Illegal Immigration: Southwest Border Strategy Results Inconclusive; More
Evaluation Needed (Letter Report, 12/11/97, GAO/GGD-98-21).

Pursuant to a legislative requirement, GAO reviewed the Attorney
General's strategy to deter illegal entry into the United States along
the southwest border, focussing on: (1) what the strategy calls for; (2)
actions taken to implement the strategy along the southwest border; (3)
whether available data confirm the strategy's hypotheses, with respect
to expected initial results from the strategies implementation along the
southwest border; and (4) the types of indicators that would be needed
to evaluate the strategy.

GAO noted that: (1) to carry out the priority to strengthen the
detection of and deterrence to illegal entry along the border, the
Attorney General's strategy called for the Border Patrol to: (a)
allocate additional Border Patrol resources in a four-phased approach
starting first with the areas of highest known illegal activity; (b)
make maximum use of physical barriers; (c) increase the proportion of
time Border Patrol agents spend on border enforcement activities; and
(d) identify the appropriate mix of technology, equipment, and personnel
needed for the Border Patrol; (2) since the strategy was issued in 1994,
the Immigration and Naturalization Service (INS) has made progress in
implementing some, but not all, aspects of the strategy; (3) at the
southwest border land ports of entry, INS has added about 800 inspector
positions since fiscal year 1994, increasing its on-board strength to
about 1,300; (4) INS and other data indicate that some of the initial
results of the strategy's implementation along the southwest border
correspond with the expected results stated in the strategy; (5)
however, sufficient data were not available for GAO to determine whether
other expected results have occurred; (6) INS data indicated that, as a
percentage of total apprehensions along the southwest border,
apprehensions of illegal aliens have decreased in the two sectors that
in 1993 accounted for the most apprehensions and received the first
influx of new resources--San Diego and El Paso; (7) the Attorney
General's strategy for deterring illegal entry across the southwest
border envisions three distinct but related results: (a) fewer aliens
will be able to cross the border illegally; (b) fewer aliens will try to
illegally immigrate into the United States; and (c) the number of
illegal aliens in the United States will decrease; (8) evaluating the
overall effectiveness of the strategy for deterring illegal entry would
require a formal, rigorous plan for: (a) collecting and analyzing
consistent and reliable data on several different indicators related to
the three expected results from strategy; and (b) examining their
interrelationships; and (9) although developing a formal evaluation plan
and implementing a rigorous and comprehensive evaluation of the strategy
may prove to be both difficult and potentially costly, without such an
evaluation the Attorney General and Congress will have no way of knowing
whether the billions of dollars invested in reducing illegal immigration
have produced the intended results.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  GGD-98-21
     TITLE:  Illegal Immigration: Southwest Border Strategy Results 
             Inconclusive; More Evaluation Needed
      DATE:  12/11/97
   SUBJECT:  Illegal aliens
             Immigration or emigration
             Human resources utilization
             Law enforcement personnel
             Data collection
             Program evaluation
             Smuggling
             Strategic planning
             Law enforcement
             Arrests
IDENTIFIER:  El Paso (TX)
             San Diego (CA)
             INS Operation Hold-the-Line
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed 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.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to the Committee on the Judiciary, U.S.  Senate and the
Committee on the Judiciary, House of Representatives

December 1997

ILLEGAL IMMIGRATION - SOUTHWEST
BORDER STRATEGY RESULTS
INCONCLUSIVE; MORE EVALUATION
NEEDED

GAO/GGD-98-21

Illegal Immigation

(183609)


Abbreviations
=============================================================== ABBREV

  COLEF - Colegio de la Frontera Norte
  EAC - Executive Associate Commissioner
  EOIR - Executive Office of Immigration Review
  CLASS - Consular Lookout and Support System
  INS - Immigration and Naturalization Service
  IRCA - Immigration Reform and Control Act
  MMP - Mexican Migration Project
  NAWS - National Agricultural Workers Survey
  OIG - Office of Inspector General
  POE - Port of Entry
  PQIC - Port Quality Improvement Committees
  SAWS - Seasonal Agricultural Workers Survey

Letter
=============================================================== LETTER


B-276689

December 11, 1997

The Honorable Orrin G.  Hatch, Chairman
The Honorable Patrick J.  Leahy
Ranking Minority Member
Committee on the Judiciary
United States Senate

The Honorable Henry J.  Hyde, Chairman
The Honorable John Conyers, Jr.
Ranking Minority Member
Committee on the Judiciary
House of Representatives

Illegal immigration has been a long-standing problem.  The U.S. 
Department of Justice, Immigration and Naturalization Service (INS)
estimated there were about 5 million illegal aliens residing in the
United States in October 1996, and their numbers increased at an
average rate of about 275,000 per year between October 1992 and
October 1996.  Congress and states with large illegal immigrant
populations have raised concerns about illegal immigrants' fiscal
impact on government programs, participation in criminal activities,
and overall effect on local economies. 

In February 1994, the Attorney General announced a broad, five-part
strategy to strengthen enforcement of the nation's immigration laws
(see app.  I).  The strategy's first priority was to deter illegal
entry by strengthening border enforcement, particularly along the
southwest border of the United States.  Since February 1994, the
Justice Department and INS--the primary Justice Department agency
charged with enforcing the nation's immigration laws--have issued
various documents and produced strategies intended to reinforce and
better define the Attorney General's overall strategy to deter
illegal entry.  Within INS, the Border Patrol is responsible for
preventing illegal entry along the border between the nation's ports
of entry, and the Inspections program is responsible for preventing
illegal entry at the ports of entry.  For ease of presentation, and
unless otherwise noted, we use the term "strategy" in this report to
refer to various documents, including policy directives and
strategies, that have been issued by the Attorney General, INS'
Border Patrol, and INS' Inspections program as well as others. 

The Illegal Immigration Reform and Immigrant Responsibility Act of
1996 (1996 Act) mandates us to track, monitor, and evaluate the
Attorney General's strategy to deter illegal entry into the United
States and to report annually for 6 years.\1 As agreed with your
committees, this report focuses on the strategy to deter illegal
entry along the southwest border.  Specifically, the report addresses
(1) what the strategy calls for; (2) actions taken to implement the
strategy along the southwest border; (3) whether available data
confirm the strategy's hypotheses, with respect to expected initial
results from the strategy's implementation along the southwest
border; and (4) the types of indicators that would be needed to
evaluate the Attorney General's strategy to deter illegal entry along
the southwest border.  Our future reports will address various other
aspects of the Attorney General's five-part strategy to deter illegal
entry. 


--------------------
\1 P.L.  104-208, sec.  107. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

To carry out the priority to strengthen the detection of and
deterrence to illegal entry along the border, the Attorney General's
strategy called for the Border Patrol to (1) allocate additional
Border Patrol resources in a four-phased approach starting first with
the areas of highest known illegal activity; (2) make maximum use of
physical barriers; (3) increase the proportion of time Border Patrol
agents spend on border enforcement activities; and (4) identify the
appropriate mix of technology, equipment, and personnel needed for
the Border Patrol.  At ports of entry along the southwest border, the
strategy called for the Inspections program to increase inspector
staff and use additional technology to increase the deterrence to and
detection of illegal entry and improve management of legal traffic
and commerce.  The strategy also called for implementing sanctions
against aliens who attempt fraudulent entry, and using technology. 

Since the strategy was issued in 1994, INS has made progress in
implementing some, but not all, aspects of the strategy.  As of
September 1997, the Border Patrol had nearly completed Phase I of its
strategy, which was to focus resources in the areas of highest known
illegal activity (the San Diego and El Paso sectors) and was moving
into phase II, which is to focus resources in the next highest known
areas (the Tucson sector and in south Texas).  Thus, the Border
Patrol has allocated additional agents in general accordance with the
strategy.  As called for in the strategy, INS had added barriers
along the southwest border, although more barriers are reportedly
still needed.  The strategy also called for increasing border
enforcement activity, as measured by the proportion of time Border
Patrol agents collectively spent on border enforcement.  The total
number of hours spent on border enforcement activities in the
southwest border sectors has been increasing since 1994 because the
number of Border Patrol agents increased.  However, the proportion of
time Border Patrol agents at the southwest border collectively spent
on border enforcement activities did not increase between 1994 and
the first half of 1997; rather the proportion remained the same at
about 59 percent.  The Border Patrol has not identified the most
appropriate mix of staffing and other resources needed for its Border
Patrol sectors as called for in the strategy.  The Border Patrol is
developing a computerized staffing model to help it identify the
right mix of staffing and technology. 

At the southwest border land ports of entry, INS has added about 800
inspector positions since fiscal year 1994, thereby increasing its
on-board strength to about 1,300, as of March 1997.  To help achieve
its goal of increasing deterrence to illegal entry, INS has also
instituted various enforcement initiatives including increased
prosecutions of aliens presenting false documents and has been pilot
testing various technologies to improve operations.  INS has also
been planning for the implementation of various border control
requirements that Congress mandated in the 1996 Act.\2

INS and other data indicate that some of the initial results of the
strategy's implementation along the southwest border correspond with
the expected results stated in the strategy.  However, sufficient
data were not available for us to determine whether other expected
results have occurred.  According to the strategy, the concentration
of resources in phases would make it more difficult and costly for
illegal aliens to cross the border, and many would ultimately be
deterred from entering altogether.  The strategy expected (1) an
eventual reduction in illegal alien apprehensions in areas where
control is gained; (2) a shift in the flow of illegal alien traffic
from sectors that traditionally accounted for most illegal
immigration activity to other sectors; (3) increased attempts to
enter illegally through the ports of entry; (4) increases in the
costs of illegal alien smuggling, and the use of more sophisticated
tactics to smuggle illegal aliens into the country; (5) a decrease in
attempted reentries by those who have previously been apprehended;
and (6) reduced violence at the border. 

INS data indicated that, as a percentage of total apprehensions along
the southwest border, apprehensions of illegal aliens have decreased
in the two sectors that in 1993 accounted for the most apprehensions
and received the first influx of new resources--San Diego and El
Paso.  In addition, INS' apprehension data and other anecdotal data
are consistent with a shift in attempted illegal entries from the San
Diego and El Paso sectors to other sectors and from the traditional
entry corridors within the San Diego, El Paso, and Tucson sectors to
other locations within those sectors.  As the Border Patrol added
resources between the ports of entry, some ports along the southwest
border experienced increases in aliens using fraudulent documents and
making false claims to United States citizenship to attempt to enter
the United States, but other ports did not, according to INS data and
INS officials.  INS intelligence reports and other data suggest that
the fees charged by smugglers and the sophistication of smuggling
methods have increased since the implementation of the strategy, all
indicating, according to INS, an increased difficulty in illegally
crossing the border.  However, these data do not indicate whether the
increased difficulty of entry has deterred the flow of illegal
entries into the country, the ultimate goal of the Attorney General's
strategy.  Data were not available for us to determine whether there
had been a decrease in attempted reentries by those who had
previously been apprehended.  On the basis of available data, we
could not determine the extent to which the strategy had reduced
border violence. 

The Attorney General's strategy for deterring illegal entry across
the southwest border envisions three distinct but related results: 
fewer aliens will be able to cross the border illegally; fewer aliens
will try to illegally immigrate into the United States; and,
consequently, the number of illegal aliens in the United States will
decrease.  The information that INS has reported on apprehensions,
attempted fraudulent entries at ports of entry and other interim
effects, and INS intelligence reports on changes in illegal alien
traffic provide only a partial picture of the effects of increased
border control.  Evaluating the overall effectiveness of the strategy
for deterring illegal entry poses complex questions that would
require a formal, rigorous plan for (1) collecting and analyzing
consistent and reliable data on several different indicators related
to the three expected results from the strategy and (2) examining
their interrelationships.  In addition to the data INS has reported,
we identified a variety of other indicators that may be useful in
providing information on some aspects of each of the results
envisioned by the strategy to illustrate the concept of how a more
comprehensive assessment of the border strategy as a whole could be
devised. 

At present, although the Justice Department believes that systematic
evaluation of major programs and initiatives is important, it has no
formal evaluation plan to systematically evaluate the effectiveness
of the Attorney General's strategy.  INS officials told us that they
are developing a list of indicators that they believe may help INS
undertake such an evaluation.  However, key aspects of a formal
evaluation plan, such as what data will be collected, by whom, and
how the overall effectiveness of the strategy will be evaluated have
yet to be determined. 

Although developing a formal evaluation plan and implementing a
rigorous and comprehensive evaluation of the strategy may prove to be
both difficult and potentially costly, without such an evaluation the
Attorney General and Congress will have no way of knowing whether the
billions of dollars invested in reducing illegal immigration have
produced the intended results.  Developing a formal evaluation plan
would be in keeping with the concepts embodied in the Government
Performance and Results Act of 1993 (the Results Act) to develop
evaluations and performance measures to gauge whether the goals and
objectives are being achieved.  Although, in response to the Results
Act, the Justice Department's draft strategic plan described some
specific program goals, strategies, and performance indicators, it
did not contain an evaluation component to explain how the Department
will assess success in meeting these goals or, more broadly, the
effectiveness of the southwest border strategy.  A formal evaluation
plan would assist Justice in identifying whether INS is implementing
the strategy as planned, what aspects of the strategy are most
effective, and, if the strategy's goals are not being achieved, the
reasons they are not.  Such information would help the agency and
Congress identify whether changes are needed in the strategy, in
policy, in resource levels, or in program management. 


--------------------
\2 Provisions in Subtitle A, Title I of the 1996 Act require INS to
take several actions to improve enforcement at the border.  Appendix
II discusses the status of the actions not discussed elsewhere in the
report. 


   BACKGROUND
------------------------------------------------------------ Letter :2

INS' overall budget has more than doubled within 5 years, from $1.5
billion in fiscal year 1993 to $3.1 billion in fiscal year 1997.  INS
has spent about $2.3 billion on border enforcement from fiscal years
1994 through 1997.  For fiscal year 1997, the combined budget for
INS' Border Patrol and Inspections programs--the two programs
responsible for deterring illegal entry along the border--was nearly
$800 million.  INS, through other INS programs, provides additional
support for the strategy by allocating funds for computer automation,
technology procurement, and construction of barriers. 

INS' Border Patrol is responsible for preventing and detecting
illegal entry along the border between the nation's ports of entry. 
The Border Patrol has 21 sectors, 9 of which are along the southwest
border. 

   Figure 1:  INS Regions and
   Border Patrol Sector
   Headquarters Along the
   Southwest Border

   (See figure in printed
   edition.)

Source:  INS. 

The Border Patrol's appropriations were $631 million for fiscal year
1997, a 69 percent increase over its $374 million expenditure for
fiscal year 1994.  As of July 21, 1997, the Border Patrol had about
6,500 agents nationwide.  About 6,000, or 92 percent, were located in
the nine sectors along the southwest border.  In fiscal year 1996,
the Border Patrol apprehended about 1.6 million aliens nationwide, of
whom 1.5 million were apprehended in sectors along the southwest
border.  (Appendix III contains detailed staffing and selected
workload data for the Border Patrol.)

INS Inspections and the U.S.  Customs Service\3 share responsibility
for inspecting all applicants for admission at the U.S.  ports of
entry.  The purpose of their inspections is to prevent the entry of
inadmissible applicants by detecting fraudulent documents, including
those representing false claims to U.S.  citizenship or permanent
residence status, and seize conveyances used for illegal entry. 
Figure 2 depicts INS' 36 land ports of entry along the southwest
border. 

   Figure 2:  INS Land Ports of
   Entry Along the Southwest
   Border

   (See figure in printed
   edition.)

Note:  Some locations have more than one port of entry. 

Source:  INS. 

As of March 30, 1997, INS Inspections had about 1,300 inspectors at
ports of entry along the southwest border.  The Inspections'
appropriation was $151 million for fiscal year 1997, a 78 percent
increase from its $85 million expenditure for fiscal year 1994.  As
of March 15, 1997, the U.S.  Customs Service had about 7,400
inspectors nationwide.  Of these 7,400 inspectors, 2,200, or 30
percent, were located along the southwest border to inspect
individuals and cargo.  In fiscal year 1996, INS and Customs
inspectors along the southwest border inspected about 280 million
people, including 84 million, or 30 percent, who were U.S.  citizens. 
(App.  III contains detailed staffing and selected workload data for
INS Inspections.)

Within the Department of Justice, the 94 offices of the U.S. 
Attorneys are responsible for prosecuting individuals charged with
committing offenses under U.S.  law, including persons who illegally
enter the United States.  Because the Justice Department determined
that it does not have the resources necessary to prosecute all
illegal entrants, the U.S.  Attorneys located in districts along the
southwest border have instituted a policy to focus criminal
prosecutions on alien smugglers, and on those aliens without legal
documentation who are linked directly to violence and crime in the
community.  The policy calls for imposing administrative, rather than
criminal, sanctions on first-time violators who do not otherwise have
criminal histories.  In October 1995, the Attorney General appointed
the U.S.  Attorney for the Southern District of California as her
Special Representative for Southwest Border Issues.  This collateral
responsibility includes coordinating the border law enforcement
activities of various Justice Department agencies, including INS, the
Drug Enforcement Administration, and the Federal Bureau of
Investigation, with the activities undertaken by the Departments of
the Treasury and Defense. 

The Department of State also has a role in deterring illegal entry
along the southwest border.  Mexican nationals who seek to visit the
United States can obtain a border-crossing card, a type of entry
document, from either State Department consulates in Mexico or from
INS at ports of entry.  According to the State Department,
insufficient staffing levels overseas, ineffective interagency
cooperation over the exchange of data, and needed computer
enhancements all contribute to a weakening of management controls in
the visa issuance function. 


--------------------
\3 Customs and INS inspectors perform inspections at the primary
inspection booths at land ports of entry.  INS and Customs inspectors
are cross-trained and cross-designated to carry out both agencies'
inspection responsibilities at U.S.  ports of entry. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :3

To determine what the Attorney General's strategy to deter illegal
entry called for, we reviewed and summarized information on border
control strategies, plans, and directives contained in a variety of
Justice Department and INS documents related to border control and,
because the Attorney General had not published a specific strategy
for the southwest border, we prepared a summary of these documents
(see app.  I).  Not all of the documents used were specifically
identified as "strategy" documents.  Justice Department officials
reviewed this summary in May 1997 and agreed that it accurately
reflected the Attorney General's strategy and its various components
at that time. 

To determine what actions had been taken to implement the strategy
along the southwest border and whether initial results expected from
the strategy's implementation have occurred, we conducted in-person
interviews with officials from six of the nine Border Patrol sectors
along the southwest border and telephone interviews with officials
from the remaining three sectors.\4 We interviewed INS officials from
the five INS district offices responsible for all of the ports of
entry along the southwest border, INS Inspections officials from
seven ports of entry, and Customs officials from five ports of
entry.\5 In addition, we analyzed INS' Border Patrol and Inspections
workload and apprehension data, reviewed documents pertaining to INS'
management priorities, and reviewed INS intelligence reports and
previous reports done by us and the Department of Justice's Inspector
General.  We did not verify the validity of INS computer generated
data related to workload and apprehension statistics.  However, we
discussed with INS officials their data validation efforts.  These
officials were confident that the data could be used to accurately
portray trends over time.  We met with the U.S.  Attorney for the
Southern District of California, who is also the Attorney General's
Southwest Border Representative, to discuss aspects of the strategy
related to prosecuting those that violate immigration laws.  We also
discussed with INS and State Department officials the status of
various border control efforts to deter illegal entry mandated by the
1996 Act.  In conjunction with one of these efforts--improvements in
border-crossing identification cards--we visited State Department
consulates in Ciudad Juarez and Tijuana, Mexico. 

To identify indicators that could be used to evaluate the
effectiveness of the strategy in deterring illegal entry, we reviewed
illegal immigration research studies and interviewed officials from
INS and the U.S.  Commission on Immigration Reform.\6 We also
convened a meeting with a panel of immigration researchers\7 to
obtain their views on a range of evaluation issues, such as
appropriate indicators of the strategy's outcome, sources of relevant
data in addition to INS, the reliability of existing data, and how
data should be analyzed. 

We did our work between December 1996 and September 1997 in
accordance with generally accepted government auditing standards.  We
requested comments on a draft of this report from the Attorney
General, the INS Commissioner, the Acting Customs Commissioner, and
the Secretary of State or their designees.  On September 16, 1997,
the INS Executive Associate Commissioner for Policy and Planning
provided us with oral comments, which are discussed near the end of
this letter.  The Customs Service had no comments on our report and
the Department of State provided technical corrections only. 


--------------------
\4 We visited the Border Patrol sectors in San Diego and El Centro,
California; Tucson, Arizona; and El Paso, Laredo, and McAllen, Texas,
and held telephone discussions with officials from the Yuma, Arizona,
and Del Rio and Marfa, Texas, sectors. 

\5 We visited the ports of entry in San Ysidro, Otay Mesa (INS only),
and Calexico (INS only) California; Nogales Arizona; and El Paso,
Laredo, and Brownsville, Texas.  These ports were chosen because they
were in locations identified in the strategy as high priority areas. 

\6 Section 141 of the Immigration Act of 1990 (P.L.  101-649)
established the Commission to examine and make recommendations
regarding the implementation and impact of U.S.  immigration policy. 

\7 The names and affiliations of the panel participants are listed in
appendix IV. 


   THE STRATEGY
------------------------------------------------------------ Letter :4

In February 1994, the Attorney General announced a five-part strategy
to strengthen enforcement of the nation's immigration laws.  The
strategy included

  -- strengthening the border,

  -- removing criminal aliens,

  -- reforming the asylum process,

  -- enforcing workplace immigration laws, and

  -- promoting citizenship for qualified immigrants. 

The strategy to strengthen the border called for "prevention through
deterrence," that is, to raise the risk of being apprehended for
illegal aliens to a point where they would consider it futile to try
to enter the United States illegally.  The strategy was to involve
concentrating new resources on the front lines at the most active
points of illegal activity along the southwest border. 

To carry out the priority to strengthen the border, the Border Patrol
was to, among other things, (1) concentrate personnel and technology
resources in a four-phased approach, starting first with the sectors
with the highest level of illegal immigration activity (as measured
by apprehensions) and moving to the areas with the least activity;
(2) make maximum use of physical barriers to deter entry along the
border; (3) increase the proportion of time Border Patrol agents
spent on border control activities; and (4) identify the appropriate
quantity and mix of technology and personnel needed to control the
border. 

Recognizing that increased enforcement by the Border Patrol might
force some aliens to try to enter the United States illegally through
the ports of entry, the strategy calls for INS' Inspections program
to increase the number of inspectors and the use of technology to
both deter and detect illegal entry and improve management of legal
traffic and commerce.  For example, to deter illegal entry the
strategy called for increasing the number of illegal aliens referred
for prosecution and testing automated fingerprint technology to
detect inadmissible aliens.  To improve management of legal traffic,
the strategy called for providing the public with more information so
they would be better prepared for the inspection process. 

In concert with INS' efforts to deter illegal entry, the strategy
calls for increasing felony prosecutions of alien smugglers and those
criminal aliens who have repeatedly reentered the United States after
having been removed. 

In addition, the 1996 Act requires the Attorney General to take
additional border control measures to deter illegal entry into the
United States.  Appendix II discusses the status of these efforts not
discussed elsewhere in this report. 


   IMPLEMENTATION OF THE STRATEGY
------------------------------------------------------------ Letter :5

INS has made progress in implementing the Attorney General's strategy
to deter illegal entry along the southwest border.  In September
1997, Border Patrol officials told us that the Border Patrol had
nearly completed phase I of the strategy, which called for allocating
Border Patrol resources to the San Diego, California, and El Paso,
Texas, sectors.  They stated that the Border Patrol was now moving
into phase II of the strategy, which called for increasing resources
in the Tucson, Arizona, sector and three sectors in south Texas--Del
Rio, Laredo, and McAllen.  INS officials told us they could not
speculate as to when they would complete the remaining aspects of the
strategy, that called for focusing resources in the remaining three
sectors along the southwest border (phase III) and the sectors along
the rest of the U.S.  land border and coastal waterways (phase IV). 

As part of phase I of the strategy, in October 1994, INS launched a
major initiative in San Diego called Operation Gatekeeper.\8 This
multiphase, multiyear operation was designed to reduce illegal
immigration into San Diego and to force alien traffic eastward to
deter and delay illegal aliens' attempts to reach urban areas. 
Resources added to the area as a result of Gatekeeper included new
Border Patrol agents and support staff, new inspectors at the San
Ysidro port of entry, new computers and technology to maximize
efficiency, and new resources for the Office of the U.S.  Attorney
for the Southern District of California to increase its capability to
prosecute criminal aliens.  Gatekeeper focused first on the area of
the greatest illegal immigration activity--the 5-mile stretch of
Imperial Beach, California.  The next phase, which began in June
1995, included intensified enforcement at the San Ysidro port of
entry and the rural parts of the San Diego sector.  In December 1994,
we reported that on the basis of initial positive results, the
strategy appeared encouraging.\9

In August 1997, INS began the next major phase of its strategy by
concentrating resources in the McAllen sector, starting first in
Brownsville, Texas.  Named Operation Rio Grande, INS plans to add
agents and equipment, such as high-powered vision scopes and stadium
type lighting, to the McAllen sector to deter illegal entry. 

The Border Patrol has generally allocated its additional resources in
accordance with the strategy and has made progress in constructing
barriers along the southwest border.  However, the agents' percentage
of time spent on border enforcement has not increased in most
southwest border sectors since 1994, and the Border Patrol has yet to
determine the best mix of agents and technology on which to base
future staffing allocations.  INS' Inspections program has deployed
additional inspectors to the southwest border and is in the process
of pilot testing various technology initiatives designed to deter
illegal entry and streamline the inspections process.  In addition,
the U.S.  Attorneys in the five districts along the southwest border
increased the number of prosecutions for certain immigration
violations. 


--------------------
\8 In September 1993, prior to the strategy being announced, INS
began Operation Hold-the-Line, in El Paso, Texas.  In addition, in
October 1994, INS began Operation Safeguard in Arizona to respond to
the increase in apprehensions there. 

\9 Border Control:  Revised Strategy Is Showing Some Positive
Results, (GAO/GGD-95-30, Dec.  29, 1994). 


      BORDER PATROL HAS GENERALLY
      ALLOCATED NEW AGENTS AND
      OTHER RESOURCES IN
      ACCORDANCE WITH STRATEGY
---------------------------------------------------------- Letter :5.1

With some exceptions, the Border Patrol was generally able to
allocate its additional resources according to the strategy,
allocating first to sectors with the highest level of known illegal
immigration activity.  In fiscal year 1993, the San Diego and El Paso
sectors had the highest levels of apprehensions of illegal
immigrants, accounting for 68 percent of all apprehensions along the
southwest border.  In fiscal year 1994, INS received funding for an
additional 350 agents and assigned these new agents to San Diego and
El Paso, the sectors with the highest priority.  Three hundred agents
were allocated to the San Diego sector and 50 to the El Paso sector. 

The strategy noted that the Border Patrol needed to be flexible to
respond to changing patterns in illegal alien traffic.  According to
INS officials, the Border Patrol began to notice almost immediately
an increase in apprehensions in other sectors, particularly Tucson
and those in south Texas (Del Rio, McAllen, and Laredo).  INS
officials attributed this increase in apprehensions in other sectors
to a "shift" in the flow of illegal alien traffic as it became more
difficult to cross illegally in San Diego and El Paso.  Consequently,
in fiscal year 1995, the Border Patrol deployed some of the
additional agents funded that year and originally planned for San
Diego and El Paso to the Tucson and south Texas sectors, the sectors
with the next highest priority after San Diego and El Paso. 

According to Border Patrol officials, deploying additional agents in
a phased manner was a new approach.  Prior to the strategy, as
additional positions became available, the Border Patrol tried to
allocate at least a few additional positions to as many of the 21
sectors as possible.  However, under the strategy, 98 percent (or
2,792) of the 2,850 new Border Patrol agent positions nationwide
authorized from fiscal year 1994 through fiscal year 1997 have gone
to 6 of the 21 Border Patrol sectors.\10 INS allocated 1,235 (about
43 percent) of these positions to the San Diego sector and 351 (about
12 percent) to the El Paso sector, sectors with the highest priority. 
Nearly all of the remaining 1,264 went to the Tucson and the south
Texas sectors, the sectors with the next highest priority (see fig. 
3). 

   Figure 3:  Allocation of New
   Agent Positions by Southwest
   Border Patrol Sector, Fiscal
   Years 1994 Through 1997

   (See figure in printed
   edition.)

Note:  Total number of new agent positions authorized for the
southwest border for fiscal years 1994 through 1997 equals 2,842. 

Source:  INS. 

As shown in figure 4, the additional allocations have resulted in an
increase in on-board staff in most southwest border sectors. 
Overall, the number of Border Patrol agents on board along the
southwest border increased 76 percent between October 1993 and July
1997. 

   Figure 4:  Number of On-Board
   Agents by Southwest Border
   Patrol Sector, October 1993
   Compared with July 1997

   (See figure in printed
   edition.)

Note:  Total on-board agents on the southwest border as of October
1993 equals 3,389.  Total on-board agents on the southwest border as
of July 1997 equals 5,957. 

Source:  INS. 

To complement the increase in staffing, southwest border sectors have
also received additional technology, equipment, and infrastructure
improvements in accordance with the Border Patrol strategy.  For
example, between October 1994 and October 1996, the San Diego sector
added almost 5 miles of permanent high-intensity lighting, 8 miles of
reinforced steel fencing, 28 infrared scopes (a night-vision device),
and 3 helicopters to detect illegal entry.  El Paso sector officials
told us they received six additional infrared scopes and expected to
receive five more soon, and in March 1997, the sector completed a 3.5
mile fencing project.  INS purchased a building for a new Border
Patrol station in Nogales, Arizona, to house the increased number of
new agents. 

In March 1997, the Border Patrol submitted a 5-year staffing plan to
Congress covering fiscal years 1996 through 2000.  The plan calls for
adding between 1,500 and 2,500 additional Border Patrol agents in
fiscal years 1998 through 2000.  This is fewer than the 3,000
additional agents whom the 1996 Act authorized INS to hire over the
3-year period because, according to INS officials, they were
concerned that their staff was growing faster than they could
properly manage and that they did not have an adequate infrastructure
(facilities, equipment, training, and supervisory capacity) to absorb
3,000 new agents.  INS plans to assign two-thirds of the new agents
to Arizona and Texas; the remainder will go to states along the
northern border and Gulf Coast. 


--------------------
\10 Of 2,850 new positions, 2,842 were authorized for the southwest
border and 8 for Puerto Rico. 


      PROGRESS IN INSTALLING
      BARRIERS BETWEEN PORTS OF
      ENTRY
---------------------------------------------------------- Letter :5.2

One of the main efforts outlined in the strategy is the "maximum
utilization of lighting, fencing, and other barriers" by all sectors,
although the strategy did not outline specific barrier projects or
miles of fencing to be built.  A 1993 border control study,
commissioned by the U.S.  Office of National Drug Control Policy
(Sandia Study), recommended fencing and in some cases vehicle
barriers in every southwest border patrol sector to deter illegal
entry.\11

While INS has not formally endorsed all of the Sandia Study
recommendations, INS Headquarters Border Patrol officials told us
that some recommendations, such as erecting 90 miles of barriers
along the southwest border, are valid.  These officials told us that,
while adding barriers is part of the strategy, INS has left it up to
each sector chief to propose where and when to build barriers.  Seven
of the nine sector plans written to carry out the strategy cite the
need for barriers to increase agents' effectiveness in apprehending
illegal aliens and reducing crime.  According to INS officials,
proposals for barrier projects are reviewed in the context of INS'
budget process; require consultation with Congress; and must be
coordinated with the U.S.  Department of Defense Joint Task Force
Six, the military unit that constructs much of the fencing for INS. 

Congress has also emphasized the need for additional fencing.  The
1996 Act requires the Attorney General, in consultation with the
Commissioner of INS, to take such actions as may be necessary to
install additional physical barriers and roads in the vicinity of the
U.S.  border to deter illegal crossings in areas of high illegal
entry into the United States.  In carrying out this provision in the
San Diego sector, the 1996 Act requires the Attorney General to
provide for second and third fences, in addition to the existing
reinforced fence, and for roads between the fences for the 14 miles
of the international land border of the United States extending
eastward from the Pacific Ocean.\12

INS has allocated $8.6 million of its fiscal year 1996 and 1997
appropriations to complete the first two phases of the triple fencing
project in the San Diego sector.  Of the $8.6 million allocated,
about $4.3 million had been obligated for expenditure as of July
1997.  According to a San Diego sector official, INS is in the
process of acquiring the property upon which to build the fencing and
conducting environmental assessment reports, before construction can
begin in certain areas.  Figure 5 depicts bollard-type (concrete
cylindrical columns set in a staggered manner) fencing\13 which is
being constructed in the area immediately inland from the Pacific
Ocean in the San Diego sector. 

   Figure 5:  Bollard Fencing
   Being Constructed in San Diego,
   California

   (See figure in printed
   edition.)

Source:  INS. 

Prior to 1994, very little substantial fencing existed--about 14
miles of reinforced steel fencing in the San Diego sector.  Since the
strategy was announced, INS has built approximately 32 miles of new
fencing in five sectors.  As of July 1997, nearly 24 miles of
additional fencing was under construction.  Most of the fencing
constructed, under construction, or planned to be built is in the San
Diego sector.  Currently, no barriers exist or are planned to be
built in four Texas sectors (Marfa, Del Rio, Laredo, and McAllen). 
However, two of these sectors (Laredo and McAllen) had indicated a
need for barriers in their sector plans.  INS officials cited the
need to overcome community concerns as a reason why they have not
built more barriers. 



                                Table 1
                
                 Status of Barriers Along the Southwest
                                 Border

                             Miles                   Miles
                             built       Miles       under
                            before       built  constructi       Miles
Border Patrol sector          1994  since 1994          on     planned
----------------------  ----------  ----------  ----------  ----------
San Diego                     14.0        18.7       5.5\a         2.0
El Centro                        0         3.0           0         3.0
Yuma                             0         4.5           0         3.0
Tucson                           0       2.0\b         3.5         2.0
El Paso                          0         3.5        14.0           0
All other sectors\c              0           0           0           0
======================================================================
Total                         14.0        31.7        23.0        10.0
----------------------------------------------------------------------
\a As mandated in the 1996 Act, 14 miles of secondary fencing to
supplement the primary border fence is also being built. 

\b Does not include 2 miles of fencing in Naco, Arizona, since INS
did not fund the construction. 

\c Includes Marfa, Del Rio, Laredo, and McAllen, Texas sectors. 

Source:  GAO analysis of INS data. 

The majority of the fencing construction along the southwest border
has been accomplished by the U.S.  Department of Defense Joint Task
Force Six and, to a lesser degree, the National Guard.  The military
provides the personnel to construct the fencing and pays for their
salaries, transportation, meals, and housing.  INS typically pays for
building materials (although steel runway landing mats are provided
at no cost because they are military surplus) and other costs, such
as equipment rental.  INS employees, such as repair and maintenance
staff, also help with the construction.  In a few instances, private
contractors have been hired by INS to construct particular types of
fencing, such as the bollard fencing project in the San Diego sector. 
However, INS officials stated that the use of private contractors is
much more costly to INS than using military assistance. 


--------------------
\11 Systematic Analysis of the Southwest Border, Sandia National
Laboratories, Albuquerque, New Mexico, January 1993. 

\12 Sec.  102(a) and (b). 

\13 According to INS officials, the fencing is divided into (1) 1.8
miles of bollard fencing; (2) 1.5 miles of steel mesh fencing, about
15 feet in height; and (3) other type of fencing still in the design
phase to cover the remainder of the 14 miles. 


      PROPORTION OF TIME SPENT ON
      BORDER ENFORCEMENT DID NOT
      INCREASE IN MOST SOUTHWEST
      BORDER SECTORS
---------------------------------------------------------- Letter :5.3

The strategy called for an increasing border enforcement activity, as
measured by the proportion of time Border Patrol agents spent on
border enforcement.  According to INS data, agents in the nine
sectors along the southwest border collectively spent 59 percent of
their total time on border enforcement in the first half of fiscal
year 1997, nearly reaching the servicewide goal of 60 percent. 
However, this proportion of time spent on border enforcement
activities was the same as that in fiscal year 1994.  Although the
proportion of time spent on border enforcement activities did not
increase during this period, the total number of hours spent by all
Border Patrol agents on border enforcement activities along the
southwest border has been increasing since fiscal year 1994, because
the overall number of Border Patrol agents assigned to the southwest
border sectors increased. 

In five of the nine southwest border sectors, the proportion of time
that Border Patrol agents spent on border enforcement in the first
half of fiscal year 1997 was within 2 percentage points (plus or
minus) of what it was in fiscal year 1994.  In two sectors, Tucson
and El Paso, the proportion of time spent on border enforcement
decreased by 7 and 13 percentage points, respectively.  In two other
sectors, San Diego and Marfa, it increased by 7 and 10 percentage
points, respectively (see fig.  6). 

   Figure 6:  Percentage of
   Agents' Time Spent on Border
   Enforcement by Southwest Border
   Sector, Fiscal Year 1994
   Compared with First Half of
   Fiscal Year 1997

   (See figure in printed
   edition.)

Source:  GAO analysis of INS data. 

INS officials reported that an increase in the amount of time spent
on program support activities such as supervision, training, and
processing apprehended aliens, as well as possible reporting errors,
could have hindered increasing the overall percentage of time spent
on border control activities.  For example, during fiscal year 1994,
the El Paso sector spent 20 percent of its time on program support
compared with 33 percent during the first half of fiscal year 1997,
an increase of 13 percentage points.  During this same time period,
the Tucson sector's percentage of time on program support increased
from 41 percent to 50 percent, an increase of 9 percentage points. 

Consistent with the strategy, southwest border sectors have
redirected Border Patrol agents from general enforcement\14
activities back to border control.  For example, the San Diego sector
reduced the proportion of time spent on general enforcement from 4
percent in fiscal year 1994 to less than 1 percent for the first half
of fiscal year 1997.  The Marfa sector reduced the proportion of time
spent on general enforcement from 14 percent to 2 percent during this
same time period.  As a result, Marfa increased its border
enforcement percentage although it did not receive additional staff
during this period. 

Servicewide, the proportion of time spent on border enforcement
declined from 56 percent in fiscal year 1994 to 50 percent in the
first half of fiscal year 1997.  Thus, the performance of INS as a
whole on this measure declined rather than increased as intended. 
INS has lowered its expectations regarding the percentage of time,
servicewide, that it expects its Border Patrol agents to spend on
border enforcement.  According to its fiscal year 1998 budget
submission, INS' 1998 servicewide goal is for the Border Patrol to
spend 56 percent of its time on border enforcement compared with the
60 percent goal for fiscal year 1997. 


--------------------
\14 General enforcement activities include checking employers for
illegal workers and visiting local jails and state prisons to locate
and process criminal aliens. 


      APPROPRIATE MIX OF AGENTS'
      STAFFING AND MATERIEL
      RESOURCES HAD NOT BEEN
      DETERMINED
---------------------------------------------------------- Letter :5.4

The strategy states that it will "seek the best mix of technology and
personnel resources to meet the long term goals," and that
"improvements in technology will make border control strategies more
effective and less resource intensive." The Border Patrol has been
increasing its supply of equipment and advanced technologies.  The
conference report for the fiscal year 1997 Department of Justice
Appropriations Act includes $27 million for infrared scopes,
low-light television systems, sensors,\15 and the replacement of
three helicopters, including upgraded forward-looking infrared
systems.\16 Since 1994, the San Diego sector alone acquired an
additional 28 infrared scopes, about 600 underground sensors, about
500 vehicles, about 600 computers, and several advanced computer
systems. 

The Border Patrol has not identified the most appropriate mix of
staffing and other resources needed for its sectors.  Headquarters
officials told us that sector chiefs may have taken current
technological assets into consideration when developing their sector
staffing proposals.  However, according to these officials, sector
staffing proposals generally did not include a discussion of the
potential impact on staffing needs of adding barriers and/or
technology.  In addition, when allocating the additional 1,000 Border
Patrol agents funded for fiscal year 1997 to the various sectors, the
Border Patrol did not formally consider how adding barriers and/or
technology would potentially affect staffing needs. 

In the 5-Year Border Patrol Deployment Plan submitted to Congress in
March 1997, the Border Patrol stated that in fiscal year 1998, it
would "assess technology improvements in sensors, scopes, biometrics
identification systems, etc., and effects on staffing requirements";
and in fiscal year 1999, it would "implement staffing changes based
on technology assessments." With the help of a contractor, the Border
Patrol is currently working on developing a computerized staffing
model to help it identify the right mix of staffing and technology. 
According to Border Patrol officials, the model will allow INS to
estimate the impact of different levels of materiel resources on
sectors' staffing levels and effectiveness in apprehending illegal
aliens.  As of June 1997, the model was being tested in the El Paso
sector.  INS officials plan to have this model operational in El Paso
by December 1997 and in all southwest border sectors by the summer of
1998. 

INS headquarters officials told us that they were also testing new
technologies, such as weight-sensitive sensors and satellite global
positioning systems, to determine their usefulness in Border Patrol
operations.  They told us that they have yet to determine how these
new technologies might be integrated into border control operations
and what their impact on agent needs might be. 


--------------------
\15 Sensors include devices buried in the ground used to detect
either foot or vehicular traffic that may have crossed the border
illegally. 

\16 Departments of Commerce, Justice, and State, the Judiciary, and
Related Agencies Appropriations Act, P.L.  104-208; H.R.  Conf.  Rep. 
104-863, at 788 (1996). 


      RESOURCES, ENFORCEMENT
      INITIATIVES, AND TECHNOLOGY
      TESTING INCREASED AT PORTS
      OF ENTRY
---------------------------------------------------------- Letter :5.5

The strategy postulated that enhanced enforcement efforts between the
ports of entry would cause an increase in port-of-entry activity,
including increased attempts to enter through fraudulent means.\17 To
handle the increased activity, Congress authorized an increase of
about 800 inspectors for southwest ports of entry since 1994 (see
fig.  7), almost doubling the number of authorized positions, from
about 865 in 1994 to 1,665 in 1997.  As of March 1997, 1,275
inspectors were on board at land ports of entry along the southwest
border (see fig.  8). 

   Figure 7:  Allocation of New
   Inspectors to Land Ports of
   Entry Along the Southwest
   Border by INS District Office,
   Fiscal Years 1994 through 1997

   (See figure in printed
   edition.)

Note:  Total number of new inspector positions authorized for fiscal
years 1994 to 1997 equals 796.  Ports of entry fall under the
administration of district offices. 

Source:  INS. 

   Figure 8:  Number of On-Board
   Inspectors at Land Ports of
   Entry Along the Southwest
   Border by INS District Office,
   September 1994 Compared With
   March 1997

   (See figure in printed
   edition.)

Note:  Total number of on-board inspectors on the southwest border as
of September 1994 equals 703.  Total number of on-board inspectors on
the southwest border as of March 1997 equals 1,275.  Ports of entry
fall under the administration of district offices. 

Source:  INS. 

INS has begun testing a number of programs and systems to increase
deterrence to illegal entry and improve and streamline the inspection
process.  These efforts are intended to prevent illegal aliens and
criminal violators from entering the United States and facilitate the
entry of legal travelers.  To accomplish these objectives, INS is
using technology to segment people seeking admission by risk category
and forming strategic partnerships with others concerned with border
management, such as the Customs Service, local communities, and the
Mexican government.  INS is also planning to measure its
effectiveness in detecting illegal entry attempts. 

According to port-of-entry officials, the additional INS inspectors
have enabled INS to staff more inspection booths at peak hours, and
allowed both INS and Customs to increase enforcement efforts and
undertake some new initiatives to deter illegal entry.  For example,
inspectors at some ports of entry are spending more time inspecting
vehicles before they reach the inspection booth to detect concealed
drugs or smuggled aliens. 

INS believes that increased sanctions will result in deterring
illegal aliens from attempting to enter the United States
fraudulently.  In response to an increase in attempted fraudulent
entries at ports of entry in the San Diego area, the Justice
Department established the first permanent immigration court facility
to be located at a port of entry.  The "port court," located at the
Otay Mesa port of entry, began as a pilot project in July 1995 and
was made permanent in October 1995.  This program was intended to
eliminate costly and time-consuming transportation of aliens to the
immigration court in downtown San Diego and allowed for the immediate
implementation of the immigration judge's order of exclusion and
deportation. 

In addition, INS inspectors in San Diego worked with the U.S. 
Attorney for the Southern District of California to develop a program
intended to increase prosecutions of persons attempting to enter the
United States fraudulently.  However, the ability to prosecute may be
hampered by the limited availability of detention space, according to
Inspections officials.  Provisions in the 1996 Act that took effect
April 1, 1997, provide for the expedited removal of certain aliens
who attempt fraudulent entry.  INS officials believe these provisions
may also help deter attempted fraudulent entry. 

INS has several pilot projects under way that use technology to try
to segregate low- and high-risk traffic and streamline the
inspections process.  In the San Diego area, a dedicated commuter
lane is in operation.  INS authorizes certain frequent crossers and
their vehicles to enter the United States through a preclearance
process.  Through an automated photo identification and card system,
registered vehicles and occupants can pass through the port of entry
quickly.\18 INS is testing other technology, including automated
vehicle license-plate readers to check vehicles against law
enforcement lookout databases and a system which uses palm prints and
fingerprints to verify the identity of individuals in order to reduce
passenger processing time. 

In addition, according to INS headquarters and most port officials we
interviewed, INS and Customs have increased cooperation between their
agencies and are working together to manage the ports in a more
efficient manner.  In 1993, we reported on the lack of cooperation
and coordination of border crossing operations as well as a long
history of interagency rivalries between INS and Customs.\19 We
recommended that the Director of the Office of Management and Budget
(OMB), working with the Secretary of the Treasury and the Attorney
General, develop and present to Congress a proposal for ending the
dual management of border inspections.  As a result of our report,
INS and Customs formed Port Quality Improvement Committees (PQIC) at
selected ports of entry, including some along the southwest border. 
A 1996 follow-up report on the Vice President's National Performance
Review indicated that the PQIC structure encouraged and strengthened
cooperation and communication among officers of all federal
inspection service personnel.\20 In addition, Customs and INS reviews
have found that better coordination and improved services have been
achieved through PQICs.  Four of the ports we visited (San Ysidro, El
Paso, Nogales, and Brownsville) had PQICs and had implemented various
cooperative initiatives to facilitate border crossing (e.g.,
increased cross-training).  We did not independently assess whether
the PQICs have resulted in better coordination and cooperation
between INS and Customs because such an assessment was beyond the
scope of this review. 

As part of its efforts to measure performance, as required by the
Results Act, INS plans full implementation in fiscal year 1998 of a
port performance measurement system, which is to include randomly
selecting applicants who seek to cross into the United States and
processing them through a more rigorous inspection.  One of the goals
of this system is to project the estimated number of immigration
related violations.  According to INS headquarters officials, this
system will ensure the effectiveness of inspections at the officer
level and will allow for evaluation of overall program performance. 


--------------------
\17 Fraudulent entries could include persons presenting counterfeit
documents or altered immigration documents; persons making a
documented false claim of U.S.  citizenship, such as presenting a
counterfeit U.S.  birth certificate; persons making oral false claims
to U.S.  citizenship; or persons being smuggled through the port of
entry concealed in vehicles. 

\18 INS performs random inspections of vehicles in the dedicated
commuter lane, and the use of the lane is not for commercial entry of
goods. 

\19 Customs Service and INS:  Dual Management Structure for Border
Inspections Should Be Ended, (GAO/GGD-93-111, June 30, 1993). 

\20 Reengineering United States Primary Passenger Processing, First
Year Summary Report, National Performance Review, 1996. 


      PROSECUTIONS
---------------------------------------------------------- Letter :5.6

The strategy called for increasing prosecutions for immigration
related violations.  As part of the strategy, U.S.  Attorneys in the
five judicial districts contiguous to the southwest border have
developed federal prosecution policies to, among other things, target
criminal aliens, smugglers, and those who attempt entry by using
false documents.  The strategy also outlines the expanded use of
administrative sanctions through immigration court orders. 

U.S.  Attorneys in the five districts along the southwest border have
increased the number of prosecutions since 1994.  For example, in
fiscal year 1994, the 5 districts filed about 1,000 cases involving
about 1,100 defendants related to 3 major immigration violations.\21
The Justice Department projected that these 5 districts would file
about 3,800 such cases in fiscal year 1997, involving over 4,000
defendants, more than tripling the number of cases and defendants.\22
The U.S.  Attorney for the Southern District of California, which
includes San Diego, projected that his office would file about 1,900
such cases in fiscal year 1997, over 6 times the 290 filed in 1994. 


--------------------
\21 These are 8 U.S.C.  1324, which provides criminal penalties for
bringing in or harboring aliens; 8 U.S.C.  1325, which provides
criminal penalties for improper entry by aliens; and 8 U.S.C.  1326,
which provides criminal penalties for reentry of an alien after
deportation. 

\22 Fiscal year 1997 numbers are a straight-line projection based on
actual data through the end of March 1997.  Fiscal year 1997 actual
statistics were not available at the end of our field work. 


   SOME ANTICIPATED INTERIM
   EFFECTS OF STRATEGY ARE
   OCCURRING
------------------------------------------------------------ Letter :6

As the strategy along the southwest border is carried out, the
Attorney General anticipated the following changes in certain
indicators would provide evidence of the interim effectiveness of the
strategy: 

(1) an initial increase in the number of illegal aliens apprehended
in locations receiving an infusion of Border Patrol resources,
followed by a decrease in apprehensions when a "decisive level of
resources" has been achieved, indicating that illegal aliens are
being deterred from entering;

(2) a shift in the flow of illegal alien traffic from sectors that
traditionally accounted for most illegal immigration activity to
other sectors as well as shifts within sectors from urban areas where
the enforcement posture was greater to rural areas;

(3) increased attempts by illegal aliens to enter illegally at the
ports of entry, as it becomes more difficult to enter between the
ports;

(4) an increase in fees charged by alien smugglers to assist illegal
aliens in crossing the border and more sophisticated smuggling
tactics;

(5) an eventual decrease in attempted reentries by those who have
previously been apprehended (recidivism); and

(6) reduced violence at the border. 

According to the strategy, changes in the predicted direction on
these indicators would be evidence that INS enforcement efforts
effectively raised the cost and difficulty of entering the United
States illegally.  Ultimately, the strategy posits that there would
be fewer illegal aliens in the United States and reduced use of
social services and benefits by illegal aliens. 

Data on the interim effects of the strategy have been collected and
reported primarily by INS, and their interpretation is not clear-cut. 
The available data suggest that some of the predicted changes have
occurred.  For example, INS data indicate that

(1) there was a period after additional resources were applied to the
San Diego sector in which Border Patrol apprehensions increased in
the sector and a subsequent period in which apprehensions decreased;

(2) there has been a shift in illegal alien traffic from sectors that
traditionally accounted for most illegal immigration activity to
other sectors as well as shifts within some sectors;

(3) there were increased numbers of illegal aliens attempting to
enter illegally at some ports of entry; and

(4) alien smuggling fees may have increased, and smuggling tactics
may have become more sophisticated. 

Data were unavailable on whether there has been a decrease in
attempted reentries by those who have previously been apprehended,
and data on violence at the border were inconclusive. 


      CHANGES IN ILLEGAL ALIEN
      APPREHENSIONS
---------------------------------------------------------- Letter :6.1

Apprehension statistics are routinely reported by INS, and they are
INS' primary quantitative indicator of the results of the strategy. 
Although an effective strategy should affect apprehensions,
apprehension data, standing alone, have limited value for determining
how many aliens have crossed the border illegally.  (A later section
discusses the limitations of apprehension data more fully.)

According to INS data for the San Diego sector, after an increase in
apprehensions, as resources were applied, sector apprehensions
eventually began to decrease.  According to an INS analysis of
seasonally adjusted San Diego sector apprehension data from October
1992 to March 1997 (see fig.  9), monthly sector apprehensions were
on a downward trend from February 1993 through December 1994.  In
January 1995, 3 months after the sector began applying Operation
Gatekeeper resources in the western part of the San Diego sector,
apprehensions began increasing.  Apprehensions continued to increase
for about 1 year.  Beginning in January 1996, apprehensions started
to decline and continued to do so through March 1997 (the end point
of the INS analysis).  The last decline in apprehensions coincided
with the addition of Border Patrol agents, barriers, and technology
to areas of the San Diego sector that were east of the original
Gatekeeper effort. 

   Figure 9:  Apprehensions
   (seasonally adjusted) in the
   San Diego Border Patrol Sector,
   October 1992 to March 1997

   (See figure in printed
   edition.)

Source:  INS. 

It is difficult to determine whether the increase in apprehensions
experienced in 1995 is due to increased enforcement or other factors. 
In December 1994, the Mexican government devalued the peso. 
According to INS officials and INS reports, apprehensions in the San
Diego sector could have increased in part due to the strategy and in
part due to an increase in illegal flow resulting from poor economic
conditions in Mexico and the associated devaluation of the peso.  It
is also difficult to determine whether the decline in apprehensions
that began in January 1996 was part of the original downward trend
predicted by the strategy or a specific result of the spring
initiative--INS' 1996 enhancement to Operation Gatekeeper--in which
additional resources were applied to the eastern parts of the San
Diego sector.\23

In El Paso, the pattern of apprehensions following implementation of
a separate border enforcement initiative, Operation Hold the Line,
differed from that of San Diego.  In this operation, begun in
September 1993, the sector redeployed its agents directly to a
20-mile section of the border in the metropolitan El Paso area
adjoining Ciudad Juarez in Mexico and maintained a high-profile
presence that was intended to deter illegal aliens from attempting to
cross the border.  According to an INS analysis of seasonally
adjusted apprehension data, apprehensions decreased in the El Paso
sector immediately after the initiative was launched, and after
declining for a period of 1 month, apprehensions began to increase
(see fig.  10).  Although the Border Patrol has continued Operation
Hold-the-Line and added new agents to the sector between fiscal years
1994 and 1997, apprehension levels began to increase in November
1993, and have generally continued to do so through March 1997,
although remaining at levels below those that existed before
September 1993. 

   Figure 10:  Apprehensions
   (seasonally adjusted) in the El
   Paso Border Patrol Sector,
   October 1992 to March 1997

   (See figure in printed
   edition.)

Source:  INS. 


--------------------
\23 In July 1996, the Department of Justice Office of the Inspector
General (OIG) began a broad investigation into allegations by members
of the Border Patrol union that Operation Gatekeeper was
characterized by the manipulation of procedures and data to create
the false impression that the initiative had successfully deterred
illegal immigration into the San Diego sector.  The OIG investigation
has been completed, although a report on the results has not yet been
issued.  In August 1997, OIG officials told us that their report on
Gatekeeper, which they expect to issue in November 1997, will not
affirm that pervasive falsification of Gatekeeper data occurred. 


      SHIFT IN ILLEGAL ALIEN
      TRAFFIC
---------------------------------------------------------- Letter :6.2

The Border Patrol strategy directed additional enforcement resources
first to the San Diego and El Paso sectors, where the majority of
illegal entries have historically occurred.  INS expected that the
flow of illegal alien traffic would shift from San Diego and El Paso
to other sectors as control was achieved.  Our analysis of INS
apprehension data indicate that a shift in apprehensions has
occurred.  As shown in figure 11, in the first 6 months of fiscal
year 1993 the San Diego and El Paso sectors accounted for 68 percent
of all southwest border apprehensions.  However, during the first 6
months of fiscal year 1997, San Diego and El Paso accounted for 33
percent of all southwest border apprehensions.  Other sectors now
account for a larger share of the apprehensions.  For example, in the
first 6 months of fiscal year 1993, the Tucson sector accounted for 7
percent of all southwest border apprehensions.  During the first half
of fiscal year 1997, the sector's share rose to 19 percent.  The
proportion of southwest border apprehensions of the three south Texas
sectors--McAllen, Laredo, and Del Rio--rose from 19 percent to 37
percent over the same period. 

   Figure 11:  Apprehensions in
   San Diego, El Paso, and Other
   Southwest Border Patrol
   Sectors, First Half of Fiscal
   Years 1993 and 1997

   (See figure in printed
   edition.)

\a Includes El Centro, Yuma, Tucson, Marfa, Del Rio, Laredo, and
McAllen. 

Source:  GAO analysis of INS data. 

The Border Patrol's enforcement posture aimed to reduce illegal
entries into large urban areas, thereby, forcing illegal alien
traffic to use rural routes where the Border Patrol believes it has a
tactical advantage.  Our analysis of INS apprehension data shows that
within the San Diego, El Paso, and Tucson sectors, apprehensions have
decreased in areas that have received greater concentrations of
enforcement resources and increased in more remote areas.  As shown
in figure 12, in the San Diego sector, the stations of Imperial Beach
and Chula Vista, which provide the shortest established routes to
urban San Diego, accounted for 61 percent of the sector apprehensions
in the first half of fiscal year 1993.\24 In the first half of fiscal
year 1997, these two stations accounted for 39 percent of the
sector's apprehensions.  Conversely, in the other stations, largely
in rural sections of the sector, the share of total apprehensions
increased from 39 percent to 61 percent over the same time period. 

   Figure 12 :  Apprehensions by
   Border Patrol Stations in the
   San Diego Sector, First Half of
   Fiscal Years 1993 and 1997

   (See figure in printed
   edition.)

Source:  GAO analysis of INS data. 

In the El Paso sector, the proportion of apprehensions in the urban
El Paso station dropped from 79 percent of all sector apprehensions
in the first 6 months of fiscal year 1993, to 30 percent in the first
6 months of fiscal year 1997, as shown in figure 13.  A 1994 study
commissioned by the U.S.  Commission on Immigration Reform concluded
that Operation Hold-the-Line appeared to have substantially deterred
some types of illegal crossers but not others.\25 Using official
statistics on apprehensions and crossings at the ports of entry
(e.g., police and crime data, birth and hospital data, education and
school attendance statistics, sales tax and general sales data) and
qualitative information from in-depth interviews with government
officials and persons at border crossing sites in El Paso and Ciudad
Juarez, the study concluded that the operation had been more
successful in curtailing illegal immigration among aliens who crossed
illegally from Ciudad Juarez to engage in illegal work or criminal
activity in El Paso (local crossers) than among aliens who crossed at
El Paso but were headed for other U.S.  destinations (long-distance
labor migrants).  According to the study, a substantial amount of
long-distance labor migration appeared to have been diverted to other
locations along the southwest border.  Border Patrol officials in the
El Paso sector told us in March 1997 that the majority of illegal
aliens entering the sector were long-distance migrants, so they
believed that their enforcement efforts were continuing to deter
local crossers. 


   Figure 13:  Apprehensions by
   Border Patrol Station in the El
   Paso Sector, First Half of
   Fiscal Years 1993 and 1997

   (See figure in printed
   edition.)

Source:  GAO analysis of INS data. 

Apprehension data also provide some support that a shift occurred in
illegal alien traffic from the urban area of Nogales, Arizona,
targeted by the Border Patrol in the first phase of its enforcement
strategy in the Tucson sector, to other stations in the sector.  As
shown in figure 14, in the first 6 months of fiscal year 1993, the
Nogales station accounted for 39 percent of all sector apprehensions. 
In the first 6 months of fiscal year 1997, the station accounted for
28 percent of all sector apprehensions.  Over this same period,
apprehensions have increased in the city of Douglas, Arizona, from 27
percent to 43 percent of all sector apprehensions. 

   Figure 14:  Apprehensions by
   Border Patrol Station in the
   Tucson Sector, First Half of
   Fiscal Years 1993 and 1997

   (See figure in printed
   edition.)

Source:  GAO analysis of INS data. 

Other information also indicates that there has been a shift in alien
traffic.  According to an INS report on Operation Gatekeeper, data
from INS' automated fingerprinting system, known as IDENT, showed
that illegal aliens were less likely to try to cross in Imperial
Beach in January 1995 than in October 1994.  According to an April
1996 report summarizing the results from an INS Intelligence
conference held in the Del Rio sector, potential illegal aliens had
been channeled away from the U.S.  urban areas on the southwest
border to more inhospitable areas.  The report stated that a sizeable
number of apprehensions were being made in extremely desolate areas
of the border, and this was taken to be an indication that illegal
aliens were trying to avoid Border Patrol deployments.  However,
according to the report, INS lacked the resources to apprehend aliens
traversing the remote areas.  The report suggested that the Border
Patrol should not abandon its high-visibility deterrent posture in
the urban areas to respond to the remote areas because it would
encourage a return to the earlier entry patterns. 

According to some INS sector officials, shifts in illegal alien
traffic have had a negative effect in their sectors.  For example,
according to a January 1997 Laredo sector report, continued media
coverage of Operations Hold-the-Line and Gatekeeper have discouraged
entries in those areas and made Laredo vulnerable to would-be
crossers.  Border Patrol officials in the Del Rio sector told us in
May 1997 that because of substantial increases in illegal alien entry
attempts and limited resources, the sector had limited success in
controlling the two main corridors within the sector.  According to
INS headquarters officials, such shifts in illegal traffic are not
failures of the strategy; rather, they are interim effects. 

The strategy has also produced some effects that INS did not
anticipate.  INS officials told us that they were in some cases
caught unaware by some of the changes in illegal alien traffic and
the tactics of illegal crossers.  For example, INS Western Region
officials told us that the increases in illegal alien traffic that
they predicted would occur in the Tucson, Arizona, and McAllen,
Texas, sectors happened earlier than they expected.  According to
these officials and San Diego sector officials, the mountains in the
eastern section of the San Diego sector were expected to serve as a
natural barrier to entry.  They were surprised when illegal aliens
attempted to cross in this difficult terrain as it became more
difficult to cross in the urban sections of the sector. 


--------------------
\24 The nine Border Patrol sectors along the southwest border are
each divided into a number of stations, with responsibility for a
section of the border or a geographic area further in the interior of
the United States.  Most sectors also maintain several checkpoints
along major highways, designed to intercept aliens who have eluded
agents along the border. 

\25 F.D.  Bean, R.  Chanove, R.G.  Cushing, R.  de la Garza, G.P. 
Freeman, C.W.  Haynes, & D.  Spener, Illegal Mexican Migration & the
United States/Mexico Border:  The Effects of Operation Hold-the-Line
on El Paso/Juarez (Austin, TX:  Population Research Center,
University of Texas at Austin, and U.S.  Commission on Immigration
Reform), July 1994. 


      CHANGES IN ILLEGAL ENTRY
      ATTEMPTS AT PORTS OF ENTRY
---------------------------------------------------------- Letter :6.3

The strategy anticipated increases in the flow of illegal traffic
through the ports of entry as it became more difficult to cross
between the ports.  According to INS regional and district officials,
some of these anticipated increases had begun to occur.  Officials
from the San Diego district told us that they had seen large
increases in attempts to enter the United States using fraudulent
documents or making false claims to U.S.  citizenship immediately
following the increase of resources to the Border Patrol in the San
Diego sector.  According to San Diego district inspections data, the
number of fraudulent documents intercepted increased about 11 percent
from fiscal year 1994 to 1995 (from about 42,000 to about 46,500),
and the number of false claims to U.S.  citizenship increased 26
percent (from about 15,400 to about 19,400).  Officials in El Paso
told us they uncovered more fraudulent documents after the initiation
of Operation Hold-the-Line.  El Paso district inspections data show a
steady increase in the number of fraudulent documents intercepted
from fiscal year 1994 (about 8,200) through fiscal year 1996 (about
11,000).  The Del Rio intelligence conference report stated that
Border Patrol enforcement efforts had forced more people to attempt
entry fraudulently through the ports of entry.  Officials from the
Phoenix district, however, told us that they had not seen any
significant change in the number of fraudulent documents intercepted
at Arizona ports of entry, as the Border Patrol resources increased
in the Tucson, Arizona sector. 

INS Western Region and San Diego District officials told us that they
had not expected the volume of people trying to enter illegally at
the ports of entry in the San Diego areas to be as large as it was
after Operation Gatekeeper or the tactics illegal aliens chose to try
to get past the inspectors.  For example, at various times during
1995, illegal aliens gathered at the entrance to the San Ysidro port
of entry and attempted to overwhelm inspectors by running in large
groups through the port.  INS created emergency response teams at the
port to deal with these unexpected tactics and added bollards and
other physical barriers to make unimpeded passage more difficult.  In
addition, INS asked Mexican government officials to help prevent
large gatherings on the Mexican side of the border.  During our visit
to San Ysidro in March 1997, INS officials told us that due to these
actions, large groups of port-runners were no longer a problem. 

It is difficult to determine whether the increases in number of
fraudulent documents intercepted and false claims to U.S. 
citizenship were a result of actual increases in illegal entry
attempts at the ports or greater efforts made to detect fraud.  The
1994 U.S.  Commission on Immigration Reform evaluation of Operation
Hold-the-Line noted that INS inspectors at the ports of entry in El
Paso began checking documents more closely after the Border Patrol
instituted the operation, which may have contributed to the increase
in recorded illegal entry attempts.\26

INS headquarters officials disagreed with these findings.  They
stated that they believed that inspectors in El Paso were exposed to
heavier traffic flows after the operation began and, therefore,
inspectors could have compensated for the increased workload by doing
more cursory reviews of documents.  These headquarters officials
stated that increases in detected fraud reflected actual increases in
fraudulent entry attempts, which were a response to heightened
enforcement between the ports of entry caused by the Border Patrol's
implementation of Operation Hold-the-Line. 


--------------------
\26 Bean et al, p.  116. 


      ALIEN SMUGGLING
---------------------------------------------------------- Letter :6.4

INS expected that if successful, its enforcement efforts along the
southwest border would make it more difficult and costly for illegal
aliens and alien smugglers to cross the border.  INS postulated that
this should be reflected in higher fees charged by alien smugglers
and more sophisticated tactics used by smugglers to evade capture by
INS. 

Fees charged by smugglers, and the sophistication of smuggling
methods have reportedly increased since fiscal year 1994.  On the
basis of testimony of the U.S.  Attorney for Southern California, INS
evaluation reports on enforcement efforts in San Diego, and INS
intelligence reports, fees paid by illegal aliens to smugglers have
increased substantially.  Fees for smuggling illegal aliens across
the southwest border have reportedly tripled in some instances and
may be as high as several thousand dollars for transportation to the
interior of the United States. 

Intelligence assessments by INS' Central Region in April and October
1996 concluded that smugglers had become more sophisticated in their
methods of operation.  The assessments said that smugglers were more
organized and were transporting aliens further into the interior of
the United States.  Similarly, officials from the U.S.  Commission on
Immigration Reform told us that their examination of alien smuggling
along the southwest border indicated a trend toward more organized
smuggling. 

INS sector officials and documents have reported that changes in
alien smuggling patterns may be having a negative impact in some
sectors.  According to a June 1996 Laredo sector intelligence
assessment, due to the success of Operation Gatekeeper and Operation
Hold-the-Line, as well as increased personnel in the McAllen sector,
alien smuggling had increased in the sector.  The assessment further
stated that limited manpower caused deterrence to be negligible
and, consequently, alien smugglers crossed virtually at any point
along the Rio Grande River beyond the area where the sector focused
its enforcement resources.  According to the report from the Del Rio
intelligence conference, the increased use of organized smuggling may
make INS' mission more difficult.  For example, most of the alien
smuggling organizations were reportedly "long-haul" groups
transporting aliens to interior work sites via interstate highways. 
In addition, according to the report, many of the smuggling
organizations that used to be headquartered in the United States
moved their operations out of the United States, which may make
prosecution of the principal leaders of these organizations more
difficult. 


      RECIDIVISM
---------------------------------------------------------- Letter :6.5

The strategy anticipated eventual reductions in apprehensions of
illegal aliens who had previously been apprehended, as control was
gained in particular locations.  According to INS, reductions in
recidivism in these locations would indicate that some illegal aliens
are being deterred from entering.  INS has plans to use IDENT, its
automated fingerprinting system, to identify recidivists and analyze
their crossing patterns.  However, due to the length of time that it
is taking INS to deploy the IDENT system in field locations,
difficulties in getting agents to use IDENT or use it properly, and
computer problems, IDENT to date has provided limited information on
recidivism. 

According to INS' fiscal year 1998 budget request, IDENT systems were
to be installed in 158 fixed locations along the southwest border by
the end of fiscal year 1997.  As of October 1, 1997, however, only
140 of these locations had an IDENT system in place.  Of the nine
southwest border sectors, only the San Diego and El Centro sectors
had an IDENT system installed in every fixed location.  Even when
IDENT has been installed, according to INS sector officials, it has
not always been put in locations where aliens are apprehended.  One
of the reasons given for not installing IDENT in these locations is
that it depends on telephone hookup to a central database, but aliens
are not necessarily apprehended where such telephone communications
exist.  As a result, many apprehended aliens have not been entered
into IDENT and whether they are recidivists cannot be readily
determined. 

Further, according to INS Headquarters officials, even when agents
have access to IDENT, they don't always use all the system
capabilities to identify recidivists.  This reportedly occurs when
agents input information on apprehended aliens into IDENT at the end
of their work shift.  In such a case, agents may input apprehended
aliens into the system but leave without determining whether the
alien is a recidivist, because doing so would require additional time
for further processing.  In addition, if agents don't use the IDENT
equipment properly (e.g., they do not clean the platen, which records
the fingerprint), they may obtain poor quality fingerprints from
apprehended aliens.  This can result in a failure to identify aliens
as recidivists.  These situations have been acknowledged as problems
by INS, but the frequency of their occurrence is not known. 

Finally, computer problems have affected the usefulness of IDENT data
and INS' ability to track recidivism over several years.  The first
year of IDENT data, which were collected in fiscal year 1995, is not
comparable with data collected in later years because of changes made
to the software at the end of 1995.  Accordingly, data collected
since January 1996 can be used as a baseline for assessing the
effects of the southwest border strategy, but data collected prior to
then cannot be used.  In addition, computer problems arose in fiscal
year 1996, affecting the ability of IDENT to accurately read
apprehension dates and times recorded by agents and to match poor
quality fingerprints within the system.  INS officials told us these
problems have been corrected and the data have been reprocessed,
validated, verified, and are more consistent.  INS officials told us
that although IDENT data gathered since January 1996 are reliable and
accurate, they have not yet been analyzed to examine trends in
recidivism. 


      CHANGES IN VIOLENCE AT THE
      BORDER
---------------------------------------------------------- Letter :6.6

The strategy anticipated a reduction in border violence as border
control was achieved.  INS officials told us that they anticipated
that crime would decline in those sections of the border where INS
invested more enforcement resources. 

The results on this indicator are inconclusive.  In November 1996,
INS officials reported that crime statistics for the San Diego area
showed that property crime rates and violent crime rates dropped
between 1994 and 1995, after the infusion of resources in the sector. 
The decreases exceeded decreases reported for the same time period
for the state of California and the nation as a whole.  However,
property and violent crime rates were decreasing in San Diego prior
to the infusion of resources.  In addition, according to FBI crime
statistics, the crime rate for Imperial Beach, the area that received
the greatest infusion of Border Patrol agents in the San Diego
sector, was 19 percent lower from January to June of 1996 compared
with the same period in 1992.  However, this decrease was smaller
than the 32 percent decrease for the San Diego region as a whole. 

According to an official in the Executive Office for the U.S. 
Attorneys, the Executive Office as well as local law enforcement
leaders believe that a more secure border is a material element in
the reduction of crime in San Diego.  In addition, according to this
official and INS officials, there has been a significant drop in
violence against aliens crossing the border illegally due to the
Mexican government establishing a special group to patrol its side of
the border.  According to the Executive Office official, enhanced
coordination between the group and U.S.  authorities has had a
profound and positive impact on the level of violence. 

The U.S.  Commission on Immigration Reform, in its 1994 evaluation of
Operation Hold-the-Line, examined statistics for a number of
different types of crimes in El Paso and the age groups involved in
committing the crimes.  The evaluation found that certain types of
petty crime and property crime committed by young adults and
juveniles had declined in El Paso after the implementation of the
operation.  However, the evaluation report said that the declines
were neither statistically significant nor greater than drops that
had occurred in previous years. 

Furthermore, linking changes in crime rates to border enforcement
efforts is problematic because there are often no data available on
whether arrested offenders have entered the country illegally. 
Without this information, it is difficult to determine what
proportion of the reported declines in crime rates may be due to
changes in the number of illegal aliens arrested for criminal
activity. 


   FORMAL EVALUATION BASED ON
   MULTIPLE INDICATORS WOULD BE
   NEEDED TO ASSESS EFFECTIVENESS
   OF THE ATTORNEY GENERAL'S
   STRATEGY
------------------------------------------------------------ Letter :7

The Attorney General's strategy for deterring illegal entry across
the southwest border envisions three distinct but related results: 
fewer aliens will be able to cross the border illegally; fewer aliens
will try to illegally immigrate into the United States; and,
consequently, the number of illegal aliens in the United States will
decrease.  However, the indicators presently used for measuring the
overall success of the strategy are not sufficiently comprehensive to
address these three distinct results, and, in many cases, data are
not being gathered systematically.  In addition, there is no overall
plan describing how these and other indicators could be used to
systematically evaluate the strategy to deter illegal entry.  To
gauge the overall success of the strategy, data would be needed to
assess each of the results envisioned by the strategy and an
evaluation plan would be needed to describe the interrelationship of
those results. 

To illustrate this point, the strategy could have a desirable effect
on the flow of illegal aliens across the southwest border and,
concomitantly, an undesirable effect on the size of the illegal alien
population in the United States.  This could occur if, for example,
in response to the strategy, aliens made fewer illegal trips to the
United States, but stayed longer on each trip, or made a legal trip
but overstayed the terms of their visas.  Information would thus be
needed on whether a possibly reduced flow across the border may be an
artifact of aliens' staying in the United States for longer periods
of time than when the border was more porous or whether aliens were
deterred from making attempts at illegal entry at the border. 

Our review of the illegal immigration research literature, interviews
with agency officials, and suggestions on evaluation strategies made
to us by our panel of immigration experts verified the importance of
a comprehensive approach for assessing the effectiveness of the
strategy.  Such an approach would require a formal evaluation plan
consisting of indicators that would serve as measures of flow across
and to the border as well as the size of the illegal alien
population.  We drew on our literature review and other consultations
to identify indicators, some noted by INS and others not, that might
provide useful information.  The indicators we identified are
discussed in the following pages and summarized in appendix V. 

It is important to keep several points in mind when contemplating
these indicators: 

  -- First, the indicators we identified address some aspects of each
     of the border strategy's intended results dealing with

the flow of illegal aliens across the southwest border,

whether aliens are being deterred from attempting to illegally
migrate into the United States, and

the number of illegal aliens residing in the United States. 

  -- Second, each indicator or result by itself would be insufficient
     to assess the effectiveness of the strategy as a whole, but
     multiple indicators drawing on a variety of methods and data
     sources could contribute to a more comprehensive evaluation of
     the effectiveness of the strategy along the southwest border. 

  -- Third, these indicators should not be viewed as an exhaustive or
     all-inclusive list.  Our purpose was to illustrate the
     significance of the multiple-indicator concept; devising an
     implementable evaluation strategy was beyond the scope of this
     review. 

  -- Fourth, the Government Performance and Results Act (Results Act)
     of 1993 is intended to improve the efficiency and effectiveness
     of federal programs by establishing a system to set goals for
     program performance and to measure results.  The Results Act
     requires that agency strategic plans contain a schedule for
     future program evaluations.  Program evaluations can be a
     potentially critical source of information in ensuring the
     reasonableness of goals and strategies and explaining program
     results.  The Results Act defines program evaluations as
     objective and formal assessments of the results, impact, or
     effects of a program or policy. 

We recognize that the results envisioned by the strategy are complex
and interrelated and that a rigorous and comprehensive approach to
evaluating the results would be challenging and potentially costly. 
However, we believe that the information that would be gained from
such an approach would be important to obtain because it could help
identify areas where Justice might not be achieving the strategy's
objectives and suggest areas where strategy and policy changes may be
necessary.  In addition, as Justice states in its September 1997
Strategic Plan,\27 "evaluation identifies and explains the linkages
between the activities and strategies undertaken and the results
achieved .  .  .  [and] enables future planning and resource
decisions to be better informed."


--------------------
\27 The United States Department of Justice Strategic Plan 1997-2002,
September 1997. 


      INDICATORS OF FLOW ACROSS
      THE BORDER
---------------------------------------------------------- Letter :7.1

Historically, inferences about the flow of illegal aliens across the
border have relied heavily on data recorded when aliens are
apprehended attempting to make illegal entry between ports of entry
or at ports of entry.  Although the number of apprehensions is a
quantifiable measure of law enforcement activity and INS can collect
apprehension information systematically and completely along the
length of the border, it is not a very good measure of the
effectiveness or results of broad strategies, such as the strategy to
deter illegal entry across the southwest border.  A major limitation
of apprehension data is that it only provides information on illegal
aliens who have been captured by INS.  Because arrested aliens are
the basis for apprehension data, such data provide little information
about illegal aliens who eluded capture, possibly by (1) changing
their crossing patterns and entering in areas where INS has not been
able to detect them, (2) using smugglers to raise the probability of
a successful crossing, (3) entering successfully at the ports of
entry with false documents, or (4) entering legally with a valid visa
or border-crossing card and then overstaying the time limits on these
documents or working illegally in the United States. 

INS has, in various contexts, made note of indicators other than
apprehensions to measure the effects of various border-control
initiatives.  These indicators include smuggling and crime statistics
that were mentioned in the previous section as indicators of the
interim effects of the strategy.  Other indicators on which INS has
collected some data in certain sectors or has reported data collected
by others include (1) the number of migrants staying in hotels and
shelters in Mexican border cities, (2) the number of "gotaways" who
are detected by the Border Patrol but not apprehended, (3) complaints
from members of U.S.  border communities about suspected illegal
aliens trespassing on their property, and (4) interview responses by
apprehended aliens on the perceived difficulty of crossing the
border.  However, INS has collected much of these data on an ad hoc
basis, not systematically and comprehensively in accordance with a
formal evaluation plan.  As a result, it is often difficult to
determine the time frame in which the data were collected, the
methodological integrity of the data collection procedures, and the
generalizability of the findings from one area of the border to
another. 

We reviewed a number of ongoing and completed research projects to
determine what indicators have been identified by others that can be
potentially used in a formal evaluation of the effects of the
strategy.  These research projects contain information gathered by
immigration researchers in Mexico and the United States on the
crossing patterns and tactics of Mexicans who have crossed or intend
to cross the border illegally.  In addition, the panel of immigration
researchers that we convened for advice on evaluating the
effectiveness of the strategy suggested that information on (1) the
social costs and effects of illegal immigration in border cities; (2)
deaths, injuries, and abuses of illegal aliens; and (3) Mexican
efforts to crack down on its border could bolster an understanding of
the strategy's potential effects on flow across the border. 

Several of the research projects we identified reported information
obtained from interviews with illegal aliens in the United States,
potential illegal aliens in Mexico, Mexicans who had been in the
United States illegally, and illegal and legal immigrants preparing
to cross the U.S.-Mexico border.  Although we cannot attest to the
quality of data collected in these research studies, nor that the
data were properly analyzed and interpreted, these research studies
have produced quantitative and qualitative information on the
crossing experiences of illegal aliens--including the number of times
individuals have been apprehended on prior illegal trips, the
locations where they have crossed the border, tactics used to cross
the border, whether smugglers were used, and the overall costs of
making an illegal trip.  One ongoing project, the Mexican Migration
Project,\28 has gathered interview data from migrants from over 30
communities in Mexico, including detailed histories of border
crossing and migratory experience in the United States covering the
past 30 years.  Two other projects conducted by a Mexican
university\29 gathered interview data from migrants who were
preparing to cross the border illegally or legally, or who had just
returned from the United States to Mexico.  The recently completed
Binational Study on Migration gathered interview data from Mexican
migrants and from alien smugglers on such issues as smuggling costs
and tactics and perceptions of United States border-control
efforts.\30 (See app.  VI for more details on these research
projects, and others listed below, including discussion of their
strengths and weaknesses.)

Members of our expert panel also suggested that evaluations of the
strategy should include indicators of the social costs and benefits
of border enforcement, such as the social and financial effects of
illegal migration on border communities in the United States.  As
mentioned earlier, the 1994 U.S.  Commission on Immigration Reform
evaluation of Operation Hold-the-Line examined a variety of social
indicators in El Paso to estimate the impact of illegal aliens on
these communities and to determine whether there were any changes in
these indicators as a result of the operation.  Future evaluation
research could examine these kinds of indicators in other cities
along the southwest border.  Such evaluations would need to take into
account differences between border cities in terms of the enforcement
strategies employed by INS, social and cultural conditions, and the
types of illegal aliens that traditionally cross through these
cities. 

The panel also recommended more systematic study of deaths, injuries,
and abuse suffered by illegal aliens in crossing the border. 
Advocacy groups have raised concerns that the strategy may divert
illegal aliens to more dangerous terrain to cross the border,
thereby, producing an unacceptably high toll in human suffering.  A
recent study by the University of Houston's Center for Immigration
Research tracked deaths of illegal aliens from such causes as
automobile-pedestrian accidents, drowning, and environmental and
weather hazards (dehydration, hypothermia, accidental falls) in
counties on the U.S.-Mexico border during the years 1993 to 1996.\31
The study did not find an increase in the overall number of illegal
alien deaths over the time period but did find an increase in the
number of deaths in the remote areas where immigrants have traveled
in an effort to avoid areas of greater enforcement along the border. 
These deaths in remote areas were purportedly caused by exposure. 
The study also noted that alien deaths were not recorded
systematically by any centralized agency, and local databases were
partial and did not use common standards. 

Immigration researchers and INS officials we spoke with agreed that
IDENT data, when more fully available, could be quite useful for
examining the flow of illegal aliens across the border.  First, IDENT
could provide an unduplicated count of the number of persons
apprehended.  Second, information on the frequency of apprehension of
individuals, the time between multiple apprehensions, and the
locations where illegal aliens are apprehended could provide a
greater understanding of how border enforcement efforts are affecting
illegal crossing patterns.  Finally, recidivism data have potential
for statistically modeling the flow of illegal aliens across the
border and the probability of apprehension.  INS officials told us in
June 1997 that they were examining different analytic techniques that
could be used to model the flow. 


--------------------
\28 This research project is co-directed by Jorge Durand of the
University of Guadalajara, Mexico, and Douglas Massey of the
University of Pennsylvania. 

\29 One project, still ongoing, is the Zapata Canyon project, which
has surveyed illegal aliens preparing to cross the southwest border
every week since September 1987.  The other project, which was
completed, is the Survey of Migration to the Northern Border, which
surveyed both illegal and legal aliens preparing to cross the border
or returning from the United States to Mexico between 1993 and 1997. 
Both projects are directed by Jorge Bustamante of the Colegio de la
Frontera Norte in Tijuana, Mexico, in collaboration with Jorge
Santibanez and Rodolfo Corona. 

\30 In March 1994, the governments of the United States and Mexico
agreed to undertake a joint study of migration between the two
countries.  National coordinators were designated for each country,
with the U.S.  Commission on Immigration Reform coordinating the work
of U.S.  researchers.  Results from the Binational Study on Migration
were released in September 1997. 

\31 K.  Eschbach, J.  Hagan, N.  Rodriguez, R.  Hernandez-Leon, and
S.  Bailey, Death at the Border (Houston:  Center for Immigration
Research, University of Houston), Working Paper #97-2, June 1997. 


      INDICATORS OF DETERRENCE
---------------------------------------------------------- Letter :7.2

The southwest border strategy is ultimately designed to deter illegal
entry into the United States.  It states that "The overarching goal
of the strategy is to make it so difficult and so costly to enter
this country illegally that fewer individuals even try."\32 INS
officials told us they had no plans, and there are no plans in the
Attorney General's strategy, to directly examine deterrence--that is,
the extent to which potential illegal aliens decide not to make an
initial or additional illegal trip to the United States or decide to
limit the number of illegal trips they had planned to take. 

According to some research, decisions to migrate illegally are
determined by a complex set of factors; among them are perceptions of
border enforcement efforts; economic conditions in the country of
origin; demand for labor in the United States; and the extent to
which social networks are established that facilitate migration from
other countries to the United States.\33

The previously mentioned Mexican Migration Project has collected some
survey information on economic conditions in each of the communities
selected for study as well as the extent of social networks that
facilitate migration from these communities to the United States. 
Although the validity of the results would depend greatly on the
quality of the underlying data and the appropriateness of the
statistical assumptions made, data from the project may help to (1)
estimate annual changes over a 30-year period in migrants' likelihood
to take an illegal trip to the United States and (2) examine the
factors contributing to decisions to take an illegal trip. 

The Binational Study also conducted interviews with Mexicans
concerning factors affecting their decisions to migrate illegally. 
Researchers interviewed residents of several different types of
communities in Mexico that differed in the extent to which patterns
of migration to the United States had taken hold.  In older sending
communities, nearly all male residents had made a trip to the United
States, and illegal migration was seen as a rite of passage.  Study
directors from the U.S.  Commission on Immigration Reform told us
that efforts at deterring illegal migration from these kinds of
communities were unlikely to be effective.  However, they said that
deterrence efforts might work better with migrants from newer sending
communities, where social networks were not yet well established. 
Some people in these communities reportedly were leaving because of
difficult economic circumstances, while others were staying because
they had heard that the trip was harder, more costly, and more
dangerous.  Commission officials told us that an evaluation should
include interviews with residents of older and newer sending
communities in Mexico to ascertain their rationales for undertaking
trips to the United States and their perceptions of border-control
efforts. 

If the border becomes harder to cross illegally, it is possible that
some migrants who might have crossed illegally may try to enter
through legal means by applying for nonimmigrant visas or
border-crossing cards.  Information available from INS and the
Department of State on the number of people applying for nonimmigrant
visas and border-crossing cards could be used to track whether
greater numbers of aliens are applying for legal means of entry. 
Trends in application rates may be difficult to interpret in the
future, however, since aliens already holding border-crossing cards
will be required to use new cards containing biometric information,
such as the cardholder's fingerprints, by October 1, 1999.\34
Application rates may then overstate the actual demand for legal
entry to the United States. 


--------------------
\32 Building A Comprehensive Southwest Border Enforcement Strategy
(Washington, D.C.:  Immigration and Naturalization Service, June
1996), p.  3. 

\33 See:  D.  S.  Massey & K.  E.  Espinosa, "What's Driving
Mexico-U.S.  Migration?  A Theoretical, Empirical, and Policy
Analysis," American Journal of Sociology, Vol.  CII, No.  4 (January
1997), pp.  939-999; and Binational Study on Migration, Binational
Study:  Migration Between Mexico & The United States (Mexico City and
Washington, D.C.:  Mexican Foreign Ministry and U.S.  Commission on
Immigration Reform, September 1997). 

\34 Section 104 of the 1996 Act provides that as of April 1, 1998,
"border-crossing identification cards," documents issued to permanent
resident aliens or aliens living in a foreign bordering territory,
include a machine-readable biometric identifier, such as the
fingerprint or handprint of the alien.  The act further mandates that
by October 1, 1999, an alien presenting a border-crossing
identification card is not to be permitted to cross over the border
into the United States unless the biometric identifier contained on
the card matches the appropriate biometric characteristic of the
alien. 


      INDICATORS OF THE NUMBER OF
      ILLEGAL ALIENS IN THE UNITED
      STATES
---------------------------------------------------------- Letter :7.3

The strategy anticipates that enforcement activities will ultimately
reduce the number of illegal aliens in the United States and,
thereby, reduce their use of benefits and social services.  INS
estimated that the number of illegal aliens residing in the United
States grew from 3.9 million in October 1992 to 5.0 million in
October 1996.\35 These estimates are difficult to use to evaluate the
impact of the strategy, however, because (1) they focus on long-term
illegal residents, rather than illegal aliens who come to the United
States for relatively short periods and return periodically to their
country of origin; (2) they do not allow for estimates of the extent
to which reductions in the flow of illegal immigration across the
southwest border may be offset by increases in aliens using legal
nonimmigrant visas but overstaying the terms of their visas; (3) they
have so far been produced too infrequently to be useful to evaluate
short-term effects of enforcement efforts; and (4) a shortage of
information about some components of the estimates makes it difficult
to estimate the number of illegal aliens without making questionable
assumptions about these components.  Some additional data sources are
available that could supplement INS data.  However, considerable
effort and research would probably be entailed in actually obtaining
adequate estimates of the number of illegal aliens in the United
States. 

There are numerous difficulties in accurately measuring the total
number of illegal aliens in the United States and in estimating how
many illegal aliens come to this country each year.  One of the major
difficulties arises because of the heterogeneity of the illegal alien
population.  Some illegal aliens, referred to as sojourners in the
illegal immigration research literature, come to the United States on
a temporary basis but consider themselves residents of a foreign
country and intend to return.  Others, referred to as settlers, come
here with the intention of staying on a more permanent basis.  In
addition, illegal aliens differ in their mode of arrival.  Some
illegally cross the southwest border; enter illegally along other
borders; or enter illegally at the land, sea, or air ports of entry. 
Other aliens enter legally with one of several types of visas and
then fail to leave within the allowed time period. 

Border enforcement efforts may have different effects on each of
these groups, some of them not intended or anticipated in the
Attorney General's strategy.  For example, successful border
enforcement may result in sojourners limiting the number of trips
they make to the United States, but staying for longer periods of
time, because of the greater difficulty in crossing the border.  In
that case, successful border control would effectively reduce the
number of illegal border crossings but could increase the average
length of time that illegal aliens reside in the United States. 
Successful border strategies may also discourage sojourners from
crossing the border illegally but may encourage them to try to enter
legally and stay longer than authorized.  A comprehensive evaluation
that seeks to determine the effect of the strategy on the number of
illegal aliens in the United States would need to have reliable and
valid data on sojourners and settlers, and on those who enter legally
as well as those who overstay.  The evaluation would also have to
account for the impact of outside factors, such as economic
conditions in countries of origin and policy changes in the United
States. 

In February 1997 INS released estimates of the overall stock of
illegal aliens residing in the United States in October 1996 and
updated earlier estimates that it had made for October 1992, using
data collected by INS and the Census Bureau.  According to the latest
estimates, 5 million illegal aliens resided in the United States as
of October 1996, up from 3.9 million as of October 1992.  These
estimates are difficult to use to evaluate the impact of enforcement
efforts at the border, for a number of reasons.  First, individuals
covered in these estimates are defined as those who have established
residence in the United States by remaining here illegally for more
than 12 months.  The estimates therefore provide little information
about the number of sojourners who come to the United States and stay
for periods shorter than 1 year.  Other kinds of data are needed to
examine the sojourner population. 

Second, uncertainty about the number of people who enter legally with
nonimmigrant visas and stay longer than authorized (overstays) makes
it difficult to assess whether a decrease in the estimated number of
illegal border crossers from Mexico due to border enforcement is
offset by an increase in Mexican overstays.  The INS overstay
estimates for 1996 contain a projection based on earlier overstay
data, thus the total estimate is subject to considerable uncertainty. 
In recent testimony to Congress, INS admitted that it has been unable
to produce estimates of overstays since 1992.\36 Indeed, because of
problems with the data system on overstays,\37 INS acknowledges that
it is unable to estimate the current size of the overstay problem and
the magnitude by country of origin.\38

Third, estimates made once every 4 years are not suited to
identifying intermediate trends that might signal effects of border
enforcement efforts.  INS has calculated an average annual growth
rate of 275,000 in the illegal resident population between 1992 and
1996, but these estimates result from averaging over 4 years and
cannot distinguish yearly changes in the illegal alien population. 

Fourth, INS acknowledges that limited information about some
components of the estimates may increase uncertainty about the size
of the illegal population.  For example, limited information about
the extent to which the Mexican-born population may be undercounted
and the number who emigrated from the United States during the period
between estimates necessitate assumptions that affect the precision
of the estimates of the illegal alien population. 

Unfortunately, data are limited on the number of illegal alien
sojourners who travel back and forth between Mexico and the United
States.  The Binational Study on Migration cited estimates from the
previously mentioned Northern Border study of the number of both
legal and illegal sojourners traveling to the United States and
returning to Mexico in 1993 and 1995.  The study found a decrease
between 1993 and 1995 in the overall number of people traveling in
each direction.  The Binational Study interpreted these findings to
mean that sojourners were likely staying longer in the United States
in 1995 than in 1993.  However, the study was unable to distinguish
the size of the illegal from the legal alien flow.  The Mexican
Migration Project data have also been used to estimate changes in net
annual illegal migration from Mexico (subtracting estimates of the
number of illegal aliens who return to Mexico from estimates of the
number of illegal aliens who migrate to the United States).  Because
neither study was based on data that represented all illegal aliens
in the United States, estimates derived from these studies may be
most useful for looking at trends, rather than absolute levels, of
illegal immigration from Mexico. 

Finally, a number of existing data sources provide estimates of the
number of illegal aliens in United States workplaces.  Since 1988,
the U.S.  Department of Labor has administered its National
Agricultural Workers Survey 3 times a year to a random sample of the
nation's crop farm workers and has asked respondents about their
immigration status, among other things.  Such data may be useful for
examining whether border enforcement efforts have reduced the number,
or proportion, of illegal aliens in a sector that has traditionally
attracted high proportions of illegal migrants, and whether the
characteristics of illegal aliens working in agriculture have
changed.  An April 1997 report prepared for the U.S.  Commission on
Immigration Reform, based on National Agricultural Workers Survey
data from 1988 to 1995, showed an increasing percentage of illegal
aliens in agriculture during this time period.\39

Many illegal aliens work in sectors other than agriculture, including
urban-sector employment in construction and services.\40 INS collects
statistics on the results of targeted and random investigations in
U.S.  workplaces.  Trends in the number of illegal aliens apprehended
at the workplace may reflect enforcement efforts at the southwest
border.  For example, if recent border enforcement efforts are
effective, the proportion of apprehended illegal aliens who have
recently arrived in the United States might decrease at various
workplaces.  However, changes in the number of illegal aliens
apprehended at U.S.  workplaces may be a result of other causes, such
as greater emphasis by INS staff on investigating whether employers
are complying with the law and INS regulations concerning aliens'
right to work in the United States. 


--------------------
\35 In 1994, INS estimated that there were 3.4 million illegal aliens
residing in the United States as of October 1992.  See:  R.  Warren,
Estimates of the Unauthorized Immigrant Population Residing in the
United States, by Country of Origin and State of Residence:  October
1992 (Washington, D.C.:  Immigration and Naturalization Service,
1994).  INS revised the October 1992 estimate, based on new data and
improvements in methodology and published the October 1996 estimate
in February 1997.  See INS Releases Updated Estimates of U.S. 
Illegal Population (Washington, D.C.:  U.S.  Immigration and
Naturalization Service News Release, February 1997). 

\36 Statement of Michael D.  Cronin, Assistant Commissioner for
Inspections, Immigration and Naturalization Service, before the
Subcommittee on Immigration and Claims, House Judiciary Committee,
June 17, 1997.  See also:  Office of the Inspector General,
Immigration and Naturalization Service Monitoring of Nonimmigrant
Overstays (Washington, D.C.:  U.S.  Department of Justice, September
1997). 

\37 The overstay estimates are based on matched records from arrival
forms, collected by INS, and departure forms, collected primarily by
airlines.  The INS nonimmigrant overstay method estimates the number
of aliens, except students, who after being legally admitted have
remained in the United States longer than they have been authorized
to stay.  These individuals have been inspected by INS at authorized
ports of entry, and a record of their arrival is known to the
government.  Border crossers from Canada and Mexico or crewmen who
enter legally through inspection are not included. 

\38 Section 110 of the 1996 Act requires INS by October 1, 1998, to
develop an automated entry-exit control system to enable the Attorney
General to monitor all entries and exits at ports of entry.  Until
such a system is in place, it is likely that considerable uncertainty
about the magnitude of overstays will remain. 

\39 R.  Mines, S.  Gabbard, & A.  Steirman, A Profile of U.S.  Farm
Workers:  Demographics, Household Composition, Income and Use of
Services, (Washington, D.C.:  Prepared for the U.S.  Commission on
Immigration Reform, April 1997). 

\40 The Binational Study notes that there is evidence of a long-term
upward trend in U.S.  urban-sector employment--particularly in
construction and services--for illegal aliens. 


      INS EVALUATION EFFORTS
---------------------------------------------------------- Letter :7.4

INS recognizes that multiple indicators are necessary, and, as part
of its fiscal year 1997 priority to strengthen border facilitation
and control, INS stated that it would "review past and current
studies and initiatives regarding measurements of success of both
enforcement and facilitation along the [Southwest] border."\41 An
objective within this priority was to make recommendations for
appropriate measures and complete implementation plans.  However, INS
was unable to provide us with any documentation that it or any other
Justice component was pursuing a broader, formal evaluation of the
southwest border strategy.  In June 1997, INS officials told us that
they had completed a review of indicators used in the past by INS and
others, but had not yet made recommendations regarding what measures
would be collected, who would be responsible for collecting the data,
or how the data would be analyzed. 

Along these lines, the Justice Department's September 1997 Strategic
Plan stated that a key element in the Department's strategic planning
process is the "systematic evaluation of major programs and
initiatives." Although the strategic plan described program goals,
strategies, and performance indicators for INS, it did not contain a
program evaluation component to explain how it will assess success in
meeting these goals and, in a broader sense, the effectiveness of the
southwest border strategy.  Justice's strategic plan acknowledged
that there has been relatively little formal evaluation of its
programs in recent years.  According to the plan, Justice intends to
examine its current approach to evaluation to determine how to better
align evaluation with its strategic planning efforts. 


--------------------
\41 FY 97 Priority:  Strengthen Border Facilitation and Control
(Washington, D.C.:  Immigration and Naturalization Service, November
1996). 


   CONCLUSIONS
------------------------------------------------------------ Letter :8

The Attorney General has a broad strategy for strengthening
enforcement of the Nation's immigration laws that places priority on
deterring illegal entry into the United States along the southwest
border.  Congress has been supporting efforts to gain control of the
southwest border by substantially increasing INS' funding for
enforcement activities.  As a result, the Department of Justice and
INS have made some strides in implementing the strategy. 

INS' data on the effects of the implementation of the strategy
indicate that a number of the interim results anticipated by the
Attorney General are occurring.  To comprehensively and
systematically assess the effects of the strategy over time, there
are a variety of other indicators that may be useful to provide
information on some aspects of each of the results envisioned by the
strategy.  INS has collected data and reported on some, but not all,
of these indicators. 

Although Justice's strategic plan recognizes the importance of
systematic evaluation of major initiatives, Justice has no
comprehensive evaluation plan for formally evaluating whether the
strategy is achieving its intended range of results--such as reducing
flow across the border, reducing flow to the border, and reducing the
number of illegal aliens who reside in the United States. 

We recognize that developing a formal evaluation plan and
implementing a rigorous and systematic evaluation of the strategy
could require a substantial investment of resources, in part because
the needed data may not be presently available, thereby possibly
requiring support for new data collection efforts.  Thus, devising
such an evaluation plan should entail determining the most important
data needs and the most appropriate and cost-effective data sources
and data collection activities as well as carefully analyzing the
relationships among various indicators to correctly interpret the
results.  Furthermore, data obtained on any set of indicators should
be interpreted in the context of economic conditions and policy
changes in the countries of origin of illegal immigrants and in the
United States to help ensure that the results are attributable to the
strategy and not to other potential causes. 

Notwithstanding the challenges in devising such a broad-based
evaluation plan, we believe that the substantial investment of
billions of dollars being made in the Attorney General's strategy
warrants a cost-effective, comprehensive evaluation to demonstrate
whether benefits commensurate to the investment have been realized. 
Such an evaluation would also be in keeping with the concepts
embodied in the Government Performance and Results Act of 1993 as
well as the Department's strategic plan to evaluate major
initiatives. 

In addition, a comprehensive evaluation would also assist Justice in
identifying whether INS is implementing the strategy as planned; what
aspects of the strategy are most effective; and, if the strategy's
goals are not being achieved, the reasons they are not.  Such
information would help the agency and Congress identify whether
changes are needed in the strategy, in policy, in resource levels, or
in program management. 


   RECOMMENDATIONS
------------------------------------------------------------ Letter :9

We recommend that the Attorney General develop and implement a plan
for a formal, cost-effective, comprehensive, systematic evaluation of
the strategy to deter illegal entry across the southwest border. 
This plan should describe (1) the indicators that would be required
for the evaluation, (2) the data that need to be collected, (3)
mechanisms for collecting the data, (4) controls intended to ensure
accuracy of the data collected, (5) expected relationships among the
indicators, and (6) procedures for analyzing the data. 


   AGENCY COMMENTS
----------------------------------------------------------- Letter :10

We requested comments on a draft of this report from the Attorney
General or her designees.  On September 16, 1997, we met to obtain
oral comments on the draft report from the INS Executive Associate
Commissioner (EAC) for Policy and Planning and other officials from
INS' Office of Policy and Planning, Field Operations, Border Patrol,
Inspections, Intelligence, Budget, General Counsel, Internal Audit,
and Congressional Relations as well as from the Justice Management
Division and the Executive Office for United States Attorneys.  INS
and the Executive Office for United States Attorneys followed up this
discussion with point sheets, which reiterated their oral comments
suggesting clarifications and technical changes.  We also requested
comments from the Secretary of State and the Acting Commissioner of
Customs.  The Customs Service had no comments on our draft report. 
We received technical comments from the Department of State.  We made
clarifications and technical changes to the draft report where
appropriate. 

Although INS chose not to provide written comments on the report or
the recommendation, INS officials told us that they recognize the
need for a comprehensive border-wide evaluation and are in the
process of designing and implementing one.  The evaluation design was
not available when we finalized this report; therefore, we are not in
a position to assess whether it contains the types of evaluation
factors discussed in our recommendation. 


--------------------------------------------------------- Letter :10.1

We are sending copies of this report to the Attorney General, the
Commissioner of the Immigration and Naturalization Service, the
Acting Commissioner of the U.S.  Customs Service, the Secretary of
State, the Director of the Office of Management and Budget, and other
interested parties.  We will also make copies available to others
upon request. 

If you or your staff have any questions concerning this report,
please contact me on (202) 512-8777.  This report was done under the
direction of Evi L.  Rezmovic, Assistant Director, Administration of
Justice Issues.  Other major contributors are listed in Appendix VII. 

Richard M.  Stana
Acting Associate Director, Administration
 of Justice Issues


ATTORNEY GENERAL'S STRATEGY TO
DETER ILLEGAL ENTRY INTO THE
UNITED STATES ALONG THE SOUTHWEST
BORDER
=========================================================== Appendix I

In February 1994, the Attorney General and Immigration and
Naturalization Service (INS) Commissioner announced a comprehensive
five-part strategy to strengthen enforcement of the nation's
immigration laws.  The strategy outlined efforts to curb illegal
entry and protect legal immigration by focusing on five initiatives: 
(1) strengthening the border; (2) removing criminal aliens; (3)
reforming the asylum process; (4) enforcing workplace immigration
laws; and (5) promoting citizenship for qualified immigrants.  The
first priority, strengthening the border, focused on immigration
control efforts along the 2,000-mile U.S.-Mexican border and is
summarized here. 

According to the 1994 strategy announced by the Attorney General and
the INS Commissioner, the first priority was to focus on the areas
between the ports of entry, which are divided into nine Border Patrol
sectors.  Recognizing the displacement effect of enhanced efforts
between the ports of entry, the ports of entry along the southwest
border were added to provide comprehensive coverage, and thus
compensate for displacement.  The new border strategy involved
"prevention through deterrence." This strategy called for
concentrating new resources on the front lines at the most active
points of illegal entry along the southwest border. 

The key objectives of the between-the-ports strategy were to (1)
close off the routes most frequently used by smugglers and illegal
aliens (generally through urban areas) and (2) shift traffic through
the ports of entry or over areas that were more remote and difficult
to cross illegally, where INS had the tactical advantage.  To carry
out this strategy, INS planned to provide the Border Patrol and other
INS enforcement divisions with the personnel, equipment, and
technology to deter, detect, and apprehend illegal aliens. 

INS also had related objectives of preventing illegal entry through
the ports, increasing prosecutions of smugglers and aliens who
repeatedly enter the United States illegally, and improving the
intelligence available to border patrol agents and port inspectors. 


   BORDER PATROL
--------------------------------------------------------- Appendix I:1

In July 1994, the Border Patrol developed its own plan to carry out
the Attorney General's between-the-ports strategy.  The plan, still
current in September 1997, was to maximize the risk of apprehension,
using increased human and physical barriers to entry to make passage
more difficult so that many will consider it futile to continue to
attempt illegal entry through traditional routes.  Consistent with
the Attorney General's strategy, the Border Patrol sought to bring a
decisive number of enforcement resources to bear in each major entry
corridor so that illegal aliens would be deterred from using
traditional corridors of entry. 

The plan called for targeting resources to the areas of greatest
illegal activity in four phases: 

  -- Phase I -Concentrate efforts in San Diego (Operation Gatekeeper)
     and El Paso (Operation Hold-the-Line), since these two areas
     historically accounted for 65 percent of all apprehensions along
     the southwest border, and contain displacement effects in the El
     Centro and Tucson sectors. 

  -- Phase II -Control the Tucson and south Texas corridors. 

  -- Phase III -Control the remainder of the southwest border. 

  -- Phase IV -Control all other areas outside the southwest border,
     and concentrate on the northern border and water avenues (for
     example, Pacific and Gulf coasts). 

The Border Patrol strategic plan directed intense enforcement efforts
at the areas of greatest illegal activity.  This strategy differed
from the previous practice of thinly allocating Border Patrol agents
along the border as new resources became available.  Each southwest
sector developed its own tactical plan detailing how it would
implement the Border Patrol's strategic plan.  According to the plan,
each sector was to focus first on controlling urban areas and
increase the percentage of border-control hours on the front lines of
the border in such duties as linewatch,\1 patrol, checkpoints, and
air operations. 


--------------------
\1 Patrolling the immediate border area. 


   INSPECTIONS AT PORTS OF ENTRY
--------------------------------------------------------- Appendix I:2

To complement the Border Patrol's strategic plan, which will be
carried out between the ports of entry, INS Inspections developed
Operations 2000+ in September 1996 to enhance operations at the ports
of entry.  This plan called for increasing the identification and
interception of criminals and illegal entrants.  To accommodate legal
entry while enhancing its efforts to deter illegal entry at the ports
of entry, INS will increase its use of technology to improve
management of legal traffic and commerce.  As part of this effort,
INS had several pilot projects ongoing to use dedicated commuter
lanes and automated entry systems to speed the flow of frequent
low-risk travelers through the ports.  The use of automated systems
could then free inspectors' time to be spent on other enforcement
activities. 


   PROSECUTIONS
--------------------------------------------------------- Appendix I:3

In concert with INS' efforts to deter illegal entry, the Attorney
General seeks to increase prosecutions of alien smugglers, coyotes,\2
and those with criminal records who repeatedly reenter the United
States after having been removed.  This program involves several
aspects regarding the federal prosecution function and is intended

  -- to develop, through the U.S.  Attorneys in the five judicial
     districts contiguous to the U.S.-Mexico border, federal
     prosecution policies to (1) support enforcement strategies
     implemented at the ports of entry by INS and U.S.  Customs and
     between the ports by the Border Patrol, (2) maximize crime
     reduction in border communities, (3) minimize border-related
     violence, and (4) engender respect for the rule of law at the
     border.  These policies include (1) targeting criminal aliens
     for 8 U.S.C.  1326\3 prosecution, (2) targeting coyotes to
     disrupt the smuggling infrastructure, (3) increasing alien
     smuggling prosecutions, (4) increasing false document vendor
     prosecutions, (5) enforcing civil rights laws, (6) prosecuting
     assaults against federal officers, and (7) prosecuting
     border-related corruption cases. 

  -- to coordinate the use of INS/Executive Office of Immigration
     Review (EOIR) administrative sanctions and the application of
     federal criminal sanctions to achieve an integrated and
     cost-effective approach to the goal of deterrence in border law
     enforcement.  This step includes (1) enforcing EOIR orders as
     predicates for federal prosecution under section 1326, and (2)
     using INS/EOIR removal orders and section 1326 provisions to
     prosecute coyotes. 

  -- to enhance cooperation between federal immigration authorities
     and state and local law enforcement agencies to (1) leverage INS
     enforcement efforts through coordination with local law
     enforcement as allowed by law; (2) identify arrestees'
     immigration status to facilitate choice at the outset of
     official processing among the options of administrative
     sanctioning, federal criminal prosecution, and state criminal
     prosecutions; and (3) ensure removal by immigration authorities
     of criminal aliens from the country, following service of state
     and local sentences after conviction.  This plan involves (1)
     instituting county jail programs, (2) employing state and local
     police to supplement INS personnel, (3) allocating cases
     effectively between federal and state prosecutors, and (4)
     integrating information/identification systems among
     jurisdictions. 

The Attorney General designated the U.S.  Attorney for the Southern
District of California as her Southwest border representative and
directed him to coordinate the border law enforcement
responsibilities of various Justice Department agencies--including
INS, the Drug Enforcement Agency, and the Federal Bureau of
Investigation--with activities undertaken by the Treasury Department
and the Department of Defense. 


--------------------
\2 Foot guides. 

\3 Statute providing criminal penalties for reentry of certain
deported aliens. 


   INTELLIGENCE
--------------------------------------------------------- Appendix I:4

Another part of the strategy to deter illegal entry across the
Southwest border was to reengineer INS' Intelligence program so the
border patrol agents and port inspectors could better anticipate and
respond to changes in illegal entry and smuggling patterns.  By the
end of fiscal year 1997, INS expected to have an approved strategic
plan for collecting and reporting intelligence information,
integrating additional assets into the Intelligence program, and
assessing field intelligence requirements.  In addition, intelligence
training and standard operating procedures for district and sector
personnel were to be developed. 


   INDICATORS OF SUCCESS
--------------------------------------------------------- Appendix I:5

As the strategy is being carried out, the Department anticipates that
its success will be measurable using more indicators than the one
traditionally used:  apprehensions.  The Department anticipates the
following measurable results to occur: 

  -- an initial increase in arrests between the ports and then an
     eventual reduction of arrests and recidivism in traditional
     corridors;

  -- shift in the flow of illegal alien entries from the most
     frequent routes (generally through urban areas) to more remote
     areas;

  -- increased port-of-entry activities, including increased entry
     attempts and increased use of fraudulent documents;

  -- increased instances of more sophisticated methods of smuggling
     at checkpoints;

  -- increased fees charged by smugglers;

  -- increased numbers of criminal aliens prosecuted for entering the
     country illegally;

  -- increased numbers of alien smugglers, coyotes, and false
     document vendors prosecuted;

  -- increased numbers of deportations of people presenting false
     documents; and,

  -- reduced violence at the border (for example, less instances of
     border banditry). 

Further, although the Department will not be able to measure them and
many complex variables other than border enforcement affect them, the
Department believes its enhanced enforcement activities will have a
positive impact on the following: 

  -- fewer illegal immigrants in the interior of the United States
     and

  -- reduction in the use of social services and benefits by illegal
     aliens in the United States. 


   DOCUMENTS USED TO PREPARE
   ANALYSIS
--------------------------------------------------------- Appendix I:6

The following documents, listed in chronological order, were used to
prepare our summary of the Attorney General's strategy to deter
illegal entry into the U.S.  along the Southwest border: 

Department of Justice, Attorney General and INS Commissioner Announce
Two-Year Strategy to Curb Illegal Immigration, Press Release,
February 3, 1994. 

U.S.  Government Printing Office, Accepting the Immigration
Challenge, The President's Report on Immigration, 1994. 

INS, Strategic Plan:  Toward INS 2000, Accepting the Challenge,
November 2, 1994. 

INS, Border Patrol Strategic Plan 1994 and Beyond, National Strategy,
U.S.  Border Patrol, July 1994. 

INS, Priority 2:  Strengthen Border Enforcement and Facilitation,
Fiscal Year 1996 Priority Implementation Plan. 

Department of Justice, Executive Office for United States Attorneys,
USA Bulletin, Vol.  44, Number 2, April 1996. 

Department of Justice, INS, Building a Comprehensive Southwest Border
Enforcement Strategy, June 1996. 

Department of Justice, INS, Meeting the Challenge Through Innovation,
September 1996. 

INS Office of Inspections, Inspections 2000+:  A Strategic Framework
for the Inspections Program, September 1996. 

Illegal Immigration Reform and Immigrant Responsibility Act of 1996. 

INS, Operation Gatekeeper, Two Years of Progress, October 1996. 

INS, Strengthen Border Facilitation and Control, Fiscal Year 1997
Priorities. 

Department of Justice, INS, Immigration Enforcement, Meeting the
Challenge, A Record of Progress, January 1997. 

In addition, we reviewed the strategies prepared in 1994 by the
Border Patrol's nine southwest border sectors. 


STATUS OF SELECTED BORDER
ENFORCEMENT INITIATIVES MANDATED
BY THE 1996 ACT
========================================================== Appendix II

Subtitle A, Title I of the 1996 Act mandates further improvements in
border-control operations.  These include improvement in the
border-crossing identification card, new civil penalties for illegal
entry or attempted entry, and a new automated entry/exit control
system.  The status of initiatives to address these provisions is
discussed below. 


   IMPROVEMENT IN THE
   BORDER-CROSSING IDENTIFICATION
   CARD
-------------------------------------------------------- Appendix II:1

Section 104 of the 1996 Act provides that documents issued on or
after April 1, 1998, bearing the designation "border-crossing
identification cards" include a machine-readable biometric
identifier, such as the fingerprint or handprint of the alien.  The
1996 Act further mandates that by October 1, 1999, an alien
presenting a border-crossing identification card is not to be
permitted to cross over the border into the United States unless the
biometric identifier contained on the card matches the appropriate
biometric characteristic of the alien. 

According to INS Inspections officials, as of September 1997,
initiatives to implement this provision were under way, but several
issues had yet to be resolved.  An INS official told us that INS and
the State Department have reached a mutual decision on the biometric
characteristics to be used on the card.  A digitized color
photograph, with potential use for facial recognition as that
technology advances, will be collected.  Live scan prints of both
index fingers will also be collected.  These fingerprints are
compatible with IDENT, INS' automated fingerprint identification
system.  However, the system to be used to verify this information at
the point of entry into the United States had not been determined. 
According to the official, resolution of this problem is necessary
since currently no viable technology available can implement what the
law requires.  INS plans to pilot test a biometric card for
pedestrians at the Hidalgo, Texas port of entry by the end of
calendar year 1997. 

The new system would also change which federal agency provides
border-crossing cards and which agency produces the cards as well as
where the cards are available.  Currently, INS Immigration Inspectors
at ports of entry and some State Department consular officers
adjudicate applications and issue border-crossing cards.  The INS
Commissioner and the Assistant Secretary of State for Consular
Affairs, on September 9 and September 18, 1997, respectively, signed
a memorandum of understanding on how the revised border-crossing card
process would work.  Under the new system, the State Department will
take over all adjudication of these cards.  Foreign services posts
will collect the fees and necessary data, including the biometric,
and will send the data electronically to INS.  INS will produce the
cards.  According to a State Department official, consular officers
will check border-crossing card applicants against the State
Department's Consular Lookout and Support System (CLASS)\1 to
identify those that may be ineligible.  The actual details of how the
system will operate are, however, still being worked out, e.g., how
to handle the volume of applicants.  INS officials said that if all
goes as expected, the new process will be in place on April 1, 1998. 


--------------------
\1 CLASS is the State Department's name-check system, which contains
information on foreign nationals who may apply for visas to the
United States. 


   NEW CIVIL PENALTIES FOR ILLEGAL
   ENTRY
-------------------------------------------------------- Appendix II:2

Section 105 of the 1996 Act mandates new civil penalties for illegal
entry.  Any alien apprehended while entering (or attempting to enter)
the United States at a time or place other than as designated by
immigration officers is subject to a civil penalty of at least $50
and not more than $250 for each such entry or attempted entry.  The
penalty is doubled for an alien who has been previously subject to a
civil penalty under this subsection.  Moreover, such penalties are in
addition to, not instead of, any criminal or other civil penalties
that may be imposed.  This provision applies to illegal entries or
attempts to enter occurring on or after April 1, 1997. 

As of September 1997, INS was still studying this issue.  According
to an INS official, a working group met in April 1997 and prepared an
option papers for INS' Policy Council.  Based on this preliminary
analysis, the council concluded that the program's administration
would probably cost more than the fines collected and the fines would
not deter individuals, since few individuals would have the money to
pay the fines.  Accordingly, the INS Policy Council requested further
study of the potential effects of the program before proceeding
further with implementation.  Such a study was on-going as of
September 1997.  No deadline was set for the study.  Further
development of the rules and regulations required to implement the
program will depend on the results of the study. 


   NEW AUTOMATED ENTRY/EXIT
   CONTROL SYSTEM
-------------------------------------------------------- Appendix II:3

Section 110 of the 1996 Act directs the Attorney General to develop
an automated entry and exit control system by October 1, 1998, to (1)
collect data on each alien departing the United States and match the
record of departure with that of the alien's arrival in the United
States and (2) enable the Attorney General to identify through
on-line searching procedures, lawfully admitted nonimmigrants who
remain in the United States beyond the time authorized. 

According to INS Inspections officials, this system is in the early
development stage.  INS is pilot testing a system for air travelers
and believes it will have a system in place for them within the
2-year mandated time period. 

However, many implementation issues remain regarding travelers who
enter and exit across the land borders.  For example, how will exit
control be done for those driving across the land borders?  Will
everyone entering the United States, including citizens, be checked? 
If so, how do you check everyone?  If not, how do you know who is an
alien and who is not?  INS expects to test a pilot project for an
arrival/departure system for pedestrian crossers at the Eagle Pass,
Texas land port of entry.  According to INS Inspections officials,
INS plans to ask Congress for more time to implement such as a system
at land ports of entry, but had not done so as of September 1997.  At
that time, INS had at the Office of Management and Budget a technical
amendment requesting more time to implement the system for land ports
of entry. 

Officials also indicated that there is some concern regarding
different mandates in the 1996 Act.  For example, while the act
specifies that the biometric is for the border-crossing card, the
provision mandating the automated entry/exit system includes no
mention of using a biometric. 


INS BORDER PATROL AND INSPECTIONS
STAFFING AND SELECTED WORKLOAD
DATA
========================================================= Appendix III



                                   Table III.1
                     
                     Authorized Border Patrol Agent Positions
                       in Southwest Border Patrol Sectors,
                             Fiscal Years 1993 -1997

                  Authorized increases each year    Total increase
                 --------------------------------  -----------------
Bor
der
Pat                                                                        Total
rol  Authorized                                                       Authorized
Sec      Agents                                               Percen   agents FY
tor     FY 93\a    FY 94    FY 95   FY 96   FY 97   FY 94-97       t          97
---  ----------  -------  -------  ------  ------  ---------  ------  ----------
San         980      300      229     428     278       1235      44        2215
 Di
 eg
 o
El          194        0        0       0      50         50       2         244
 Ce
 nt
 ro
Yum         178        0        0       0       0          0       0         178
 a
Tuc         281        0      128     246     228        602      21         883
 son
El          602       50       93     101     107        351      12         953
 Pa
 so
Mar         131        0        0       0       0          0       0         131
 fa
Del         290        0      100       0      52        152       5         442
 Rio
Lar         347        0       75       0      49        124       4         471
 edo
McA         386        0       75      25     228        328      12         714
 ll
 en
================================================================================
Tot        3389      350      700     800   992\b       2842     100        6231
 al
 So
 ut
 hw
 es
 t
 Bo
 rd
 er
--------------------------------------------------------------------------------
\a According to INS officials, the number of on-board agents as of
September 30, 1993, is considered as fiscal year 1993 authorized
border patrol staffing levels for comparison purposes. 

\b Does not include eight agents deployed to Puerto Rico for the
Attorney General's Puerto Rico Anti-Crime initiative. 

Source:  INS. 



                                   Table III.2
                     
                        Apprehensions by Southwest Border-
                     Patrol Sector, Fiscal Year 1992 Through
                          First Half of Fiscal Year 1997

Bo
rd
er
Pa
tr
ol
se
ct                                                                    First half
or        FY 92        FY 93        FY 94        FY 95        FY 96        FY 97
--  -----------  -----------  -----------  -----------  -----------  -----------
Sa      565,581      531,689      450,152      524,231      483,815      172,339
 n
 D
 i
 e
 g
 o
El       29,852       30,058       27,654       37,317       66,873       57,438
 C
 e
 n
 t
 r
 o
Yu       24,892       23,548       21,211       20,894       28,310       15,376
 ma
Tu       71,036       92,639      139,473      227,529      305,348      140,077
 c
 s
 o
 n
El      248,642      285,781       79,688      110,971      145,929       69,362
 P
 a
 s
 o
Ma       13,819       15,486       13,494       11,552       13,214        6,622
 r
 f
 a
De       33,414       42,289       50,036       76,490      121,137       61,567
 l
 R
 i
 o
La       72,449       82,348       73,142       93,305      131,841       75,475
 r
 e
 d
 o
Mc       85,889      109,048      124,251      169,101      210,553      131,643
 A
 l
 l
 e
 n
================================================================================
To    1,145,574    1,212,886      979,101    1,271,390    1,507,020      729,899
 t
 a
 l
 s
 o
 u
 t
 h
 w
 e
 s
 t
 b
 o
 r
 d
 e
 r
--------------------------------------------------------------------------------
Source:  INS. 



                                   Table III.3
                     
                     Border Patrol Hours by Southwest Border
                     Patrol Sector, Fiscal Year 1992 Through
                           First Half Fiscal Year 1997

Border
Patrol                                                                First half
Sector         FY 92       FY 93       FY 94       FY 95       FY 96       FY 97
--------  ----------  ----------  ----------  ----------  ----------  ----------
San Diego
--------------------------------------------------------------------------------
General       55,687      59,050      72,240      43,884      26,695       7,457
 enforce
 ment
Program      720,097     917,149     933,549   1,030,319   1,255,970     766,895
 support
Border       866,500   1,067,872   1,004,232   1,361,983   1,756,647   1,021,266
 enforce
 ment
================================================================================
Total      1,642,284   2,044,071   2,010,021   2,436,186   3,039,312   1,795,618


El Centro
--------------------------------------------------------------------------------
General       27,619      27,396      24,871      19,252      14,648       5,376
 enforce
 ment
Program      139,131     149,095     138,529     143,259     148,241      73,628
 support
Border       280,491     246,289     249,180     248,206     251,101     109,240
 enforce
 ment
================================================================================
Total        447,241     422,780     412,580     410,717     413,990     188,244


Yuma
--------------------------------------------------------------------------------
General       33,772      34,645      29,259      20,087      16,511       5,100
 enforce
 ment
Program       94,731     104,368     102,467     112,633     105,227      51,941
 support
Border       279,030     255,330     258,137     249,374     257,871     118,558
 enforce
 ment
================================================================================
Total        407,533     394,343     389,863     382,094     379,609     175,599


Tucson
--------------------------------------------------------------------------------
General       21,898      25,418      20,756      15,768      17,557       5,513
 enforce
 ment
Program      219,951     264,022     250,294     316,267     395,867     305,885
 support
Border       295,429     315,541     344,813     365,489     500,349     298,793
 enforce
 ment
================================================================================
Total        537,278     604,981     615,863     697,524     913,773     610,191


El Paso
--------------------------------------------------------------------------------
General       88,711      84,074      37,365      50,331      34,967      17,318
 enforce
 ment
Program      391,691     414,460     273,088     361,939     477,792     268,626
 support
Border       707,507     809,921   1,023,146     964,533   1,028,846     516,186
 enforce
 ment
================================================================================
Total      1,187,909   1,308,455   1,333,599   1,376,803   1,541,605     802,130


Marfa
--------------------------------------------------------------------------------
General       63,492      46,467      42,755      36,854      13,757       2,003
 enforce
 ment
Program       89,312      99,873      93,757      89,693      91,170      40,558
 support
Border       168,120     172,732     173,084     167,585     157,773      83,746
 enforce
 ment
================================================================================
Total        320,924     319,072     309,596     294,132     262,700     126,307

Del Rio
--------------------------------------------------------------------------------
General       57,079      30,854      26,073      22,077      17,742       3,456
 enforce
 ment
Program      217,709     226,377     222,617     233,946     310,483     160,515
 support
Border       453,309     396,972     382,607     399,761     501,760     252,388
 enforce
 ment
================================================================================
Total        728,097     654,203     631,297     655,784     829,985     416,359


Laredo
--------------------------------------------------------------------------------
General       30,279      25,342      24,199      19,884      19,664       6,036
 enforce
 ment
Program      182,963     188,928     165,711      75,005     210,034     105,349
 support
Border       485,064     566,219     568,169     536,463     615,189     317,902
 enforce
 ment
================================================================================
Total        698,306     780,489     758,079     731,352     844,887     429,287


McAllen
--------------------------------------------------------------------------------
General       40,662      47,543      45,016      39,511      23,261      10,395
 enforce
 ment
Program      279,164     358,910     355,743     353,511     418,905     214,308
 support
Border       378,562     482,576     470,671     470,868     551,726     274,431
 enforce
 ment
================================================================================
Total        698,388     889,029     871,430     863,890     993,892     499,134


Southwest
Border
--------------------------------------------------------------------------------
General      419,199     380,789     322,534     267,648     184,802      62,654
 enforce
 ment
Program    2,334,749   2,723,182   2,535,755   2,816,572   3,413,689   1,987,705
 support
Border     3,914,012   4,313,452   4,474,039   4,764,262   5,621,262   2,992,510
 enforce
 ment
================================================================================
Total      6,667,960   7,417,423   7,522,360   7,848,482   9,219,753   5,042,869
--------------------------------------------------------------------------------
Note:  General Enforcement includes duties indirectly related to
deterrence, detection, and apprehensions of illegal aliens, such as
employer sanctions, criminal aliens and antismuggling activities. 
Program Support includes hours that support enforcement duties such
as, air operations, intelligence, administrative/supervisory, special
operations and training.  Border Enforcement includes duties directly
related to the deterrence, detection and apprehension of illegal
aliens and interdiction of drugs and other contraband along the
border, such as patrolling the immediate border area, traffic check,
transportation check, and boat and air patrol. 

Source:  INS. 



                                   Table III.4
                     
                      Authorized Inspector Positions by Land
                     Ports of Entry and INS District Offices
                     Along the Southwest Border, Fiscal Years
                                    1994 -1997

                                                                         Total
                                                                        authoriz
                                                                           ed
                         Authorized increases each                      inspecto
                                    year                                   rs
                        ----------------------------                    --------
INS district  Authoriz                                   Total
and land            ed                                increase
ports of      inspecto                                  FY 95-  Percen
entry         rs FY 94     FY 95   FY 96\a     FY 97        97       t     FY 97
------------  --------  --------  --------  --------  --------  ------  --------
San Diego          301        64       220        80       364      46       665
 District
San Ysidro         199        64       142        24
 (2)
Tecate               4         0        19         0
Calexico            94         0        45        48
Andrade              4         0        14         8
Phoenix            104         8       106         8       122      15       226
 District
San Luis            20         0        11         5
Sasabe               3         0        15         0
Nogales (2)       57\b         8        33         0
Naco                 5         0        13         0
Lukeville            4         0        20         0
Douglas             15         0        14         3
El Paso            163        11       108         0       119      15       282
 District
Columbus             8         0         4         0
Santa Teresa         0         0        10         0
El Paso (2)      134\b        11        65         0
Fabens               7         0         5         0
Fort Hancock         3         0         1         0
Presidio            11         0         8         0
Ysleta               0         0        15         0
San Antonio        143         6        56        14        76      10       219
 District
Laredo (3)        93\b         6        42         5
Eagle Pass          26         0         7         6
Del Rio (2)         24         0         7         3
Harlingen          158        21        46        48       115      14       273
 District
Roma                24         0         2         0
Rio Grande           0         0         2         6
 City
Progresso           19         0         6         0
Pharr                7         0         6         4
Los Indios           0         0         1         6
Los Ebanos           0         0         2         1
Hidalgo             46         0        12         4
Falcoln              0         0         0         3
 Heights
Brownsville       62\b        21        15        24
 (2)
================================================================================
Total              869       110       536       150       796     100      1665
 southwest
 border
--------------------------------------------------------------------------------
Note:  Some locations have more than one port of entry. 

\a Includes 136 inspector positions authorized in fiscal year 1994
and funded by land-border fees.  According to INS officials, these
inspectors were not hired until fiscal year 1996, due to delays in
getting land-border fee regulations finalized. 

\b Includes one staff assigned to the district office. 

Source:  INS. 



                                   Table III.5
                     
                      INS Inspections, Selected Workload and
                       Enforcement Data by Southwest Border
                        District Offices, Fiscal Year 1994
                       Through First Half Fiscal Year 1997

INS District                                                       First half FY
Office                     FY 94           FY 95           FY 96              97
----------------  --------------  --------------  --------------  --------------
San Diego
--------------------------------------------------------------------------------
U.S. citizens         28,808,845      26,776,081      26,342,930      14,249,519
 inspected\a
Aliens                63,152,638      59,387,840      57,278,208      30,065,972
 inspected\a
Total persons         91,961,483      86,163,921      83,621,138      44,315,491
 inspected\a
Pedestrians                           16,744,829      19,420,202       9,969,126
 inspected
Vehicles                              28,094,028      26,152,680      13,941,527
 inspected
Fraudulent                41,974          46,515          41,221          20,270
 documents
 intercepted
Oral false                15,403          19,368          12,633           5,292
 claims to U.S.
 citizenship
Individuals                6,565           5,377           6,581           4,207
 smuggled

Phoenix
--------------------------------------------------------------------------------
U.S. citizens         11,717,738       8,299,518       8,313,959       4,332,991
 inspected\a
Aliens                23,268,288      23,072,110      23,035,160      11,897,583
 inspected\a
Total persons         34,986,026      31,371,628      31,349,119      16,230,574
 inspected\a
Pedestrians                            7,806,361       7,657,480       9,969,126
 inspected
Vehicles                               8,951,615       8,984,091      13,941,527
 inspected
Fraudulent                 7,599           6,367           5,338           3,003
 documents
 intercepted
Oral false                   825             686             772             500
 claims to U.S.
 citizenship
Individuals                  279             253             427             241
 smuggled

El Paso
--------------------------------------------------------------------------------
U.S. citizens         27,230,751      24,975,450      22,481,928      11,180,053
 inspected\a
Aliens                42,033,208      39,821,598      36,339,994      18,248,564
 inspected\a
Total persons         69,263,959      64,797,048      58,821,922      29,428,617
 inspected\a
Pedestrians                            4,578,798       4,571,073       2,213,165
 inspected
Vehicles                              18,143,401      17,529,375       8,833,279
 inspected
Fraudulent                 8,173           8,754          11,034           6,720
 documents
 intercepted
Oral false                 4,567           5,887           6,725           4,092
 claims to U.S.
 citizenship
Individuals                  751             599             611             388
 smuggled

San Antonio
--------------------------------------------------------------------------------
U.S. citizens         18,668,021      13,979,554      14,355,119       7,379,551
 inspected\a
Aliens                39,567,895      41,165,203      39,708,171      21,462,035
 inspected\a
Total persons         58,235,916       55144,757      54,063,290      28,841,586
 inspected\a
Pedestrians                            7,289,530       5,219,599       2,914,118
 inspected
Vehicles                              11,454,275      13,026,781       6,383,029
 inspected
Fraudulent                 6,338           6,100           7,636           5,040
 documents
 intercepted
Oral false                 1,467           1,284           1,541           1,008
 claims to U.S.
 citizenship
Individuals                1,559           1,224           1,349             706
 smuggled

Harlingen
--------------------------------------------------------------------------------
U.S. Citizens         20,831,853      12,920,985      12,849,749       7,073,072
 inspected\a
Aliens                40,564,252      39,536,870      40,258,253      20,735,209
 inspected\a
Total persons         61,396,105      52,457,855      53,108,002      27,808,281
 inspected\a
Pedestrians                            7,348,097       1,943,636       4,189,713
 inspected
Vehicles                              14,412,630      15,195,266       7,811,192
 inspected
Fraudulent                 9,176           7,100           9,681           5,386
 documents
 intercepted
Oral false                 1,264           1,549           1,702             773
 claims to U.S.
 citizenship
Individuals                1,394             897           1,299             762
 smuggled

Southwest Border
--------------------------------------------------------------------------------
U.S. citizens        107,257,208      86,951,588      84,343,685      44,215,186
 inspected\a
Aliens               208,586,281     202,983,621     196,619,786     102,409,363
 inspected\a
Total persons        315,843,489     289,935,209     280,963,471     146,624,549
 inspected\a
Pedestrians                           43,767,615      44,811,990      23,824,075
 inspected
Vehicles                              81,055,949      80,888,193      41,553,145
 inspected
Fraudulent                73,260          74,836          74,910          40,419
 documents
 intercepted
Oral false                23,526          28,774          23,373          11,615
 claims to U.S.
 citizenship
Individuals               10,548           8,350          10,267           6,304
 smuggled
--------------------------------------------------------------------------------
\a Estimate based on periodic sampling of the number of occupants per
vehicles entering the port of entry. 

Source:  INS. 


EXPERT PANEL PARTICIPANTS
========================================================== Appendix IV

The following experts in research on illegal immigration issues
paticipated in a panel discussion held at GAO in Washington, D.C.  on
March 28, 1997.  They also reviewed portions of the draft report. 

Deborah Cobb-Clark
Professor of Economics
Research School of Social Sciences
The Australian National University
Canberra, Australia

Sherrie Kossoudji
Professor of Economics
The School of Social Work
University of Michigan
Ann Arbor, MI

B.  Lindsay Lowell
Director of Policy Research
U.S.  Commission on Immigration Reform
Washington, D.C. 

Philip Martin
Professor of Economics
Department of Agricultural and Resource Economics
University of California, Davis
Davis, CA

Douglas Massey
Professor of Sociology
Population Studies Center
University of Pennsylvania
Philadelphia, PA


INDICATORS FOR MEASURING THE
EFFECTIVENESS OF THE STRATEGY TO
DETER ILLEGAL ENTRY ALONG THE
SOUTHWEST BORDER
=========================================================== Appendix V

                                        Predicted outcome                                                                        Are there plans to
                                        if AG's strategy                                                 Is indicator currently  measure indicator on
Indicator           Source of data      is successful       Findings             Limitations             being measured?         ongoing basis?
------------------  ------------------  ------------------  -------------------  ----------------------  ----------------------  --------------------
A. Flow across the Southwest border
-----------------------------------------------------------------------------------------------------------------------------------------------------
Border Patrol       INS                 1. Displacement of  Mixed (see letter).  1. Measures event, not  Currently collected     Yes
apprehensions.                          flow in accordance                       individuals.            along the entire
                                        with Border Patrol                                               southwest border.
                                        strategy.                                2. Unclear
                                                                                 relationship to actual
                                        2. Increasing                            flow of illegal
                                        apprehensions at                         aliens.
                                        first; eventual
                                        reductions as                            3. No information on
                                        control is gained.                       those aliens who elude
                                                                                 capture.

Detection of        Border Patrol       Fewer aliens        Not determined.      No systematic           Yes, in some places.    Yes
illegal aliens      sector data:        detected.                                methodology for
crossing the                                                                     measuring entries and
border.             1. Sensor hits.     Fewer "gotaways."                        gotaways; likely
                    2. Sign cuts.                                                differs across
Number of           3. Infrared and                                              sectors.
"gotaways."         low-light
                    television                                                   Measures event, not
                    cameras.                                                     individuals.



Recidivism (repeat  INS IDENT system.   Attempted           None yet             1. Limited IDENT        Yes, but not            Yes
apprehensions of                        reentries will                           implementation along    completely implemented
aliens).                                decrease over                            southwest border.       along southwest
                                        time.                                                            border.
                                                                                 2. No usable data
                                                                                 preceding AG strategy,
                                                                                 precluding a time-
                                                                                 series analysis of its
                                                                                 effectiveness.

                                                                                 3. Not yet clear how
                                                                                 recidivism data will
                                                                                 be used to model the
                                                                                 flow. Several possible
                                                                                 analytic models can be
                                                                                 tested.

Probability of      Multiple sources:   An increase in the  Analyses of MMP      Household surveys       Yes                     Yes
apprehension.                           probability of      data show a          conducted in Mexico
                    1. IDENT            apprehension.       decrease in the      (e.g., the MMP) and
Difficulty in       recidivism data.                        probability of       surveys of migrants at
crossing border.                                            apprehension         points of entry and
                    2. Mexican                              through 1995.        exit between the U.S.
                    Migration Project                                            and Mexico (e.g., the
                    (MMP) provides                          Binational study     Survey of Migration to
                    annual data from                        data show            the Northern Border)
                    Mexican migrants                        increasing           sample different
                    on the number of                        probability of       subpopulations of
                    times they were                         apprehension in      migrants, who may
                    captured on trips                       recent               differ in their
                    to the U.S. and                         years.               likelihood of
                    places where they                                            apprehension.
                    were apprehended.
                                                                                 MMP data are only
                    3. Colegio de la                                             collected annually,
                    Frontera Norte                                               and retrospective
                    (COLEF) surveys of                                           accounts of
                    border crossers                                              apprehensions may be
                    (Zapata Canyon                                               biased.
                    Project and Survey
                    of Migration to                                              There isn't consensus
                    the Northern                                                 on the methodology for
                    Border).                                                     estimating probability
                                                                                 of apprehension.
                    4. Focus group
                    interviews in
                    Mexico (Binational
                    Study on
                    Migration).

Estimated number    INS INTEX system -  An increase in the  INTEX not yet        The causes of illegal   No                      Yes
of illegal entry    -random checks of   number of aliens    implemented.         POE entries may be
attempts at U.S.    travelers at POEs.  attempting to                            difficult to
ports of entry                          enter illegally at                       disentangle. For
(POE).                                  POEs.                                    example, increased
                                                                                 illegal attempts may
                                                                                 reflect more effective
                                                                                 Border Patrol
                                                                                 strategies between the
                                                                                 ports, an increase in
                                                                                 the flow of illegal
                                                                                 migrants, or simply a
                                                                                 more effective
                                                                                 inspection strategy at
                                                                                 the POE.

                                                                                 INTEX methodology is
                                                                                 still being tested.

Alien smuggling:    Multiple sources:   An increase in      According to INS     Changes in smuggling    Yes                     Yes
                                        smuggling fees and  Intelligence,        fees and tactics may
1. Smuggling        1. INS              sophistication of   smuggling fees are   indicate that border
usage.              intelligence data   smugglers.          increasing.          crossing is more
                    as well as          Smugglers may try                        difficult. However it
2. Smuggling        intelligence data   other means of      Binational study     does not necessarily
costs.              from other federal  delivering aliens   data suggest a rise  indicate that migrants
                    and                 to destinations     in more organized    are less successful at
3. Tactics of       state agencies.     (e.g., using vans,  smuggling rings.     entering.
smugglers.                              small trucks, and
                    2. MMP data .       tractor trailers).

                    3. COLEF data.

                    4. Binational
                    study.

Number of people    INS intelligence    An increase in the  Mixed.               Number of people in     Not clear how           Yes
in hotels and       information.        number of people                         hotels and shelters     systematically data
shelters in                             in hotels and       INS reports that     may be affected by      are collected across
Mexican border      Periodic surveys    shelters in         migrant shelters in  economic opportunities  the southwest border.
cities.             by interested       Mexican border      Mexican border       in border cities.
                    organizations.      cities.             cities are filling
                                                            up, and people are
                    Binational study                        staying longer.
                    site visits.
                                                            Binational study
                                                            site visits found
                                                            no "backup" in
                                                            Tijuana as result
                                                            of enforcement
                                                            efforts in San
                                                            Diego.

Crime in U.S.       Local crime data.   Less crime in U.S.  There was a          Crime may be dropping   Not clear how           Unknown
border cities.                          border cities.      reduction in         overall, regardless of  systematically data
                                                            "nuisance" crimes    effects of border       are collected across
                                                            by juveniles and     strategy. Need to       the southwest border.
                                                            young adults after   control for such
                                                            implementation of    effects using time-
                                                            Operation Hold-      series analyses.
                                                            the-Line in El
                                                            Paso, according to   However, drops in
                                                            1994 Commission on   specific types of
                                                            Immigration Reform   crime (e.g., property
                                                            Study.               crime, car thefts) may
                                                                                 be best indicators of
                                                            According to an INS  drops in crimes
                                                            analysis, property   committed by aliens.
                                                            and violent crime
                                                            rates dropped
                                                            between 1994 and
                                                            1995 in San Diego.

Use of public       State and local     A decrease in       Commission on        Need to control for     Not clear how           Unknown
services in U.S.    sources.            public service      Immigration Reform   long-term trends in     systematically data
border cities.                          usage in border     found some           these indicators,       are collected across
                    Hospitals, local    cities.             reductions in        using time-series       the southwest border.
                    school districts,                       public service       analyses.
                    local welfare                           usage in the first
                    departments.                            few months after
                                                            implementation of
                                                            Operation Hold-
                                                            the-Line in El
                                                            Paso.

Deaths of aliens    County death        Depends on how      Center for           Can be difficult to     Data not collected on   Unknown
attempting entry.   records.            enforcement         Immigration          identify if deceased    systematic basis along
                                        resources are       Research Study       is an undocumented      southwest border.
                    Death records in    allocated. In some  (1997) found no      migrant.
                    Mexico.             cases, deaths may   overall increase in
                                        be reduced or       the number of        Problems of
                    University of       prevented (by       deaths of illegal    undercount--some
                    Houston Center for  fencing along       aliens, but found    proportion of deaths
                    Immigration         highways, for       increasing number    in remote areas will
                    Research reports    example). In other  of deaths from       go undiscovered.
                    (1996, 1997).       cases, deaths may   environmental
                                        increase (as        exposure (falls,
                                        enforcement in      hypothermia,
                                        urban areas forces  dehydration).
                                        aliens to attempt
                                        mountain or desert
                                        crossings) .

Assaults against    INS statistics.     A higher incidence  Assaults on Border   Violence may increase   Yes                     Yes
INS agents.                             of violence         Patrol agents up     if INS is able to
                                        against INS agents  40% from fiscal      frustrate drug
                                        as crossing         year 1994 to fiscal  traffickers. But drug
                                        efforts of illegal  year 1996. Numbers   trafficking may be
                                        aliens are          vary considerably    only somewhat related
                                        frustrated.         across sectors.      to alien smuggling.
                                                                                 Frustrating drug
                                                                                 traffickers does not
                                                                                 necessarily imply
                                                                                 success in deterring
                                                                                 illegal alien traffic.

Abuses of aliens    DOJ.                May vary,           U.S. Commission on   Potential for           Yes                     Yes
by INS officers.                        depending on type   Immigration Reform   underreporting of
                    Advocacy groups.    of enforcement      found reductions in  abuse incidents by
                                        effort.             complaints against   victims.
                                                            Border Patrol
                                                            agents after
                                                            implementation of
                                                            Operation Hold-
                                                            the-Line in El Paso
                                                            (1994).

                                                            Human rights
                                                            organizations have
                                                            complained about
                                                            higher abuse rates
                                                            in other sectors.


B. Flow to Southwest border (deterrence)
-----------------------------------------------------------------------------------------------------------------------------------------------------
Probability of      MMP surveys of      Decreasing          Some published       The probability of      Yes                     Yes
taking a first      migrant             probability of      research shows       taking first or
illegal trip to     communities in      first-timers        probability of       additional trip is
the U.S.            Mexico.             migrating without   migration is not     influenced by a number
                                        documents.          strongly correlated  of factors besides INS
Probability of      Results of focus                        with INS             efforts at the border,
taking additional   groups conducted    Decreasing          enforcement          including economic
illegal trips.      in Mexico           probability that    efforts. INS         conditions in Mexico
                    (Binational Study   experienced         enforcement may      and the U.S., social
                    on Migration).      migrants will take  make migrants        networks in U.S., and
                                        additional trips.   already in the       availability of legal
                                                            migration stream     modes of entry.
                                                            more likely to take
                                                            trips to the U.S.,   MMP Data are not
                                                            to recoup losses     representative of the
                                                            from earlier         immigrant population
                                                            apprehensions.       in the U.S. and may
                                                                                 underrepresent new
                                                                                 sending areas in
                                                                                 southern Mexico.
                                                                                 Retrospective survey
                                                                                 design. Data only
                                                                                 collected annually.



Trends in           INS.                An increase in      Not determined.      Application rates may   Yes                     Yes
applications for                        applications for                         simply be indicators
temporary visas     U.S. Department of  border-crossing                          of greater economic
and border-         State.              cards and                                difficulties in Mexico
crossing cards.                         nonimmigrant                             and other sending
                                        visas.                                   countries, rather than
                                                                                 indicators of
                                                                                 deterrence of migrants
                                                                                 from attempting
                                                                                 illegal modes of
                                                                                 entry.

                                                                                 The 1996 Act mandates
                                                                                 that border-crossing
                                                                                 cards be replaced by
                                                                                 biometric border-
                                                                                 crossing cards within
                                                                                 3 years. This may make
                                                                                 trends difficult to
                                                                                 interpret.

Departures and      Where available     Roughly equal       Not determined.      Data are not            Not determined.         Unknown
arrivals at         from Border Patrol  number of                                consistently
airports in         officials in each   departures and                           available.
Mexican cities      sector.             arrivals in border
along the                               areas where Border                       Data do not
Southwest border.                       Patrol has                               distinguish between
                                        successfully                             trips made to Northern
                                        deterred entry                           Mexico in order to
                                        (more arrivals                           cross border with
                                        than departures in                       trips made in order to
                                        areas where the                          work (in maquiladora
                                        border has not yet                       industries, for
                                        been secured).                           example) or settle.

Changes in traffic  Mexican             Reduced traffic of  Not determined.      Depends on reliability  Not determined.         Unknown
through Mexico of   intelligence        aliens from other                        and generalizability
illegal aliens      sources.            countries through                        of intelligence
from countries                          Mexico or longer                         information.
other than Mexico.  INS                 durations of stays
                    Intelligence        in Mexico by these
                    information         aliens.


C. Number of undocumented aliens in
-----------------------------------------------------------------------------------------------------------------------------------------------------
Estimates of the    Published analyses  An eventual         INS estimated 5      Current surveys         Yes, periodically.      Unknown
stock of illegal    based on INS and    reduction in the    million illegal      generally measure only
residents in the    Census              overall stock of    residents in 1996,   long-term residents
United States.      Bureau data.        illegal residents   up from 3.9 million  (those who have been
                                        in the U.S.         in 1992.             in the U.S. for at
                                                                                 least 1 year).

                                                                                 Estimates differ based
                                                                                 on assumptions made
                                                                                 about undercount and
                                                                                 the emigration of
                                                                                 illegal residents.

                                                                                 Difficult to determine
                                                                                 the specific causes of
                                                                                 changes in the stock
                                                                                 (e.g., better border
                                                                                 enforcement, better
                                                                                 worksite enforcement,
                                                                                 shrinking potential
                                                                                 migrant populations,
                                                                                 etc.).

Number of           Published MMP       A decreased flow    Unclear. COLEF data  Unrepresentativeness    Yes                     Yes
undocumented        analyses.           of illegal          suggest that inflow  of samples.
sojourners                              sojourners to the   and return flow of
                    COLEF Survey of     U.S.                both documented and  Successful deterrence
                    Migration to the                        undocumented         may have perverse
                    Northern                                migrants might have  results--with aliens
                    Border.                                 decreased in recent  spending longer time
                                                            years, but analyses  in the U.S. on any one
                                                            do not distinguish   illegal visit. Fewer
                                                            between the two      border crossings, but
                                                            groups.              an overall increase in
                                                                                 illegal aliens in the
                                                                                 U.S.

Estimates of        INS analyses;       An increase in the  According to INS,    GAO has reported on     INS has not had         Unknown
overstays of legal  Nonimmigrant        number of aliens,   about 40 percent of  some problems in        capability to produce
visas.              Overstay Method     specifically        undocumented         estimating visa         reliable estimates
                    (matches record     Mexican nationals,  immigrants in the    overstays; estimates    since 1992.
                    of arrival forms    who try to enter    U.S. enter with a    are likely to be more
                    with                legally and         nonimmigrant visa.   accurate for those who
                    departure forms).   overstay, in an                          enter and leave by
                                        effort to get                            air.
                    By fiscal year      around border
                    1999, INS is        enforcement.
                    required to have
                    in place
                    a system to track
                    all
                    entries and exits
                    from the U.S.

Labor supply and    National            A labor shortage    The proportion of    Growing proportions in  Yes                     Yes
wages in sectors    Agricultural        or higher wages in  illegal aliens in    agriculture may
of the economy      Workers Survey      an industry that    agriculture has      reflect replacement of
where illegal       (NAWS).             has traditionally   increased between    special agricutural
aliens tend to                          hired undocumented  1988 and 1995.       workers legalized
work.                                   workers.                                 after passage of the
                                                                                 Immigration Reform and
                                                                                 Control Act.

                                                                                 NAWS data are not
                                                                                 generalizable to
                                                                                 sectors other than
                                                                                 agriculture.

Number of           INS.                A decrease in the   Not determined.      The number of           Yes                     Yes
undocumented                            number of                                undocumented workers
aliens found in                         undocumented                             discovered is also a
INS employer                            workers who are                          function of the amount
inspections.                            recent arrivals to                       of emphasis by INS on
                                        the U.S. found in                        worksite enforcement.
                                        INS employer
                                        inspections.
-----------------------------------------------------------------------------------------------------------------------------------------------------
Source:  GAO analysis. 


DESCRIPTION OF SELECTED
IMMIGRATION RESEARCH PROJECTS
========================================================== Appendix VI

Several ongoing and recently completed research projects have
collected data on the flow of undocumented migrants across the
border, as well as information on migrants' background
characteristics, labor market experience in country of origin and the
United States, and potential migrants' intentions to take
undocumented trips to the United States. 


   THE MEXICAN MIGRATION PROJECT
-------------------------------------------------------- Appendix VI:1

The Mexican Migration Project was funded by the National Institute of
Child Health and Human Development to create a comprehensive
binational dataset on Mexican migration to the United States.  The
project is directed by Jorge Durand of the University of Guadalajara
(Mexico), and Douglas S.  Massey of the University of Pennsylvania. 
Two to five Mexican communities were surveyed each year during
December and January of successive years using simple random sampling
methods.  The sample size was generally 200 households unless the
community was under 500 residents, in which case a smaller number of
households were interviewed.  If initial fieldwork indicated that
U.S.  migrants returned home in large numbers during months other
than December or January, interviewers returned to the community
during those months to gather a portion of the 200 interviews. 

These representative community surveys yielded information on where
migrants went in the United States, and during the months of July and
August interviewers traveled to those U.S.  destinations to gather
nonrandom samples of 10 to 20 out-migrant households from each
community.  The U.S.-based samples thus contain migrants who have
established their households in the United States. 

The communities were chosen to provide a range of different sizes,
regions, ethnic compositions, and economic bases.  The sample thus
includes isolated rural towns, large farming communities, small
cities, and very large metropolitan areas; it covers communities in
the states of Guanajuato, Michoacan, Jalisco, Nayarit, Zacatecas,
Guerrero, Colima, and San Luis Potosi; and it embraces communities
that specialize in mining, fishing, farming, and manufacturing, as
well as some that feature very diversified economies. 

The study's questionnaire followed the logic of an ethnosurvey, which
blends qualitative and quantitative data-gathering techniques.  A
semi-structured instrument required that identical information be
obtained for each person, but question wording and ordering were not
fixed.  The precise phrasing and timing of each query was left to the
judgment of the interviewer, depending on circumstances.  The design
thereby combined features of ethnography and standard survey
research. 

The ethnosurvey questionnaire proceeded in three phases, with the
household head serving as the principal respondent for all persons in
the sample.  In the first phase, the interviewer gathered basic
social and demographic information on the head, spouse, resident and
nonresident children, and other household members, including age,
birthplace, marital status, education, and occupation.  The
interviewer then asked which of those enumerated had ever been to the
United States.  For those individuals with migrant experience, the
interviewer recorded the total number of U.S.  trips as well as
information about the first and most recent U.S.  trips, including
the year, duration, destination, U.S.  occupation, legal status, and
hourly wage.  This exercise was then repeated for first and most
recent migrations within Mexico. 

The second phase of the ethnosurvey questionnaire compiled a
year-by-year life history for all household heads, including a
childbearing history, a property history, a housing history, a
business history, and a labor history.  The third and final phase of
the questionnaire gathered information about the household head's
experiences on his or her most recent trip to the United States,
including the mode of border-crossing, the kind and number of
accompanying relatives, the kind and number of relatives already
present in the United States, the number of social ties that had been
formed with U.S.  citizens, English language ability, job
characteristics, and use of U.S.  social services. 

Data at the community and municipio (county) levels were also
collected, using several sources.  First, since 1990, interviewers
have used a special community questionnaire to collect and compile
data from various sources in each community.  Next, data were
compiled from the Anuario Estadistico of 1993, published in Mexico by
INEGI (el Instituto Nacional de Economia, Geografia e Informatica). 
Finally, data were compiled from the published volumes of the Mexican
censuses for 1930, 1940, 1950, 1960, 1970, 1980, and 1990.  These
decennial census data were then interpolated for years between the
censuses and extrapolated for years after 1990. 


      STRENGTHS
------------------------------------------------------ Appendix VI:1.1

Strengths include the large number and diversity of households and
communities represented in the samples and the breadth of information
collected.  The retrospective nature of the survey allows for
analysis of migration flows over a long period of time.  The survey
is a representative sample of communities in states in Western Mexico
that have been the traditional sending areas of Mexico; it captures
both migrants that are living in the United States as well as persons
who have been in the United States but live in Mexico. 


      LIMITATIONS
------------------------------------------------------ Appendix VI:1.2

One limitation is the representativeness of the sample.  The Mexican
sample underrepresents those states in Mexico that have only recently
become sources of U.S.  migration, including states in southern
Mexico and states bordering the United States.  The U.S.  subsample
may undersample people with little connection to their community of
origin, or those who have moved to nontraditional locations in the
United States.  In addition, information on other household members
collected from the head of the household may be inaccurate, and
retrospective data on migration experience may be biased in that
respondents may attribute events that occurred to the incorrect time
period or recall more recent events more accurately than past events. 


   THE ZAPATA CANYON PROJECT
-------------------------------------------------------- Appendix VI:2

Since September 1987, the ongoing Zapata Canyon Project has conducted
personal interviews with randomly selected undocumented immigrants
preparing to cross the border between Mexico and the United States. 
Interviews are conducted 3 days per week--usually weekends--at
habitual crossing points of undocumented immigrants in the Mexican
border cities of Tijuana, Mexicali, Ciudad Juarez, Nuevo Laredo, and
Matamoros.  The survey employs a short, standardized questionnaire
specifically designed not to take too much time from someone who is
in the process of entering the United States illegally.  The project
is directed by Jorge Bustamante of El Colegio de la Frontera Norte
(COLEF), in collaboration with Jorge Santibanez and Rodolfo Corona. 

The surveys include information on the demographic characteristics of
migrants, prior labor experience in Mexico and the United States,
location of border crossing, cost of the trip to the border, whether
a smuggler was used for crossing, how many times migrants were
apprehended by the INS, and reasons for making the trip. 


      STRENGTHS
------------------------------------------------------ Appendix VI:2.1

The survey provides extensive time series information on illegal
border crossing and can be used for examining changes in
sociodemographic characteristics of border crossers. 


      LIMITATIONS
------------------------------------------------------ Appendix VI:2.2

The Zapata Canyon data may not come from a representative sample of
undocumented migrants.  Thus, it is not possible to estimate the
volume of the undocumented flow from Mexico.  The survey does not
focus on the extent of return migration from the United States to
Mexico.  A limited number of questions can be asked of migrants who
are in a hurry to make a clandestine crossing into the United States. 


   SURVEY OF MIGRATION TO THE
   NORTHERN BORDER
-------------------------------------------------------- Appendix VI:3

This survey, also known as EMIF (Encuesta sobre Migracion en la
Frontera Norte de Mexico), was funded by the World Bank through the
Mexican Ministry of Labor and the Ministry of the Interior's National
Council for Population.  A team of researchers from El Colegio de la
Frontera Norte, directed by Jorge Bustamante, designed and
administered the survey.  The survey was designed to produce direct
estimates of the volume of documented and undocumented migration
flows from Mexico to the United States as well as return migration
from the United States to Mexico. 

The survey methodology derives from the theoretical concept of
circular migration and is based on an adaptation of what biological
statisticians call the "sampling of mobile populations." The team
identified the "migratory routes" through which circular migration
occurred and defined empirical points of observation where migration
could be viewed, including bus stations, airports, railroad stations,
and customs and immigration inspection places along highways.  In
these places, systematic counts of the number of migrants were made
over specific periods of time.  The survey was conducted continuously
from March 28, 1993, to March 27, 1994; from December 14, 1994, to
December 13, 1995; and from July 14, 1996, to July 13, 1997. 

The survey sampled four groups of migrants:  (1) illegal migrants
voluntarily deported from the United States by the INS; (2) illegal
and legal migrants preparing to cross the border from Mexico to the
United States; (3) Mexican nationals who were permanent residents of
the United States and were returning to Mexico; and (4) permanent
residents of Mexico who have been in the United States legally or
illegally and were returning to Mexico. 

The survey includes questions on the demographic characteristics of
migrants, prior labor experience in Mexico and the United States,
prior legal and illegal border crossing experience (including number
of previous crossings, whether documents were used to cross, number
of times apprehended by INS, location of border crossings, whether
smugglers were used for crossing, and how much was paid to them), and
reasons for making the current trip. 


      STRENGTHS
------------------------------------------------------ Appendix VI:3.1

The survey provides information on legal and illegal flow in both
directions--from Mexico to the United States, and from the United
States to Mexico. 


      LIMITATIONS
------------------------------------------------------ Appendix VI:3.2

The survey focuses only on labor migrants; nonlabor migrants are not
interviewed.  Data collection has recently been completed; however,
the study directors have applied for additional funding to continue
the survey. 


   THE UNITED STATES-MEXICO
   BINATIONAL STUDY ON MIGRATION
-------------------------------------------------------- Appendix VI:4

After a meeting of the Migration and Consular Affairs Group of the
Mexican-United States Binational Commission in March 1995, the
governments of Mexico and the United States decided to undertake a
joint study of migration between the two countries.  The Binational
Study was funded by both the United States and Mexican governments in
conjunction with private sector funding in both countries.  The main
objective of the Binational Study was to contribute to a better
understanding and appreciation of the nature, dimensions, and
consequences of migration from Mexico to the United States.  National
coordinators were designated for each country, with the Commission on
Immigration Reform coordinating the work of U.S.  researchers.  The
Binational Study was released on September 2, 1997. 

The research was conducted by a team of 20 independent researchers,
10 from each country, who reviewed existing research, generated new
data and analyses, and undertook site visits and consulted with
migrants and local residents to gain a joint understanding of the
issues raised in this study.  The researchers participated in five
teams studying different aspects of migration, including (1) the size
of the legal and illegal Mexican-born population in the United
States, and the size of the migration streams crossing the border;
(2) demographic, educational, and income characteristics of
Mexican-born migrants; (3) factors influencing migration from Mexico;
(4) economic and social effects of migration on both the United
States and Mexico; and (5) societal responses to migration in both
the United States and Mexico, including legislation and policy
responses, court decisions, advocacy from the private sector, and
public opinion. 


      STRENGTHS
------------------------------------------------------ Appendix VI:4.1

The study analyzed a broad number of issues relating to illegal and
legal immigration.  Its use of a combination of data from both the
United States and Mexico enhanced understanding of Mexican migration
and its impacts on both countries.  The study made a number of
specific recommendations for needed research. 


      LIMITATIONS
------------------------------------------------------ Appendix VI:4.2

Data limitations often constrained the study teams' abilities to draw
firm conclusions on many issues.  Data collection and analysis has
been completed, and there are no mechanisms in place for follow-up. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix VII

GENERAL GOVERNMENT DIVISION,
WASHINGTON, D.C. 

Evi L.  Rezmovic, Assistant Director, Administration of Justice
Issues
Barry J.  Seltser, Assistant Director, Technical/Methodological
 Assistance Group
Barbara A.  Stolz, Senior Evaluator
Katherine M.  Wheeler, Publishing Advisor

OFFICE OF THE GENERAL COUNSEL,
WASHINGTON, D.C. 

Ann H.  Finley, Senior Attorney

LOS ANGELES FIELD OFFICE

Michael P.  Dino, Evaluator-In-Charge
Tom Jessor, Senior Evaluator
Nancy K.  Kawahara, Senior Evaluator
James C.  Geibel, Evaluator


BIBLIOGRAPHY
=========================================================== Appendix 0

Bean, Frank D., Roland Chanove, Robert G.  Cushing, Rodolfo de la
Garza, Gary P.  Freeman, Charles W.  Haynes, and David Spener. 
Illegal Mexican Migration and the United States/Mexico Border:  The
Effects of Operation Hold the Line on El Paso/Juarez.  Austin, TX: 
Population Research Center, University of Texas at Austin and U.S. 
Commission on Immigration Reform, July 1994. 

Binational Study on Migration.  Binational Study:  Migration Between
Mexico & The United States.  Mexico City and Washington, D.C.: 
Mexican Foreign Ministry and U.S.  Commission on Immigration Reform,
September 1997. 

Bustamante, Jorge A.  "Mexico-United States Labor Migration:  Some
Theoretical and Methodological Innovations and Research Findings."
Paper prepared for the 23rd General Population Conference of the
International Union for the Scientific Study of Population
(unpublished), October 1997. 

Bustamante, J.A., Jorge Santibanez, and Rodolfo Corona. 
"Mexico-United States Labor Migration Flows:  Some Theoretical and
Methodological Innovations and Research Findings," Migration and
Immigrants:  Research and Policies.  Mexico's Report for the
Continuous Reporting System on Migration (SOPEMI) of the Organization
for Economic Cooperation and Development (OECD), November 1996. 

Chavez, Leo R., Estevan T.  Flores, and Marta Lopez-Garza.  "Here
Today, Gone Tomorrow?  Undocumented Settlers and Immigration Reform."
Human Organization, Vol.  XLIX (1990), pp.  193-205. 

Cornelius, Wayne A., and J.A.  Bustamante (eds.).  Mexican Migration
to the United States:  Process, Consequences, and Policy Options.  La
Jolla, CA:  Center for U.S.  - Mexican Studies, University of
California, San Diego, 1990. 

Cornelius, W.A.  "Impacts of the 1986 U.S.  Immigration Law on
Emigration from Rural Mexican Sending Communities." Population and
Development Review, Vol.  XV, No.  4, (1989). 

Crane, Keith, Beth J.  Asch, Joanna Z.  Heilbrunn, and Danielle C. 
Cullinane.  The Effect of Employer Sanctions on the Flow of
Undocumented Immigrants to the United States.  Santa Monica, CA, and
Washington, D.C.:  The RAND Corporation and The Urban Institute,
April 1990. 

Donato, Katharine M., Jorge Durand, and Douglas S.  Massey. 
"Stemming the Tide?  Assessing the Deterrent Effects of the
Immigration Reform and Control Act." Demography, Vol.  XXIX, No.  2
(May 1992), pp.  139-157. 

Durand, Jorge and Douglas S.  Massey.  "Mexican Migration to the
United States:  A Critical Review." Latin American Research Review,
Vol.  XXVII (1992), pp.  3-42. 

Eschbach, Karl, Jacqueline Hagan, Nestor Rodriguez, Ruben
Hernandez-Leon, and Stanley Bailey.  "Death at the Border." Houston,
TX:  University of Houston Center for Immigration Research (Working
Paper #97-2), June 1997. 

Espenshade, Thomas J.  "Using INS Border Apprehension Data to Measure
the Flow of Undocumented Migrants Crossing the U.S.-Mexico Frontier."
International Migration Review, Vol.  XXIX, No.  2 (Summer 1995), pp. 
545-565. 

Espenshade, Thomas J., and Dolores Acevedo.  "Migrant Cohort Size,
Enforcement Effort, and the Apprehension of Undocumented Aliens."
Population Research and Policy Review, Vol.  XIV (1995), pp. 
145-172. 

Espenshade, Thomas J.  "Does the Threat of Border Apprehension Deter
Undocumented U.S.  Immigration?" Population and Development Review,
Vol.  XX, No.  4 (December 1994), pp.  871-891. 

Espenshade, Thomas J.  "Policy Influences on Undocumented Migration
to the United States." Proceedings of the American Philosophical
Society, Vol.  CXXXVI, No.  2 (1992), pp.  188-207. 

Espenshade, Thomas J., Michael J.  White, and Frank D.  Bean. 
"Patterns of Recent Illegal Migration to the United States." Future
Demographic Trends in Europe and North America:  What Can We Assume
Today?  W.  Lutz (ed.).  London:  Academic Press, 1991, pp.  301-336. 

Heyman, Josiah McC.  "Putting Power in the Anthropology of
Bureaucracy:  The Immigration and Naturalization Service at the
Mexico-United States Border." Current Anthropology, Vol.  XXXVI, No. 
2 (April 1995), pp.  261-287. 

Johnson, Hans.  Undocumented Immigration to California:  1980-1993. 
San Francisco:  Public Policy Institute of California, September
1996. 

Kossoudji, Sherrie A.  "Playing Cat and Mouse at the U.S.-Mexican
Border." Demography, Vol.  XXIX, No.  2 (May 1992), pp.  159-180. 

Lowell, B.  Lindsay (ed.).  Temporary Migrants in the United States. 
Washington, D.C.:  U.S.  Commission on Immigration Reform, 1996. 

Lowell, B.  Lindsay and Zhongren Jing.  "Unauthorized Workers and
Immigration Reform:  What Can We Ascertain from Employers?"
International Migration Review, Vol.  XXVIII, No.  3 (1994), pp. 
427-429. 

Martin, Philip L.  "Good Intentions Gone Awry:  IRCA and U.S. 
Agriculture." ANNALS AAPSS, Vol.  DXXXIV (July 1994), pp.  44-57. 

Massey, Douglas S., and Kristin E.  Espinosa.  "What's Driving
Mexico-U.S.  Migration?  A Theoretical, Empirical and Policy
Analysis." American Journal of Sociology, Vol.  CII, No.  4 (January
1997), pp.  939-999. 

Massey, Douglas S., and Audrey Singer.  "New Estimates of
Undocumented Mexican Migration and the Probability of Apprehension."
Demography, Vol.  XXXII, No.  2 (May 1995), pp.  203-213. 

Massey, Douglas S., and Felipe Garcia Espana.  "The Social Process of
International Migration." Science, Vol.  CCXXXVII (1987), pp. 
733-738. 

Mines, Richard, Susan Gabbard, and Anne Steirman.  A Profile of U.S. 
Farm Workers:  Demographics, Household Composition, Income and Use of
Services.  Washington, D.C.:  Prepared for the U.S.  Commission on
Immigration Reform, April 1997. 

Reyes, Belinda I.  Dynamics of Immigration:  Return Migration to
Western Mexico.  San Francisco:  Public Policy Institute of
California, January 1997. 

Singer, Audrey, and Douglas S.  Massey.  "The Social Process of
Undocumented Border Crossing." Paper presented at the Meetings of The
Latin American Studies Association, Guadalajara, Mexico
(unpublished), April 19, 1997. 

U.S.  Commission on Agricultural Workers.  Report of the Commission
on Agricultural Workers.  Washington, D.C.:  Government Printing
Office, 1992. 

U.S.  Commission on Immigration Reform.  U.S.  Immigration Policy: 
Restoring Credibility.  Washington, D.C., September 1994. 

U.S.  Department of Justice Office of the Inspector General. 
Immigration and Naturalization Service Monitoring of Nonimmigrant
Overstays.  Washington, D.C.:  U.S.  Department of Justice, September
1997. 

U.S.  Department of Labor.  U.S.  Farmworkers in the Post-IRCA
Period.  Washington, D.C.:  U.S.  Department of Labor Research Report
No.  4, March 1993. 

U.S.  Immigration and Naturalization Service.  Operation Gatekeeper: 
Landmark Progress.  Washington, D.C., October 1995. 

U.S.  Immigration and Naturalization Service.  Operation Gatekeeper: 
Two Years of Progress.  Washington, D.C., October 1996. 

Warren, Robert.  Estimates of the Unauthorized Immigrant Population
Residing in the United States, by Country of Origin and State of
Residence:  October 1992.  Washington, D.C.:  U.S.  Immigration and
Naturalization Service (unpublished), 1994. 




RELATED GAO PRODUCTS
============================================================ Chapter 1

Border Patrol:  Staffing and Enforcement Activities (GAO/GGD-96-65,
Mar.  11, 1996). 

Illegal Immigration:  INS Overstay Estimation Methods Need
Improvement (GAO/PEMD-95-20, Sept.  26, 1995). 

Border Control:  Revised Strategy Is Showing Some Positive Results
(GAO/T-GGD-95-92, Mar.  10, 1995). 

Border Control:  Revised Strategy Is Showing Some Positive Results
(GAO/GGD-95-30, Dec.  29, 1994). 

Border Management:  Dual Management Structure at Entry Ports Should
End (GAO/T-GGD-94-34, Dec.  10, 1993). 

Illegal Aliens:  Despite Data Limitations, Current Methods Provide
Better Population Estimates (GAO/PEMD-93-25, Aug.  5, 1993). 

Customs Service and INS:  Dual Management Structure for Border
Inspections Should Be Ended (GAO/GGD-93-111, June 30, 1993). 

Border Patrol:  Southwest Border Enforcement Affected by Mission
Expansion and Budget (GAO/T-GGD-92-66, Aug.  5, 1992). 

Border Patrol:  Southwest Border Enforcement Affected by Mission
Expansion and Budget (GAO/GGD-91-72BR, Mar.  28, 1991). 

*** End of document. ***