Food Stamp Overpayments: Thousands of Deceased Individuals Are Being
Counted as Household Members (Letter Report, 02/11/98, GAO/RCED-98-53).

Pursuant to a congressional request, GAO reviewed the Food Stamp
Program, focusing on: (1) how many deceased individuals were included as
members of households that received food stamp benefits and the
estimated value of improper benefits that were issued to the households;
(2) how these individuals could be included without being detected; and
(3) whether computer matching or other methods could effectively
identify such individuals.

GAO noted that: (1) GAO identified nearly 26,000 deceased individuals in
the four states GAO reviewed who were included in households receiving
food stamps for the 2-year period 1995 through 1996; (2) these
households improperly collected an estimated $8.5 million in food stamps
benefits; (3) the inclusion of deceased individuals in food stamp
households goes undetected because agencies rely primarily on unverified
information on household membership provided by food stamp applicants
and participants; (4) states are not required to match
applicant-provided Social Security numbers with the Social Security
numbers of deceased individuals; (5) however, several state agencies
match information on the applicant's household members with information
on deceased individuals from their state's vital statistics agency; (6)
these states' efforts have had mixed success because the states have not
always had verified, comprehensive death information; (7) while the
Social Security Administration (SSA) makes information on its deceased
beneficiaries available to state agencies through its State Verification
and Exchange System, this information is limited to the recipients of
specific SSA benefits; (8) states' computer matching of individuals in
food stamp households with data in SSA's more comprehensive Death Master
File would provide a cost-effective mechanism to accurately and
independently identify deceased individuals included in food stamp
households; (9) it would be even more cost-effective, however, for SSA
to notify the states when a food stamp participant dies, rather than
having the states conduct computer matches, because SSA already has a
system in place to identify deceased individuals who received food stamp
benefits but not Social Security benefits; and (10) furthermore, some
states place restrictions on the use of the death data they provide to
SSA; the agency currently does not have the authority to disclose
restricted death information to other states administering federal
benefit programs.

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

 REPORTNUM:  RCED-98-53
     TITLE:  Food Stamp Overpayments: Thousands of Deceased Individuals 
             Are Being Counted as Household Members
      DATE:  02/11/98
   SUBJECT:  Welfare benefits
             Erroneous payments
             Social security number
             Social security benefits
             Food relief programs
             State-administered programs
             Program abuses
             Vital records
             Computer matching
             Federal/state relations
IDENTIFIER:  Food Stamp Program
             California
             Texas
             Florida
             New York
             SSA Death Master File
             
******************************************************************
** 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 Chairman, Committee on Agriculture, Nutrition, and
Forestry, U.S.  Senate

February 1998

FOOD STAMP OVERPAYMENTS -
THOUSANDS OF DECEASED INDIVIDUALS
ARE BEING COUNTED AS HOUSEHOLD
MEMBERS

GAO/RCED-98-53

Food Stamp Households With Deceased Members

(150273)


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

  FNS - Food and Nutrition Service
  IEVS - Income and Eligibility Verification System
  RSDI - Retirement, Survivor and Disability Insurance
  SSA - Social Security Administration
  SSI - Supplemental Security Income
  SVES - State Verification and Exchange System
  USDA - U.S.  Department of Agriculture

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


B-278830

February 11, 1998

The Honorable Richard G.  Lugar
Chairman, Committee on Agriculture,
 Nutrition, and Forestry
United States Senate

Dear Mr.  Chairman: 

During numerous hearings over the last several years, the Congress
has expressed its strong desire to reduce the level of fraud, waste,
and abuse in the Food Stamp Program, which is administered by the
U.S.  Department of Agriculture (USDA).  We recently reported that in
calendar year 1995, despite federal regulations prohibiting inmates
of correctional institutions from participating in the Food Stamp
Program, more than 12,000 inmates in four states were included in
households that improperly collected an estimated $3.5 million in
benefits.\1 In this context, you asked us to determine whether other
ineligible individuals were similarly being counted as members of
households receiving food stamp benefits.  As agreed, this report
focuses on deceased individuals who were counted as household members
for the purpose of calculating benefits. 

We determined (1) how many deceased individuals were included as
members of households that received food stamp benefits and the
estimated value of improper benefits that were issued to the
households, (2) how these individuals could be included without being
detected, and (3) whether computer matching or other methods could
effectively identify such individuals. 

To identify deceased individuals counted as members of food stamp
households, we conducted a computer match comparing 1995 and 1996
food stamp rolls in four large states (California, Florida, New York,
and Texas) with the Social Security Administration's (SSA) Death
Master File.\2 These four states account for 35 percent of the
nation's participants in the Food Stamp Program.\3 (See app.  I.)
SSA's records contained all the individuals reported as deceased
nationwide from 1937 through 1996.  To provide the highest level of
confidence that our analyses resulted in valid matches, we used only
those matches that (1) met the most reliable and stringent match
criteria used by SSA to verify Social Security numbers, (2) showed a
date of death that preceded the dates that food stamp benefits were
issued to the household, and (3) showed that the individual was
deceased at least a full month and that sufficient time had elapsed
for the household to notify the state of the change and for the state
to take action.\4


--------------------
\1 Food Stamps:  Substantial Overpayments Result From Prisoners
Counted as Household Members (GAO/RCED-97-54, Mar.  10, 1997). 

\2 SSA's Death Master File compiles death information from a wide
variety of sources and is considered the most comprehensive list of
death information available in the federal government. 

\3 Because California does not maintain statewide participation
information, we used state eligibility information in our match.  We
determined that for California, eligibility for the program is
representative of actual participation, as described in app.  II. 

\4 Program regulations generally give a household 10 days to notify
the state of changes in composition and the state 10 days from the
date of notification to adjust benefits. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

We identified nearly 26,000 deceased individuals in the four states
we reviewed who were included in households receiving food stamps for
the 2-year period 1995 through 1996.  These households improperly
collected an estimated $8.5 million in food stamp benefits. 

The inclusion of deceased individuals in food stamp households goes
undetected because agencies rely primarily on unverified information
on household membership provided by food stamp applicants and
participants.  States are not required to match applicant-provided
Social Security numbers with the Social Security numbers of deceased
individuals.  However, several state agencies match information on
the applicant's household members with information on deceased
individuals from their state's vital statistics agency.  These
states' efforts have had mixed success because the states have not
always had verified, comprehensive death information.  While the
Social Security Administration makes information on its deceased
beneficiaries available to state agencies through its State
Verification and Exchange System, this information is limited to the
recipients of specific Social Security Administration benefits. 

States' computer matching of individuals in food stamp households
with data in the Social Security Administration's more comprehensive
Death Master File would provide a cost-effective mechanism to
accurately and independently identify deceased individuals included
in food stamp households.  It would be even more cost-effective,
however, for the Social Security Administration to notify the states
when a food stamp participant dies, rather than having the states
conduct computer matches, because the Social Security Administration
already has a system in place to identify deceased individuals who
received Social Security benefits.  However, such notifications would
require the agency to make some adjustments to its current data
system to identify deceased individuals who received food stamp
benefits but not Social Security benefits.  Furthermore, some states
place restrictions on the use of the death data they provide to the
Social Security Administration; the agency currently does not have
the authority to disclose restricted death information to other
states administering federal benefit programs. 


   BACKGROUND
------------------------------------------------------------ Letter :2

The Food Stamp Program is designed to promote the general welfare and
to safeguard the health and well-being of the nation's population by
raising the nutrition levels of low-income families.  Recipients use
their food stamp benefits to purchase allowable food products from
authorized retail food merchants. 

Eligibility for food stamp benefits is determined on a household
basis.  A household can be an individual, a family, or another group
who lives together and customarily purchases and prepares food in
common.  Applicants must provide a Social Security number for each
household member.  The value of the food stamp benefits for a
household is determined by the number of eligible household members
and their income, adjusted for assets and such costs as expenses for
shelter and utilities.  Nationally, the average monthly value of the
benefit per household member is about $73.  The household's monthly
food stamp allotment increases with each additional member, provided
income limits are not exceeded.  Households that receive food stamps
are required to report changes in household membership, such as the
loss of a member, to the administering state agency. 

Within USDA, the Food and Nutrition Service (FNS) administers the
Food Stamp Program through agreements with state agencies.  FNS is
responsible for approving state plans for operation and for ensuring
that the states are administering the program in accordance with
regulations.  The states are required to establish a performance
reporting system to monitor the program, including a quality control
review process to help ensure that benefits are issued only to
qualifying households and that the benefit amounts are correct. 
State agencies are responsible for imposing penalties for violations
of program requirements and are responsible for recovering food stamp
overpayments.\5 The federal government pays all of the benefit costs
and one-half of the administrative costs for each state.  In 1996,
USDA paid the states about $1.7 billion to administer the program. 

The program is administered by states.  Eligibility workers in
service centers work directly with clients to certify household
eligibility and determine benefit amounts at the time of application
and at least annually thereafter. 

With respect to deceased individuals, SSA's information is more
comprehensive than that gathered by state agencies.  SSA compiles
information in its Death Master File from a wide variety of sources,
including the Department of Veterans Affairs, the Health Care Finance
Administration, funeral directors, and its own internal claims and
postentitlement processes.  SSA also includes in the Death Master
File death information that it purchases from the states through
agreements negotiated with each state.  About half of these
agreements restrict SSA's ability to disclose this information to
agencies in other states.  Each Social Security number in the Death
Master File has been verified to ensure that it is associated with
the appropriate name, date of birth, and gender for each deceased
individual.  SSA updates its file monthly.  In contrast, state
information regarding deceased individuals is initially compiled at
the local level by, for example, county departments of health, which
gather the information from local funeral homes and other sources,
including coroners.  The counties forward the information to a
designated state agency, such as the state bureau of vital
statistics. 


--------------------
\5 The Personal Responsibility and Work Opportunity Reconciliation
Act of 1996 doubled the disqualification period for intentional
program violations from 6 months to 1 year for the first violation
and from 1 year to 2 years for the second violation. 


   INCLUSION OF DECEASED
   INDIVIDUALS IN FOOD STAMP
   HOUSEHOLDS COSTS MILLIONS IN
   OVERPAYMENTS
------------------------------------------------------------ Letter :3

During calendar years 1995 and 1996, about $8.5 million in food stamp
benefits were improperly collected by households that included
deceased individuals--nearly 26,000 in total--in the four states we
examined.  (See table 1.) While these individuals were counted in
food stamp households for an average of 4 months, we identified 177
instances in which deceased individuals were counted as beneficiaries
for the full 2 years covered by our review.  We also identified 20
instances in which such individuals were counted in more than one
state during the period of our review. 



                                Table 1
                
                    Deceased Individuals Counted as
                Household Members, Number of Months for
                  Which Individuals Were Counted, and
                   Estimated Overpayments, by State,
                      Calendar Years 1995 and 1996

                             Number of
                              deceased
                           individuals    Total months
                            counted as       for which
                             household     individuals       Estimated
States examined                members    were counted    overpayments
----------------------  --------------  --------------  --------------
California\a                     1,646           6,293        $415,859
Florida                          5,794          21,748       1,709,245
New York                        11,976          61,893       5,012,315
Texas                            6,465          18,078       1,341,598
======================================================================
Total                           25,881         108,012      $8,479,016
----------------------------------------------------------------------
\a The lower number of deceased individuals included in food stamp
households in California in comparison with the other states in our
review can be attributed largely to the state's "cash-out" policy,
under which the state provides a cash supplement in lieu of food
stamps to Supplementary Security Income recipients.  Consequently,
many elderly individuals, who accounted for a large portion of the
deceased individuals who were counted as household members in other
states, were not on California's food stamp rolls. 

Sources:  For California, California Department of Public Social
Services; for Florida, Florida Department of Children and Families;
for New York, New York State Department of Social Services; for
Texas, Texas Department of Health Services; FNS; and SSA. 

The deceased individuals whom we identified in our match were members
of households of varying sizes, some with multiple members and some
with a single member--the deceased individual was the household.  For
households with multiple members, the household may have included a
deceased individual as a member at application, or the member may
have passed away while on the food stamp rolls.  But benefits
continued to be issued on the presumption that the individual was
present in the home.  For single-member households, some other
individual obtained the benefits.  In these cases, the food stamps
could have been issued either to a person designated as the deceased
individual's authorized representative\6 or to someone who
fraudulently represented himself or herself as the deceased
individual. 

Food stamp benefits are issued either as coupons or via electronic
benefit transfer systems.  For coupons, issuance procedures require
clients to present various items of identification, such as Food
Stamp Program cards bearing their signature, in order to pick up food
stamps from a service center or other outlet.  A small number of the
coupons are mailed to the clients.  Under an electronic benefit
transfer system, the state agency issues access cards (similar to
debit cards) and personal identification numbers to clients who
obtain benefits through point-of-sale terminals in stores.  However,
the effectiveness of the issuance procedures to ensure that only
eligible participants receive benefits depends on how rigorously the
procedures are followed by the responsible staff and clients. 


--------------------
\6 All food stamp clients have the right when they apply to specify
an authorized representative to act on their behalf, including
receiving their stamps. 


   UNVERIFIED AND INCOMPLETE
   HOUSEHOLD INFORMATION ALLOWS
   INCLUSION OF DECEASED
   INDIVIDUALS IN HOUSEHOLDS TO GO
   UNDETECTED
------------------------------------------------------------ Letter :4

Deceased individuals are included in food stamp households because
(1) local agencies use unverified and incomplete information to
determine the composition of a household when calculating benefits
and (2) recipients of benefits sometimes do not notify the agencies
of changes in household composition.  Some states have attempted to
verify the information that food stamp clients provide on household
composition through computer matching of this information with deaths
reported to the state, although the program does not require such
matching.  The success of these efforts has been mixed, largely
because the states do not have complete, accurate, and timely data. 
Currently, even the information SSA reports to the states on deaths
is incomplete--SSA provides information only on deceased individuals
who received both SSA benefits and food stamp benefits, not just food
stamp benefits. 


      STATES RELY ON
      CLIENT-PROVIDED HOUSEHOLD
      INFORMATION
---------------------------------------------------------- Letter :4.1

Most agencies primarily rely on food stamp applicants to provide
accurate information on household composition and to report
subsequent changes, such as the loss of a household member.  In
general, the Food Stamp Program takes this approach because it has to
balance the issues of clients' convenience, administrative
simplicity, and payment accuracy; consequently, controls over
determining household composition are not as rigorous as they could
be.  A household that wishes to receive benefits must present an
application listing members and their Social Security numbers and
provide information about income and other eligibility factors. 
Eligibility workers review this information, obtain identification
documents such as a driver's license, interview a household
representative, and certify eligibility.\7 In addition, they
recertify the household at least annually.  However, at no time are
all household members required to appear and present identification. 
Furthermore, clients are responsible for identifying changes in
household composition.\8

According to FNS, which identifies error rates for each state by
reviewing a random sample of cases, clients' errors or
misrepresentations contribute significantly to incorrectly issued
benefits, particularly when overpayments occur.  In 1996, FNS
reported that about 7 percent of the benefits issued nationwide were
overpayments and that 57 percent of the overpayments were
attributable to intentional or unintentional inaccuracies in
client-provided information. 

Nevertheless, FNS' regulations do not require the verification of
client-provided information on household composition, unless the
eligibility workers deem the information "questionable." The
regulations allow each state agency to develop guidance for
identifying questionable information.  In the states we visited, the
guidance defined questionable information as applicants' statements
that were contradictory or did not agree with the information that
was in the case record or otherwise available to the eligibility
workers.  FNS has not required the states to match Social Security
numbers of deceased individuals with the numbers provided for the
Food Stamp Program. 

When the eligibility workers in the states we visited suspected
questionable or fraudulent information, they could refer the
application to investigators before granting aid.  Investigators in
each state told us that they try to verify questionable information
on household composition by visiting homes and making collateral
contacts to confirm information with friends, neighbors, or
landlords.  According to the investigators, these techniques are
hit-or-miss, time-consuming, costly undertakings and provide
information that is only as reliable as its source.  Furthermore,
investigative resources are generally very limited. 


--------------------
\7 Sometimes the eligibility workers fill out the application on the
basis of the interview. 

\8 Almost all households in California must submit monthly reports on
changes.  Households in Florida and Texas are required to report
changes in circumstances, including changes in household composition,
within 10 days of becoming aware of the change.  Households with
earned income in New York must report changes quarterly; all others
must report within 10 days of a change. 


      STATES' MATCHING EFFORTS
      HAVE HAD MIXED RESULTS
---------------------------------------------------------- Letter :4.2

Although some state agencies have employed their own computer matches
as a means of identifying deceased individuals who are included in
food stamp households, the practice does not appear to be widespread. 
Of the states we visited, California, Florida, and New York have
established a matching program between the state's public assistance
rolls (for all assistance programs, not just food stamps) and the
state's death records from its vital statistics agency.  According to
FNS, only two other states, Illinois and Ohio, currently perform a
monthly computer match between state records of deceased individuals
and their food stamp rolls, and two other states, Maryland and
Pennsylvania, have conducted matches, but not routinely.\9

The California Department of Health Services conducts a monthly match
of its assistance eligibility rolls, including that for food stamps,
with state death records from its vital statistics agency.  However,
because the agency's Social Security numbers are not verified, and
the information is limited to California, mismatches may occur, and
some deceased individuals may not be identified. 

While Florida officials consider their match to be very useful, the
state's reliance on unverified Social Security numbers and other
identifying information reduces its effectiveness.  The officials
appreciate the positive effects of their match on the overall
integrity of assistance programs (food stamps, Medicare, and others)
because it eliminates ineligible individuals from the rolls and
discourages fraud and abuse.  However, because the agency's match is
based solely on Social Security numbers, and these numbers are not
verified in the state's record of deceased individuals, invalid
matches are produced.  During the first year of the match (initiated
in December 1995), 4,528 matches were acted on by eligibility workers
in the following way:  1,439 cases were updated, 21 cases were
referred for fraud investigation, and 57 cases were determined to be
duplicative.  According to a state official, the remaining 3,011
matches required no action because (1) the case file had already been
revised or (2) the match was invalid because although a Social
Security number matched, other data-- such as date of birth, gender,
and race--did not.  Access to SSA's death information, in which the
Social Security number, name, date of birth, and gender have been
verified, would reduce the number of invalid matches. 

The New York State matching effort was suspended in June 1997 because
the outdated death information used in the match resulted in invalid
matches.  The match, which had been in place since the late 1970s,
was conducted using only the state's vital statistics data from the
previous year.  Eligibility workers stopped following up on the match
results because they often found that appropriate action had already
been taken as a result of more recent information provided by other
sources, such as household members or obituary listings.  In 1996, a
new state law required the state to conduct a computer match to
identify deceased individuals included in households receiving
federal and state assistance, primarily so that liens could be placed
on the estates of individuals owing funds to the state.  In response
to this requirement, the state is redesigning the matching program
and establishing a process with the state's vital statistics bureau
to ensure that current data are used.  SSA's death information, if
available to the state, would provide the most comprehensive and
current source of information. 


--------------------
\9 In addition, every month, New York City matches public assistance
rolls against death records from its health department. 


      SSA-PROVIDED INFORMATION
      EXCLUDES DATA ON INDIVIDUALS
      WHO ONLY RECEIVED FOOD
      STAMPS
---------------------------------------------------------- Letter :4.3

SSA provides limited information to the states on deceased
beneficiaries of its own programs, Retirement, Survivor and
Disability Insurance (RSDI) and/or Supplemental Security Income
(SSI), through its State Verification and Exchange System (SVES). 
SVES is a computer system that the states use to verify the Social
Security numbers of applicants to the Food Stamp Program and to
determine if applicants are receiving RSDI or SSI benefits.  However,
SVES does not automatically notify the states of the deaths of food
stamp recipients who did not also receive RSDI or SSI benefits. 

For SSA beneficiaries, SVES will automatically alert the states if
RSDI or SSI benefits have been terminated or altered because of death
or other conditions, and will also provide the SSA-verified date of
death and changes in benefits.  However, for individuals who do not
receive SSA benefits but who have applied for food stamp benefits,
SSA only verifies the accuracy of the Social Security number(s),
name(s), and date(s) of birth submitted in the application.  SSA does
not notify the states if those Social Security numbers belong to
deceased individuals. 


   PROVIDING STATES WITH
   COMPREHENSIVE INFORMATION WOULD
   ALLOW MORE EFFECTIVE
   IDENTIFICATION OF OVERPAYMENTS
------------------------------------------------------------ Letter :5

Matching SSA's comprehensive Death Master File with states'
information on food stamp participants would be an effective and
inexpensive way of identifying deceased individuals included in food
stamp households.  Because SSA already has a system in place for
providing death information to the states, it would be even more
cost-effective for SSA to routinely provide death information on food
stamp participants to the states rather than having the states
conduct computer matches of the food stamp rolls and their death
records.  However, in order for SSA to provide this additional
information, it would have to (1) modify its SVES and (2) be given
the authority to provide the data to the states.  The availability of
such information would provide ancillary benefits, such as helping
state agencies identify deceased individuals in other assistance
programs. 


      STATE-CONDUCTED COMPUTER
      MATCHING IS A COST-EFFECTIVE
      WAY TO IDENTIFY DECEASED
      INDIVIDUALS INCLUDED IN FOOD
      STAMP HOUSEHOLDS
---------------------------------------------------------- Letter :5.1

Although the lack of reliable information adversely affected the
success of matching programs in California, Florida, and New York,
both an FNS study and our own experiences demonstrate that automated
data matches by states using food stamp records provide a
cost-effective means of reducing fraud and improving program
integrity.  A 1995 FNS study of the Income and Eligibility
Verification System (IEVS), which compares wage, benefit, and other
payment information reported by food stamp clients with records in
six government agencies' databases, demonstrated the
cost-effectiveness of programs using computer matches to verify
beneficiary information.\10

In the two sample states it reviewed, FNS found that all of the IEVS
matching programs had a cost-effectiveness ratio (program savings
compared with the costs of the match, follow-up, and claims
collection) greater than 1, indicating that every dollar spent on
IEVS returned more than a dollar in savings to the program. 

We found during this and our prior review of prisoners' participation
that developing the computer programs to identify participation by
ineligible individuals did not require a large investment of
programmers' time.  Our programmers required an average of about 20
days to develop a program for each state.  This effort required them
to become familiar with the files, resolve any apparent data
problems, and create and run each program.  State programmers may
require less time because they are already familiar with the food
stamp data in their state.  The cost of having states conduct
computer matching is relatively low for the return generated: 
According to the 1995 study of IEVS, data-processing costs were
approximately 2 cents per case, and investigative follow-up and claim
collection costs were about $5 to $7 for all matches, whether or not
there were overpayments. 

Although the states we reviewed that matched their vital statistics
data with the food stamp rolls had mixed success, their success rate
could be improved if they used the more comprehensive data maintained
by SSA.  However, states wishing to conduct such matches may have
difficulties because of the size of the SSA database--which includes
57 million records.  Moreover, SSA would have to provide its data
files to all the states that wanted to conduct such matches, which
would be a costly and inefficient method of matching deceased
individuals with food stamp rolls. 


--------------------
\10 The Income and Eligibility Verification System (IEVS) Targeting
Demonstration, Findings and Guidelines for State Food Stamp IEVS
Programs, Final Report, USDA, Food and Consumer Service [now the Food
and Nutrition Service], Office of Analysis and Evaluation (Apr.  1,
1995). 


      HAVING SSA IDENTIFY FOOD
      STAMP RECIPIENTS WHO HAVE
      DIED WOULD BE A MORE
      COST-EFFECTIVE AND
      COMPREHENSIVE WAY TO ENSURE
      THAT DECEASED INDIVIDUALS
      ARE EXCLUDED FROM FOOD STAMP
      HOUSEHOLDS
---------------------------------------------------------- Letter :5.2

Computer matching of SSA's Death Master File information with
information on food stamp recipients would yield more comprehensive
results than states' efforts to match such beneficiary and death
information in their own state records.  Unlike states' information,
the Death Master File includes, for example, participants who would
not be included in a state's or local agency's vital statistics
information because they died outside of the state or of the local
agency's jurisdiction.  In addition, SSA's information, including
each individual's Social Security number, name, gender, and date of
birth, is verified, while this information is not verified in states'
vital statistics records. 

As previously discussed, SVES currently verifies information on
Social Security numbers and on SSA benefits provided to food stamp
applicants for the states, and notifies the states of any changes in
SSA benefit payments for participants in the Food Stamp Program. 
While SVES provides an established means for this exchange of data,
two problems would have to be addressed if SVES were also to provide
information on deceased individuals whose Social Security numbers are
being used only for food stamp benefits. 

First, SSA officials stated that the agency would have to make a
minor modification to SVES in order for it to identify individuals
who are deceased and received food stamp benefits but not any SSA
benefits.  According to SSA officials, while SVES is not currently
configured to query the Death Master File with state-submitted
information, the software needed to do so has been developed and
validated.  The officials stated that SVES was originally designed to
include this capability.  Developed by SSA about 6 years ago, the
software would have to be updated and activated to provide this
information.  According to SSA officials, while the cost and time
required to do this programming is unknown, it is not expected to be
prohibitive. 

Second, and more significantly, SSA does not have the authority to
provide death information obtained from one state to other states. 
Under section 205 (r)(1) of the Social Security Act,\11 SSA has the
authority to enter into agreements with the states to obtain states'
death information.  In addition, section 205 (r)(3) of the Social
Security Act\12 authorizes SSA to provide death information to
federal and state agencies administering federal benefit programs to
"the extent feasible." However, some of the state agreements restrict
SSA's ability to disclose such information.  According to SSA, over
half of the agreements between SSA and the states contain such
restrictions. 

This problem of disclosure has been recognized for some time.\13 In
1991, we recommended that the Congress remove these restrictions.  In
1993, the Congress passed legislation to remove such restrictions on
disclosure to federal agencies to ensure that federal payments or
other benefits are not erroneously paid to deceased individuals (26
U.S.C.  6103 (d)(4)).  However, the legislation did not cover state
agencies administering such federal programs.  One of the Department
of Health and Human Services' recommendations for the National
Performance Review would have resolved this problem.  The
recommendation called for establishing a federal clearinghouse for
death information that could have been used by federal and state
agencies.  Legislation to establish such a clearinghouse was
introduced in the 103rd Congress but was not enacted. 

In addition to providing accurate, comprehensive match data to
identify deceased individuals in the Food Stamp Program, access to
SSA's Death Master File information would provide opportunities to
states to identify deceased individuals in other assistance programs
and to improve the integrity of their food stamp data.  The states
use SVES to verify information for applicants and current recipients
of all assistance programs, including the Food Stamp Program,
Medicare, and the states' programs of aid to families and children
and general relief. 

SSA's information on deceased individuals could also be used to
identify agency errors.  We identified a couple of cases in which the
household reported the loss of a member, but, for some reason, the
state or local agency did not remove that person from the benefit
calculation, causing an overpayment.  A notice of a match with SSA's
data sent to the eligibility worker would provide another opportunity
for the case to be looked at and adjusted to reflect the change in
household size. 

SSA's death information could also help agencies to profile the
circumstances under which benefits are erroneously issued to deceased
and other ineligible individuals.  For example, an examination on the
basis of household size of instances in which identified deceased
individuals counted as members of food stamp households could be
useful in determining if the problem is predominantly caused by (1)
multimember households' not reporting deceased members or (2)
individuals' fraudulently using the name, date of birth, and Social
Security number of deceased individuals.  An agency could then take
appropriate steps to address the problem, such as reinforcing the
responsibility of households to report changes in composition to the
agency.  Similarly, examining the method of food stamp issuance
(electronic benefit transfer, mail, or clients' pickup) could alert
an agency to the issuance method under which having deceased
individuals counted as members of food stamp households is most
prevalent and help it to select appropriate countermeasures. 
Similarly, identifying the eligibility workers responsible for a
disproportionate number of cases with deceased individuals could
identify the need for additional guidance or training. 


--------------------
\11 42 U.S.C.  405 (r)(1). 

\12 42 U.S.C.  405 (r)(3). 

\13 Agencies Need Death Information From Social Security to Avoid
Erroneous Payments (GAO/HRD-91-3, Feb.  6, 1991) and Status of Agency
Use of SSA Death Information (GAO/HRD-93-31R, July 20, 1993). 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

The inclusion of deceased individuals in households receiving food
stamp benefits compromises the integrity of the Food Stamp Program
and results in overpayments.  Conventional methods for detecting such
individuals have not been fully effective, resulting in overpayments
in the states we reviewed.  A match by states using SSA's Death
Master File information or, preferably, a modification to SSA's
computerized verification system that provides SSA's deceased
beneficiary information to the states could be a cost-effective
method for identifying such individuals in food stamp households and
reducing overpayments.  To make this possible, SSA would need to be
given the authority to provide this information to the states. 


   RECOMMENDATION TO THE CONGRESS
------------------------------------------------------------ Letter :7

In order to ensure the integrity of the Food Stamp Program by
preventing deceased individuals from being counted as household
members, we recommend that the Congress enact legislation to enable
SSA to disclose all information from its Death Master File to the
states administering the Food Stamp Program. 


   RECOMMENDATIONS TO THE
   SECRETARY OF AGRICULTURE
------------------------------------------------------------ Letter :8

Until the Congress enacts legislation to enable SSA to disclose all
information from its Death Master File to the states administering
the Food Stamp Program, we recommend that the Secretary of
Agriculture work with the Commissioner of the Social Security
Administration to encourage the states to voluntarily allow such
disclosures. 

We further recommend that the Secretary of Agriculture direct the
Food and Nutrition Service to emphasize to the states the need to
identify deceased individuals and remove them from the rolls of food
stamp households. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :9

We provided copies of a draft of this report to the U.S.  Department
of Agriculture and the Social Security Administration for review and
comment.  Their comments and our responses are in appendixes III and
IV, respectively. 

In commenting on the draft report, the Department agreed with the
report's recommendations to improve the availability and use of data
from the Social Security Administration so that deceased individuals
would not be counted as members of food stamp households.  The
Department stated that the integrity of the Food Stamp Program would
be improved with better access to Social Security's death record
information.  While agreeing with our report's overall findings and
recommendations, the Department believed that our report was too
definite about the magnitude of the overpayments and that our
methodology did not provide an exact determination of the extent to
which issued benefits were actually used.  As our report recognizes,
without a detailed review of each of the 26,000 cases we identified,
it is not possible to precisely calculate the level of the associated
overpayments.  Instead, to demonstrate the general magnitude of the
dollar values involved, we prepared an estimate using average monthly
payments per person in the states involved.  Our estimating approach,
which we believe was reasonable, could understate or overstate the
actual values.  In any event, we believe the key observation remains: 
The large number of deceased individuals counted in food stamp
households that we identified suggests a systemic problem needing
immediate attention. 

In commenting on the draft report, the Social Security Administration
agreed with our conclusion that the use of its Death Master File
could improve efforts to identify deceased individuals counted as
members of food stamp households.  Social Security stated, however,
that our report's recommendation to the Congress concerning Social
Security's authority to release information in its Death Master File
was not needed.  It said that while some of this information was
restricted and could not be released to the U.S.  Department of
Agriculture's Food and Nutrition Service and the states administering
the Food Stamp Program, Social Security already has the authority to
provide 95 percent of the death information it possesses.  We believe
there is no sound reason why the U.S.  Department of Agriculture and
the states should not have the benefit of all the information
possessed by Social Security as they strive to reduce fraud, waste,
and abuse in the Food Stamp Program.  Furthermore, we do not believe
the data now shared by Social Security can address 95 percent of the
problem; available data from the Food and Nutrition Service indicate
that a significant number of deceased individuals now included in
food stamp households cannot be identified with the information
Social Security now provides.  Congressional action to enable Social
Security to disclose all the death information it possesses to the
Food and Nutrition Service and the states would remove the only
legitimate obstacle to having the Social Security Administration
share information that would help another federal agency reduce
fraud, waste, and abuse in its program. 


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

We conducted our work from March 1997 through January 1998 in
accordance with generally accepted government auditing standards. 
Our detailed methodology is presented in appendix II. 

We are providing copies of this report to appropriate congressional
committees, interested Members of Congress, and other interested
parties.  We will also make copies available to others on request. 

As arranged with your office, unless you publicly announce its
contents earlier, we will make no further distribution of this report
until 30 days from the date of this letter.  At that time, we will
send copies to the Secretary of Agriculture, the Commissioner of the
Social Security Administration, and other interested parties. 

Major contributors to this report are listed in appendix V.  If you
have any questions about this report, please contact me at (202)
512-5138. 

Sincerely yours,

Robert A.  Robinson
Director, Food and
 Agriculture Issues


INFORMATION ON THE COST OF FOOD
STAMP BENEFITS AND PARTICIPATION
IN FOUR STATES
=========================================================== Appendix I

In 1996, California, Florida, New York, and Texas represented almost
36 percent of the cost of benefits in the Food Stamp Program and
approximately 35 percent of the nation's participants. 



                               Table I.1
                
                 1996 Food Stamp Program's Benefit Cost
                       and Participants by State

                                     Costs of food
State                                       stamps  State participants
------------------------------  ------------------  ------------------
California                          $2,554,862,013           3,143,390
Florida                              1,295,526,302           1,371,352
New York                             2,053,944,448           2,098,561
Texas                                2,140,168,464           2,371,958
======================================================================
Total                                8,044,501,227           8,985,261
======================================================================
Program total                      $22,441,331,729          25,533,302
----------------------------------------------------------------------
Source:  U.S.  Department of Agriculture's Food and Nutrition
Service. 


OBJECTIVES, SCOPE, AND METHODOLOGY
========================================================== Appendix II

In response to the Congress's strong interest in reducing the level
of fraud, waste, and abuse in the Food Stamp Program, we reviewed
food stamp beneficiaries' data to determine whether deceased
individuals, who are not eligible for food stamps, were
inappropriately included as members of households receiving food
stamps.  Specifically, we determined (1) how many deceased
individuals were included as members of households that received food
stamp benefits and the estimated value of improper benefits that were
issued to the households, (2) how these individuals could be included
without being detected, and (3) whether computer matching or other
methods could effectively identify such individuals. 

To determine if deceased individuals were included as members of
households that received food stamp benefits and the estimated value
of benefits that were issued to the households, we matched the food
stamp records of the four states with the largest benefits in the
Food Stamp Program and the Social Security Administration's (SSA)
Death Master File.  Specifically: 

  -- Florida, New York, and Texas state welfare agencies provided us
     with computer files containing information on all members of
     households and the amount of food stamp benefits issued to these
     households during calendar years 1995 and 1996.  The data
     provided personal identifiers, including name, Social Security
     number, date of birth, gender, and the months in which food
     stamp benefits had been issued to the household of which each
     individual was a member.  The state agencies had verified the
     Social Security numbers for the data on food stamp beneficiaries
     through SSA's Enumeration Verification System. 

  -- In California, which only maintains eligibility information for
     households at the state level, we determined that eligibility
     was predictive of issuance in two counties, so we used
     eligibility information in lieu of issuance data.\1

  -- SSA provided a copy of its Death Master File, which has compiled
     information since 1937.\2 The data provided the same personal
     identifiers as obtained for food stamp beneficiaries and listed
     the date of death for each individual. 

We matched the verified Social Security numbers of deceased
individuals with the verified Social Security numbers in the states'
records of membership in food stamp households.  For those deceased
individuals identified as members of households, we determined the
periods in which food stamp issuance occurred after the date of
death.  We estimated the dollar value of food stamps issued to
households with deceased members by applying the state's average
monthly issuance per individual recipient from 1995 and 1996 to each
period in which issuance occurred after the date of death.  Food
stamp benefits are calculated for households, not for individuals. 
As such, it is difficult to determine the exact value of benefits
issued to a deceased individual included in a household, unless he or
she is the only member of a household.  Even then, the amount will
vary from individual to individual, on the basis of such factors as
income, assets, and the cost of shelter.  Therefore, we relied on the
average monthly benefit issued per person in the locations we
reviewed, which ranged from a high of $82 in New York to a low of $65
in California. 

In recognition of the notification and processing time frames that
allow 10 days for clients to report household changes and 10 days for
the state agency to take action, we did not consider any issuance in
the month of death or in the following month, where appropriate, to
be an overpayment. 

Because of the quality control program operated by the U.S. 
Department of Agriculture's Food and Nutrition Service (FNS) and the
states' ongoing quality assurance efforts, we accepted the
computerized food stamp data as reliable.  To provide additional
confidence in the data's accuracy, we reviewed a limited number of
food stamp case files at social service centers in six large
metropolitan areas:  Los Angeles County, California; Orange County,
California; San Diego County, California; San Antonio, Texas; Tampa,
Florida; and New York City, New York.  We compared the name, Social
Security number, and date of birth of individuals in the computerized
data with information in the relevant case files and found the
differences to be insignificant.  We did not attempt to verify dates
of benefit issuance because many case files did not contain complete
information. 

The reliability of SSA's death information was confirmed in our prior
report, which found that nearly all the information was accurate,\3
and in a 1994 Internal Revenue Service review that reported that up
to 98 percent of the data was accurate.\4

To determine why the inclusion of deceased individuals in food stamp
households was not detected, we visited state agency officials in
Sacramento, California; Tallahassee, Florida; Albany, New York; and
Austin, Texas; to discuss and review policy and procedures for
verifying applicant data and any subsequent changes, including the
loss of a household member.  We discussed fraud detection and
computerized data matching efforts, quality control and assurance
efforts, and methods of food stamp issuance with state officials.  In
addition, at the social service centers in the six large metropolitan
areas we selected for review, we discussed the processes for food
stamp application, data verification, certification, and
recertification.  We discussed local fraud detection and computerized
data matching efforts.  While reviewing case files at each location,
we validated to the extent possible the death of a participant. 

To determine whether computer matching or other methods could
effectively identify deceased individuals included in food stamp
households, we discussed with agency officials in each of the states
we visited their opinions regarding the value of such a match.  We
contacted state officials who had implemented a match to identify
such deceased individuals to determine the cost, quality, savings,
and barriers to matching.  To determine the effort associated with
data matching to identify such individuals, we calculated the time
used by our programmer to develop and implement the match programs
and reviewed studies performed for FNS and the Office of Management
and Budget regarding the costs and effectiveness of matching
routines. 



(See figure in printed edition.)Appendix III

--------------------
\1 In a match between the state eligibility information and the Los
Angeles County and Orange County, California, food stamp issuance
information, we found that more than 89 and 75 percent, respectively,
of the eligible individuals participated. 

\2 While SSA's Death Master File is considered the most comprehensive
list of death information available in the federal government, some
deaths may not be reported; hence, it cannot be considered a complete
listing of all deaths. 

\3 Social Security:  Most Social Security Death Information Accurate
But Improvements Possible (GAO/HEHS-94-211, Aug.  29, 1994). 

\4 Review of the Use of Deceased Taxpayer Data, Department of
Treasury, Internal Revenue Service, Internal Audit Report, Dec.  14,
1994. 


COMMENTS FROM THE U.S.  DEPARTMENT
OF AGRICULTURE
========================================================== Appendix II



(See figure in printed edition.)


The following are GAO's comments on the U.S.  Department of
Agriculture's letter dated January 15, 1998. 

GAO COMMENTS

1.  We agree with USDA's observation that our methodology did not
allow an exact determination of the dollar value of benefits issued
on behalf of deceased household members and the extent to which the
benefits issued were actually used.  Our estimates of overpayments
are based on the number of deceased individuals we identified and
FNS' per person average monthly value of food stamp benefits provided
in each state.  Given the various factors that determine specific
benefit amounts--including household income, assets, costs of
housing, and the number of individuals in the household--actual
overpayments could not be determined without a detailed analysis of
the 26,000 cases we identified--the actual overpayments may have been
lower or may have been higher.  Our estimates are presented to show
the general magnitude of the problem.  The fact that almost 26,000
deceased individuals were included in food stamp households indicates
a significant, systemic problem that needs to be addressed. 

2.  Concerning USDA's view that data matching will not allow the
recovery of all overpayments, the states are required to recover
overpayments by establishing claims against households, and they have
several years to take action on those claims.  In addition, the
states can recover overpayments by reducing future food stamp
benefits.  While state agencies may be unsuccessful in recovering
overpayments from some households whose membership includes deceased
individuals, such as single-member households, the use of SSA's
comprehensive death information will still yield savings to the
program by (1) identifying deceased individuals in the application
process before any benefits are issued and (2) preventing additional
issuance once the deceased individual is identified.  Furthermore,
such efforts have the effect of reinforcing the requirement that
clients report changes in household composition. 

3.  Concerning USDA's view that the report does not discuss all the
costs of identifying and following up on information regarding
deceased individuals in food stamp households, we noted in this
report the results of a 1995 Food and Nutrition Service study of the
Income and Eligibility Verification System (IEVS), which is used to
identify unreported and underreported income.  The study found that
data-processing costs were approximately 2 cents per case, and
investigative followup and claim collection costs were about $5 to $7
per case for all cases, whether or not they were overpayments.  The
cost of identifying and following up on deceased individual cases
would be similar to those incurred under IEVS.  As such, our
recommendation to use the existing SSA system was made with the goal
of maximizing efficiency and minimizing costs by simply adding a
vital piece of information to reports that are already routinely used
by eligibility workers.  We continue to believe that we have
recommended the most cost-effective method for identifying deceased
individuals on food stamp rolls--one that would require state
agencies, as part of the normal course of their work, to follow up on
the information provided. 

4.  We have modified the report title to better capture the key issue
identified in this report; namely, that thousands of deceased
individuals are being inappropriately included in food stamp
households. 




(See figure in printed edition.)Appendix IV
COMMENTS FROM THE SOCIAL SECURITY
ADMINISTRATION
========================================================== Appendix II



(See figure in printed edition.)


The following are GAO's comments on the Social Security
Administration's letter dated January 16, 1998. 

GAO COMMENTS

1.  We continue to believe that our recommendation to the Congress
has merit.  As SSA concedes in its comment, the agency has death
information that if shared with USDA and the states would help reduce
the number of deceased individuals inappropriately included in food
stamp households.  However, SSA states that it does not have the
authority to disclose such information to USDA or the states.  In
this context, we believe that USDA could better ensure the integrity
of the Food Stamp Program if it had all the information already in
the hands of another federal agency.  If the Congress enacted
legislation enabling SSA to share this information with the states,
this legislation would help USDA reduce fraud, waste, and abuse in
this program. 

Such legislation could, but would not have to, amend the Social
Security Act.  The Congress has partially addressed the issue of
restricted state death data in a 1993 amendment to the Internal
Revenue Code.  Under this provision, the Internal Revenue Service may
not disclose federal tax information to a state unless that state
allows federal agencies to use restricted state death data for the
purpose of ensuring that federal benefits and other payments are not
erroneously paid to deceased individuals.  26 U.S.C.  6103 (d) (4). 
While this provision covers USDA, it does not include state agencies
administering federal benefit programs, such as the Food Stamp
Program.  If the Congress wished, this provision could be amended to
cover such state agencies. 

2.  We do not share SSA's view that making roughly 95 percent of its
death information available to the states administering the Food
Stamp Program is adequate to solve the problem of deceased
individuals being counted as members of food stamp households.  For
the four states included in our review, we identified almost 26,000
deceased individuals that were being counted as members of food stamp
households.  While in this instance we cannot determine what
proportion of these individuals were identified from SSA's data that
are currently withheld from USDA and the states, previous GAO work
related to this issue has demonstrated that using SSA's data was
extremely useful in identifying and reducing overpayments in other
federal benefit programs.\1 As part of that work, we estimated that
millions could be saved annually if agencies had access to the
restricted death information. 

In addition, FNS data regarding the characteristics of food stamp
households provides evidence that a significant number of deceased
individuals included in food stamp households may be excluded from
the data that SSA currently makes available to states.  As is noted
in this report, SSA only provides information through the State
Verification and Exchange System (SVES) on deceased individuals who
were Supplemental Security Income (SSI) or social security
beneficiaries.  In 1995, FNS reported that 83% of individuals in food
stamp households did not receive SSI, and that 87% did not receive
social security income.  In the final analysis, therefore, we don't
believe there is merit in SSA's view that the information it already
provides is good enough.  We believe there is no sound reason why
USDA and the states should not have the benefit of all the
information possessed by SSA as they strive to reduce fraud, waste
and abuse in the Food Stamp Program. 

3.  We have modified the footnote to table 1 to reflect SSA's
comment. 


--------------------
\1 Status of Agency Use of SSA Death Information (GAO/HRD-93-31R,
July 20, 1993). 


MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V

Keith Oleson, Assistant Director
David Moreno, Project Leader
Leo Acosta
Brad Dobbins
Don Ficklin
Jerry Hall
Alan R.  Kasdan


*** End of document. ***