This is the accessible text file for GAO report number GAO-04-411 
entitled 'Federal Law Enforcement: Information on Use of Investigation 
and Arrest Statistics' which was released on March 19, 2004.

This text file was formatted by the U.S. General Accounting Office 
(GAO) to be accessible to users with visual impairments, as part of a 
longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov.

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

Report to Congressional Requesters:

United States General Accounting Office:

GAO:

March 2004:

Federal Law Enforcement:

Information on Use of Investigation and Arrest Statistics:

GAO-04-411:

GAO Highlights:

Highlights of GAO-04-411, a report to congressional requesters

Why GAO Did This Study:

The 21st Century Department of Justice Appropriations Authorization Act 
(P.L. 107-273) requires GAO to report on how investigation and arrest 
statistics are reported and used by federal law enforcement agencies. 
This report provides information on (1) the guidance and procedures 
followed by federal law enforcement agencies regarding counting 
investigations and arrests, (2) how investigation and arrest statistics 
are used, and (3) whether multiple agencies are counting and reporting 
the same investigations and arrests. GAO selected six agencies for 
review: the Drug Enforcement Administration (DEA), the Federal Bureau 
of Investigation (FBI), the former Immigration and Naturalization 
Service (INS), the U.S. Marshals Service, the former U.S. Customs 
Service, and the U.S. Postal Inspection Service (USPIS).

What GAO Found:

Guidance and procedures for counting investigations, or cases, are 
generally consistent among the six agencies GAO reviewed. Agencies 
pursue investigations into crimes that have a nexus to their missions, 
such as drug trafficking for the DEA, mail theft for USPIS, and illegal 
aliens for the former INS. Once agents have made the decision to open a 
case, the cases are to be reviewed and approved by a supervisor, and 
details of the case are then entered into the agencies’ case management 
tracking systems. GAO also found agency guidance and procedures for 
counting arrests to be generally consistent among all six agencies. In 
addition, the agencies required supervisory review of the 
justifications for the arrests before they were entered into the 
agencies’ data tracking systems and officially counted.

In general, agencies use investigation and arrest statistics as 
indicators of agency work and as output measures in performance plans, 
budget justifications, and testimonies. In some cases, these data are 
considered in making promotion, bonus, and award determinations. 
However, investigation and arrest statistics are not emphasized in any 
of these activities, but are one of many factors that are considered.

All of the agencies GAO reviewed counted the same investigations and 
arrests when more than one of them participated in the investigative 
and arresting activities. This practice seems appropriate because many 
investigations and arrests would not have occurred without the 
involvement and cooperation of all the agencies that participated. If 
agencies were not allowed to count investigations and arrests in which 
they participated, agencies would be less likely to work together, 
cases would be much smaller, and the desired disruption of high-level 
criminal organizations would be hampered.

The Departments of Justice and Homeland Security, and USPIS reviewed a 
draft of this report and generally agreed with GAO’s findings. 
Technical comments were incorporated as appropriate.


www.gao.gov/cgi-bin/getrpt?GAO-04-411.

To view the full product, including the scope and methodology, click on 
the link above. For more information, contact Paul Jones at (202) 
512-8777 or jonesp@gao.gov.

[End of section]

Contents:

Letter:

Background:

Results:

Concluding Observations:

Agency Comments:

Appendix I: Scope and Methodology:

Appendix II: Drug Enforcement Administration Profile:

Counting Investigations and Arrests:

Use of Investigation and Arrest Statistics:

Appendix III: FBI Profile:

Counting Investigations and Arrests:

Use of Investigation and Arrest Statistics:

Appendix IV: U.S. Postal Inspection Service Profile:

Counting Investigations and Arrests:

Use of Investigation and Arrest Statistics:

Appendix V: U.S. Customs Service Profile:

Counting Investigations and Arrests:

Use of Investigation and Arrest Statistics:

Appendix VI: Immigration and Naturalization Service Profile:

Counting Investigations and Arrests:

Use of Investigation and Arrest Statistics:

Appendix VII: U.S. Marshals Service Profile:

Counting Investigations and Arrests:

Use of Investigation and Arrest Statistics:

Appendix VIII: Drug Trafficking Investigation:

Appendix IX: Child Pornography Investigation:

Appendix X: Counterterrorism Investigation:

Appendix XI: Comments from the U.S. Postal Inspection Service:

Appendix XII: GAO Contacts and Staff Acknowledgments:

GAO Contacts:

Staff Acknowledgments:

Figures:

Figure 1: Arrests for Federal Offenses, by Arresting Federal Law 
Enforcement Agency:

Figure 2: Investigations Closed--Fiscal Years 1998-2001:

Figure 3: Total Arrests--Fiscal Years 1998-2001:

Figure 4: Investigations Closed--Fiscal Years 1998-2001:

Figure 5: Total Arrests--Fiscal Years 1998-2001:

Figure 6: Investigations Closed--Fiscal Years 1998-2001:

Figure 7: Total Arrests--Fiscal Years 1998-2001:

Figure 8: Cases Closed--Fiscal Years 1998-2001:

Figure 9: Total Arrests--Fiscal Years 1998 to 2001:

Figure 10: Investigations Opened and Closed and Arrests--Fiscal Years 
1998-2001:

Figure 11: Warrants Closed by the Marshals Service and by Other 
Agencies--Fiscal Years 1998-2002:

Abbreviations:

ASAC: Assistant Special Agent in Charge:

BJS: Bureau of Justice Statistics:

CAR: Case Activity Report:

CAST: Case Status Subsystem:

DEA: Drug Enforcement Administration:

DHS: Department of Homeland Security:

DSS: Defendant Statistical System:

FBI: Federal Bureau of Investigation:

FMP: Field Management Plans:

ICMS: Investigations Case Management System:

INS: Immigration and Naturalization Service:

ISRAA: Integrated Statistical Reporting and Analysis Application:

ISDBIS: Inspection Service Database Information System:

JTTF: Joint Terrorism Task Force:

PAS: Performance Analysis System:

PTS: Prisoner Tracking System:

SAC: Special Agent in Charge:

SOD: Special Operations Division:

TECS: Treasury Enforcement Communications System:

USPIS: U.S. Postal Inspection Service:

[End of section]

United States General Accounting Office:

Washington, DC 20548:

March 19, 2004:

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

The Honorable F. James Sensenbrenner, Jr.: 
Chairman: 
The Honorable John Conyers, Jr.: 
Ranking Member: 
Committee on the Judiciary: 
House of Representatives:

Congress has expressed interest in issues concerning how federal law 
enforcement agencies count investigations and arrests and about how the 
pursuit of such statistics may be affecting agencies' priorities and 
management and administrative practices. The issues include whether 
multiple agencies are counting and reporting the same investigations 
and arrests; whether agencies are using duplicative investigation and 
arrest statistics to support their requests for more resources; and 
whether agencies are using the statistics to support promotions, 
bonuses, and other awards for investigative personnel. Staff from your 
offices suggested that additional information on the extent of 
agencies' reporting of duplicative investigations and arrests would be 
helpful to them in making budgetary or policy decisions.

The 21st Century Department of Justice Appropriations Authorization 
Act[Footnote 1] required us to report on how investigation and arrest 
statistics are reported and used by federal law enforcement agencies. 
After consulting with your offices, we agreed that our objectives for 
this report would be to determine (1) what guidance and procedures are 
followed by federal law enforcement agencies regarding counting 
investigations and arrests, (2) how investigation and arrest statistics 
are used, and (3) whether multiple agencies are counting and reporting 
the same investigations and arrests.

To address these objectives, we reviewed and analyzed data and 
documentation and interviewed officials at six federal agencies that 
perform investigations and arrest suspects. The agencies selected were 
the Drug Enforcement Administration (DEA), the Federal Bureau of 
Investigation (FBI), Immigration and Naturalization Service (INS), and 
the U.S. Marshals Service, all from the Department of Justice 
(Justice); the U.S. Customs Service (Customs), from the Department of 
the Treasury; and the U.S. Postal Inspection Service (USPIS).[Footnote 
2] We selected DEA, FBI, Customs, INS, and the Marshals Service because 
these agencies reported the highest number of federal arrests, 
according to the Bureau of Justice Statistics (BJS) Compendium of 
Federal Justice Statistics, 2000, the most recent published report 
available at the time of our selection. We selected USPIS because it 
reported the highest number of federal arrests from non-Justice or non-
Treasury agencies. We also reviewed information related to three joint 
investigations involving several of the selected agencies to see how 
they counted arrests while participating with other federal law 
enforcement agencies.

We conducted our review between March 2003 and February 2004 in 
accordance with generally accepted government auditing standards. 
Additional details on our scope and methodology may be found in 
appendix I. Detailed information on the six agencies' use of 
investigation and arrest statistics can be found in appendixes II 
through VII. Detailed information about the three investigations can be 
found in appendixes VIII through X.

Background:

Investigations and arrests are core functions of all federal law 
enforcement agencies. According to BJS, as of June 2000 (the latest 
available information), 69 federal agencies employed 88,000 full-time 
law enforcement officers.[Footnote 3] Of these, Justice employed more 
than half (58 percent) and the Department of the Treasury employed 21 
percent. During fiscal year 2001, according to BJS, these two agencies 
accounted for the highest percentage of arrests for federal offenses, 
with 71 percent made by Justice components and 12 percent by Treasury 
components. In addition, state and local agencies made 4 percent of the 
arrests, and 7 percent of the arrests were made after the suspects 
voluntarily reported to the Marshals Service following a summons. The 
remaining 6 percent of the arrests were made by other agencies or were 
undesignated in the statistics.

Suspects arrested by federal agencies for federal crimes are 
transferred to the custody of the Marshals Service for processing, 
transportation,and detention. According to BJS, in 2001 the Marshals 
Service received 118,896 suspects from federal law enforcement 
agencies, including those the Marshals arrested themselves. Of these 
arrests, 28 percent were for drug offenses; 21 percent for immigration 
offenses, 16 percent for supervision violations; 14 percent for 
property offenses (such as embezzlement, forgery, burglary, and motor 
vehicle theft); 8 percent were for public-order offenses;[Footnote 4] 5 
percent for weapons offenses; 4 percent for violent offenses; and 3 
percent to secure and safeguard a material witness.

Figure 1 shows the number of suspects arrested only for federal 
offenses for the agencies we reviewed. Some suspects arrested by 
federal agents are transferred to state and local jurisdictions for 
prosecution for nonfederal crimes. For example, according to BJS, in 
fiscal year 2001 DEA arrested 11,778 suspects for federal offenses who 
were booked by the Marshals Service. However, DEA's statistical 
reporting system recorded over 35,000 arrests that same year; the 
additional suspects were turned over to state and local authorities and 
were not booked through the Marshals Service, according to BJS. 
Similarly, BJS reported that USPIS arrested 1,226 suspects for federal 
offenses who were booked by the Marshals Service in fiscal year 2001, 
while USPIS told us that they had made 4,698 federal arrests that same 
year.

Figure 1: Arrests for Federal Offenses, by Arresting Federal Law 
Enforcement Agency:

[See PDF for image]

Note: BJS statistics only include suspects arrested for federal 
offenses who are booked by the Marshals Service. Agencies' arrest 
totals are higher than those shown in figure 1.

[End of figure]

Results:

Guidance and Procedures for Counting Investigations and Arrests Are 
Generally Consistent among Agencies:

The guidance and procedures for counting investigations, or "cases" as 
they are sometimes referred to, are generally consistent among the six 
agencies we reviewed.[Footnote 5] Agencies pursue investigations into 
crimes that have a nexus to their respective missions, such as drug 
trafficking for DEA, mail theft for USPIS, and illegal aliens for INS. 
Generally, according to their guidance and procedures, agencies open 
cases that result from tips or leads received from confidential 
informants or other sources, they may be invited to help in other 
agencies' cases, or they may participate in task force investigations. 
Once the agents have made the decision to open a case, the cases are to 
be reviewed and approved by a supervisor, and details of the case are 
then entered into the agencies' case management tracking systems and 
counted.

We also found agency guidance and procedures for counting arrests to be 
generally consistent among all six agencies. That is, agents must be 
directly involved in the arrest, assist other law enforcement personnel 
in making the arrest, or provide information that leads to an arrest. 
For example, according to DEA's Agents Manual, agents are to count 
drug-related arrests only when DEA is directly involved in the arrest. 
Similarly, USPIS inspectors are to count arrests when physically 
present or providing assistance. According to USPIS's Inspection 
Service Manual, inspectors are to count an arrest when:

* an inspector participated personally in making an arrest or 
contributed significantly to an investigation resulting in an arrest 
made by another law enforcement agency;

* an inspector's investigative efforts with another law enforcement 
agency motivate and materially contribute to the identity and arrest of 
a person for a postal crime event though the inspector was not present 
at the time; or:

* an inspector's investigation of a postal offense develops additional, 
significant evidence that is brought to the prosecutor's attention that 
leads to prosecution for an additional offense.

The determination of material contribution, as used in the Inspection 
Service Manual, is left to the judgment of a supervisor. For example, 
according to USPIS officials, if a postal inspector alerted the highway 
patrol to an individual suspected of mail theft, or if the inspector 
was conducting an ongoing investigation on the suspect and the highway 
patrol made the arrest, the arrest would be claimed by USPIS even 
though the postal inspectors did not make the physical arrest.

The other agencies in our review used similar criteria for counting 
arrests. In addition, the agencies required supervisory review of the 
justifications for the arrests before they were entered into the 
agencies' data tracking systems and officially counted.

In addition to their guidance and procedures for counting 
investigations and arrests, three of the agencies in our review--DEA, 
FBI, and USPIS--have an inspection process to, among other things, 
review the appropriateness of investigations and arrests that are made. 
DEA, for example, told us that its Inspections Division periodically 
validates a sample of arrests and screens them for any type of 
questionable activity, such as "piggy backing" arrests. According to a 
DEA official, piggy backing is when state and local law enforcement 
agencies perform the investigative work and a DEA agent goes along for 
the arrest, writes it up, and claims credit for the arrest. The 
official said that the Inspections Division has consistently found a 
very low percentage of questionable arrests, but that a database of 
questionable arrests has not been accumulated. The official gave one 
example from DEA's New York Field Office, where 2 or 3 questionable 
arrests out of over 8,000 were found. The official indicated that 
questionable arrests are mostly isolated incidents and are not part of 
any systemic problems. The official concluded that if questionable 
arrests were found, those arrest statistics would be removed from the 
agency database. FBI and USPIS officials said that no questionable 
arrests were found during their reviews.

Investigation and Arrest Statistics Are Used for a Variety of Purposes:

Agency officials with whom we spoke told us that investigation and 
arrest statistics are used for many purposes, depending on the 
circumstances. In general, the officials said that statistics serve as 
indicators of agency work and as output measures in performance plans, 
budget justifications, testimonies, and for some agencies, are 
considered in making promotion, bonus, and award determinations. 
Officials at the agencies we reviewed said that investigation and 
arrest statistics are not emphasized in any of these activities, but 
are one of many factors that are considered when reporting agency 
results or when making personnel decisions.

Agencies Reported Investigation and Arrest Statistics in Their Budget 
Justifications, Congressional Testimonies, and Other Public Documents:

We found that agencies generally reported investigation and arrest 
statistics in their budget justifications, congressional testimonies, 
and/or other public documents. These statistics, however, were not the 
only criteria used as indicators of agency workload and productivity.

We reviewed FBI budget requests for fiscal years 2003 and 2004, for 
example, and found numbers of investigations and arrests listed in the 
documents, as well as numbers of indictments and convictions. DEA's 
budget requests to Congress also included investigation and arrest 
statistics. For example, in its fiscal year 2003 budget request, DEA 
reported on an operation that resulted in 38 arrests, and a table in 
DEA's fiscal year 2004 budget request entitled "Domestic Enforcement" 
showed the number of national/local investigations, investigations 
completed, and total investigations.

Conversely, INS did not cite investigation and arrest statistics in its 
budget justification documents. Investigation and arrest activities 
were discussed, however, in the very broadest terms. For example, in 
its 2003 budget justification documentation, INS indicated that it 
would initiate high-priority investigations, conduct asset seizures, 
and present individuals for prosecution for alien smuggling-related 
violations to disrupt the means and methods that facilitate alien 
smuggling. And, in its 2002 documentation, INS noted that, as a result 
of its efforts, many alien smugglers, fraud organizations, and 
facilitators were arrested and presented for prosecution; assets were 
seized; and aliens with a nexus to organized crime, violent gangs, drug 
trafficking gangs, or who have terrorist-related affiliations, were 
apprehended.

Concerning inclusion of arrest statistics in congressional testimony, 
DEA's Administrator's testimonies to Congress on the agency's budget 
requests often contained references to successful cases that resulted 
in arrests. For example, in the fiscal year 2003 budget request, the 
Administrator said that an operation resulted in 14 arrests. The 
Administrator's testimony for fiscal year 2004 also included similar 
examples, such as DEA disrupting 30 drug trafficking organizations and 
dismantling 15 others. In addition, in congressional testimony on the 
fiscal year 2002 budget request, the FBI Director said that overall, 
during fiscal year 2000, FBI investigations contributed to the 
indictment of over 19,000 individuals, the conviction of over 21,000 
individuals, and the arrest of more than 36,000 individuals.

As an example of using investigation statistics in public documents, on 
September 2, 2003, DEA listed on its Web site 37 major operations that 
it had been involved in from 1992 to 2003. Many of these listings 
detailed major investigations involving joint operations with other 
federal, state, and local law enforcement agencies that resulted in 
disruptions and dismantlement of narcotics trafficking operations and 
in numerous arrests. In addition, many of the listings gave credit to 
the other participating agencies for their work on the same cases.

Agency officials said, however, that the investigation and arrest 
statistics are only one of many factors used as indicators of agency 
workload and productivity and are not emphasized in reporting results 
of agencies' workload performance. For example, DEA officials told us 
that instead of pursuing numbers of investigations and arrests, their 
focus is on:

* targeting, disrupting, and dismantling major drug trafficking 
organizations;

* working cooperatively and closely with other federal, state, and 
local law enforcement agencies; and:

* making an impact on reducing the flow of narcotics and dangerous 
drugs into the United States.

Numbers of Investigations and Arrests Are Used as Workload Indicators 
and, to a Limited Extent, Used in Promotion, Bonus, and Award 
Decisions:

Agency officials with whom we spoke told us that investigation and 
arrest statistics are used as measures of productivity and indicators 
of workload activity, but only to a limited extent in personnel 
management activities such as promotions, bonuses, and awards. USPIS 
officials, for example, said that investigation and arrest statistics 
are only one of many indicators of an individual's performance and are 
not required for making promotion and other personnel decisions. The 
knowledge, skills, and abilities for promotion do not list "number of 
arrests" as a competency. For example criteria for promotion to the 
manager level, GS-15, is based on competencies including customer 
focus, interpersonal skills, problem identification and analysis/
decision making, strategic leadership, and oral and written 
communication.

USPIS officials said that awards and bonuses are usually given for 
performances above and beyond normal expectations, not just for making 
arrests. An inspector or team, for example, that makes a large number 
of arrests as culmination of an investigation could receive an award, 
according to the officials.

The other agencies we reviewed generally followed similar criteria on 
the use of investigation and arrest statistics in performance 
management decisions as that described earlier by the USPIS officials. 
For example, INS's promotion criteria considered such factors as job 
experience, decision-making, managerial writing, and job simulation.

Multiple Agencies Count the Same Investigations and Arrests in Their 
Individual Agency Databases and Reports:

All of the agencies we reviewed counted the same investigations and 
arrests when more than one of them participated in the investigative 
and arresting activities. In two of the three reviews of joint 
investigations that we performed, agencies reported that they each 
counted some of the same arrestees involved in the investigations. (See 
apps. VIII and IX.) Agency officials told us that they believe that 
this practice is appropriate, because in their opinion, many 
investigations and arrests would not have occurred without the 
involvement and cooperation of all the agencies that participated. If 
agencies were not allowed to count investigations and arrests in which 
they participated, the officials said that agencies would be less 
likely to work together, cases would be much smaller, and the desired 
disruption of high-level criminal organizations would be hampered.

In general, the agency press releases and Web sites we reviewed gave 
credit to one another when they jointly participated in major 
investigations that resulted in a number of arrests. We found several 
examples of this practice, but did not find any overall federal 
database that would identify joint investigations and arrests that were 
conducted by multiple federal law enforcement agencies. Several of the 
agencies' internal databases, however, are capable of identifying joint 
investigations and arrests, while others could possibly be so modified, 
according to agency officials. For example, DEA's statistical database 
was able to identify arrests made by DEA unilaterally, as well as those 
made jointly with other federal, state, and local law enforcement 
agencies, as shown in figure 3.

Concluding Observations:

Federal law enforcement agencies are in business to enforce the 
nation's law and regulations, investigate the activities of criminal 
organizations, and arrest individuals suspected of criminal activity. 
Increasingly, federal law enforcement agencies do not pursue these 
activities in a vacuum. All involved agencies do count the same 
investigations and arrests resulting from joint operations, present 
these statistics in their public documents and budget justifications, 
and consider their actions justified. There is no central federal 
repository of joint investigations and arrests conducted by the 
agencies we reviewed. Moreover, not all of the agencies currently 
distinguish between unilateral and joint arrests and investigations 
within their databases. Making this distinction may help guide Congress 
when making budget decisions about these agencies. Also, the agencies 
can provide, or if instructed, modify their databases to reflect more 
refined information. However, we did not evaluate what cost, if any, 
would be associated with requiring agencies to do so.

Agency Comments:

We provided a draft copy of this report to Justice, DHS, and USPIS. 
Justice and DHS indicated that they had no further comments on our 
draft; however, technical clarifications were provided during our exit 
meetings. USPIS agreed with our report's overall finding that federal 
law enforcement agencies are generally consistent in the way they 
report and make use of investigation and arrest statistics. However, 
USPIS provided technical comments which we have incorporated, as 
appropriate. USPIS's written comments are reproduced in appendix XI.

We are providing copies of this report to the Attorney General, the 
Secretary of Homeland Security, and the Postmaster General. We will 
also make copies available to others on request. In addition, the 
report will be available at no charge on GAO's Web site at http://
www.gao.gov.

If you or your staffs have any questions regarding this report, please 
contact Darryl W. Dutton at (213) 830-1086 or me at (202) 512-8777. Key 
contributors to this report are listed in appendix XII.

Paul L. Jones: 
Director, Homeland Security and Justice Issues:

[End of section]

Appendix I: Scope and Methodology:

Overall, to address our objectives, we selected six federal agencies 
that perform investigations and arrest suspects. The agencies selected 
were the Drug Enforcement Administration (DEA), the Federal Bureau of 
Investigation (FBI), and the U.S. Marshals Service from the Department 
of Justice (Justice); the former U.S. Customs Service (Customs) and 
Immigration and Naturalization Service (INS), now part of the 
Department of Homeland Security (DHS);[Footnote 6] and the U.S. Postal 
Inspection Service (USPIS). We selected DEA, FBI, Customs, INS, and the 
Marshals Service because these agencies reported the highest number of 
federal arrests, according to the Bureau of Justice Statistics (BJS) 
Compendium of Federal Justice Statistics, 2000, the most recent data 
available at the time of our selection. We selected USPIS because it 
reported the highest number of federal arrests from non-Justice or non-
Treasury agencies. In addition, our review focused on fiscal years 1998 
through 2001 as mandated by the 21st Century Department of Justice 
Appropriations Act.[Footnote 7] Our review focused on agencies' 
policies and procedures used to count investigations and arrests, not 
on the number of investigations conducted and arrests made. Therefore, 
we did not perform reliability assessments of data systems at the 
selected agencies.

To identify the guidance and procedures followed by federal law 
enforcement agencies regarding counting and reporting investigation and 
arrest statistics, we reviewed agency mission statements, policies, and 
applicable manuals concerning investigations and arrests. We also 
obtained information about agency investigation and arrest statistical 
tracking systems -for example, DEA's Case Status Subsystem; FBI's 
Integrated Statistical Reporting and Analysis database; and USPIS's 
Inspection Service Database Information System. We also obtained (1) 
overall statistics of investigations and arrests by federal law 
enforcement agencies compiled by BJS and used in its Compendium of 
Federal Justice Statistics, for fiscal years 1998, 1999, 2000, and 
2001, the latest Compendium available at the time of our review, and 
(2) selected Office of Inspector General and internal agency inspection 
reports concerning the use of investigation and arrest statistics. We 
also interviewed officials from each agency who were responsible for 
reporting, compiling, analyzing, and disseminating investigation and 
arrest statistics.

To determine how investigation and arrest statistics are used, we 
reviewed selected agency budget justifications that were submitted to 
the Congress, congressional testimonies used to justify congressional 
appropriations, and internal agency manuals and policies for use of 
investigation and arrest statistics. We also reviewed guidance on 
issues such as promotion, bonus, and award criteria for agents and 
interviewed officials who used investigation and arrest statistics in 
their administrative and management systems. Our review was performed 
primarily at the agencies' headquarters office in Washington, D.C. 
However, to obtain the perspective of field staff regarding the use of 
investigation and arrest statistics for administrative and management 
purposes, we spoke with key DEA, USPIS, Customs, and Marshals Service 
staff at their Los Angeles offices.

To determine if multiple agencies are reporting the same investigations 
and arrests, we obtained, when available, information from agency 
statistical systems - such as DEA's Defendant Statistical System. We 
wanted to know whether (1) other law enforcement agencies were involved 
in investigations or arrests as part of joint investigations, and (2) 
the individual agencies could be distinguished from each other. We 
searched for and obtained from congressional testimony and agencies' 
Web sites examples of major investigations involving more than one 
agency and analyzed selected agency budget justifications (e.g., FBI 
and INS) and performance reports to determine how investigation and 
arrest statistics were reported to the Congress. We also interviewed 
agency officials and obtained documents to explain the reasons for 
either counting or not counting investigations and arrests when other 
federal, state, or local law enforcement agencies were involved in the 
investigations.

We also conducted assessments of three joint investigations to 
determine the extent to which agencies were or were not counting the 
same arrests: a drug trafficking investigation, a child pornography 
investigation, and a counterterrorism investigation. For the drug 
trafficking investigation, we searched our selected agencies' Web 
sites, where available, for joint operations and found Operation 
Marquis on DEA's Web site. Operation Marquis was a DEA-led 
investigation that involved the FBI and several other federal law 
enforcement agencies. We subsequently asked DEA and FBI for lists of 
arrestees from Operation Marquis and matched them to determine whether 
both agencies were counting the same arrestees. (See app. VIII for 
additional information on Operation Marquis.) For the child pornography 
case, we asked Customs whether it had a joint investigation that 
included one of the other federal agencies among our selected agencies. 
Customs recommended that we use Operation Bayou Blaster, which also 
involved USPIS. Again, we asked both Customs and USPIS to provide us 
with lists of arrestees associated with Operation Bayou Blaster. 
Customs provided us with a list, but USPIS was unable to generate a 
list.[Footnote 8] Consequently, we asked USPIS to crosscheck its 
database to the list of arrestees provided by Customs. (See app. IX for 
additional information on the child pornography investigation.) For the 
counterterrorism joint investigation, we discussed a Joint Terrorism 
Task Force operation with FBI and INS officials. The FBI and INS 
provided us with names of arrestees associated with the operation and 
included in agency arrest statistics for comparison purposes. (See app. 
X for additional information on the Counterterrorism Joint Task Force 
investigation.):

We conducted our work at selected agency headquarters in Washington 
D.C., and at DEA, USPIS, Customs, and Marshals Service offices in Los 
Angeles, California.

[End of section]

Appendix II: Drug Enforcement Administration Profile:

The Drug Enforcement Administration's (DEA) mission is to, among other 
things,

* enforce the controlled substances laws and regulations of the United 
States and:

* investigate and prepare for the prosecution of major violators of 
controlled substance laws operating at interstate and international 
levels.

To perform its mission, in fiscal year 2003, DEA had a total of 9,629 
employees including 4,680 special agents operating in 225 offices in 
the United States and in 80 other offices throughout the world. DEA's 
budget was $1.5 billion in fiscal year 2003.

Counting Investigations and Arrests:

A DEA investigation is referred to as a "case," and involves targeting 
organizations or businesses suspected of illegal narcotics trafficking. 
Any given case can include one or multiple organizations or 
individuals, but it is counted only as one case.

* DEA's Case Status Subsystem (CAST) is the system used to track cases. 
CAST identifies, among other things, the target (e.g., criminal 
organization) of the case and whether other agencies are involved. When 
an agent has sufficient cause to open a case, he or she enters general 
information about the case into CAST, such as the file number, agent's 
name, entity under investigation, date opened, and identification 
number.

DEA pursues investigations into drug trafficking organizations in 
several ways. DEA special agents may open cases that result from tips 
or leads received from confidential informants or other sources, may be 
invited to help in other agencies' cases, and may participate in 
interagency task force investigations. When DEA agents initiate their 
own cases, they may also elicit help from other federal, state, and 
local law enforcement agencies. Agents' cases are approved by a 
supervisor and are entered into CAST.

DEA officials told us that their emphasis, instead of pursuing numbers 
of investigations and arrests, is on:

* targeting, disrupting,[Footnote 9] and dismantling[Footnote 10] major 
drug trafficking organizations;

* working cooperatively and closely with other federal, state, and 
local law enforcement agencies; and:

* making an impact on reducing the flow of narcotics and dangerous 
drugs into the United States.

On September 2, 2003, DEA listed on its Web site 37 major operations 
that it had been involved in from 1992 to 2003. Many of these listings 
detailed major investigations involving joint operations with other 
federal, state, and local law enforcement agencies that resulted in 
disruptions and dismantlement of narcotics trafficking operations and 
in numerous arrests. In addition, many of the listings gave credit to 
the other participating agencies for their work on the cases.

DEA provided us with the numbers of cases closed between fiscal years 
1998 and 2001, including DEA unilateral cases and those performed 
jointly with other law enforcement agencies as shown in figure 2.

Figure 2: Investigations Closed--Fiscal Years 1998-2001:

[See PDF for image]

[A] Other is the total of other federal agency referral, unknown, and 
missing cases.

[End of figure]

According to DEA's Agents Manual, agents are to claim (i.e., count) 
drug-related arrests only when DEA is directly involved in the arrest. 
DEA's process for counting and recording arrests also includes having a 
supervisory agent review and sign each Form 202, Personal History 
Report, used to document personal information on each person arrested. 
By signing the form, the DEA supervisor attests that DEA directly 
participated in the arrest or, in the case of a foreign arrest, 
provided substantial assistance. An Associate Deputy Chief, Office of 
Inspections, said that DEA's decision about whether to count an arrest 
is contingent on several factors, including a clear nexus to a drug 
offense, involvement of a DEA informant or DEA monies, or the physical 
presence of DEA agents at the time of the arrest, and/or a significant 
role by DEA agents.

DEA's Defendant Statistical System (DSS) tracks the number of arrests 
counted by DEA. Each Form 202--there is one Form 202 for each person 
arrested--is completed and the information entered into the DSS. An 
arrest can only be entered in DSS once. When the Form 202 information 
is entered into DSS, duplicate arrests, if any, will be identified and 
appropriate divisions will be notified to fix the problems. In 
addition, DEA has a manual system, the Drug Enforcement Arrest Log, 
which is used as a check against arrests entered into DSS.

As with investigation statistics, DEA was able to provide us with 
numbers of unilateral and joint arrests by fiscal year from 1998 to 
2001. Figure 3 shows numbers of arrests and agencies that were involved 
in the arrests but does not identify the agency(s) making the actual 
physical arrest.

Figure 3: Total Arrests--Fiscal Years 1998-2001:

[See PDF for image]

[A] Other is the total of other federal agency referral, unknown, and 
missing cases.

[End of figure]

According to an Associate Deputy Chief, Office of Inspections, DEA 
conducts on-site inspections about every 3 years in its domestic 
offices. As part of the inspection, the division or office is assessed 
to determine if it is successfully achieving DEA's objectives and 
priorities, including a validation of claimed arrests.

Field Management Plans (FMP) are used in the inspections process. FMPs 
describe the priorities set by the Special Agent in Charge (SAC) of a 
field division to counter the field office's unique drug threats and 
delineate the methods for accomplishing the agency's mission and 
priorities. The FMP outlines major operations in the field division 
that have been completed, the number of major drug trafficking 
organizations disrupted or dismantled, and other field division 
highlights. The Office of Inspections assesses the field division's 
adherence to the FMP and its success in achieving the goals in the FMP. 
Case files and the "appropriateness" of arrests are also reviewed.

A DEA Office of Inspections official told us that periodically a sample 
of arrests are validated and screened for appropriateness and for any 
type of questionable activity, such as "piggy backing" arrests. Piggy 
backing is when a state or local law enforcement agency does the 
investigative work and DEA goes along for the arrest, writes it up, and 
claims credit for the arrest. The official said that Inspections has 
consistently found a very low percentage of questionable arrests at the 
offices reviewed, but they do not accumulate or maintain a database of 
questionable arrests. The official gave one example of the New York 
office, where 2 or 3 questionable arrests out of over 8,000 were found. 
The official indicated that questionable arrests are mostly isolated 
incidents, and are not part of any systemic problems. If questionable 
arrests are found, the officials said that the arrest statistics would 
be removed from the agency database.

In another example, out of approximately 2,200 arrests made by the Los 
Angeles office in 1999, the Office of Inspections found 16 that were 
questionable. The Los Angeles SAC said the arrests were questionable 
because the agents had not adequately documented DEA's participation in 
the arrests to qualify for counting them. The Office of Inspections 
finally judged only 3 of the 16 as not being justified. These 3 
involved cases where the Highway Patrol found drugs in suspects' 
vehicles and called DEA out to establish probable cause to make the 
arrest. This is a gray area, according to the SAC, and DEA must show 
significant participation in order to claim the arrest. In the 3 cases, 
significant participation was not shown, and the arrests were not 
allowed, according to the SAC.

The SAC also said that in addition to the periodic Office of Inspection 
evaluations, the Los Angeles office conducts its own yearly self-
inspections, including reviews of the appropriateness of claimed 
arrests.

DEA officials also told us that it is appropriate for each 
participating agency to claim the investigations and arrests that 
result from joint operations. The officials said that if only one 
agency could claim the investigations and arrests, agencies would not 
work together, cases would be much smaller, and the desired outcome of 
disrupting major drug trafficking organizations would not happen.

Use of Investigation and Arrest Statistics:

Overall, DEA officials said that investigation and arrest statistics 
serve as indicators of agency work, to help determine whether or not 
something is being accomplished. DEA Domestic Operations officials said 
that from a managerial standpoint, DEA's focus is more on who and what 
priority target organizations have been disrupted or dismantled. 
Statistical data are useful, the Domestic Operations officials said, 
because they provide a picture of activity; they are the evidence that 
validates the work performed. DEA officials from the Evaluation and 
Planning Section said investigations are DEA's top output measure and 
provide basic information on workload. For example, if a group of 10 
DEA agents and a supervisor had only 7 arrests for a year, management 
would want to look at the group to question its level of activity. It 
could be that the group was engaged in a very long, complicated wiretap 
case, which would not lend itself to many arrests. Arrests, however, 
would only be used as an indicator of activity, according to a SAC and 
Assistant Special Agents in Charge (ASACs).

Investigation and arrest statistics are not used as performance 
indicators in various DEA-related materials. For example, in Department 
of Justice Performance Reports,[Footnote 11] we were unable to find any 
investigation and arrest statistics related to DEA. DEA, however, did 
use numbers of investigations and arrests in its budget justifications 
and congressional testimonies. For example, in the fiscal year 2003 
budget request, DEA reported on an operation that resulted in 38 
arrests. For the fiscal year 2004 budget request, DEA also provided 
tables that included numbers of national and local investigations, 
investigations completed, and total investigations.

According to a SAC and ASACs, investigation and arrest statistics are 
not used for making agent promotion, award, and performance 
determinations. The officials said that promotions, awards, and 
performance ratings are based on many factors, such as levels of 
violator disruption; coordination efforts with interagency task forces; 
cooperative activities with other federal, state, and local law 
enforcement agencies; and furthering DEA's mission. For example, two 
GS-14 field office group supervisors told us that statistics, 
particularly arrest statistics, do not play an important part in 
promotions, bonuses, or other awards.

Promotion criteria used to evaluate agents for promotion to GS-14 and 
GS-15, for example, is based on the following competencies, which do 
not include investigation and arrest statistics:

* acting as a model,

* gathering information and making judgments/decisions,

* interacting with others,

* monitoring and guiding,

* oral communication, and:

* planning and coordinating.

A DEA Career Board official told us that when looking for the best-
qualified applicant for a GS-14 or GS-15 position, a SAC could consider 
investigation and/or arrest statistics. For example, if a reference for 
the applicant is contacted, the reference may say that they highly 
recommend the applicant because of the applicant's work on a certain 
case.

A DEA official in Employee Relations said that awards and/or bonuses 
are distributed based on performance or a special act. A special act 
could include an agent's involvement on a significant arrest or drug 
seizure. An agent's supervisor writes the justification for an award 
and/or bonus. The division head and the Chief of Operations at DEA 
headquarters review the justification.

DEA Domestic Operations officials said that performance evaluations 
that are used as a basis for promotions may well indicate that an agent 
"maintained a high level of cases" or "participated in several 
significant cases," but performance decisions are not justified based 
on sheer numbers alone. The two field office group supervisors also 
said that if DEA has successful investigations, arrests will naturally 
follow but are not emphasized in promotion, bonus, and other award 
decisions. Domestic Operations officials also said that management's 
"tone at the top," which is emphasized throughout DEA, is not on how 
many people were arrested, but on what drug trafficking organizations 
were disrupted and dismantled.

[End of section]

Appendix III: FBI Profile:

The Federal Bureau of Investigation's (FBI) mission is to:

* uphold the law through the investigation of violations of federal 
criminal law;

* protect the United States from foreign intelligence and terrorist 
activities; and:

* provide leadership and law enforcement assistance to federal, state, 
local, and international agencies.

As of January 31, 2002, approximately 11,000 special agents and 16,000 
professional support personnel were located at the FBI's Washington, 
D.C., headquarters and in 56 field offices, approximately 400 satellite 
offices, 4 specialized field installations, and over 40 foreign liaison 
posts. The FBI's budget was $4.6 billion in fiscal year 2003.

Counting Investigations and Arrests:

An FBI investigation is referred to as either preliminary or full 
field. Officials of the Inspection Division told us that facts and 
circumstances have to rise to a certain level to justify opening either 
type of investigation and that the determination is somewhat 
judgmental. Full field investigations are initiated when there is 
information that raises a reasonable suspicion that a crime has been 
committed. If the information received is not deemed sufficient to 
predicate the opening of a full field investigation, but is determined 
to warrant further inquiry on a limited basis in order to determine the 
credibility of an allegation of criminal activity and the need for a 
more in-depth investigative effort, a preliminary investigation can be 
opened. The more sensitive investigations, such as foreign counter 
intelligence, are usually opened as preliminary investigations. 
Preliminary investigations can proceed to full field investigations. 
The officials told us that the amount of information needed to initiate 
an investigation is the same whether the FBI is working alone or is 
involved in a joint investigation.

The Automated Case Support database is the FBI's overall case 
management system that is used to capture information and data 
pertaining to each investigation. An agent initiates an investigation, 
either an FBI investigation or a joint investigation, by opening a hard 
copy investigative file, using an Investigative Summary Form 302. A 
supervisor must then approve the initiation. Information pertaining to 
the investigation is subsequently entered into database.

We asked the FBI to provide us with information on investigations 
closed in fiscal years 1998-2001. The data provided included 
investigations pertaining to drugs, violent crime, white-collar crime, 
counterterrorism, counterintelligence, and cyber crime. These data are 
displayed in figure 4. Finance Division officials were unable to 
distinguish between investigations performed solely by the FBI and 
those performed jointly with other agencies because this information is 
not captured in the database.

Figure 4: Investigations Closed--Fiscal Years 1998-2001:

[See PDF for image]

[End of figure]

The FBI counts an arrest when the subject is taken into custody with a 
warrant, complaint, or indictment or, if arrested for probable cause, 
after the judicial paperwork is obtained. The FBI reports the arrest as 
federal, local, or international.

* Federal arrests are those in which FBI agents acting alone or with 
other law enforcement officers arrest the subject. The FBI does not 
count the arrest if the subject of an FBI investigation is arrested by 
another law enforcement agency without any assistance from the FBI. 
However, if the arrest is part of an FBI-led task force, the FBI does 
count the arrest even if no FBI agent is present.

* Local arrests are those where the FBI supplied information or other 
assistance to a local agency that significantly contributed to the 
probable cause supporting an arrest warrant for an individual who was 
not the subject of an FBI investigation, and FBI agents were not 
involved in making the arrests.

* International arrests are those where the FBI supplied information or 
other assistance to another country that significantly contributed to 
the probable cause supporting an arrest warrant for an individual who 
either was or was not the subject of an FBI investigation, and FBI 
agents were not involved in making the arrests.

Arrests are reported on the Form FD-515, "Accomplishment Report." This 
form captures such information as arrests, convictions, and the various 
investigative techniques used in an investigation. The agent prepares 
the FD-515; the supervisor reviews and approves the form, thereby 
attesting to the arrest as a valid accomplishment. The field office 
then enters the information into the Integrated Statistical Reporting 
and Analysis Application (ISRAA) database. Arrests, as well as other 
accomplishments, are to be entered into the system within 30 days. A 
variety of edit checks are performed to help ensure the reliability of 
the data input into ISRAA, and each field office completes an annual 
audit of the data.

We asked the FBI to provide us with information on arrests in fiscal 
years 1998-2001. These data are displayed in figure 5. While the 
participation of other agencies is noted on the FD-515, this 
information is not entered into ISRAA and, as a result, the FBI could 
not distinguish between arrests made solely by the FBI and those made 
jointly with other agencies.

Figure 5: Total Arrests--Fiscal Years 1998-2001:

[See PDF for image]

[End of figure]

In addition to the supervisory reviews of investigations and arrests 
already noted, supervisory agents are to perform periodic investigative 
file reviews on all investigations being worked by their agents. These 
reviews occur about every 90 days for investigations led by experienced 
agents and about every 30 days for investigations led by less 
experienced agents. During these file reviews, supervisors are to 
monitor the progress of investigations by reviewing investigative work 
completed; verifying compliance with any applicable policies and 
procedures, including those pertaining to any arrests that have been 
made; and assessing the validity of continuing with the investigation. 
FBI officials also told us that field offices' Assistant Special 
Agents-in-Charge periodically are to check supervisory file reviews to 
ensure the adequacy of the review process.

In addition, the FBI's Inspection Division is responsible for reviewing 
FBI field offices and program divisions to ensure compliance with 
applicable laws and regulations and the efficient and economical 
management of resources. The Inspection Division is to inspect all FBI 
units at least once every 3 years. An Inspection Division official told 
us that the division's reviews include information pertaining to 
arrests and that, if it were determined that an arrest had been 
inappropriately counted, action would be taken, including correction of 
the data in ISRAA. The officials also said that they had not found 
evidence of the inappropriate counting of arrests over the period of 
our review, fiscal years 1998-2001.

Inspections Division officials told us that all agencies involved in a 
joint investigation count investigations and arrests and that this is a 
long-standing and accepted practice that is part of interagency 
cooperation. If each agency involved could not count the statistics, 
there would be more competition among agencies and less participation 
in joint investigations, according to the officials.

Use of Investigation and Arrest Statistics:

Investigation and arrest statistics are used as indicators of the FBI's 
work for a variety of purposes, including management of field offices. 
For example, a field office's numbers for a particular type of crime 
might spike upward. Finance Division officials said that if a spike 
occurred, a determination would be made about what happened to account 
for the spike and changes might be made in how the office performs 
investigations or where it focuses its resources. The officials told us 
that field offices use numbers of investigations and arrests as part of 
justifications for resources, along with other factors such as the type 
of crimes investigated by the office and whether there has been 
continual growth in the numbers. Numbers alone are not determinative, 
however, according to Inspection Division officials. The officials said 
that FBI headquarters also looks at trends, where agents are placing 
their investigative emphasis, political factors such as what 
legislation might be on the horizon, and problems agents are 
encountering that might indicate a need for new technology.

Investigation and arrest statistics are also used as performance 
indicators in various FBI-related materials. For example, in Department 
of Justice Performance Reports,[Footnote 12] the number of terrorism 
investigations is reported, along with the number of related 
convictions. The FBI has also used numbers of investigations and 
arrests, as well as indictments and convictions, in its budget requests 
and congressional testimonies. For example, in the fiscal year 2003 
request, fiscal years 2000 and 2001 actual numbers of investigations 
pending, opened, and closed are given, along with numbers of arrests, 
indictments, and convictions. In addition, in congressional testimony 
on the fiscal year 2002 budget request, the FBI Director said that 
overall, during fiscal year 2000, FBI investigations contributed to the 
indictment of over 19,000 individuals, the conviction of over 21,000 
individuals, and the arrest of more than 36,000 individuals.

Concerning performance management measures, an official from the 
Executive Development and Selection Program told us that certain 
numbers of investigations and arrests are not required for promotions, 
bonuses, or awards. The official told us that the statistics are used 
only as indicators of an agent's performance and that it is necessary 
to look at the work behind the statistics. For example, an agent may 
have zero arrests but be involved in a complex investigation that has 
not yet resulted in arrests. Or, an agent may have a high number of 
arrests resulting from relatively simple investigations.

Special agent promotions are scheduled at regular intervals from GS-10 
or GS-11 to GS-13 and are contingent upon the satisfactory work record 
of the individual. Promotions to GS-14 and GS-15 are competitive, but 
the official told us the vacancy announcements are not yet 
standardized. Agents applying for promotion must describe how they meet 
each of the qualifications listed on the announcement. A new system was 
implemented in January 2004 that will emphasize competencies for each 
vacancy. The first four competencies will be core competencies and the 
last three will be specialized. For example, if the vacancy were in the 
counterterrorism unit, experience with counterterrorism would be 
listed. When applying for a promotion, an agent will complete a form 
and address his or her education and training, pre-FBI experience, FBI 
background, and give two examples of how he or she meets each of the 
required qualifications.

The official from the Executive Development and Selection Program said 
that the quality of work experience, rather than quantity, is 
emphasized in the promotion process. There is no baseline for the 
number of investigations or arrests that an agent must demonstrate. 
There is no place provided on the application for investigation or 
arrest statistics, though officials said that an agent could provide 
these in the narrative should he or she choose to do so. For example, 
an agent could indicate that he or she demonstrated leadership by being 
the agent on 25 investigations resulting in 52 convictions; the agent 
could also discuss a specific investigation as an example of 
leadership.

Agent performance is evaluated annually on seven critical elements 
(which do not include investigation and arrest statistics) using a 
meets or does not meet expectation system. A narrative justification is 
required only if the agent does not meet expectations. The critical 
elements include the following:

* investigating, decision making, and analyzing;

* organizing, planning, and coordinating;

* relating with others and providing professional service;

* acquiring, applying, and sharing job knowledge;

* maintaining high professional standards; and:

* communicating orally and in writing.

An officer of the FBI Agents Association confirmed that investigation 
and arrest statistics are not used in the performance appraisal 
process.

The official from the Executive Development and Selection Program told 
us that a noteworthy accomplishment might be used as the basis for a 
special award. The FBI uses its awards program to motivate employees to 
increase productivity and creativity. To receive an award, an agent 
must be shown to have significantly exceeded the requirements of his or 
her position. Field offices can give awards differently. One office 
might give an award for outstanding performance on one investigation. 
Another office might give an award based on sustained success--for 
example, continuous, outstanding performance with making arrests or 
obtaining convictions.

[End of section]

Appendix IV: U.S. Postal Inspection Service Profile:

The U.S. Postal Inspection Service's (USPIS) mission is to protect the 
U.S. Postal Service, its employees, and customers from criminal attack 
and to protect the nation's mail system from criminal misuse.

In fiscal year 2003, USPIS had 1,955 Postal Inspectors operating in 18 
field divisions in the United States with a budget of $521.7 million.

Counting Investigations and Arrests:

USPIS enforces over 200 federal laws in investigations of crimes that 
may adversely affect or fraudulently use the U.S. mail, the postal 
system, or postal employees. USPIS cases involve crimes that have a 
nexus to the Postal Service. For example, postal-related violations 
including mail theft, identity fraud, child exploitation, illegal 
drugs, and money laundering are investigated by USPIS.

A USPIS investigation is referred to as a "case." To open a case, an 
inspector fills out a request for case, Form 623. A Form 623 must be 
approved and signed by a supervisor before a case is opened. The case 
is then entered into the Inspection Service Database Information System 
(ISDBIS). In the narrative section of Form 623, inspectors at their 
discretion may note whether the case is a joint investigation and 
whether any other agencies are involved; however, indicating whether a 
case is a task force operation or whether other agencies are involved 
is optional. A data entry operator will enter this information into the 
narrative section of the case in ISDBIS, which cannot currently be 
retrieved to identify USPIS cases alone or USPIS as part of a joint 
operation.

A USPIS official in the Information Technology Division, however, told 
us that USPIS plans to launch a new database, which will be fully 
operational in Fall 2004 and will identify cases as joint task force 
operations when applicable and the other law enforcement agencies 
involved. However, it would still be optional for inspectors to 
indicate whether a case is a joint task force operation. The new system 
will be able to run reports listing how many cases are joint task force 
cases and identify the agencies that are participating. For example, a 
report could be generated naming all cases USPIS is working on with the 
FBI.

USPIS provided us with numbers of cases closed from fiscal years 1998 
to 2001, as shown in figure 6.

Figure 6: Investigations Closed--Fiscal Years 1998-2001:

[See PDF for image]

[End of figure]

According to USPIS's Inspection Service Manual, inspectors are to claim 
(i.e., count) an arrest when:

* an inspector participated personally in making an arrest or 
contributed significantly to an investigation resulting in an arrest 
made by another law enforcement agency;

* an inspector's investigative efforts with another law enforcement 
agency motivate and materially contribute to the identity and arrest of 
a person for a postal crime even though the inspector was not present 
at the time; or:

* an inspector's investigation of a postal offense develops additional, 
significant evidence that is brought to the prosecutor's attention.

The determination of material contribution is left to the supervisor's 
judgment. For example, if a postal inspector alerted the highway patrol 
to an individual suspected of mail theft, or if the inspector was 
conducting an ongoing investigation on the suspect and the highway 
patrol made the arrest, the arrest would be claimed by USPIS even 
though postal inspectors did not make the physical arrest.

An inspector fills out a Case Activity Report (CAR) to report case 
statistics and to summarize significant developments in the case, 
including arrests. A supervisor must approve each CAR and the arrests 
before entering into ISDBIS. After a CAR is submitted and entered into 
ISDBIS by a data entry operator, a Case Summary Report is printed and 
sent to the originating inspector to be verified for accuracy. ISDBIS 
tracks the number of arrests counted by USPIS. Arrests made as a result 
of joint operations are counted the same as those that result from 
investigations involving USPIS only. ISDBIS currently cannot sort 
arrests made as a result of joint operations and those in which only 
USPIS was involved. According to an official in the Information 
Technology Division, the new database, scheduled to be implemented 
early next year, can separate arrests resulting from joint operations 
from those involving USPIS alone.

USPIS provided us with the total number of arrests made from fiscal 
years 1998 to 2001 as shown in figure 7.

Figure 7: Total Arrests--Fiscal Years 1998-2001:

[See PDF for image]

[End of figure]

Prior to fiscal year 2003, USPIS had an Office of Inspections Division 
that had overall responsibility for conducting quality assurance 
reviews of field divisions and headquarters groups/divisions. As part 
of a USPIS reorganization in fiscal year 2003, the quality assurance 
review responsibility was assigned to the Strategic Planning and 
Performance Management Group at USPIS headquarters. A USPIS official in 
the Strategic Planning and Management Group told us that the group is 
to review each field office every 3 years for compliance to USPIS 
policies and procedures. In addition, according to a Deputy Chief 
Inspector, USPIS field offices are to conduct annual comprehensive 
self-assessments and that the contents of case files are to be reviewed 
for accuracy during that process. For example, arrests counted and 
hours worked on an investigation are two items that are reviewed for 
accuracy. The official said that overall, USPIS has found minimal 
problems with the accuracy of its case files and has not found any 
problems with a specific USPIS field division counting incorrect case 
and arrest statistics.

A Deputy Chief Inspector and an Assistant Chief Inspector told us that 
double counting investigation and arrest data is acceptable and vital 
in showing the results of USPIS efforts. For example, the officials 
said that if arrests generated by a joint operation could be claimed by 
only one of the involved agencies, turf battles would result. In 
addition, one agency may end up "looking better than another." This 
would result in law enforcement agencies refusing to work with one 
another and there would be no more task forces, according to the 
officials. The officials also said that task forces are needed because 
cases are often complex in nature, and joining forces with other law 
enforcement agencies streamlines, economizes, and makes operations more 
efficient.

Use of Investigation and Arrest Statistics:

A Deputy Chief Inspector and an Assistant Chief Inspector told us that 
Inspectors in Charge of field divisions determine where resources are 
needed by several indicators, including case and arrest statistics. 
They told us that a "briefing book" is prepared monthly using case and 
arrest data from ISDBIS. The briefing book contains an overview of each 
functional area (i.e., fraud, dangerous mailings, and child 
exploitation) and an analysis on whether USPIS is meeting its goals. 
The data in the briefing book are shared with all field offices. The 
officials said that looking at indicators and determining where 
employees are assigned is a part of their management system.

A Deputy Chief Inspector also told us that investigation and arrest 
statistics are used as indicators in performance measurement and 
planning. The performance plan provides the basis for performance 
agreements, or field division contracts, and an annual performance 
report. The performance plan is divided into operational objectives 
that support USPIS's strategic goals. Each operational objective has an 
indicator(s), which measures how closely USPIS met the objective, and a 
target to meet in the upcoming fiscal year. Each field division is 
evaluated on its results as measured against its objectives for each 
fiscal year; for example:

* Operational objective: Identify and resolve domestic and 
international in-transit mail theft.

* Indicators: Major domestic and international airport mail theft 
problems resolved (arrests count towards resolving mail theft cases).

* Target: Thirty major domestic and international airport mail theft 
problems resolved.

USPIS also issues an annual report of investigations, which is a 
document on the fiscal year's activities, to the Postmaster General, 
the Postal Service Board of Governors, and postal managers and 
employees. The report details USPIS's investigative activities in 
criminal areas, such as mail theft and robbery. The number of arrests 
and convictions in each criminal area are also listed in the report.

USPIS officials also told us that data on an inspector's numbers of 
investigations and arrests are not required for making promotion, 
bonus, and award determinations. The knowledge, skills, and abilities 
requirements for promotion to the GS-13 and GS-14 levels, for example, 
do not list "number of arrests" as a competency. The officials said 
that promotion criteria to the manager level, GS-15, is based on 
competencies, including:

* customer focus,

* interpersonal skills,

* problem identification and analysis/decision making,

* supervisory/management skills,

* strategic leadership,

* planning and organizing,

* project/program management,

* oral communication,

* written communication,

* ability to work autonomously, and:

* ability to be flexible in diverse situations.

USPIS officials said that awards and bonuses are usually given for 
performances above and beyond normal expectations, not just for making 
arrests. An inspector or team, for example, that makes a large number 
of arrests as culmination of an investigation could receive an award 
for the arrests, according to the officials.

Concerning inspector performance measurement, a Deputy Chief Inspector 
and an Assistant Chief Inspector said that investigation and arrest 
statistics are one of many indicators of an individual's performance. 
Performance management focuses on linking an inspector's goals to 
national goals, rather than on arrest quotas. Team leaders in the field 
are responsible for helping individual inspectors set goals for the 
year through creating a performance achievement plan.

A Deputy Chief Inspector told us that case and arrest data are not used 
in their budget process. USPIS's budget is historically based--meaning 
it is based on the previous year's budget. It is dependent on U.S. 
Postal Service finances and is not subject to the congressional 
appropriations process.

[End of section]

Appendix V: U.S. Customs Service Profile:

The U.S. Customs Service was responsible for ensuring that all goods 
and persons entering and exiting the United States did so legally and 
was to, among other things,[Footnote 13]

* assess and collect Customs duties, excise taxes, fees, and penalties 
due on imported merchandise;

* interdict and seize contraband, including narcotics and illegal 
drugs; and:

* protect American business and labor and intellectual property rights 
by enforcing U.S. laws intended to prevent illegal trade practices.

To accomplish its mission, in fiscal year 2002, Customs had a workforce 
of over 20,754 employees, 3,031 of which were Special Agent Criminal 
Investigators. Customs' budget was $3.6 billion in fiscal year 2002.

Counting Investigations and Arrests:

From fiscal years 1998 to 2001, the terms investigations and cases 
meant the same thing and were initiated by special agents working from 
information received from various sources, tips, or confidential 
informants. Cases could have also been initiated by other federal, 
state, and local law enforcement agencies requesting assistance or 
through joint task force operations. Customs could also have requested 
other agencies' assistance on its cases, and most cases did involve 
other federal, state, and local law enforcement agencies, according to 
Customs Office of Investigation officials.

Customs officials also said that in fiscal years 1998 to 2001, cases 
were entered into the Case Management System within the Treasury 
Enforcement Communications System (TECS) when agents, with first line 
supervisor's approval, originated them.[Footnote 14] A case number was 
created, which included the office identifier (e.g., Los Angeles), the 
type of case (e.g., money laundering, narcotics, etc.), and information 
on how the case got started and who originated it. Other federal agency 
participation was usually mentioned in the case summary, but was not 
identified in a separate TECS field used for generating case 
statistics, according to the officials.

As shown in figure 8, Customs provided us with the numbers of cases 
closed between fiscal years 1998 and 2001.

Figure 8: Cases Closed--Fiscal Years 1998-2001:

[See PDF for image]

[End of figure]

According to Customs officials, Customs agents could have counted an 
arrest if they:

* physically made the arrest;

* assisted in making the arrest; or:

* discovered the violation, but the arrest was made by other law 
enforcement officers.

During fiscal years 1998 to 2001, numbers of arrests were captured in 
TECS when agents filled out Reports of Investigation and entered the 
details into the system. Customs officials said that agents were to 
provide details of their participation in the arrests in the case file. 
The narrative was to be reviewed and approved by a first line 
supervisor before the arrests would be counted. TECS data fields 
required agents to record whether they were the arresting officers, and 
if a Customs agent was not the arresting officer, then the arresting 
officer's name and agency was to be input into the system. Customs 
agents would not count as an arrest the stopping or detaining of an 
individual for questioning. Numbers of arrests were not formally 
audited for questionable claims, and it was up to the first and second 
line supervisors to check the integrity of the investigations and 
subsequent arrests that were counted. Figure 9 shows numbers of arrests 
and agencies that were involved in the arrests.

Figure 9: Total Arrests--Fiscal Years 1998 to 2001:

[See PDF for image]

[End of figure]

Customs officials said that TECS data entry systems are to preclude two 
or more Customs offices from claiming the same investigations or 
arrests, so there was no double counting within Customs. Whether other 
agencies counted investigations and arrests that they worked on with 
Customs was not clear to the officials, but they assumed that they did, 
and rightfully so. The officials said that cooperation and trust in 
working together would be destroyed if participating agencies were not 
allowed to count investigations and arrests in which they participated.

Use of Investigation and Arrest Statistics:

From fiscal years 1998 to 2001, Customs officials told us that 
investigation and arrest statistics were used as one way to measure 
productivity. The officials said, however, that arrests were only one 
element in a number of performance management measures, including 
numbers of cases, seizures, indictments, prosecutions, and successful 
wiretaps, for example, that were used to gauge an office's or agent's 
performance throughout the year. Officials said that numbers of arrests 
were reported from the field to headquarters twice a year as one way of 
showing what had been accomplished. The statistics were readily 
available, according to the officials, but special agents were not told 
that they did not have enough arrests or would need to increase the 
number of their arrests.

According to Customs Human Resources officials, investigation and 
arrest statistics were not used for special agent promotion purposes. 
The officials said that promotions up to the journeyman level were 
based primarily on the recommendation of first line supervisors. The 
journeyman level was to the GS-12 level through 2000, but was raised to 
the GS-13 level beginning in 2001.

Before 2000, promotions beyond the journeyman level were competitive 
and were based on applicants responding to a series of set questions 
regarding the type, complexity, and variety of investigations, not the 
quantity of investigations and arrests. The questions were weighted, a 
score was generated, and a roster of applicants eligible for promotion 
was developed. Beginning in 2000, Customs initiated the SA14 Promotion 
Test System. This system was for promotion to the GS 14 level and 
included three tests--critical thinking skills, job knowledge, and an 
assessment of administrative and planning skills. Once the applicants 
had passed these tests, they were further assessed through a structured 
interview, which assessed additional leadership skills via situational 
questions about how the applicants would handle the various situations. 
Customs officials said that promotion to GS-15 is based on a merit 
promotion system, which used knowledge, skills, and abilities that were 
developed for the specific position.

Also in fiscal years 1998 to 2001, Customs officials said that 
investigation and arrest statistics were used to some degree in award 
and bonus decisions, but so were other factors, such as successful 
court appearances and prosecutions.

[End of section]

Appendix VI: Immigration and Naturalization Service Profile:

The Immigration and Naturalization Service's (INS) primary mission was 
to administer and enforce the nation's immigration laws.[Footnote 15] 
Among other things, INS activities included:

* determining the admissibility of persons seeking to enter the United 
States through an inspections process,

* facilitating entry:

* processing and granting immigration-related benefits,

* patrolling the borders,

* deterring and investigating illegal employment,

* providing information to employers and benefit providers to prevent 
illicit employment or benefit receipt, and:

* disrupting and dismantling organizations engaging in document and 
benefit fraud and alien smuggling.

In addition, INS apprehended, detained, and removed aliens present in 
the United States without lawful status and/or those who have violated 
U.S. criminal laws. As individual aliens engaging in criminal activity 
and organizations facilitating illegal immigration are often associated 
with other criminal activity, INS also played a role in enforcing U.S. 
criminal laws. To perform its mission, in fiscal year 2002, INS had a 
total of 36,117 employees with a budget of $6.2 billion. The mission 
was accomplished through INS's operational offices located on the 
border, in the interior, and overseas and through numerous special 
facilities (e.g., detention centers, applications processing centers, 
and national records repository) throughout the United States.

Counting Investigations and Arrests:

INS's Investigations Division was the enforcement arm of the INS 
charged with investigating violation of the criminal and administrative 
provisions of the Immigration and Nationality Act and other related 
provisions of the United States Code. For INS, the investigative case 
process began with the receipt of a complaint or other lead by the 
Investigations Division that provided a "reason to believe" that a 
violation of law may have occurred. An investigation could have been 
opened as either a preliminary or a full field investigation. In either 
case, supervisory approval was required to initiate an investigation.

A preliminary investigation was opened when a lead or allegation was 
not sufficient enough to warrant a full investigation. In those 
instances, limited investigative activities would have been conducted 
solely for the purpose of providing enough additional information on 
which to make an informed judgment as to appropriate disposition of the 
matter at hand. Preliminary inquiries were ordinarily assigned for a 
period not to exceed 30 days. At the end of that period, a decision was 
to be made whether to:

* close the investigation without further action,

* extend the inquiry for no more than an additional 30 days, or:

* assign the matter for a full field investigation.

A full field investigation may have been opened on the basis of 
sufficient, articulable facts that were in existence at the time of 
initial review, developed during the conduct of a preliminary inquiry, 
or assigned as a headquarters-designated case.[Footnote 16] Full field 
investigations consisted of all investigative or enforcement activities 
necessary to bring an investigation to its logical conclusion.

Under INS's Investigations Case Management System (ICMS),[Footnote 17] 
a Form G-600 was prepared when an investigation was opened. The G-600 
was basically an index card used to track and document the progress or 
termination of investigations. Information about the investigation, 
such as the case number, date opened, agent assigned, etc., was 
initially recorded on the G-600. Additional information would have been 
added to the G-600 as the investigation progressed. First line 
supervisors maintained the G-600s for the investigations by agents in 
their units.

Each investigative unit (e.g., field office, port of entry, or border 
patrol office) prepared a detailed monthly report, called an 
Investigations Activity Report of Field Operations (G-23 Report). The 
G-23 Report was a record of the number of cases opened or completed, 
the number of hours worked, and the results of the investigations. The 
report was a matrix of rows and columns, with the columns showing the 
number of cases received, opened, completed, etc., and the rows showing 
the category of cases, for example, trafficking, criminals, or 
employers. Hard copies of the G-23 Reports were maintained at the unit 
level and required supervisory signature.

At the beginning of each month, the data from the G-23s for the 
previous month were keyed into the Performance Analysis System (PAS). 
Each office was to close out its monthly statistical reporting on the 
last working day of each month. They then had 8 working days to 
consolidate the unit workload counts into office level totals and key 
the data into PAS. After the eighth day, the PAS system was to be 
locked down and no further data entry would have been possible by the 
field offices. The data were strictly numbers of activities and did not 
identify individual investigations or agents.

As shown in figure 10, INS provided us with the number of 
investigations opened and closed for fiscal years 1998 through 2001.

Figure 10: Investigations Opened and Closed and Arrests--Fiscal Years 
1998-2001:

[See PDF for image]

[End of figure]

According to INS's manual, The Law of Arrest, Search, and Seizure for 
Immigration Officers, an arrest occurred when a reasonable person in 
the suspect's position would conclude that he or she was under arrest. 
An arrest did not depend solely on whether the officer had announced 
that the suspect was under arrest. An arrest was to be supported by 
probable cause to believe that the person had committed an offense 
against the United States. Probable cause is knowledge or trustworthy 
information of facts and circumstances that would lead a reasonable, 
prudent person to believe that an offense had been committed or was 
being committed by the person to be arrested.

An INS officer was authorized to make arrests for both administrative 
(civil) and criminal violations of the Immigration and Nationality Act. 
According to INS officials, a Form I-213, Record of Deportable Alien, 
was to be completed to record administrative arrests, which were the 
bulk of INS arrests. A Form G-166, Report of Investigation was to be 
completed when a criminal arrest was made. Supervisory review and 
signature at the bottom of the forms verified that the arrest occurred.

Arrests were counted and recorded into the PAS system in the same 
manner as the investigations. That is, at the end of each month, a 
manual count of the arrest forms would be made and then support staff 
would enter the data into PAS.

Officials told us that INS's Office of Internal Audits conducted 
reviews of investigation files. Each district office was to be reviewed 
about every 3 years. For each district, a representative number of 
investigations would be reviewed to determine whether INS policies and 
procedures were followed and that all investigative documentation was 
complete. While the purpose of these reviews was to ensure that INS 
procedures were followed, an INS official said that the review could be 
considered as verification that the documentation of the arrests were 
valid and proper.

Use of Investigation and Arrest Statistics:

INS officials told us that, after the data have been entered into PAS, 
INS's Operational Analysis Branch (OAB) printed out a Monthly 
Statistics Report, which was distributed to INS upper level management. 
OAB also printed a Workload Summary Report on a quarterly basis. 
According to INS officials, the reports were an accounting of the work 
INS performed and could have been used to assist in making resource 
allocations and staffing decisions. The officials said that the data 
could also have been reviewed to see if there were any trend indicators 
about shifts in criminal activity.

INS Human Resources officials said that INS employee performance 
evaluations were not based on investigation and arrest statistics. 
Rather, employees were evaluated on how well they performed their jobs. 
INS agents were evaluated annually, and their supervisor wrote up 
narratives about how well an agent was performing. There were no set 
job elements that had to be covered, and the supervisor determined what 
should be evaluated and how well the agent was performing.

Concerning promotions, the Human Resources officials said that 
promotions up to the journeyman level were based primarily on the 
recommendation of first line supervisors. Promotions beyond the 
journeyman level were competitive and based on a scored assessment, 
which covered the following four critical factors:

* Job experience: This factor was worth 30 percent and described the 
assignments the individual had had and other collateral duties.

* Decision making: This factor was worth 30 percent and tested the 
individual's decision-making process and problem solving abilities by 
asking a series of questions about hypothetical situations.

* In-basket job simulation: This factor was worth 20 percent and tested 
the individual's administrative skills in organizing work, setting 
priorities, delegating work, etc. Individuals were given 45 minutes to 
review a series of documents and then given 45 minutes to answer 50 
questions about how to deal with certain events on the basis of the 
documents they reviewed.

* Managerial writing: This factor was worth 20 percent and tested the 
individual's writing skills and knowledge of proper grammar syntax, 
paragraph structure, and report organization.

INS provided, via the Department of Justice, budget requests to 
Congress each fiscal year. We reviewed INS's budget requests for fiscal 
years 2003 and 2002 to determine the extent, if any, investigation and 
arrest statistics were used as justification for increase resources. 
Investigation and arrest statistics were not used, in either table or 
narrative forms, as a basis for justifying an increase in resources. 
Investigation and arrest activities were discussed, however, but in the 
very broadest terms, for example:

* "Although an eventual reduction in arrests is a primary indicator of 
illegal entry attempts (and therefore deterrence), other critical 
indicators include decrease in border related crime, decrease in 
recidivism, shifting of illegal activity to non-traditional points of 
entry and through non-traditional methods, increase smuggling fees, 
increase in property values and commercial and public development along 
the border, etc.":

* "INS will initiate high priority investigations, conduct asset 
seizures, and present individuals for prosecution for alien smuggling 
related violations to disrupt the means and methods that facilitate 
alien smuggling.":

* "As a result of INS efforts, many alien smugglers, fraud 
organizations, and facilitators were arrested and presented for 
prosecution; assets were seized; and aliens with a nexus to organized 
crime, violent gangs, drug trafficking gangs, or who have terrorist-
related affiliations, were apprehended.":

[End of section]

Appendix VII: U.S. Marshals Service Profile:

The U.S. Marshals Service's mission is to protect the federal courts 
and the judicial system, apprehend federal fugitives, and manage seized 
assets. Regarding federal fugitives, the Marshals Service's 
responsibilities are to:

* locate and arrest federal fugitives, including prison escapees, bail 
jumpers, and parole and probation violators;

* enforce bench warrants issued by federal judges and warrants issued 
at the request of other federal agencies; and:

* serve as the "booking agent" for suspects arrested for federal 
offenses.

To perform its mission, in fiscal year 2002, the Marshals Service had a 
total of 4,134 employees, of which about 2,700 were U.S. Marshals and 
Deputy U.S. Marshals. The Marshals Service's budget was $676.5 million 
in fiscal year 2002. The Director, Deputy Director, and 94 U.S. 
Marshals direct the activities of 95 district offices and personnel 
stationed at more than 350 locations throughout the 50 states, Guam, 
Northern Mariana Islands, Puerto Rico, and the Virgin Islands.

Counting Investigations and Arrests:

For the Marshals Service, an investigation consists of locating and 
arresting a federal fugitive. The Marshals Service initiates fugitive 
investigations in response to two basic scenarios. In the first, an 
individual has already been in the federal criminal justice system and 
has subsequently become a fugitive. The fugitive may have failed to 
make a court appearance, escaped from custody, or violated the terms of 
parole or supervised release. In each of these instances, the court 
issues a warrant and the Marshals Service is responsible for 
investigating, apprehending, and arresting the fugitive. In the second 
scenario, another law enforcement agency has investigated an 
individual, the individual has been indicted and a warrant issued, and 
the agency requests that the Marshals Service make the apprehension. 
Unlike other law enforcement agencies that investigate the commission 
of a crime, the Marshals Service investigations primarily consist of 
locating (tracking down) and arresting federal fugitives.

The Marshals Service uses its Warrant Information Network to track the 
number of fugitive warrants received and closed. The network is a 
computer-based automated system that manages records and information 
collected during investigations of fugitives. The system can also 
provide data for analyses that are used to report information to 
Congress or for management purposes, for example, to provide a listing 
of active warrants for a specific offense or for a district or 
suboffice. Figure 11 shows the number of warrants closed for fiscal 
years 1998-2001.

Figure 11: Warrants Closed by the Marshals Service and by Other 
Agencies--Fiscal Years 1998-2002:

[See PDF for image]

[End of figure]

The Marshals Service takes custody of all federal prisoners arrested by 
all federal officials empowered to make arrests. The Marshals Service 
Prisoner Tracking System (PTS) maintains a record of all suspects 
arrested for federal offenses and booked by the Marshals Service. The 
Marshals Service claims arrests on its workload statistics if a deputy 
marshal actually makes the arrest, based on a federal fugitive warrant. 
If another law enforcement agency brings a prisoner to the Marshals 
Service for booking, that agency would be recorded as the arresting 
agency in the PTS.

When either a deputy marshal or an agent from another federal agency, 
such as the DEA, FBI, or Customs Service, presents a federal prisoner 
for booking by the Marshals Service, the following procedures are 
followed:

* A Marshals Service Form 312 (Prisoner Personal History) is filled 
out, and the prisoner is fingerprinted and photographed. The form 
contains background information on the prisoner, the charges, case 
number, and which agency brought the prisoner in for booking.

* The agent (e.g., DEA) that fills out the Marshals Service Form 312 is 
called the lead agent and that agency will be credited with the arrest 
in the PTS. Only one agency is listed as the arresting agency even 
though many agencies may have participated in a joint operation through 
a task force. However, there is a space for indicating whether the case 
was a joint operation, but it is not mandatory to fill it out.

* The Form 312 information is entered into the PTS.

Use of Investigation and Arrest Statistics:

Marshals Service officials told us that investigation and arrest 
statistics are used as workload measures; for example, to show how many 
prisoners were produced for court appearances. With these statistics, 
they said that they could show workload projections to justify budget 
requests. The officials also said that investigation and arrest 
statistics are used to manage programs, set policies, and allocate 
funds and positions. For example, on the basis of an assessment of 
workload statistics, the number of positions at a particular courthouse 
was decreased in fiscal year 2002.

Marshals Service officials also said that investigation and arrest 
statistics are not used for making promotion, bonus, or award 
decisions. Criteria that are considered for promotions, for example, 
include time in grade, technical knowledge, analytical/problem solving 
ability, time management, and interpersonal relationships. For higher 
grades, management skills--including organization and planning, budget 
management, and human resource management--are also considered for 
promotions.

[End of section]

Appendix VIII: Drug Trafficking Investigation:

We found Operation Marquis on the Drug Enforcement Administration's 
(DEA) Web site when we searched for drug trafficking investigations 
involving multiple federal law enforcement agencies. The Web site 
indicated that Operation Marquis was coordinated by DEA's Special 
Operations Division (SOD)--a joint Department of Justice (Justice), 
DEA, the Federal Bureaus of Investigation (FBI), U.S. Customs Service 
(Customs), and the Internal Revenue Service program--and was conducted 
in 1999, 2000, and 2001. Attorneys from Justice's Criminal Division, 
and agents and analysts from participating law enforcement agencies 
staffed the investigation.

Operation Marquis targeted a Mexico-based drug trafficking organization 
responsible for putting tens of millions of dollars worth of cocaine 
and marijuana on the streets of at least a dozen U.S. cities. According 
to DEA's Web site, over 300 individuals were arrested as a result of 
the operation. In addition to arrests, the investigation resulted in 
the seizure of 8,645 kilograms of cocaine, 23,096 pounds of marijuana, 
50 pounds of methamphetamine, and $13 million in U.S. currency.

On DEA's Web site, several Justice and Customs officials commented on 
the success of Operation Marquis, including:

* "These law enforcement activities will have a measurable impact on 
drug trafficking across our Southwest Border. The work completed in 
this case emphasizes the importance of interagency cooperation in 
targeting and investigating drug trafficking organizations."--from an 
FBI assistant director.

* "This investigation demonstrates what can be achieved when law 
enforcement efforts are coordinated and resources are pooled. Operation 
Marquis shut down a sprawling criminal network that plagued communities 
throughout the country."--from the then-Acting Customs Commissioner.

We asked DEA and the FBI to provide us with the names of individuals 
that they had counted in their arrest statistics for Operation Marquis. 
Both agencies independently generated a list of individuals they 
counted as having been arrested as part of Operation Marquis. 
Specifically, DEA counted 331 arrests and the FBI counted 154 arrests. 
After comparing the lists, we were able to match eight names as having 
been counted as arrests by each agency.

We also asked DEA and FBI officials whether they had been designated as 
the lead or assist agency, how many special agents they had assigned 
throughout the investigation and at various times during the progress 
of the investigation, and whether their agents were physically present 
during the arrests. In addition, we asked if the amount of time their 
agents spent on the investigation could be determined.

DEA officials told us that the investigation was initiated through its 
SOD and that DEA was the lead agency. DEA told us that there were about 
46 lead criminal investigators assigned to Operation Marquis; however, 
because their statistical data systems do not record such information, 
DEA officials could not tell us whether DEA agents were present at the 
arrests counted by DEA. A DEA official, however, did tell us that 
agents logged 217,937 work hours on investigations that comprised 
Operation Marquis.

FBI Criminal Investigation Division officials also told us that 
Operation Marquis was a DEA-initiated investigation and that DEA was 
the lead agency. Seven FBI field offices were involved: San Antonio, 
Houston, Dallas, Memphis, Charlotte, Kansas City, and Little Rock. 
Because their statistical data systems do not record such information, 
the FBI officials said that there is no way to determine exactly how 
many agents participated in some manner with Operation Marquis. 
However, they said SOD investigations typically have two FBI staff 
coordinators assigned to an investigation. In addition, each field 
office typically assigns one lead agent, who works an investigation 
full-time, and possibly a co-case agent, who would work the 
investigation part-time. There also could have been several agents in 
any given field office working on or assisting with parts of the 
investigation on a part-time basis, for example, helping to setup or 
monitor a wiretap.

Also, because their statistical data systems do not record information 
by operation, FBI officials said that there is no way to tell how many 
agents participated in the physical arrest of individuals for Operation 
Marquis.. Usually, at least one or two FBI agents make or assist in an 
arrest, but the officials could not tell us the role played by their 
agents with regard to any of the individual arrests. Moreover, FBI 
officials could not tell us how many hours agents spent on the 
investigation because agents do not record their time by case number, 
but rather by the type of work performed, such as bank robbery:

We asked DEA and FBI officials whether, as a result of the numbers of 
arrests, special agents were given awards, promotions, bonuses, etc. 
DEA officials said that they did not know how many, if any, agents 
received awards, promotions, or bonuses for their work on Operation 
Marquis. The FBI Criminal Investigation Division officials said they 
knew of no awards, promotions, or bonuses given as a result of 
Operation Marquis.

[End of section]

Appendix IX: Child Pornography Investigation:

A U.S. Customs Service (Customs) Special Agent in Lake Charles, 
Louisiana, initiated Operation Bayou Blaster on October 1, 1994. The 
Customs agent developed a plan to set up an undercover operation that 
targeted individuals involved in the sexual exploitation of children 
via the Internet. Customs and U.S. Postal Inspection Service (USPIS) 
officials told us that a fake child pornography Web site was used to 
arrest individuals who ordered material from the Web site. USPIS was 
asked to participate by making some of the deliveries of the 
pornographic material, but not all.

All operational and undercover activity related to Operation Bayou 
Blaster ceased on February 23, 2001. Customs arrested 100 individuals 
through the efforts of more than 400 special agents who were involved 
at various times over the 6-year duration of the operation.

We asked Customs and USPIS to provide us with the names of individuals 
included in agency arrest statistics for Operation Bayou Blaster. 
Customs provided us with a list of the 100 individuals that were 
arrested, but USPIS was not able to generate a list of arrests that 
resulted from Operation Bayou Blaster. However, at our request, USPIS 
crosschecked its database with the 100 names provided by Customs. USPIS 
was able to match 30 arrestees, who had the same dates of birth, year 
arrested, and location arrested, in their database as the Customs list. 
USPIS officials noted that USPIS was asked to assist on only 30 of the 
deliveries.

We also asked Customs and USPIS officials whether they had been 
designated as the lead or assist agency, how many special agents they 
had assigned throughout the operation and at various times during the 
progress of the operation, whether their agents were physically present 
during the arrests, and their roles in the arrests. In addition, we 
asked if the amount of time their agents spent on the operation could 
be determined.

Customs officials said that Customs was the lead agency for the 
operation. They told us that of the over 400 agents involved through 
the more than 6 years of the operation, many of them participated in 
the arrests. However, because their statistical data systems do not 
capture such information, they were unable to tell us whether the 
agents were physically present at the arrests, or what exactly were 
their roles in the arrests. The officials provided a list showing that 
Customs agents had charged over 78,000 hours to this operation.

USPIS officials told us that Customs was the lead agency for the 
operation, and USPIS had no lead special agent for Operation Bayou 
Blaster. The officials said that Customs calls on USPIS all the time to 
assist in the delivery of pornographic materials. A USPIS official said 
20 inspectors were involved with the 30 arrests claimed by USPIS; 
however, because their statistical data systems do not capture such 
information, the official was unable to say whether the inspectors were 
physically present at the arrests or what role the inspectors played in 
the arrests. The official told us that inspectors logged over 1,700 
hours on Operation Bayou Blaster.

We asked Customs and USPIS officials whether, as a result of the 
arrests, special agents were given awards, promotions, bonuses, etc. As 
far as Customs officials knew, there were no awards, promotions, or 
bonuses given to the agents as a result of the Operation. The officials 
also said that with the large number of Customs agents involved in the 
Operation, it would be difficult to determine whether any awards, 
bonuses, or promotions were a direct result of the Operation. USPIS 
officials could not tell us if inspector participation in Operation 
Bayou Blaster resulted in any awards, bonuses, or promotions.

[End of section]

Appendix X: Counterterrorism Investigation:

We discussed a closed counterterrorism operation with Federal Bureau of 
Investigation (FBI) and Immigration and Naturalization Service (INS) 
officials. Between October 11, 2001, and April 17, 2002, the Joint 
Terrorism Task Force (JTTF) in New Orleans conducted this operation 
based on information that a telephone number associated with one 
subject in the United States had been contacted by a pay phone known to 
be used by the Taliban/al Qaeda in Afghanistan.

Shortly after the operation began, an INS agent assigned to the JTTF 
was asked to obtain INS files on six subjects to determine their 
immigration status in the United States. It was found that the six 
subjects had been released from INS custody while seeking asylum. As a 
result of reports of suspicious activity, the INS district director 
decided to revoke their paroles and, on January 12, 2002, the six 
individuals were taken back into custody while their asylum 
applications or appeals were pending. At the conclusion of their cases, 
four of the individuals were removed from the country, and, as of 
February 6, 2004, two were being held for removal.

We asked the FBI and INS to provide us with the names of individuals 
included in agency arrest statistics for this investigation. The FBI 
provided us with a list of the six individuals whose arrests were 
included in its statistics. INS provided us with a list of the six 
individuals, but told us that even though these six individuals were 
taken into custody, INS did not include them in agency arrest 
statistics because INS had previously arrested these individuals in 
1999; these arrests were counted for statistical purposes at that time. 
Comparison of the lists of names provided by the FBI and INS revealed 
that the six individuals arrested by INS in 1999 and taken back into 
custody in 2002, and the six individuals counted as arrests by the FBI 
in 2002, were the same six individuals.

While apprehending these individuals, FBI and INS agents encountered 
and arrested two other aliens whose immigration documents were no 
longer valid. INS counted these two subjects as arrests for, 
respectively, a nonimmigrant overstaying his or her visa and illegal 
entry into the United States.

We asked FBI and INS officials whether they had been designated as the 
lead or assist agency, how many special agents they had assigned 
throughout the operation and at various times during the progress of 
the operation, and whether their agents were physically present during 
the arrests. In addition, we asked if the amount of time their agents 
spent on the operation could be determined.

FBI Counterterrorism Division officials told us that both the FBI and 
INS initiated the operation, with the FBI taking the lead in the 
operation. FBI officials said two FBI agents were assigned full-time 
and numerous others helped with such matters as surveillance. The 
officials could not tell us how many hours agents spent on the 
operation, since agents do not record their time by operation, but 
rather by the type of work performed, such as bank robbery.

INS officials also told us that the FBI initiated the operation and 
that INS became involved when the FBI asked INS agents on the JTTF for 
their assistance. INS officials told us that no INS agents were 
specifically assigned to the operation; one of the two INS agents in 
the JTTF assisted in this operation. After the apprehensions on January 
12, 2002, the INS agent assisted the FBI in the continued operation--
checking the status of other aliens whose names appeared in records, 
reviewing INS files, assisting with interviews, etc. INS officials 
could not tell us how many hours agents spent on the operation because 
agents do not record their time by operation, but rather by the type of 
work performed.

FBI Counterterrorism Division officials said that both the FBI and INS 
participated in the physical arrest of the six individuals; however, 
they could not tell us the exact number of FBI agents present at the 
arrests. INS officials also told us the same thing; that both the FBI 
and INS participated in the apprehension of the six individuals. 
Specifically, seven INS agents, two INS supervisory special agents, and 
four INS deportation officers participated in the apprehension, 
according to the officials.

We asked FBI and INS officials whether, as a result of the arrests, 
special agents were given awards, promotions, bonuses, etc. FBI 
Counterterrorism Division officials said two agents received $500 
awards for their performance on this operation. INS officials told us 
no INS agents received awards for their participation in the operation.

[End of section]

Appendix XI: Comments from the U.S. Postal Inspection Service:

UNITED STATES POSTAL INSPECTION SERVICE:

Deputy Chief Inspector: 
Headquarters Operations:

March 5, 2004:

Mr. Tim Outlaw: 
Analyst-in-Charge: 
United States Government Accounting Office: 
441 G Street, NW - Room 6Q26 
Washington, DC 20548:

Dear Mr. Outlaw:

Thank you for providing us with a copy of your office's proposed report 
entitled Federal Law Enforcement. Information on Use of Investigation 
and Arrest Statistics (GAO-04-411). We appreciate the opportunity to 
have been a part of your work and agree with your overall findings that 
federal law enforcement agencies are generally consistent in the way 
they report and make use of investigation and arrest statistics.

We have reviewed the draft report in its entirety and offer the 
following comments:

Arrests for Federal Offenses, by Arresting Law Enforcement Agency - 
Figure 1:

While we recognize the statistics used in Figure 1 to portray the 
number of federal arrests between 1998 - 2002 were obtained through the 
Bureau of Justice Statistics (BJS), and you do state in the report that 
agencies" arrest totals are higher than the chart indicates, I am 
providing you with a more accurate accounting of federal arrests made 
by Postal Inspectors during that time period.

1998: 4,319 Federal Arrests.

1999: 4,530 Federal Arrests.

2000: 4,633 Federal Arrests.

2001: 4,698 Federal Arrests.

2002: 4,927 Federal Arrests.

Guidance and Procedures for Counting Investigations and Arrests Are 
Generally Consistent among Agencies - Page 5:

The first of the three bullet items taken from our Inspection Service 
Manual (ISM) regarding claimable arrests is inaccurate. That item 
should read, "an inspector participated personally in making an arrest 
or contributed significantly to an investigation resulting in an arrest 
made by another law enforcement agency", not "....made by another USPIS 
inspector". The draft report correctly states that item on page 31.

Appendix I: Scope and Methodology - Page 14:

In the middle of the first paragraph, we understand your report is 
merely stating the question asked of the Customs Service to identify a 
"joint investigation" involving a child pornography case, however, the 
impression the reader takes from this is that Operation Bayou Blaster 
was, in fact, a "joint investigation". If possible, we would ask that 
you insert language stating the Postal Inspection Service does not 
consider, nor did it treat, Operation Bayou Blaster as "joint 
investigation", thus the reason we were unable to provide you with a 
"list of arrestees". Postal Inspectors provided assistance, when 
requested by Customs, on a case by case basis. Your explanation of this 
so-called "joint investigation" later in the report (Appendix IX: Child 
Pornography Investigation, pages 53-54) more accurately describes the 
role of the Postal Inspection Service in this Customs Service 
investigation.

Again, it was our pleasure to work with your office on this project. If 
you have any questions or require further explanation, please contact 
Assistant Chief Inspector Lawrence E. Maxwell at (202) 268-5015.

Signed by: 

James Rowan, Jr. 
Deputy Chief Inspector 

Cc: ACI Maxwell:

[End of section]

Appendix XII: GAO Contacts and Staff Acknowledgments:

GAO Contacts:

Paul Jones (202) 512-8777 Darryl Dutton (213) 830-1086:

Staff Acknowledgments:

In addition to those named above, the following individuals contributed 
to this report: Tim Outlaw, Doris Page, Carolyn Ikeda, Alison Heafitz, 
Christine Davis, and Amy Bernstein.

FOOTNOTES

[1] P.L. 107-273 (Nov. 2, 2002), section 11026.

[2] On March 1, 2003, functions of several border and security 
agencies, including INS and Customs, were transferred into the 
Directorate of Border and Transportation Security within the Department 
of Homeland Security. As part of this transition, these agency 
functions were reorganized into the Bureau of Immigration and Customs 
Enforcement. Several primary program areas comprise Immigration and 
Customs Enforcement, including customs and immigration investigations. 
Because most of the investigation and arrest activity reviewed for this 
study occurred prior to Homeland Security's reorganization, for this 
report, we will refer to the former Customs and INS agency structures.

[3] Compendium of Federal Justice Statistics, U.S. Department of 
Justice, Office of Justice Programs, Bureau of Justice Statistics, 
November 2003, NCJ 201627.

[4] Public order offenses include offenses concerning tax law, perjury, 
and racketeering and extortion.

[5] Some agencies refer to investigations as cases. For this report, we 
will use investigations and cases interchangeably.

[6] On March 1, 2003, functions of several border and security 
agencies, including Customs and INS, were transferred into the 
Directorate of Border and Transportation Security within DHS. As part 
of this transition, these agency functions were reorganized into the 
Bureau of Immigration and Customs Enforcement. Several primary program 
areas comprise Immigration and Customs Enforcement, including customs 
and immigration investigations. Because most of the investigation and 
arrest activity reviewed for this study occurred prior to DHS's 
reorganization, for this report, we will refer to the former Customs 
and INS agency structures.

[7] P.L. 107-273 (Nov. 2, 2002), section 11026.

[8] USPIS said that it did not consider Operation Bayou Blaster a 
"joint" investigation, but provided assistance when requested by 
Customs. For this reason, USPIS said it was unable to provide us with a 
list of arrestees for this operation. 

[9] Disruption occurs when the normal effective operation of an 
identified organization is significantly impacted so that it is 
temporarily unable to conduct criminal operations for a significant 
period of time. The disruption must be the result of an affirmative law 
enforcement action, including, but not limited, to the arrest, 
indictment, and conviction of the organization's leadership or a 
substantial seizure of the organization's assets.

[10] Dismantlement occurs when an identified organization is 
eviscerated and no longer capable of operating as a coordinated 
criminal enterprise. The dismantlement must be the result of an 
affirmative law enforcement action, including, but not limited, to the 
arrest, indictment, and conviction of all or most of its principal 
leadership.

[11] FY 2001 Performance Report and FY 2002 Revised Final Performance 
Plan, FY 2003 Performance Plan; and FY 2002 Performance Report and FY 
2003 Revised Final Performance Plan, FY 2004 Performance Plan.

[12] FY 2001 Performance Report, FY 2002 Revised Final Performance 
Plan, FY 2003 Performance Plan; and FY 2002 Performance Report, FY 2003 
Revised Final Performance Plan, FY 2004 Performance Plan.

[13] On March 1, 2003, the functions of the U.S. Customs Service were 
transferred to the Department of Homeland Security. Because most of the 
investigation and arrest activity reviewed for this study occurred 
before the transfer, for this report, we will refer to the former 
Customs structure, policies, and procedures.

[14] An Immigration and Customs Enforcement official said that TECS 
would continue to be the case management system within the Department 
of Homeland Security for the foreseeable future, at least until 2005 
when a new system under development is scheduled for roll out.

[15] On March 1, 2003, the functions of the INS were transferred into 
the Department of Homeland Security. Because most of the investigation 
and arrest activity reviewed for this study occurred before the 
reorganization, for this report, we will refer to the former INS agency 
structure.

[16] INS also had a category for "Auxiliary Investigations," which were 
investigations done in one district's jurisdiction at the request of 
another district. An investigation conducted under this situation must 
have met the basic definition of an investigation, and a completed 
auxiliary investigation was given the same status and credit as a full 
field investigation.

[17] The ICMS was not a computer-based system; it was a hard copy, 
manual documentation system. The Criminal Investigations Reporting 
System was implemented in INS on October 1, 2002; it is a computer-
based system that replaced the ICMS for the initiating and tracking of 
cases.

GAO's Mission:

The General Accounting Office, the investigative arm of Congress, 
exists to support Congress in meeting its constitutional 
responsibilities and to help improve the performance and accountability 
of the federal government for the American people. GAO examines the use 
of public funds; evaluates federal programs and policies; and provides 
analyses, recommendations, and other assistance to help Congress make 
informed oversight, policy, and funding decisions. GAO's commitment to 
good government is reflected in its core values of accountability, 
integrity, and reliability.

Obtaining Copies of GAO Reports and Testimony:

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

Each day, GAO issues a list of newly released reports, testimony, and 
correspondence. GAO posts this list, known as "Today's Reports," on its 
Web site daily. The list contains links to the full-text document 
files. To have GAO e-mail this list to you every afternoon, go to 
www.gao.gov and select "Subscribe to e-mail alerts" under the "Order 
GAO Products" heading.

Order by Mail or Phone:

The first copy of each printed report is free. Additional copies are $2 
each. A check or money order should be made out to the Superintendent 
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 
more copies mailed to a single address are discounted 25 percent. 
Orders should be sent to:

U.S. General Accounting Office

441 G Street NW,

Room LM Washington,

D.C. 20548:

To order by Phone: 	

	Voice: (202) 512-6000:

	TDD: (202) 512-2537:

	Fax: (202) 512-6061:

To Report Fraud, Waste, and Abuse in Federal Programs:

Contact:

Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov

Automated answering system: (800) 424-5454 or (202) 512-7470:

Public Affairs:

Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800 U.S.

General Accounting Office, 441 G Street NW, Room 7149 Washington, D.C.

20548: