Lead
Development Centers Do Not Significantly Contribute to Increases in Legal
Source Cases
September
2003
Reference
Number: 2003-10-201
This report has cleared the Treasury
Inspector General for Tax Administration disclosure review process and
information determined to be restricted from public release has been redacted
from this document.
September
25, 2003
MEMORANDUM FOR CHIEF, CRIMINAL INVESTIGATION
FROM: Gordon C. Milbourn III /s/ Gordon C.
Milbourn III
Assistant Inspector General
for Audit (Small Business and
Corporate Programs)
SUBJECT: Final Audit Report - Lead Development
Centers Do Not Significantly Contribute to Increases in Legal Source Cases (Audit
# 200210039)
This
report presents the results of our review of the Criminal Investigation (CI)
Lead Development Centers (LDC). The overall objective was to
evaluate the operational readiness of the LDCs and the ability of the LDCs to
timely and effectively develop legal source income tax cases. This review was a follow-up to a prior
Treasury Inspector General for Tax Administration audit report that raised
concerns with the accelerated rollout of the LDCs and their ability to meet
their planned objectives to refocus resources to investigate legal source tax
crimes.
In
summary, we found that while the LDCs are fully operational, their workload has
consisted mainly of researching investigative leads provided by the CI field offices and
conducting limited research on the Compliance functions’ fraud referrals. The LDCs have not yet had a significant
effect on identifying new legal source income tax leads. In our opinion, the LDCs are also not adding
significant value to the fraud referral process due to the very limited
research procedures established for these referrals.
In
addition, LDC resources are not always effectively used. CI field offices did not clearly communicate
project objectives or provide sufficient information on allegations, LDC
research did not always adequately address the allegations, and the LDCs may
have conducted unnecessary research based on the CI field offices’ subsequent
closing actions and time expended.
Further, the LDC management information system is not adequate to
evaluate balanced measures or workload, and funding for a critical research
database remains uncertain.
We recommended that
the Chief, CI, include applicable General Investigations in the LDC inventory
and return the fraud referral process to the CI field offices. We also recommended the Chief, CI, establish
procedures to improve the research guidance for the LDCs, emphasize the need
for effective communication between the LDCs and CI field offices, and
implement a review process to evaluate the impact on LDC resources when CI
field offices close cases with little additional investigative time. In addition, the Chief, CI, should ensure
that data in the LDC database are consistent and consider redirecting any cost
savings from reducing the use of an existing research database to provide
continued funding for a more critical research database.
Management’s Response: The Chief, CI, generally agreed with the recommendations included in this report. The CI function recently completed an internal study focusing on the redesign and realignment of the LDC organizational structure. The redesign evaluation provides for a more efficient and effective LDC structure that will focus on the mission of case development and deliver more strategic use of LDC resources. The proposed redesign recommends that 5 of the existing 12 LDCs establish specific General Investigation responsibilities based on the CI function’s operational priorities. In addition, the Chief, CI, is considering implementing a standardized research protocol, and stated the LDCs will no longer be involved in the fraud referral process. However, the Chief, CI, stated that CI management does not have the decision-making authority to redirect any cost savings associated with renegotiating information systems contracts with vendors. In addition, the Chief, CI, disagreed with our proposed savings amounts presented in Appendix IV. Management’s complete response to the draft report is included as Appendix VI.
Office of Audit Comment: While CI management generally agreed with our recommendations, their corrective actions did not always fully address some of the issues we raised. Unless CI management addresses the inadequacies and missing data from the LDC database, we believe they will not be able to determine if the increase in the quality of the detailed allegation narrative results in a corresponding decrease in LDC research time. The continued use of this database until implementation of the CI function’s management information system upgrade will hinder CI management’s ability to evaluate the workload and effectiveness of the LDCs. In addition, we believe the CI field offices should evaluate those additional factors that affect the ability to fully investigate the lead before it is sent to the LDC for research. We believe it is still necessary for the CI function to implement a review process on cases closed with little or no time charged to assist in determining how effectively the CI function is using LDC resources. Concerning the funding for research services, we understand that CI management’s ability to direct savings from one contract to another is limited. However, we believe that any savings from the LexisNexis contract provide an opportunity for the CI function to present a business case to obtain corporate funding for ChoicePoint.
CI
management did not agree with our outcome measures. CI management responded that the removal of the fraud referral
program from the LDCs to the field offices would result only in a shifting of
the staff days. While we agree with
this statement, our outcome measure, entitled “Inefficient Use of Resources,”
quantifies the benefits of making specific programs more efficient, most
notably by reassigning personnel to other programs or restructuring duties
within a program. Removing the fraud
referral process from the LDCs will allow the LDCs to redirect these resources
to better accomplish one of its goals – identifying legal source income tax
cases.
In addition, CI management maintained that we present an erroneous
claim that a Primary Investigation could not be effectively evaluated within 8
hours of receipt of the completed LDC research package. However, we believe that the manner in which
the field offices closed these cases, with little or no time charged, warrants
closer, periodic program oversight to ensure that LDC resources are being
effectively used. In some instances,
the closing action should have been taken by the field office without involving
the LDC. In other instances, we believe
that it would be difficult for the field office, based on the closing action, to
have effectively evaluated the information provided by the LDC and to have made
an investigative decision with little or no time charged. We realize that there are many variables
that could affect actual resource savings.
Our intent was to show that an analysis of the subsequent time spent by
the field offices, and type of closing action taken, could serve as an
indication of how effectively the CI function uses LDC resources.
Therefore, we will not be
adjusting the potential outcome measures in this report. Where appropriate, we have included in
the report our comments related to management’s response.
Copies
of this report are also being sent to the Internal Revenue Service managers who
are affected by the report recommendations.
Please contact me at (202) 622-6510 if you have questions or Daniel R.
Devlin, Assistant Inspector General for Audit (Headquarters Operations and
Exempt Organizations Programs), at (202) 622-8500.
The Lead Development Centers Are Operational
Lead Development Centers Are Not Using Resources Effectively
Funding for the ChoicePoint Database Remains Uncertain
Appendix I – Detailed Objective, Scope, and Methodology
Appendix II – Major Contributors to This Report
Appendix III – Report Distribution List
Appendix IV – Outcome Measures
Appendix V – Description of Research Tools
Appendix VI – Management’s Response to the Draft Report
The mission of the
Criminal Investigation (CI) function is to serve the American public by
investigating potential criminal violations of the Internal Revenue Code
(I.R.C.) and related financial crimes in a manner that fosters confidence in
the tax system and compliance with the law.
Over the last 20 years, the Congress and the
Department of the Treasury have expanded the CI function’s jurisdiction to
cover offenses not only under the I.R.C. but also under money laundering and
currency reporting statutes. This
resulted in the CI function playing a major role in investigations with no obvious
direct connection with tax compliance.
In July 1998, the Commissioner of the Internal Revenue Service (IRS) appointed Judge William Webster to perform an independent review of the CI function’s operations. The Webster Report confirmed that the CI function’s focus had drifted from its primary mission of investigating potential violations of the Federal internal revenue laws concerning legal sources of income.
The Webster Report also concluded that the Compliance and CI functions needed to reinvigorate the fraud referral program by identifying legal source tax violations. The Webster Report subsequently made recommendations to strengthen and support the CI function in accomplishing its primary mission. In response, the CI function established the Lead Development Centers (LDC) to address the Webster Report’s recommendation to increase legal source cases and reinvigorate the fraud referral program.
The LDCs were designed
to act as a consolidated research center for the field offices and serve as the
focal point of case development activities for legal source income cases
developed from investigative leads received from a variety of sources. The LDCs receive the leads, conduct the
necessary research from a variety of databases, and prepare a written analysis
and summary of the data that is sent to the CI field offices. The goal is to help free up special agents’
time by having investigative analysts research and analyze databases to fully
develop leads and fraud referrals more efficiently.
We conducted our audit between November 2002 and July 2003 at the CI Office of Strategy in Washington, D.C., and the Atlanta, Baltimore, and Philadelphia LDCs. The audit was conducted in accordance with Government Auditing Standards. Detailed information on our audit objective, scope, and methodology is presented in Appendix I. Major contributors to the report are listed in Appendix II.
The CI function began implementing the LDC concept in
October 2000, with the initial goal of conducting a 1-year pilot to determine
the optimal organizational structure, site location, automated data processing
needs, procedures and protocols, and workload requirements. However in December 2000, the Chief, CI, decided to accelerate the
rollout of the remaining 11 LDCs prior to the completion of the pilot
study. The last LDC became fully
functional in June 2002. We based our
results on work the LDCs completed during Fiscal Year (FY) 2002.
The Treasury Inspector
General for Tax Administration (TIGTA) issued an audit report in
February 2002 that raised concerns with the rollout of the LDCs. We cautioned that the CI function may be at
risk of not meeting its planned objective since the necessary research tools
were not in place and staffing levels were not validated prior to nationwide
implementation.
The exact cost to operate the LDCs is unknown; however, the CI function estimates that it costs $7.2 million annually. The 12 LDCs had an authorized staff of 132 as of September 30, 2002, and received 4,027 leads during FY 2002. The LDCs sent 3,258 of the 4,027 leads to CI field offices for further investigation.
Although the LDCs are operational, they have not yet
achieved their primary mission of increasing legal source cases or
reinvigorating the fraud referral program. We believe resources are not being effectively used to increase
the LDCs’ contribution to legal source cases.
Also, as discussed in the previous TIGTA report, funding issues with one of the external databases still exist, and CI
management has not yet validated the staffing levels, due in part to
inadequacies in the LDCs’ management information system.
In addition to the specific items cited later in this report, CI management advised us of some additional factors that may have contributed to the LDCs’ inability to increase legal source cases. For example, many of the LDCs became operational during the first half of FY 2002 and were overloaded with research requests from the CI field offices. In addition, LDC resources were affected when they subsequently joined forces with the CI field offices in the fight against terrorism.
The CI function’s intent was that the LDCs would eventually conduct independent research projects to develop legal source leads. Although legal source Primary Investigation cases increased from 3,386 in FY 2000 to 3,565 in FY 2002, an increase of about 5 percent, we could not confirm that this was due to the operations of the LDCs.
The LDCs’ current role is predominantly to research investigative leads that are provided by the CI field offices and fraud referrals that come from the Compliance functions. These are the same sources that existed prior to the creation of LDCs and are external to the LDCs; thus, the LDCs can only respond to incoming referrals and cannot influence the growth of legal source cases. In our opinion, the LDCs could have a greater impact on increasing legal source cases if they became more involved in working general research projects. Also, the role of the LDCs in regard to the fraud referral program should be reevaluated to help redirect resources to identify legal source cases.
The LDCs have an opportunity to be more proactive in
identifying leads
The LDCs are beginning to conduct some work on national research projects such as offshore trusts and foreign bank credit cards, which will help increase legal source cases. However, the LDCs can become more involved with general research projects.
We analyzed the Criminal Investigation Management Information System (CIMIS) as of September 30, 2002, and identified 251 open projects in the 9 CI field offices covered by the 3 LDCs we visited. Although many of these were administrative-type projects, we identified 39 projects that, based on the description in the allegation field, could be conducive to LDC research.
For example, one field office opened a project in March 2002 to identify potential underreported income tax cases. The type of information being gathered for this project is financial in nature and includes data from various databases. Although this is the type of work that the LDCs were designed to conduct, the field office did not use LDC resources to perform the research. Instead the field office opened the project and, as of September 30, 2002, had not charged any time to it. By not involving the LDC, the CI function may have missed an opportunity to timely work the project and identify new legal source cases.
We believe using LDC resources on these types of general research projects is efficient and will free up the special agents to pursue active criminal cases. Also, by having the LDCs perform this research, the CI function could improve the timeliness of working these projects and increase the number of legal source cases.
The LDCs are not
adding significant value to fraud referrals
In response to the Webster Report and to address the declining number of fraud referrals, both the CI and Compliance functions took steps to reinvigorate the fraud referral program. These included:
·
Establishing 64
fraud referral specialist positions in the Compliance functions.
Fraud referrals represent
about 9 percent of the Primary Investigations numbered during FY 2002. Table 1 shows some comparable
statistics.
Table 1: Analysis of Fraud Referrals
|
|
FY 2000 |
FY 2002 |
Increase
(Decrease) |
Percent Change |
|---|---|---|---|---|
|
Fraud Referrals Received |
436 |
526 |
90 |
20.6% |
|
Referrals Accepted |
229 |
326 |
97 |
42.4% |
|
Referrals
Rejected |
222 |
192 |
(30) |
(13.5%) |
|
% Accepted |
50.8 |
62.9 |
12.1 |
23.8% |
Source: TIGTA analysis of data from CIMIS Report 11 and the CI function’s Business Performance Review.
Although the number of fraud referrals accepted for investigation has increased since the LDCs became operational, the LDCs’ role was not, in our opinion, a significant factor in this increase. As currently operating, the LDCs do not develop the referrals in any significant way. They are essentially serving as a pass-through entity for fraud referrals.
One reason the LDCs contribute so little to developing fraud referrals is the limited information they receive on the Referral Report of Potential Criminal Fraud Cases (Form 2797). These forms contain only a summary of information about the allegation. When submitting the Forms 2797 to the LDCs, the Compliance functions do not include detailed supporting workpapers.
Also, on fraud referrals, the LDCs conduct only limited database research for criminal history checks and suspicious financial transaction activity. The LDCs limit their research in this manner pursuant to an understanding between the CI and Compliance functions that the originators would research available databases before sending referrals to the LDCs. However, the Compliance functions do not have access to the criminal history and suspicious financial transaction systems.
Verbal instructions from a Director of Field Operations to not reject any fraud referrals, regardless of their merits, contributed to the LDCs’ minimal involvement in the fraud referral process. The LDCs rejected only 10 of 436 referrals (2.3 percent) processed in FY 2002 and did not reject any of the 125 processed referrals that were received between July and September 2002. These results reaffirm our opinion that the LDCs are simply serving as a pass-through entity and are not developing fraud leads as expected.
Although the LDCs are not adding significant value to fraud referrals, their involvement consumes time and resources. The LDCs processed 436 referrals during FY 2002, taking an average of 6.9 workdays to process each referral, even though their goal was to process referrals within 2 workdays. The LDCs expended an average of 14 staff hours on each referral.
Considering the limited value being added by the LDCs, we believe it would be more efficient for the CI field offices to process fraud referrals. Prior to the establishment of the LDCs, fraud referrals were sent directly to the CI field offices. Currently, many of the field offices have a special agent that serves as a fraud referral coordinator. This coordinator is a liaison with the Compliance functions, providing training and giving presentations. CI field offices also have compliance assistants who conduct research and have access to research databases. Based on the current volume of referrals, we believe the CI field offices have the capacity to absorb this additional workload. Working fraud referrals in the field offices would eliminate the delays and inefficiencies in the LDCs and allow the fraud referral coordinators to more effectively meet with the Compliance referral specialists and review workpapers, if necessary.
By removing the LDCs from the fraud referral process, the CI function would more effectively use its LDC resources to perform general research projects to identify additional legal source cases. We estimate that at least 676 staff days could be redirected to help identify legal source cases.
1.
The Chief, CI, should consider increasing the use of LDC
resources on general research projects by:
·
Establishing
procedures to ensure that, before a CI field office approves a general research
project, it determines if the LDC could contribute to its development.
· Reviewing the current inventories of general research projects to determine if the LDCs could assist in any future research.
Management’s Response: The Chief, CI, agreed with the recommendation and stated that the proposed redesign of the LDCs recommends establishing specific general investigation responsibilities based on CI priorities outlined in the Strategy and Program Plan and the Annual Business Plan.
2.
The
Chief, CI, should reevaluate the role of the LDCs in the fraud referral process,
consider removing the fraud referral process from the LDCs, and have:
·
The fraud referral
coordinators in the CI field offices receive, number, and assign fraud
referrals.
· The compliance assistants in the field offices conduct the limited law enforcement research now done by the LDCs.
Management’s Response: The Chief, CI, agreed with the recommendation and stated that the CI function’s proposed redesign of the LDCs recommends that the LDCs no longer be involved in the fraud referral process. Fraud referrals will be made directly from the fraud referral specialists to the field office. However, the Chief, CI, disagreed with the entire amount of proposed savings, contending that there is only a shifting of the staff days from the LDCs to the field offices.
Office of Audit Comment: We agree that there is only a shifting of resources with respect to moving the fraud referral program from the LDCs. However, our outcome measure, entitled “Inefficient Use of Resources,” quantifies the benefits of making specific programs more efficient, most notably by reassigning personnel to other programs or restructuring duties within a program. In the case of the LDCs, processing fraud referrals, and in some cases duplicating research performed by the Compliance functions, resulted in the inefficient use of LDC resources. Removing the fraud referral process from the LDCs will allow the LDCs to redirect these resources to better accomplish one of its goals – identifying legal source income tax cases.
To increase legal source cases, the CI function needs to make more effective use of LDC resources by improving procedures and communications between the LDCs and the CI field offices. Specifically, we determined that:
· CI offices did not clearly communicate project objectives or provide sufficient information on the allegations.
· The LDCs performed research that did not always adequately address the allegations or was not used by CI field offices.
CI offices did not clearly communicate project objectives or provide sufficient information on allegations
In 2 projects involving a total of 14 leads, CI management did not adequately communicate the scope, criteria, and objectives of the project or provide sufficient information on research already conducted, to enable the LDCs to effectively develop the leads further. For one project, the project coordinator did not provide the LDCs with information on how the leads were identified or what to look for, or with detailed guidance on how to work the leads. As a result, some LDCs closed the investigations without conducting any research. For the second project, CI staff reviewed suspicious financial transaction activities of taxpayers but did not provide the LDCs with the research results; the LDCs then researched the same databases.
We also determined that in 18 of 42 leads submitted by the CI field offices to the LDCs, the field offices did not provide sufficient information as to the nature of the allegations. Either the leads were vague or the special agent did not include any supporting information, such as the memorandum of an informant’s communication, a copy of a newspaper article, or a referral from another agency. The following two examples illustrate the limited information provided to the LDCs:
In both examples, the allegations could have been improved if the agents included such information as a description of the subject’s lifestyle (e.g., the type of property owned or income level). In the second example, the agent could have provided the name and address of the business and the estimated amount of unreported income.
The LDCs performed research that did not always adequately address the allegation or was not used by CI field offices
Most investigative leads originate in the CI field offices from a variety of sources. A special agent prepares the lead with a description of the allegation. A field manager is responsible for discussing the merits of the investigation with the agent and should consider factors such as the adequacy of the allegation, reliability of the source, prosecution criterion, availability of resources, jury appeal, and ability to establish willful intent and responsibility. When other Federal, state, or local law enforcement agencies are involved, the CI field office should also coordinate with appropriate authorities to determine which agency will prosecute the case. After review, the field manager sends the lead to the LDC for numbering and further research.
Overall, the LDCs appeared to adequately research the allegations. The LDCs conducted research of the available data sources and provided the results to the field. However, in 4 of the 24 leads with an adequate description of the allegation, the LDC research did not fully pursue the allegation in a logical manner. For example, in one lead, a public source alleged that the subject was involved in a kickback scheme and failed to withhold employees’ taxes from paychecks. The LDC spent 52 hours on this lead but did not pursue such fundamental concerns as whether the subject had filed the proper employment tax return(s) to report the withheld taxes.
In 2 of the 42 leads submitted by the CI field offices, the LDC analysts unnecessarily expended time researching the lead after initial research demonstrated that the allegation lacked merit. For example, in one lead it was alleged that the taxpayer filed a form to improperly exempt wages from Federal withholding tax. LDC research of tax account information determined that the taxpayer had filed the proper returns and satisfied all tax obligations. Since this research proved the allegation was false, additional research was not necessary. However, the LDC spent 19 hours on this lead conducting unnecessary research (e.g., property searches, motor vehicle records, and currency transactions).
Additionally, we determined that in some cases the LDCs are performing research that the CI field offices do not use. We identified 10 Primary Investigations where the CI field offices’ closing actions apparently discounted the LDCs’ research results. For example, an LDC spent 123 hours researching an allegation that a subject and his or her sons spent over $1 million on property. Although the LDC identified numerous property transactions, the field office closed the case due to “lack of pattern,” after charging only 3 additional hours to the investigation.
Further, in looking for similar characteristics, our analysis of the CIMIS showed that, during FY 2002, the CI field offices closed 121 Primary Investigations without opening a more intensive criminal investigation, expending no more than 8 hours of additional investigative time on these cases. The closing codes used by the field offices on these 121 cases also indicated that they might not have considered factors such as the availability of resources or the ability to establish responsibility prior to submitting these leads to the LDCs. In our opinion, it would be difficult for the field offices to make an investigative decision on the case with less than 8 hours charged, or the decision should have been made before the CI field office referred the lead to the LDC. The LDCs spent about 252 staff days researching these cases, which the field offices then closed with minimal additional time and apparently without properly considering either their ability to fully investigate the cases or the information provided by the LDCs.
Our limited discussions with various levels of CI field office management revealed a mixed opinion of the relative value that LDCs provide to the investigative process. Some managers believe the LDCs are contributing to the investigative process and are improving; other managers believe the LDCs are not significantly contributing to the quality and timeliness of investigations.
The LDC concept is continuing to evolve, and prior work processes may have contributed to ineffectively using LDC resources and not having clear communication among the various CI offices. Prior to establishment of the LDCs, the field office special agents and managers discussed the merits of the leads and controlled the research on the leads, so there was little need to prepare a detailed explanation of the allegation. However, without the benefit of this interaction, the LDCs generally will require more information to effectively research the lead and provide sufficient information to prove or refute the allegation.
The LDCs generally conduct the same research on each lead regardless of the source. CI management did not disseminate procedures regarding research guidance or consider that different types of leads may require different types of research, depending on what information is available or the results of previous research. We are also not aware of any CI management program reviews that analyze the results of the LDC research or the CI field offices’ subsequent actions on Primary Investigations to determine whether the leads had merit to begin with and if the LDCs’ resources were effectively used.
3.
The Chief, CI, should issue procedures for the CI field
offices to follow when submitting leads or research projects to the LDCs. These procedures should emphasize the need
for effective communication between the offices and the LDCs and require a
detailed description of the allegation to allow the LDCs to focus their
research. The offices should include
supporting documentation, as necessary.
Management’s Response: The Chief, CI, agreed with this recommendation but did not provide specific corrective action. The Chief, CI, in a narrative following his response, stated that, in theory, if the quality of the detailed description of the allegation is increased, there should be a corresponding decrease in the amount of LDC research time. The decrease in applied time will be tracked through the existing LDC management information system database and eventually through the proposed upgraded CIMIS.
Office of Audit Comment: Management’s response indicates that they will initially use the existing LDC database to track the time applied by the LDCs. However, as reported on page 16 of this report and in Recommendation 7, the LDC management information system is not adequate to evaluate balanced measures or workload. The Chief, CI’s, response does not provide for any interim actions to correct the inadequate and missing data from this database. The CI function should be mindful of the need to remedy the problems with its database before it can begin to accurately track the time applied by the LDCs.
4.
The Chief,
CI, should issue procedures for the LDC analysts to follow when researching
cases. These procedures should provide
for an incremental approach to the research to address the allegation and
provisions to contact the referring agent if additional questions arise or
research is needed.
Management’s Response: The Chief, CI, agreed with this recommendation and, as part of
the LDC redesign, is considering a standardized research protocol for the
redesigned LDC operations to allow flexibility in expanding the research scope
when needed.
5.
The Chief, CI, should issue guidance to reemphasize the
need for field offices to evaluate additional factors that affect the ability
to fully investigate the lead (e.g., resources, jury appeal, prosecution by
another agency, etc.) before the lead is sent to the LDC.
Management’s Response: The Chief, CI, agreed with this recommendation and stated that, in theory, if the Supervisory Special Agents (SSA) in the field offices conduct the required screening on research requests to the LDCs, then the number of Primary Investigations closed using the closing codes we mentioned should decrease. Further, the LDCs should also realize a decrease in the number of research requests as the SSAs apply those additional evaluation factors we identified.
Office of Audit Comment: CI management’s response does not address the
recommendation. While we agree with the
Chief, CI’s, theoretical statement, we believe CI management still needs to
reemphasize this guidance to the field offices.
6.
The Chief,
CI, should consider implementing, as part of a program review, a process to
analyze cases closed by the field offices with little or no time and the impact
on LDC resources.
Management’s Response: The Chief, CI, stated that the CI function’s proposed LDC redesign recommends that LDCs no longer be involved in the initiation of Primary Investigations. Their report recommends that the authority to initiate Primary Investigations be delegated back to the CI field offices based on the allegation.
In addition, the Chief, CI, did not agree with the total amount of proposed savings. The CI function doubts that the field office Special Agents in Charge could effectively and efficiently evaluate a Primary Investigation to closure without the LDC research. The CI function also maintained that we presented an erroneous claim that a Primary Investigation could not be effectively evaluated within 8 hours of receipt of the completed research package.
Office of Audit Comment: CI management’s
response does not address the recommendation.
Although the LDCs will not initiate the Primary Investigation, they will
still conduct research on behalf of the field offices. Therefore, there is still a need for the CI
function to implement a review process on cases closed with little or no time
charged.
As to the proposed savings
associated with the potential inefficient use of LDC resources, we believe that
the manner in which the field offices closed these cases, with little or no
time charged, warrants closer, periodic program oversight to ensure more
effective use of LDC resources. In some
instances, the closing action should have been taken by the field office
without involving the LDC. In other
instances, we believe that it would be difficult for the field office, based on
the closing action, to have effectively evaluated the information provided by
the LDC and to have made an investigative decision with little or no time
charged. We realize that there are many
variables that could affect actual resource savings. Our intent was to show that an analysis of the subsequent time
spent by the field offices, along with the type of closing action taken, could
serve as an indication of how effectively LDC resources are used.
In response to the prior TIGTA report, CI management stated
that they would conduct a study in FY 2002 to determine the proper staffing
level and resource allocation. However,
CI management later realized the need to have a uniform database and thus
established 1 database structure for all 12 LDCs as of October 2002. CI management stated they would begin their
staffing and workload analysis after obtaining 6 months of data from the uniform
database.
CI
management provided the LDC managers with a demonstration of the uniform
database. However, they did not issue
any instructions on its use. In
addition, there were no reviews conducted to ensure that the uniform database
would capture data that were consistent and necessary for performing any
meaningful program analyses.
In February 2003, the CI function established business results quantity measures as part of its balanced measures. One measure records the average hours applied each month for each Primary Investigation on which research was completed during the month. Information relating to this performance measure is obtained from the LDC database.
Based on our observations and analysis, the CI function will be unable to effectively use the information in this database to measure its performance, establish a staffing pattern, or make other management decisions because the LDCs did not uniformly record data and time spent working leads, conducting research projects, or supporting criminal case development.
For instance, the LDC database contains a field to record
the source of the investigation and a field to record the number of staff hours
the LDC takes to process a lead. Our
analysis of the source field for FY 2002 showed that the 12 LDCs used 287
different referral sources. One LDC
used the name of the special agent.
This LDC recorded 115 different sources, while another LDC consolidated
the sources down to 5. We also analyzed
the available records from the consolidated LDC database for the first quarter
of FY 2003 and determined that, while the LDCs had made some improvements, they
were still inconsistently recording the source of the leads. A factor contributing to the inconsistencies
is that the source field is not restricted to allow only the entry of
predefined information.
Additionally, the LDCs are not consistently recording the
time spent working leads or research projects.
During FY 2002, the LDCs did not record the time spent on 38 percent of
the Primary Investigations. Further, we
determined through our analysis of LDC analysts’ time reports for Calendar Year
2002 that the analysts spent about 1,650 hours on research projects and 190
hours in support of criminal case development; this time was also not included
in the LDC database.
An effective management information system is necessary to
measure program results, evaluate workload, and make management decisions. The data recorded in this system should be
accurate and consistent. Without
accurate and consistent information, CI management may not be able to determine
if they are on target to meet program goals.
The CI function is in the process of redesigning the CIMIS
to include the time the LDC analysts spend on leads. This should ensure the recording and availability of more
consistent and reliable data. The CI
function expects to complete this redesign in January 2004; however, funding
and other delays could make it later.
In the meantime, the CI function needs to ensure that the LDC database
is accurate to properly staff, monitor, and evaluate the LDCs.
7. The Chief, CI, should ensure that the data in the LDC database are consistent and issue instructions on how to use the LDC database. This should include restricting the sources to predefined codes and recording all time spent on leads, research projects, and in support of criminal case development. These improvements should also be considered in the CIMIS redesign, as applicable.
Management’s Response: The Chief, CI, agreed with our recommendation that the LDCs should have an adequate management information system. The expectation is that the LDCs will continue to use the current database until the full implementation of the CIMIS III scheduled for October 2004.
Office of Audit Comment: Although management agreed with our recommendation, the response does not provide for any interim actions pending the implementation of CIMIS III in October 2004. As mentioned in response to Recommendation 3, management plans to use the existing LDC database to monitor the time the LDCs apply to developing research packages. However, without any further interim actions, the existing LDC management information system will not provide complete or accurate data.
The prior TIGTA report stated that the LDCs do not have sufficient research tools to accomplish the overall objective of providing the highest quality leads for legal source tax investigations. The report stated that if the LDCs are to function as envisioned, the CI function would need access to the Integrated Data and Retrieval System (IDRS) and the Electronic Fraud Detection System (EFDS) through the CI network, faster Internet connectivity to enhance research capabilities, and access to the law enforcement-enhanced ChoicePoint commercial database.
The CI function has resolved the issues concerning access to the IDRS, EFDS, and Internet. However, continued funding issues remain with the purchase of ChoicePoint. Because the IRS did not fund ChoicePoint, the CI function funded ChoicePoint in FY 2002 with its own operating funds and then used $1.4 million from the Treasury Enforcement Office of Asset Forfeiture (TEOAF) Fund to fund ChoicePoint during FY 2003.
The CI function also uses LexisNexis to conduct research, which cost about $1.4 million for the first 8 months of FY 2003. This research service provides data similar to that provided by ChoicePoint. The CI function conducted a study and determined that when the contract for LexisNexis expires, it will reduce the number of users from 4,000 to about 300 and use ChoicePoint as its primary research tool.
CI management is concerned about the pending loss of TEOAF funds and the inability to obtain funding from the IRS due to the process of prioritizing and funding the many information technology needs. The CI function could redirect the savings realized from renegotiating the LexisNexis contract to fund the ChoicePoint database.
8.
The Chief, CI, should request that any cost savings generated
from renegotiating the LexisNexis contract be provided to continue funding for
ChoicePoint.
Management’s Response: The CI function is limited in its ability to direct the application of savings from negotiation of contracts with vendors such as LexisNexis. The control and authority for the application of savings rests within the corporate IRS. Therefore, the CI function cannot perform any corrective action for this recommendation.
Office of Audit Comment: We understand that the CI function’s ability to direct savings from one contract to another is limited. However, any savings from the LexisNexis contract provide an opportunity for the CI function to present a business case to obtain corporate funding for ChoicePoint.
Appendix I
Our overall objective
was to evaluate the operational readiness of the Criminal Investigation (CI)
Lead Development Centers (LDC) and the ability of the LDCs to timely and
effectively develop legal source income tax cases. To accomplish our objective, we analyzed the Criminal
Investigation Management Information System (CIMIS) and the LDC data as of
September 30, 2002. We validated the
data received from the CIMIS by comparing the results of various queries to the
CI function’s management information reports.
However, we did not verify the accuracy of the data entered in the
system to original source documents. We
selected the Atlanta, Baltimore, and Philadelphia LDCs based upon a combination
of the following factors: the
operational date, proximity of the field office function and Small
Business/Self-Employed Division fraud referral specialists, and location of the
tax fraud hotline call site function.
We also considered the LDCs’ workloads (i.e., the number of cases opened,
number of cases in inventory, and average days in inventory). To accomplish our objective, we:
I.
Determined if the
LDCs are properly staffed and equipped to accomplish their mission.
A.
Obtained the numbers of authorized staffing and
on-rolls staff at each LDC as of September 30, 2002.
B.
Reviewed the CI
function’s staffing pattern assumptions regarding the number of leads to be
worked and the time to work each lead and determined if they had been
validated.
C.
Analyzed the
Atlanta, Baltimore, and Philadelphia investigative analysts’ time reports for
Calendar Year 2002 to determine how they used their time.
II.
Determined the
effectiveness and economy of the databases used by the LDCs to develop
investigative leads.
A. Identified the different research database sources used by the CI function and the various LDCs and determined whether “best practices” were established and uniformly used.
B. Determined whether access and security issues identified in the prior audit report regarding the Integrated Data Retrieval System (IDRS), Electronic Fraud Detection System (EFDS), and Internet had been resolved.
C. Judgmentally selected 58 of 168 Primary Investigations (PI) initiated during June and July 2002 from the Baltimore and Philadelphia LDC inventories. Also, we selected 14 of 80 investigative PIs that were initiated from the Atlanta LDC during Fiscal Year (FY) 2002 and were subsequently closed by the CI field office function by the time of our visit. We used judgmental sampling techniques due to the small universe and because we did not plan to project our results.
D. Reviewed 56 of the 72 leads selected and determined the adequacy of the allegation, whether the research addressed the allegation, and the subsequent actions taken by the CI field office. We eliminated 16 fraud referral leads from this analysis because the LDC performed either limited or no research.
E.
Compared
ChoicePoint and LexisNexis features to other available resources to identify
the cost and potential duplication of information.
III.
Determined if the
source and mix of work coming into the LDCs can or will provide for an increase
in the legal source tax cases sent to the field.
A.
Interviewed the LDC
and the CI field personnel at the Atlanta, Baltimore, and Philadelphia locations
about their views and satisfaction with the LDCs.
B.
Analyzed CI and LDC
management information data and identified:
1.
The percentage of
legal source cases opened and fraud referrals accepted by the CI function in
FYs 1999 through 2002.
2.
The number of leads processed, the time to process, the
number of hours charged by the LDCs during FY 2002, and the subsequent actions
by the CI field offices.
3.
The number of open
General Investigations as of September 30, 2002.
C.
Evaluated how the
work was prioritized to develop legal source tax cases.
Appendix II
Major Contributors to This Report
Daniel R. Devlin, Assistant Inspector General
for Audit (Headquarters Operations and Exempt Organizations Programs)
John R. Wright, Director
Ronald F. Koperniak, Audit Manager
Diana M. Tengesdal, Audit Manager
Michael J. Hillenbrand, Senior Auditor
Ahmed Tobaa, Senior Auditor
Janis Zuika, Auditor
Appendix III
Commissioner C
Deputy Commissioner
for Services and Enforcement SE
Director, Strategy
SE:CI:S
Director, Field Operations, Midatlantic Area SE:CI:FO:MA
Director, Field Operations, North Atlantic Area SE:CI:FO:NA
Director, Field Operations, Southeast Area SE:CI:FO:SE
Chief Counsel CC
National Taxpayer Advocate TA
Director, Office of Legislative Affairs CL:LA
Director,
Office of Program Evaluation and Risk Analysis
RAS:O
Office of Management Controls OS:CFO:AR:M
Audit Liaison: Criminal Investigation SE:CI:S:PS
Appendix IV
This appendix presents detailed information on the measurable impact that our recommended corrective actions will have on tax administration. These benefits will be incorporated into our Semiannual Report to the Congress.
Type and Value of Outcome Measure:
· Inefficient Use of Resources – Potential; 676 staff days (see page 3).
Methodology Used to Measure the Reported Benefit:
We analyzed data from the 12 Lead Development Center (LDC) stand-alone databases during Fiscal Year (FY) 2002 to determine how many fraud referrals were processed (through research and analysis) and the time spent by the LDCs processing fraud referrals. According to the information recorded in the LDC databases, the LDCs expended a total of 5,049 hours to research and analyze 436 fraud referrals during FY 2002. The staff day savings of 676 was calculated by dividing the total hours by the number of hours in a normal workday (5,049/8). By adopting our recommendation to send fraud referrals to the field offices, the Internal Revenue Service could use these resources more effectively to identify legal source cases through the use of General Investigations.
Type and Value of Outcome Measure:
· Inefficient Use of Resources – Potential; 252 staff days (see page 9).
Methodology Used to Measure the Reported Benefit:
We analyzed data from the Criminal Investigation Management Information System and the 12 LDC stand-alone databases during FY 2002. The LDCs returned 3,258 leads to the field offices for investigation. As of September 30, 2002, the field offices had closed 960 of the 3,258 leads without opening a criminal case. We identified 279 cases closed by the field offices with less than 8 hours of time charged. The field offices closed 121 of these cases using 7 closing codes that, in our opinion, would make it difficult for the field office to make an investigative decision on the case with less than 8 hours charged, or in which the decision should have been made before the CI field office referred the case to the LDC for research. For example, before a CI field office sends a lead to an LDC, it should determine if it has the resources to work the case. Leads closed due to inadequate resources after the LDC conducted research would indicate an ineffective use of LDC resources.
According to information recorded in the LDC databases, the LDCs expended a total of 2,014 hours conducting research that was not effectively used on these 121 cases. The staff day savings of 252 was calculated by dividing the total hours by the number of hours in a normal workday (2,014/8).
Appendix V
ChoicePoint - A commercial
database that provides access to more than 14 billion current and historical
records on individuals and businesses.
With as little information as a name or Social Security Number, it
cross-references public and proprietary records, including identity
verification information; relatives and associates; corporate information; real
property records; deed transfers; liens; bankruptcies; Uniform Commercial Code
filings; licensed firearms; explosives/weapons dealers; and county, state, and
Federal courthouses.
Currency and Banking Retrieval
System - An
on-line database that contains data received from financial and nonfinancial
institutions, casinos, and businesses of any financial transactions in excess
of $10,000, suspicious activities, and persons carrying cash or monetary
instruments in excess of $10,000 into or out of the country.
Department of Motor Vehicles - Available in each state and provides
driver’s license and vehicle registration information.
Electronic Fraud
Detection System - An Internal Revenue Service (IRS) stand-alone system that accesses
all electronically filed returns that have been scored for potential fraud by
the Electronic Filing programs. A
comparison of online prior year data, including Wage and Tax Statement (Form
W-2) information, with the current tax return allows the Criminal Investigation
(CI) function to make better decisions on cases it believes are potentially
fraudulent.
Information
Return Processing -
An IRS computer system that consists of payer and payee information such as wages, interest, dividends, Social Security
benefits, and early distribution penalty tax.
Integrated Data Retrieval System - An
IRS computer system capable of retrieving or updating stored information; it
works in conjunction with a taxpayer’s account records.
LexisNexis - A commercial database that
provides a nationwide search of over 1,400 different data sources on legal,
corporate, government, and academic markets and publishes legal, tax, and
regulatory information. The service
searches information on people, including bankruptcies, liens and judgments,
prison inmate records, professional licenses, medical professions directories,
voter registration records, and civil and criminal court records.
Suspicious Activity Report - All financial institutions operating in the United States (U.S.),
including insured banks, savings associations, savings association service
corporations, credit unions, bank holding companies, nonbank subsidiaries of
bank holding companies, and U.S. branches and agencies of foreign banks, are
required to make this report following the discovery of insider abuse involving
any amount, violations aggregating $5,000 or more where a suspect can be
identified, violations aggregating $25,000 or more regardless of a potential
suspect, or transactions aggregating $5,000 or more that involve potential
money laundering or violations of the Bank Secrecy Act.
Treasury Enforcement and
Communication System (TECS) - A computerized information system designed to identify
individuals and businesses suspected of, or involved in, violation of Federal
law. Individual records available from
the TECS database come from the U.S. Customs Service; Bureau of Alcohol,
Tobacco, and Firearms; Treasury Inspector General for Tax Administration; and
CI function. The information available
includes vehicles, vessels, aircraft, organizations’ articles, firearms, wanted
persons and fugitives, felons and dishonorably discharged veterans who have
requested relief to own firearms and/or explosives, violent felons, gangs,
terrorists, and nonresident delinquent taxpayers.
Appendix VI
The response was removed due to its size. To see the response, please go to the Adobe PDF version of the report on the TIGTA Public Web Page.