Retail Payments Issues: Experience With Electronic Check Presentment
(Letter Report, 07/14/98, GAO/GGD-98-145).

Pursuant to a congressional request, GAO reviewed the current role of
electronic check presentment (ECP) in the collection process for
interbank checks, focusing on: (1) identifying and describing the ECP
services offered to U.S. banks; (2) determining the ECP volume in the
United States for 1995 through 1997; (3) determining whether ECP affects
the length of time that it takes for a dishonored check to be returned
to the depositary bank; (4) identifying any factors that may limit ECP
use; and (5) determining how ECP may affect banks' risk of check fraud.

GAO noted that: (1) the Federal Reserve System is the leading ECP
provider in the United States; it offers three ECP services nationwide;
(2) for these services, the collecting banks send the paper checks to
the appropriate Federal Reserve check office, which then presents the
checks electronically to paying banks; (3) these ECP services differ in
how the paper checks are handled after they have been electronically
presented to paying banks; (4) delivery of the magnetic ink character
recognition (MICR) line data serves as presentment, and the checks are
truncated at the Federal Reserve check office serving the paying bank;
(5) ECP volume accounts for a small, but growing, percentage of the
overall U.S. interbank check volume; (6) from 1995 to 1997, ECP volume
increased 114 percent; (7) GAO's analysis of return items that were
initially electronically presented suggests that use of the Federal
Reserve ECP services may not have a substantial effect on the percentage
of dishonored local checks that are returned to depositary banks within
the 2-day hold period; (8) for these ECP services, the percentage of
local checks returned within 2 days during this period was only
marginally higher than the paper-check presentment method; (9)
interviews with regulatory and banking officials identified several
factors that deter banks from accepting the electronic presentment of
checks, including: (a) the concern that ECP may increase a paying bank's
vulnerability to check fraud; (b) the lack of a clear economic incentive
to use electronic presentment; and (c) a perceived consumer preference
for receiving cancelled checks; (10) in addition, certain types of state
laws have been identified as having the effect of impeding ECP because
of the laws' reliance on paper checks; (11) while ECP may allow paying
banks to identify checks that might not be honored sooner and deter
certain types of check fraud, banking officials expressed concerns that
ECP could make paying banks more vulnerable to other types of check
fraud; (12) officials at five large banks told GAO that forged
signatures and endorsements, along with counterfeit checks, have created
the highest check fraud losses in the period since 1995; and (13)
because the receipt of the MICR line data does not provide a paying bank
with adequate information for identifying forged signatures, banking
officials said they continue to insist on paper checks.

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

 REPORTNUM:  GGD-98-145
     TITLE:  Retail Payments Issues: Experience With Electronic Check 
             Presentment
      DATE:  07/14/98
   SUBJECT:  Electronic funds transfer
             Electronic data interchange
             Clearinghouses (banking)
             Bank deposits
             Check disbursement or control
             Lending institutions
             Fraud

             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


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


Report to the Subcommittee on Financial Institutions and Consumer
Credit, Committee on Banking and Financial Services, House of
Representatives

July 1998

RETAIL PAYMENTS ISSUES -
EXPERIENCE WITH ELECTRONIC CHECK
PRESENTMENT

GAO/GGD-98-145

Electronic Check Presentment

(233528)


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

  ABA - American Bankers Association
  Basic - Basic MICR Presentment Service
  ECCHO - Electronic Check Clearing House Organization
  ECCS - Electronic Check Clearing Service
  ECP - Electronic Check Presentment
  EFAA - Expedited Funds Availability Act
  FRBB - Federal Reserve Bank of Boston
  FRBM - Federal Reserve Bank of Minneapolis
  MICR - Magnetic Ink Character Recognition
  MICRPlus - MICR Presentment Plus Service
  NYCHA - New York Clearing House Association
  Truncation - Check Truncation Service
  UCC - Uniform Commercial Code

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


B-277713

July 14, 1998

The Honorable Margaret Roukema
Chairwoman
The Honorable Bruce Vento
Ranking Minority Member
Subcommittee on Financial Institutions
 and Consumer Credit
Committee on Banking and Financial Services
House of Representatives

The Expedited Funds Availability Act (EFAA), enacted in 1987, limits
the time that banks\1 can hold funds deposited into customer accounts
before the funds are to be made available for withdrawal.  In 1996,
the Board of Governors of the Federal Reserve System (Federal Reserve
Board) recommended that the maximum permissible "hold period" for
local checks be extended from 2 to 3 business days after the day of
deposit to give banks adequate time to identify fraudulent checks
that should be dishonored.\2 Consumer groups disagreed with that
recommendation and countered that giving banks an additional day
would create hardships for consumers.  The 2-day hold period for
local checks remains in effect. 

Concerns about both check fraud risk and depositors' access to funds
have heightened interest in how technology can help the
check-collection process by potentially speeding up the return of
checks dishonored due to insufficient funds, fraud, or other reasons,
while continuing to allow ready access to deposited funds.  According
to the Federal Reserve Board, the experience of depositary banks (the
first bank at which a check is deposited) in receiving returned
checks is one of the most significant factors to consider in
assessing uniform time schedules under which bank deposits should be
made available to depositors. 

You requested that we provide information about the current role of
electronic check presentment (ECP) in the collection process for
"interbank checks"--checks for which the depositary bank and the
paying bank are not the same entity.  With ECP, the check is
presented electronically to the paying bank by transmission of the
magnetic ink character recognition (MICR) line data at the bottom of
the check, which identify the routing number of the paying bank, the
amount of the check, the number of the check, and the account number
of the customer.  With ECP, the paying bank must decide whether to
honor the check, or have the check returned to the depositary bank,
upon receipt of the electronic data rather than the paper check. 

As agreed with your offices, our objectives were to (1) identify and
describe the ECP services offered to U.S.  banks; (2) determine the
ECP volume in the United States for 1995 through 1997; (3) determine
whether ECP affects the length of time that it takes for a dishonored
check to be returned to the depositary bank; (4) identify any factors
that may limit ECP use; and (5) determine how ECP may affect banks'
risk of check fraud. 


--------------------
\1 In this report, the term "bank" includes any depository financial
institution, such as a commercial bank or savings bank. 

\2 Checks are "dishonored" for reasons that include being written on
a closed account or one with insufficient funds to cover the check,
showing evidence of check counterfeiting or forged signatures, or
other evidence of fraud. 


   BACKGROUND
------------------------------------------------------------ Letter :1

In 1996, an estimated 64 billion checks were written in the United
States.  Approximately 30 to 35 percent of these checks were "on-us"
checks--that is, checks for which the collecting bank (any bank
except the paying bank handling the item for collection) and the
paying bank (the bank on which the check is drawn) are the same
entity.  On-us checks never enter the interbank check-collection
process.  The remaining 65 to 70 percent (about 45 billion checks in
1996) of the checks written were interbank checks.\3

Interbank checks are cleared and settled through an elaborate
check-collection process.  The check-collection process includes,
among other steps, check presentment and the final settlement of
checks.\4 Check presentment occurs when the checks are delivered to
the paying banks for payment and the paying banks must decide whether
to honor or return the checks.  Final settlement of checks occurs
when the collecting banks are credited and the paying banks are
debited, usually through accounts held at either Federal Reserve
Banks or correspondent banks.\5

In the check-collection process, depositary banks generally sort
deposited checks by destination and dispatch them for collection. 
Depositary banks can physically present paper checks to paying banks
through several methods: 

  -- direct presentment of the paper checks to the paying banks;

  -- exchange of paper checks at clearing houses in which the
     depositary banks and the paying banks are members;

  -- presentment of the paper checks through intermediaries, such as
     correspondent banks or Federal Reserve check offices; or

  -- some combination of the above methods, such as depositary banks
     using correspondent banks to collect checks and the
     correspondents collecting those checks through local clearing
     houses or Federal Reserve check offices. 

When paying banks decide not to pay certain checks, they must return
the dishonored checks to the depositary banks.\6 Under the Uniform
Commercial Code (UCC), the paying bank generally has until midnight
of the day following presentment ("midnight deadline") to return
dishonored checks or send notices of dishonor.  (The UCC and other
laws and regulations governing the check-collection processes are
described below.) Banks use this period to decide whether or not to
pay checks.  Dishonored checks are returned for several reasons, such
as insufficient funds in the check writer's account, issuance of a
stop-payment order on the check, or because the check was written on
a closed account.  Banks vary considerably in their policies for
making pay/no pay decisions and may view these decisions as a matter
of customer service.  Thus, even if banks have earlier information
indicating that some checks may have to be returned, using the entire
period permitted by the UCC may enable the bank to honor the check. 
The paying banks may return dishonored checks, commonly referred to
as return items, directly to the depositary banks, through clearing
house associations, or to any returning bank (a bank handling a
returned check), including the Federal Reserve Banks. 

Although most paper checks are physically presented to the paying
banks, checks can be presented electronically by agreement with
paying banks.  When checks are electronically presented, the MICR
data are electronically transmitted by the presenting bank to the
paying banks and the receipt of these data constitutes presentment,
rather than the paying banks' receipt of the paper checks.  The paper
checks may be sent to the paying banks at a later time or may be
"truncated" or held, at some point in the collection process,
depending upon the terms of the applicable electronic presentment
agreement.  When checks are truncated, the check writers do not
receive the cancelled checks.  Checks can be truncated at depositary
banks or at intermediary banks, such as Federal Reserve Banks or
correspondent banks. 

The check-collection process is regulated by a complex system of laws
and regulations.  The primary laws affecting checks are Articles 3
and 4 of the UCC, as adopted in each state; EFAA; and the Federal
Reserve Board's Regulations CC and J.  The UCC is a set of model laws
that govern commercial and financial activities.  Efforts have been
made to encourage each of the 50 states to enact UCC articles in a
uniform manner.  The UCC's Articles 3 and 4 govern negotiable
instruments, including checks and bank deposits, and collections,
respectively.  Article 4 was revised in 1990, in part, to promote the
efficiency of the check-collection process by making the provisions
of Article 4 more compatible with the needs of an automated system
and to facilitate the adoption of programs allowing for the
presentment of checks by electronic transmission of information. 
Although the UCC allows for the use of electronic check presentment,
paying banks still have the legal right to insist on paper
presentment.  EFAA, which is implemented through Regulation CC,
limits the time that banks can hold funds deposited into customer
accounts before these funds must be made available for withdrawal. 
Regulation CC, which preempts the UCC or other state law to the
extent that either is inconsistent with Regulation CC, also governs
the collection and return of checks.  Regulation J governs checks
collected through the Federal Reserve System. 

EFAA and Regulation CC establish the specific funds availability
schedules for checks.  Generally, funds are to be made available for
withdrawal within 2 business days after the day of deposit for local
checks and within 5 business days for nonlocal checks.  Local checks,
in general, are those for which the depositary bank and the paying
bank are located in the same Federal Reserve check-processing region. 
Nonlocal checks, in general, are those for which the depositary bank
and the paying bank are located in different Federal Reserve
check-processing regions. 

EFAA and Regulation CC also require that the first $100 of a
depositor's aggregated checks and certain other types of checks, such
as Treasury, state and local government, or cashier's checks, are to
be made available for withdrawal by the next business day after the
day of deposit.  Under EFAA and Regulation CC, banks are permitted to
hold checks longer for, among other things, checks over $5,000 or
suspicious checks.  In certain circumstances, such as when banks have
reasonable cause to believe that certain checks are uncollectible or
when an account is new, the banks may delay the availability of
funds.  In such a case, the banks must notify the customers, explain
the delay, and indicate when the funds will be available. 


--------------------
\3 The figure used for the interbank check market includes share
drafts drawn on credit unions. 

\4 This process is described in greater detail in Payments,
Clearance, and Settlement:  A Guide to the Systems, Risks, and Issues
(GAO/GGD-97-73, June 20, 1997). 

\5 A correspondent bank is a bank that holds deposits owned by other
banks and performs banking services, such as check collection. 

\6 In providing check clearing services, Federal Reserve Banks are
collecting banks. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

The Federal Reserve System is the leading ECP provider in the United
States with broadly available services; it offers three ECP services
nationwide.  For these services, the collecting banks send the paper
checks to the appropriate Federal Reserve check office, which then
presents the checks electronically to paying banks.  These ECP
services differ in how the paper checks are handled after they have
been electronically presented to paying banks.  For example, the
Federal Reserve's basic ECP service provides the paper checks to the
paying bank on the same day as presentment.  However, with another
Federal Reserve ECP service, checks are not delivered to the paying
bank.  Delivery of the MICR line data serves as presentment, and the
checks are truncated at the Federal Reserve check office serving the
paying bank.  In addition to the three nationwide services, the
Federal Reserve Bank of Minneapolis (FRBM) offers the Electronic
Check Clearing Service (ECCS) to banks located in the Upper Peninsula
of Michigan.  The fundamental characteristic of ECCS that
distinguishes it from other ECP services is that depositary banks
electronically transmit the MICR data to the FRBM, rather than
transporting the paper checks. 

ECP volume accounts for a small, but growing, percentage of the
overall U.S.  interbank check volume.  From 1995 to 1997, ECP volume
increased 114 percent.  In 1997, ECP volume was about 2.2 billion
checks, almost 5 percent of the estimated U.S.  interbank check
volume.  The 2.2 billion ECP checks represented approximately 14
percent of the almost 16 billion checks that the Federal Reserve
collected in 1997.\7

Our analysis of a limited sample of return items that were initially
electronically presented suggests that use of the three Federal
Reserve ECP services that were nationally available may not have a
substantial effect on the percentage of dishonored local checks that
are returned to depositary banks within the 2-day hold period.  For
these three ECP services, the percentage of local checks returned
within 2 days during this period at the Federal Reserve check offices
was only marginally higher than the percentage returned within 2 days
using the paper-check presentment method. 

For ECCS, the percentage of local checks returned within the 2-day
hold period was substantially higher than the percentage of paper
checks that were physically presented and returned.  This comparison
of ECCS and paper-presentment methods showed that substantial
improvements in check-return performance occurred for those banks
when the paper checks were held at the depositary bank until the
checks were finally paid.  Given the small sample size of this
comparison and the particular conditions of the Upper Peninsula of
Michigan (such as the remote location of banks and a relatively low
check volume), it is not possible to know whether these improvements
could be achieved in other geographical areas or nationwide. 

Our interviews with regulatory and banking officials identified
several factors that, in their views, deter banks from accepting the
electronic presentment of checks.  These factors include (1) the
concern that ECP may increase a paying bank's vulnerability to check
fraud, (2) the lack of a clear economic incentive to use electronic
presentment, and (3) a perceived consumer preference for receiving
cancelled checks.  In addition, certain types of state laws have been
identified as having the effect of impeding ECP because of the laws'
reliance on paper checks. 

While ECP may allow paying banks to identify checks that might not be
honored sooner and deter certain types of check fraud, banking
officials we talked with expressed concerns that ECP could also make
paying banks more vulnerable to other frequently occurring types of
check fraud.  In interviews and written responses, officials at five
large banks told us that forged signatures and endorsements, along
with counterfeit checks, have created the highest check fraud losses
in the period since 1995.  Because the receipt of the MICR line data
does not provide a paying bank with information viewed as adequate
for identifying forged signatures, these banking officials said they
have continued to insist on paper presentment. 


--------------------
\7 Since the Federal Reserve check offices process checks through
reader sorters from which they maintain an accounting of the
processed checks, the volume figures for the Federal Reserve's check
collection services should be considered exact. 


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

To identify and describe the ECP services offered to U.S.  banks, we
interviewed officials from the Federal Reserve Board and FRBM and
reviewed Federal Reserve Board documents that described the Federal
Reserve's ECP services.  We also interviewed officials at the New
York Clearing House Association (NYCHA), the Electronic Check
Clearing House Organization (ECCHO), and the American Bankers
Association (ABA). 

To determine the volume of checks that were electronically presented
in the United States from 1995 to 1997, we collected data from the
Federal Reserve Banks on their ECP services.  To calculate the
percentage of the ECP volume in the U.S.  interbank check-collection
market, we used the number of interbank checks cited in the January
1998 report by the Committee on the Federal Reserve in the Payments
Mechanism.\8 The number of interbank checks is an estimate because
the actual number of checks that are cleared through clearing houses
and correspondent banks, and by direct presentment, as well as the
number of on-us checks, is unknown. 

To determine how ECP affects the length of time it takes for a
dishonored check to be returned to a depositary bank, we used data
from a survey coordinated by the Federal Reserve Board and conducted
by four Federal Reserve offices.  At our request, the Federal Reserve
collected information on the return cycle time for a small sample of
items that were initially presented by either paper check or
electronic transmission and that were returned as dishonored through
four Federal Reserve check offices.  This information was collected
for a sample of 2,258 return items, primarily during the week of
January 12 through 16, 1998.\9 The four check offices were selected
because they had higher-than-average ECP volumes.  Also, the Federal
Reserve Board applied the same ECP return item survey to FRBM's
additional ECP service, ECCS, which is offered only to banks located
in the Upper Peninsula of Michigan.  For comparison purposes, we also
collected comparable check-return data from 10 judgmentally selected
banks in the Upper Peninsula of Michigan that do not participate in
ECCS. 

Because the sizes of the samples were small and not selected on a
random basis, we could not make any statistical generalizations
regarding the entire banking industry or the Federal Reserve's ECP
services on the basis of these surveys.  Since the samples were
selected during an 1-week period at 4 Federal Reserve check offices,
we could not ascertain whether our results would differ if the
samples were taken over a longer period of time or collected at more
of the Federal Reserve's 45 check offices.  Furthermore, our results
may not fully reflect differences among paper presentment and ECP
services because the checks are not assigned to these processes on a
random basis, but rather on the basis of each paying bank's
decisions.  It is reasonable to expect that paying banks make
decisions to accept electronic presentment based on a number of
criteria, such as price and availability of services.  Thus, the
returned checks collected from the different presentment methods may
well have different characteristics; these could, in turn, affect the
time required for their return if they are dishonored by the paying
bank.  This nonrandom selection process limited our ability to
determine whether ECP services available nationwide had any effect on
check return cycles. 

To determine potential legal and operational impediments to ECP, we
interviewed officials at several large banks, the Federal Reserve
Board, the Federal Reserve Banks of Minneapolis and Boston, and
NYCHA.  We asked these officials how certain provisions of the UCC's
Articles 3 and 4, EFAA, and the Federal Reserve Board's Regulation CC
and other issues may impede increased ECP use. 

To determine how ECP may affect a bank's risk of check fraud, we
interviewed officials at the Federal Reserve Bank of Boston (FRBB),
NYCHA, and three large commercial banks in New York City.  In
addition, we sent a survey of written questions regarding ECP and
check fraud risk to six large commercial banks that are members of
ECCHO.  Five of the six banks responded to our survey questions. 

We obtained written comments on a draft of this report from the
Federal Reserve Board.  Their comments are discussed near the end of
this report and are reprinted in appendix II.  We did our work from
July 1997 to April 1998 in Washington, D.C.; Minneapolis, MN; and New
York City in accordance with generally accepted government auditing
standards. 


--------------------
\8 The Federal Reserve in the Payments Mechanism, Committee on the
Federal Reserve in the Payments Mechanism, Federal Reserve System,
January 1998. 

\9 See appendix I for a more detailed discussion on the methodology
of the ECP return item survey. 


   THE FEDERAL RESERVE OFFERS ECP
   SERVICES NATIONWIDE, WHILE THE
   PRIVATE CHECK CLEARING HOUSES
   OFFER ELECTRONIC CHECK
   INFORMATION SERVICES
------------------------------------------------------------ Letter :4

The Federal Reserve Banks offer banks three ECP services:  basic MICR
presentment (Basic), MICR presentment plus (MICR Plus), and check
truncation (Truncation).  In addition, FRBM offers ECCS to banks
located in the Upper Peninsula of Michigan.  For all four ECP
services, Federal Reserve check offices transmit the electronic data
to the paying banks and receipt of these data constitutes
presentment.  Under the three nationally available ECP services, the
collecting banks deposit checks with the Federal Reserve check
offices, but the three ECP services differ in how the Federal Reserve
check offices handle the paper checks after they have been
electronically presented to paying banks.  The return process for
Basic items is the same process used to return checks that were
physically presented.  The other two nationally available ECP
services, MICR Plus and Truncation, use different processes to return
checks than those used to return physically presented checks. 

The fundamental characteristic of ECCS, which distinguishes it from
other Federal Reserve ECP services, is that banks using the service
do not deliver paper checks to FRBM; instead, they electronically
transmit the checks' MICR data to FRBM.  Since depositary banks keep
the deposited checks until the checks are finally paid, they can pull
any paper checks that are dishonored by the paying banks.  Under
ECCS, depositary banks receive notices of dishonored checks from
paying banks through FRBM. 

Table 1 lists the four Federal Reserve ECP services and explains (1)
how collecting banks deposit checks with the Federal Reserve check
offices and (2) how paper checks and return items are handled after
electronic presentment. 



                                     Table 1
                     
                     Descriptions of the Four Federal Reserve
                                   ECP Services

            How banks collect checks
ECP         using Federal Reserve       How paper checks and return items are
service     check offices\a             handled after electronic presentment
----------  --------------------------  ----------------------------------------
Basic       Collecting bank deposits    Federal Reserve check office delivers
            the paper checks with the   paper checks to the paying bank on the
            Federal Reserve check       same day as electronic presentment.
            office.                     Paying bank returns items as it would
                                        under paper presentment.

MICR Plus   Collecting bank deposits    Federal Reserve check office serving the
            the paper checks with the   paying bank holds paper checks for 1-3
            Federal Reserve check       days after electronic presentment and
            office.                     pulls and delivers return items from
                                        stored checks at the paying bank's
                                        request. After 1-3 days, the remaining
                                        paper checks are delivered to the paying
                                        bank.

Truncation  Collecting bank deposits    Federal Reserve check office serving the
            the paper checks with the   paying bank holds the paper checks after
            Federal Reserve check       electronic presentment and pulls and
            office.                     delivers the return items from the
                                        stored checks at the paying bank's
                                        request. Generally, after 60 days, the
                                        paper checks are destroyed.\b An image
                                        of each paper check (front and back) is
                                        stored at the Federal Reserve check
                                        office for 7 years.

ECCS\c      Collecting bank provides    The depositary bank stores the paper
            MICR data electronically    checks for 1 day. The paying bank
            to FRBM.                    returns items electronically through the
                                        FRBM to the depositary bank, which pulls
                                        the checks. The depositary bank delivers
                                        the remaining paper checks to the FRBM.
                                        FRBM sorts the paper checks by the
                                        paying bank and either delivers the
                                        checks to the paying bank or truncates
                                        the checks.
--------------------------------------------------------------------------------
\a The collection of checks (ECP or otherwise) may involve two
Federal Reserve check offices. 

\b Some physical checks are retained longer as established by
agreement between the paying bank and the Federal Reserve Bank. 

\c FRBM offers ECCS to banks located in the Upper Peninsula of
Michigan. 

Source:  Federal Reserve Board. 


      PRIVATE CHECK CLEARING
      HOUSES' ELECTRONIC SERVICES
      DO NOT CONSTITUTE
      PRESENTMENT
---------------------------------------------------------- Letter :4.1

Private clearing houses offer electronic check information services
to banks.  These services transmit the MICR data to the paying banks,
but they do not constitute electronic presentment of checks.\10
Industry officials we interviewed noted that, while agreements
between or among banks that would enable them to present checks
directly and electronically are possible under the UCC, few such
agreements exist.  NYCHA offers electronic check informational
services to banks in which the MICR data are transmitted to paying
banks with the paper presentment to follow.  Banks perceive some
advantages to receiving MICR data before the paper checks are
physically presented.  One banking official told us that electronic
transmission of MICR data, used as an informational service, may
assist a paying bank with its cash management services or allow the
bank to debit a customer's account earlier than it otherwise could. 
Such information may also provide for earlier identification of
checks that cannot be paid.  For instance, checks written on closed
accounts cannot be paid and thus are returned to the depositary bank. 
NYCHA estimated that in 1997 its members electronically transmitted
data on 53 million checks, or approximately 19 percent of the total
volume of checks exchanged at NYCHA.  Additionally, several
commercial banks formed ECCHO in 1989 to coordinate the use of
electronic check information among banks and to design standards for
the use of electronic check information.  In 1997, ECCHO reported
that its members transmitted data on 567 million checks.\11


--------------------
\10 In addition to ECP products, the Federal Reserve Banks also offer
an electronic check information product.  This product is similar to
the electronic check information products offered in the private
sector.  In 1997, the Reserve Banks' MICR information volume totaled
1.4 billion items. 

\11 According to NYCHA officials, ECCHO's report of 567 million
checks may include check data transmitted by NYCHA members. 


   ECP REPRESENTED A SMALL BUT
   GROWING PERCENTAGE OF U.S. 
   INTERBANK CHECK VOLUME IN 1997
------------------------------------------------------------ Letter :5

In 1997, 2.2 billion checks were electronically presented in the
United States--almost 5 percent of the estimated U.S.  interbank
check volume (about 45 billion checks).  According to a recent
Federal Reserve report, the Federal Reserve Banks accounted for about
35 percent (about 16 billion checks) of the total interbank
check-collection market in 1996.  The remaining 65 percent of the
market (about 29 billion checks) were presented directly or through
private, local clearing houses.  The 2.2 billion checks
electronically presented by the Federal Reserve in 1997 represented
14 percent of the total number of checks the Federal Reserve
collected that year. 

Electronically presented checks as a percentage of the Federal
Reserve's overall check collection volume grew from 1995 to 1997. 
During this period, the Federal Reserve's ECP volume increased 114
percent, from slightly more than 1 billion checks to 2.2 billion
checks.  From 1996 to 1997, the Federal Reserve's ECP volume
increased 56 percent, from 1.4 billion to 2.2 billion checks. 

Although all 12 Federal Reserve Banks offered ECP services, ECP
volume was concentrated primarily in the southern and midwestern
Federal Reserve Districts.  As shown in table 2, the check offices
comprising the Federal Reserve Bank of Atlanta electronically
presented the highest volume of checks (420 million).  The Federal
Reserve Bank of Boston electronically presented the lowest number of
checks among the Federal Reserve Districts in 1997 (46 million). 



                                Table 2
                
                The Federal Reserve ECP Volume in 1997,
                      by Federal Reserve District

                                                                   ECP
                                         ECP       Total    percentage
                                    volume\a      checks      of total
                                   (millions   collected        checks
Federal Reserve District                   )  (millions)     collected
---------------------------------  ---------  ----------  ------------
Atlanta                                  420       2,366         17.8%
Kansas City                              316       1,161          27.2
Cleveland                                234       1,104          21.2
St. Louis                                222         903          24.6
Minneapolis                              203         973          20.9
Dallas                                   192       1,205          15.9
Richmond                                 169       1,732           9.8
San Francisco                            164       2,030           8.1
Chicago                                  113       1,688           6.8
New York                                 113       1,226           9.2
Philadelphia                              49         718           6.9
Boston                                    46         862           5.3
======================================================================
System total                           2,241      15,949         14.0%
----------------------------------------------------------------------
\a ECP volume for each Federal Reserve District includes Basic, MICR
Plus, and Truncation.  The volume for Minneapolis also includes ECCS. 

Source:  Federal Reserve Board data. 

The most commonly used Federal Reserve ECP service was Basic
presentment.  In 1997, about 1.5 billion checks were presented using
the Basic presentment service.  Truncation was used to present about
558 million checks, and MICR Plus was used to present about 204
million checks. 


   A LIMITED SAMPLE OF ECP RETURN
   ITEMS SUGGESTED THAT ECP
   SERVICES AVAILABLE NATIONALLY
   HAD LITTLE EFFECT ON THE
   PERCENTAGES OF CHECKS RETURNED
   WITHIN PERMISSIBLE HOLD PERIODS
------------------------------------------------------------ Letter :6

Our analysis of a limited sample of returned checks collected in four
Federal Reserve check offices during the week of January 12 through
16, 1998, indicated that the use of ECP services that are available
nationwide did not have a substantial effect on the percentage of
dishonored local checks returned to depositary banks within the 2-day
hold period.  On the basis of our analysis of the 1-week sample, the
percentage of dishonored checks initially presented through Basic,
MICR Plus, and Truncation services and then returned to depositary
banks within the 2-day hold period was only marginally higher than
the percentage of returned checks under paper presentment.  However,
our analysis did show that the percentage of local checks returned
within the 2-day hold period using ECCS was substantially higher than
the comparable percentage of physically presented paper checks.  Our
comparison of ECCS and paper-check presentment methods indicated that
substantial improvements in check-return performance occurred when
the paper check was held at the depositary bank until the expiration
of the "midnight deadline" specified in the UCC.  Given the small
sample size of this comparison and the particular conditions of the
Upper Peninsula of Michigan (severe winter weather conditions and the
low volume of checks electronically presented through ECCS), however,
it is not possible to know whether these improvements could be
achieved in other settings or nationwide. 


      RESULTS SHOWED MARGINALLY
      BETTER LOCAL CHECK-RETURN
      PERFORMANCE FOR ECP SERVICES
      COMPARED WITH PAPER-CHECK
      PRESENTMENT
---------------------------------------------------------- Letter :6.1

Our analysis of a 1-week sample of returned checks indicated that the
percentages of dishonored local checks returned within the holding
period were only marginally higher for checks electronically
presented through Basic, MICR Plus, and Truncation services compared
with checks presented in paper form.  Almost 15 percent of the
electronically presented local checks in our sample were returned
within 2 business days after they were deposited compared with only 6
percent of the physically presented local paper checks. 



                                     Table 3
                     
                      Dishonored Local Checks in Sample That
                       Were Returned Within 2 or 3 Business
                           Days, By Presentment Method

                                        Dishonored local checks in
                                          sample returned within
                                      ------------------------------
                                        2 business      3 business
                                           days            days
                                      --------------  --------------
                                                                           Total
                                                                          sample
                                              Percen          Percen       items
Presentment method                    Number       t  Number       t    reviewed
------------------------------------  ------  ------  ------  ------  ----------
ECP services offered nationwide
--------------------------------------------------------------------------------
Basic                                     53     14%     311     82%         381
MICR Plus                                  0       0     228      85         269
Truncation                               122      22     472      87         544
Paper                                     17       6     201      69         291
--------------------------------------------------------------------------------
Source:  Federal Reserve Board. 

The majority of the electronically presented checks were not received
by the depositary bank until the third business day.  (See table 3.)
Truncation produced the best 2-day cumulative percentage of the
electronically presented checks:  22 percent.  However, at 3 business
days, paper-check presentment's cumulative percentage was 69 percent,
which was less than the cumulative percentages of Basic, MICR Plus,
and Truncation services at 82 percent, 85 percent, and 87 percent,
respectively. 

According to the Federal Reserve Board, the time required to return
checks handled by the Federal Reserve likely represents the upper
bound of overall check-return cycle times.  In a Federal Reserve
Board survey, Federal Reserve Banks indicated that, in 1995, they
delivered about 15 percent of local returned checks that they
processed to the depositary bank by the second business day.\12 This
check-return performance is low compared with the bank average
reported by the Board in the same survey.  Depositary banks received
about 48 percent of all local returned checks within 2 business days,
according to the survey.  The differences in these check-return
performances appear to be an indicator of the significance of
depositary banks' direct presentment of paper checks to paying banks
and exchanges of paper checks at clearing houses.  The Federal
Reserve Board said that checks returned through one or more
intermediary banks, such as Federal Reserve Banks, were not received
by depositary banks as quickly as checks returned directly to
depositary banks by paying banks.  Federal Reserve Board officials
said that the Federal Reserve's lower percentage of returned checks
delivered within the 2-day hold period reflects, in part, the Federal
Reserve's obligation to provide services to handle checks that cannot
be easily processed by direct presentment or through a clearing
house. 


--------------------
\12 Report to the Congress on Funds Availability Schedules and Check
Fraud at Depository Institutions, Board of Governors of the Federal
Reserve System, October 1996, p.  5. 


      RESULTS SHOWED BANKS USING
      ECCS HAD A BETTER
      CHECK-RETURN PERFORMANCE
      THAN BANKS PRESENTING PAPER
      CHECKS
---------------------------------------------------------- Letter :6.2

We found that ECCS, the ECP service available only to banks in
Michigan's Upper Peninsula, provided a better check-return
performance than the paper-check presentment method used by banks in
the same area.  The basis for this finding was a comparison of the
percentage of returned checks for which ECCS banks received a return
notification within 2 business days after the checks were deposited
and the percentage of returned checks received by paper presenting
banks within the same 2-day period.  (As previously noted, depositary
banks using ECCS keep deposited checks until the expiration of the
"midnight deadline" specified in the UCC and pull returned checks
when they receive a return notice through FRBM from a paying bank.)

This comparison of the ECCS and paper-check presentment methods
indicated that substantial reductions in return cycle times are
possible when the paper check is held at the depositary bank.  For
the week of January 12 through 16, 1998, the 8 banks that used ECCS
received a return notification on 131 local dishonored checks, while
the 10 judgmentally selected Upper Peninsula banks, which constituted
our paper presentment comparison group, received 276 dishonored local
checks.  We found that the return notifications for the 131 ECCS
dishonored checks were provided to depositary banks within 2 business
days after the checks were deposited.  (See table 4.) Moreover, 82
percent of the 131 return notices were received by depositary banks
within 1 business day after the checks were deposited.  In contrast,
only about 8 percent of the 276 dishonored checks that were
physically presented were returned to depositary banks within 2
business days after the checks were deposited. 



                                     Table 4
                     
                      Dishonored Local Checks in Sample That
                       Were Returned Within 2 or 3 Business
                      Days, by Presentment Method Within the
                                 Upper Peninsula

                                       Dishonored local checks in
                                         sample returned within
                                     ------------------------------
                                       2 business      3 business
                                          days            days
                                     --------------  --------------
                                                                           Total
                                                                          sample
                                             Percen          Percen        items
Presentment method                   Number       t  Number       t     reviewed
-----------------------------------  ------  ------  ------  ------  -----------
ECCS (ECP service offered only in       131    100%     N/A     N/A          131
 the Upper Peninsula of Michigan)
Paper                                    21       8     113      41          276
--------------------------------------------------------------------------------
Legend

N/A = Not applicable. 

Source:  Federal Reserve Board data and our analysis of data from 10
banks that did not participate in ECCS. 

Conditions in the Upper Peninsula region, such as relatively low
check volume and banks that are remotely located, and other factors
may well have played a role in showing ECCS to its best advantage in
comparison with paper presentment.  As a result, it is not likely
that the savings in the time between ECCS and paper presentment may
be achieved in the banking industry as a whole.  First, several
regional conditions complicated the return of dishonored checks
within the 2-day hold period for local checks using the paper
presentment process.  These conditions included severe winter weather
conditions, the distances between the banks and the most commonly
used private check clearing house and the Federal Reserve check
office in Minneapolis, and the two time zones dividing the banks in
the Upper Peninsula area. 

In addition, the composition of ECCS participating banks and the low
volume of checks electronically presented through ECCS may have
contributed to its better performance.  Banks participating in ECCS
had to store a fairly limited number of checks and retrieve a small
number of return items.  All of the ECCS participating banks were
small, with assets ranging from $52 million to $350 million.  In
1997, ECCS volume was nearly 1.9 million checks.  This volume
constituted about 0.2 percent of the 973 million checks that the FRBM
check office collected in 1997. 


      RESULTS SUGGESTED SIMILAR
      NONLOCAL CHECK RETURN
      PERFORMANCE FOR ALL CHECK
      PRESENTMENT METHODS
---------------------------------------------------------- Letter :6.3

In our analysis, we also compared return times for electronic and
paper presentment of nonlocal checks.  Our data showed that the
check-return performance against the funds availability schedules of
electronically presented nonlocal checks was not very different from
that of paper-presentment checks.  Again, we evaluated the return
cycle times of nonlocal checks using the applicable availability
schedule specified in EFAA for nonlocal checks--that is, the
5-business-day funds availability.  We found that a very high
percentage of dishonored checks were returned to depositary banks
within 5 business days after the day the checks were deposited,
regardless of whether the checks were electronically or physically
presented.  Approximately 90 percent of the electronically presented
nonlocal checks were returned within 5 business days to depositary
banks, while 89 percent of the physically presented checks were
returned within 5 business days. 

Our data showed that a considerable number of electronically
presented nonlocal checks were returned to depositary banks within 3
business days after the checks were deposited.  As shown in table 5,
43 percent of the checks presented through the Basic service and 35
percent of the checks presented through the Truncation service were
returned to depositary banks within 3 business days.  This compares
with 14 percent of the nonlocal checks returned under paper
presentment. 



                                     Table 5
                     
                       Dishonored Nonlocal Checks in Sample
                       That Were Returned Within 3, 4, or 5
                       Business Days, by Presentment Method
                                    Nationwide

                      Dishonored nonlocal checks in sample returned
                                          within
                      ----------------------------------------------
                        3 business      4 business      5 business
                           days            days            days
                      --------------  --------------  --------------
                                                                           Total
                                                                          sample
                              Percen          Percen          Percen       items
Presentment method    Number       t  Number       t  Number       t    reviewed
--------------------  ------  ------  ------  ------  ------  ------  ----------
ECP services offered nationwide
--------------------------------------------------------------------------------
Basic                     81     43%     146     78%     176     94%         188
MICR Plus                  0       0      32      43      61      82          74
Truncation               105      35     210      71     272      92         297
Paper check               12      14      52      63      74      89          83
--------------------------------------------------------------------------------
Source:  Federal Reserve Board data. 


   SEVERAL FACTORS CURRENTLY LIMIT
   ECP USE
------------------------------------------------------------ Letter :7

While some observers believe that ECP may potentially contribute to a
more efficient check-collection process, several factors were
mentioned as currently limiting ECP use.  As previously noted,
although the UCC permits the use of electronic presentment agreements
between the presenting bank (the last bank in the check-collection
process demanding payment from the paying bank) and paying banks, few
such agreements have been negotiated, and private clearing houses do
not currently offer ECP services.  In interviews with regulatory
officials and bankers, we identified several factors that may play a
role in discouraging banks from agreeing to accept an electronically
presented check, including

  -- the lack of a clear economic incentive to use electronic
     presentment;

  -- laws and regulations that require maintaining cancelled checks
     in certain situations;

  -- a perceived consumer preference for receiving cancelled checks
     with monthly bank statements and state laws requiring that
     cancelled checks be offered;

  -- UCC and Regulation CC's requirement that paying banks generally
     must physically return a dishonored check to depositary banks;

  -- operating and business factors that limit banks' adoption of
     ECP; and

  -- a concern that ECP may increase banks' vulnerability to forged
     signatures and counterfeit checks. 

While there are current impediments to expanded ECP use, the Federal
Reserve Banks and some check-collection service providers are working
to create a more favorable environment for ECP growth, including
possible ways to promote ECP use and establishing standards for using
ECP. 


      BANKS LACK A CLEAR ECONOMIC
      INCENTIVE TO USE ELECTRONIC
      PRESENTMENT
---------------------------------------------------------- Letter :7.1

For a bank, deciding on the relative advantages and disadvantages of
ECP use requires weighing the potential gains in its role as a
depositary bank against the potential losses in its role as a paying
bank.  In its January 1998 report, the Committee on the Federal
Reserve in the Payments Mechanism concluded that paying banks have
little incentive to speed up the presentment process because they
receive benefits from "float." Float occurs in the check-collection
process because of the time it takes to process a check--that is, the
time between when the check is accepted for payment and when the
paying bank deducts the amount of the check from the check writer's
account.  Since the paying bank has use of the funds during that
time, it has little incentive to speed up the process.  When a bank
is the depositary bank, however, it has a clear incentive to collect
the check--and receive payment--as quickly as possible and at the
lowest cost. 


      LAWS AND REGULATIONS REQUIRE
      THE RETENTION OF THE
      CANCELLED CHECK IN SPECIFIC
      SITUATIONS
---------------------------------------------------------- Letter :7.2

Many states have laws or regulations that require certain
individuals, businesses, and organizations to maintain cancelled
checks.  Cancelled checks also serve as documentation for federal and
state tax authorities and as proof of payment for certain categories
of commercial transactions.  The Federal Reserve Bank of Boston
(FRBB) compiled a survey (dated 1997) of federal and state laws and
regulations that relate to the retention of an original cancelled
check.  On the basis of our analysis of the survey, 41 states, plus
the District of Columbia, had at least 1 law or regulation that
required individuals or organizations to retain their cancelled
checks for various purposes, including documentation for state and
local governments and by certain businesses.  The remaining nine
states--Alaska, Georgia, Iowa, Maryland, Michigan, Nebraska, Oregon,
West Virginia, and Utah--were not listed as having any laws or
regulations that required the retention of cancelled checks. 

An example, cited in the survey, of a state law requiring the
retention of cancelled checks was a Florida law that provides, in
general, that county officials are required to retain cancelled
checks as a part of the permanent record of the applicable office. 
Another example was a California regulation requiring investment
advisors to retain cancelled checks for at least 6 years--during the
first 2 years of which, the cancelled checks are to be held in a
location that is easily accessible. 


      CUSTOMER PREFERENCE AND
      STATE LAWS REGARDING
      CANCELLED CHECKS
---------------------------------------------------------- Letter :7.3

The perceived consumer preference for receiving cancelled checks is
another operational deterrent to ECP use.  If consumers are to
receive their cancelled checks in their monthly statements, banks
cannot use check truncation anywhere in the collection process. 
Federal Reserve Board officials and banking officials with whom we
spoke expressed a belief that many consumers want their cancelled
checks returned. 

Additionally, two states have laws that guarantee bank consumers the
right to receive cancelled checks.  According to the FRBB's survey,
New York and Massachusetts have such laws.  A New York law requires,
in general, that banking institutions offering consumer accounts
offer a consumer account through which cancelled checks are returned
to the customer with a periodic statement of the account.  In
Massachusetts, a state law requires that, if depositors request their
cancelled checks, the bank must return the cancelled checks without
charging a fee. 


      REGULATION CC REQUIRES THAT
      A DISHONORED CHECK BE
      PHYSICALLY RETURNED
---------------------------------------------------------- Letter :7.4

Regulation CC, which governs funds availability and the collection
and return of checks, requires that paying banks must physically
return the original dishonored check to the depositary bank if it is
available.\13 This requirement applies to all checks, regardless of
whether they were electronically or physically presented.  The
Federal Reserve ECP services were structured so that paper checks can
be returned to depositary banks. 

Federal Reserve Board officials told us that Regulation CC could be
amended to permit banks to return dishonored checks using images in
lieu of the paper check but that before Regulation CC could be
modified, certain operational and legal issues would need to be
resolved. 


--------------------
\13 While Regulation CC provides that, in certain circumstances, a
bank may send a copy or notice in place of a check, such notice in
lieu of return is permitted only when a bank does not have and cannot
obtain possession of the check or must retain possession of the check
for protest.  Regulation CC commentary provides that a check is not
unavailable for return if it is merely difficult to retrieve from a
filing system or from storage by a keeper of checks in a truncation
system. 


      BUSINESS FACTORS COULD
      IMPEDE ECP GROWTH
---------------------------------------------------------- Letter :7.5

On the basis of our interviews with banking officials, we also
identified business factors that could play a role in discouraging
banks from choosing electronic check presentment and truncation,
including the current concentration of resources on ensuring that
banks' computers are able to handle dates after the year 2000 and the
potential for depositary banks to secure access to information on the
paying banks' customers. 

Banks must ensure their computer systems are able to handle dates in
2000 and beyond.  However, many banks may not have sufficient
resources to both ensure that their computers are able to handle this
date change and make necessary investments to use ECP.  Officials of
large banks told us that they have allocated significant resources
toward ensuring that their computers are able to handle these dates. 
As a result, any significant changes in check clearing technology are
not likely to be considered until after 2000 in many banks. 


      BANKS ARE CONCERNED THAT ECP
      USE MAY INCREASE THEIR
      VULNERABILITY TO FORGED
      SIGNATURES AND COUNTERFEIT
      CHECKS
---------------------------------------------------------- Letter :7.6

From our interviews and from written responses from banking officials
at five large banks, we learned that these banks do not consider MICR
data adequate information for detecting certain types of fraud,
particularly forged signatures and counterfeit checks.  (The
potential effects of ECP use on check fraud are discussed in the next
major section of this report.)


      FUTURE PLANS FOR ECP GROWTH
---------------------------------------------------------- Letter :7.7

Despite the current limitations previously mentioned, the Federal
Reserve Banks, some private organizations, and several large banks in
the United States are working to create a more favorable environment
for ECP growth.  Two of these efforts include (1) the establishment
of a working group comprising officials from banks, the Federal
Reserve Banks, and check collection service providers and (2) the
expected creation of the first multilateral ECP organization, that
is, an ECP organization established by an agreement involving more
than two banks. 

In its January 1998 report, the Committee on the Federal Reserve in
the Payments Mechanism concluded that "it is not yet clear" whether
the whole check system would benefit from moving toward ECP and
truncation.  The report recommended that a working group comprising
banks, the Federal Reserve, and other check-collection process
participants should be convened to determine the cost and feasibility
of further ECP use and truncation.  The report outlined possible
steps that the Federal Reserve might take if this working group
concludes that such a coordinated move is both feasible and
advisable.  For example, the report stated that the Federal Reserve
could amend Regulation CC to permit banks to return dishonored
checks, or return items, using images or electronic transmissions in
place of the original checks. 

In addition, several commercial banks, with the assistance of NYCHA,
are forming the first national multilateral ECP organization, The
Small Value Payments Company, L.L.C., which was expected to begin
operations in mid-1998.  Thirteen of the largest U.S.  banks are to
become owners in The Small Value Payments Company.  Under current
plans, the organization would be established in stages.  In the
initial stage, banks are to transmit electronic check information
among themselves, but the receipt of the paper check would still
constitute presentment.  According to an NYCHA official, the initial
phase would provide The Small Value Payments Company with time to
ensure that ECP standards are established and that banks are
complying with the standards.  As currently proposed, the final phase
would be realized when the transmission and receipt of the MICR data
constitute presentment and when the checks are truncated at
depositary banks.  The same NYCHA official acknowledged that The
Small Value Payments Company faces obstacles in completing the final
phase of its operation.  One of these obstacles is state laws that
guarantee consumers the right to elect to receive cancelled checks. 
But the NYCHA official predicted that the final phase may be
completed within 4 to 5 years from the initial operating date of The
Small Value Payments Company. 


   BANKERS SAID THAT ECP USE MAKES
   BANKS MORE VULNERABLE TO FORGED
   SIGNATURES AND COUNTERFEIT
   CHECKS
------------------------------------------------------------ Letter :8

In interviews and written responses to our questions, officials at
five commercial banks indicated that forged signatures and
endorsements, as well as counterfeit checks, have created their
highest check fraud losses in the period since 1995.  Under the UCC,
banks may be responsible for recrediting the account of a customer if
the bank charges the account for a check that is not properly
payable, such as a check with a forged signature.  According to these
banking officials, perpetrators of fraud are becoming more
sophisticated in committing check fraud.  A banking official told us
that large banks in New York City and San Francisco face a greater
potential for check fraud than small banks located in rural areas. 
Of the five commercial banks that responded to our written survey
questions, four listed forgeries of either signatures or endorsements
as the type of check fraud that has caused them their highest dollar
losses since 1995.  The fifth bank reported that counterfeit checks
constituted its highest fraud losses for the same period. 

Banking officials, in interviews and written responses to our
questions, informed us that they do not consider the receipt of the
MICR data as providing adequate information for detecting certain
types of fraud, particularly forgeries and counterfeit checks. 
According to one banking official, banks need access to the paper
checks to detect potential forged signatures or counterfeit checks. 
Because the receipt of the MICR data does not provide paying banks
with information viewed as adequate for identifying forged
signatures, these banking officials said they have continued to
insist on paper presentment.  As previously noted, although the UCC
permits the use of electronic presentment agreements between the
presenting bank and the paying bank, very few agreements have been
negotiated. 

A potential alternative to paper checks is a digitized image of
checks.  According to banking officials, transmitting a digitized
image of the paper check along with the MICR data would allow banks
to verify signatures.  Specifically, a digitized check image would
allow paying banks to inspect the image as they would inspect a check
for possible fraudulent signatures. 

However, we were told that both business and technological
considerations currently limit widespread use of digitized check
images.  First, according to one banking official, banks are unlikely
to invest their financial resources in check imaging technology
because customers expect to have their cancelled paper checks
returned.  Second, check images cannot currently be economically
transmitted.  One check image consists of a massive amount of digital
information--that is, about 50,000 bytes of information, compared
with 94 bytes of information for a MICR data transmission.  The
Committee on the Federal Reserve in the Payment Mechanism, in its
January 1998 report, stated that imaging technology would require
both additional investment and ongoing costs for banks.  The
committee also noted in the report that "there is some risk that
changes in technology and the evolution of the retail payments system
will overtake the cost and benefit issues" associated with check
imaging.  For example, while "there is little evidence to suggest
that the volume of checks is likely to drop substantially over the
next several years," check imaging could become obsolete if other
electronic payment methods emerge and lead to a decreased use of
checks. 

ECP use may enhance a bank's ability to identify checks that cannot
be paid, such as checks written on closed accounts, and potentially
deter some types of check fraud.  Fraudulent practices like
"paperhanging" and "check kiting" can be created from closed accounts
and insufficient funds, respectively.\14 If the MICR data were
presented faster to the paying banks than paper checks are, this
action might facilitate the identification of a check written on a
closed account or a check written on an account with insufficient
funds. 

However, a Federal Reserve official and a banking official noted that
both ECP and paper presentment methods have some limits in detecting
some types of check fraud, particularly forged endorsements, and that
it would be difficult to judge which method is superior.  A banking
official noted that forged endorsements often are not detected until
the proper payee of a check notifies the check writer that a payment
is missing. 


--------------------
\14 The term paperhanging refers to checks that are deliberately
written on closed accounts.  Check kiting may take many forms, but it
often involves the writing of checks on two or more banks for the
purpose of fraudulently obtaining interest-free, unauthorized loans. 


   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :9

We requested comments on a draft of this report from the Federal
Reserve Board.  The Board's comments are reprinted in appendix II. 
The Board agreed with the facts as stated in our report.  In
addition, the Board provided technical comments, which we have
incorporated where appropriate. 


---------------------------------------------------------- Letter :9.1

We are sending copies of this report to the Chairman and the Ranking
Minority Member of the House Committee on Banking and Financial
Services; the Chairman and the Ranking Minority Member of the Senate
Committee on Banking, Housing, and Urban Affairs; the Chairman of the
Board of Governors of the Federal Reserve System; the Presidents of
the Federal Reserve Banks of Boston and Minneapolis; and others upon
request. 

This report was prepared under the direction of James M.  McDermott,
Assistant Director, Financial Institutions and Markets Issues.  Other
major contributors are listed in appendix III.  Please contact me on
(202) 512-8678 if you have any questions about this report. 

Susan S.  Westin
Associate Director, Financial Institutions
 and Markets Issues


METHODOLOGY OF THE ECP RETURN ITEM
SURVEY
=========================================================== Appendix I

To determine how ECP use affects the length of time that it takes for
a dishonored check to be returned to a depositary bank, that is, the
return cycle time, we used data from surveys conducted by us and the
Federal Reserve Board.  The survey instruments we and the Federal
Reserve Board used captured the return cycle times and other
characteristics of a small sample of returned checks that were
initially presented by paper check and by electronic transmission. 
The Federal Reserve, at our request, sampled dishonored checks as
they were being returned by the paying bank to four Federal Reserve
check offices (Kansas City, Jacksonville, Minneapolis, and Richmond),
primarily during the week of January 12 through 16, 1998.\15 These
four check offices were selected on the basis of the amount of their
ECP volume and the number of banks that acquire ECP services from the
four Federal Reserve check offices.  Since ECP is a paying bank
service, the samples were drawn at the Federal Reserve office that
served the paying bank. 

Also, the Federal Reserve Board applied the same ECP return item
survey to FRBM's additional ECP service, ECCS, which is offered only
to banks located in the Upper Peninsula of Michigan. 

To collect a sample of paper checks comparable to those ECCS items
obtained from the Minneapolis check office, we judgmentally selected
10 banks located on the Upper Peninsula of Michigan, all of which are
in the Minneapolis check processing region, where ECCS is offered. 
Then, we asked the 10 banks, which did not participate in ECCS, to
record return-time data, using the same survey forms and instructions
used by the Federal Reserve Board, for all checks returned to them
during the study week of January 12 through 16, 1998. 


--------------------
\15 Survey data collected by the Federal Reserve Board included 213
items that were outside of the study period out of a total sample of
2,258 items. 


      RETURN ITEM SAMPLE DESIGN
------------------------------------------------------- Appendix I:0.1

For the Basic, MICR Plus, and Truncation ECP services, the 4 Federal
Reserve check offices were instructed to select 25 return items daily
for each presentment service from a list of return items received
from paying banks and to record the return time data.  In addition,
they were instructed to collect 25 paper items each day during the
5-day study period.  Therefore, the weekly sample size of each
service (Basic, MICR Plus, Truncation, and Paper) was expected to be
125 items at each of the 4 check offices, and the total sample size
for each presentment service was to be 500 items.  For determining
the return cycle time of ECCS, the Minneapolis check office was
instructed to select 75 return items daily, or all items if there
were fewer than 75 returns for any day. 

The Federal Reserve check offices were instructed to decrease the
likelihood of bias in the sample of return items by allocating the
sample across all the cash letters (i.e., a listing of return items)
received from banks of various sizes on a particular day, to the
extent possible. 

The 10 Michigan banks that we selected were instructed to select all
of the return items (i.e., dishonored checks) returned to them, as
depositary banks, on each day of the study week. 

Because the check offices, Michigan banks, and check samples were not
selected on a random basis, we cannot make any statistical
generalizations regarding the entire banking industry, or the Federal
Reserve's ECP services, on the basis of this analysis. 


      SURVEY DATA COLLECTION
------------------------------------------------------- Appendix I:0.2

Return items from 515 paying banks were selected by the 4 Federal
Reserve check offices.  Data from 2,258 return items were recorded by
the Federal Reserve Board, after those items that had been sent
through the collection process more than once had been discarded.  A
depositary bank may potentially send a dishonored check back through
the collection process multiple times before a check is paid.  The
inclusion of those items might have skewed our results because it is
very difficult to determine the endorsement and return dates for
checks that have been presented more than once.  Table I.1 shows the
actual sample size for each presentment service. 



                               Table I.1
                
                Total Size of ECP Return Item Sample, by
                           Presentment Method


                                                        Number
                                                            of
                                                         items
                                                        presen
                                             Number of     ted
                                                return    more   Total
                                            items used    than  sample
Presentment services                          in study    once      \a
------------------------------------------  ----------  ------  ------
ECP services offered nationwide
----------------------------------------------------------------------
Basic                                              569      72     641
MICR Plus                                          343      32     375
Truncation                                         841     143     984
ECCS (ECP service offered only in the              131       0     131
 Upper Peninsula of Michigan)
Paper check                                        374      47     421
======================================================================
Total                                            2,258     294   2,552
----------------------------------------------------------------------
\a Includes both local and nonlocal checks. 

Source:  Federal Reserve Board. 

The number of paper items in the overall sample was lower than the
500 items originally planned.  We were advised by Federal Reserve
Board staff that, because Federal Reserve Bank staff collected the
sample of paper items during the busy processing cycle, they were
generally unable to collect as many paper items as instructed. 

Moreover, the MICR Plus sample was also smaller than originally
planned.  Return items that were initially presented by MICR Plus
were sampled at three of the four Federal Reserve check offices. 
Federal Reserve Board officials advised us that one check office did
not have any MICR Plus customers from which they could take a sample
of items.  However, the other three offices did achieve a total
sample size of 421 MICR Plus return items. 

The actual number of ECCS items received by the FRBM fell below the
sample quota of 75 per day, and all items were selected, resulting in
a total of 131 items for the study week.  Items that were presented
through ECCS were all local paper checks that were not presented more
than once.  Therefore, to compare the return cycle times of checks
that were presented through ECCS with those that were physically
presented, we included only the 276 physically presented local paper
checks that had not been represented by the 10 Michigan banks in our
analysis.  (See table II.2.)



                               Table II.2
                
                Total Size of Upper Michigan Bank Sample
                            for Paper Checks

                                                                 Total
Paper check type                                                sample
-----------------------------------------------------------  ---------
Local                                                              276
Nonlocal                                                            73
Represented                                                        127
======================================================================
Total                                                              476
----------------------------------------------------------------------
Source:  GAO. 

We did not verify that this data-collection process was followed at
the Federal Reserve check offices or at the 10 Michigan banks. 
However, we examined check endorsement and presentment and return
dates for out-of-range values and logical consistency.  We also
resolved, to the extent possible, data coding and entry problems with
respondents at the banks and officials at the Federal Reserve Board. 


      DATA ELEMENTS COLLECTED BY
      THE RETURN-ITEM SURVEY
------------------------------------------------------- Appendix I:0.3

We and the four Federal Reserve check offices used the same data
collection instrument to record return cycle times and other check
characteristics.  This instrument was similar to one developed and
administered by the Federal Reserve Board in previous check return
time surveys.  The ECP return item survey instructed the Federal
Reserve Bank staff and officials at the 10 Michigan banks to collect
certain data elements located on paper checks.  These data elements
included (1) the endorsement date or the date the check was
deposited, (2) the date the check was presented to the paying bank,
and (3) the date the check was returned to the depositary bank.  In
addition, the survey instrument collected the ABA numbers of both the
depositary banks and paying banks so that checks could be categorized
as local or nonlocal.  Also recorded were the reasons the checks were
returned, whether the checks had been represented, and whether
intermediaries were involved in the return process. 


      CALCULATING THE RETURN CYCLE
      TIME
------------------------------------------------------- Appendix I:0.4

To ascertain the return cycle times for both local and nonlocal
checks, we calculated the number of days from the endorsement date
(the date the check was deposited at the depositary bank) to the
return date (the date the dishonored check was received by the
depositary bank).  For the checks sampled at the Federal Reserve
check offices, we added 1 day to the return cycle for nonlocal checks
because the samples were drawn and the final date was recorded at the
Federal Reserve offices that served the paying banks.  Since the
samples were not drawn at the depositary banks, we needed to adjust
the recorded return dates to include the time required for
transporting the return items from the Federal Reserve check offices
to the depositary banks. 

According to the Federal Reserve Board official who supervised the
ECP Return Item Survey data collection, because the samples were
drawn at the Federal Reserve check office serving the paying bank,
any nonlocal return items would have been delivered overnight to the
Federal Reserve check office serving the depositary bank.  The
official acknowledged that there may have been instances in which the
return item took more than 1 day to actually reach the depositary
bank, but these instances would have been rare and nearly impossible
to identify.  We agreed that the Federal Reserve Board's approach of
adding 1 day to the return cycle time for nonlocal checks seemed to
be reasonable, given that the sample could not be drawn at the
Federal Reserve check office serving the depositary bank and that the
understatement of return cycle times would be small. 

For local checks, no adjustments were made to the calculated return
cycles because the same Federal Reserve check office served both the
depositary bank and the paying bank.  Again, we felt that any
understatement of return cycle times would be small. 

Because the dates on which the 10 Michigan banks recorded their check
data also represented the actual return dates for checks so recorded,
no return cycle time adjustment was necessary. 

Return dates for checks were recorded, and cycle times were
calculated, in whole days.  Weekends and holidays were excluded from
the elapsed times.  Some of the return dates fell outside of the
study week.  The 213 such cases in the Federal Reserve data were
included in the analysis, while cases that were outside of the study
period were excluded from the Michigan bank data.  For those return
items with return dates falling on a weekend date, the dates were
adjusted to have been returned on the next business day because
checks cannot be collected or returned on a nonbusiness day, weekend
date, or holiday. 




(See figure in printed edition.)Appendix II
COMMENTS FROM THE BOARD OF
GOVERNORS OF THE FEDERAL RESERVE
SYSTEM
=========================================================== Appendix I


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix III

GENERAL GOVERNMENT DIVISION

Nancy Eibeck, Evaluator-in-Charge
Carl Ramirez, Senior Social Science Analyst
Catherine M.  Hurley, Computer Specialist
Jerry Sandau, Social Science Analyst
Desiree W.  Whipple, Communications Analyst

OFFICE OF THE GENERAL COUNSEL

Geoffrey Hamilton, Senior Attorney


GLOSSARY
=========================================================== Appendix 0


      CLEARING HOUSE ASSOCIATION
------------------------------------------------------- Appendix 0:0.1

A voluntary association formed by banks that establishes a meeting
place for the exchanging of checks drawn on those banks. 


      COLLECTING BANK
------------------------------------------------------- Appendix 0:0.2

Any bank except the paying bank handling a check for collection. 


      CORRESPONDENT BANK
------------------------------------------------------- Appendix 0:0.3

A bank that holds deposits owned by other banks and performs banking
services, such as check collection. 


      DEPOSITARY BANK
------------------------------------------------------- Appendix 0:0.4

The first bank at which a check is deposited. 


      FLOAT
------------------------------------------------------- Appendix 0:0.5

The time between when the check is accepted for payment and when the
paying bank deducts the amount of the check from the check writer's
account. 


      INTERBANK CHECKS
------------------------------------------------------- Appendix 0:0.6

Checks for which the depositary bank and the paying bank are not the
same entity. 


      INTERMEDIARY BANK
------------------------------------------------------- Appendix 0:0.7

A bank other than the depositary or the paying bank to which a check
is transferred in the course of collection. 


      MAGNETIC INK CHARACTER
      RECOGNITION (MICR) LINE
------------------------------------------------------- Appendix 0:0.8

The line at the bottom of a check that identifies the routing number
of the paying bank, the amount of the check, the number of the check,
and the account number of the customer. 


      MIDNIGHT DEADLINE
------------------------------------------------------- Appendix 0:0.9

Midnight on the next banking day following the banking day on which
the paying bank received the check. 


      ON-US CHECKS
------------------------------------------------------ Appendix 0:0.10

Checks for which the collecting bank and the paying bank are the same
entity. 


      PAYING BANK
------------------------------------------------------ Appendix 0:0.11

The bank on which the check is drawn. 


      PRESENTING BANK
------------------------------------------------------ Appendix 0:0.12

The last bank in the check-collection process demanding payment from
the paying bank. 


      RETURNING BANK
------------------------------------------------------ Appendix 0:0.13

A bank handling a returned check. 


*** End of document. ***