t> ()\)f_RTl c \We
\ RR('I.R l
-tf~P.,R 2 4 \.-83
\)e\los\\orf
Highlights of
activities in several
States and information
on USDA's new
Operation Awareness.
A report on
efforts to reduce
fraud, waste, and
abuse in the Food
Stamp Program
United States
Department of
Agriculture
Food and
Nutrition
Service
FNS-237
March 1983
Reducing Errors
in Certification
One of the major problems in certification
is that some food stamp applicants
inaccurately report their
incomes in order to qualify for
stamps. To counter this, some
States are working hard to make
food stamp recipients more aware
of their reporting responsibilities.
Other States are experimenting
with and perfecting various techniques
to certify households for
food stamps and reduce errors.
Project Integrity
South Carolina and Kentucky,
working with FNS' Southeast Regional
Office launched Project Integrity.
Through the media and letters
sent to each recipient household,
State agencies announced that review
teams were using sophisticated
methods to check the
accuracy of information on applications
and in food stamp case files.
Among these methods were home
visits and wage matching the applicant's
reported income with records
at the State employment
security commission. Recipients
were encouraged to update their
2
Introduction
Welcome to the Food and Nutrition
Service's ( FNS) first report on
strategies to reduce fraud, waste,
and abuse in the Food Stamp Program.
In the past months we have all witnessed
a number of instances
where Federal, State, and local efforts
have effectively increased the
integrity of the Food Stamp Program.
Your efforts to improve program
management are sincerely appreciated.
I hope that we will have
even more success in the future as
we continue and intensify our efforts.
We want to recognize these efforts.
"State to State" is just one
of many projects planned to acknowledge
FNS, State, and local
achievements under "Operation
Awareness," a national effort to
emphasize accountability and
strengthen public confidence in the
Food Stamp Program.
This report shares some States'
food stamp records before review
teams began their work. Many recipients
did update their re,cords as
a result.
The project's intent was twofold:
to reduce the amount of errors in reporting,
and to make recipients
more aware of their responsibility
"Project Integrity" press conference held in
Charleston, South Carolina in July 1982. Responding
to the media inquiries are: FNS
Southeast Regional Administrator David B. Alspach,
Bill Hall of the South Carolina Depart-successful
or promising actions to
combat fraud, waste, and abuse,
and improve the quality of the
Food Stamp Program. We hope
other State and local agencies will
be able to incorporate some of
these initiatives into their own programs.
We also report information
from the national front. We will try
to keep you informed on new initiatives
as they occur both within
States and on the Federal level.
We encourage you to help us exchange
ideas "State to State."
Send any information or comments
you would like to share to the Deputy
Administrator for Family Nutrition
Programs, Food and Nutrition
Service, 3101 Park Center Drive, Alexandria,
Virginia 22302.
Virgil L. Conrad
Deputy Administrator
for Family Nutrition Programs
Food and Nutrition Service
to properly report their circumstances.
In Charleston, South Carolina, reviewers
found problems in 98 of
the 376 cases they reviewed.
They found that in 26 percent of
the cases, households had been
issued a total of $66,525 worth of
ment of Social Services Office of Audit,
Harrison Rearden of South Carolina's Department
of Social Services Office of Economic
Services, and Charleston County Food
Stamp Director Bill Knowles.
food stamps that were above what
they were entitled to receive.
In Jefferson County, Kentucky,
1 09 cases of the 346 recipient
households reviewed contained
allotment errors in the amount of
$48,988. As an additional part to
Jefferson County's Project Integrity,
the FNS Compliance Branch investigated
35 retail stores because of
suspected violations of the program.
Of these, 22 had not complied with
the regulations, which resulted in
the disqualification of 8 firms.
In both South Carolina and Kentucky,
recipient claims have been established
and are being pursued.
Publicity resulting from the above activities
is continuing to have a positive
effect on compliance with
program regulations.
• For more information contact" Bill
Bradshaw, South Carolina Department
of Social Services (803)
758-7169, and Barbara Anderson,
Kentucky Department of Human Resources
(502) 564-7514.
Social Services Director Bill Knowles of
Charleston County, South Carolina answers
questions on "Project Integrity" from Joan
Mack of WCBD-TV.
Nevada Ranks Best in
Quality Control
Over the last 3 years Nevada
has maintained the lowest Quality
Control (QC) error rate of any State
in the Nation. While the national
rate for erroneous overissuance of
food stamps has hovered around
1 0 percent of food stamp dollars issued,
Nevada has consistently had
an overissuance error rate no
higher than 3. 7 percent. Its most recent
overissuance error rate was
2.8 percent, which earned enhanced
administrative funds from
USDA of $309,340.
Nevada attributes much of its success
in maintaining a low QC error
rate to an ambitious system of supervisory
review of cases and the linking
of the results of those reviews
to very detailed performance standards
for eligibility workers. Supervisors
in each local office review a
minimum of 35 cases per worker
per month. The average caseload
for a Nevada eligibility worker is
220 cases per worker. Job ratings
hinge on the results of these
supervisory reviews.
The parts of a case the supervisor
reviews are determined by
QC errors. Most emphasis is on
correct income and deduction
determinations and other important
eligibility criteria. L<ess important errors,
such as minor documentation
mistakes, while they are discussed
with workers, do not affect performance
assessments.
Nevada finds that cases in which
there is some type of income are
most prone to error. Therefore,
supervisors are required to review
at least the income and deduction
portion of all such cases before the
first month's benefits are issued.
Error rates are tabulated monthly
and become the basis for an
annual performance assessment of
the worker. An error rate over 4 percent
is considered unsatisfactory
and a repeated error rate over 6 percent
can lead to dismissal. The
worker is cited for all errors over
$5.00 similar to the QC system.
Monthly reviews always let workers
know where they stand regarding
their performance.
Given all of the pressure to perform
accurately, the State recognizes
an obligation to ensure that
the worker has the support necessary
to do the job. Training is the
key. Not only are new employees
given comprehensive training when
they start their tenure, but Nevada
maintains an ongoing training program
designed to update workers
and keep State administrative policy
consistent.
Robert E. Leard
'' What we want to do is
communicate our antifraud,
waste, and abuse efforts.
And when I say our, I mean
all of us. We need to do this
because we need to tell the
public that we are getting
the Food Stamp Program
under control by rooting out
fraud, waste, and abuse
and by tightening admin-istration.'
'
Robert E Leard,
Acting Administrator of FNS
Reno, Nevada
September 22, 1982
Workers having unacceptably
high error rates are counseled and
trained in their problem areas.
Only as a last resort will a worker
be fired, and only after several written
warnings that outline the
worker's problem area. However,
the good communication between
staff and supervisors that exists in
Nevada has resulted in very few adverse
personnel actions and low
turnover.
Nevada has made other efforts
to support its staff. The State
keeps its food stamp manual as
streamlined and understandable as
3
possible. It has also implemented
a modern new computer system.
This has freed both the workers
and clerical staff from many actions
that in the past were routine
yet very time consuming.
Finally, Nevada takes into account
the fact that about half of
the defects in the State's food
stamp cases are client caused. To
reduce client errors, Nevada
checks all cases against wage records
from its Department of Employment
Security for unreported
earnings or unreported receipt of unemployment
benefits. With the exception
of expedited service cases,
these records can usually be
checked before the first month's
food stamps are issued.
While Nevada has placed great
emphasis on running a tight program
and reducing its error rate, it
is still trying to balance the considerations
of accountability in the program
against the considerations of
client access to the program. Employee
performance standards are
not limited to eligibility determinations.
All Federal processing requirements,
including expedited
service, are monitored and are part
of the performance standards.
•For more information contact:
June Young, State Department of
Human Resources, Nevada (702)
885-5904.
New Jersey's Wage
Matching System
New Jersey operates a statewide,
centralized wage matching
system. It is an integrated system
including Aid to Families with Dependent
Children, General Assistance,
Medicaid and all food stamp households.
Wage matching for food
stamp cases has been in effect for
about a year. Matching is done
quarterly by checking social security
numbers with wage and unemployment
information on file in the
State treasury's computer system.
New Jersey is a wage request State.
Plans are being made to match
some New Jersey and Pennsylvania
wage information with New
Jersey data.
Only cases in which a discrepancy
of $1 01 or more exists between
earnings listed on applications
and earnings reported by
employers are followed up. The
4
Michael Greene (kneeling) and Keith Turner
(seated) of the New Jersey Department of Human
Services check social security numbers
Two verification specialists discuss a policy issue.
From left to right are Frances Hamlin
counties are provided a list of possible
fraudulent cases, and are required
to check out all information.
County prosecutors and other officials
put cases in order of priority
with wage and unemployment information
on file in the State treasury's computer system.
and James Ledford of Mecklenburg County,
N.C.
for action, depending on the size of
the actual discrepancy discovered.
County action is closely monitored
by the State agency.
New Jersey believes wage match-
1
J
ing has a deterrent effect on fraud.
When the system was first publicized,
approximately 3,258 welfare
recipients dropped off the rolls.
The State projects an annual savings
of $5,772,000 in the Food
Stamp Program alone. Approximately
31,575 discrepancies were
followed up the first year for all programs.
Approximately 6,368 cases
were referred to county fraud investigation
units for investigation and
possible prosecution. As a result
of wage matching, 1 ,399 food
stamp cases were disqualified, and
3,307 cases were closed or
adjusted.
• For more information contact: Michael
Greene, Department of Human
Services, (609) 890-9500 ext.
240.
Grants Helping to Reduce Errors
The following States are studying
various ways to reduce errors in
certifying households as eligible to
receive food stamps under yearlong
grants from the Food and Nutrition
Service.
New Mexico
This State's goal is to produce
reliable data on various techniques
that have been effective in verifying
household circumstances. The
long-range goal is to reduce quality
control errors.
The project involves applying vari-ous
verification techniques on
previously certified cases that are
considered to be error prone.
These techniques include 1) a historical
review of the case record, 2) a
collateral contact with a person designated
by the participant, 3) a collateral
contact initiated by the
eligibility worker, and 4) a home
visit.
The final phase of the project involves
a practical application test,
in local offices, of all or some of
the project results. Products from
this project include:
1) A final report that will
discuss-
• cost effectiveness of verification
techniques relative to
yielding information on the household's
circumstances.
• results from the practical
application test.
2) The procedures used to develop
the State error-prone profile.
3) A "how to" verification strategy/
interviewing techniques manual.
The final report is due February
1983.
• For more information contact"
Marilyn Manges, Department of Human
Services, (505) 827-2441.
North Carolina
Two counties in North Carolina
and the Research Triangle Institute
(RTI) are studying whether it would
be cost effective to use verification
specialists to interview and certify
"error-prone" households for food
stamps. The probability of error in
a household is predicted from an
error-prone profile analysis of State
quality control statistics. The profiles
used are statistically valid.
The team of verification specialists
work on fewer cases than are
usually done. All of their cases are
predicted to be "error-prone." The
interviews are about two to three
times longer than normal and are
more investigative, in an effort to understand
the household's total economic
profile.
Verification specialists help
recipients verify more information
on their economic circumstances
so the benefits they receive will be
accurate.
The study will measure the impact
of using intensified verification
on error rates. The study will also
assess how reliable State errorprone
profiles are in determining
how cases are treated in a particu
·lar local office.
Results should be available early
in 1983.
•For more information contact·
Ben Watts, Department of Human
Resources, (919) 733-9940.
New Hampshire
New Hampshire is comparing the
cost, timeliness, and accuracy of different
ways to receive periodic
financial information from food
stamp applicants and recipients in
two district offices.
The State agency asks applicants
and recipients to report information
to their case workers in one
of three ways-by mail, telephone,
or face to face. They are also
asked to report this information
over different time periods-every
month or every 2 or 4 months.
The State hopes to show how
they can best assign cases to each
reporting method and recertification
schedule to reduce dollar errors.
The State will then be able to
weigh how effective its action will
be against the administrative cost
from data it has gathered. Results
should be available in the spring of
1983.
• For more information contact"
Barry Bode/, Department of Health
and Welfare, (603) 271-4316.
5
Better Ways to Issue
Food Stamps
Several States are continuing various
efforts to test alternate issuance
systems in search for ways to
reduce the incidence of mail loss,
authorization-to-participate (ATP)
card loss, and duplicate issuance.
Arizona Improves Mail Loss Rate
Arizona has dramatically reduced
the number of food stamps that
are lost in the mail by expanding
its use of certified mail and implementing
postal changes. These
steps have helped cut losses from
1.22 percent from January to
March of 1982 to 0.58 percent in
April 1982. The State agency is
working with the U.S. Postal Service
to improve security of food
stamps in the mail system. The following
are steps that have been
taken:
• In Phoenix, the State agency
has begun giving the post office a
computer printout of all certified mailings
to provide an audit trail for all
certified envelopes put into the
mail system.
• With FNS' permission, the State
agency mails replacements for lost
food stamps sent via certified mail after
15 days, not 1 0 days. This
gives the State 5 more days to
check with the post office to verify
the loss. The extra 5 days allow
the Postal Service time to search
for certified mail receipts.
• The Postal Service has successfully
tested the use of presorted
(by zip code) and sealed mailings
to reduce the possibility of theft by
postal clerks. The State agency is
now considering expanding the presorted
mailings to the entire caseload.
•For more information contact"
Rick Burr, Department of Economic
Security, (602) 225-5639.
Michigan's On-line
Issuance System
Tremendous savings have resulted
from an on-line food stamp
issuance system that Michigan implemented
in the fall of 1981 in
Wayne County, which includes Detroit
and its suburbs.
Under this new system, a certified
food stamp recipient receives
by mail a plastic card embossed
with the household's name and
case number. This card, which
has a magnetic strip on it, replaces
the A TP card that was used before
the on-line system began.
The reciRient then takes the plastic
card to an issuance site where
the card is inserted into a reader,
which scans the magnetic strip and
transmits an inquiry to the statewide
computer system in Lansing.
The inquiry concerns the household's
eligibility status and whether
participation has already occurred
that month. If the recipient is eligible
and has not already participated,
the computer terminal prints
an authorization sheet that lists the
household's certification period, the
amount of its food stamp allotment,
and the book denominations the
household should receive. Partici-
---- -
Arizona is working with the U.S. Postal Serv- proving successful is using presorted (by zip
ice on ways to improve security of food code) and sealed mailings of food stamps.
stamps in the mail system. One way that's
6
Mary C. Jarratt
''Today, as never before,
those of us who administer
public assistance programs
are expected to run tight,
efficient programsaccountable
programs.
With fewer dollars to work
with, it is imperative for us
to make sure those benefits
are reaching the people
for whom they were intended.
Lost money-lost
through errors, mail loss,
cheating, theft-are benefits
that never reach the
families who deserve them.
That is a kind of 'budget'
cut that can be avoided!' '
Mary C. Jarratt,
Assistant Secretary of Agriculture
El Paso, Texas
October 24, 1982
pation for that month is then immediately
reconciled to the host system
in Lansing.
Advantages of the new on-line
system include:
• Dramatic decline in duplicate issuance
of food stamps.
• Elimination of undelivered, lost,
and stolen ATP's.
• Immediate information on food
stamp participation at any point in
the month and immediate reconciliation.
• Tighter controls on issuing and redeeming
food stamps.
• Reduced caseworker time on
resolving duplicate issuances.
• More timely and accurate reports
of the household's participation
history.
The cost savings for the first
year of on-line issuance (September
1981 to September 1982) compared
to the last year of A TP
issuance stack up like this:
• The postage costs decreased
from about $420,000 under the
ATP system to about $40,000 under
the on-line issuance system,
for a savings of about $380,000 a
year.
• The cost for replacements decreased
from about $2.2 million for
replacement ATP's to about
$160,000 for on-line issuance replacements,
for a savings of over
$2 million a year.
• The savings from not printing
A TP's and an A TP replacement report
amount to about $11,000 a
year.
• Under the ATP system, duplicate
authorizations averaged about
1 ,000 per month at a cost of about
$95,000. The on-line system averages
30-40 replacements a month
at a monthly cost of $3,000-$4,000.
Michigan has been so pleased
with the results of the new system
that it is planning to switch all its
counties still using the A TP systems
to the on-line system by December
1983. Other areas that
are using on-line issuance systems
are New York City, Florida, and
New Mexico.
• For more information contact:
Dave Wigent, State Department of
Social Services, (517) 373-0865.
Illinois' "Vendor
Chargeback System"
Since 1977, the Chicago area
(Cook County) has used a "vendor
chargeback system" that has resulted
in relatively low incidence of
overissuance of food stamp benefits.
The system places full financial liability
on issuance agents for illegally
redeemed A TP cards.
The Illinois Department of Public
Aid (IDPA) has an issuance agreement
with the Lincoln National
Bank located in the Chicago area.
The bank then subcontracts with local
currency exchanges to issue
food stamps. The issuance agents
are required by contract to witness
recipients' signing of ATP cards
and view valid identification. The
value of any ATP transactions redeemed
by means of a forgery is
charged back to the vendor who is
responsible for the transaction.
In Cook County, about $30 million
in benefits are issued each
month to approximately 265,000
households. Of this total, only 210
duplicate issuances of food stamp allotments
occur on an average each
month. This figure represents only
0.0126 percent of the total ATP
transactions. The IDPA has determined
that of these 210 duplicate allotments
issued, only 35 were to
people who were not authorized to
cash A TP cards. The remaining
175 involved household fraud,
which resulted in claims to recover
from households the extra food
stamps to which they were not entitled.
Beginning in January 1983, the
IDPA will streamline issuance of
food stamps in'Cook county by eliminating
the use of ATP cards and
by instituting a system of direct delivery
of food stamps to currency exchanges.
Households choose
which currency exchange they
want their food stamps to be de-
Currency exchanges in Chicago serve as
convenience bill paying and check cashing facilities.
This currency exchange in Chicago's
Loop also issues food stamps and,
under the "vendor chargeback system," is financially
responsible for any illegally redeemed
ATP cards.
livered to and, on a specific date,
they will go there to pick up their
stamps.
The household will no longer receive
an ATP card in the mail and
take it to a currency exchange for redemption.
With the elimination of
A TP cards, duplicate issuances are
expected to decrease even further
than the present low average of
210 a month.
However, the IDPA still plans to
maintain its "vendor chargeback system"
for those few cases when
food stamps are issued to unauthorized
persons or to households that
have committed fraud. With the
elimination of A TP cards and the institution
of the direct delivery system,
IDPA expects to save about
$2.3 million a year, 50 percent of
which represents Federal matching
funds.
• For more information contact"
Tim H. Grace, Illinois Department
of Public Aid, (217) 782-1355.
Claims-Getting Back
What Recipients Owe
State agencies can now keep 50
percent of their fraud claims collections
and 25 percent of their recipient
error claims collections. These
incentives, in conjunction with new
regulatory changes on fraud disqualification
and claims procedures,
should help to increase overall recipient
claims collection amounts.
The following States and territories
have had better than average
records of collecting claims. Their
simple, inexpensive, but efffective,
efforts are highlighted below:
Nonfraud Claims
• Some States like Virginia and
Ohio rely on personal contact with
a household to establish payment
schedules and encourage recipients
to pay back claims. Other
States also indicate that personal
contact with a household yields positive
results.
• In Guam, employees may make
home visits to find households that
have not responded to letters from
the State agency demanding the
household repay the benefits they
were overissued. Once households
are located, they must come
into the local office to set up a repay-
7
ment schedule. This system resulted
in one of the highest rates
of collection of nonfraud claims in
the first quarter of fiscal year 1982.
• Sacramento and Orange
Counties in California both followup
their State's "demand" letters for
nonfraud claims with their own
stronger letters. For instance, Sacramento
County's letter threatens
legal action for failure to respond.
Both counties report that response
to the letter is high.
• Although Puerto Rico no longer issues
food stamps, its collection record
on food stamps had been
noteworthy. In the first quarter of fiscal
year 1982, approximately
$35,000 in nonfraud claims was
collected by offsetting restored benefits
against outstanding claims.
Agency officials identified the casefiles
of participants with claim determinations
by flagging them with
green labels. Claims technicians
were thus alerted to the existence
of a claim and could offset the
claim before restored benefits were
issued.
Fraud Claims
• A State law lets Oregon collect
some fraud claims by deducting the
claim amount from the household's
State tax refund. Once each year,
a list of food stamp recipients with
outstanding debts is compared with
a list of people projected to receive
a refund from the State Comptroller's
Office. The State agency then
decides which households will have
the debt subtracted from any refund
they are due. Montana also
uses a tax refund offset system to
collect claims.
•For more information contactLouise
Eldridge, FNS National Office,
(703) 756-3496.
On the National Front
Logo for Operation Awareness
On December 6, 1982, Secretary
of Agriculture John R. Block was
briefed on Operation Awareness,
the FNS plan to enhance communication
among States in order to highlight
and continue efforts to
improve management of the program.
At the briefing, the symbol representing
the identity of Operation
Awareness was adopted. The de-
8
FNS' Deputy Administrator for Family
Nutrition Programs Virgil Conrad explains
some background materials on Operation
Awareness during a briefing for Secretary
Block held last December.
sign reflects the image of the
liberty bell that is on all food stamps.
FNS will be using the logo on helpful
information, correspondence,
and publications related to Operation
Awareness activities.
USDA Funds Help States
Combat Fraud
As of October 15, 1982 FNS was
reimbursing 44 State agencies for
75 percent of the funds they spend
to investigate fraud in the Food
Stamp Program. Normal funding levels
are at 50 percent.
To be eligible for the extra
money, a State must submit to
FNS a revised budget and certain
other information, including detailed
descriptions of the organizations
and activities to be claimed at the
75-percent rate. States can also obtain
75 percent of their expenses
for holding hearings and prosecuting
people accused of fraudulently
receiving food stamps.
New Jersey is one State benefiting
from these funds. In Mercer
County, for example, in fiscal year
1981, 1,593 cases were submitted
for investigations. Of this amount,
1 ,062 required further action, including
645 that were sent either to the
county prosecutor or prepared for
civil action. Food stamp fraud
claims totaled $437,500 for fiscal
year 1981. Thus far, $42,500 has
been collected and $1,800 recouped
through allotment reduc-tions.
- ,, .
New Jersey received $1'.6 ·million
from FNS in fiscal year 1981 for 75
percent of its investigation and prosecution
expenses.
• For more information contact: Michael
Greene, New Jersey Department
of Human Services, (609)
890-9500.
New Coupons to Deter
Counterfeiters
Recent efforts by local, State,
and Federal officials have turned
up incidents of counterfeited food
stamps. In Chicago, for instance, local
police officers cooperating with
a Federal task force arrested 40 people
who were trafficking in
counterfeit food stamps.
To help deter such counterfeiters,
last fall FNS began distributing
food stamps with enhanced
security features. The 1 0 dollar
stamps have a feature called a latent
image that can be detected by
looking at it from a certain angle.
All food stamps-1, 5, and 1 0 dollar
values-have a new microscopic
line that reads U.S.
Department of Agriculture.
These printing devices make
food stamps more difficult to
counterfeit and give investigators another
tool to find the source of the
counterfeiting. Grocers, bankers,
State and local law enforcement officials
can quickly determine if food
stamps are genuine. FNS plans to
continue to develop more effective
ways to deter counterfeiting.
• For more information contactAsher
Bryte, FNS National Office
(703) 756-3545.
Reports on Fraud,
Waste, and Abuse
• "Fraud in Government Benefit
Programs: Suggested State/Local
Prevention Strategies." Publication
number 81671. Office of Public Affairs,
National Institute of Justice.
(301) 251-5000, Publication, $6.95.
• "Development and Application
of Error Prone Profiles in the Food
Stamp Program." Research Triangle
Institute. Free copies are available
from Demonstration Project
Section, FNS (703) 756-3387.
This is an equal opportunity program. If
you believe you have been discriminated
against because of race, color, national origin,
sex, age, handicap, religious creed or
political beliefs, write immediately to the Secretary
of Agriculture, Washington, D.C.
20250.
Home Economics and Household Collection
