Bureau of Labor Statistics: Making the CPI More Reflective of Current
Consumer Spending (Testimony, 04/29/98, GAO/T-GGD-98-115).

GAO discussed whether the Consumer Price Index (CPI) published by the
Bureau of Labor Statistics (BLS) accurately reflects consumer spending,
focusing on the: (1) need for and advantages of more frequent updating
of the CPI expenditure weights; (2) nature of the work GAO is doing with
regard to the CPI; and (3) BLS' coverage of CPI improvement efforts in
its strategic plan and fiscal year (FY) 1999 annual performance plan.

GAO noted that: (1) BLS uses expenditure weights to aggregate market
basket items into the overall index number; (2) the preponderance of
evidence GAO obtained pointed to the need for and advantages of more
frequent updates; (3) this evidence included the: (a) views of
professionals knowledgeable about the CPI; (b) practices of other
countries; (c) results of research that show that the age of expenditure
weights affects the CPI; and (d) the sizable effect more frequent
updates could have on the federal budget in comparison to the relatively
small costs associated with updates; (4) BLS has said it agrees with
GAO's recommendation for more frequent updates and is considering the
appropriate update frequency; (5) in examining BLS' FY 1999 performance
plan, GAO found that it was partially successful in providing a picture
of BLS' intended performance to improve the CPI's quality; (6) further,
the plan did not fully portray how BLS' strategies and resources would
help achieve the performance goals for improving CPI quality or how BLS
would ensure that the data it uses to assess its performance are
credible; (7) GAO found linkages between some, but not all, of the
commission's recommendations and the plan's performance goals and
indicators; (8) neither BLS' strategic plan nor its FY 1999 performance
plan discusses such linkages or the lack of them; (9) BLS and the
Department of Labor question the usefulness of discussing
recommendations contained in particular reports, such as the Boskin
commission report, in long-range planning documents; (10) however, GAO
believes such a discussion would enhance the plans' usefulness and
credibility to CPI stakeholders given the great interest shown in the
Boskin commission recommendations by Congress and others.

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

 REPORTNUM:  T-GGD-98-115
     TITLE:  Bureau of Labor Statistics: Making the CPI More Reflective 
             of Current Consumer Spending
      DATE:  04/29/98
   SUBJECT:  Price indexes
             Economic analysis
             Strategic planning
             Inflation
             Agency missions
             Reporting requirements
             Economic policies
             Foreign governments
IDENTIFIER:  Consumer Price Index
             Japan
             Italy
             Germany
             France
             Canada
             United Kingdom
             
******************************************************************
** 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


Before the Subcommittee on Human Resources Committee on Government
Reform and Oversight
House of Representatives

For Release on Delivery
Expected at
11:00 a.m.  EDT
Wednesday
April 29, 1998

BUREAU OF LABOR STATISTICS -
MAKING THE CPI MORE REFLECTIVE OF
CURRENT CONSUMER SPENDING

Statement of
Bernard L.  Ungar, Director
Government Business Operations Issues
General Government Division

GAO/T-GGD-98-115

GAO/GGD-98-115T


(410314)


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

  BLS - x
  CES - x
  CPI - x
  CBO - x
  OMB - x

BUREA OF LABOR STATISTICS:  MAKING
THE CPI MORE REFLECTIVE OF CURRENT
CONSUMER SPENDING
====================================================== Chapter Summary

The principal source of information on trends in consumer prices and
inflation in the United States is the Consumer Price Index (CPI),
which is published by the Bureau of Labor Statistics (BLS).  To
determine the level of inflation in consumer prices, the CPI tracks
prices for a fixed "market basket" of goods and services that urban
consumers purchase for day-to-day living. 

As requested by the Subcommittee, GAO's testimony today addresses
certain topics related to the CPI.  The first is the need for BLS to
make the CPI expenditure weights more current by updating them more
frequently than at 10-year intervals, as GAO reported in October
1997.  BLS uses expenditure weights to aggregate market basket items
into the overall index number.  The preponderance of evidence GAO
obtained pointed to the need for and advantages of more frequent
updates.  This evidence included the (1) views of professionals
knowledgeable about the CPI, (2) practices of other countries, (3)
results of research that show that the age of expenditure weights
affects the CPI, and (4) the sizable effect more frequent updates
could have on the federal budget in comparison to the relatively
small costs associated with updates.  BLS has said it agrees with
GAO's recommendation for more frequent updates and is considering the
appropriate update frequency. 

GAO also examined how well those elements of BLS' strategic plan and
performance plan that focus on improving the CPI would meet the
criteria in the Government Performance and Results Act of 1993 and
related guidance, even though the BLS plans are not required to
comply with the statutory requirements.  In examining BLS' fiscal
year 1999 performance plan, GAO found that it was partially
successful in providing a picture of BLS' intended performance to
improve the CPI's quality.  Further, the plan did not fully portray
how BLS' strategies and resources would help achieve the performance
goals for improving CPI quality or how BLS would ensure that the data
it uses to assess its performance are credible. 

GAO also reviewed the linkage between the goals in BLS' strategic
plan relating to improving the CPI and the 1996 Boskin commission
recommendations to BLS.  GAO found linkages between some, but not
all, of the commission's recommendations and the plan's performance
goals and indicators.  Neither BLS' strategic plan nor its fiscal
year 1999 performance plan discusses such linkages or the lack of
them.  BLS and the Department of Labor question the usefulness of
discussing recommendations contained in particular reports, such as
the Boskin commission report, in long-range planning documents. 
However, GAO believes such a discussion would enhance the plans'
usefulness and credibility to CPI stakeholders given the great
interest shown in the Boskin commission recommendations by Congress
and others. 


BUREA OF LABOR STATISTICS:  MAKING
THE CPI MORE REFLECTIVE OF CURRENT
CONSUMER SPENDING
==================================================== Chapter Statement

Mr.  Chairman and Members of the Subcommittee: 

I am pleased to be here today to assist the Subcommittee in its
deliberations with the Bureau of Labor Statistics (BLS) on the
Consumer Price Index (CPI), one of the most important indexes
produced by the federal government.  The Subcommittee has asked us to
address several topics related to making the CPI more reflective of
current consumer spending. 

According to BLS, the CPI is the principal source of information on
trends in consumer prices and inflation in the United States and
affects nearly all Americans because of the many ways it is used. 
For example, it is often used to increase wages and benefit payments
to adjust for the erosion of consumer purchasing power due to
inflation.  In 1997, the federal government adjusted the benefits of
43.6 million Social Security beneficiaries and 21.4 million food
stamp recipients because of the growth in consumer prices as reported
by the CPI.  In addition, federal tax brackets are adjusted
automatically by the CPI to prevent inflation-induced increases in
tax rates. 

To gauge the level of inflation in consumer prices, the CPI tracks
prices for a fixed "market basket" of goods and services that urban
consumers purchase.  These purchases are for food, clothing, shelter,
fuels, transportation, medical care, entertainment, and other goods
and services that people buy for day-to-day living.  Only those
expenditures made by consumers (not including businesses) are
captured in the CPI. 

Congress has been concerned with whether the CPI accurately reflects
consumer spending, and the Senate several years ago chartered the
Advisory Commission to Study the Consumer Price Index (also referred
to as the Boskin commission).  The commission concluded in its
December 1996 report that the fixed market basket CPI (1) becomes
less and less representative of consumer spending over time as
consumers respond to price changes and new consumer choices and (2)
overstates the true cost of living.\1 The commission estimated the
size of the overstatement for the next few years to be 1.1 percentage
points per year, and it made recommendations to BLS and to the
President and Congress.  Within the community of professional
economists and statisticians, the commission's report is supported by
some and opposed by others. 

As requested, I will discuss three topics today:  (1) the need for
and advantages of more frequent updating of the CPI expenditure
weights, (2) the nature of the work we are currently doing with
regard to the CPI, and (3) BLS' coverage of CPI improvement efforts
in its strategic plan and fiscal year 1999 annual performance plan. 


--------------------
\1 The Boskin commission's December 1996 report to the Senate
Committee on Finance was titled Toward a More Accurate Measure of the
Cost of Living. 


   NEED FOR AND ADVANTAGES OF MORE
   FREQUENT UPDATING OF THE CPI
   EXPENDITURE WEIGHTS
-------------------------------------------------- Chapter Statement:1

In an October 1997 report, we concluded that the CPI expenditure
weights should be updated more frequently than at the current rate of
approximately every 10 years to make the fixed market basket CPI more
timely in its representation of consumer expenditures.\2

The CPI is based on a fixed market basket of goods and services
purchased by urban consumers.  These items are assigned weights by
BLS to give proportionate emphasis for price changes of one item in
relation to other items in the CPI.  The weights used to aggregate
the items into the overall index number for the CPI are based on the
Consumer Expenditure Survey, which BLS uses to determine what goods
and services consumers are buying.  Historically, BLS has changed the
expenditure weights only when making major revisions to the CPI. 
Since 1940, that has occurred about every 10 years. 

My comments on the need for and advantages of more frequent updating
of the CPI expenditure weights are based largely on our October
report and will provide information on (1) the views of individuals
knowledgeable about the CPI on the issue of updating the expenditure
weights between major revisions to the CPI, (2) the updating
practices of other industrial countries, (3) the cost to update the
expenditure weights, (4) the possible effect on the federal budget of
more frequent updates, and (5) BLS' response to our recommendation
that it should update the expenditure weights more frequently. 


--------------------
\2 Consumer Price Index:  More Frequent Updating of Market Basket
Expenditure Weights Is Needed (GAO/GGD/OCE-98-2, Oct.  9, 1997). 


      MORE FREQUENT UPDATING
      DEEMED DESIRABLE BY
      INDIVIDUALS KNOWLEDGEABLE
      ABOUT THE CPI
------------------------------------------------ Chapter Statement:1.1

We spoke with 10 individuals who were former officials of BLS or who
had otherwise studied the CPI, and they were unanimous in stating
that 10 years between updates of the expenditure weights was too
long.  However, there was less agreement among the individuals on
exactly how often the updating should occur.  Five, including the
chairman and three members of the Boskin commission, said more
frequent updating of expenditure weights was less important than
other ways of making the CPI more reflective of current consumer
spending.  Dr.  Boskin told us that if there were no changes in
existing products or no new products in the economy, then updating
the expenditure weights would be the only step that BLS would need to
take.  However, since the economy does change, with new products
being developed and product improvements occurring constantly, he
believed more frequent updating was only one step among a number that
should be taken to improve the CPI. 


      PRACTICES OF OTHER
      INDUSTRIAL COUNTRIES
------------------------------------------------ Chapter Statement:1.2

We also obtained information on the updating practices in Japan,
Italy, Germany, France, Canada, and the United Kingdom, which along
with the United States have made up the Group of Seven (G-7)
countries.\3 All of them track consumer prices through a market
basket of goods and services and weight the prices of the items in
the market basket.  According to information provided by BLS and
contained in international publications, all six update their
consumer price indexes more often than the United States.  Two
updated the weights of their consumer price indexes annually, and the
other 4 did so approximately every 5 years.  However, BLS officials
noted that the updates by the six countries do not comprise the same
level of detail as used for the CPI; for example, some do not use
consumer expenditure surveys as does the United States. 


--------------------
\3 The G-7 countries have met to coordinate economic and monetary
policy. 


      COST TO UPDATE THE
      EXPENDITURE WEIGHTS IS
      ESTIMATED TO BE RELATIVELY
      SMALL
------------------------------------------------ Chapter Statement:1.3

The estimated cost of updating the expenditure weights is relatively
small in comparison to the cost of major revisions.  To estimate the
cost for our October 1997 report, we assumed that an update to the
expenditure weights would have occurred in 1992 and would occur in
2003, which is 5 years after major revisions to the CPI market
basket.  On the basis of data supplied by BLS, the estimated cost to
have updated the weights in 1992 would have been $2.4 million spread
over 3 years.  According to BLS, the estimated cost to update the
expenditure weights in 2003 would be $3.1 million over a 3-year
period.  In comparison, BLS reported that the major CPI revision in
1987 cost $47 million over 5 years and, at the time of our report,
expected the cost of the 1998 major revision to be about $66 million
over 6 years.  BLS explained this difference in cost by pointing out
that an update of the expenditure weights would exclude many
activities that are included in major CPI revisions. 


      MORE FREQUENT UPDATES COULD
      AFFECT THE FEDERAL BUDGET
------------------------------------------------ Chapter Statement:1.4

Because federal tax brackets and federal payments, such as those to
Social Security beneficiaries, are adjusted for inflation, a CPI that
more accurately measures inflation could affect the federal budget. 
BLS estimated the annual range of change in the CPI, if the
expenditure weights were updated on a 5-year cycle, from 0 (zero)--no
change--to a decrease of 0.2 percentage point.  We asked the
Congressional Budget Office (CBO) to use the midpoint of BLS' range
(a decrease of 0.1 percentage point) to estimate the effect on the
federal budget.  CBO estimated that, assuming no other changes in
policy or economic assumptions, if updating the weights in 2003 (5
years after the 1998 major revision) reduced CPI growth by 0.1
percentage point annually, the projected budget surplus would be
increased by a cumulative total of $16.5 billion over the 5-year
period of 2004 through 2008.\4

Most of the impact of such a reduction in the CPI would be on federal
outlays--such as reduced payments to Social Security beneficiaries,
which account for most of the outlays--and most of the impact would
occur in the later years.  For example, according to estimates by SSA
actuaries using an annual 0.1 percentage point reduction in CPI
growth, the average monthly benefit check for retired workers in 2004
would be reduced by $0.91, from $939.94 to $939.03; by the fifth year
(2008), the average monthly check would be reduced by $4.86, from
$1,065.98 to $1,061.12. 


--------------------
\4 In our October 1997 report, we gave a 4-year CBO projection of
$10.8 billion, which was based on a March 1997 baseline.  CBO used a
March 1998 baseline to make the 5-year projection. 


      BLS' RESPONSE TO OUR OCTOBER
      1997 RECOMMENDATION
------------------------------------------------ Chapter Statement:1.5

In our October 1997 report, we recommended that as long as a fixed
market basket CPI is published, the Commissioner of BLS should update
the expenditure weights of the CPI's market basket of goods and
services more frequently than every 10 years.  BLS plans to publish,
starting in 2002, a CPI index based on the concept of superlative
index formulas, which would enable that index to be more current than
the fixed market basket index in reflecting consumer spending.  The
Boskin commission recommended use of a superlative index formula to
construct the CPI, although BLS believes the recommendation did not
specify how a true superlative index might be constructed.  Moreover,
BLS does not view the superlative index as a replacement for the
fixed market basket index, and said it would continue to publish the
fixed market basket CPI. 

In commenting on GAO's recommendation, the BLS Commissioner said she
supports more frequent updates of the expenditure weights.  However,
she said neither economic theory nor empirical evidence demonstrates
the superiority of any particular update interval.  This has been the
principal reason why BLS has not updated the expenditure weights
between major revisions to the CPI.  Other reasons cited by BLS
included difficulties in obtaining funds to bring about change to the
CPI and concern with what would be the best approach to improve the
CPI. 

Although theoretical guidance is not available on all facets of
updating expenditure weights, such as exactly how often updates
should occur, the preponderance of the data we reviewed supported the
need for updating expenditure weights more frequently than
approximately every 10 years.  Recognizing that the data are not
perfect and do not isolate the effects of using outdated expenditure
weights, comparisons of price indexes with old and new weights that
go back to those made for the first revision in 1940 indicate that
price indexes computed with more current weights were always
different from indexes computed with older weights.  These
comparisons and more recent research conducted by BLS tend to show
lower rates of inflation with indexes using newer weights. 

Since last fall when we received the Commissioner's comments, the
Commissioner has told this Subcommittee that BLS is committed to more
frequent updating of the weights used to calculate the CPI.  In March
1998, the Commissioner said that, although BLS had not yet decided on
the optimal frequency of weight updates, it was clear that a
reduction in the current (approximately 10-year) period between
updatings was warranted. 


   CURRENT GAO WORK ON THE CPI
-------------------------------------------------- Chapter Statement:2

At the request of the Ranking Minority Member of the Senate Committee
on Governmental Affairs, the Honorable John Glenn, we are gathering
information from BLS related to the quality adjustment issues raised
by the Boskin commission.  To better inform policymakers and other
commentators on this issue, we have been asked to describe the
policies, procedures, and practices that BLS currently uses to
account for differences between a new item and the item it replaced. 
This replacement process is necessary when an item is no longer
available for BLS to price.  Within this process, BLS determines if
the new item is comparable to (i.e., the same as) the old item that
disappeared.  We are in the process of collecting information to
describe what BLS does when the two items are not comparable.  We
anticipate issuing our report early next year. 


   BLS' COVERAGE OF CPI
   IMPROVEMENT EFFORTS IN ITS
   STRATEGIC PLAN AND 1999
   PERFORMANCE PLAN
-------------------------------------------------- Chapter Statement:3

For this testimony, you asked us to review certain planning documents
BLS prepared for improving the accuracy of the CPI.  The model for
such planning documents is the Government Performance and Results Act
of 1993, also referred to as the Results Act.  The Act seeks to
improve the efficiency, effectiveness, and accountability of federal
programs by establishing a system to set goals for programs based on
their intended purposes, measure the performance of those programs,
and use performance information to improve results.  As you know, if
successfully implemented, the Results Act will help agencies focus on
how to improve their programs' performance in achieving desired
results. 

The approach to performance-based management and accountability
envisioned by the Results Act is a dynamic and iterative process in
which one stage builds on and reinforces the progress made at earlier
stages.  Under the Results Act, agencies first were to prepare
long-term strategic plans that set the general direction for their
efforts.  The strategic plans agencies prepared were to be the
starting point for agencies to set performance goals for programs in
their annual performance plans. 

The Results Act does not require component agencies of departments,
such as BLS, to prepare strategic and annual performance plans. 
However, some component agencies have prepared those plans as BLS has
done at the direction of the Department of Labor.  Although the
strategic and annual performance plans prepared by component agencies
such as BLS are not required to comply with the requirements of the
Results Act, these component agencies may use the statutory
requirements as guidance in developing their strategic and
performance plans.  In addition, the Department of Labor instructed
BLS to follow OMB's guidance to federal agencies for implementing the
Results Act in preparing its annual performance plan.  Moreover,
Labor included the entire BLS strategic plan as part of the
department's strategic plan. 

BLS' strategic plan includes an agency strategic goal to "improve
accuracy, efficiency, and relevancy of our economic measures and
program outputs through increased application of state-of-the-art
statistical techniques, economic concepts, technology, and management
processes." The CPI is one of these economic measures, and one of the
strategic plan's objectives is to improve the quality of the CPI. 
The strategic plan provides performance goals and performance
indicators for this objective (see app.  I for those for fiscal year
1999). 

BLS' fiscal year 1999 annual performance plan lists the same agency
performance goals and performance indicators for the CPI that are
listed for fiscal year 1999 in BLS' strategic plan.  Thus, the
performance goals and indicators in the 1999 annual performance plan
are linked to those in the strategic plan. 


      OBSERVATIONS ON BLS' FISCAL
      YEAR 1999 ANNUAL PERFORMANCE
      PLAN
------------------------------------------------ Chapter Statement:3.1

As you requested, we reviewed BLS' performance plan for fiscal year
1999, focusing on the plan's elements that related to the objective
of improving the quality of the CPI.  To do this review, we used
criteria in the Results Act, Office of Management and Budget's (OMB)
guidance on developing the plans (Circular A-11, part 2), our
February 1998 guidance for congressional review of the plans,\5 our
April 1998 guidance for assessing agencies annual performance
plans,\6 and the December 17, 1997, letter to OMB Director Raines
from eight congressional leaders that includes suggestions for how to
make the performance plans more useful.  For purposes of our
analysis, we collapsed the six requirements for annual performance
plans in the Results Act and the related guidance into three core
questions:  (1) To what extent does the agency's performance plan
provide a clear picture of intended performance across the agency? 
(2) How well does the performance plan discuss the strategies and
resources the agency will use to achieve its performance goals?  (3)
To what extent does the agency's performance plan provide confidence
that its performance information will be credible? 

Overall, BLS' fiscal year 1999 performance plan would partially meet
the criteria in the Results Act and related guidance, although, as
stated previously, we recognize that the plan was not required to
comply with the statutory requirements.  Considering that this is the
first Results Act performance plan that BLS has produced, the plan
contains a great deal of useful information to inform Congress and
others about how BLS intends to accomplish its mission.  We expect
that as BLS gains experience, future performance plans will build
upon this initial effort and become increasingly more useful to
Congress and the public.  Specifically with respect to the CPI, BLS'
fiscal year 1999 performance plan provides a partial picture of BLS'
intended performance relative to its goals aimed at improving the
quality of the CPI.  Further, the plan does not fully portray how
BLS' strategies and resources will help achieve BLS' performance
goals for improving CPI quality or how BLS will ensure that
information it uses to assess actual performance against CPI
improvement goals is accurate, complete, and consistent. 

Under the criteria in the Results Act and related guidance, agency
annual performance plans are to contain goals for each of the program
activities identified in the agency's budget and should express them
in objective, quantifiable, and measurable form to allow comparison
between actual and planned performance.  Agencies can use their
discretion in determining their performance goals.  Agency plans
should also contain indicators to be used in measuring or assessing
the relevant outputs, service levels, and outcomes of each program
activity.  In our view, BLS' performance goals associated with
improving the quality of the CPI, which are shown in appendix I, are
linked to its budget and are objective and measurable.  As BLS notes,
the goals are generally expressed as significant milestones expected
to be accomplished during the given year. 

Because of the way in which the goals are expressed, BLS' performance
indicators for accomplishing the milestones are activity or output
oriented rather than outcome oriented.  BLS notes this in its plan
and says that, over time, it will show how its indicators relate to
desired outcomes, such as improving the accuracy of the CPI.  The use
of output-oriented goals is allowed under OMB's guidance for
implementing the Results Act.  The guidance notes that, although
outcomes should be used whenever possible, outputs may be used for
several reasons, including when outcomes may not be scheduled for
achievement in the fiscal year covered by the plan.  However, it
appears to us that the goals and indicators BLS cites in its
performance plan are, for the most part, changes to the CPI that BLS
expects to complete in fiscal year 1999 that are aimed at improving
the CPI's quality.  Thus, it would appear that BLS' performance plan
would be more fully responsive to OMB's guidance if the goals or
indicators included an outcome dimension to reflect the results of
BLS' activities.  Such outcome-oriented goals or indicators could
relate to improvements in the accuracy, efficiency, or relevancy of
the CPI. 

Furthermore, BLS' goals and indicators could be more specific and
quantifiable to enable better stakeholder comparisons of actual
results with established goals.  For example, BLS' indicator for its
performance goal to expand the Consumer Expenditure Survey (CE)
sample is:  "Begin data collection and processing of larger sample."
In its justification for its submission for the President's budget,
BLS requested resources to expand the sample size of the CE by
approximately 50 percent.  The performance goal as set forth in the
annual plan is not as specific, however, and does not inform the
reader that BLS plans to expand the CE by 50 percent.  Also, if the
baseline measure cited the prior CE sample size, which it does not
do, then CPI stakeholders could determine whether BLS has achieved
its goal to expand the CE by 50 percent when it reports its results. 
We also note that none of the performance goals to improve the
quality of the CPI provide baseline measures. 

Under the criteria in the Results Act and related guidance, annual
performance plans should briefly describe the agencies' strategies
and the human, capital, financial, and other resources they will use
to achieve their performance goals.  BLS' performance plan indicates
that state-of-the-art statistical techniques, economic concepts,
technology, and management processes will be applied to achieve its
performance goals aimed at improving the quality of the CPI.  It also
mentions the operational processes and technological advances that
BLS is making.  Further, the plan points out that BLS is requesting
additional funding for fiscal year 1999 of about $9.1 million and 36
full-time equivalent employees to undertake initiatives needed to
improve the CPI's timeliness and accuracy. 

However, BLS' performance plan does not fully portray how BLS'
strategies and resources will help achieve the BLS performance goals. 
We believe such information would better enable CPI stakeholders to
see how BLS intends to achieve each goal and the associated resources
it will need to do so.  BLS' performance plan does link each
performance goal aimed at improving the quality of the CPI with BLS'
program activities listed in the President's budget and shows the
increased amount of funding and full-time equivalent staff associated
with achieving all such goals.  However, BLS' performance plan does
not show the funding level associated with the individual program
activities identified in its performance plan, as provided in OMB's
Results Act guidance. 

Finally, under the criteria in the Results Act and related guidance,
agencies should describe the means to be used to verify and validate
performance data.  According to this guidance, the means used should
be credible and specific to ensure that performance information is
sufficiently complete, accurate, and consistent.  BLS' performance
plan for fiscal year 1999 does not specify the procedures BLS intends
to use to verify and validate the information related to its
performance goals for improving CPI quality.  However, in its
performance plan for fiscal year 1999, BLS states that over time it
will validate its indicators by showing how they relate to desired
outcomes, such as improving accuracy.  In addition, we believe it
would be more useful to CPI stakeholders if the plan included
information to demonstrate the credibility of the data BLS will use
to measure its performance. 


--------------------
\5 Agencies' Annual Performance Plans Under the Results Act:  An
Assessment Guide to Facilitate Congressional Decisionmaking
(GAO/GGD/AIMD 10.1.18, Feb.  1998). 

\6 The Results Act:  An Evaluator's Guide to Assessing Agency Annual
Performance Plans (GAO/GGD-10.1.20, April 1998). 


      OBSERVATIONS ON LINKAGE
      BETWEEN BOSKIN COMMISSION'S
      RECOMMENDATIONS AND BLS' CPI
      PERFORMANCE GOALS AND
      INDICATORS
------------------------------------------------ Chapter Statement:3.2

As you requested, we also compared the extent to which certain BLS
performance goals and performance indicators contained in its
strategic plan that relate to improving the CPI reflected the
recommendations made in December 1996 by the Boskin commission. 

Overall, we found linkages between BLS' performance goals and
indicators relating to improving the CPI and three Boskin commission
recommendations (see app.  II).  The commission referred to these
specific recommendations as short-run recommendations--that is, those
that the commission thought could be implemented immediately, with
little additional resources or no data collection initiatives.\7
These recommendations were to develop a monthly index that would
adopt a superlative index formula at the upper level of aggregation
in the index, to be called the CPI, and replace the current fixed
market basket CPI;\8 use a geometric means formula at the lower level
of aggregation in the CPI;\9

and expand the use of regression techniques.  We based the linkages
on concepts that were identified in the recommendations and the
performance goals and indicators.  For example, we made a linkage
between the Boskin recommendation that said BLS should move to a
geometric means at the elementary aggregates level and the
performance goal and related indicators that mentioned geometric
means. 

BLS' performance goals and indicators that we linked did not always
address the full scope of the Boskin recommendations.  For example,
BLS has identified a series of performance indicators for the
performance goal "develop alternative measures of change in living
costs" that would result in published superlative indexes by fiscal
year 2002, which we linked to the third Boskin commission
recommendation.  We note that the plan's discussion of this
performance goal and its indicators does not address several aspects
of that Boskin recommendation, such as abandoning the current fixed
market basket CPI.  In addition, this performance goal and related
indicators indirectly relate to the two overall Boskin commission
recommendations for BLS to establish a cost-of-living index as its
objective and to develop and publish a monthly index and an annual
index.  If the strategic plan provided more detailed information, it
would help inform stakeholders about the frequency (monthly and/or
annually) with which the planned alternative index measures will be
published and would allow CPI stakeholders to make appropriate
linkages to the Boskin commission's recommendations. 

We found no linkage between the performance goals and indicators and
the eight intermediate and longer-run recommendations of the Boskin
commission.  To determine BLS' views on these recommendations, we
relied upon a written statement BLS sent to the Chairman of the Joint
Economic Committee in June 1997.\10 BLS has provided Congress with
its views on the Boskin commission report on several occasions; for
example, in addition to the June 1997 statement, it also provided
testimony to the Senate Budget Committee in January 1997 and to the
House Budget Committee in March 1997.  BLS informed us that any
Boskin commission recommendation for which linkages cannot be
identified in its strategic plan were those that either restate
current BLS/CPI policy, are presently in development, or have been
explicitly rejected by BLS.  Further, BLS said that it had initiated
work related to many of the areas covered by the Boskin commission's
recommendations prior to the issuance of the commission's report. 

BLS told us that both it and the Department of Labor question the
usefulness of discussing recommendations contained in particular
reports, such as the Boskin commission report, in long-range planning
documents, including BLS' strategic plan and fiscal year 1999 annual
performance plan.  However, given the high degree of interest that
Members of Congress, the Federal Reserve, and others who are
concerned about the accuracy of the CPI have shown in the Boskin
commission's report and recommendations, we believe that a discussion
of the relationship of the goals in BLS' performance plan for fiscal
year 2000 to the Boskin commission recommendations, or an explanation
of the absence of such relationship, would add to the plan's
credibility and usefulness to CPI stakeholders. 


--------------------
\7 The Boskin commission placed its recommendations to BLS into three
categories:  short run, intermediate run, and longer run.  We placed
two unclassified recommendations into a general/overall category. 

\8 The Boskin commission referred in its recommendation to a
particular form of superlative index but BLS does not agree that the
form referred to should be characterized as a superlative index. 

\9 On April 16, 1998, BLS announced that, beginning in January 1999,
it will use a geometric mean formula to aggregate the lower-level
categories for approximately 61 percent of total consumer spending
represented in the CPI. 

\10 BLS, "Measurement Issues in the Consumer Price Index," paper
prepared in response to a letter from Representative Jim Saxton,
Chairman of the Joint Economic Committee, June 1997. 


------------------------------------------------ Chapter Statement:3.3

Mr.  Chairman, this concludes my prepared statement.  I would be
pleased to respond to any questions that you or other Members of the
Subcommittee may have. 


CPI-RELATED OBJECTIVES,
PERFORMANCE GOALS, AND FISCAL YEAR
1999 PERFORMANCE INDICATORS
=========================================================== Appendix I



                               Table I.1
                
                  CPI-Related Objectives, Performance
                Goals, and Fiscal Year 1999 Performance
                 Indicators for BLS' Strategic Goal to
                   Improve Accuracy, Efficiency, and
                     Relevancy of Economic Measures

                                    Fiscal year 1999 performance
Performance goal                    indicator
----------------------------------  ----------------------------------
Objective: Improve the quality of the Consumer Price Index (CPI)
----------------------------------------------------------------------
Improve the CPI sample design and   Use new sample design and
estimation methodology used for     estimation methodology for rent
CPI rent and Owner's Equivalent     and REQ in published CPI.
(REQ) estimates.

Update the CPI market basket's      Updated Housing sample used in
Housing sample to reflect current   published CPI.
demographic and geographic
population characteristics.

Update the CPI market basket's      Published CPI includes
Commodities and Services sample to   certain previously unpriced
reflect current geographic          services,
population and expenditure           prices collected for new goods
patterns of all goods and           and services entering the market
services.                           place,
                                     quality adjustment for certain
                                    goods and services based on
                                    supplemental sample of prices and
                                    characteristics, and
                                     new item samples based on
                                    Telephone Point of Purchase
                                    Survey.

Develop alternative measures of     Produce updated superlative
change in living costs.             indexes in an enhanced research
                                    environment utilizing 1997
                                    Consumer Expenditure Survey data.

Expand the Consumer Expenditure     Begin data collection and
Survey sample.                      processing of larger sample.

Test an experimental CPI using      Incorporate changes into the
geometric means.                    official CPI.


Objective: Improve service sector coverage in major BLS programs:
Consumer Price Index, Producer Price Index, International Price
Program, and Productivity.
----------------------------------------------------------------------
Increase the coverage of service    Published CPI includes certain
industries' price and productivity  previously unpriced services.
indexes.
----------------------------------------------------------------------
Source:  BLS' strategic plan and fiscal year 1999 performance plan. 


OUR ANALYSIS OF THE LINKAGE
BETWEEN THE BOSKIN COMMISSION'S
RECOMMENDATIONS AND BLS' STRATEGIC
PLAN FOR FISCAL YEARS 1998 THROUGH
2002
========================================================== Appendix II

                                Performance indicator, by the end of fiscal year
                                ------------------------------------------------
Boskin          BLS'
commission's    performance
recommendation  goal            1998            1999            2002\a
--------------  --------------  --------------  --------------  ----------------
General/overall
--------------------------------------------------------------------------------
1. BLS should   \b              \b              \b              \b
establish a
cost-of-
living index
as its
objective in
measuring
consumer
prices.

2. BLS should   \b              \b              \b              \b
develop and
publish two
indexes--one
published
monthly and
the other
published and
updated
annually and
revised
historically.


Short run--recommendations that the commission thought could be implemented
immediately, with little additional resources or new data collection
initiatives.
--------------------------------------------------------------------------------
3. The timely   Develop         Begin research  Produce         Publish
monthly index   alternative     on design       updated         production
should          measures of     issues.         superlative     quality
continue to be  change in                       indexes in an   superlative
called the CPI  living costs.   Produce         enhanced        indexes to
and should                      updated         research        supplement the
move toward a                   superlative     environment     official CPI.
cost-of-                        indexes         utilizing 1997
living concept                  utilizing 1996  Consumer
by adopting a                   Consumer        Expenditure
"superlative"                   Expenditure     Survey data.
index formula                   Survey data.
to account for
changing
market
baskets,
abandoning the
pretense of
sustaining the
Laspeyres
formula.

                Expand the      Prepare for     Begin data      \d
                Consumer        sampling; hire  collection and
                Expenditure     and train       processing of
                Survey          interviewers.   larger sample.
                sample.\c

4. BLS should   Test an         Determine       Incorporate     \d
move to         experimental    which CPI       changes into
geometric       CPI using       basic indexes   the official
means at the    geometric       are best        CPI.
elementary      means.          calculated
aggregates                      with the
level.                          geometric
                                means formula.

BLS should      Update the CPI  \f              Prices          \d
greatly expand  market                          collected for
the use of      basket's                        new goods and
hedonic         Commodities                     services
regression      and Services                    entering the
techniques to   sample to                       market place.
deal with       reflect
quality         current                         Quality
change.\e       geographic                      adjustment for
                population and                  certain goods
                expenditure                     and services
                patterns of                     based on
                all goods and                   supplemental
                services.                       sample of
                                                prices and
                                                characteristic
                                                s.


Intermediate run--reforms that the commission said are feasible within the
current state-of-the-art but that would require new data collection,
reorganization of activities, and changes in the detail of the various sub-
indexes produced by the CPI.
--------------------------------------------------------------------------------
5. BLS should   \f              \f              \f              \f
study the
behavior of
the individual
components of
the index to
ascertain
which
components
provide most
information on
the future
longer-term
movements in
the index and
which items
have
fluctuations
that are
largely
unrelated to
the total and
emphasize the
former in its
data
collection
activities.

6. BLS should   \f              \f              \f              \f
change the CPI
sampling
procedures to
de-emphasize
geography,
starting first
with sampling
the universe
of commodities
to be priced
and then
deciding,
commodity by
commodity,
what is the
most efficient
way to collect
a
representative
sample of
prices from
which outlets,
and only later
turn to
geographically
clustered
samples for
the economy of
data
collection.

7. BLS should   \f              \f              \f              \f
investigate
the impact of
classification
, that is item
group
definition, on
the price
indexes, to
improve the
ability of the
index to fully
capture item
substitution.

8. There are a  \f              \f              \f              \f
number of
additional
conceptual
issues that
require
attention. The
price of
durables, such
as cars,
should be
converted to a
price of
annual
services,
along the same
lines as the
current
treatment of
the price of
owner-
occupied
housing. Also,
the treatment
of "insurance"
should move to
an ex-ante
consumer price
measure rather
than the
currently used
ex-post
insurance
profits based
measure.

9. BLS needs a  \f              \f              \f              \f
more permanent
mechanism for
bringing
outside
information,
expertise, and
research
results to it.


Longer run--recommendations the commission identified as
emphasizing topics and areas that need additional research and
attention.
--------------------------------------------------------------------------------
10. BLS should  \f              \f              \f              \f
develop a
research
program to
look beyond
its current
"market
basket"
framework for
the CPI.

11. BLS should  \f              \f              \f              \f
investigate
the
ramifications
of the
embedded
assumption of
price
equilibrium.

12. BLS should  \f              \f              \f              \f
develop a
number of new
data
collection
initiatives to
make some
progress along
these lines.
--------------------------------------------------------------------------------
Note:  The strategic plan includes additional agency performance
goals and performance indicators, such as those related to the 1998
major revision of the CPI, which are excluded from this table because
they cannot be directly linked with the stated Boskin commission
recommendations.  For a detailed response by BLS to the Boskin
commission's recommendations, see "Measurement Issues in the Consumer
Price Index," a paper prepared in response to a letter from
Representative Jim Saxton, Chairman of the Joint Economic Committee,
June 1997. 

\a For the agency performance goals listed, there were no performance
indicators listed for fiscal years 2000 and 2001. 

\b To the extent that BLS addressed the short-run recommendations, we
view these general/overall recommendations as having an indirect
linkage to BLS' performance goals and performance indicators. 

\c As part of its fiscal year 1998 budget enhancements proposal, BLS
stated that an expanded Consumer Expenditure Survey would permit the
CPI program to produce a superlative form index.  We therefore linked
this goal to the third recommendation.  It could also be linked to a
Boskin recommendation that was addressed to Congress--"Congress
should provide the additional resources necessary to expand the CES
sample and the detail collected, to make the POPS survey more
frequent, and to acquire additional commodity detail from alternative
national sources, such as industry surveys and scanner data."

\d According to BLS officials, BLS expects to complete its milestones
for the performance goal before fiscal year 2002. 

\e This recommendation was added to the 12 stated recommendations
because, according to commission members, "...we should have listed
it as a numbered explicit recommendation in the last section, since
at least one eminent scholar in the field interpreted its exclusion
as a downweighting of its significance.  We meant the opposite--it
was so obvious (to us) as not to even require repetition as an
explicit formal recommendation." (See Michael J.  Boskin et al.,
"Consumer Prices, the Consumer Price Index, and the Cost of Living,"
Journal of Economic Perspectives, Vol.  12, No.  1, Winter 1998, ft. 
no.  11, p.  14.) We also classified it among the short-run
recommendations since BLS has incorporated it in its strategic plan
for fiscal year 1999. 

\f No linkage was found between the Boskin recommendation and BLS'
performance goals and indicators. 

Source:  BLS and Toward a More Accurate Measure of the Cost of
Living, Final Report to the Senate Finance Committee from the
Advisory Commission to Study the Consumer Price Index, December 4,
1996. 


*** End of document. ***