the place of

employment rather than at the border. These technologies include increased use of

identification with non-duplicatable biometrics, together with the real time query of

databases that report the legal status of the worker or job applicant. These technologies, if

used, will probably make possible a more effective control of undocumented labor supply

than is possible through the cat-and-mouse game currently played out between the Border

Patrol and the undocumented migrants at the border. One implication of this effectiveness

is that the United States may be forced to make its decisions about the use of unskilled

foreign labor more transparent. If the best evidence is that more eight million illegal

immigrants are a functioning part of the labor force of the United States, it is likely that

18

policies that would effectively deport this segment of the labor force could create

significant dislocations in the American economy.

Regulating foreign labor supply through legal channels has several advantages

with respect to the imperative for national security. One of the advantages is that under a

regulated system authorized migrants present themselves for inspection at the border, in

contrast to the present system of clandestine entry. A second advantage to regulating the

labor supply through legal channels is that this would put an end to the substantial

financial costs associated with the current cat and mouse game along the border.

Moreover, it would release resources for the Border Patrol and other agencies in the new

Department of Homeland Security to concentrate their efforts on the interdiction of

diseases, drugs and terrorists, rather than on ineffective efforts to disrupt labor flows.

Among the most important benefit of regulating migration through legal channels is that

such a policy would put an end to the tragic and needless deaths of undocumented

migrants.

19

References Andreas, P.

2000 Border Games: Policing the U.S.-Mexico Divide. Ithaca, New York: Cornell

University Press.

Andreas, P.

2001 “The Transformation of Migrant Smuggling across the U.S. –Mexico Border.” In

Global Human Smuggling: Comparative Perspectives. David Kyle and Rey

Koslowski, eds. Baltimore, MD: The John Hopkins University Press.

Eschbach, K., J. Hagan, N. Rodriguez, R. Hernandez and S. Bailey

1999 “Death at the Border,” International Migration Review 33(2): 430-454.

Eschbach, K., J. Hagan and N. Rodriguez

2001 “Causes and Trends in Migrant Deaths along the U.S.-Mexico Border, 1985-1998,”

Center for Immigration Research, University of Houston. .

Espenshade,T.J., J.L. Baraka, and G.A. Huber

1997 “Implications of the 1996 Immigration Reforms,” Population and Development

Review 23(4):769-801.

Hanson, G. and A. Spilimbergo

1999 “Illegal Immigration, Border Enforcement, and Relative Wages: Evidence from

Apprehensions at the U.S.-Mexico Border.” American Economic Review 89:1337-

1357.

Kossoudji, S.

20

1992 “Playing Cat and Mouse at the U.S.-Mexico Border,” Demography, 29(2):159-180.

Massey, D., S., J. Durand, and N.J. Malone

2002 Beyond Smoke and Mirrors: Mexican Immigration in an Era of Economic

Integration. New York, New York: Russel Sage Foundation

Passel, J.S. and Fix, M. E

2001 “U.S. Immigration at the Beginning of the 21st Century,” Testimony before the

Subcommittee on Immigration and Claims Hearing on “The U.S. Population and

Immigration,” Committee on the Judiciary, U.S. House of Representatives, Urban

Institute, Washington, DC, 2001.

United States General Accounting Office

1997 Illegal Immigration: Southwest Border strategy Results Inconclusive; More

Evaluation Needed. GAO/GAD-98-164, July 31. Washington, DC: U.S.

Government Printing Office.

United States General Accounting Office

1999 Illegal Immigration: status for southwest border strategy Implementation,

GAO/GGD-99-44, May 19. Washington, DC: U.S. Government Printing Office.

United States General Accounting Office

2001 INS’ Southwest border strategy: resource and Impact Issues Remain After

Seven Years, Washington, D.C. U.S. Government Printing Office

Reyes, B., H.P. Johnson and R. Van Sweargen

2002 “Has Increased Border enforcement Reduced Unauthorized Migration?” Research

21

Brief, Public Policy Institute of California.

Suro, R.

1998 “Tightening controls and changing Flows: Evaluating the INS Border Enforcement

Strategy,” Research Perspectives on Migration 2(1). Carnegie Endowment for

International Peace.

22

Figure 1. Percentage of Apprehensions by Border Patrol Sector, 1985-2002

San Diego

El Paso

El Centro

Yuma

Tucson

Marfa

DelRio

Laredo

McAllen

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

23

Figure 2. Vital registration deaths of foreign transients and undocumented migrants

24

Figure 3. Total number of deaths, 1985-2002

0

50

100

150

200

250

300

350

400

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

U Houston

Border Patrol

25

Figure 4. Deaths by Cause

Train4%

Drowning22%Suffocation

0%

Motor Vehicle17%

Homicide16%

All Other5%

Unknown12%

Environment12%

Pedestrian12%

26

Figure 5. Trends in death by cause and year

PedestrianHomicide

Environment

Unknown

Drowning

Motor Vehicle

TrainSuffocation

All Other

0

50

100

150

200

250

300

350

400

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000