Proposed Program Evaluation Act of 1977, S. 2

Published by the Government Accountability Office on 1977-09-28.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                         DCCUNENT FESUME
03749 -   B2773984]

[Proposed Program Evaluation Act of 1977, S. 21. September 28,
1977. 11 pp.
Testimony before the Senate Committee on Rules and
Administration; by Harry S. Havens, Director, Program Analysis
Issue Area: Program Evaluation Systems (2600).
Contact: Program Analysis Div.
Budget Function: General Government: Executive Direction and
    Management (802); General Government: General Property and
    Records anagement (804).
Congressional Relevance: Senate Committee on Rules and
Authority: Program Evaluation Act cf '77; S. 2; 95th Cong.),

          The proposed "Program Evaluation Act of 1977" would
require that the Congress periodically reauthorize related
Federal programs. The bill would establish a schedule of
termination dates for nearly all programs grouped by Dudget
subfunctions. The key ingredient necessary to achieve effective
review of programs is legislative commitment to the oversight
process, ccupled with some form of discipline to trigger the
oversight activity, and the careful study of programs, policy
issues, and alternatives for improvement. Congress should
consider alternative ways of articulating this legislative
commitment and triggering the review process tc avoid the
problems associated with the threat of actual program
termination. The set of programs that are reviewed together
should depend on the issues involved and the objectives of the
review rather than on an arbitrary determination of budget
  'bfunction. It will be difficult, if not impossible, to conduct
the kind of comprehensive evaluations anticipated within the
time-frames established in the bill. Periodic review and
reconsideration of tax expenditures should also be included, as
should transition provisions for programs which are to be
terminated. GAO should be given the responsibility for compiling
and maintaining an inventory of Federal programs. (SC)
United States eneral Accounting Of ice
        Washington, D.C. 20546

                                   L O

                         £FU RLEAbL ON DLLIVniv
                         Expected at 10:OG a.m. Ll
                         Weanesday, September 2, 177

                Statement of
             Harry S. Havens
Director of the Program Analysis Division

 of t'le U.S. General Accounting Oftice

                Before the
 Committee on    ules and    dninistation
         United States Senate

the Program Evaluation Act of 177, S.2
hr. Chairman ana   embers of the Committee:
     We are pleased to be here today to outline our views
on   .2 "Program Evaluation Act of 177."          The bill. if
enacted, would represent a major reform to the congres-

sional authorization process, requiring that the Congress

periodically reauthorize related Federal programs.          The
bill would establish a schedule of termination dates for

nearly all programs grouped by budget subfunction.

     As you know, this office has long supported efforts
to strengthen ad improve the effectiveness ano account-
ability of Federal programs and we agree completely with

the objectives of S. 2.   We are concerned, however, with
certain   spects of the bill.         Although S.2 as reported by
the Senate Committee on Governmental Affairs includes many

improvements over earlier versions of the bill on which
we have testified, we believe that the legislation can

be further improved and we would like to take this oppor-
tunity to offer our suggestions for making the bill as

workable and effective as possible.

  Concerns with the
  Sunset Mechanism
        whether the threat of automatic termination
  'sunset" of programs by   u6det subfuncton is desirable,
 necessary and sufficient to enc*urage
                                       effective revie.
 and reconsideration of programs   y the Congress is a
 matter of considerable aebate which
                                      the Congress itself
 must determine. However, we are concerned
                                             with some
 of the implications of tnis aspect
                                    of the proposea

      By establishing statutory termination
                                             dates, S.2
 would force the Congress to take
                                  positive authorizing
action in order to continue a program
                                       after the program's
termination date. The principal argument
                                            for tnis re-
quirement is that without the "action-forcing"
of termination, program review requirements
                                              would be
meaningless. Opponents of the automatic
mechanism argue that sunset could
                                   not prevent the pro-
forma reauthorization of programs
                                   and that forced action --
before the Congress is ready -- on
                                    a reauthorization in
order to avoid termination will not
                                     necessarily be well
considered action.

      In our opinion, the key ingredient
                                         necessary to achieve
effective review of programs is legislative
                                             commitment to
the oversight process, coupled with
                                     some form of discipline
to trigger the oversight activity,
                                    and the careful study
of programs, policy issues, and alternatives
                                              for improvement.
he suggest that you consider alternative ways of articulating

this legislative commitment and triggering the review process
(perhaps using a mechanism sucn as the Program Evaluation

Resolution) which would avoid the problems associated with

the threat of actual program termination.

       he make this suggestion because we are zoncerned with
some of the implications of this concept of the sunset
mechanism.    For example, where termination of    program is
not a realistic alternative, sunset could      ead to emphasis
on the wrong questions about the program and could introduce

unnecessary concerns for business, consumers and others.

The threat of program termination could also discourage the
careful examination of programs, policy issues, ano alter-

natives for improvement, encouraging instead "rhetorical"
evaluations aimed at justifying the need for continuing a
program and stressing the dire consequences of termination.

Sunset could also contribute to unrealistic expectations
concerning budget reductions and the number of programs

to b    obolished.
Scheduling Limitations

       S. 2 contains a statutory schedule of termination dates
for programs grcuped by budget subfunction.     Niowever, any
schedule for the comprehensive reconsideration of Federal
programs involves a compromise between two important and
conflicting objectives:    1   enabling the Congress to legis-
late in broad policy areas at one time by grouping programs

 with similar goals for reconsideration and
                                            2) distributing
 the review, reconsideration, and reauthorization
 for each committee over the review cycle
                                          as evenly as possible.
      Since related programs are often under the
 of the same authorizing committee, it will
                                            be difficult
 to meet both objectives in a comprehensive
                                            schedule. Also,
 by grouping programs for reconsideration
                                          according to budget
 subfunctions, additional problems are likely
                                              to arise be-
 cause of the inherent limitations in any
                                          program or budget
 classification system. Programs often have
                                              multiple objec-
 tives which are related to more than one
                                          budget subfunction.
      The bill's schedule fox programs in the health
area illustrates these difficulties. Programs
                                                 in the
dealth Planning and Construction subfunction
                                               (554) woulG
be reconsidered in 1982, Health Care Services
                                                (551) and
Health Research and Education (552) programs
                                               would be
reconsidered in 184 and programs in the
                                           Prevention ana
Control of Health Problems (553) subfunction
                                              would be re-
considered in 1986. While successfully
                                        distributing the
workload in the health budget function
                                        (550s) over three
Congresses, the schedule would make it difficult
                                                  to review
our overall health policy in a comprehensive
     If Congress wished to comprehensively reconsider
Federal Government's efforts and role in
                                          the health policy
area, programs in each of the health subfunctions,
                                                    as well
as the health aspects of programs in other
 (e.g.,   veterans Hospital ana   iedical Care (703) anu Pollution
 Control and Abatement (304)) might also neea to be considered.

      Let me emphasize that thcse (or similar) problems

 would be found in any fixed, comprehensive schedule.
comments are intended to illustrate the problems, not
criticize the details of the schedule in Title I of
                                                    the bill.
      In our opinion, the set of programs that are revieweo

together should depend on the issues involved and tne
tives of the review.    To assure that congressional policy
objectives    ominate the process (rather than the schedule)

additional scheduling flexibility seems necessary.       At a
minimum, we believe the bill should leave room for reviewing

multi-purpose programs in more than one context if necessary.

This would permit each of the health subfunctions to
reconsidered as scheduled in 1962,     14   anda 195, for ex-
ample, but would also allow the Congress to reconsider
entire Federal effort in the health policy area at one
Program Evaluation Requirements

     In response to concerns over the workload that would

be required to rigorously reevaluate all Federal programs

every five or six years, S.2 was amended to allow the

Congress to focus review and reconsideration efforts
they will be most effective.      Title III of the   ill would
establish a procedure allowing each Ho-se of Congress
select certain programs for comprehensive evaluation.

Section 303 of the bill lists twelve criteria to be used
in conducting such "comprehensive" evaluations.      Although
we agree with the intent of this Title, we are concerned

with the practicality of its specific provisions.

     We believe that it will be difficult. if not im-
possible, to conduct the kind of comprehensive evalua-

tions defined in section 303 wit. n the timeframes esta-

blished by Title III, as the Prcsident would have, at most,
e< ht and one half months for the conduct of any comprehensive

evaluation.   Depending on such factors as the size and com-
plexity of a program, the availability of data and the type
of analysis required, a full-scale evaluation may properly

take several years to complete.      When this is the case, the
imposition of short time frames runs a risk of degrading

the validity of the results.

    The twelve criteria for comprehensive evaluation listed
in Section 303 can be sorted into two broad categories:

     (1) those that deal with measuring the actual performance,

        impacts, and results of continuing programs (program

        review or performance monitoring criteria) and
    (2) those that deal with identifying and assessing

        alternative ways of improving the Federal government's

        performance in the future (program reconsideration
        or policy analysis criteria).

       We believe that Title III could be
                                           made more effective
  if a distinction is made between these
                                          two activities. The
 process of reviewing programs (that
                                       is, monitoring and
 measuring what is happening) should
                                      go on all the time.
 The information collectea in this
                                    way, together witn in-
 formation from other sources, cr
                                    then provide a basis for
 deciding which programs should be
                                    comprehensively recon-
 siaered. In this context, the reconsideraticn
                                                  process woulo
 focus on alternatives --   .ti:nuation,   major modification
 in program design, or termination.

     We believe this distinction should
                                         be made because the
process of measuring actual program
                                    performance usually
entails long term research, fact-finding
                                          and data collection
efforts -- efforts which likely
                                cannot be completed within
 the eight and one-half month timeframe
                                         o Title III. However,
program reviews produce the feedback
                                       and knowledge of Federal
program performance, results ano impacts
                                           needed to support
comprehensive reconsideration efforts.
                                           e strongly believe
that the Congress should encourage
                                    the develcpment of infor-
mation on the actual operations and
                                     performance of programs
to serve as the basis for its reconsideration
                                                efforts. iithout
feedback on how programs really work,
                                       effective analysis
is impaired and important, potentially
                                       answerable questions
will be left unanswered.

      The amended version of S.2 recognizes quite appro-
priately that the Congress has many sources of feedback

on how well programs work (e.g., constituents, interest

groups, newspapers, etc).    The perceived quality cf these
sources of feedback will, of course, affect the Congress'

aemand for frmal program evaluation studies.     However,
because systeratic and quality feedback is needed to effec-

t,   ly manage and incremently improve programs, we recommend

that the Congress, when enacting major authorizing legisla-

tion, include a requirement for agencies to review their

programs.   To the extent possible, authorizing legislation
should also indicate the kinds of oversight questions the
Congress wants answered by the responsible agency.    by
establishing statutory feedback requirements in the context

of the authorizing decisions, reviews can be tailored to
the nature of the programs and to the specific oversight
interests of the Congress.

     One way of encouraging this sort of oversight planning
would be to revise Section 102(a) to require committees,

when reporting authorizing legislation, te include an over-

sight or program review section in the legislation (or
in committee reports) to assure that program performance

information is gathered and made available for the congres-

sional review and reconsideration process.

     Kith program performance information
                                          developed by the
 agency review efforts over the entire
                                       review cycle (5 to 6
 years) available for inclusion in any
                                       reconsideration reports
 required under Title III, the selection
                                         procedure in Sec    -
 tion 302 coula e used to select programs
                                           for comprehensive
 reconsideration.     e would encourage the Congress to specify
 in the resolution (or supporting committee
                                             reports) aopteo
 pursuant to Section 302, reconsideration
                                           issues, questions
 and options it wishes the President to
                                         address in each of
 the program areas selected for comiprehensive
      If a distinction is not made between
                                           program review
ano program reconsideration, then at
                                      a minimum, the bill's
procedure for selecting programs for
                                      comprehensive evaiiation
should be odified to enable committees
                                         to specify areas
of review for future Congresses, thus
                                       allowing more than
eight and one-half months for the conduct
                                           and reporting of
evaluations by the executive branch.
Tax Expenitures
Soulo hot Be xclude

      he are also concerned that S.2, as reporteo,
                                                   does not
provide for the periodic review and reconsideration
                                                     of tax
expenditures. We believe that if the
                                       Congress wishes to
comprehensively reconsider the Federal
                                        effort and role in
broad policy areas, it will be necessary
                                          for the full range
of programs --   including direct loans, loan guarantees,
tax expenditures ano other subsidies,
                                      grants and direct
operations -- to be reviewed and reconsidered.

  Transition rovisions for
  Programs that are Terminated

      S.2, as reported, contains no transition
 for programs that are allowed to
                                  terminate because to o
 so, according to the sponsors of
                                  the bill, "would severely
 limit the option.  available to the Congress when it
 not to reauthorize a program." If
                                      the Congress comes to
 the point of allocwing a program to
                                      terminate, sone sort of
 legislation will usually be needed
                                      to provide for the oroerly
transition of personnel and residual
                                        authority. For example,
 in the case of a loan rogram, the
                                     Congress would likely
wish to terminate loan making but
                                    continue the loan collection
activities of the program.     e suggest that the bill's evalu-
ation provisions be modifiea to include
                                           some consideration
of what could be cone in the event
                                   of program trmination.
Program Inventory

      In Title II of the bill, as reported,
                                             the Congres-
sional udget Office (C8O) is given
                                      responsibilities for
compiling and maintaining an inventory
                                         of Federal programs.
In fulfillment of our responsibilities
                                         under Title VIII
of the Budget Act, during the past
                                    three vears we have in-
vestea subst-antial resources in
                                 the development of a data
base of ederal outlay and regulatory
                                       programs. Thus, S,2,
by assigning program inventory responsibility
                                                to the CL.,
will likely result in some duplication
                                       of effort with our
program inventory work. Consequently,
                                       we believe that
 assigning the responsibility to GAO would be a more efficient

 ana economical approach.     However, if the bill should be
 enacted in its present form we will, of course, prcvioe

 CLG    with whatever assistance we can from our continuing
 work unaer Title VlIl    of the Congressinal   udget Act.
 bAG Assistance in
 laentifying Programs
 for heconsiaeration

        he believe that our office could make a substantial
contribution to the process of identifying programs in need
of comprehensive evaluation. A the versight arm of the

Congress, our staff has substantial knowledge of the
operations of Federal programs.      A large part of our work
is now classified as program evaluation and most of our work

(including that which is not classified as program evaluation)

is directly related to the process of reviewing and improving

the operations of government programs.      e believe that this
knowledge would be put to good use if our office, along with

the President and congressional committees, were specifically
identified as a source of recommendations in the selection

of programs for which comprehensive evaluation would appear

to oe worthwhile.

        Our office, of course, will do all that it can to assist
committees in carrying our their responsibilities under this
legislation. hr. Chairman, this concludes our prepared state-
ment.    We would be happy to respond to any questions.