FAS | Intelligence | GAO Reports |||| Index | Search |


Money Laundering: Needed Improvements for Reporting Suspicious Transactions Are Planned

(Chapter Report, 05/30/95, GAO/GGD-95-156)


Pursuant to a congressional request, GAO provided information on money
laundering activities, focusing on: (1) how suspicious transactions are
reported; (2) how currency transactions reports are used by law
enforcement agencies; and (3) whether the reporting process can be
improved.

GAO found that: (1) financial institutions file reports of suspicious
transactions each year on various forms to various agencies, leading to
the initiation of major investigations into various types of criminal
activity; (2) there is no way of ensuring that the information is being
used to its full potential, since there is no overall control or
coordination of the reports; (3) the form that is filed most frequently
is filed with the Internal Revenue Service (IRS) and kept on a
centralized database, but the form is only useful in providing
additional information on an investigation that has already been
initiated; (4) other forms used to report suspicious transactions
contain more useful information but, since they are filed with six
different agencies and are not kept on a centralized database, they
cannot be used on a reactive basis; (5) IRS has not developed agencywide
procedures for managing suspicious transaction reports, resulting in
varied use of the reports among 35 district offices; (6) 9 of the 15
states that receive copies of suspicious transaction reports use the
information to initiate criminal investigations; and (7) the Department
of the Treasury and IRS have agreed to substantial changes regarding how
suspicious transactions are to be reported, how the information is to be
used, and how to improve the reports' contributions at both the federal
and state levels.

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

 REPORTNUM:  GGD-95-156
     TITLE:  Money Laundering: Needed Improvements for Reporting 
             Suspicious Transactions Are Planned
      DATE:  05/30/95
   SUBJECT:  Money laundering
             Investigations by federal agencies
             Law enforcement agencies
             White collar crime
             Federal forms
             Reporting requirements
             Financial institutions
             Federal/state relations
             Law enforcement
IDENTIFIER:  Treasury Currency Transaction Report
             Arizona
             Treasury Financial Crimes Enforcement Network
             Treasury Project Gateway
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


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


Report to the Ranking Minority Member
Permanent Subcommittee on Investigations, Committee on Governmental
Affairs, U.S.  Senate

May 1995

MONEY LAUNDERING - NEEDED
IMPROVEMENTS FOR REPORTING
SUSPICIOUS TRANSACTIONS ARE
PLANNED

GAO/GGD-95-156

Suspicious Transaction Reports


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

  CID - Criminal Investigation Division
  CRF - Criminal Referral Form
  CTR - Currency Transaction Report
  FinCEN - Financial Crimes Enforcement Network
  IRS - Internal Revenue Service

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


B-259790

May 30, 1995

The Honorable Sam Nunn
Ranking Minority Member
Permanent Subcommittee
  on Investigations
Committee on Governmental Affairs
United States Senate

Dear Senator Nunn: 

This report was prepared in response to your request to determine how
suspicious transactions that might involve money laundering are
reported.  The report discusses the various forms used by financial
institutions to report suspicious transactions and how the reports
are used by law enforcement agencies at the state and federal levels. 

As arranged with the Subcommittee, unless you announce its contents
earlier, we plan no further distribution of this report until 30 days
from its issue date.  At that time, we will send copies to other
congressional committees, various bureaus and offices within the
Department of the Treasury, and other interested parties.  Copies
will be made available to others upon request. 

The major contributors to this report are listed in appendix IV. 
Please contact me on (202) 512-8777 if you or your staff have any
questions concerning this report. 

Sincerely yours,

Laurie E.  Ekstrand
Associate Director, Administration
  of Justice Issues


EXECUTIVE SUMMARY
============================================================ Chapter 0


   PURPOSE
---------------------------------------------------------- Chapter 0:1

Money laundering is the disguising or concealing of illicit income in
order to make it appear legitimate.  Financial institutions such as
banks, savings and loan associations, and credit unions are in a
unique position to help identify money launderers by reporting
suspicious transactions to federal and state law enforcement
authorities.  The Permanent Subcommittee on Investigations, Senate
Committee on Governmental Affairs, asked GAO to determine (1) how
suspicious transactions are reported, (2) how the reports are used by
law enforcement agencies, and (3) whether the process can be
improved. 


   BACKGROUND
---------------------------------------------------------- Chapter 0:2

Federal law enforcement officials estimate that between $100 billion
and $300 billion is laundered in this country each year.  While
illegal drug trafficking accounts for much of the funds being
laundered, other criminal activities, including tax evasion, also
account for an extensive amount. 

In the past two decades, federal law enforcement efforts to combat
money laundering have focused on requiring financial institutions to
report currency transactions that exceed $10,000.  Beginning in 1988,
these reports have been supplemented by reports of suspicious
transactions.  Many of the transactions reported as suspicious
involve individuals who appear to be attempting to avoid the $10,000
reporting requirement.  However, any activity that deviates from the
norm for a particular account can be considered suspicious. 

The Right to Financial Privacy Act, enacted in 1978, raised questions
as to whether financial institutions were authorized to report
suspicious transactions.  To address these concerns, legislation has
been enacted to provide protection against civil liability for
institutions reporting suspicious transactions. 


   RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3

Banks and other financial institutions report tens of thousands of
suspicious transactions each year.  The reports have led to the
initiation of major investigations into various types of criminal
activity.  However, because there is no overall control or
coordination of the reports, there is no way of ensuring that the
information is being used to its full potential. 

Financial institutions report suspicious transactions on a variety of
different forms that provide different types of information and that
are filed with different law enforcement and regulatory agencies. 
The form that is filed most frequently is filed with the Internal
Revenue Service (IRS) and kept on a centralized database.  However,
the form does not contain any information describing the suspicious
activity that would allow law enforcement agencies to evaluate the
usefulness of the information on the basis of the form alone. 
Moreover, some institutions have been filing these forms erroneously. 
IRS and other federal and state law enforcement agencies use the
database on a reactive basis; that is, to provide additional
information on an investigation that has already been initiated. 

Other forms used to report suspicious transactions do describe the
activity so that the information can be evaluated.  However, these
forms are filed with six different federal financial regulatory
agencies.  Because the forms are not maintained on a centralized
database, they are not used on a reactive basis.  Financial
institutions filing this form are required to send a copy of it to
the nearest district office of IRS' Criminal Investigation Division. 
However, IRS has not developed any guidance or directives as to how
the information is to be managed as an intelligence resource.  Use of
the reports to initiate investigations varies among the 35 district
offices. 

GAO identified 15 states that receive copies of suspicious
transaction reports filed on one or both of these two forms.  Nine of
these states told GAO that they use the information to initiate
criminal investigations. 

The Department of the Treasury, the financial regulatory agencies,
and IRS have recently agreed to substantial changes regarding how
suspicious transactions are to be reported and how the information is
to be used.  These proposals, which were made with input from the
financial community, have the potential for significantly improving
the contribution that suspicious transaction reports make to law
enforcement at both the federal and state levels. 


   GAO'S ANALYSIS
---------------------------------------------------------- Chapter 0:4


      MANY REPORTS OF SUSPICIOUS
      TRANSACTIONS HAVE LIMITED
      UTILITY FOR LAW ENFORCEMENT
-------------------------------------------------------- Chapter 0:4.1

In 1990 the Department of the Treasury modified the Currency
Transaction Report (CTR) form, which financial institutions use to
report currency transactions exceeding $10,000 to IRS.  A block was
added to the form that could be checked to indicate that the
transaction was considered suspicious.  In addition, the instructions
for preparing the CTR were amended so that the form could be used to
report suspicious transactions of any dollar amount. 

Of the 10.2 million CTRs filed in 1993, 63,536 were marked
suspicious.  Most of these suspicious CTRs--47,083--identified
transactions that were for $10,000 or less.  The average amount of
these transactions was $7,117.  (See p.  19.) Although CTRs are the
form financial institutions use most frequently to report suspicious
transactions, it is the least useful for providing intelligence
information on possible criminal activity.  Because there is no means
for indicating on the form the reason why the institution considered
the transaction suspicious, it is difficult for law enforcement
personnel to evaluate the information and assess its potential. 

Most financial institutions do not use CTRs to report suspicious
transactions.  Of the 26,029 institutions that filed CTRs in 1993,
less than 18 percent marked one or more of the CTRs as suspicious. 
Over 25 percent of the suspicious CTRs were filed by 20 institutions,
including 3 institutions that erroneously filed the suspicious CTRs
for transactions that they did not consider to be suspicious.  (See
pp.  20 and 21.)


      OTHER FORMS ARE ALSO USED TO
      REPORT SUSPICIOUS
      TRANSACTIONS
-------------------------------------------------------- Chapter 0:4.2

Since 1988 the financial regulatory agencies have required banks,
savings and loan associations, and credit unions to report suspected
money laundering on a Criminal Referral Form (CRF).  Although each
regulatory agency requires the use of its own form, each form
contains essentially the same information, including a description of
the transaction.  Of the 80,340 CRFs that were filed in 1993, 13,220
reported suspected money laundering.  The remaining CRFs reported
other criminal activity, such as credit card fraud, employee theft,
and check kiting.  (See p.  23.)

There is no central depository of CRFs.  They are to be filed with
the cognizant regulatory agency and, for those that report suspected
money laundering, a copy is to be sent to the nearest district office
of IRS' Criminal Investigation Division (CID). 

CID offices in IRS' western region have been receiving suspicious
transaction reports on various forms since the late 1980s, when
special agents began soliciting reports on their own initiative.  At
first the reports were made exclusively on a locally designed form. 
Currently, even though financial institutions throughout the country
use CTRs and/or CRFs to make suspicious transaction reports, some
institutions in the western part of the country continue to use the
locally designed forms in addition to the CTR and/or CRF forms.  (See
p.  24.)

Arizona has had its own form for banks in that state to use for
reporting suspicious transactions since 1985.  State officials told
GAO that in 1993 financial institutions in Arizona filed
approximately 1,200 of these forms with the state Attorney General's
Office.  (See p.  24.)


      REPORTS OF SUSPICIOUS
      TRANSACTIONS ARE HELPFUL TO
      LAW ENFORCEMENT BUT THEIR
      USE IS INCONSISTENT
-------------------------------------------------------- Chapter 0:4.3

CID special agents GAO spoke with provided numerous examples of major
investigations, many conducted by other agencies, that were initiated
on the basis of suspicious transactions reported to IRS.  These cases
involved different types of criminal activity, such as drug
trafficking, food stamp fraud, and theft of government property. 
Many of the cases involved millions of dollars in illicit proceeds. 
(See pp.  26 and 27.)

IRS does not have agencywide policies or procedures for managing
suspicious transaction reports.  Consequently, the extent to which
special agents in the 35 CID district offices solicit, process, and
evaluate the reports is up to the discretion of the district CID
chief and varies significantly among districts.  The percentage of
investigations initiated on the basis of suspicious transaction
reports also varies significantly among districts.  From October 1990
to June 1994 CID initiated 21,507 investigations nationwide.  About 4
percent of the cases were initiated as a result of a suspicious
transaction report.  Among the district offices, however, the
percentage varied from 0 to over 18 percent.  GAO believes that the
varying rates are an indication that use of the reports may not be
emphasized to the same extent among the districts.  (See p.  30.)

Almost all of the states have been authorized direct access to the
Treasury database of CTRs which includes those CTRs that have been
marked as suspicious.  However, states must use a specific name to
search the database and can access only the suspicious CTRs filed on
the individual or business named.  Consequently, states do not have
the capability of using the database to target subjects for further
investigation.  (See p.  31.)

GAO identified 15 states that receive copies of suspicious CTRs
and/or CRFs that have been filed by financial institutions within
their state.  Nine of these states told GAO they use the information
to target subjects for further investigation.  (See pp.  31 through
34.)


      RECENT INITIATIVES FOR
      IMPROVING SUSPICIOUS
      TRANSACTION REPORTING
-------------------------------------------------------- Chapter 0:4.4

The Department of the Treasury and the financial regulatory agencies,
with input from the financial community, have recently proposed
several major changes regarding suspicious transaction reporting. 
These include using a single form for the reports that would provide
a description of the transaction, establishing a centralized database
for the reports, and facilitating access to the information by the
states.  In addition, IRS plans to develop and issue policies and
procedures to ensure that the reports are used and managed on a
consistent basis.  GAO believes the actions planned by Treasury and
IRS, if properly implemented in a timely manner, will significantly
improve how suspicious transactions are reported and used.  (See pp. 
34 through 36.)


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5

GAO is not making any recommendations in this report. 


   AGENCY AND INDUSTRY COMMENTS
---------------------------------------------------------- Chapter 0:6

GAO provided a draft of this report to the American Bankers
Association, Treasury's Financial Crimes Enforcement Network, and
IRS.  Their written comments are discussed in chapter 3 and are
contained in appendixes I, II, and III.  Each of these organizations
generally agreed with the information in the report. 


INTRODUCTION
============================================================ Chapter 1

Money laundering is the disguising or concealing of illicit income in
order to make it appear legitimate.  Over the past two decades,
federal law enforcement efforts to detect money laundering have
evolved into a strategy that is heavily dependent upon the reporting
of large currency transactions and tactical and strategic
intelligence analysis of the collected data. 

In 1988 the Department of the Treasury began to encourage banks and
other financial institutions to supplement reports of large currency
transactions with reports of suspicious transactions of any amount. 
Since then, the suspicious transaction reports have taken on a number
of different formats that are filed with various law enforcement and
regulatory agencies at both the state and federal levels.  This
report describes how these reports are made and how they are being
used. 


   CRIMINALS FACE PROBLEMS IN
   DEALING IN LARGE AMOUNTS OF
   CASH
---------------------------------------------------------- Chapter 1:1

Federal law enforcement officials estimate that between $100 billion
and $300 billion in U.S.  currency is laundered each year.  While
narcotics traffickers are the largest single block of users of money
laundering schemes, numerous other types of activities typical of
organized crime--for example, illegal gambling or
prostitution--create an appreciable demand.  In addition, violations
of tax laws often accompany laundering schemes that conceal the
existence of an illegal source of income.  Money laundering is also a
factor in many cases of tax fraud involving income from a legitimate
source. 

Although the process of money laundering has been broken down into a
number of steps, it is generally agreed by law enforcement and
regulatory officials that the point at which criminals are most
vulnerable to detection is "placement." Placement is the concealing
of illicit proceeds by

  converting the cash to another medium that is more convenient or
     less suspicious for purposes of exchange, such as property,
     cashier's checks, or money orders; or

  depositing the funds into a financial institution account for
     subsequent disbursement. 

Because of the problems associated with converting and concealing
large amounts of cash, placement is perhaps the most difficult part
of money laundering and is currently the primary focus of U.S.  law
enforcement, legislative, and regulatory efforts to attack money
laundering. 


   U.S.  EFFORTS TO COMBAT MONEY
   LAUNDERING CENTER ON REPORTS OF
   LARGE CASH TRANSACTIONS
---------------------------------------------------------- Chapter 1:2

Federal efforts to detect large cash deposits were significantly
enhanced with the passage of the Bank Secrecy Act in 1970.  The act
requires individuals as well as banks and other financial
institutions to report large foreign and domestic financial
transactions to the Department of the Treasury.  The act has been
amended to provide substantial criminal and civil penalties for
institutions who fail to file the required reports and for
individuals who deliberately evade certain reporting requirements. 
Although the implementing regulations of the act require four types
of reports, the report filed most frequently is the Currency
Transaction Report (CTR). 

Financial institutions\1 are required to file a CTR for each deposit,
withdrawal, exchange of currency, or other payment or transfer, by,
through, or to such institutions that involves a transaction in
currency of more than $10,000.  CTRs are filed on an Internal Revenue
Service (IRS) form 4789, which is to be sent to the IRS Detroit
Computing Center in Michigan. 

The volume of CTRs being filed has increased substantially in the
past several years.  In May 1993 we testified before the House
Banking Committee that since 1987 the annual filings of CTRs had
increased at an average rate of 12.7 percent.\2 Increased efforts by
federal regulatory and law enforcement agencies, as well as enhanced
cooperation by the banks themselves, have significantly improved bank
compliance with the reporting requirements.  The substantial increase
in the volume of currency transaction reports being filed has
increased the importance of identifying those transactions thought to
be suspicious. 


--------------------
\1 As defined by Treasury regulation, these include banks, federally
regulated security brokers, currency exchange houses, funds
transmitters, check cashing businesses, and persons subject to
supervision by state or federal bank supervisory authority.  In this
report, unless otherwise noted, the term "financial institutions"
will refer to banks, credit unions, and savings and loan
associations. 

\2 Money Laundering:  The Use of Bank Secrecy Act Reports by Law
Enforcement Could Be Increased (GA0/T-GGD-93-31, May 26, 1993). 
Legislation enacted in September 1994 contains several provisions for
reducing the volume of CTRs filed. 


   SUSPICIOUS TRANSACTIONS ARE
   LOOSELY DEFINED
---------------------------------------------------------- Chapter 1:3

Although U.S.  financial institutions have been reporting suspected
money laundering for a number of years, specific criteria for
determining whether a transaction is suspicious have never been
developed.  Consequently, institutions generally have a wide degree
of latitude in deciding what constitutes suspicious activity. 

Financial institutions have developed a number of means designed to
help ensure that they are not being used to launder illicit proceeds. 
Chief among these is a policy commonly referred to as "know your
customer." Among other things, the policy calls for financial
institutions to verify the identity of individuals and businesses
that are account holders and to be familiar enough with their banking
practices so that transactions that are outside the norm can be
readily identified.  Officials from the Department of the Treasury
and the American Bankers Association told us that most, if not all,
financial institutions have implemented a know your customer policy
and treat any transaction not typically associated with an account as
suspicious.  Moreover, guidance from regulatory agencies generally
encourages institutions to use the policy in this manner. 

Although suspicious activity generally depends upon the customer,
certain types of transactions are suspicious in and of themselves.  A
common type of suspicious transaction is structuring.  Structuring
occurs when a person conducts currency transactions in amounts of
$10,000 or less for the purpose of evading the reporting requirements
of the Bank Secrecy Act. 

In September 1992 the Association of Reserve City Bankers (now known
as the Bankers Roundtable) published the results of a survey of
suspicious transaction reporting by the nation's major banking
institutions.\3 The report included more than 200 profiles of
suspicious transactions that had been reported by 60 of the nation's
largest banking institutions.  The majority of the transactions that
were reported as suspicious (85 percent) involved structuring.  The
report found that the most common method of structuring involved cash
deposits but also included check cashing, cash withdrawals, and the
purchase of monetary instruments. 

Other transactions that were reported as suspicious included

  customers changing the dollar amount of the transaction or
     cancelling the transaction when informed of the reporting
     requirement,

  unusually large purchases of money orders and cashier's checks,

  unusually large cash deposits, and

  wire transfers of funds to a foreign country. 


--------------------
\3 Survey of Suspicious Transactions, Final Report of the
Subcommittee and Task Force on Money Laundering, Payments System
Committee, Association of Reserve City Bankers. 


   LEGISLATION HAS FACILITATED
   SUSPICIOUS TRANSACTION
   REPORTING BY FINANCIAL
   INSTITUTIONS
---------------------------------------------------------- Chapter 1:4

The Department of the Treasury has identified 8 countries in addition
to the United States that, as of July 1993, require the reporting of
currency transactions that exceed a specified amount.\4 However, many
other countries require the recording of transactions over some
specified threshold.  These records can then be made available to law
enforcement under the terms of that country's bank secrecy laws. 

Many countries also either require or encourage financial
institutions to report those transactions considered to be
suspicious.  In the United States, financial institutions have been
encouraged for some time to report suspicious account activity that
might be indicative of criminal activity.  However, certain
provisions in the Right to Financial Privacy Act (P.L.  95-630) of
1978 generated questions in the banking community about the type of
customer information that could be disclosed in reporting a
suspicious transaction, as well as concerns of potential liability
for such disclosure.  Subsequent legislation addressed these issues
by, among other things, providing certain protections against civil
liability for institutions reporting suspicious transactions. 

The Money Laundering Control Act of 1986 (P.L.  99-570) amended the
Right to Financial Privacy Act to explicitly define the specific
types of account information that financial institutions could
disclose without customer permission, subpoena, summons, or search
warrant.  The intent was to strike a balance between the privacy
rights of customers while allowing financial institutions to give
government investigators enough information about the nature of
possible violations in order for such investigators to determine
whether there was a basis to proceed with a summons, subpoena, or
search warrant for additional information.  The 1986 amendments also
established a limited "good faith" defense whereby financial
institutions and their employees, when making a disclosure of certain
specified information, would be shielded from civil liability to the
customer for such disclosure or for any failure to notify the
customer of such disclosure. 

Despite this provision, many banks were concerned that they might
still be liable under the Right to Financial Privacy Act for
disclosures made on a voluntary basis.  Nothing in the statutory
language required a financial institution to initiate a disclosure to
a government agency of a suspected transaction, and some questioned
whether the government would intervene on their behalf should a civil
action be initiated against them. 

This situation was remedied, to some extent, by the promulgation of
regulations by the Comptroller of the Currency and other federal
agencies charged with the responsibility to monitor U.S.  financial
institutions.  Comptroller of the Currency Regulation 12 C.F.R. 
Section 21.11 and corresponding regulations issued by the other bank
regulatory agencies now require financial institutions to report
suspected money laundering. 

Nonetheless, there was still concern over the possibility of civil
suits because of reporting suspicious transactions.  In 1992, under
the Annunzio-Wylie Anti-Money Laundering Act (P.L.  102-550),
financial institutions and their employees reporting suspicious
transactions were given broadened immunity from civil liability under
any state or federal law or regulation, such as the Right to
Financial Privacy Act.  The act also prohibits financial institutions
from notifying persons involved in a suspicious transaction that the
transaction has been reported. 


--------------------
\4 These are Australia, Brazil, Costa Rica, Ecuador, Norway,
Paraguay, Uruguay, and Venezuela. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:5

We were requested by the then Chairman of the Permanent Subcommittee
on Investigations, Senate Governmental Affairs Committee, to review
the manner in which suspicious activities that relate to possible
money laundering are reported by determining

  how banks and other financial institutions report suspicious
     transactions,

  to whom the transactions are reported,

  the volume of reports made,

  how the reports are used, and

  whether the process can be improved. 

To respond to the request, we reviewed pertinent laws and regulations
and published material such as academic and periodical literature. 
We also reviewed reports prepared by federal and state agencies,
private research associations, and other experts.  We interviewed
officials at the Internal Revenue Service, the Department of the
Treasury, the Federal Reserve Board, and the American Bankers
Association.  We also used the results of our previous reports
dealing with money laundering that are cited in the text. 

In order to determine the volume and characteristics of suspicious
transaction reports filed on Currency Transaction Reports, we used
data from the computer database at the IRS Detroit Computing Center
and relied upon IRS for the necessary computer programming.  At our
request, IRS identified the 20 institutions that filed the largest
volume of Currency Transaction Reports that had been marked
suspicious in calendar year 1993.  In order to determine what factors
might influence some institutions to mark a high percentage of CTRs
as suspicious, we contacted the seven institutions of these 20 that
had marked more than 8 percent of the CTRs filed as suspicious.  This
percentage was arbitrarily selected and has no statistical basis. 

To ascertain what states require suspicious transaction reporting and
how the reports are used, we telefaxed a single-page questionnaire to
bank regulatory officials in each state.  All of the states
responded.  For those states that indicated there was a requirement,
we conducted telephone interviews with regulatory and law enforcement
personnel.  We interviewed officials and observed operations at the
state facility in Sacramento, California, that processes data for
that state's Office of the Attorney General.  We chose this one state
operation to visit because of its proximity to San Francisco,
California, where we were reviewing IRS district operations. 

In order to determine the extent to which suspicious transaction
reports are used to initiate investigations by IRS' Criminal
Investigation Division (CID), we used data provided us from a
management information system at IRS headquarters.  We visited or
contacted by telephone a total of 10 CID district offices.  The San
Francisco office was selected because of its recognized role as an
innovator in using suspicious transaction reports.  The other
district offices were judgmentally selected so as to include offices
that had initiated a relatively high percentage of cases based on
suspicious transaction reports as well as those that had initiated a
low percentage. 

We provided a draft of this report to the American Bankers
Association, Treasury's Financial Crimes Enforcement Network, and
IRS.  Their comments are discussed on pages 37 and 38 and reproduced
in full in appendixes I, II, and III. 

We did our review from April through December 1994 in accordance with
generally accepted government auditing standards. 


REPORTS OF SUSPICIOUS TRANSACTIONS
ARE MADE ON DIFFERENT FORMS AND
WITH DIFFERENT AGENCIES
============================================================ Chapter 2

Over the past several years, different forms have been developed for
financial institutions to use in reporting transactions that might
involve money laundering.  Each of these forms has evolved from a
recognized need, but the forms differ as to the amount and detail of
the information provided and where the form is filed.  Because of the
concurrent development and implementation of the forms, the reports
overlap one another.  Consequently, the same suspicious activity may
be reported two or more times, on two or more different forms, and to
several different agencies. 

This chapter describes how suspicious transactions are reported and
to whom they are reported.  Chapter 3 discusses how the various
reports are used by different law enforcement agencies. 


   CURRENCY TRANSACTION REPORTS
   CAN BE USED TO IDENTIFY
   SUSPICIOUS TRANSACTIONS
---------------------------------------------------------- Chapter 2:1

As previously discussed, financial institutions are required to
report certain transactions that exceed $10,000 on a Currency
Transaction Report (CTR).  Beginning in 1990, CTRs have also been
used by some institutions to identify suspicious transactions. 
Although this means of identifying suspicious transactions produces
the largest volume of reports, most financial institutions do not use
the CTR form to report suspicious transactions.  Using the CTR for
this purpose does not provide any information about the nature of the
suspicious activity.  Moreover, the validity of some of the
suspicious transaction reports filed on a CTR is questionable because
some have been filed erroneously. 


      SUSPICIOUS CTRS ARE
      VOLUNTARY AND PROVIDE LITTLE
      INFORMATION
-------------------------------------------------------- Chapter 2:1.1

After it had received inquiries from financial institutions about
whether suspicious transactions should be reported and what
information should be reported, the Department of the Treasury issued
Administrative Ruling 88-1 on June 22, 1988.  The ruling encourages
but does not require financial institutions to report those
transactions that might be "...relevant to a possible violation of
the Bank Secrecy Act or its regulations or indicative of money
laundering or tax evasion" to the local Criminal Investigation
Division (CID) office of the Internal Revenue Service. 

Immediately after Administrative Ruling 88-1 was released, Treasury
officials began to notice that financial institutions were reporting
suspicious transactions by filing CTRs with the word "suspicious"
written across the form.  To facilitate this type of reporting,
Treasury issued a revised CTR form in January 1990 with a block that
could be checked to indicate that the transaction was suspicious.  In
addition, the instructions for the form were amended to read "This
form may be filed for any suspicious transaction, even if it does not
exceed $10,000." Although the revised form makes it possible to
identify transactions that have been designated as suspicious, the
form does not provide for a description of the transaction. 
Consequently, there is no way to determine why the transaction was
considered suspicious. 

All CTRs are to be filed with the IRS Detroit Computing Center, where
they are processed and entered onto a computer database along with
other reports required by the Bank Secrecy Act.  Once a week, staff
at the Center are to distribute copies of those CTRs that are marked
suspicious to the CID district office that has jurisdiction over the
state where the CTR was filed. 

In 1993 more than 10 million CTRs were filed, 63,536 of them marked
suspicious.\1 As figure 2.1 demonstrates, the number of suspicious
CTRs filed since 1990 has remained relatively constant despite a
substantial increase in the volume of CTRs filed. 

   Figure 2.1:  CTRs Filed, 1990
   Through 1993

   (See figure in printed
   edition.)


--------------------
\1 Approximately 126,000 CTRs were filed for amounts of $10,000 or
less but were not marked suspicious.  IRS has not determined the
reason why these reports were filed. 


      MOST FINANCIAL INSTITUTIONS
      DO NOT FILE SUSPICIOUS CTRS
-------------------------------------------------------- Chapter 2:1.2

Of the 35,131 institutions filing CTRs in 1993, about 75 percent
identified themselves as banks, credit unions, or savings and loan
associations.  Of these 26,029 financial institutions, only
4,473--about 17 percent--marked 1 or more of the CTRs they filed as
suspicious.  Overall, less than 1 percent of the CTRs filed by all
financial institutions were marked suspicious.  Table 2.1 provides
additional information on institutions that filed CTRs and suspicious
CTRs in 1993.  Table 2.2 provides additional data on the suspicious
CTRs filed in 1993. 



                                    Table 2.1
                     
                      CTRs and Suspicious CTRs Filed During
                                Calendar Year 1993

                                       Number of                         Percent
                 Number of            institutio   Percent               of CTRs
                institutio    Number   ns filing    filing   Number of    marked
Type of          ns filing   of CTRs  suspicious  suspicio  suspicious  suspicio
institution           CTRs     filed        CTRs   us CTRs  CTRs filed        us
--------------  ----------  --------  ----------  --------  ----------  --------
Bank                20,132  9,894,70       3,410      16.9      49,498       0.5
                                   7
Savings and          2,295   143,026         503      21.9       4,624       3.2
 loan
 association
Credit union         3,602    32,326         560      15.5       2,488       7.7
================================================================================
Subtotals           26,029  10,070,0       4,473      17.2      56,610       0.6
                                  59
Securities             311     2,405          13       4.2         296      12.3
 broker or
 dealer
Other                2,738   130,649         307      11.2       5,755       4.4
Unknown              6,053    44,011         345       5.7         875       2.0
================================================================================
Totals              35,131  10,247,1       5,138      14.6      63,536       0.6
                                  24
--------------------------------------------------------------------------------
Source:  IRS Detroit Computing Center. 



                                    Table 2.2
                     
                      Suspicious CTRs Filed During Calendar
                      Year 1993 by Type of Filer and Amount

                  Suspicio
                        us
                      CTRs
                     filed                        Suspicious
                       for    Number              CTRs filed    Number
Type of            $10,000        of   Average      for more        of   Average
institution        or less    filers    amount  than $10,000    filers    amount
----------------  --------  --------  --------  ------------  --------  --------
Bank                37,001     2,632    $7,340        12,497     1,849   $27,030
Savings and loan     3,858       407     6,881           766       247    22,734
 association
Credit union         1,938       431     5,931           550       262    23,085
================================================================================
Subtotals           42,797     3,470    $7,235        13,813     2,358   $26,635
Securities              27         7    $5,736           269        11   $35,318
 broker or
 dealer
Other                3,553       214     6,003         2,202       155    15,268
Unknown                706       267     5,622           169       110    20,206
================================================================================
Totals              47,083     3,958    $7,117        16,453     2,634   $25,189
--------------------------------------------------------------------------------
Source:  IRS Detroit Computing Center. 

We discussed suspicious transaction reporting with officials of the
two largest (ranked by total assets) banks in the country.  Both
banks have a policy of not filing suspicious CTRs.  The reasons given
for this policy were a concern over inadequate internal review and
evaluation of the reports, possible civil liability for violating a
customer's right to privacy, and the lack of space on the form to
describe why the transaction was considered to be suspicious. 


      A DISPROPORTIONATE NUMBER OF
      SUSPICIOUS CTRS ARE FILED BY
      A FEW INSTITUTIONS
-------------------------------------------------------- Chapter 2:1.3

Although a total of 5,138 institutions filed suspicious CTRs in 1993,
over a quarter of the 63,536 suspicious CTRs filed were filed by 20
institutions.  At the request of the Department of the Treasury, we
are not revealing the identity of these institutions.  However, table
2.3 provides additional information concerning these institutions. 



                                    Table 2.3
                     
                     Selected Statistical Information on the
                        20 Institutions Filing the Largest
                        Volume of Suspicious CTRs in 1993

                            Percent of
               Number of           all                  Total CTRs       Percent
              suspicious    suspicious    Cumulative      filed by        marked
Filer         CTRs filed    CTRs filed    percentage   institution    suspicious
----------  ------------  ------------  ------------  ------------  ------------
1                  3,079          5.04          5.04         5,486         56.12
2                  1,891          3.10          8.14        69,035          2.74
3                  1,620          2.65         10.79        72,287          2.24
4                  1,501          2.46         13.25        11,967         12.54
5                  1,047          1.71         14.96       125,036          0.84
6                    779          1.28         16.24        19,606          3.97
7                    758          1.24         17.48         1,651         45.91
8                    717          1.17         18.65        87,516          0.82
9                    674          1.10         19.76        11,184          6.03
10                   535          0.88         20.63        58,512          0.91
11                   515          0.84         21.48         1,093         47.12
12                   449          0.74         22.21         7,376          6.09
13                   424          0.69         22.90           987         42.96
14                   382          0.63         23.53         5,726          6.67
15                   379          0.62         24.15           780         48.59
16                   375          0.61         24.76         7,680          4.88
17                   373          0.61         25.38        68,227          0.55
18                   366          0.60         25.97        11,226          3.26
19                   347          0.57         26.54         4,270          8.13
20                   330          0.54         27.08         4,788          6.89
================================================================================
Totals            16,541         27.08
--------------------------------------------------------------------------------
Source:  GAO analysis of IRS data. 


      REASONS FOR FILING
      SUSPICIOUS CTRS VARY, AND
      SOME ARE FILED ERRONEOUSLY
-------------------------------------------------------- Chapter 2:1.4

In order to determine the factors that influence some institutions to
mark a high percentage of CTRs as suspicious, we contacted the seven
institutions that had marked more than 8 percent of CTRs filed as
suspicious.  This percentage was arbitrarily selected and has no
statistical basis.  We found a variety of reasons for why
institutions filed suspicious CTRs, and we also identified several
instances where suspicious CTRs were filed erroneously. 

The institution filing the largest number of suspicious CTRs--over 5
percent of those filed nationwide--is not a financial institution but
a large corporation that provides money transmitting services at
thousands of locations nationwide.  Under procedures developed by the
company, all transactions over a specified dollar threshold set by
the company are to be monitored at a central location where the
decision is made about whether or not to file a suspicious CTR. 
Company officials we spoke with told us that because of the nature of
their business, they are inclined to regard many cash transactions as
suspicious even though the amount might be relatively small compared
to typical transactions at a financial institution. 

The eleventh largest filer of suspicious CTRs is also not a financial
institution but a liquor store that operates a check cashing service. 
Staff from the store told us that they had been filing the suspicious
CTRs erroneously because of incorrect instructions they had received. 
An IRS agent had informed them that a CTR was to be filed whenever a
customer's total transactions exceeded $10,000.  (Although Treasury
regulations do call for aggregating transactions, the time period
specified is 1 business day.) The store maintains records on
individual customers so that any time the transaction total exceeded
$10,000, which may have taken several months or longer, a CTR would
be filed on each subsequent transaction no matter what the amount
was.  According to store personnel, they had also been told by IRS to
classify these CTRs as suspicious since none of the other transaction
descriptions on the form were appropriate to describe the
transaction. 

According to officials we spoke with at the institution filing the
fifteenth largest volume of suspicious CTRs--a small bank--most, if
not all, of its suspicious CTRs were filed at the request of IRS.  We
were told that IRS had informed the bank that an account holder was
under investigation and that deposit activity for the account was
generally under the $10,000 reporting threshold for a CTR.  The bank
officials also told us that IRS requested the bank to file a
suspicious CTR for every transaction no matter how much the amount
was so that IRS would be able to monitor the account activity. 
Consequently, the suspicious CTRs were not being filed because the
bank considered the transactions to be suspicious but in order to
allow IRS to monitor activity that would not otherwise be reported. 

The seventh largest filer of suspicious CTRs is a small bank located
in one of the nation's largest cities.  We were told that after the
bank initially determines that an account has had a single suspicious
transaction, its policy is to file suspicious CTRs on all subsequent
transactions for that account.  Bank officials told us that many of
the suspicious CTRs filed by the bank are likely to be erroneous
since not all subsequent transactions might be considered suspicious. 

We were also told that the bank had been heavily fined in the past
for Bank Secrecy Act violations.  As proof of its willingness to
comply with the spirit as well as the letter of the law, the bank has
implemented a policy that encourages employees to file suspicious
CTRs whenever there is any questionable activity.  This "when in
doubt, file" philosophy was echoed by the remaining three banks of
the seven that we spoke with. 


   CRIMINAL REFERRAL FORMS ARE
   ALSO USED TO REPORT SUSPECTED
   MONEY LAUNDERING
---------------------------------------------------------- Chapter 2:2

Federal regulations require financial institutions to file Criminal
Referral Forms or Reports of Apparent Crime (CRF) to report known or
suspected crimes, such as credit card fraud, employee theft, and
check kiting.  In 1988 the activity to be reported was broadened to
include suspected structuring of transactions to evade the CTR
reporting requirements, other violations of the Bank Secrecy Act, and
money laundering.  (See ch.  1, p.  14)

Each of the financial regulatory agencies requires its own form be
used for the report.  The different forms, however, provide
essentially the same information about the identity of the reporting
institution and the individual or business that is the subject of the
report.  Each form also differs substantially from the CTR in that
each has space for a description of the transaction or activity that
is being reported as suspicious. 

The directions for filing the reports require the financial
institution to send the original to the cognizant regulatory agency
and copies to the nearest office of the United States Attorney, the
closest office of the Federal Bureau of Investigation, and the
Department of the Treasury.  The instructions also specify that when
suspected money laundering and/or Bank Secrecy Act violations are
being reported, a copy of the report is to be sent to the local
office of the IRS Criminal Investigation Division.  Table 2.4 shows
the CRFs filed with the regulatory agencies, including those CRFs
reporting suspected money laundering and/or Bank Secrecy Act
violations. 



                          Table 2.4
           
             Criminal Referral Forms Filed During
                      Calendar Year 1993

                                            CRFs  Percent of
                                       reporting        CRFs
                                       suspected   reporting
                          Total CRFs       money       money
Regulatory agency           received  laundering  laundering
------------------------  ----------  ----------  ----------
Federal Reserve                5,799       1,585        27.3
Office of Thrift              10,401       3,230        31.1
 Supervision
Federal Deposit               15,657       2,170        13.9
 Insurance Corporation
Resolution Trust                 375          24         6.4
 Corporation
National Credit Union            761          11         1.4
 Administration
Office of the                 47,347     6,200\a        13.1
 Comptroller of the
 Currency
============================================================
Totals                        80,340      13,220        16.5
------------------------------------------------------------
\a Estimated by agency personnel. 

Source:  GAO poll of regulatory agencies. 


   SOME IRS DISTRICTS RECEIVE
   OTHER TYPES OF SUSPICIOUS
   TRANSACTION REPORTS
---------------------------------------------------------- Chapter 2:3

As discussed in chapter 1, the Money Laundering Control Act of 1986
amended the Right to Financial Privacy Act with provisions that
authorized financial institutions to disclose certain specified
account information.  Recognizing the potential value of information
and reports of suspicious transactions that could now be obtained
from financial institutions, special agents with the IRS Criminal
Investigation Division (CID) in the San Francisco, California,
district office began a local initiative in 1987 to capitalize on the
legislation. 

Under the initiative, financial institutions--primarily banks--in
IRS' western region were asked to report suspicious transactions
directly to the local CID office.  At first, the reports were taken
over the telephone.  As cooperation by the banks increased and the
volume of telephone calls became difficult to manage, the financial
institutions began filing the reports on a one-page form that was to
be mailed to the CID district office.  The form is shorter than the
multipage Criminal Referral Form used by financial regulatory
agencies but, similar to the CRF, has space for a narrative
description of the suspicious nature of the transaction. 

Currently, even though financial institutions throughout the country
use CTRs and/or CRFs to make suspicious transaction reports, some
institutions in the western part of the country continue to file an
additional report directly with the local CID district office.  The
reports are to be evaluated and researched at the district, sent to
an IRS computer service center in California, transcribed onto
computer tape, and mailed to the IRS Detroit Computing Center in
Detroit, where they are to be put on a database.  As of July 1994 a
total of 68,111 reports had been filed with CID district offices in
IRS' western region--mostly with the San Francisco office. 

In calendar year 1993 a total of 91 financial institutions filed
20,940 reports with the CID district offices, again mostly with the
San Francisco office.  Some of the reports that were filed were
copies of CRFs that were filed with the CID office in accordance with
the filing instructions.  Others, however, were the one-page form
that some banks continue to use.  IRS officials do not believe that
any were copies of suspicious CTRs. 


   ARIZONA HAS A SEPARATE FORM FOR
   REPORTING SUSPICIOUS
   TRANSACTIONS
---------------------------------------------------------- Chapter 2:4

Before federal agencies developed forms for reporting suspicious
transactions, Arizona was using its own form.  In 1985 the Arizona
Attorney General's Office developed a voluntary, informal reporting
system relating to possible money laundering activity through
financial institutions.  Suspected money laundering and suspicious
transactions were to be reported to the state Attorney General on a
one-page form that requested identifying information concerning the
customer and the nature of the transaction.  By 1990 the state was
receiving approximately 150 reports of suspicious transactions a
month. 

In 1991 a state law was passed requiring any state or federally
chartered institution to file with the state copies of various
reports made to the Department of the Treasury.  The state law also
provides that the timely filing of a report with the appropriate
federal agency shall be deemed compliance with the state requirements
if such reports are already being supplied to the state.  Arizona had
been receiving copies of suspicious CTRs filed by state financial
institutions since August 1989.\2

Consequently, financial institutions were excused from filing copies
of suspicious CTRs but were required to file copies of CRFs.  The
state Attorney General's Office accepted a copy of a CRF filed in
lieu of the state form. 

Officials with the Arizona Attorney General's Office told us that in
1993 the state received an average of 300 reports of suspicious
transactions a month, not including those copies of suspicious CTRs
received on computer tape from IRS.  About two-thirds of the reports
were copies of CRFs filed with financial regulatory agencies.  The
remaining reports were made on the state form, which, we were told,
some financial institutions used for situations they felt did not
warrant a CRF. 


--------------------
\2 Arizona is one of several states that have agreements with
Treasury to receive copies of all CTRs filed within the state on
computer tapes from the IRS Detroit Computing Center.  See Money
Laundering:  State Efforts To Fight It Are Increasing But More
Federal Help Is Needed (GAO/GGD-93-1, Oct.  15, 1992). 


REPORTS OF SUSPICIOUS TRANSACTIONS
ARE NOT CONSISTENTLY USED OR
MANAGED
============================================================ Chapter 3

Information concerning suspicious transactions can be an effective
means of identifying a wide variety of criminal activity.  Even so,
use of the information by law enforcement at federal and state levels
is limited and inconsistent. 

No federal agency has been designated as responsible for developing
and administering a program that would manage these resources with a
focused, nationwide perspective.  Although the Internal Revenue
Service is the primary recipient of the reports, the use of the
reports is a local initiative and varies among offices.  Several
states have recognized the value of the reports but their ability to
use the information also differs because access to the data varies
among the states. 


   IRS HAS HAD SUCCESS IN USING
   REPORTS OF SUSPICIOUS
   TRANSACTIONS
---------------------------------------------------------- Chapter 3:1

As discussed in chapter 2, district offices of IRS' Criminal
Investigation Division receive reports of suspicious transactions in
several different ways.  CID agents we spoke with at both the
headquarters and district levels described suspicious transactions
reports from financial institutions as extremely valuable
intelligence leads.  IRS does not keep records or data to measure the
value of the reports.  However, agents we spoke with at the field
level related numerous examples of major investigations that had been
initiated on the basis of suspicious transaction reports made by
financial institutions.  These examples include the following: 

  In March 1994 a Texas funeral director was indicted along with
     three other individuals in U.S.  District Court on charges that
     they accepted $4.9 million in drug proceeds during a 5-week
     period in 1989.  The investigation originated when a banker
     became suspicious of large cash deposits being made into the
     account of the funeral home and telephoned the CID district
     office in Dallas. 

  In June 1994 a technical engineer with the Bureau of Engraving and
     Printing in Washington, D.C., was arrested and charged with the
     theft of $1.7 million worth of newly printed hundred-dollar
     bills.  Tellers at a bank in Annapolis, Maryland, became
     suspicious and telephoned the CID district office in Baltimore
     after the individual made several deposits just under the
     $10,000 reporting threshold. 

  In November 1993 the San Francisco CID district office received a
     Criminal Referral Form regarding possible structuring of
     deposits in order to avoid having a CTR filed.  On the basis of
     a subsequent investigation by CID and the U.S.  Postal Service,
     a Post Office employee has been charged with embezzling over
     $600,000 from the Postal Service over the past several years. 

  Telephone calls from two different banks to the Richmond, Virginia,
     CID district office reporting that an individual was purchasing
     cashier's checks with cash in amounts just under $10,000
     resulted in a major narcotics ring being exposed.  Eventually,
     14 individuals were convicted and over $1.5 million worth of
     cash, vehicles, and real estate was seized. 

  The Houston, Texas, CID district office received a report from a
     bank that an individual had deposited more than $12 million in
     cash during a 4-month period claiming that the money was to be
     used to open a chain of 13 stores to sell beauty and clothing
     products.  A subsequent investigation by IRS and the Drug
     Enforcement Administration resulted in the indictment of four
     individuals for trafficking in cocaine. 

  In New York City, a 2-year investigation by several federal law
     enforcement agencies resulted in the indictment in September
     1994 of 30 grocery store owners accused of food stamp fraud. 
     The case was initiated on the basis of a suspicious CTR. 

  A telephone call from a bank to CID agents in Oklahoma City began a
     joint investigation that, 2 years later, led to the seizure of
     over 26 pounds of heroin at that city's airport.  The estimated
     value of the drugs was $20 million.  The suspicious transaction
     originally reported involved two individuals using cash to
     purchase cashier's checks for less than $10,000. 


   IRS' MANAGEMENT AND USE OF
   SUSPICIOUS TRANSACTION REPORTS
   VARY AMONG DISTRICT OFFICES
---------------------------------------------------------- Chapter 3:2

Reports of suspicious transactions are a source of intelligence data
for CID special agents throughout IRS.  As discussed above, districts
have used the reports to initiate a number of major investigations. 
However, the reports are not managed from an agencywide perspective. 
The extent to which agents in IRS' 35 CID district offices solicit,
process, evaluate, and use the reports is up to the discretion of the
district CID chief and varies from one district to another.  As a
result, IRS cannot be certain the reports are being used to their
full potential throughout the agency. 

There are no IRS procedures or policies as to how suspicious
transaction reports are to be managed at the district level.  The CID
Investigative Handbook offers only the following guidance:  "The
[district CID chief] should consider designating specific special
agents to be responsible for responding to financial institutions
that provide information on suspicious currency transactions and for
evaluating the information received to determine if a criminal
investigation is warranted."

During our review, CID management at the national office surveyed the
35 district CID offices to determine local policies regarding the
receipt and evaluation of suspicious transaction reports.  The
results of the survey indicated that the districts differ
significantly as to the level of effort spent evaluating the reports
and the amount of emphasis given the initiative by district
management. 

CID officials told us that some districts place much more emphasis on
agents establishing a close, working relationship with financial
institutions than do other districts.  In these districts, for
example, one agent is designated to spend much of his or her time
personally contacting financial institutions and trade associations
to explain the importance of the suspicious transaction reports.  We
were told that, typically, the institutions in these districts will
often call the agent personally even before a report is prepared. 

According to CID officials, many of the districts maintain a
localized computer database of every report that is received.  This
database is then checked for prior reports when newly received
reports are evaluated.  Not all of the districts maintain such a
database, however, so that IRS does not know how many reports have
been received nationwide.  Without this information, IRS cannot
assess the management of the reports from an agencywide perspective. 

The CID districts also differ on how individual reports are
evaluated.  We were told that some districts assign the reports to
agents who decide if further investigation is warranted on the basis
of the information in the report.  Other districts have a policy of
researching every report against databases both internal and external
to IRS before deciding if an investigation should be opened. 

The districts vary widely on the role the reports play in the
initiation of investigations.  From October 1990 to June 1994 CID
district offices initiated over 21,000 cases.  On an agencywide
basis, about 4 percent of the cases were initiated as a result of
reports of suspicious transactions.  Among individual districts,
however, the rate varied from 0 to over 18 percent.  CID officials
said that they did not know why the rates varied.  In our opinion,
the variance in the rates is an indication that the reports could be
receiving different amounts of emphasis among the districts.  Table
3.1 shows the rates for all of the CID district offices. 



                          Table 3.1
           
             IRS CID Investigations Initiated by
           Suspicious Transaction Reports, October
                   1990 Through June 1994.

                                           Initiated
                                   Total          by
                             investigati  suspicious
                                     ons  transactio  Percen
CID district office            initiated    n report       t
---------------------------  -----------  ----------  ------
Puerto Rico\a                         74          14   18.92
Dallas, TX                           867          99   11.42
Sacramento, CA\b                     344          39   11.34
San Francisco, CA                    258          27   10.47
Oklahoma City, OK                    420          42   10.00
Manhattan, NY                      1,016          90    8.86
Las Vegas, NV                        269          20    7.43
Laguna Niguel, CA                    513          37    7.21
Anchorage, AK\b                      135           8    5.93
\Brooklyn, NY                        785          43    5.48
Wilmington, DE\b                      55           3    5.45
San Jose, CA\b                       306          16    5.23
Cleveland, OH                        613          31    5.06
Philadelphia, PA                     604          30    4.97
Los Angeles, CA                      652          32    4.91
Seattle, WA                          332          16    4.82
Ft. Lauderdale, FL                   727          35    4.81
Richmond, VA                         582          28    4.81
Newark, NJ                           693          29    4.18
Baltimore, MD                        458          19    4.15
Phoenix, AZ                          389          14    3.60
Denver, CO                           620          22    3.55
Honolulu, HI\b                       114           4    3.51
Pittsburgh, PA                       363          12    3.31
Greensboro, NC                       754          23    3.05
Indianapolis, IN                     455          12    2.64
Buffalo, NY                          410          10    2.44
Boise, ID\b                           86           2    2.33
Hartford, CT                         387           9    2.33
St. Louis, MO                        499          11    2.20
St. Paul, MN                         365           8    2.19
Boston, MA                           481          10    2.08
Parkersburg, WV\b                    155           3    1.94
Louisville, KY\b                     207           4    1.93
Des Moines, IA\b                     108           2    1.85
Atlanta, GA                          877          16    1.82
Austin, TX                           831          14    1.68
Jacksonville, FL                     663          10    1.51
New Orleans, LA                      418           6    1.44
Detroit, MI                          767          11    1.43
Wichita, KS\b                        148           2    1.35
Houston, TX                          480           6    1.25
Portland, OR\b                       187           2    1.07
Chicago, IL                          698           7    1.00
Nashville, TN                        610           5    0.82
Cincinnati, OH\b                     343           2    0.58
Omaha, NE\b                           95           0    0.00
Aberdeen, SD\b                        28           0    0.00
Milwaukee, WI                        266           0    0.00
============================================================
Totals                            21,507         885    4.11
------------------------------------------------------------
\a In addition to conducting investigations in the Commonwealth, the
Puerto Rico office is also responsible for conducting investigations
overseas. 

\b These CID district offices were closed prior to October 1993 and
their operations consolidated into the remaining 35 CID district
offices. 

Source:  GAO analysis of IRS records. 

CID officials told us that the majority of CID district offices share
suspicious transaction reports with the Examination function in IRS. 
Under these procedures, if CID does not initiate a criminal
investigation on a report, the information will be passed on to tax
examiners to use in identifying tax fraud.  IRS does not keep records
on how useful suspicious transaction reports have been in this
regard. 


   USE OF REPORTS OF SUSPICIOUS
   TRANSACTIONS BY THE STATES
   VARIES SUBSTANTIALLY
---------------------------------------------------------- Chapter 3:3

Several states have recognized the value of suspicious transaction
reports.  The type of report these states receive, however, differs
among the states so that the information available for state law
enforcement agencies to work with varies considerably.  Moreover, no
state has access to the reports on the same basis as do federal
authorities. 


      MOST STATES HAVE LIMITED USE
      OF SUSPICIOUS CTRS ON FILE
      WITH IRS
-------------------------------------------------------- Chapter 3:3.1

In July 1993 Treasury Department officials announced the initiation
of "Project Gateway," a program that would allow authorized personnel
in every state direct access to the database containing all of the
Currency Transaction Reports, including those marked suspicious, at
IRS' Detroit Computing Center.  Under the program, authorized
personnel in each state would be able to access the data through
computer terminals linked to the Center.  As of November 1994, 47
states as well as the District of Columbia had entered into
agreements with the Department of the Treasury to participate in
Project Gateway, and a total of 40 states had already begun
operations.  Treasury officials told us that agreements with the
remaining 3 states were in the final stages of negotiation. 

Although access to the data is now direct, states are limited as to
what CTRs--including those marked suspicious--can be accessed.  Under
Project Gateway, state analysts must use a specific name to search
the database and can access only those reports filed on the
individual or business named.  Consequently, states can use the data
only on a reactive basis--that is, when they already have the name of
a suspect.  They cannot use the data on a proactive basis, as CID is
able to, for targeting individuals for investigation on the basis of
suspicious transaction reports having been filed. 


      SOME STATES HAVE EXPANDED
      USE OF SUSPICIOUS CTRS AND
      OTHER TYPES OF SUSPICIOUS
      TRANSACTION REPORTS
-------------------------------------------------------- Chapter 3:3.2

In response to our survey, 15 states said that they require financial
institutions to report suspicious transactions that might involve
money laundering.  Nine of these states said they use the information
to initiate criminal investigations. 

   Figure 3.1:  States with
   Suspicious Transaction
   Reporting Requirements

   (See figure in printed
   edition.)

Five of the 15 states that require suspicious transaction
reporting--Colorado, Connecticut, Idaho, Indiana, and Oklahoma--said
they require financial institutions to file a copy of any Criminal
Referral Form filed with the federal regulatory agencies.  Officials
in these states told us that the primary reason for receiving copies
of the CRFs is to monitor reports of criminal activity occurring
within the institutions.  They said that they do not use the reports
of suspicious customer transactions as a basis for initiating
criminal investigations. 

As mentioned in chapter 2, several states have agreements with
Treasury that allow them to receive copies of all CTRs filed within
the state on computer tapes from the IRS Detroit Computing Center. 
Six states--Arizona, California, Florida, Illinois, New York, and
Texas--are currently receiving CTRs, including those marked
suspicious, on computer tape.\1 The use of suspicious CTRs by these
states varies. 

In Arizona, as previously discussed, the Attorney General's Office
also receives reports of suspicious transactions on CRFs as well as
on the state's own form.  State officials told us that all of the
reports are entered into the state's own database and used on both a
reactive and proactive basis. 

Florida is somewhat similar to Arizona in that it requires
state-chartered banks to forward copies of CRFs filed to the state
banking department.  Florida officials said that suspicious CTRs
received from IRS are put on a state database and used on a reactive
basis.  The CRFs, however, are researched and sent to local law
enforcement agencies for further investigation at their discretion. 

New York officials said that they also receive copies of CRFs from
state-chartered financial institutions.  These are evaluated along
with suspicious CTRs and those reports of suspicious transactions
that merit further attention are routed to the appropriate law
enforcement agency. 

California does not require financial institutions to send copies of
CRFs to the state.  However, the state is receiving photocopies of
the special reports provided by California financial institutions to
IRS' CID in the western region (see p.  p.  23).  California enters
these reports onto a database along with the suspicious CTRs it
receives from the Detroit Computing Center.  All of the suspicious
transaction reports are used on a both a reactive and proactive basis
by the state. 

Illinois and Texas officials told us that neither state receives
copies of CRFs.  However, both receive CTRs on magnetic tape from
IRS.  According to state officials, each state removes those marked
suspicious and researches and evaluates them.  The resulting leads
are sent to law enforcement units in the field for further
investigation at their discretion. 

Other states receive copies of suspicious CTRs from sources other
than IRS.  Although Georgia, Nebraska, and Utah do not receive CTRs
on computer tapes from IRS, each uses suspicious CTRs to some extent
to target individuals for further investigation.  Each of these
states has a law requiring banks to provide the state with copies of
CTRs filed with IRS.  In addition to receiving copies of all CTRs
filed, Georgia requires financial institutions to telefax copies of
those CTRs marked suspicious to the state banking department. 
Although Nebraska does not have the capability to process CTRs filed
on magnetic media, the state police receive copies from financial
institutions of those filed on paper and review them for those marked
suspicious.  Similarly, an analyst with the Utah state police scans
all CTRs received to identify those marked suspicious.  Law
enforcement officials from each of these three states told us that
the reports are reviewed and evaluated and, where warranted, sent to
field units for further investigation. 


--------------------
\1 Although Maryland used to receive CTRs on computer tapes from IRS,
it no longer does so. 


   RECENT INITIATIVES WOULD
   SIGNIFICANTLY CHANGE HOW
   SUSPICIOUS TRANSACTION REPORTS
   ARE MADE AND USED
---------------------------------------------------------- Chapter 3:4

Concurrent with our review, the Department of the Treasury and the
financial institution regulatory agencies were in the process of
reviewing various aspects of the federal government's efforts to
combat money laundering.  Similarly, as discussed above, IRS'
Criminal Investigation Division had initiated a survey of how
suspicious transaction reports are used and managed at the district
office level.  By December 1994, as we were preparing this report,
these efforts had resulted in a number of proposals and agreements
that could have a substantial impact on suspicious transaction
reporting by financial institutions. 

For the past several years, a group known as the Interagency Bank
Fraud Working Group\2 has been attempting to consolidate the six
separate CRF forms being used onto a single, standardized form that
would be filed with a single recipient.  As previously discussed,
financial institutions use the forms to report several types of
criminal activity, including suspected money laundering and/or
attempts to evade currency reporting requirements.  Under current
procedures, the institution filing the CRF is also responsible for
sending copies of the form to a number of regulatory and law
enforcement agencies.  The purpose for consolidating the forms and
designating a single recipient was to ease the reporting burden on
the financial institutions and to place responsibility for ensuring
correct dissemination of the reports with the government rather than
with the reporting institution. 

In August 1991 the six regulatory agencies signed a Memorandum of
Understanding with Treasury's Financial Crimes Enforcement Network
(FinCEN)\3 that authorized FinCEN to design, develop, implement, and
maintain a computerized database containing the standardized CRFs. 
Under the agreement, financial institutions would file CRFs directly
with FinCEN. 

In the interim, the Department of the Treasury, in conjunction with a
requirement in 1992 legislation, formed the Bank Secrecy Act Advisory
Group composed of 30 individuals from various state and federal
agencies as well as the private sector.  The Advisory Group, which
first met in April 1994, was charged with assessing all of the
reporting and recordkeeping requirements of the act as well as other
facets of the government's efforts to combat money laundering.  One
of the issues discussed during the three meetings held in 1994 was
how to facilitate the reporting of suspicious transactions by
financial institutions. 

In December 1994, as we were preparing this report, we were informed
by representatives from FinCEN, the Bank Fraud Working Group, and the
Bank Secrecy Act Advisory Group that the following agreements had
been reached: 

  The "suspicious transaction" block would be removed from the
     Currency Transaction Report and the form would no longer be used
     to report suspicious transactions.  This action had been taken
     as part of a general effort to simplify the form by reducing the
     amount of information to be reported on the form. 

  A standardized version of the Criminal Referral Form was being
     prepared that could be filed either on paper or electronically. 
     The filing instructions for the form would specify that only one
     form would be filed, with FinCEN, rather than copies sent to
     various federal agencies. 

  IRS' Detroit Computing Center would provide processing services for
     the new CRF and also develop and maintain a centralized database
     of the reports.  FinCEN would serve as database administrator
     and assure that the appropriate federal law enforcement agencies
     have access to the CRF database.  CRFs reporting suspected Bank
     Secrecy Act violations and/or money laundering would be made
     available to the appropriate district offices of IRS' Criminal
     Investigation Division. 

  FinCEN was exploring the feasibility of making available to the
     states those CRFs reporting money laundering and/or Bank Secrecy
     Act violations.  The reports would be made available to the
     states on the same basis as state access to the reports required
     by the Bank Secrecy Act. 

  The database containing the consolidated CRF would be fully
     operational by September 1995. 

Also in December 1994 we were informed by officials of IRS' Criminal
Investigation Division that procedures were being prepared to address
how suspicious transaction reports were to be managed at the district
level.  IRS officials said that these procedures would be
incorporated into the CID Investigative Handbook and would help
ensure consistent treatment and use of the reports.  Among the areas
to be emphasized were the importance of

  developing and maintaining a working relationship with financial
     institutions,

  promptly evaluating the reports received, and

  performing a minimum level of additional research on the reports. 


--------------------
\2 Members of the Group include representatives from the six
financial regulatory agencies as well as the Departments of Justice
and the Treasury, the Federal Bureau of Investigation, and the Secret
Service. 

\3 FinCEN is a relatively small Treasury agency that was established
in April 1990 to support law enforcement agencies by analyzing and
coordinating financial intelligence. 


   CONCLUSIONS
---------------------------------------------------------- Chapter 3:5

Financial institutions are in a unique position to assist law
enforcement at the federal and state levels by reporting suspicious
transactions that might indicate money laundering.  Reports of
suspicious transactions have led to the initiation of a number of
major investigations dealing with a wide range of criminal activity. 
However, the lack of overall direction and control over the reporting
of suspicious transactions has led to a situation where reports are
filed with different agencies on different forms that vary as to the
amount of useful information they contain. 

Although IRS has successfully used the reports to initiate a number
of investigations, the management of--and emphasis given-- the
information varies among district offices.  IRS has no agencywide
policies or procedures regarding how best to solicit, process, and
utilize the information.  Because IRS cannot be certain the
information is used and managed consistently, it has no assurance
that the information is being used to its full potential throughout
the Service. 

Several states have recognized the value of suspicious transaction
reports as a criminal intelligence resource.  However, use of the
information by these states is limited compared to federal
authorities because the type of information available to the states
differs. 

Recent agreements and proposals made by the Department of the
Treasury, IRS, and others are an indication that the problems
associated with how suspicious transactions are reported are being
addressed.  We believe that the actions planned, if properly
implemented in a timely manner, will do much to provide for the
consistent and centralized management of the reports that has been
lacking. 


   AGENCY AND INDUSTRY COMMENTS
---------------------------------------------------------- Chapter 3:6

A draft of this report was provided to the American Bankers
Association, FinCEN, and IRS for comment.  The Association provided
written comments on the report (see app.  I) in which it said that it
believes financial institutions have an excellent record of
cooperating with law enforcement on the reporting of possible
violations of law.  It added that this cooperation should improve
even more with the anticipated changes in suspicious transaction
reporting discussed in this report because bankers will be better
equipped to focus on reporting potential criminal violations rather
than routine transactions. 

FinCEN provided written comments (see app.  II) stating that it found
the report to be comprehensive and accurate. 

IRS also provided written comments on the report (see app.  III) and
said that it generally agreed with the report's findings.  The
comments noted that, although CID should be allowed maximum
flexibility in the use of its resources, national guidelines are
being developed to ensure consistency in the evaluation and
processing of suspicious transaction reports.  IRS also noted that
changes are being made to a CID management information system that
will enable CID to better ensure the proper use of suspicious
transaction reports and track its accomplishments in the area. 

IRS did take exception with a statement in the executive summary of
the report that describes the use of the IRS database of CTRs and
suspicious CTRs as being reactive.  IRS did not believe that the
statement recognizes the proactive value of the data in identifying
new targets or initiating new investigations.  In clarifying these
comments with IRS officials, we were informed that, although the word
"database" was used, IRS was actually referring to the individual
suspicious CTRs and not the computer database on which they are
maintained. 

It was not our intention to portray suspicious CTRs as not having any
proactive value.  The statement in question refers specifically to
the database and not to the individual reports on the database.  As
noted in chapter 2 (see p.  17), IRS procedures call for staff at the
Detroit Computing Center to distribute copies of CTRs that have been
marked suspicious to the appropriate CID district offices on a weekly
basis.  However, as we point out in chapter 3 (see p.  27), the
extent to which these suspicious CTRs--as well as suspicious
transactions reported on Criminal Referral Forms--are used
proactively is up to the discretion of the district CID chief. 




(See figure in printed edition.)Appendix I
COMMENTS FROM THE AMERICAN BANKERS
ASSOCIATION
============================================================ Chapter 3




(See figure in printed edition.)Appendix II
COMMENTS FROM THE FINANCIAL CRIMES
ENFORCEMENT NETWORK
============================================================ Chapter 3




(See figure in printed edition.)Appendix III
COMMENTS FROM THE INTERNAL REVENUE
SERVICE
============================================================ Chapter 3

supplementing those in the report text appear at the end of this
appendix. 



(See figure in printed edition.)


The following is GAO's comment on the Internal Revenue Service's
letter dated April 26, 1995. 


   GAO COMMENT
---------------------------------------------------------- Chapter 3:7

As explained in the report (see p.  28), the percentages of cases
initiated are based on all reports of suspicious transactions no
matter what form was used to make the report.  Because IRS is not
able to determine if the reports were made on a Currency Transaction
Report or on a Criminal Referral Form, there is no way to quantify
the proactive value of suspicious Currency Transaction Reports. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix IV


   GENERAL GOVERNMENT DIVISION,
   WASHINGTON, D.C. 
-------------------------------------------------------- Appendix IV:1

Darryl Dutton, Assistant Director
Michael L.  Eid, Senior Evaluator
Chas.  Michael Johnson, Evaluator
Donna M.  Leiss, Communications Analyst
Michelle D.  Wiggins, Secretary