Unemployment Insurance: Millions in Benefits Overpaid to Military
Reservists (Letter Report, 08/05/96, GAO/HEHS-96-101).

Pursuant to a congressional request, GAO determined the amount of
unemployment insurance (UI) paid to military reservists, focusing on:
(1) why UI claimants do not report reserve income; (2) the
administrative and legislative options available to prevent future trust
fund losses; and (3) how these options will affect reservists' retention
rates.

GAO found that: (1) active UI claimants did not report more than $7
million in reserve income for fiscal year 1994; (2) the average amount
of nonreported income varied from $273 to $959 per claimant, and
resulted in UI benefit overpayments of $3.6 million; (3) most UI benefit
overpayments went to Army Reserve personnel; (4) federal trust fund
losses from the Unemployment Compensation for Ex-Servicemen Program
totalled $1.2 million; (5) the UI system paid over $25 billion in
benefits and received over $26 billion in state and federal unemployment
tax revenues; (6) the integrity of the UI system is adversely affected
by improperly paid benefits; (7) these overpayments hinder the UI
system's ability to provide unemployment benefits, contribute to high
state employer payroll taxes and federal outlays, and lower claimants'
benefit levels; (8) UI claimants do not report their reserve income
because they do not understand the reporting requirements, receive
improper information regarding their reporting responsibilities, and
have incentives not to report reserve income; (10) claimants are rarely
penalized for not reporting their reserve income; (11) states can
withhold a portion of a reservists' future benefits until applicable
overpayments are repaid; (12) it is difficult to verify reservists'
benefit levels without online access to federal wage data; and (13)
nonreporting of reserve wage income will not affect the military's
retention rates.

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

 REPORTNUM:  HEHS-96-101
     TITLE:  Unemployment Insurance: Millions in Benefits Overpaid to 
             Military Reservists
      DATE:  08/05/96
   SUBJECT:  Financial management systems
             Overpayments
             Military pay
             Unemployment insurance
             Military reserve personnel
             Unemployment insurance benefits
             Reporting requirements
             Payroll records
             Noncompliance
             Unemployment compensation programs
IDENTIFIER:  Unemployment Insurance Program
             California
             Colorado
             Georgia
             Massachusetts
             Pennsylvania
             Unemployment Compensation for Federal Employees Program
             Unemployment Compensation for Ex-Servicemen Program
             Federal/State Extended Unemployment Compensation Program
             District of Columbia
             Puerto Rico
             Virgin Islands
             Florida Unemployment Insurance Program
             Texas Unemployment Insurance Program
             Unemployment Insurance Trust Fund
             
******************************************************************
** 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 Honorable
Patricia Schroeder, House of Representatives

August 1996

UNEMPLOYMENT INSURANCE - MILLIONS
IN BENEFITS OVERPAID TO MILITARY
RESERVISTS

GAO/HEHS-96-101

UI Overpayments

(205275)


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

  EB - extended benefits
  ETA - Employment and Training Administration
  DFAS - Defense Finance and Accounting Service
  DMDC - Defense Manpower Data Center
  IRS - Internal Revenue Service
  UCFE - Unemployment Compensation for Federal Employees
  UCX - Unemployment Compensation for Ex-Servicemen
  UI - Unemployment Insurance
  USCG - United States Coast Guard

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


B-258100

August 5, 1996

The Honorable Patricia Schroeder
House of Representatives

Dear Ms.  Schroeder: 

The Congress established the national Unemployment Insurance (UI)
system in the 1930s to provide partial income assistance to many
temporarily unemployed workers with substantial work histories. 
Today, UI is the major federal program providing assistance to the
unemployed.  Many workers covered by the UI system are also among the
1.1 million personnel currently participating in the national Reserve
forces.\1 The Reserve augments the active military services,
providing trained troops for active duty in time of war or national
emergency or at such other times as national security requires. 

Most UI claimants are required to report the income they receive
while in the Reserve so that state UI programs can reduce their
benefits accordingly.  You were concerned that some claimants may not
be complying with this requirement.  Because noncompliance can result
in UI benefit overpayments and may contribute to a weakening of the
fiscal integrity of the UI system, you requested that we determine
the following: 

  -- How much Reserve income is not reported by UI claimants, and
     what are the implications in terms of federal and state UI trust
     fund losses? 

  -- Why do UI claimants not report Reserve income? 

  -- What administrative or legislative options are available to
     prevent future losses? 

  -- How will these options affect Reserve force retention, if at
     all? 

To estimate the amount of nonreported Reserve income, benefit
overpayments, and losses to federal and state UI trust funds, we
matched fiscal year 1994 UI claimant data from seven selected states
with Reserve force payroll and personnel data.\2 To identify
explanations for the nonreporting of Reserve income and possible
options to enhance reporting, we interviewed UI Service officials
from the Department of Labor's Employment and Training Administration
(ETA) to obtain information on the operation and structure of the
federal-state UI system.  We also held discussions with state UI
program officials to obtain information on claimant eligibility
procedures and their treatment of claimants with income from the
Reserve.  In addition, we met with Department of Defense officials
representing numerous Reserve operations, Department of
Transportation officials representing the Coast Guard Reserve, and
officials from the state National Guard Bureaus to obtain their views
on Reserve personnel compliance with UI program requirements and the
potential impact of reporting enhancement options on Reserve force
retention.  We conducted our audit work between August 1994 and May
1996 in accordance with generally accepted government auditing
standards.  For additional information on our methodology, see
appendix I. 


--------------------
\1 The Reserve has seven components across all the major military
services and the Coast Guard:  the Army National Guard, Army Reserve,
Naval Reserve, Marine Corps Reserve, Air National Guard, Air Force
Reserve, and the Coast Guard Reserve.  Reservists receive income for
participation in monthly weekend Reserve drills, week-long annual
training duty, and the deployment of Reserve units in the event of a
national crisis or domestic emergency. 

\2 The seven states were California, Colorado, Florida, Georgia,
Massachusetts, Pennsylvania, and Texas.  We selected these states
based on their average state unemployment rate during fiscal year
1994, the percentage of the nation's total Reserve personnel based in
each state, their geographic balance, and their previous experience
with matching the benefit and payroll records of state UI program
claimants with income from Reserve force participation.  We did not
independently verify the data provided to us by the states or by
federal agencies.  (See app.  I.)


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

Our analysis of benefit and Reserve data from seven states shows that
some Reserve personnel are receiving improper benefit payments from
state UI programs.  In the seven states included in our analysis, we
estimate that UI claimants who were active participants in the
Reserve failed to report over $7 million in Reserve income in fiscal
year 1994.  This led to UI benefit overpayments of approximately $3.6
million, of which federal trust fund losses were about $1.2 million. 
We expect that the federal and state trust fund losses from all UI
programs are much greater, because the seven states we reviewed
account for only 27 percent of all reservists. 

Program officials from the seven states we visited and Labor
identified several reasons why some claimants who are reservists did
not comply with program reporting requirements.  First, many UI
claimants participating in the Reserve may be unaware of their
reporting duties.  For example, some UI claimants may believe that
Reserve participation is a civic duty rather than employment covered
by the UI program.  In addition, many UI programs do not inform UI
applicants or claimants about their reporting requirements.  Federal
and state UI program officials told us that in most states, program
procedures, application forms, and informational materials either do
not inquire specifically about Reserve income or do not inform
claimants about these requirements. 

State programs we visited also do not affirmatively attempt to
discover that a claimant has nonreported Reserve income.  Because
reporting Reserve income will often reduce their weekly UI benefit
amounts, knowledgeable claimants may be reluctant to volunteer this
information.  However, those few states that have attempted to
systematically detect overpayments of UI benefits to reservists have
had difficulty verifying claimant reports of Reserve income because
they cannot obtain automated Reserve payroll data from Defense.  This
requires them to resort to more labor-intensive manual audits.  Our
own matching effort suggests that this inability to effectively
obtain automated data may be related to possible deficiencies in
Defense's payroll information system. 

Among the options identified by Labor and state UI officials for
improving the reporting of Reserve income are better informing
program claimants of their reporting responsibilities and improving
state UI program access to Reserve personnel and payroll data.  Most
Reserve officials we interviewed did not believe that increased
efforts to identify nonreported income would affect service retention
rates. 


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

The UI system is a federal-state partnership.  Within overall federal
guidelines, states operate their own UI programs, levy and collect
their own payroll tax, and determine the level and duration of
benefits and the conditions for benefit eligibility.  However, the
federal government, through the UI Service, a part of ETA, is
responsible for maintaining the fiscal integrity of the system,
including the individual state UI program trust funds.  The UI
Service provides information, guidance, and technical assistance to
programs in the 50 states, the District of Columbia, Puerto Rico, and
the Virgin Islands.  ETA officials reported that the UI Service also
monitors state methods for implementing administrative procedures,
comments on revised state procedures and revisions to the states'
Handbook of Operating Procedures, and reviews one-third of the
states' programs annually for compliance with program requirements
for the payment of benefits to exservice personnel and other federal
employees. 

The UI system includes several programs that cover most public and
private sector workers.  The regular UI program provides for up to 26
weeks of benefits to qualifying unemployed private sector and state
and local government employees.  Under this program, each state
maintains an account in the U.S.  Treasury, which is funded by state
payroll taxes on most private sector employers and by payments from
state and local government and some nonprofit employers reimbursing
the fund for benefits paid to former employees.  The Unemployment
Compensation for Federal Employees (UCFE) program is completely
federally financed and provides benefits to qualifying unemployed
civilian federal employees.  The Unemployment Compensation for
Ex-Servicemen (UCX) program is also completely federally financed and
provides benefits to service members after their discharge from
active duty as well as Reserve and National Guard personnel who have
been on active duty for 90 days of continuous service before their
release from active duty.  The UCFE and UCX programs are administered
by the states under agreements with the Secretary of Labor.\3

Eligible claimants may receive benefits from any one or a combination
of these funds on the basis of the nature of the claimant's prior
employment.  Most states reduce or offset the weekly benefit amounts
paid to a claimant by a percentage of any income earned by the
claimant during that week of program-covered unemployment.  The type
and amount of income used to offset benefits varies by state but
typically consists of wages from part-time employment, including
income received from a claimant's active participation in the
Reserve.  In 1994, 48 states,\4 the District of Columbia, Puerto
Rico, and the Virgin Islands offset UI benefits by a portion of the
income claimants received from Reserve service. 

Under current state UI laws, only private employers-- because they
pay a tax on some fraction of each employee's wages--are required to
regularly report wage information on their employees to states.  In
addition, states can impose financial penalties on private employers
who fail to comply with wage-reporting requirements.  Unlike private
employers, however, because federal, state, and local government and
some nonprofit employers are not assessed a payroll tax but rather
reimburse state UI programs for any benefits disbursed to their
former employees, they are not required to report employee wage
information to the states and other jurisdictions that operate UI
programs. 

Personnel and payroll data for the Reserve forces are maintained at
six different locations.  Each of the four military services (Army,
Air Force, Naval Reserve, and Marine Corps) collects and processes
personnel and payroll information from their respective active and
Reserve units at one of the four centers operated by the Defense
Finance and Accounting Service (DFAS) throughout the nation; the
Coast Guard maintains its data at the Department of Transportation's
United States Coast Guard (USCG) Pay and Personnel Center.  Although
the DFAS centers and the USCG Pay and Personnel Center are
responsible for collecting and maintaining current data, aggregate
historical payroll and personnel information for all services is
maintained by a central repository at the Defense Manpower Data
Center (DMDC) in Seaside, California. 

Some of our previous audits have repeatedly identified problems
associated with Defense's military pay systems.  In particular, we
have cited the payroll systems operated by the DFAS centers as
inaccurate, unreliable, and duplicative, resulting in a waste of
federal resources and an impairment of Defense operations.\5 For
example, we found that the DFAS payroll system for active duty Army
military personnel did not accurately summarize and report payroll
information.  Thus, DFAS has been unable to ensure that it reports
accurate information to the Army, much less to the Internal Revenue
Service (IRS) and other federal and state agencies.  Part of the
problem may stem from each service maintaining a unique and
independent payroll system. 

As a result of our earlier findings and recommendations in reports
citing weaknesses in Defense's financial management systems, DFAS has
been tasked with integrating the individual services' personnel and
payroll systems into one standardized system.  However, the Military
Pay Directorate at DFAS stated that the Marine Corps is the only
service that has made any progress toward this goal and that this
integration effort will take several years to implement. 
Furthermore, this effort may still not include the standardization of
all the payroll and personnel systems once it is completed. 


--------------------
\3 In addition, claimants may be eligible for extended UI benefits
under the Federal/State Extended Unemployment Compensation
program--commonly called the extended benefits (EB) program.  EB was
enacted by the Congress in 1970 to provide for a permanent extended
benefits program.  This program provides 13 weeks of additional
benefits to UI claimants following the exhaustion of the 26 weeks of
regular UI benefits.  States pay EB at the same rate as the weekly
benefit amount received by claimants under state law.  However, EB is
financed equally by federal funds and state tax revenues. 

\4 Oregon and Maine are the only states that exempt Reserve pay from
their UI offset provision in all instances.  In May 1995, Colorado
exempted Reserve pay for weekend drill and annual training
participation from its UI program offset requirements, leaving offset
coverage only in cases of participation during a national emergency. 

\5 See Financial Management:  Defense's System for Army Military
Payroll Is Unreliable (GAO/AIMD-93-32, Sept.  30, 1993) and Financial
Management:  Financial Control and System Weaknesses Continue to
Waste DOD Resources and Undermine Operations
(GAO/T-AIMD/NSIAD-94-154, Apr.  12, 1994). 


   NONREPORTING OF RESERVE INCOME
   RESULTS IN LARGE UI BENEFIT
   OVERPAYMENTS
------------------------------------------------------------ Letter :3

Our analysis of Reserve payroll and UI benefit data for seven states
that account for 27 percent of all Reserve personnel shows that UI
claimants who have been active participants in the Reserve did not
report over $7 million in fiscal year 1994 program-covered Reserve
income.\6 This nonreporting resulted in estimated UI benefit
overpayments of $3.6 million to over 11,500 Reserve personnel during
fiscal year 1994.\7 Thirty-two percent or $1.2 million were federal
trust fund losses primarily from the UCX program.\8

According to Labor officials, this suggests that exservice personnel
who were active in the Reserve accounted for a disproportionate
amount of nonreported income detected compared with reservists
separated from private or other public employers. 

Although we did not examine programs in the remaining states, the
District of Columbia, Puerto Rico, and the Virgin Islands, there are
several reasons to expect that the total federal trust fund losses
are much higher than $1.2 million.  First, these remaining programs
cover about 73 percent of the nation's reservists.  Most of these
programs treat Reserve income under their laws and program procedures
in a manner similar to the seven states we reviewed.  In addition, we
used an extremely conservative method for estimating the nonreporting
of Reserve income and associated overpayments in the seven states. 
For example, we excluded from our analysis all reservists who
reported any earnings.  Thus, we did not estimate the nonreported
income and benefit overpayments generated by claimants in the Reserve
who may have declared income to their UI program from some source but
who did not report all or part of their Reserve income (that is,
declared income from their annual training period but not their
weekend drills).  For more information on our methodology, see
appendix I. 

Although the amount of nonreported income and associated trust fund
losses may be small in comparison with the billions of dollars in
total annual program benefits,\9 the existence of overpayments is
enough to raise concerns about the effectiveness of the fiscal
control exercised over the UI system.  According to state and federal
program officials we interviewed, the integrity of the UI system is
adversely affected whenever claimants are improperly paid benefits,
either through oversight or fraud.  These unnecessary payments erode
the UI system's ability to provide benefits to those workers who are
unemployed through no fault of their own.  They contribute, if only
marginally, to higher state employer payroll taxes and federal
outlays and possibly lower claimant benefit levels than would
otherwise prevail. 


--------------------
\6 Among the seven states we examined, the average amount of
nonreported income per claimant varied from $273 in Texas to $959 in
California. 

\7 California had the largest average benefit overpayment per
claimant, $543.  Although a larger portion of total overpayments went
to members of the Army Reserve, probably because the Army comprises
64 percent of all reservists, the average overpayment among all
reservists did not vary significantly across services. 

\8 Although we calculated the amount of nonreported income and the
associated benefit overpayments for all members of the Naval, Marine,
and Coast Guard Reserves in the seven states, neither DMDC nor DFAS
was able to provide us with accurate data for the Army and Air Force
Reserve components.  After a delay of 8 months, we developed a
methodology for obtaining data on a sample of cases to complete our
analysis.  Thus, the estimates for the Air Force Reserve and National
Guard and the Army Reserve and National Guard are based on a
statistical sample.  (See app.  I.)

\9 In 1994, the UI system paid over $25 billion in benefits and
received over $26 billion in state and federal unemployment tax
revenues. 


   WHY CLAIMANTS DO NOT ALWAYS
   REPORT RESERVE INCOME
------------------------------------------------------------ Letter :4

State officials cited various reasons why claimants may not be
reporting their Reserve income while receiving UI benefits. 
According to state officials, the claimants may not understand the
reporting responsibilities, are not specifically informed of these
responsibilities, and may have incentives not to report all Reserve
income--incentives that are amplified by the states' limited ability
to detect nonreporting. 


      MANY UI CLAIMANTS MAY BE
      UNAWARE OF REQUIREMENTS TO
      REPORT RESERVE INCOME
---------------------------------------------------------- Letter :4.1

Many claimants may be unaware or have misconceptions about their UI
program's income-reporting requirements.  Some federal and state UI
program officials told us that many claimants believe that Reserve
participation does not affect their ability to seek work or to fill
their "regular hours of duty"--a key condition for UI benefit
eligibility.  Thus, they may believe that Reserve participation does
not constitute employment with reportable income.\10 For most
claimants, income from Reserve service is often earned part-time on
weekends, rather than through full-time employment Monday through
Friday.  Thus, reservists may also believe that the small amount of
earnings from Reserve participation they receive relative to their
primary employment earnings may not be regarded as reportable income. 
State program officials also believe that some participants view
their Reserve service as an instance of civic duty and patriotism
rather than employment; thus, they do not consider compensation
received from Reserve participation to be reportable income.  DFAS
officials noted that they could help notify Reserve personnel of
their income-reporting responsibilities regarding state UI benefits
by informing them of their duties in a note on their leave and
earnings statements. 


--------------------
\10 In most states, to maintain eligibility, UI claimants must be
able to do and be available for full-time work, registered with the
state's job service, and actively seeking work. 


      UI CLAIMANTS NOT
      SPECIFICALLY ASKED TO REPORT
      RESERVE INCOME
---------------------------------------------------------- Letter :4.2

Most UI programs throughout the nation require prospective claimants
to report all expected earnings--including Reserve income--received
during the benefit period as well as all earnings received during the
base period.\11 However, state program claims processors in the
states included in our review told us that they do not specifically
ask claimants whether they are receiving Reserve income and most do
not inform claimants of the Reserve income-reporting requirement in
writing.  Federal and state program officials we interviewed believe
that procedures in these seven UI programs were typical of the
procedures and materials used in UI programs generally, with most
programs not providing explicit information to claimants about their
reporting responsibilities regarding Reserve income. 

None of the UI application forms in the seven states inquired about
applicants' receipt of Reserve income.  For example, although the
application forms for five of the seven states contained questions
relating to military service, none included a question regarding
Reserve income.  California and Georgia forms asked if the
prospective claimant had served in the Armed Forces during the past
18 months.  Florida and Texas forms asked if the prospective claimant
was in the military service.  Massachusetts asked if the prospective
claimant was a veteran and about active duty.  Colorado and
Pennsylvania forms did not contain any questions regarding military
service. 

Though all seven states provided illustrations in their UI program
brochures and handbooks of the type of earnings claimants must
report, only two states--Colorado and Massachusetts--provided any
material that explicitly mentioned that claimants must report Reserve
income.  In addition, the UI handbook, a key source of program
information provided to all prospective claimants by the UI offices
in all the states we reviewed, generally did not elaborate on the
types of income that need to be reported.  Only the Massachusetts
handbook specifically addressed Reserve pay, stating that this income
must be reported.  To maintain continuation of their benefits, state
UI offices routinely ask claimants to recertify their unemployment
status and to report any income they receive during their benefit
period.  None of the recertification forms for the seven states we
visited specifically asks claimants whether they are receiving
Reserve income. 

Some state UI officials explained that they do not believe the
application forms are a reason for reservists not reporting income. 
They said that these application forms have been streamlined over the
years and that they believe the current questions on active military
service are sufficient to remind an applicant to report Reserve pay. 
However, most of them also agreed that their handbooks could be more
specific in instructing the applicants to report Reserve income.  ETA
officials believed that including questions on the application forms
that only refer to military service is not sufficient to identify
Reserve income.  In particular, they were concerned that state UI
workers may incorrectly assume that claimants include information on
Reserve income when they answer the written application questions
regarding active military duty and not pursue this issue further when
initially screening prospective claimants.  ETA officials were also
concerned that because Reserve income is not derived from active
military duty, applicants may knowingly or unknowingly evade
disclosure of Reserve income when answering the questions on military
service. 

ETA officials regarded the application procedures and guidance used
in the seven states we visited as typical of most states and thought
that most programs' procedures should be more specific to elicit
information about Reserve income.  They believed that more states
should list Reserve pay specifically as a type of income that needs
to be reported at the time of application and that this income should
be listed in the UI handbook provided to prospective claimants. 


--------------------
\11 The base period is typically the year before the year in which
the initial benefit claim is made.  Income earned during a claimant's
base period determines a claimant's weekly benefit amount. 


      INCENTIVES NOT TO REPORT
      RESERVE INCOME AMPLIFIED BY
      THE STATES' LIMITED ABILITY
      TO DETECT NONREPORTING
---------------------------------------------------------- Letter :4.3

In most states, income earned by UI claimants above a minimum level,
including wages from Reserve service, offsets or reduces their weekly
UI benefit amounts.  Because claimants will receive reduced UI
benefits by reporting Reserve income, there is an incentive not to
report this income.  For example, a claimant who is eligible to
receive the maximum weekly benefit amount of $250 under the Florida
UI program and who also receives $100 in weekend Reserve drill pay
would see his or her weekly UI benefit reduced by $66 and receive
$184 for that week.\12 In addition, in Florida as in most states, a
claimant participating for a full week of annual Reserve training
would be completely ineligible for UI benefits during that week. 

Further, claimants appear to face little risk of detection if they do
not report Reserve income.  Despite state penalties for fraud,
including reduction or loss of benefits,\13 state UI officials
believed that in many states claimants face little risk of detection
if they do not report Reserve income.  To enforce such penalties,
state UI programs must match reported claimant income with Reserve
earnings.  However, unlike private employers who must routinely
report quarterly wage and employment information on all employees
when they remit their payroll taxes, federal employers such as the
Reserve have no such requirements because they reimburse states for
UI benefits paid as they occur.  Thus, although UI programs have
on-line access to private sector employee wage data to verify benefit
levels and duration, they have no comparable access to federal wage
records.  Without ongoing access to federal wage data for all
reservists, states must conduct periodic matches\14 of UI claimant
data with Reserve personnel and payroll information to detect the
nonreporting of income.\15


--------------------
\12 The Florida program's offset provision is triggered when earnings
exceed $34 per week, specifying a dollar-for-dollar benefit reduction
for earnings above that level. 

\13 State UI laws provide for penalties against claimants who
fraudulently or otherwise fail to report all covered income earned. 
For example, in many states, once it is determined that claimants
have been overpaid benefits as a result of their nonreporting of
income, the state can withhold a portion or all of any future
benefits owed the claimants until the overpayments have been repaid. 
In other states, claimants found to be reporting only a part of their
earnings may lose their current benefits and become ineligible for
future benefit claims for some time.  See table II.2. 

\14 A state program would typically match its UI claimant list with
the roster of active Reserve personnel for the same time period.  For
all claimants who are also identified as active reservists, the state
program would then review payroll data on those claimants for the
period in which benefits and Reserve participation overlap.  For
Reserve income that was not reported, the state UI program could then
recalculate the weekly benefit amount and associated overpayment. 
(See app.  I.)

\15 Although it was outside the scope of this report, the
nonreporting of claimant income appears to be a broader problem
involving all UI claimants who were former federal civilian and
military employees, rather than just those participating in the
Reserve.  Officials from many of the state programs we visited
reported general difficulties in monitoring reported income from
claimants who were former federal employees. 


         STATE PROGRAM MATCHING
         EFFORTS CONSTRAINED BY
         LIMITED ACCESS TO RESERVE
         PAYROLL DATA
-------------------------------------------------------- Letter :4.3.1

In our discussions with state UI program officials we found that
states face several obstacles to conducting effective matching
operations.  These obstacles include UI programs' lack of awareness
of the availability of automated Reserve personnel records,
difficulties in obtaining Reserve payroll records from the DFAS
centers, and limited assistance from Labor.  DFAS' inability to
provide payroll information may actually reflect deficiencies in its
automated payroll data system. 

Few states have attempted to detect the nonreporting of Reserve
income in any systematic manner.  Our discussions with federal and
state UI program officials identified only three state
programs--Colorado, Pennsylvania, and Texas--that have conducted such
efforts in recent years and no program that has attempted to detect
nonreported Reserve income on a routine basis. 

Two of the state programs that matched UI program and Reserve
personnel and payroll data indicated that they were unaware of the
availability of automated information from DMDC that could have
expedited their effort.  Without this information, these states
resorted to slower, manual matches of information.  Pennsylvania and
Colorado--unaware that automated Reserve personnel rosters were
available from DMDC--requested printed personnel rosters.  State
program analysts then manually matched Reserve personnel rosters with
state UI claimant files--a time-consuming and labor-intensive
procedure.  (See app.  II.)

State officials also told us of their difficulties in obtaining
payroll records from DFAS.  For example, Colorado requested automated
payroll records for all reservists from DFAS but never received them. 
The state then asked the individual DFAS centers responsible for Army
and Air Force payment data for Reserve files; eventually the state
received printed leave and earnings statements, which necessitated a
manual file match.  The Texas state UI program experienced similar
difficulties.\16

States' experiences in working with DFAS in the past are similar to
our own efforts to obtain Reserve payroll information to match fiscal
year 1994 state UI claimant data with Reserve personnel and payroll
records for seven states.\17 Although the centers for the Coast Guard
and the Marines Corps provided us with complete Reserve wage data
within 1 month, we experienced great difficulty obtaining comparable
data for the other Reserve services.  It took almost 5 months before
we received limited data on Naval Reserve personnel; the Cleveland
DFAS center was unable to provide the actual wages for the Naval
Reserve personnel.  The center provided us with the dates worked and
the military pay grade for each reservist.  Because actual wages
earned were not provided by the Cleveland DFAS center, we had to
reconstruct the amount of wages, using pay charts that showed monthly
wages by pay grade, to determine the amount of wages that reservists
earned on a specific date.  Despite three attempts over 8 months,
DFAS was never able to provide us with automated wage information for
the Army and Air Force--about 70 percent of total Reserve personnel. 
The Denver and Indianapolis DFAS centers together were unable to
create for us a data tape linking the amount of wages earned by Army
and Air Force reservists to any of the specific dates we requested. 
For this reason, we ultimately had to rely on data that the DFAS
centers' staff manually extracted from printed and microfiche payroll
records.  The Military Pay Directorate at DFAS stated that the lack
of an integrated personnel and payroll system contributed to its
difficulties in providing us accurate and timely wage data for the
time periods and individual reservists we identified.\18

According to DFAS officials, we did not receive payroll data for the
Army and Air Force Reserve personnel requested because they were no
longer available.  Officials said that the centers typically maintain
the payroll information we requested for about 400 days, after which
time it is downloaded onto microfiche.  Although we had requested our
information well within the period during which the information was
accessible, by the time DFAS said it was prepared to provide the data
to us, officials stated that Reserve data files had already been
downloaded.  We ultimately obtained payroll data for a sample of
matches, which DFAS center personnel constructed from microfiche and
printed records. 

Labor has generally not facilitated the matching of Reserve and UI
data and does not assist states in obtaining payroll data from the
DFAS centers or DMDC.  ETA officials told us that they have not
received requests from states to assist in matching Reserve income
and UI benefit data and they have concentrated on providing
assistance in other areas, such as compliance with federal
requirements regarding the payment of UCFE claims.  However, they
acknowledged that this is an important area and one where they could
provide additional assistance to the state programs. 

Although the three states conducting matches were ultimately able to
detect nonreported Reserve income and UI benefit overpayments, the
difficulties they encountered in doing so led to their decision to
discontinue these efforts.\19 Although none of the officials we
talked to in these states had plans to continue matching, they said
they would likely reconsider their decision if either better guidance
and assistance from Labor or increased responsiveness from DFAS was
forthcoming. 


--------------------
\16 Pennsylvania did not attempt to obtain files from the DFAS
centers.  (See app.  II.)

\17 The DFAS centers provided payroll information for the Air Force,
Army, Naval Reserve, and the Marine Corps.  We received Coast Guard
Reserve wage data from the USCG Pay and Personnel Center in Topeka,
Kansas, that maintains payroll and personnel information for the
Department of Transportation.  The Coast Guard Reserve accounts for
about 1 percent of all Reserve personnel. 

\18 Although DMDC provided us with automated lists of Reserve
personnel, the agency only maintains aggregate monthly payroll data. 
Because UI benefits are generally paid on a weekly or biweekly basis,
we had to ask DFAS to provide us with comparable Reserve wage
information. 

\19 Colorado also discontinued its manual matching effort because in
May 1995 the state legislature exempted from the UI definition of
reportable wages income earned from weekend Reserve drill and annual
training exercises.  However, officials remarked that the matching
effort was extremely labor-intensive and if they had initially
realized the level of difficulty involved they would have
reconsidered the project for that reason alone. 


   REDUCING NONREPORTING OF
   RESERVE INCOME COULD PREVENT
   FUTURE LOSSES
------------------------------------------------------------ Letter :5

State and federal UI program officials and Defense officials
suggested several options to reduce the nonreporting of Reserve
income, which could prevent future trust fund losses.  These options
focus on more effective ways to inform claimants about their
reporting responsibilities and proposals to improve the detection of
nonreported income.  Most federal and state program officials believe
that these options could be implemented administratively. 

Defense, Labor, and state UI officials suggested several ways to
improve claimant awareness of their responsibility to report
Reserve-related income.  For example, DFAS officials suggested that
their agency could help notify Reserve personnel of their
income-reporting responsibilities administratively by informing them
of their duties in a note on their leave and earnings statements. 
Labor and state UI officials mentioned that programs could improve
claimants' awareness of their reporting responsibilities by revising
their application forms and handbooks to specify clearly their
programs' treatment of Reserve pay. 

State UI program officials generally agree that acquiring access to
Reserve personnel and payroll data could facilitate the detection of
nonreported Reserve income, although they identified a variety of
suggestions on the best way to obtain such access.  For example, some
officials suggested that states obtain automated records of Reserve
personnel and payroll data annually from DFAS to enable matching on a
regular basis.  Other officials believe that such regular data access
is unnecessary as long as the appropriate Defense agencies--DMDC and
DFAS--respond quickly when the states request wage information.  The
most frequent alternative suggested by federal and state officials is
to require Defense to report Reserve payroll and personnel data to
states on a quarterly basis, as private sector employers are required
to do, to permit verification of claimant income on a regular basis. 
Officials agree that this change could be implemented as an
administrative action; no legislative change would be needed. 

Some Labor officials believe that providing states with wage records
should be a requirement for all federal employers.\20 They believe
that the nonreporting of program-covered income by federal employees
generally is a far greater and more serious problem than nonreporting
by reservists alone.  Thus, state access to federal wage and
personnel information could significantly reduce the amount of
nonreported income and the associated benefit overpayments by
claimants separated from any federal employer. 


--------------------
\20 As part of Labor's efforts to improve UCFE claims processing, ETA
contracted with the Information Technology Support Center, a
public-private consortium, to explore the feasibility of providing
states with on-line access to federal wage data.  Under the project,
states will be able to electronically request claimant payroll
information.  The U.S.  Postal Service, one of the participants, has
projected a potential annual cost savings of about $1 million from
the effort.  Most of the savings accrue from freeing up resources now
allocated to respond to the large number of state program requests
for wage and employment verification. 


   REDUCING NONREPORTING OF
   RESERVE INCOME IS NOT EXPECTED
   TO AFFECT PERSONNEL RETENTION
   LEVELS
------------------------------------------------------------ Letter :6

Almost all the Department of Defense officials we interviewed,
including those representing the various Reserve components, do not
believe that the reporting of Reserve wage income for UI benefit
computation purposes would have a detrimental effect on their ability
to recruit and retain effective Reserve forces.  However, they also
prefer that states exempt Reserve income from any UI offset
requirements.  Despite the revenue loss to state UI programs, they
believe that reservists should not be penalized through the reduction
of UI benefits paid for an otherwise legitimate claim, because
reservists are performing an important national public service.  They
note that although most states have not exempted Reserve income in
calculating UI benefits, some states have; Colorado, for instance,
has exempted Reserve weekend drill and annual training income from
state UI program offset provisions. 


   CONCLUSIONS AND RECOMMENDATIONS
------------------------------------------------------------ Letter :7

The nonreporting of Reserve income results in the annual loss of
millions of dollars in state and federal UI benefit overpayments. 
Some nonreporting is attributed to claimants being unaware of their
reporting responsibilities.  To better inform claimants of their
reporting responsibilities, we recommend that the Secretary of
Defense direct the four DFAS centers to notify all reservists of
their income-reporting responsibilities with respect to state UI
benefits in a message included on their leave and earnings statement. 

We also recommend that the Secretary of Transportation direct the
USCG Pay and Personnel Center to notify all reservists of their
income-reporting responsibilities with respect to state UI benefits
in a message on their leave and earnings statement. 

ETA, in meeting its oversight responsibility for the financial
integrity of the UI system and in providing guidance and technical
assistance to the state UI programs to enhance their operations, can
also help to improve compliance with state income-reporting
requirements.  In particular, we recommend that the Secretary of
Labor direct ETA's UI Service to provide assistance and encourage
state UI programs to review the administrative forms or procedures
used to gather information about a prospective or continuing
claimants' wages, making revisions as necessary to clearly identify
to claimants the types of Reserve income they must report for the
offset of benefits. 

To reduce income nonreporting and the associated benefit overpayments
effectively, states also need better and more timely access to
Reserve payroll and personnel data.  Obtaining such data could help
detect nonreported Reserve income.  In addition, Transportation,
Defense, and state UI program officials believe that providing states
with Reserve data would have little or no impact on service retention
rates. 

We recognize that to be successful in this effort Defense agencies
must be able to provide accurate payroll and personnel information in
a timely fashion.  For this reason, we recommend that the Secretary
of Defense direct the DMDC and the four DFAS centers to develop a
process for giving states Reserve personnel and payroll data in a
timely, economical, and efficient manner.  In doing so, they should
coordinate with Labor's UI Service to identify states' needs. 

In addition, we recommend that the Secretary of Transportation direct
the USCG Pay and Personnel Center to develop a process for giving
states Coast Guard Reserve personnel and payroll data in a timely,
economical, and efficient manner.  In doing so, it should coordinate
with Defense's DMDC and with Labor's UI Service to identify states'
needs. 


   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :8

We obtained comments on our draft report from the Departments of
Labor, Defense, and Transportation.  Labor and Defense provided
written comments, which appear in appendixes III and IV. 

Labor generally agreed with the information provided in the report
and noted that it was already taking steps to implement our
recommendation to assist states in their review of procedures
identifying prospective or continuing claimant's wages.  The
Assistant Secretary of Labor also provided technical comments that
have been incorporated into the background section of this report. 
The Department of Labor did question how improper benefit payments
could contribute to higher state employer payroll taxes and possibly
lower claimant benefit levels.  Although we stated that such effects
would be quite marginal, to the extent that overpayments were
reduced, state legislatures could choose to use those savings to
incrementally raise claimant benefit levels or reduce employer taxes. 

The Department of Defense also generally concurred with the findings
and agreed to take actions to implement our recommendations and
provided methodologies and completion dates for accomplishing these
actions. 

Department of Transportation officials representing the Coast Guard
did not take issue with the overall findings of the report.  The
Transportation Program Manager for the USCG Pay and Personnel System
and other officials agreed that some steps could be taken to assist
states in detecting overpayments to reservists.  They stressed,
however, that actions taken in response to the recommendations should
be cost-effective.  This concern is addressed in our recommendation
to the Secretary of Transportation to direct the USCG Pay and
Personnel Center to develop a process for giving states Coast Guard
Reserve personnel and payroll data in a timely, economical, and
efficient manner.  Transportation's suggested approach could address
the concerns raised by our recommendations. 

Transportation officials also indicated that on the basis of the data
presented in the draft of our report, it could be inferred that no UI
reporting concerns were identified for about 95 percent of all
reservists.  Officials suggested that the report could explicitly
mention the cooperation of the vast majority of reservists with the
UI program.  One Transportation official also stated that the report
does not offer a basis of comparison for the unfamiliar reader to
evaluate and understand the relative significance of the issues
identified.  For example, he stated that the data presented indicate
that the UI overpayments of $3.6 million identified in our sample
were spread over about 275,000 reservists in the seven states we
reviewed.  As a result, the average overpayment per reservist is
about $13 per year.  A similar calculation for the federal share of
UI overpayments results in an average share of about $4.25 per
reservist per year.  He stated that these calculations are not to
trivialize the significance of the overpayments, but rather to
provide perspective that could be useful in identifying appropriate
remedial actions. 

Regarding reservists' compliance with UI program reporting
responsibilities, we focused only on those reservists receiving UI
benefits who did not report any Reserve income.  We did not analyze
those reservists receiving UI benefits who only partially reported
Reserve income nor those in full compliance with income-reporting
requirements.  Consequently, although it is likely that most
reservists are in compliance, our data do not permit us to say that
no UI reporting concerns would be identified for the remaining
reservists. 

Finally, we believe our findings of $7 million in nonreported income
and $3.6 million in overpayments represent absolute amounts and
actually understate the loss because the seven states account for
only 27 percent of all reservists.  Also, we do not believe that the
average overpayment per reservist is a meaningful statistic for
assessing the significance of the problem.  The existence of
overpayments is enough to raise concern about the effectiveness of
the fiscal control exercised over the UI program.  Failure to rectify
the problem erodes the integrity of the UI program and it is
important that action be taken to correct the problem. 


---------------------------------------------------------- Letter :8.1

We are sending copies of this report to the Secretaries of Labor,
Transportation, and Defense and UI program directors in California,
Colorado, Florida, Georgia, Massachusetts, Pennsylvania, and Texas. 
Major contributors to this report are listed in appendix V.  If you
have any questions concerning this report, I can be reached at (202)
512-7014. 

Sincerely yours,

Carlotta C.  Joyner
Director, Education and
 Employment Issues


METHODOLOGY FOR MATCHING STATE UI
CLAIMANT DATA WITH RESERVE PAYROLL
DATA
=========================================================== Appendix I

We matched fiscal year 1994 UI claimant data from seven selected
states with Reserve force payroll and personnel data to estimate the
amount of nonreported Reserve income and benefit overpayments and
associated losses to the federal and state UI trust funds.  Although
we did not independently verify the accuracy of the data provided to
us by DMDC, the DFAS centers, the USCG Pay and Personnel Center, or
the seven state UI programs, we believe that this does not affect our
results.  The data sources we used were the only ones available, and
state UI programs would rely on these data sources to calculate any
benefit overpayments.  We obtained data on all persons who received
some UI benefit payment (regular, UCX, or UCFE) between October 1,
1993, and September 30, 1994,\21

from seven state UI programs--California, Colorado, Florida, Georgia,
Massachusetts, Pennsylvania, and Texas.\22 We judgmentally selected
these states using a variety of characteristics including

  -- a high unemployment rate during fiscal year 1994 (California),

  -- a large number of reservists (Georgia, Texas, Pennsylvania,
     Florida, and California),

  -- previous experience with matching UI claimants and Reserve force
     data (Colorado, Pennsylvania, and Texas),\23 and

  -- geographic balance. 

We requested most of the states' UI claimant tape files before our
site visits to these states.  To assist them in their preparation of
these tapes, we held discussions with state officials and provided
them a structured protocol listing the types of data and the magnetic
tape format they would need to provide us.  The tapes we received
contained the names of UI claimants, their social security numbers,
and other information related to the benefits paid. 

We created tapes containing only the social security number extracted
from the original state data tapes.  Consistent with the requirements
of the Computer Matching and Privacy Protection Act of 1988,\24 we
sent the tapes to and worked with officials from DMDC in Seaside,
California, to match the seven state data tapes of UI claimants with
Defense's Reserve personnel and payroll records.\25

DMDC agreed to match these tapes (with our on-site supervision) with
Reserve personnel employment data to identify UI claimants who were
in the Reserve.  Before the visit, we worked with DMDC technicians to
coordinate tape format between DMDC and the seven state programs.  We
developed a computer program that would identify those persons who
received UI benefits for the same period as they received Reserve pay
but who did not declare such income for benefit offset. 

Once matches were identified, DMDC segregated the data and placed the
information on tapes according to the military service branch in
which reservists were employed.  Because DMDC does not maintain
payment data on the dates of service for when payments are earned, to
facilitate our match, it sent the tapes to the particular payroll
centers responsible for the payroll of these personnel's respective
service branch.  In total, we requested payroll information on the
number of personnel in each Reserve component from the payroll
centers listed in table I.1. 



                         Table I.1
          
          Number of UI Claimants Who Concurrently
            Received Reserve Income, by Reserve
          Component at Each Payroll Center, Fiscal
                         Year 1994

                             Claimants
                            matched to
                               Reserve
Reserve component         payroll data  Payroll center
----------------------  --------------  ------------------
Army National Guard              7,459  Indianapolis DFAS
 and                                     center
 Army Reserve
Air National Guard               2,002  Denver DFAS center
 and
 Air Force Reserve
Naval Reserve                    1,569  Cleveland DFAS
                                         center
Marine Corps Reserve               416  Kansas City DFAS
                                         center, Missouri
Coast Guard Reserve                 83  USCG Pay and
                                         Personnel Center,
                                         Topeka, Kansas
==========================================================
Total                           11,529
----------------------------------------------------------
Most of the state UI programs were able to provide us with the tapes
within a month and DMDC was able to perform the matches within a few
weeks after our request.  In helping us match fiscal year 1994 state
UI claimant data with Reserve personnel and payroll for seven states,
only the USCG Pay and Personnel Center and the Kansas City DFAS
center were able to provide complete automated information for the
Coast Guard and the Marine Corps components, respectively.\26
However, it took almost 5 months for the Cleveland DFAS center to
respond to our request for fiscal year 1994 Naval Reserve personnel
payroll data.  Then, the DFAS center was unable to provide the actual
wages for the Naval Reserve personnel and instead provided pay scales
that we had to convert to wage amounts.  Despite three attempts over
an 8-month period, the Denver and Indianapolis DFAS centers
representing the Air Force and Army Reserve components were never
able to provide us with accurate and complete payroll information on
personnel matched.  Thus, to complete our assignment we developed an
alternative methodology relying on a sample of these Reserve
components and based on data that DFAS center staff manually
extracted from printed and microfiche payroll records. 

To estimate the amount of nonreported income and benefit overpayments
to UI claimants in the Army and Air Force Reserve components, we
selected a random sample of matched personnel--UI claimants from each
of the seven states who also were employed by the Army or Air Force
Reserve during the same period of time.  We then provided the social
security numbers and dates of wages received for each of these
reservists to the DFAS centers that manually reconstructed, by
searching microfiche, the corresponding payroll information. 


--------------------
\21 Texas only provided data on claims initiated during fiscal year
1994, excluding information on claimants who were receiving benefits
during fiscal year 1994 but who initiated their claim during fiscal
year 1993.  Thus the Texas file is a proportionately smaller file
than the other states we reviewed. 

\22 See app.  II for a detailed discussion of the characteristics of
each of the seven state programs and the interviews we had with state
UI program officials. 

\23 We chose states--Colorado, Pennsylvania, and Texas--with past
experience in matching UI claimant data with Reserve payroll
information because we believed that they could provide a greater
understanding of the problems of matching such data and of the
options available to address them.  We chose states with generally
higher numbers of reservists and higher unemployment rates during
fiscal year 1994 because we believed that these factors might
increase the likelihood that we would detect the nonreporting of
Reserve income. 

\24 The Computer Matching and Privacy Protection Act of 1988 (5
U.S.C.  552 (a)) authorizes agencies to disclose records to GAO.  The
act also imposes safeguards on the federal government's use of
computer records pertaining to individuals' income, employment, and
other personal matters.  As a third party, GAO cannot compare various
sets of data and release the results of such comparisons for purposes
of enforcement actions against individuals. 

\25 The National Guard office of each respective state also maintains
current personnel records for the Reserve components in that state
but does not collect payroll information. 

\26 These two components represent only 5 percent of the Reserve
force. 


   OVERPAYMENT CALCULATION METHOD
--------------------------------------------------------- Appendix I:1

We appended the associated DFAS or USCG centers' payroll data to each
matched UI claimant (a particular Reserve member identified as
receiving both UI benefits and Reserve income for a certain time
period).  We then applied the appropriate state UI program offset
provision to calculate the amount of nonreported program-covered
income and, using the claimant's eligible weekly benefit amount, we
estimated the magnitude of the overpayment.  Through this process, we
estimated the cumulative amount of nonreported Reserve income and UI
benefit overpayments for fiscal year 1994. 

From our analyses, we estimated that the seven states we reviewed
made millions of dollars in UI overpayments to UI claimants who were
active Reserve participants.  Table I.2 shows the breakdown of
nonreported income and overpayments by branch of service. 



                         Table I.2
          
          Nonreported Income and Overpayments for
             UI Reserve Personnel in the Seven
                  States, Fiscal Year 1994

                                       Federal
                        Overpaymen        loss
            Nonreporte   ts (total      (UCFE,
Reserve      d reserve          UI    UCX, and  State loss
component       income   benefits)       EB)\a      (UI)\b
----------  ----------  ----------  ----------  ----------
Army        $4,291,282  $2,181,163    $750,426  $1,430,737
 National
 Guard and
 Army
 Reserve
Naval          903,521     494,108     138,570     355,538
 Reserve
Marine         142,005      61,722      14,771      46,951
 Corps
 Reserve
Air          1,610,025     871,317     259,233     612,084
 National
 Guard and
 Air Force
 Reserve
Coast           62,245      34,773       4,700      30,073
 Guard
 Reserve
==========================================================
Total       $7,009,078  $3,643,083  $1,167,700  $2,475,383
----------------------------------------------------------
Note:  Because the overpayment amount and nonreported income for the
Army and Air Force were estimated from samples, these amounts and the
associated estimates of federal and state loss are subject to
sampling error.  The sampling errors, calculated at the 95-percent
confidence level, range from ï¿½ 16.7 percent for federal losses due to
Army overpayments to ï¿½ 5.2 percent for federal losses due to Air
Force overpayments. 

\a We included extended benefits (EB) overpayments for three states
(California, Georgia, and Massachusetts) that provided extended
benefits during fiscal year 1994.  We recognize that a portion--less
than $73,000-- should be attributed to state losses but the data did
not permit us to apportion these funds exactly. 

\b We included in the state fund losses joint UI/UCX and UI/UCFE
overpayments of which a portion should be attributed to the federal
losses. 


   SOME MATCHES EXCLUDED FROM
   ANALYSES
--------------------------------------------------------- Appendix I:2

Although we did not examine programs for the remaining states, the
District of Columbia, Puerto Rico, and the U.S.  territories, there
are several reasons to expect that the total federal trust fund
losses are much higher than $1.2 million.  First, these remaining
programs cover 73 percent of the nation's reservists.  Most of these
programs treat Reserve income under their laws and program procedures
in a manner similar to the seven states we reviewed.\27

In addition, we used an extremely conservative method for estimating
the nonreporting of Reserve income and associated overpayments in our
seven states.  In calculating the amounts of nonreported earnings and
UI overpayments, we excluded from our analysis all Reserve claimants
who reported any earnings.  About one-third of the reservists
receiving UI during fiscal year 1994 reported some earnings. 
However, we were unable to determine if these earnings were Reserve
pay or other types of income because state UI program data files do
not include the source of income listed on the application form. 
Thus, we did not estimate the nonreported income and associated
benefit overpayments generated by claimants in the Reserve who may
have declared income to their UI program from some source but who did
not report all or part of their Reserve income (that is, declared
income from their annual training period but not their weekend
drills). 


--------------------
\27 We have little reason to believe that reservists' behavior
regarding the reporting of income is systematically different in
these remaining programs compared with the seven states we reviewed. 
In addition, although 1994 state unemployment rates--a primary factor
influencing the total number of program claimants and thus the pool
of possible claimants with nonreported income--tend to be higher in
the seven states we reviewed, they are fairly high in many of the
other UI programs as well. 


SUMMARY OF CASE STUDIES OF STATE
UI PROGRAMS
========================================================== Appendix II

To identify explanations for the nonreporting of Reserve income and
possible options to enhance reporting, we spoke with UI officials
from seven states:  California, Colorado, Florida, Georgia,
Massachusetts, Pennsylvania, and Texas.\28 These officials include UI
program directors and administrators, benefit payment and quality
control unit staff,\29 application clerks and reviewers, and computer
staff who conduct matches of UI program information with claimant
income and other data.  In addition, state officials provided us with
copies of their programs' benefit applications and continued claims
forms as well as handbooks and other publications used to explain
program eligibility and benefit payment requirements. 


--------------------
\28 We judgmentally selected these states according to a combination
of criteria, including a state's average unemployment rate, its
percentage of total Reserve personnel, and its previous matching
experiences.  For a discussion of our methodology, see app.  I. 

\29 Among the responsibilities of benefit payment and quality control
staff are claimant enrollment and eligibility determination, benefit
certification, and the detection of benefit overpayments. 


   STATES GENERALLY REDUCE UI
   BENEFITS BY A PORTION OF
   RESERVE INCOME
-------------------------------------------------------- Appendix II:1

Reservists receive pay for several types of activities, including
monthly weekend drill sessions and a 2-week annual training session. 
In addition, reservists can be activated for indefinite periods of
time during a designated national crisis or domestic emergency. 

States offset UI weekly benefit amounts by certain types of income
earned, including Reserve wages earned during the period of benefit
receipt.  Of the UI programs in the 50 states, the District of
Columbia, the Virgin Islands, and Puerto Rico, only Oregon and Maine
completely exclude all Reserve wages from benefit computation.\30
State program requirements, including those of the seven states we
visited, vary in the type of Reserve pay claimants must report and in
the formula used to offset this income against weekly benefit
payments (see table II.1).  Although all seven states we visited
offset weekly benefit amounts by a claimant's earnings for the 2-week
annual training session, only five states offset benefits for income
earned from the monthly weekend drill sessions.  For instance,
although California requires that claimants report all income
received from Reserve participation, the claimant's benefit amount is
not offset by income from monthly weekend sessions, according to
state officials. 



                         Table II.1
          
              State Program Variations in the
             Treatment of Reserve Income and in
          Benefit Calculations, as of January 1995

              Type of         Earnings
              Reserve pay     allowed before       Maximum
              included for    UI benefit            weekly
              offset          offset               benefit
State         calculation\a   calculation\b       amount\c
------------  --------------  --------------  ------------
California    Annual          $25 up to               $230
              training,       $100, then 25%
              national        of wages over
              crisis, and     $100
              domestic
              emergency

Colorado      All\d           1/4 of the              $267
                              weekly benefit
                              amount

Florida       All             $34 flat                $250
                              amount

Georgia       Annual          $30 flat                $195
              training,       amount
              national
              crisis, and
              domestic
              emergency

Massachusett  All             1/3 of the              $336
s                             weekly benefit
                              amount

Pennsylvania  All             40% of the              $340
                              weekly benefit
                              amount

Texas         All             25% of the              $252
                              weekly benefit
                              amount
----------------------------------------------------------
\a Reserve income includes pay for (1) monthly weekend drills, (2)
annual training, and (3) national crisis or domestic emergency
deployments. 

\b The amount of income from employment earned while receiving
benefits that a claimant can earn before benefit offset is
calculated. 

\c As of January 1995; these figures do not include any dependent
allowance benefits. 

\d Colorado excluded weekend Reserve drill and annual training pay
from its UI program offset provision beginning in May 1995. 

States also allow claimants to earn a certain amount of income from
part-time employment before reducing their UI benefits, disregarding
certain amounts of part-time income in their offsetting of benefits. 
The exact amount varies by state; California, for example, disregards
$25 of the first $100 per week of income earned and excludes 25
percent of earnings above $100 per week.  Thus, a California UI
claimant who received weekly part-time earnings of $150 would have
$38 disregarded from any offset to his or her UI benefits (that is,
$25 of the first $100, plus 25 percent, or $12.50--rounded to $13--of
the remaining $50, or a total disregard of $38); thus his or her
benefits would be reduced by $112 ($150-$38). 


--------------------
\30 In May 1995, Colorado exempted Reserve pay for weekend drill and
annual training participation from its UI program offset
requirements. 


   EXPLANATIONS FOR THE
   NONREPORTING OF RESERVE INCOME
-------------------------------------------------------- Appendix II:2

Officials from all seven state UI programs and the Department of
Labor told us that the nonreporting of claimant income from Reserve
participation was a serious problem, even though they did not know
the magnitude of the total dollars involved.  These officials believe
that they are responsible for preventing improper payments of UI
benefits to claimants who are ineligible for such benefits or to
claimants whose benefits should be reduced because of nonreported
earnings.  Accordingly, they believe that the integrity of the UI
program is adversely affected when claimants receive improper
benefits.  These officials view their efforts to detect such
overpayments as a means to deter future program abuses. 

State UI officials identified several explanations for the
nonreporting of Reserve income and subsequent benefit overpayment. 
First, many claimants may not understand their responsibility to
report Reserve income.  Second, state efforts to inform claimants
about UI program income-reporting requirements may be inadequate. 
Third, claimants may have incentives not to report all Reserve
income. 

State officials also acknowledged that many UI programs could take
additional steps to ensure that claimants are aware of program
requirements regarding Reserve earnings.  Most recognize that they
have an opportunity to inform claimants of reporting responsibilities
during the initial benefit application and the weekly or biweekly
claim recertification.  When applying for benefits, most applicants
are required by states to report all expected earnings, including
Reserve wages, to be received during the benefit period.  However,
during the application interview, state program officials generally
do not ask claimants whether they are receiving income from Reserve
participation.  Also, because such wages are not included in any
determination of an initial weekly benefit amount, applicants may
leave UI interviews believing that they do not have to report Reserve
income during subsequent periods for which they will be receiving
weekly benefits.  Nevertheless, only two states--Colorado and
Massachusetts--explicitly asked that Reserve income be reported.\31
In the remaining states we visited, no mention was made on the
application form, recertification notice, handbook, or any of the
other information or guidance given to the claimant of the
requirement or need to report Reserve income. 

UI program officials also noted the financial incentives for
claimants not to report any Reserve income, because such reporting
will reduce weekly UI benefits in most UI programs.  In cases where
UI claimants deliberately fail to comply with state reporting
requirements, states may invoke fraud statutes that allow them to
attach financial penalties to money owed them.  (See table II.2.)
Once states determine that claimants have been overpaid as a result
of their nonreporting of income, most states can withhold a portion
or all of any future benefits owed the claimants until the
overpayments have been paid back.  Besides reclaiming overpayments
from future benefits, one state--California--attaches a 30-percent
penalty to fraudulent overpayments, and claimants are expected to pay
this penalty in cash.  Another state--Texas--discontinues benefit
payments to claimants for the remainder of the benefit period and can
disqualify claimants for benefits for up to 52 weeks after
discovering that an overpayment has been made due to fraud. 



                         Table II.2
          
          State UI Program Penalties for Fraud and
           Nonreporting of Program-Covered Income

            Nonfraud penalty
            for failure to    Fraud penalty for failure to
State       report income     report income
----------  ----------------  ----------------------------
California  25% deducted      100% deducted from weekly
            from weekly       benefit payments until
            benefit payments  repaid plus 30% of
            until             overpayment amount as a
            overpayment       penalty to be paid in cash
            repaid

Colorado    100% deducted     100% deducted from remaining
            from remaining    weekly benefit payments
            weekly benefit    until repaid plus a 50%
            payments until    penalty of overpayments; in
            overpayment       addition, forfeiture of 4
            repaid            weeks of benefits for every
                              week of overpayment

Florida     100% deducted     100% deducted from remaining
            from remaining    weekly benefit payments plus
            weekly benefit    a 52-week disqualification
            payments until    from future benefits
            overpayment
            repaid

Georgia     50% deducted      100% deducted from remaining
            from remaining    weekly benefits until repaid
            weekly benefit    plus a 10% penalty;
            payments until    forfeiture of up to 52 weeks
            overpayment       of future benefits
            repaid

Massachuse  50% deducted      Total overpayments repaid
tts         from remaining    plus 12% annual interest
            weekly benefit    charge on overpayment amount
            payments until
            overpayment
            repaid

Pennsylvan  30% deducted      100% deducted from remaining
ia          from remaining    weekly benefit payments
            weekly benefit    until repaid; a lien can be
            payments until    attached
            overpayment
            repaid

Texas       100% deducted     Remainder of benefit period
            from remaining    forfeited; disqualified from
            benefit payments  benefits for 52-week period
            until
            overpayment
            repaid
----------------------------------------------------------
Despite penalties for nonreporting of income, state UI program
officials and Labor officials reported that enforcement was difficult
because Reserve wage and employment information is not readily
available to states for use in verifying claimants' earnings.  Unlike
private employers, who must routinely report quarterly wage and
employment information for all employees (thus permitting states to
determine whether claimants are accurately reporting their employment
status and wages for UI benefit calculation), federal employers,
including the Department of Defense and other civilian federal
employers, are not required to do so.  Without this information,
states are unable to identify income being received from a federal
employer when claimants apply for or receive UI benefits. 


--------------------
\31 In Colorado, a handbook prepared for UCX claimants cited Reserve
pay as income to be reported to the UI office at the time of
application or when such pay is earned while the claimant is
receiving benefits.  In the informational brochure distributed to
applicants, Massachusetts lists Reserve drill pay as income that must
be reported. 


   STATES USE MATCHING TO ENSURE
   PROGRAM'S FISCAL INTEGRITY
-------------------------------------------------------- Appendix II:3

In an effort to reduce overpayments and identify program fraud and
abuse, all seven states we visited matched UI program data with other
sources of claimant earnings information.  For example, they all
conducted computer matches of UI claimant files with state wage
record files submitted by private employers.  In addition, several
states, including Florida, Texas, and Pennsylvania, conduct targeted
matches to certain claimant groups.  For example, Texas matches UI
program data with private sector wage information on longshore
workers and employees who work for large manufacturers who
periodically initiate large layoffs.  States also conduct quality
assurance reviews to validate the continued eligibility of UI
claimants generally.\32 However, according to state quality assurance
officials, such reviews are unable to identify the nonreporting of
Reserve income unless claimants have already listed this type of
employment on their initial applications or continued claims
processing forms. 

State program officials told us that matches are an essential
internal control mechanism for maintaining the financial integrity of
their programs.  Although most of the state officials we interviewed
said that they have not conducted explicit cost-benefit analyses of
their matching efforts, they viewed these efforts as an effective
tool to deter future program abuse.  However, UI officials from
California reported that the revenue recovered from overpayments
identified by their automated matches was greater than the cost of
detecting and recovering these overpayments.  Most program officials
from other states told us that the costs of conducting automated
matches themselves were fairly small, though the costs of actually
recovering overpayments were much higher. 


--------------------
\32 During a quality assurance review, a state typically selects a
sample of claims from UI, UCX, and UCFE programs for an extensive
review, including an interview with the claimant or a questionnaire
that is sent to all of the employers listed on the claimant's
application forms. 


   STATES ENCOUNTERED DIFFICULTIES
   IN MATCHING UI BENEFIT DATA
   WITH RESERVE PERSONNEL AND
   PAYROLL RECORDS
-------------------------------------------------------- Appendix II:4

Three of the seven states we visited--Colorado, Pennsylvania, and
Texas--have conducted matches to identify the nonreporting of Reserve
income received by UI claimants.  Each of the states had initiated
the matches after receiving reports of UI claimants not reporting
their Reserve income.  However, the UI programs' lack of awareness of
the availability of automated Reserve personnel records, difficulties
in obtaining Reserve payroll records from DFAS, and limited
assistance from Labor hampered the progress of each match.  By June
1995, all three states had discontinued their matching efforts before
they were completed. 

Two of the state programs we visited that matched UI program and
Reserve personnel and payroll data indicated that they were unaware
of the availability of automated information from DMDC that could
have expedited their efforts.  Although Colorado and Pennsylvania
both determined that DMDC had records of Reserve personnel, neither
state was aware it could obtain this information in automated form,
which would have expedited the matching process considerably.  For
example, after obtaining printed Reserve personnel listings
containing over 13,000 names, Colorado employed about six full-time
staff to match those names with its UI claimants list.  The manual
matching continued for more than a year before the state closed the
project. 

The Texas state UI program, after some delay, received automated
personnel lists from DFAS through the state's National Guard unit. 
However, because it did not believe that the state of Texas had
authority to request such information, DFAS would not release the
payroll information on Reserve personnel that the UI program needed
for matching.  Because of the lack of payroll information, Texas
corresponded with the individual reservists receiving UI benefits and
requested their voluntary reporting of Reserve income.  In many
cases, Texas sent several requests before claimants provided the
appropriate wage information and because the Texas UI program was
unable to verify the information, the accuracy of responses and the
results of the match were questionable.\33

State officials told us that they have also received little
assistance from Labor's ETA with their efforts to identify
nonreported Reserve income.  Texas officials did report that ETA
regional officials successfully interceded with the Texas state
National Guard in obtaining automated Reserve personnel records. 
However, while ETA hired a contractor to develop a technical
assistance guide to help state benefit payment control units develop
matching techniques, the contractor provided an inadequate
description of how these units could identify the nonreported Reserve
income.  ETA officials, on the other hand, claim that they have not
generally received any requests from states asking for their
assistance in conducting matches; the agency has concentrated on
providing assistance in other areas, such as compliance with federal
requirements regarding the payment of UCFE claims.  Officials from
Pennsylvania and Texas have reported that they would not initiate
future matches without access to automated Reserve personnel and
payroll records and assistance from Labor.\34

Because each state discontinued its matching efforts before
completion, the amount of detected nonreported Reserve income and
associated benefit overpayments was very incomplete.  Nevertheless,
each state identified overpayments.  For example, state officials
reported that Colorado identified over $280,000 in benefit
overpayments on the basis of about 200 cases with nonreported Reserve
income; Pennsylvania projected about $96,000 in overpayments for 339
cases; and Texas detected $124,000 for 416 cases.  The Texas and
Pennsylvania overpayment totals were derived from information
self-reported by claimants, which likely understated benefit
overpayments. 


--------------------
\33 Officials reported that some claimants might be hesitant to
respond because claimants believed that they could be liable for UI
benefit overpayments. 

\34 In May 1995, Colorado's legislature exempted Reserve weekend
drill and annual training income from offset considerations, thereby
eliminating the need for the state to identify this type of
nonreported income. 


   SUMMARY OF STATES' MATCHING
   EFFORTS
-------------------------------------------------------- Appendix II:5


      COLORADO
------------------------------------------------------ Appendix II:5.1

Started in August 1993, Colorado's matching effort initially included
all Reserve components and was later narrowed to the Army and Air
Force National Guard.  Unaware of DMDC's automated records, Colorado
asked DMDC for a printed roster of Reserve personnel and conducted a
manual match of the Reserve personnel roster with state UI records. 
DFAS failed to provide Colorado with payroll records on matched
personnel, requiring data requests to the individual DFAS centers. 
The state manually matched each UI case file against the Reserve
payroll record, a time-consuming, labor-intensive effort.  Colorado
identified overpayment cases and initiated recovery actions.  The
state stopped its matching effort in May 1995 after passing
legislation to eliminate inclusion of Reserve income in the offset of
UI benefits. 


      PENNSYLVANIA
------------------------------------------------------ Appendix II:5.2

Initiated in May 1994, Pennsylvania's matching effort included all
Reserve components.  The state learned about the matching procedure
from Colorado officials.  Like Colorado, it requested a printed
roster of Reserve personnel from DMDC instead of automated files. 
Pennsylvania conducted a time-consuming manual match of the Reserve
personnel roster to state UI records.  It did not attempt to obtain
payroll records from DFAS.  As of May 1995, only one of its eight
regional offices had completed its time-consuming investigations and
the state suspended the initiative in June 1995. 


      TEXAS
------------------------------------------------------ Appendix II:5.3

Initiated in February 1990, Texas' matching effort included the Air
Force and Army National Guard.  Texas asked the state National Guard
to coordinate personnel information collection from DMDC, which did
so.  Labor's ETA facilitated procurement of automated Reserve
personnel records for the state's Air Force and Army National Guard. 
Texas conducted automated matches of its UI claimant files with
Reserve personnel records.  DFAS was unwilling to provide Texas with
automated or printed payroll records for matched files. 
Consequently, Texas relied on personal UI claimant responses for the
verification of Reserve income.  It has not compiled complete results
and has no plans to do so or to conduct future matches. 

Although all state program officials identified better monitoring and
matching of claimants' earnings as a solution, such efforts have been
seriously hindered, they told us, by a lack of automated payroll and
personnel information on reservists who receive UI benefits. 




(See figure in printed edition.)Appendix III
COMMENTS FROM THE DEPARTMENT OF
LABOR
========================================================== Appendix II




(See figure in printed edition.)Appendix IV
COMMENTS FROM THE DEPARTMENT OF
DEFENSE
========================================================== Appendix II



(See figure in printed edition.)


GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
=========================================================== Appendix V

GAO CONTACTS

Charles A.  Jeszeck, Assistant Director, (202) 512-7036
Christine M.  McGagh, Evaluator-in-Charge, (617) 565-7547
C.  Robert DeRoy, Assistant Director (Data Analysis and Evaluation
Support), (202) 512-6859

ACKNOWLEDGMENTS

In addition to those named above, the following individuals made
important contributions to this report:  J.  William Hansbury, Steven
R.  Machlin, Lori Rectanus, and Carol L.  Patey. 


*** End of document. ***