Food Assistance: Observations on Reducing Fraud and Abuse in the Food
Stamp Program (Testimony, 04/23/98, GAO/T-RCED-98-167).

GAO discussed fraud and abuse in the Food Stamp Program, focusing on
the: (1) nature and extent of the problem; (2) reasons it often goes
undetected; and (3) ways computerized information can be used to
identify and reduce it.

GAO noted that: (1) fraud and abuse in the Food Stamp Program generally
occurs in the form of either overpayments to food stamp recipients or
trafficking; (2) overpayments occur when ineligible persons are provided
food stamps, as well as when eligible persons are provided more than
they are entitled to receive; (3) overpayments are caused by inadvertent
and intentional errors made by recipients and errors made by state
caseworkers; (4) the latest available information indicates that
overpayments in 1996 totalled about $1.5 billion, or about 7 percent of
the food stamp benefits issued that year; (5) errors also result in
underpayments; in fiscal year 1996, such underpayments totalled about
$518 million; (6) with regard to trafficking, the Department of
Agriculture (USDA) estimated that in 1993 about $815 million in food
stamps, approximately 4 percent of the food stamps issued, were traded
for cash at retail stores; (7) no one knows the extent of trafficking
between individuals before the food stamps are redeemed at authorized
retailers; (8) participation in the Food Stamp Program by ineligible
recipients occurs, and often goes undetected, because the information
used to determine a household's eligibility and benefit amount for the
program is not always accurate; (9) state agencies that administer the
program determine household membership on the basis of unverified
information provided by food stamp applicants; (10) food stamp
trafficking takes place when recipients collaborate with unscrupulous
retailers to convert food stamp benefits to cash or other non-food
items; (11) these retailers make a profit by giving the recipients a
discounted cash payment for the stamps, then redeeming the stamps at
full face value to the government; (12) while USDA has reduced the
overpayment rate in recent years, further reductions could result if
food stamp rolls were matched against computerized information held by
various sources; (13) computer matching provides a cost-effective
mechanism to accurately and independently identify ineligible
participants; (14) some states already conduct data matching programs;
(15) by taking a leading role in promoting the use and sharing of
information among federal and state agencies, USDA can enhance the
states' effectiveness in identifying ineligible participants and
reducing overpayments; and (16) in addition, the congressionally
mandated use of electronic benefit transfers, while not the answer to
eliminating all fraud, has the potential to reduce trafficking by
providing an electronic trail of transactions.

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

 REPORTNUM:  T-RCED-98-167
     TITLE:  Food Assistance: Observations on Reducing Fraud and Abuse 
             in the Food Stamp Program
      DATE:  04/23/98
   SUBJECT:  Program abuses
             Overpayments
             Welfare benefits
             Eligibility determinations
             Federal/state relations
             Food relief programs
             Fraud
             Electronic funds transfer
             Retail facilities
             Computer matching
IDENTIFIER:  Food Stamp Program
             USDA Electronic Benefit Transfer System
             
******************************************************************
** 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


Before the Committee on Agriculture,
Nutrition, and Forestry
U.S.  Senate

For Release
on Delivery
Expected at
9:00 a.m.  EDT
Thursday
April 23, 1998

FOOD ASSISTANCE - OBSERVATIONS ON
REDUCING FRAUD AND ABUSE IN THE
FOOD STAMP PROGRAM

Statement of Robert A.  Robinson, Director
Food and Agriculture Issues,
Resources, Community, and Economic
Development Division

GAO/T-RCED-98-167

GAO/RCED-98-167T


(150288)


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

  EBT -
  FNS -
  USDA -

============================================================ Chapter 0

Mr.  Chairman and Members of the Committee: 

Thank you for the opportunity to present our observations on reducing
fraud and abuse in the Food Stamp Program.  The Food Stamp Program is
one of the nation's largest welfare programs and the largest single
program administered by the U.S.  Department of Agriculture (USDA). 
In fiscal year 1997, over $21 billion in food stamps were provided to
about 23 million recipients,\1 down somewhat from recent years.  The
program is the principal component of the government's food
assistance safety net.  Any program of this magnitude will be
susceptible to fraud and abuse, and the Food Stamp Program is no
exception.  It has been subject to both the participation of
ineligible recipients and the improper use of benefits; however, USDA
has been able to reduce the overpayment error rate in recent years
and is taking actions to address food stamp trafficking--that is,
exchanging food stamps for cash or other non-food items. 

We have reported recently on the improper inclusion of prisoners and
deceased individuals in food stamp households and on the extent of
trafficking.  Today, our discussion of fraud and abuse in the Food
Stamp Program will address the nature and the extent of the problem,
the reasons it often goes undetected, and the ways computerized
information can be used to identify and reduce it. 

In summary, we found the following: 

  -- Fraud and abuse in the Food Stamp Program generally occurs in
     the form of either overpayments to food stamp recipients or
     trafficking.  Overpayments occur when ineligible persons are
     provided food stamps, as well as when eligible persons are
     provided more than they are entitled to receive.  Overpayments
     are caused by inadvertent and intentional errors made by
     recipients and errors made by state caseworkers.  The latest
     available information indicates that overpayments in 1996
     totaled about $1.5 billion, or about 7 percent of the food stamp
     benefits issued that year.  Errors also result in underpayments;
     in fiscal year 1996 such underpayments totalled about $518
     million.  With regard to trafficking, USDA estimated that in
     1993 (the latest year of available data) about $815 million in
     food stamps, approximately 4 percent of the food stamps issued,
     were traded for cash at retail stores.  No one knows the extent
     of trafficking between individuals before the food stamps are
     redeemed at authorized retailers. 

  -- Participation in the Food Stamp Program by ineligible recipients
     occurs, and often goes undetected, because the information used
     to determine a household's eligibility and benefit amount for
     the program is not always accurate.  State agencies that
     administer the program determine household membership primarily
     on the basis of unverified information provided by food stamp
     applicants.  Food stamp trafficking takes place when recipients
     collaborate with unscrupulous retailers to convert food stamp
     benefits to cash or other non-food items.  These retailers make
     a profit by giving the recipients a discounted cash payment for
     the stamps, then redeeming the stamps at full face value to the
     government.  In 1995, we reported that food stamp trafficking
     occurred and/or went undetected because USDA (1) admitted some
     retailers to the program that did not meet eligibility criteria
     because they lacked sufficient time and resources to make
     necessary pre-authorization visits and (2) lacked timely and
     accurate information on retailers' total food sales and food
     stamp redemptions to effectively detect retailers that violate
     program regulations.\2

  -- While USDA has reduced the overpayment rate in recent years,
     further reductions could result if food stamp rolls were matched
     against computerized information held by various sources. 
     Computer matching provides a cost-effective mechanism to
     accurately and independently identify ineligible participants. 
     Some states already conduct data matching programs, such as
     matches with the rolls of other states to find participants
     receiving duplicate benefits.  By taking a leading role in
     promoting the use and sharing of information among federal and
     state agencies, USDA can enhance the states' effectiveness in
     identifying ineligible participants and reducing overpayments. 
     In addition, the congressionally mandated use of electronic
     benefit transfers, while not the answer to eliminating all
     fraud, has the potential to reduce trafficking by providing an
     electronic trail of transactions. 


--------------------
\1 For this testimony, food stamps refers to the benefits provided in
the form of coupons or through electronic benefit transfer. 

\2 Food Assistance:  Reducing Food Stamp Benefit Overpayments and
Trafficking (GAO/RCED-95-198, June 23, 1995). 


   BACKGROUND
---------------------------------------------------------- Chapter 0:1

The Food Stamp Program provides a safety net to the millions of
low-income individuals and families nationwide who do not otherwise
have the means to obtain a healthy diet.  Food stamp benefits are
calculated to ensure that households have the resources needed to
purchase a model diet plan based on the National Academy of Sciences'
Recommended Dietary Allowances.  USDA's Food and Nutrition Service
(FNS) administers the program in partnership with the states, funding
all of the program's benefits and about 50 percent of the states'
administrative costs.  FNS develops program policy and guidance, such
as nationwide criteria for determining who is eligible for assistance
and the amount of benefits recipients are entitled to receive, and
oversees the states' activities.  The states are responsible for the
day-to-day operation of the program, including meeting with
applicants and determining their eligibility and benefit levels. 

Food stamp recipients must use their benefits only to purchase
allowable food products from retail food stores that FNS authorizes
to participate in the program.  Recipients use food stamp coupons or
an electronic benefit transfer (EBT) card to pay for these items. 
EBT systems use the same electronic funds transfer technology that
many grocery stores use for their debit card payment systems.  The
Personal Responsibility and Work Opportunity Reconciliation Act of
1996 mandates that all states implement EBT systems by October 1,
2002, unless USDA waives the requirement.  As of March 1998, 16
states had implemented EBT systems statewide, with all other states
in some earlier stage of implementation.  Collectively, about 40
percent of all food stamp benefits are now delivered through EBT
systems. 


   NATURE AND EXTENT OF FRAUD AND
   ABUSE
---------------------------------------------------------- Chapter 0:2

Fraud and abuse in the Food Stamp Program generally occurs in the
form of either overpayments to food stamp recipients or trafficking. 
Overpayments occur when ineligible persons are provided food stamps,
as well as when eligible persons are provided more than they are
entitled to receive.  In 1996, the latest year for which data are
available, the states overpaid recipients an estimated $1.5 billion,
or 7 percent of the approximately $22 billion in food stamps issued. 
Approximately 57 percent of the overpayments were caused by recipient
errors (36 percent unintentional and 21 percent intentional), and 43
percent were caused by caseworkers' errors.  It should also be noted
that recipient and caseworker errors can result in underpayments. 
According to FNS' data, food stamp recipients were underpaid by about
$518 million in fiscal year 1996. 

In March 1997, we reported on one specific kind of food stamp
overpayment--payments to households that included inmates of
correctional institutions.\3 Federal regulations prohibit prisoners
from participating in the Food Stamp Program.  By matching automated
food stamp records and prison records in four states--California,
Florida, New York, and Texas--we identified over 12,000 inmates who
were included in the households receiving food stamps in calendar
year 1995.  These households improperly collected an estimated $3.5
million in food stamps in 1995.  Subsequently, in August 1997, the
Balanced Budget Act of 1997 (P.L.  105-33, Aug.  5, 1997) included a
provision directing the states to ensure that individuals who are
under federal, state, or local detention for more than 30 days are
not participating in the Food Stamp Program. 

In February 1998, we reported on another type of food stamp
overpayment--payments made to households that included deceased
individuals as members.\4 By matching automated food stamp records
from the four states previously mentioned with death information from
the Social Security Administration's (SSA) Death Master File, we
identified nearly 26,000 deceased individuals who were included in
households receiving food stamps in 1995 and 1996.  These households
improperly collected an estimated $8.5 million in food stamp
benefits.  Subsequently, in March 1998, you, Mr.  Chairman,
introduced legislation that would require the Commissioner of SSA to
use all of SSA's death information to notify state agencies when an
individual receiving food stamp benefits is deceased. 

At your request, Mr.  Chairman, we are currently reviewing the extent
to which individuals are included in food stamp households in more
than one state during the same time period, referred to as "duplicate
participation." While some states conduct matches of their food stamp
rolls with neighboring states, our review is focussed on
non-neighboring states.  For example, we will determine whether
duplicate participation occurs between New York and Florida or
between California and Texas.  Our preliminary results indicate that
such duplicate participation is occurring on a significant scale and
that there is no national mechanism in place to identify and
eliminate it. 

Regarding trafficking--the second main area of fraud and abuse in the
Food Stamp Program--a 1995 FNS study estimated that up to $815
million,\5
or about 4 percent of the food stamps issued, was exchanged for cash
by authorized retailers during fiscal year 1993.  The study found
that the trafficking rate was highest, 13 percent of food stamps
redeemed, among small, privately owned food retailers that generally
do not stock a full line of food.  In contrast, supermarkets and
large grocery stores had an average trafficking rate of less than 2
percent of the benefits redeemed.  Data on the extent to which food
stamps are exchanged between individuals prior to reaching authorized
retailers are unavailable. 

In our March 1998 report on food stamp trafficking, conducted at your
request, Mr.  Chairman,\6 we found that retail store owners and
retail store clerks share almost equal responsibility for the food
stamp trafficking problem.  Specifically, in the 432 trafficking
cases we reviewed, store owners alone were caught trafficking in
about 40 percent of the cases, clerks alone were involved in 47
percent of the cases, and store owners and clerks together were
caught in 13 percent of the trafficking cases.  FNS took
administrative action against all the store owners that were
trafficking but took no actions against the store clerks because it
lacks the authority to do so.  However, some clerks were subject to
court-ordered actions, including financial penalties or jail
sentences. 


--------------------
\3 Food Stamps:  Substantial Overpayments Result From Prisoners
Counted as Household Members (GAO/RCED-97-54, Mar.  10, 1997). 

\4 Food Stamp Overpayments:  Thousands of Deceased Individuals Are
Being Counted as Household Members (GAO/RCED-98-53, Feb.  11, 1998). 

\5 The Extent of Trafficking in the Food Stamp Program, U.S. 
Department of Agriculture, Food and Nutrition Service, Aug.  19,
1995. 

\6 Food Stamp Program:  Information on Trafficking Food Stamp Program
Benefits (GAO/RCED-98-77, Mar.  26, 1998). 


   LARGE NUMBER OF PARTICIPANTS
   AND ADMINISTRATIVE PROCESSES
   MAKE PROGRAM SUSCEPTIBLE TO
   FRAUD AND ABUSE
---------------------------------------------------------- Chapter 0:3

The Food Stamp Program is inherently susceptible to some level of
fraud and abuse because of the sheer number of program participants
(about 23 million in fiscal year 1997), the basic approach used to
determine a household's eligibility and benefit amount, and the
process used to authorize and monitor a sufficient number of
retailers to accept food stamps.  In making eligibility decisions,
state caseworkers rely on applicants to provide accurate information
on, among other things, household composition, and to report
subsequent changes, such as the loss of a household member.  Only
"questionable" cases are investigated.  In general, the agencies take
this approach in an effort to make the program convenient for clients
and simple to administer, and to ensure accurate payments;
consequently, controls over determining household composition are not
as rigorous as they could be.  For example, FNS' regulations do not
require caseworkers to verify client-provided information on
household composition, unless such information is deemed
"questionable," as defined by the state agency.  Investigators
attempt to verify this information through techniques such as
visiting the home and/or contacting neighbors and landlords, however,
they characterize these efforts as time-consuming, costly, and often
unreliable. 

With respect to trafficking, it is difficult to track the flow of
food stamps after they are issued to recipients.  Federal and state
officials agree that food stamps have essentially become a second
currency exchanged by some recipients for cash or non-food items. 
Trafficked food stamps may change hands several times, but all food
stamps must eventually pass through an FNS-authorized retailer
because only such a retailer can redeem food stamps for cash from the
government.  FNS is responsible for monitoring program compliance by
the approximately 185,000 stores that currently are authorized to
redeem food stamps.  Our 1995 report found that, at that time, FNS'
controls and procedures for authorizing and monitoring the retailers
that participate in the Food Stamp Program did not deter or prevent
retailers from trafficking in food stamps.  Since our report, FNS has
initiated several actions to reduce trafficking in the program, such
as contracting with different companies to make 35,000 to 40,000
store visits by the end of fiscal year 1998.  These visits are being
made to verify that the stores are bonafide grocery operations.  In
addition, FNS is improving its Store Tracking and Redemption System
by, for example, developing a profile that enables FNS to better
identify stores that may be trafficking. 


   OVERPAYMENT LEVELS HAVE
   DECLINED, BUT ADDITIONAL
   ACTIONS TO FURTHER REDUCE FRAUD
   AND ABUSE WOULD BE WORTHWHILE
---------------------------------------------------------- Chapter 0:4

USDA's data show that overpayments in the Food Stamp Program have
declined since 1993.  According to the data, the overpayment error
rate at the national level has decreased from 8.27 percent of the
total benefits provided in fiscal year 1993 to 6.92 percent in fiscal
year 1996, the lowest error rate ever achieved in the program.  With
the support of the Congress, FNS has increased its emphasis on
achieving payment accuracy and has employed various initiatives to
assist the states in reducing the number of errors.  For example, FNS
sponsored national, regional, and state conferences, provided direct
technical assistance to the states, and facilitated the exchange of
state information on effective strategies for determining accurate
payments.  While these efforts have been useful in reducing fraud and
abuse, we believe that FNS could achieve even greater success by
taking a leading role in promoting the use and sharing of automated
information by state agencies. 

Given the program's strong reliance on applicants, clients and
retailers to comply with program regulations and provide accurate and
timely information, state agencies need to have access to information
that will allow them to independently and cost effectively verify the
information they are provided and identify noncompliance.  Our
reviews have demonstrated that useful information can be obtained
from (1) matching state food stamp rolls against other state
databases, such as prisoner rolls, and (2) reconfiguring existing
federal databases to provide additional useful information to state
agencies, such as death notices.  Additional opportunities to use
computerized resources to verify information exist, as seen in our
ongoing review of duplicate participation in non-neighboring states. 

Both an FNS study and our own experiences demonstrate that automated
data matches by the states using food stamp records provides a
cost-effective means of reducing fraud and improving program
integrity.  The cost of conducting computer matches is relatively low
for the return generated, which includes identifying ineligible
individuals in the application process before any benefits are issued
and preventing additional issuance once an ineligible participant is
identified. 

State agencies have already implemented computerized matches on their
own initiative, such as neighboring state matches for duplicate
participation.  Two state agencies we visited have taken steps to
obtain information from credit reporting services to ensure that
applicants are eligible for benefits.  In addition to recouping
overpayments, matching efforts help the program realize savings by
identifying erroneous information during the application process,
according to states.  Furthermore, the states said that these efforts
have a deterrent effect on applicants who may be considering
fraudulent activities. 

Relatedly, while EBT will not eliminate all types of fraud, it shows
promise as a means to identify redemption patterns that indicate
potential trafficking.  By eliminating paper coupons that may be
lost, stolen, or sold without any record of sale and creating an
electronic record of transactions, EBT can help identify and reduce
food stamp trafficking.  However, because EBT systems are simply
another vehicle for distributing benefits, they cannot correct fraud
and abuse that occurs during the process of determining eligibility
and benefit levels.  Also, like any computer system, food stamp EBT
systems can be susceptible to security breaches that result in new
forms of fraud and abuse. 

FNS can further expand on its recent successes in reducing
overpayments by actively encouraging the states to identify ways to
use computerized information to verify information provided by
applicants and to encourage states to share their techniques and
information with each other.  FNS can demonstrate its leadership in
this regard by identifying sources of information that would be
useful to the states and ensuring that states have access to that
information. 


-------------------------------------------------------- Chapter 0:4.1

Thank you again for the opportunity to appear before you today.  We
would be pleased to answer any questions you may have. 


*** End of document. ***