Pension Plans: Stronger Labor ERISA Enforcement Should Better Protect
Plan Participants (Letter Report, 08/08/94, GAO/HEHS-94-157).

The Department of Labor's Pension and Welfare Benefits Administration
(PWBA) is responsible for enforcing provisions of the Employee
Retirement Income Security Act of 1974 (ERISA), the federal program to
protect an estimated 200 million participants and beneficiaries of
private pension and welfare plans, as well as the $2.5 trillion in
assets held by those plans.  A review of Labor's enforcement program
shows improvements since 1986, but also the need to strengthen
enforcement by taking steps to ensure maximum use of investigative
resources.  PWBA has never evaluated its current enforcement strategy;
such an evaluation is needed to determine whether PWBA is focusing on
the right issues and whether the strategy produces the greatest results.
In addition, PWBA has done little to assess the effectiveness of
computer targeting programs developed to systematically select pension
and welfare plans for investigation of potential fiduciary violations.
The enforcement program also can be strengthened by increasing the use
of penalties authorized by ERISA to deter plans from violating the law.

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

 REPORTNUM:  HEHS-94-157
     TITLE:  Pension Plans: Stronger Labor ERISA Enforcement Should 
             Better Protect Plan Participants
      DATE:  08/08/94
   SUBJECT:  Pension plan cost control
             Funds management
             Law enforcement
             Internal controls
             Fines (penalties)
             Legal information systems
             Compliance
             Reporting requirements
             Prioritizing
             Regulatory agencies
IDENTIFIER:  PWBA Case Management System
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


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


Report to the Secretary of Labor

August 1994

PENSION PLANS - STRONGER LABOR
ERISA ENFORCEMENT SHOULD BETTER
PROTECT PLAN PARTICIPANTS

GAO/HEHS-94-157

Labor's ERISA Enforcement


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

  ERISA - Employee Retirement Income Security Act of 1974
  ESOP - employee stock ownership plan
  IRS - Internal Revenue Service
  MEWA - multiple employer welfare arrangement
  OCA - Office of the Chief Accountant
  PWBA - Pension and Welfare Benefits Administration

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


B-225405

August 8, 1994

The Honorable Robert B.  Reich
The Secretary of Labor

Dear Mr.  Secretary: 

This report describes the results of our review of certain aspects of
the Department of Labor's enforcement of the Employee Retirement
Income Security Act of 1974 (ERISA).  By strengthening the
enforcement program, Labor can better protect an estimated 200
million plan participants and beneficiaries and $2.5 trillion in
assets held by private pension and welfare plans. 

Labor's Pension and Welfare Benefits Administration (PWBA) enforces
ERISA's prohibited transaction and fiduciary requirements that ensure
private pension and welfare plans operate in the best interest of
plan participants, reporting and disclosure requirements that ensure
plans provide financial and other information to the federal
government and plan participants, and bonding requirements.  Our
report discusses the need to improve Labor's (1) enforcement
strategy, (2) methodology for targeting pension and welfare plans for
investigation, and (3) use of penalties to increase ERISA compliance. 

To perform our work, we reviewed selected PWBA enforcement documents;
analyzed information from PWBA's Case Management System and PWBA's 10
area offices; and interviewed PWBA headquarters, area office, and
other officials.  For more details on our review scope and
methodology, see appendix I. 


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

While Labor's enforcement program has improved since 1986, it can be
strengthened by taking steps to ensure maximum use of investigative
resources.  PWBA has never evaluated its current enforcement
strategy, which requires the allocation of a substantial percentage
of resources to investigate "significant issue" cases involving
financial institutions and service providers with a high potential
for ERISA violations.  Such an evaluation is needed to determine
whether PWBA is focusing on the right issues and whether the
allocation formula produces the greatest results, as measured by such
quantitative indicators as dollars recovered and participants
impacted. 

Furthermore, PWBA has done little to assess the effectiveness of
computer targeting programs developed to systematically select
pension and welfare plans for investigation of potential fiduciary
violations.  Investigators need this information so they can use
programs with the highest probability of identifying plans with
violations.  However, PWBA has no plans to test the programs in 1994
and has taken few steps to correct weaknesses in the procedures used
in the past to test programs. 

The enforcement program also can be strengthened by increasing the
use of penalties authorized by ERISA to deter plans from violating
the law.  Opportunities to identify and penalize plans for not filing
required annual reports are probably being missed since PWBA, because
of a lack of staff, does not routinely follow up on Internal Revenue
Service (IRS) referrals of plans that report financial and other
information one year but not the next.  Because of a lack of legal
resources in the Labor Solicitor's Office, PWBA has not been able to
take legal action to penalize welfare and certain pension plans
involved in prohibited transactions.  Also, given the restrictive
legal requirements that have limited the use of penalties for
violations of ERISA fiduciary requirements, PWBA needs to determine
whether additional administrative guidance, changes to the law, or
both are needed to remedy confusion associated with the penalty and
enhance PWBA's penalty enforcement. 


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

Labor and IRS have primary responsibility for enforcing ERISA
requirements.  PWBA enforces prohibited transaction, fiduciary,
reporting, disclosure, and bonding requirements.  IRS enforces
participation, vesting, and funding requirements. 

Labor established its current ERISA enforcement strategy in December
1986 after being criticized by us and others for not having a
comprehensive, consistent long-term strategy for enforcing the law
and selecting plans for investigation.  Since that time, the strategy
has focused on investigating "significant issue" cases with a high
potential for fiduciary violations, such as untimely cash deposits or
other imprudent management practices.  These cases involve financial
institutions that hold or manage plan assets, such as banks and trust
companies, and service providers that provide plans with dental,
vision, legal, accounting, and other services.  Through program year
1993 (October 1, 1992, to September 30, 1993), the strategy called
for PWBA to allocate at least 50 percent of its investigative
resources to significant issue cases, with no less than 20 percent to
be spent on either financial institution or service provider cases. 
The remaining resources were to be devoted to investigating general
cases. 

The goal of PWBA's strategy is to achieve the greatest possible ERISA
compliance by using resources effectively.  PWBA believes that
investigations of significant issue cases have a broader impact than
investigations of individual cases because financial institutions and
service providers often serve many plans and many participants. 
Consequently, when a fiduciary violation by a financial institution
or service provider is corrected, dollar recoveries and the number of
plans and participants affected are usually larger than when a
violation by an individual plan is corrected. 

PWBA has implemented the strategy by setting forth specific
requirements for its 10 area offices through its annual planning
process.  For example, PWBA's 1994 program year planning guidance
required a balance among types and sizes of plans selected for
general investigations, with a general rule that no more than 5
percent of all such cases should involve plans with fewer than 50
participants.  The guidance also emphasized investigations of
multiple employer welfare arrangements (MEWA) and employee stock
ownership plans (ESOP) as significant issue cases. 

PWBA uses several methods to select financial institutions, service
providers, and pension and welfare plans for investigation.  The
methods include referrals from IRS and other agencies, complaints
from participants, manual review of financial and other information
on plans' annual Form 5500 series reports,\1 spinoffs from other
investigations, special area office projects, and computer targeting. 

In 1990, PWBA developed a number of unique computer targeting
programs that search automated Form 5500 series information for
characteristics that PWBA believes indicate a high potential for
ERISA violations.  PWBA initially developed 95 programs, but reduced
the number to 81 for 1994 and subsequent program years to eliminate
duplicate and other unproductive programs.  The computer targeting
programs are used primarily to identify pension and welfare plans for
investigation, although some programs can be used to identify
financial institutions and service providers.  PWBA believes the
targeting helps focus its resources. 

The type of investigation conducted depends on the circumstances of
the case.  For example, PWBA must open a limited review for all plans
chosen to test the computer targeting programs.  Such reviews inquire
into one or more specific aspects of a plan or other entity operation
to quickly determine whether a complete fiduciary or criminal
investigation is warranted.  In addition, PWBA may open a fiduciary
or criminal investigation without conducting a limited review when
the available evidence is sufficient to justify doing so. 

When violations are identified, ERISA authorizes Labor to assess
civil penalties against the violators.  Labor may assess a penalty of
up to $1,000 per day against a plan administrator who fails or
refuses to file a Form 5500 series report or whose report is rejected
for failing to include material information.  When welfare plans or
pension plans that do not qualify for tax exemption are found to have
violated ERISA's prohibited transaction requirements, Labor may
assess parties in interest\2 a penalty up to 5 percent of the
prohibited transaction and up to 100 percent if the transaction is
not corrected within 90 days.  Labor must, with certain exceptions,
assess a penalty against a fiduciary or any person who knowingly
participated in a fiduciary breach that occurred on or continued
after December 19, 1989.  The fiduciary penalty is equal to 20
percent of the recovery amount agreed to in a settlement agreement
with Labor or contained in a court order. 


--------------------
\1 Most pension and some welfare plans annually file with IRS a Form
5500, 5500-C, or 5500-R report that contains financial and other
information.  Information from these forms is computerized by IRS and
shared with Labor. 

\2 As related to ERISA welfare or pension plans, a party in interest
includes fiduciaries, employees, persons who provide services to a
plan, employers or employee organizations whose members are covered
by a plan, or others (29 U.S.C.  1002 (14)). 


   ENFORCEMENT PROGRAM MAY NOT
   MAXIMIZE USE OF RESOURCES
------------------------------------------------------------ Letter :3

Since 1986, PWBA has increased its enforcement activities and
improved its enforcement results, as measured by such quantitative
indicators as dollar recoveries and participants impacted.  However,
the enforcement program may not ensure that scarce investigative
resources will be used as effectively and efficiently as possible in
the future.  Most area offices reported that allocating fewer
resources to significant issue cases and more resources to general
cases would improve enforcement results.  Officials in three area
offices said that the universe of financial institutions or service
providers with a high potential for ERISA violations may be
diminishing.  Furthermore, area offices may not be using computer
targeting programs with the highest probability of identifying plans
with ERISA violations because PWBA has not determined the
effectiveness of the programs.  In addition, area offices waste some
resources handling and tracking cases referred to and received from
IRS and other agencies upon which no action is taken. 


      ENFORCEMENT ACTIVITIES
      INCREASED AND RESULTS
      IMPROVED
---------------------------------------------------------- Letter :3.1

PWBA's enforcement activities have increased since program year
1986--the last full program year before the current strategy was
implemented.  PWBA Case Management System data show that by program
year 1993, the number of cases opened had more than doubled from
1,420 in 1986 to 3,250 and the number of cases closed had increased
from 1,674 to 2,998.  These enforcement increases were partly caused
by an increase in the number of limited reviews needed to test
computer targeting programs.  For example, 1,480 of the 2,998 cases
closed in program year 1993 had been opened to test computer
targeting programs.  The increase was also influenced by a 37-percent
increase in area office enforcement staff, which grew from 266 in
1986 to 365 in 1993. 

Enforcement results also have improved since program year 1988--the
first full program year under the significant issue strategy.  PWBA
uses a number of quantitative indicators to measure the effectiveness
of its enforcement program.  They include dollar recoveries, number
of plans impacted, number of participants impacted, number of cases
with monetary recoveries,\3 number of cases with other fiduciary
results (fiduciaries removed, diversification, investments stopped,
and administrative practices changed), number of cases with
nonfiduciary results (reporting, disclosure, and bonding violations),
number of cases with criminal indictments, number of individuals
indicted, number of cases referred for litigation (civil cases to
Labor's Solicitor and criminal cases to the Justice Department), and
number of cases with litigation filed.  When assessing enforcement
efforts, PWBA also considers such qualitative effects as presence in
the community and changes in industry practices and behavior.  Case
Management System data show that all but one quantitative indicator
improved between program years 1988 and 1993, as shown in table 1. 



                           Table 1
           
             Comparison of Selected Quantitative
           Enforcement Indicators for Cases Closed
                     During 1988 and 1993

Indicator                                 1988          1993
--------------------------------  ------------  ------------
Dollars recovered (millions)              $103          $183
Plans impacted                          33,824        72,199
Participants impacted (millions)          7.16         21.00
Cases with monetary recoveries             244           303
Cases with other fiduciary                  82           125
 results
Cases with nonfiduciary results            227           187
Cases with criminal indictments              9            29
Individuals indicted                        12            41
Cases referred for litigation               57           117
Cases with litigation filed                 27            38
------------------------------------------------------------
The improved results stemmed from both significant issue and general
case investigations.  Significant issue cases, which are generally
more complex and time consuming than general cases, accounted for 29
percent of all cases opened in program year 1993, 33 percent of all
cases closed, and 46 percent of total investigative time. 

At the same time, as illustrated in table 2, significant issue case
investigations closed during program year 1993 produced higher
monetary recoveries; affected more plans and participants; and
resulted in more cases with criminal indictments, individuals
indicted, and cases referred for litigation.  General case
investigations identified more plans with fiduciary and nonfiduciary
violations, and resulted in more cases with litigation filed.  With
the exception of cases with criminal indictments and individuals
indicted, similar results occurred in most program years since 1988. 



                           Table 2
           
             Comparison of Selected Quantitative
            Enforcement Indicators for Significant
            Issue and General Case Investigations
                        Closed in 1993

                                   Significant       General
Indicator                          issue cases         cases
--------------------------------  ------------  ------------
Dollars recovered (millions)              $107           $76
Plans impacted                          69,678         2,521
Participants impacted (millions)         17.08          3.91
Cases with monetary recoveries             107           196
Cases with other fiduciary                  44            81
 results
Cases with nonfiduciary results             48           139
Cases with criminal indictments             18            11
Individuals indicted                        27            14
Cases referred for litigation               79            38
Cases with litigation filed                 18            20
------------------------------------------------------------
Information obtained through a questionnaire we developed and
distributed to PWBA's 10 area offices indicated that some of the
improved enforcement results since program year 1986 are attributed
to the significant issue strategy.  At least eight area offices
reported increases in dollar recoveries, plans and participants
covered, and indictments between program years 1986 and 1992.  Half
the area offices said the number of plans with fiduciary violations
(that is, cases with monetary recoveries and other fiduciary results)
had increased.  Moreover, at least half of these offices said the
increases resulted in part from the strategy.  (See question 1 in
app.  II.)


--------------------
\3 Monetary recoveries include assets that were restored to the plans
and prohibited transactions that were corrected. 


      CHANGING THE SIGNIFICANT
      ISSUE STRATEGY MAY FURTHER
      IMPROVE ENFORCEMENT RESULTS
---------------------------------------------------------- Letter :3.2

Despite the success of the significant issue strategy since 1986,
most area offices believe that refocusing investigative resources
would further improve enforcement results.  In response to our
questionnaire, six area offices said that fewer resources should be
allocated to significant issue cases and seven said that more
resources should be allocated to general cases.  Most of these area
offices said that reallocating investigative resources would enable
them to maximize dollar recoveries, court-ordered corrections,
voluntary corrections, and criminal convictions, while increasing the
pension industry's awareness of Labor's enforcement role.  In
addition, nine area offices wanted more discretion over the size of
the plans they investigate.  (See questions 2 through 4 in app.  II.)

Six of the seven area offices we visited indicated that it may be
time to change the significant issue strategy.  Four of the six said
that the resource allocation requirement should be reduced.  Two of
these four and one other office said that the universe of financial
institutions or service providers was diminishing.  Four offices
offered suggestions for refocusing resources, including spending more
time on small plans, defined contribution plans,\4 defined benefit
plans sponsored by bankrupt sponsors, and plans that offer life
insurance. 

PWBA headquarters officials noted enforcement difficulties associated
with some of these suggested alternatives.  For example, civil
recoveries from plans with bankrupt sponsors are often limited
because typically there are insufficient assets available.  Criminal
prosecution in cases involving small dollar amounts is sometimes
difficult because U.S.  Attorney offices do not have the resources to
pursue all cases.  Some area offices have worked with state and local
prosecutors on small cases the U.S.  Attorney offices might have
opted not to pursue. 

In addition, the PWBA Field Focus Group established by Labor in 1993
as part of President Clinton's effort to "reinvent" the government
recommended that the ERISA enforcement strategy be revised.  In its
October 5, 1993, report to the Assistant Secretary for Pension and
Welfare Benefits Administration, the Group said that the strategy may
no longer focus enforcement efforts properly; they noted that
enforcement efforts should change as issues change and should respond
to new developments.  The Group recommended that no more than 20
percent of available time be devoted to any particular strategy
component and that strategy issues, projects, and emphasis be chosen
annually and run no more than 2 years. 

PWBA revised its resource allocation formula for program year 1994. 
It reduced the requirement for significant issue cases from 50
percent to 40 percent and removed the requirement that at least 20
percent of investigative resources be spent on either financial
institution or service provider cases. 

While reducing the resource allocation formula was partially
responsive to the Field Focus Group recommendations and may be viewed
favorably by area offices that told us the prescribed formula was too
high, the change does not respond to the Focus Group's recommendation
that strategy issues be chosen annually and last no more than 2
years.  Moreover, PWBA had no empirical data showing that 40 percent
is the resource allocation formula that produces the greatest
enforcement results.  Information provided by two of the seven area
offices we visited indicated that the staffing resources needed may
differ by area.  One area office suggested that the resource
allocation should be between 25 and 33 percent and one suggested that
it should be between 30 and 40 percent; five did not suggest a
formula. 

In commenting on a draft of this report, Labor said that it never
represented that the formula would produce the greatest enforcement
results.  Rather, Labor said, the formula was established to make
certain that area offices paid sufficient attention to significant
issues and that policy priority was firmly established.  Labor now
believes that specific direction from headquarters on the formula is
probably unnecessary. 


--------------------
\4 Defined contribution plans are pension plans that provide an
individual account for each participant and base benefits on
accumulated contributions, earnings, and forfeitures to the account. 


      LITTLE DONE TO DETERMINE
      EFFECTIVENESS OF COMPUTER
      TARGETING PROGRAMS
---------------------------------------------------------- Letter :3.3

Results of a PWBA consultant's tests of computer targeting programs
available from 1990 to 1993 do not provide sufficient information to
determine the effectiveness of those programs.\5 The consultant
reported that 14 of 19 programs tested the first 2 years were
"general successes" based on PWBA's interim criteria.\6 Under these
criteria, a program was deemed successful if limited reviews required
by area offices for test purposes were converted to full
investigations more than 5 percent of the time.\7 This 5 percent rate
is PWBA's estimated "baseline" conversion rate for investigations
resulting from manual review of Form 5500 series data.  Instead of
actual results, conversion rates were used as success indicators
because PWBA wanted timely results and some investigation results are
not known for 3 years.  Ideally, test results would be based on
completed investigations and actual violations identified. 

PWBA revised its computer targeting programs for program year 1994,
but made no specific plans to test the programs.  In April 1994,
headquarters officials told us that PWBA would be reevaluating
computer targeting program testing and would consider the
recommendations in our September 30, 1993, letter to the Assistant
Secretary for Pension and Welfare Benefits Administration.  (See app. 
III.) In that letter, we pointed out the following weaknesses in
testing procedures used to evaluate the predecessor programs: 

  Because PWBA did not randomly select plans for investigation (it
     selected the highest ranking plans), test results were not
     representative of all plans identified by the targeting
     programs. 

  By using a more appropriate sampling formula, PWBA could test the
     programs with fewer investigations than required in the past. 

These weaknesses should be corrected if PWBA continues to test one
program at a time.  In addition, we continue to believe that PWBA may
be able to target plans more effectively and efficiently by using
multivariate analysis to analyze computer targeting programs in
combination with one another rather than individually. 


--------------------
\5 Mathematica Policy Research, Inc., Evaluation of Selected Computer
Targeting Programs (May 22, 1992) and Evaluation of Selected Plan
Year 1989 Computer Targeting Programs (Feb.  28, 1994). 

\6 Programs were tested to determine whether they were general and
specific successes.  The general success rate included targeted cases
that identified any type of violation.  The specific success rate
included targeted cases that identified the specific violation the
program was designed to discover. 

\7 To be considered successful with Labor's desired level of
confidence, the conversion rate for sampled cases had to be at least
5 percentage points higher than the 5 percent baseline rate.  That
is, the conversion rate had to be greater than 10 percent. 


      STAFF RESOURCES ARE WASTED
      HANDLING REFERRALS ON WHICH
      NO ACTION IS TAKEN
---------------------------------------------------------- Letter :3.4

PWBA and other federal agencies waste scarce staff resources handling
referrals of cases with suspected fiduciary violations on which no
action is taken.  For example, data reported by area offices to PWBA
headquarters showed that PWBA did not investigate over 90 percent of
the 3,894 IRS referrals it received between 1990 and 1993, usually
because the referrals involved small plans (1,595 cases), plans that
had already corrected the identified violation (681 cases), or plans
with bonding (504 cases) or other nonfiduciary violations.  At least
nine area offices said that they had not opened investigations on
some referrals from IRS and other agencies in program year 1992 for
these same reasons.  (See questions 10 through 13 in app.  II.)

While information on the cost associated with such referrals is not
readily available, area offices estimate that the cost is small. 
About half of PWBA's area offices estimated costs of handling such
referrals.\8 These offices estimated that they spent about $4,600 in
fiscal year 1992 on 535 referrals from IRS and other agencies on
which the area offices took no action.  They also estimated that they
spent about $1,600 in fiscal year 1992 to refer 115 cases to other
agencies that took no action on the cases.  (See questions 20 through
21 in app.  II.)

A Labor/IRS memorandum of understanding governing coordination
between the two agencies requires that IRS refer to PWBA all plans
with identified ERISA violations.  PWBA area office and IRS district
office officials said that the agreement leaves IRS no discretion in
deciding which cases to refer.  As a result, referrals are made
regardless of whether corrective action has already been taken or
whether the plan has terminated.  Moreover, IRS refers cases that
PWBA does not investigate because of PWBA's focus on large plans. 
PWBA program planning guidance limits the time area offices can
expend investigating plans with fewer than 50 participants, but the
bulk of IRS referrals involve such plans.  PWBA headquarters
officials told us that despite the inefficiency, they want all
identified violations referred to the area offices so the offices can
decide whether to open an investigation. 

PWBA's Field Focus Group suggested revising the Labor/IRS memorandum
of understanding.  Among other things, the Group recommended that
existing procedures be revised to eliminate IRS referrals of plans
with fewer than 10 participants and to require IRS to refer reporting
violations directly to PWBA's Office of the Chief Accountant (OCA),
which enforces ERISA reporting requirements. 


--------------------
\8 The estimates were for administrative costs and costs of staff
time used to decide on and prepare documentation related to
processing referrals. 


   OPPORTUNITIES FOR INCREASED USE
   OF PENALTY AUTHORITY
------------------------------------------------------------ Letter :4

PWBA has not fully used the penalty authority provided by ERISA. 
PWBA may be missing opportunities to identify and penalize plans that
violate reporting requirements because PWBA does not routinely follow
up on all IRS referrals.  Lack of legal resources has hindered PWBA
from assessing more penalties for prohibited transactions by welfare
plans and pension plans that do not qualify for tax exemption.  In
addition, restrictive legal requirements have limited the assessment
of penalties for fiduciary violations. 


      OPPORTUNITIES FOR INCREASED
      USE OF REPORTING PENALTIES
---------------------------------------------------------- Letter :4.1

OCA does not routinely follow up on IRS service center referrals of
plans that file a Form 5500 series report one year but not the
next.\9 Officials in the IRS Memphis Service Center told us that in
September 1992 the center sent OCA the names of approximately 4,000
to 5,000 plans that had not filed a Form 5500 series report in
program year 1992 but had filed in a previous year.  Service center
officials said that while some plans may not have filed because they
terminated, merged with another plan, or had some other valid reason,
other plans likely had no valid reason for not filing.  An OCA
headquarters official told us that no action had been taken on these
or similar referrals from other IRS service centers because OCA does
not have enough staff. 

OCA has had several projects to identify plans with reporting
violations.  To encourage compliance by late filers and nonfilers,
PWBA offered a "grace period" from March through December 1992 during
which plan administrators who had failed to file annual reports could
file those reports and pay a reduced penalty.  Before 1992, PWBA made
a special effort to identify plans with over 100 participants that
failed to include an accountant's report with their annual reports. 
In late 1993, PWBA began a program to identify nonfilers from
information in master trust filings with Labor. 


--------------------
\9 Form 5500 series reports are sent to one of four IRS service
centers. 


      ASSESSMENT OF PROHIBITED
      TRANSACTION PENALTIES
      HINDERED BY LACK OF LEGAL
      RESOURCES
---------------------------------------------------------- Letter :4.2

From 1990 through 1993, PWBA assessed penalties on 11 of 48 welfare
or unqualified pension plans found to have prohibited transaction
violations.  Officials in PWBA headquarters and the Labor Solicitor's
Office said that the other violators were not assessed penalties
because plans often challenge the penalty in court and the
Solicitor's Office did not have enough staff to pursue all these
cases.  In addition, Labor pointed out in its comments on a draft of
this report that such cases often involve small dollar amounts, but
consume substantial litigation resources. 

Officials in five area offices told us that PWBA should be in a
better position to pursue such cases when the Labor Solicitor's
Office is decentralized.  Under a memorandum of understanding signed
by PWBA and the Solicitor's Office in fall 1993, civil penalty
litigation under ERISA was decentralized to four regions that had not
previously handled litigation.  Among other changes, the agreement
called for two additional staff in the Solicitor's Offices for each
of the four regions. 


      RESTRICTIVE LEGAL
      REQUIREMENTS LIMIT
      ASSESSMENT OF FIDUCIARY
      PENALTIES
---------------------------------------------------------- Letter :4.3

Restrictive legal requirements have limited PWBA's ability to assess
penalties against fiduciaries or other persons who knowingly
participate in a fiduciary breach.  Penalties may be assessed only
against fiduciaries or knowing participants in a breach who, by court
order or settlement agreement, restore plan assets.  If (1) there is
no settlement agreement or court order or (2) plan assets are
returned by someone other than a fiduciary or knowing participant,
the penalty may not be assessed. 

Area office officials told us that without settlement agreements they
could not assess penalties against some violators.  In some cases,
area offices said, violators restored plan assets after receiving a
PWBA letter requesting voluntary compliance, but avoided penalties by
claiming that the asset restoration was independent of PWBA's
enforcement efforts or by simply not replying to the letter.  Area
offices noted other cases where asset restoration and associated
settlement agreements were delayed when violators refused to take
corrective action unless PWBA forgave the penalty.  Area office
officials also said they were unable to assess penalties in cases in
which a party other than the fiduciary or knowing participant
restored the plan assets. 

Many area offices told us that the fiduciary penalty is complicated
and confusing despite guidance from headquarters.  Temporary guidance
issued in April 1991 states that PWBA must assess a fiduciary penalty
if the fiduciary or knowing participant restores assets in response
to a voluntary compliance letter.  The guidance also states that if
the fiduciary or knowing participant restores assets after receiving
such a letter but claims the action was voluntary, PWBA may not
assess the penalty because there is no settlement agreement.  The
guidance further states that if the fiduciary or knowing participant
restores assets after receiving a voluntary compliance letter but
remains silent on the reason for the action, a penalty should be
assessed unless special circumstances indicate that it should not be
assessed.  Area office officials said that they need additional
guidance regarding which fiduciary actions in response to voluntary
compliance letters constitute settlement agreements and what parties
are subject to the penalty. 

The importance of the fiduciary penalty and associated problems is
growing.  For example, the number of cases approved for penalty
assessment letters increased from 10 in 1991 to 105 in 1993. 
Moreover, area office officials believe the numbers will increase as
more cases are identified with fiduciary violations occurring after
the effective date of the penalty--December 19, 1989. 


   CONCLUSIONS
------------------------------------------------------------ Letter :5

Labor's ERISA enforcement program has improved since 1986, but
further strengthening the program would better protect millions of
private pension plan participants and trillions of dollars in assets
held by those plans.  To ensure that it is maximizing the use of
scarce investigative resources, PWBA needs to determine whether
financial institutions and service providers warrant continued
enforcement emphasis and, more specifically, whether 40 percent of
its investigative staffing resources should be devoted to such
efforts.  Moreover, PWBA needs to determine whether the amount of
resources allocated to significant issue investigations should be
tailored for each area office. 

PWBA also needs to determine the effectiveness of its computer
targeting programs to identify plans with ERISA violations so area
offices can use the most successful programs as soon as possible. 
However, PWBA's targeting testing procedures have been inadequate. 
Furthermore, at the time of our review, PWBA had no specific plans to
test its current programs and had taken few steps to correct
weaknesses in the procedures used to test the predecessor programs. 

The ERISA enforcement program can also be strengthened by making
better use of penalties authorized by ERISA.  By not routinely
following up on referrals from IRS, PWBA is likely to miss
opportunities to identify and penalize plans that have violated ERISA
reporting requirements.  Lack of legal resources has hindered PWBA's
use of penalties against welfare and unqualified pension plans
involved in prohibited transactions, but decentralizing legal
services may allow area offices to more aggressively pursue such
penalties.  Given the restrictive legal requirements that have
limited the use of the fiduciary penalty, the expected growth in
cases involving this penalty, and the decentralization of legal
services, PWBA needs to determine whether additional administrative
guidance, changes to the law, or both are needed to remedy confusion
associated with the penalty and enhance PWBA's penalty enforcement. 


   RECOMMENDATIONS TO THE
   SECRETARY OF LABOR
------------------------------------------------------------ Letter :6

We recommend that you direct the Assistant Secretary for Pension and
Welfare Benefits Administration to take the following actions to
strengthen the ERISA enforcement program: 

  Evaluate the significant issue strategy to determine (1) whether
     financial institutions and service providers continue to be the
     issue areas with the greatest potential for achieving maximum
     ERISA enforcement results and (2) whether 40 percent is the
     resource allocation formula that will provide the greatest
     enforcement results or whether the formula should be tailored
     for each area office. 

  Begin testing the revised computer targeting programs as soon as
     possible.  If PWBA opts to test each individual program using
     the same criteria described in our September 30, 1993, letter to
     the Assistant Secretary, PWBA should (1) randomly select plans
     for testing so results can be projected and programs properly
     validated and (2) use a formula to set a sample size that will
     require less calendar and staff time to test each program.  PWBA
     should also test the feasibility of using multivariate analysis
     to target plans for investigation. 

  Increase the use of penalties authorized by ERISA by establishing
     procedures to routinely review referrals of potential reporting
     violators from IRS service centers and using decentralized legal
     staff to help assess prohibited transaction penalties when
     warranted.  PWBA should also determine whether additional
     administrative guidance, changes to the law, or both are needed
     to remedy confusion associated with the penalty and enhance
     PWBA's penalty enforcement. 


   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :7

In providing comments on our draft report, the Assistant Secretary
for Pension and Welfare Benefits Administration agreed that Labor's
ERISA enforcement program should maximize the use of resources, but
generally disagreed with many of our recommendations.  Labor said
that its current strategy and computer targeting leverage enforcement
resources, which are extremely limited in view of the large universe
of participants and plans covered by ERISA.  Labor also said that
recent and anticipated changes should improve the program.  We
commend Labor for striving to improve its ERISA enforcement program,
but continue to believe that implementing our recommendations will
further strengthen the program. 


      ENFORCEMENT STRATEGY
---------------------------------------------------------- Letter :7.1

In commenting on our recommendation to evaluate the significant issue
strategy, Labor said that it has reviewed and refined its enforcement
strategy over the last 4 years and is in the process of making
substantial changes that should further improve the program.  It is
also planning to delegate more authority to area offices to establish
criteria for opening investigations.  In addition, Labor is looking
to reduce the time mandated for national enforcement initiatives and
allow more latitude for area office initiatives and local conditions. 

Our report discusses some recent strategy refinements, such as
decentralization of legal services and reduction of the significant
issue resource allocation formula.  Labor said that these changes
were brought about by the collective judgment of agency officials and
years of experience with its enforcement strategy.  We also revised
our report to mention Labor's belief that providing specific
headquarters direction on the amount of resources to be allocated to
significant issues will probably no longer be necessary.  As part of
our continuing oversight of the enforcement program, we plan to
monitor Labor's progress in implementing these strategy changes and
any others resulting from needed changes identified during our
recommended evaluation of the significant issue strategy. 


      COMPUTER TARGETING
---------------------------------------------------------- Letter :7.2

Labor believes, and we agree, that computer targeting testing
protocol should use enforcement staff and resources judiciously. 
However, Labor generally disagrees with our recommendations to (1)
randomly select plans for testing so results can be projected and (2)
use a selection formula that lowers the number of plans needed to
test programs.  Labor also raised concerns about the resources that
would be required to carry out our multivariate analysis
recommendation. 

We continue to believe that PWBA should randomly select plans for
testing if it wants to validate the programs as currently defined. 
If, as Labor believes, the purpose of testing is to identify the
levels of success of investigations of the highest ranking plans
identified by the programs one year and project those success levels
to the highest ranking plans in future years, random sampling would
not be necessary.  On the other hand, as noted in our September 30,
1993, letter to the Assistant Secretary for Pension and Welfare
Benefits Administration, this approach would not validate each of
PWBA's 81 computer targeting programs because the success rate for
the highest ranking plans may not be the same as the rate for other
plans identified by the programs. 

We also believe that implementing our recommendation to use a
sampling formula that requires fewer plans to test programs would
help Labor achieve its goal of making judicious use of staff and
resources.  Resources freed by investigating fewer plans could be
used to test more of the 81 individual programs each year so area
offices could use the most successful programs as soon as possible. 
Alternatively, these resources could be used to carry out other
enforcement activities.  Of course a larger number of plans would be
needed to test programs if, as Labor suggests, the purpose of testing
is to determine the relative success of the programs or the level of
confidence desired by PWBA is higher than originally sought. 

Further, we believe that implementing our recommendation that Labor
test whether multivariate analysis is a more effective and efficient
way to target investigations is consistent with Labor's desire for
judicious use of resources.  The test, at a minimum, should determine
whether the additional resources needed to assess all applicable
characteristics for each plan reviewed would be offset by the
decreases in resources needed to review a smaller total number of
plans.  Test results should address Labor's concern about the
availability of limited resources to fulfill its enforcement
responsibilities. 


      USE OF PENALTIES
---------------------------------------------------------- Letter :7.3

In commenting on our recommendation to make greater use of civil
penalties as an enforcement tool, Labor said that it seeks to deter
would-be violators through aggressive litigation and substantial
penalties.  For example, Labor said that it had prioritized the use
of limited resources in a highly effective way to carry out its
reporting compliance responsibility.  It said that a program
initiated in 1994 to identify nonfilers, among other things, will
include following up on the welfare plan listing received from IRS. 
Labor also said that its 1992 "grace period" program encouraged late
filers and nonfilers to voluntarily submit previously unfiled annual
reports.  We are reviewing this program as part of another assignment
and cannot comment on its merits at this time.  However, we revised
our report to recognize these efforts. 

Labor also said that decentralizing legal services to area offices
may result in more opportunities to assess penalties for prohibited
transactions involving welfare plans or nonqualified pension plans. 
Labor pointed out that the usefulness of such penalties may be
limited because cases often involve small recovery amounts and
require substantial resources to litigate.  We revised our report to
explain Labor's position on decentralized legal services.  We also
recognize that when deciding whether to assess this optional penalty,
Labor should consider the expected costs and revenues, as well as
other potential benefits such as the deterrent effect. 

Labor acknowledged difficulties with applying the fiduciary penalty,
noting that the authorizing law provides fixed, narrow restrictions
rather than broad flexibility.  We revised our report to emphasize
the difficulty PWBA area offices have had assessing the penalty and
highlight the growing importance of the fiduciary penalty.  Based on
our work and Labor's comments, we believe that headquarters needs to
determine whether additional administrative guidance, changes to the
law, or both are required to remedy confusion and enhance PWBA's
penalty enforcement. 


      OTHER MATTERS
---------------------------------------------------------- Letter :7.4

In addition to commenting on the specific sections of our report,
Labor raised some general concerns about our draft report.  Labor
said that it was concerned about the reliability of many conclusions
drawn from our area office questionnaire because some questions were
subject to interpretation and responses would reflect a range of
variables depending on the circumstances of the area office. 

In developing the area office questionnaire, we recognized that some
questions might be subject to interpretation and took steps to
minimize the subjectivity.  Before sending the questionnaire to
PWBA's 10 area offices, we discussed its contents and language with
officials in PWBA headquarters and 2 area offices and incorporated
their suggestions and comments as appropriate.  After the
questionnaire was completed, we visited five area offices to discuss
their responses and made changes, as necessary, to ensure we reported
them properly. 

In addition, Labor said that it was concerned about the impact of
differences in perspectives between area office directors and the
agency's policy leadership regarding the results of certain
enforcement efforts.  Citing two national initiatives as examples,
Labor pointed out that enforcement efforts may be viewed as
successful in qualitative terms, such as changing industry practices,
but not in quantitative terms, such as the number of fiduciary
violations.  Such differences, Labor said, may have affected area
office answers and had a major impact on the validity of our
conclusions.  We revised our report to clarify that while PWBA
considers "qualitative" effects when assessing its enforcement
efforts, we analyzed only quantitative indicators. 

Finally, Labor said that it would have been useful for us to discuss
our questionnaire results with top-level officials before preparing
our draft report to determine whether there were any national policy
issues that should have been taken into consideration.  Before
completing this report, we provided PWBA's Deputy Assistant Secretary
for Program Operations with a detailed briefing on the results of our
work.  However, no national policy issues were raised by Labor
officials at this meeting. 


---------------------------------------------------------- Letter :7.5

As you know, the head of a federal agency is required by 31 U.S.C. 
720 to submit a written statement of actions taken on our
recommendations to the Senate Committee on Governmental Affairs and
the House Committee on Government Operations no later than 60 days
after the date of the report.  A written statement must also be
submitted to the House and Senate Committees on Appropriations with
the agency's first request for appropriations made more than 60 days
after the date of the report. 

We are providing copies of this report to the Chairman and Ranking
Minority Members of the Senate Committee on Labor and Human Resources
and House Committee on Education and Labor, various other Senate and
House committees and subcommittees, interested Members of the
Congress, and other interested parties.  We also will make copies
available to others upon request. 

If you or your staff have any questions about this report, please
call me on (202) 512-7215.  Major contributors to this review are
listed in appendix IV. 

Sincerely yours,

Joseph F.  Delfico
Director, Income Security Issues


DETAILED SCOPE AND METHODOLOGY
=========================================================== Appendix I

We used information from several sources to understand Labor's
enforcement program.  We reviewed our previous reports; applicable
sections of the Employee Retirement Income Security Act of 1974, as
amended, and the Internal Revenue Code; selected federal regulations;
and the Pension and Welfare Benefits Administration's Enforcement
Strategy Implementation Plan, program planning guidance for program
years 1986 to 1994, Sources of Cases Studies for program years 1987
through 1990, and the Field Focus Group report sent to the Assistant
Secretary for Pension and Welfare Benefits Administration on October
5, 1993.  We interviewed officials in Labor's PWBA headquarters and
seven area offices, Office of the Solicitor, and Office of the
Inspector General, as well as representatives of selected outside
interest groups.  We visited IRS's Memphis Service Center to
determine firsthand how Form 5500 series reports are processed.  And
we sent a questionnaire to PWBA's 10 area offices (see app.  II). 

To assess PWBA's current enforcement strategy, we analyzed selected
statistical data from PWBA's Case Management System reports for
program years 1987 through 1993 and information from the
questionnaires received from PWBA's 10 area offices.  We discussed
the enforcement strategy with officials at PWBA headquarters and in
seven area offices.  To determine whether PWBA's enforcement activity
changed, we compared the number of cases opened and closed during the
last complete program year before the strategy was implemented, 1986,
with the number opened and closed during the latest complete program
year, 1993.  To determine the effect of PWBA's enforcement program,
we compared selected quantitative indicators for the first complete
program year after the strategy was implemented, 1988, with
comparable indicators for program year 1993.  To determine the effect
of the strategy, we compared selected quantitative indicators for
significant issue investigations with comparable indicators for
general case investigations for program year 1993.  The indicators
used for these analyses included dollar recoveries, number of plans
impacted, number of participants impacted, number of cases with
monetary recoveries, number of cases with other fiduciary results,
number of cases with nonfiduciary results, number of cases with
criminal indictments, number of individuals indicted, number of cases
referred for litigation, and number of cases with litigation filed. 
We did not attempt to analyze qualitative effects, such as changes in
industry practices or behaviors. 

To assess PWBA's computer targeting efforts, we reviewed Mathematica
Policy Research, Inc.'s planning document and reports on computer
targeting test results dated May 22, 1992, and February 28, 1994.  We
also tabulated the area office questionnaires to determine which
methods of targeting area office officials viewed as the most
effective.  In addition, we discussed the computer targeting
methodology with PWBA headquarters and area office officials. 

To assess PWBA's case referral procedures, we discussed the
procedures with officials in PWBA headquarters and seven area
offices, as well as with officials in IRS headquarters and six IRS
key district offices.  We reviewed the memorandum of understanding
between Labor and IRS.  We also analyzed tracking documents and
selected headquarters reports to ascertain whether Labor or IRS acted
on referrals and, if not, why. 

To analyze Labor's use of ERISA penalty authority, we analyzed
information in PWBA's Case Management System and tabulated the area
office questionnaire results.  We reviewed selected case files.  With
officials in the Labor Solicitor's Office and in PWBA headquarters
and seven area offices, we also discussed barriers to the use of
penalties. 

Our work was done primarily between April 1992 and December 1993,
with selected information updated as of mid-June 1994.  The work was
performed in accordance with generally accepted government auditing
standards.  However, we did not independently verify all information
provided by the area offices. 


RESULTS OF MAIL QUESTIONNAIRE TO
PWBA AREA OFFICES
========================================================== Appendix II



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)




(See figure in printed edition.)Appendix III
SEPTEMBER 30, 1993, GAO LETTER TO
ASSISTANT SECRETARY FOR PENSION
AND WELFARE BENEFITS
ADMINISTRATION
========================================================== Appendix II



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix IV

Jane L.  Ross, Associate Director, (202) 512-7215
Donald C.  Snyder, Assistant Director, (202) 512-7204
Byron S.  Galloway, Evaluator-in-Charge, (202) 512-7247
Albashar A.  Abdullah
Daniel R.  Garcia
Joel I.  Grossman
George A.  Scott
Roger J.  Thomas
