Public Pensions: State and Local Government Contributions to Underfunded
Plans (Letter Report, 03/14/96, GAO/HEHS-96-56).

Pursuant to a congressional request, GAO reviewed the status of public
pension plan funding, focusing on the basic pension plans of state and
local governments.

GAO found that: (1) states and localities with underfunded pension plans
run the risk of reducing future pension benefits to taxpayers or raising
revenues; (2) unfunded liabilities for all state and local pension plans
totalled $200 billion in 1992; (3) contributions to pension funds in
1992 fell short of the actuarially required amounts by 60 percent; (4)
75 percent of state and local pension plans involved in a Public Pension
Coordinating Council (PPCC) survey were underfunded; (5) more than half
of the pension plan sponsors surveyed continued to make payments to pay
off their unfunded liabilities; (6) between 1990 and 1992, 20 percent of
the plans were both underfunded and not receiving required sponsor
contributions; and (7) of 117 plans with complete data in 1990 and 1992,
90 were underfunded.

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

 REPORTNUM:  HEHS-96-56
     TITLE:  Public Pensions: State and Local Government Contributions 
             to Underfunded Plans
      DATE:  03/14/96
   SUBJECT:  Actuarial tables
             Pension plan cost control
             Financial management
             Civil service pensions
             Retirement pensions
             Accounting procedures
             State-administered programs
             Employee retirement plans
             Underpayments
IDENTIFIER:  New York
             New Jersey
             California
             
******************************************************************
** 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 Congressional Requesters

March 1996

PUBLIC PENSIONS - STATE AND LOCAL
GOVERNMENT CONTRIBUTIONS TO
UNDERFUNDED PLANS

GAO/HEHS-96-56

State and Local Pension Funding

(105686)


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

  AAL - actuarial accrued liability
  ERISA - Employee Retirement Income Security Act of 1974
  GASB - Governmental Accounting Standards Board
  PBGC - Pension Benefit Guaranty Corporation
  PBO - projected benefit obligation
  PPCC - Public Pension Coordinating Council

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


B-262220

March 14, 1996

The Honorable Nancy L.  Johnson
Chairman, Subcommittee on Oversight

The Honorable Sam M.  Gibbons
Ranking Minority Member
Committee on Ways and Means
House of Representatives

Over 13 million state and local government employees expect to
receive the pension benefits they have earned.  If their employers
are not currently on track to fully funding these pensions, some
employees may ultimately receive less than expected unless additional
revenues are provided. 

Oversight or other protections for beneficiaries of state and local
pension plans derive principally from the state and local laws that
created such plans, their administration by state and local agencies,
and ultimately their enforcement by the courts.  Although federal
pension laws impose funding requirements on private pension plans,
they impose no such requirements on state and local plans. 

This report responds in part to your request that we review the
status of public pension plan funding.  It focuses on the basic
pension plans of state and local governments.  Another report
examines state and local supplemental retirement programs, such as
section 401(k) or 457 plans, and a third report examines federal
pension plans.\1

Regarding state and local pension plans, you asked us to provide
information on the implications of underfunding.  In addition, you
asked us to provide recent data about the (1) funding status of such
plans; (2) status of contributions to such plans, particularly
underfunded plans; and (3) changes in the funding and contribution
status over time. 

To address your concerns, we researched and examined studies and
literature on state and local government funding of pension plans. 
In particular, we analyzed data assembled by the Public Pension
Coordinating Council (PPCC) from a survey representing 451 state and
local pension plans that covered 76 percent of the active
participants of such plans.  Data for 1992 were the most recent
available; we compared these data with 1990 data from an earlier PPCC
survey.  We did not make any adjustments to the data reported by the
plan sponsors.  We also interviewed officials of selected plans.  We
performed our work from June 1995 through November 1995 in accordance
with generally accepted government auditing standards.  (See app.  I
for more information on our scope and methodology.)


--------------------
\1 See our forthcoming report on section 457 plans (GAO/HEHS-96-38);
Public Pensions:  Summary of Federal Pension Plan Data
(GAO/AIMD-96-6, Feb.  16, 1996). 


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

Although state and local governments rarely become insolvent or cease
to operate, those with underfunded pension plans may face difficult
budget choices in the future if they do not work toward full funding. 
Their future taxpayers will face a liability for benefits earned by
current and former government workers, leaving these governments to
choose between reducing future pension benefits or raising revenues. 

Funding of state and local pension plans has improved substantially
since the 1970s.  After adjusting for inflation, the dollar amount of
the unfunded liability has decreased by roughly half.  Still, in
1992, 75 percent of state and local government pension plans in the
PPCC survey were underfunded; 38 percent were less than 80-percent
funded. 

Sponsors of slightly more than half of plans in the PPCC survey made
contributions that kept on schedule for paying off any unfunded
liability.  One-third of the pension plans, however, were both
underfunded in 1992 and not receiving the actuarially required
sponsor contributions.  Of all plans with complete data, one-fifth
were both underfunded and not receiving full contributions in both
1990 and 1992. 


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

In late 1992, about 13 million workers and retirees participated in
state and local government pension plans.  Eighty-seven percent of
state and local full-time employees participated in defined benefit
plans, while 9 percent participated in defined contribution plans. 
Some employees participated in both; 93 percent participated in one
or the other. 

In a defined benefit plan, benefits are established by a formula that
is generally based on such factors as years of employment, age at
retirement, and salary level.  Employers and employees in most state
and local plans contribute to a fund from which these defined
benefits will be paid.  Actuaries calculate the size of the fund that
will be needed to pay these benefits on the basis of projections of
fund investment earnings, mortality, and other factors.  If the
fund's assets are less than the projected liabilities, the plan is
generally considered to be underfunded.  Actuaries also calculate the
contribution amount needed to cover the liability that accrues each
year and to pay an installment on any unfunded liability.  Thus, if a
plan sponsor is making these actuarially required contributions, the
plan can be underfunded yet still on track toward full funding. 

In contrast, a defined contribution plan, such as a 401(k) or 457
supplemental retirement account, sets the amount contributed to
individual worker accounts.  The balance in such an account at
retirement, reflecting total contributions and investment earnings,
determines the worker's retirement benefit.  Thus, by definition,
such a plan cannot be underfunded in an actuarial sense. 


      OVERSIGHT OF STATE AND LOCAL
      PLANS
---------------------------------------------------------- Letter :2.1

For private-sector pension plans, the Employee Retirement Income
Security Act of 1974 (ERISA) ensures that most promised employee
benefits will be paid.  If a company goes out of business and leaves
its defined benefit pension plan without adequate funds, the business
no longer has earnings with which to make further contributions to
cover its pension obligations.  Among other things, ERISA insures
pension benefits against insufficient funding; the Pension Benefit
Guaranty Corporation (PBGC) assumes the liability and assets of
terminated private pension plans and pays the retirement benefits,
subject to certain limits.  ERISA also requires that plan investments
be diversified and funded on a sound actuarial basis and that plan
fiduciaries adhere to certain standards of conduct.  Further, ERISA
establishes a framework for enforcing its provisions involving three
federal agencies:  the Department of Labor, the Internal Revenue
Service, and the PBGC. 

ERISA does not apply, however, to state and local pension plans, nor
does federal law impose funding requirements on them.\2 Instead,
state and local pension plans are created and governed by laws of
their respective governments, which specify any funding or other
requirements.  These laws, their administration by state and local
government agencies, and ultimately their enforcement by the courts
provide any protections the beneficiaries may have.  For example, in
all 50 states, statutes include provisions for fiduciary standards;
in about half of the states, these provisions are similar to ERISA's. 
Also, about half restrict the types of investments that can be made. 
In addition, annual contributions to 56 percent of state and local
pension plans are required to be actuarially based; for 40 percent of
these plans, statutes set a specific contribution level, which in
most cases is periodically adjusted to achieve actuarial balance,
according to a state pension official.  In addition, the plans are
subject to review by state and local governmental audit agencies and
legislative oversight committees. 

Moreover, the Governmental Accounting Standards Board (GASB) sets
accounting and reporting standards for state and local governmental
entities, including pension plans.  Most state and local governments
adhere to these standards.  GASB will soon require the reporting of a
6-year funding and contribution history for pension plans and, for
plan sponsors, the reporting of a measure of the difference between
actuarially required and actual contributions.  According to a PPCC
analyst, such reporting should provide further impetus to improve
funding and contributions. 

At times, perceptions arise that a state or local government is
redirecting funds from a pension plan to meet other budgetary needs. 
For example, since 1992 California has attempted to delay its annual
contributions of roughly $500 million by more than a year, costing
its Public Employees' Retirement System as much as $50 million per
year in interest.  In other states, such as New York and New Jersey,
the legislatures have attempted to change certain actuarial
assumptions to lower contributions.\3 Also, several state
legislatures have encouraged pension managers to invest some portion
of plan assets in ways that will promote economic development as long
as these investments are sound.  Some critics raise concerns that
such targeted investments often are not sound, citing certain bad
investments that lost millions.\4 For example, in 1990, Connecticut's
pension plan invested about $25 million in Colt Industries, which in
1992 declared bankruptcy; the pension plan lost $21 million. 

State and local government plans nevertheless operate in public view,
and some plan fiduciaries and others have filed suits against plan
sponsors.  For example, according to pension officials and trade
publications, in the California case, a superior court judge ordered
the state to make the delayed contribution with accrued interest; the
case is pending on appeal.  Similarly, New York's highest court ruled
that the New York law changing the actuarial methodology for the
public employees' retirement system violated the New York State
Constitution, and the state has agreed to a payment schedule that
will make full restitution of missed contributions by 1999.  As of
this writing, the other cases involving New York and New Jersey are
still pending in their respective courts.  Regarding targeted pension
investments, while some such investments do not earn competitive
returns, others do. 


--------------------
\2 However, participants of government pension plans do enjoy tax
deferral on contributions and investment earnings.  To qualify for
this tax deferral, plans must adhere to federal rules, including ones
on coverage, participation, nondiscrimination, integration with
Social Security, benefit distribution, and operating for the
exclusive benefit of plan members. 

\3 The pension plan sponsor's actuarially determined annual
contribution is particularly sensitive to changes in the underlying
assumptions.  For example, a small change in the assumed rate of
return on plan investments can produce a large change in calculated
pension liabilities and, in turn, in the annual contribution needed
from the employer; a 1-percentage point increase in the assumed
investment return rate, with other assumptions remaining the same,
could result in a 20- to 25-percent reduction in the required annual
contribution.

It is noteworthy, however, that changing actuarial assumptions is not
necessarily inappropriate.  After assessing pension fund performance
and other factors, actuaries may change assumptions each time they
make a new valuation and recalculate contributions required from plan
sponsors. 

\4 Public Pension Plans:  Evaluation of Economically Targeted
Investment Programs (GAO/PEMD-95-13, Mar.  17, 1995). 


   UNDERFUNDING MAY PRESENT FUTURE
   PROBLEMS
------------------------------------------------------------ Letter :3

Although incidents of insolvency or termination of state and local
pension plans are rare, underfunding of such plans may present
governments with difficult budget decisions in the future.  If the
actuarial value of a pension plan's liabilities exceeds its assets, a
plan may still have enough funds to pay benefits for many years.  If
such underfunding persists, however, eventually--perhaps years in the
future--the plan may lack enough funds to pay benefits.  At that
time, the sponsoring government will need to have made additional
contributions to the pension fund.  Or, a government may change the
law to reduce benefits or postpone benefit increases that offset
inflation, depending on the law that created the plan and the state
constitution or municipal charter that governs lawmaking.  Thus, the
ability of state and local governments with underfunded plans to meet
their pension obligations at some future date will depend on
balancing competing budgetary demands and possibly on the willingness
of their taxpayers to meet the cost. 

Moreover, if pension benefits are not fully funded, the fiscal burden
of providing for them can grow quickly as a share of the budget under
various circumstances.  Benefit costs can increase rapidly if the
number of retirees surges.  Also, government revenues can grow slowly
if the tax base decreases or tax rates are cut, even though promised
benefits have already been determined for years to come.  In fact,
the ratio of active workers to retirees is declining in the state and
local sector,\5 which means that the cost of paying benefits for
previous years' employees is growing relative to the cost of paying
current employees.  If benefits are not fully funded, the relative
fiscal burden of providing for them will grow as well.  Also, the
prospect of such budgetary pressures can significantly affect the
sponsoring government's bond ratings. 

In addition to concerns that full benefits might not be paid as
promised or that the fiscal burden of doing so might be excessive,
underfunding of state and local plans implies that the cost of
government has been partially shifted from one generation of
taxpayers to another.  This year's cost of government includes the
cost of pension benefits that employees earn with this year's work. 

Underfunding can arise when pension contributions do not fully cover
the cost of pension benefits that workers earn in a given year. 
Underfunding can also arise for other reasons, however, such as
pension plan investments' not earning as high a return as projected. 
Actuarially required contributions include an installment on the
amount needed to amortize the underfunded amount.  Thus,
undercontributing arises when the sponsoring government is not paying
enough either to cover the pension liability incurred this year or to
amortize this year's share of the unfunded liability or both. 

A number of federal pension plans also have unfunded liabilities;\6
however, arrangements have been made to fund the liabilities in the
future.  More importantly, the funding status of federal pension
plans does not have the same implications as that of state and local
plans.  Unlike state and local pension assets, the vast majority of
federal pension assets must by law be invested in nonmarketable U.S. 
Treasury securities.  In effect, federal pension assets largely
represent government promises to pay benefits, rather than
investments that can be converted to cash.  When the Treasury pays
benefits now or in the future, it must obtain the money either from
tax revenues or borrowing, regardless of the plans' technical funding
status. 


--------------------
\5 1990 Comparative Study of Major Public Employee Retirement
Systems, State of Wisconsin Retirement Research Committee, Staff
Report No.  79 (1990). 

\6 Public Pensions (GAO/AIMD-96-6, Feb.  16, 1996). 


   STATE AND LOCAL PENSION
   UNDERFUNDING HAS DECREASED, BUT
   UNDERFUNDED PLANS REMAIN
------------------------------------------------------------ Letter :4

Although underfunding of state and local pensions has decreased
considerably since the mid-1970s, underfunded plans remain.  The
unfunded liabilities of all state and local government pension funds
then totaled $150 to $175 billion,\7 according to the U.S.  House of
Representatives Pension Task Force; adjusting for inflation, this
equals about $400 billion in 1992 dollars.  In 1992, unfunded
liabilities totaled roughly $200 billion, according to our estimate
from the PPCC sample.  Thus, the unfunded liability has decreased by
about half in constant dollars.  Also, in the mid-1970s, the funding
ratio was roughly 50 percent.  The 1992 funding ratio for plans in
the PPCC sample was 82 percent (see table 1).\8 (The funding ratio is
the proportion of pension liability\9

covered by the value of plan assets.)



                                Table 1
                
                   Funding Status of State and Local
                          Pension Plans, 1992


                    Funding             Funding
                     ratio               ratio               Funding
           Number   (percent   Number   (percent   Number     ratio
          of plans     )      of plans     )      of plans  (percent)
--------  --------  --------  --------  --------  --------  ----------
Underfun     81        79       198        73       279         77
 ded
 plans
Fully        25       117        72       116        97        117
 funded
 and
 overfun
 ded
 plans
======================================================================
All         106        85       270        77       376         82
 plans
----------------------------------------------------------------------
\a Local plans include all plans not sponsored by a state. 

Source:  GAO Analysis of 1993 PPCC Survey. 

Funding ratios vary widely, however.  Of the plans that were
underfunded in 1992, 38 percent were less than 80-percent funded (see
fig.  1).  For underfunded plans alone in the PPCC sample, total
assets equaled 77 percent of liabilities. 

   Figure 1:  Distribution of
   State and Local Pension Plans
   by Funding Ratio, 1992

   (See figure in printed
   edition.)

Source:  GAO Analysis of PPCC Survey. 


--------------------
\7 The 1975 estimate is based on a different, more limited sample, so
comparing it with our 1992 estimate has limitations.  Still, the
change in funding status is great. 

\8 Data are from the 1993 Survey of State and Local Government
Employee Retirement Systems, PENDAT survey for 1993, prepared by
PPCC.  Plans with complete funding data for 1992 represented 95
percent of all employees and 94 percent of all assets for the plans
responding to the PPCC survey.  (See app.  I for more detail.)

\9 We used the actuarial accrued liability to measure plans' pension
liability.  It is noteworthy that funding ratios provide only general
indications of funding status because public pension plans can use
different actuarial methods and assumptions, such as the assumed rate
of return on plan investments.  (See app.  I.)


   STATE AND LOCAL GOVERNMENT
   CONTRIBUTIONS HAVE FALLEN SHORT
   OF ACTUARIALLY REQUIRED AMOUNTS
------------------------------------------------------------ Letter :5

In 1992, state and local government contributions to their pension
funds fell short of the actuarially required amounts; the
contribution ratio was 88 percent for all plans in the sample with
complete contribution data.\10 The contribution ratio is the
proportion of the actuarially required contribution covered by actual
contributions.  (See table 2.)



                                Table 2
                
                 Contribution Status of State and Local
                          Pension Plans, 1992


                    Contribu            Contribu
                      tion                tion
                     ratio               ratio              Contributi
           Number   (percent   Number   (percent   Number    on ratio
          of plans     )      of plans     )      of plans  (percent)
--------  --------  --------  --------  --------  --------  ----------
Received     50        69       100        73       150         69
 less
 than
 full
 contrib
 utions
Received     46       105       151       106       197        105
 full
 contrib
 utions
 or more
======================================================================
All          96        84       251        99       347         88
 plans
----------------------------------------------------------------------
\a Local plans include all plans not sponsored by a state.  Local
plans cover 26 percent of all employees in state or local plans and
11 percent of employees in plans that received less than full
contributions.  Thus, the contribution ratio for local plans does not
have a large effect on the ratio for all plans combined. 

Source:  GAO Analysis of 1993 PPCC Survey. 

Underfunding of state and local government plans will not likely
improve if contributions fall short of actuarially required amounts,
which are calculated to cover currently accruing liabilities and also
to help pay off any existing unfunded liability.  Inadequate
contributions over the long term could seriously erode the financial
status of some plans, especially those underfunded by large amounts. 

Contribution ratios also varied widely in 1992.  While 57 percent of
plans received full contributions, the remaining 43 percent had a
combined contribution ratio of just 69 percent, or nearly $4 billion
less than the actuarially required amount.  About 15 percent of plans
received less than 60 percent of required amounts.  For state plans
alone, over half received less than full contributions, with nearly
one-fourth below 60 percent of required amounts. 

Examining only the plans that were underfunded, 44 percent received
less than full contributions and 16 percent received less than 60
percent of the required amount.  About 55 percent of underfunded
state plans received less than full contributions, compared with 40
percent of underfunded local plans.  (See table 3.)



                                Table 3
                
                   Contribution Ratios of Underfunded
                              Plans, 1992


                                                        Local\     All
Contribution ratio                               State       a   plans
----------------------------------------------  ------  ------  ------
Less than 40                                      13.0     4.0     6.6
40 to less than 60                                10.1     9.7     9.8
60 to less than 80                                10.1    10.9    10.7
80 to less than 99                                21.7    15.4    17.2
99 or more                                        44.9    60.0    55.7
======================================================================
Total                                            100.0   100.0   100.0
----------------------------------------------------------------------
Note:  Numbers in table may not add to the total due to rounding. 

\a Local plans include all plans not sponsored by a state. 

Source:  GAO Analysis of 1993 PPCC Survey. 


--------------------
\10 Plans with complete contribution data for 1992 represented 84
percent of all employees and 85 percent of all assets for the plans
responding to the PPCC survey. 


   RECENT CHANGES IN FUNDING AND
   CONTRIBUTION STATUS
------------------------------------------------------------ Letter :6

According to our analysis of the PPCC survey data, the funding status
of state and local pension plans improved between 1990 and 1992.  The
contribution status of these plans worsened slightly, however, and a
significant share of plans underfunded in both years also received
less than full contributions in both years.  Although the survey had
complete data for both years for only a subset of all survey
respondents, this analysis illustrates the plans' varied experience
and what can happen in the worst cases. 

For the plans with complete funding data for both years, the funding
ratio increased from 80 to 83 percent.  For underfunded plans alone,
the funding ratio increased from 72 to 78 percent.\11

For plans with complete contribution data for both years, the
contribution ratio decreased from 93 to 85 percent.  For plans
receiving less than full contributions, the contribution ratio
decreased from 62 to 60 percent. 

Examining the distribution of plans by funding and contribution
status better reveals the potential for funding problems since each
pension fund must meet its own obligations.  Of the 156 plans with
complete funding data for both years, 97 plans increased their
funding ratios, and the number of underfunded plans dropped from 122
to 118.  For the 143 plans with complete contribution data for both
years, the number of plans receiving less than full contributions
increased from 53 to 57. 

Still, it may be perfectly appropriate for an overfunded plan to
undercontribute; underfunded plans that do so are the primary
concern.  Of the 117 plans that had complete data for both years,\12
90 were underfunded in both years.  Sponsors undercontributed to 28
of these in one of the two years but not both and undercontributed to
25 in both years.  (See table 4.) Of the 25, 8 plans had funding
ratios that decreased between 1990 and 1992.  Another three had level
funding ratios.  Thus, nearly half of the 25 showed no improvement in
their funding status. 



                                Table 4
                
                     Recent Changes in Funding and
                   Contribution Status of Plans With
                    Complete Data, 1990 Versus 1992


                                                         Fully
                                                Underf  funded
                                                 unded  1 year
                                                  both      or     All
Receiving full contributions                     years    both   plans
----------------------------------------------  ------  ------  ------
Neither year                                        25       3      28
1990 or 1992                                        28      10      38
1990 and 1992                                       37      14      51
======================================================================
Total                                               90      27     117
----------------------------------------------------------------------
Source:  GAO Analysis of 1991 and 1993 PPCC Surveys. 

Even with undercontributing, funding ratios may improve for various
reasons, including strong returns on pension fund investments.  Also,
while most of the underfunded plans without full contributions
nevertheless improved their funding status, their sponsors may not
have been paying off the unfunded liability exactly on schedule. 
Conversely, making the full actuarially required contribution,
including partial payment of the unfunded liability, may not have
been sufficient to improve funding ratios; for example, investments
may have returned less than estimated.  Of the 37 plans that were
underfunded yet fully contributing in both years, 9 nevertheless had
decreasing funding ratios.  This illustrates the value of full
funding in buffering against poor investment returns or other
temporary strains on the pension fund. 


--------------------
\11 Of the 376 plans with complete funding data in 1992, 156 also had
complete data in 1990.  The funding ratio for the 376 plans was 82
percent in 1992 compared with 83 percent for the 156 plans.  Thus,
the plans with missing data tended to be similar in funding status.

The 376 plans covered 95 percent of the employees in plans that
responded to the survey, while the 156 plans covered 62 percent of
them.  Contribution data had a similar pattern, though 143 plans had
complete contribution data in both years, covering 54 percent of
employees.  (See app.  I for more detail on the data set.)

\12 Of the 504 plans responding in either year, 117 plans had both
complete funding and contribution data for both years.  These 117
plans covered 42 percent of the employees covered by the 504 plans. 


   CONCLUSIONS
------------------------------------------------------------ Letter :7

The funding status of state and local government pension funds has
improved substantially in the past 15 years.  More than half of
underfunded plan sponsors are contributing enough to reduce their
unfunded liability, while the other plans are not. 

Most significantly, one-third of state and local pension plans were
both underfunded in 1992 and receiving less than the actuarially
required sponsor contributions.  Sponsors of underfunded plans who
consistently undercontribute will leave their plans with little
buffer against possible deterioration in the plans' financial status. 
Such a deterioration could arise, for example, from an increase in
the number of retirees or poor investment performance.  As a result,
sponsors create the potential for difficult budget choices in the
future and may implicitly shift to future taxpayers part of the
burden for paying today's government workers. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :8

The Chair of the PPCC's Survey Committee and the administrator of the
PPCC database provided comments on a draft of this report.  The PPCC
represents associations of finance and retirement officials from
state and local government.  (See app.  I for more detail.) Neither
individual disputed the accuracy of the data we presented, but both
disagreed with some of our specific conclusions. 

The Survey Committee Chair commented that the report's tone is not
balanced and would likely lead readers to think that public pension
underfunding is a larger problem than it really is.  We do not agree
that this report is biased in tone or content.  We clearly
acknowledge that on the whole funding has improved substantially. 
However, we also attempt to focus attention on those underfunded
state and local plans that may face problems if undercontributing
persists. 

The administrator of the PPCC database suggested that only
contribution ratios of less than 90 percent be considered significant
undercontributing.  He feels that relatively small levels of
undercontributing often may reflect differences between actual
experience and actuarial projections and that sponsors may compensate
with overcontributions in other years.  We acknowledge that a small
level of undercontributing in one year may not significantly erode
funding levels for a given plan, but we do not believe we can
arbitrarily specify a numerical value at which undercontributing
becomes significant in isolation from other factors. 

Ultimately, the critical question is whether undercontributing for a
given plan persists from year to year and whether its funding level
improves or worsens.  Our 2-year analysis attempts to address this
question, but, unfortunately, the PPCC Survey does not yet have
complete data for enough plans for enough years to draw firm
conclusions.  Regarding this analysis, PPCC's database administrator
also commented that it is not appropriate to generalize from the
relatively small number of plans that had complete data for both
years.  We agree and, in fact, were careful not to generalize from
this 2-year analysis; we presented this analysis only to illustrate
and focus on the implications of persistent undercontributing. 

In general, both commenters stated their view that the funding status
of state and local pensions is improving.  One noted that GASB
reporting rules may provide further impetus to improve state and
local plan funding and contributions.  Both commenters also had some
technical comments, which we have incorporated where appropriate. 


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

As arranged with your office, unless you publicly announce its
contents earlier, we plan no further distribution of this report
until 30 days from the date of this letter.  At that time, copies
will be made available to others upon request. 

If you have questions concerning this letter, please call me on (202)
512-7215.  Other GAO contacts and staff acknowledgments are listed in
appendix II. 

Jane L.  Ross
Director, Income Security Issues


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

To analyze the current status of state and local pension plan funding
and contributions, we used data from the Public Pension Coordinating
Council (PPCC).  Members of the PPCC include the

  Government Finance Officers Association,

  National Association of State Retirement Administrators,

  National Conference on Public Employee Retirement Systems, and

  National Council on Teacher Retirement. 

The most recently available PPCC data were from their 1993 survey
representing the financial status primarily for fiscal year 1992.  We
compared these data with a similar survey PPCC conducted in 1991,
representing the financial status primarily for fiscal year 1990. 

Despite some limitations, the PPCC data are the best available, and
respondents to the 1993 survey represent 83 percent of the assets of
all state and local government plans and 76 percent of active plan
members.  Following are the data limitations:  (1) survey responses
represent the financial status for the fiscal year with the most
recent actuarial valuation and thus do not all represent the same
fiscal year's financial status; (2) the samples are not random and
therefore limit any generalization of results to nonrespondents; (3)
many responses lack complete funding or contribution information; (4)
the 1991 survey had fewer respondents than the 1993 survey; and (5)
the data represent the plans' own estimates using varied actuarial
cost methods and assumptions. 

For the 1991 survey, 73 percent of responses had data from fiscal
year 1990, 18 percent from 1989, and the remainder from other years. 
For the 1993 survey, 68 percent of responses had data from fiscal
year 1992, 19 percent from 1991, 8 percent from 1993, and the
remainder from other years. 

PPCC sent its survey to all members of two of its member associations
and to a representative sample of members of the other two
associations.  PPCC reported that one of its associations was
especially helpful in ensuring responses.  Due to the nonrandom
nature of this sample and the resulting potential for bias, no
analysis can offer any generalizations about nonrespondents.  Nor can
confidence intervals be calculated.  Nevertheless, the survey covered
a substantial majority of pension plan members and assets.  Thus the
analysis describes the funding status of a large and important
portion of all plans and members. 

As noted in footnotes throughout the report, many respondents did not
provide complete funding data or contribution data or both.  We can
calculate the number of employees represented by plans with complete
data, but we cannot generalize anything about the plans with
incomplete answers or assess any resulting bias in the results. 
Still, as the footnotes detail, the plans with complete data
generally represented a substantial share of the employees in
responding plans. 



                                    Table I.1
                     
                         Information on PPCC Survey Data


                                          Number          Number          Number
                                              of              of              of
                                          employ          employ          employ
                                  Number     ees  Number     ees  Number     ees
                                      of  repres      of  repres      of  repres
                                   plans   ented   plans   ented   plans   ented
--------------------------------  ------  ------  ------  ------  ------  ------
Responding to survey                 271  9,140,     451  9,945,     504  10,825
                                             285             551            ,468

With complete data for
--------------------------------------------------------------------------------
Funding                              185  7,205,     376  9,420,     156  6,702,
                                             524             708             879
Contributions                        188  6,553,     347  8,393,     143  5,818,
                                             254             893             174
Both                                 150  5,240,     318  7,976,     117  4,597,
                                             075             721             509
--------------------------------------------------------------------------------
The limitations of incomplete data were greatest for comparisons
between 1990 and 1992.  Therefore, we present this analysis primarily
for illustration. 

We did not adjust the PPCC data to standardize actuarial cost methods
and assumptions.  State and local governments may have many
legitimate reasons for choosing various cost methods and assumptions,
and we did not evaluate their choices.  For example, among various
investment restrictions, some plans are not allowed to invest in
stocks while others are; therefore, plans' assumed rate of return on
investments should differ. 


      CHANGE IN ACTUARIAL MEASURES
      USED
------------------------------------------------------- Appendix I:0.1

In our previous analysis of this database,\13 we used a measure
called projected benefit obligation (PBO) for the pension plans'
liabilities.  At the time, the Governmental Accounting Standards
Board (GASB) required that state and local plans report this measure. 
In 1994, GASB changed its policy to require a measure called
actuarial accrued liability (AAL), primarily responding to many
requests from plans that did not use the PBO measure.  Also, many
officials felt that the PBO underestimates plan liabilities. 

In accordance with GASB's change in policy, our current analysis also
used the AAL measure; the PPCC database includes both PBO and AAL
data, when reported.  Our analysis confirmed that PBO generally
yields higher funding ratios and therefore suggests a lower degree of
underfunding.  Since our previous analysis used the PBO measure and
our current analysis used the AAL measure, funding ratios and related
statistics should not be compared between the two reports.  The trend
analysis in this report, however, does make a valid comparison over
time, using AAL for both the 1990 and 1992 data. 


--------------------
\13 Underfunded State and Local Pension Plans (GAO/HRD-93-9R, Dec. 
3, 1992). 


GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
========================================================== Appendix II

CONTACTS

Donald C.  Snyder, Assistant Director, (202) 512-7204

ACKNOWLEDGMENTS

The following individuals made major contributions to this report: 
Dea M.  Crittenden, Evaluator-in-Charge; Kenneth C.  Stockbridge,
Evaluator; Lawrence Charron, Evaluator; and Sharon Fucinari, Computer
Specialist. 


*** End of document. ***