Economic Statistics: Status Report on the Initiative to Improve Economic
Statistics (Letter Report, 07/07/95, GAO/GGD-95-98).

Pursuant to a congressional request, GAO reviewed the actions taken by
federal agencies to implement the Economics Statistics Initiative (ESI),
focusing on: (1) the Council of Economic Advisers' (CEA) recommendations
that comprise ESI and the context of their development; (2) the
agencies' ESI implementation plans and the associated budget resources
requested and received for fiscal year (FY) 1990 through FY 1994; (3)
ESI implementation actions through May 1994; (4) other completed or
ongoing major program improvements outside of ESI; and (5) FY 1995
budget resources requested and received for economic statistical
improvements and the agencies' views on what is needed to make further
statistical improvements.

GAO found that: (1) CEA made 38 near-term ESI recommendations to address
well-known problems in economic statistics, particularly in measuring
output and quality improvements to goods and services in a rapidly
changing economy; (2) the federal agencies responsible for ESI
implementation made plans to implement most of the ESI recommendations
and requested more than $95 million and received about $50 million from
FY 1990 through FY 1994 to implement their plans; (3) as of May 1994,
most of the agencies' implementation plans were still in progress and
agencies cited budget limitations for delaying their development; (4)
some agencies considered other ongoing nonESI efforts to improve
economic statistics responsive to ESI and did not make new plans; (5)
the agencies believed further efforts were needed to improve economic
statistics, since ESI was never meant to address all the problems with
economic statistics and many ESI plans were not completed; (6) the
agencies have requested $38 million for additional statistics
improvements in FY 1995 and, as of December 1994, $18 million in
appropriations had been approved; and (7) senior policy officials
believe that more funding and better leadership is needed to further
improve economic statistics.

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

 REPORTNUM:  GGD-95-98
     TITLE:  Economic Statistics: Status Report on the Initiative to 
             Improve Economic Statistics
      DATE:  07/07/95
   SUBJECT:  Economic indicators
             Statistical data
             Industrial statistics
             Labor statistics
             Economic analysis
             Projections
             Economic policies
             Appropriations
             Statistical methods
IDENTIFIER:  Economic Statistics Initiative
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


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


Report to the Ranking Minority Member, Committee on Banking and
Financial Services, House of Representatives

July 1995

ECONOMIC STATISTICS - STATUS
REPORT ON THE INITIATIVE TO
IMPROVE ECONOMIC STATISTICS

GAO/GGD-95-98

Economic Statistics Initiative


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

  ACES - annual capital expenditure survey
  ACS - Automated Commercial System
  AEA - American Economic Association
  BEA - Bureau of Economic Analysis
  BEL - Business Establishment List
  BLS - Bureau of Labor Statistics
  CATI - computer-assisted telephone interview
  CEA - Council of Economic Advisers
  CES - current employment statistics
  CNSTAT - Committee on National Statistics of the National Research
     Council
  CPI - consumer price index
  CPS - current population survey
  ECI - employment cost index
  ECPC - Economic Classification Policy Committee
  ESI - Economics Statistics Initiative
  GDP - gross domestic product
  GNP - gross national product
  IMF - International Monetary Fund
  IRS - Internal Revenue Service
  IPP - international price program
  NABE - National Association of Business Economists
  NAICS - North American Industry Classification System
  NAS - National Academy of Sciences
  NASS - National Agricultural Statistics Service
  NBER - National Bureau of Economic Research
  NIPA - National Income and Product Accounts
  NSF - National Science Foundation
  NTDB - National Trade Data Bank
  OMB - Office of Management and Budget
  OTA - Office of Technology Assessment
  PPI - producer price index
  QFR - Quarterly Financial Report
  SIC - standard industrial classification
  SIPP - survey of income and program participation
  SNA - system of national accounts
  SSA - Social Security Administration

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


B-260038

July 7, 1995

The Honorable Henry B.  Gonzalez
Ranking Minority Member
Committee on Banking and Financial Services
House of Representatives

Dear Mr.  Gonzalez: 

You requested that we describe actions taken by federal agencies to
implement the Economics Statistics Initiative (ESI), a group of
recommendations announced in 1990 and 1991 by the Council of Economic
Advisers (CEA) to improve the quality of U.S.  economic statistics. 
This report (1) identifies the recommendations that compose ESI and
describes the context within which they were developed, (2) describes
agency plans developed to implement ESI and the associated budget
resources requested and received for fiscal years 1990 through
1994,\1 (3) describes agencies' actions through May 1994 to implement
the ESI recommendations, (4) describes other major program
improvements outside the ESI efforts that were completed or ongoing
during this time, and (5) summarizes fiscal year 1995 budget
resources requested and received for economic statistical
improvements and the agencies' views on what is needed to make
further improvements.\2 As agreed with your office, we have
separately reported on examples of the effect of selected economic
statistics on the federal budget and economic policymaking.\3


--------------------
\1 The federal agencies responsible for implementing the ESI
recommendations were the U.S.  Department of Commerce, including its
Bureau of the Census and the Bureau of Economic Analysis (BEA); the
U.S.  Department of Labor's Bureau of Labor Statistics (BLS); the
Board of Governors of the Federal Reserve System (Federal Reserve);
the Office of Management and Budget (OMB); the National Science
Foundation (NSF); and the Department of Agriculture's National
Agricultural Statistics Service (NASS). 

\2 The agencies ceased referring to the improvements as responding to
ESI for fiscal year 1995. 

\3 Economic Statistics:  Measurement Problems Can Affect the Budget
and Economic Policymaking (GAO/GGD-95-99, May 2, 1995). 


   BACKGROUND
------------------------------------------------------------ Letter :1

Economic statistics are used by both government and industry to
understand current economic conditions and to predict future
conditions.  Such statistics are part of the process by which
government policies and laws are developed, implemented, and
evaluated; and substantial components of the federal budget are
adjusted to reflect changes in various economic statistics, such as
measures of price changes.  Business and individual decisions on
spending and investment are influenced by economic statistics. 
Economic statistics also feed into academic and other research, which
in turn often influences public policy debate and private sector
decisionmaking. 

Experts in both the public and private sector, including many
government agency officials responsible for producing economic
statistics, have long-standing concerns about the accuracy and
adequacy of these statistics.  Among the most common concerns are
that the statistics inadequately reflect the increased importance of
international transactions to the U.S.  economy, that they do not
provide sufficient data to explain the productivity of the economy,
and that they poorly measure investment and saving. 

In response to these concerns, in April 1989, the Chairperson of CEA
chaired a working group, including both producers and users of
economic statistics within the federal government, that was charged
with assessing these concerns and proposing recommendations to
address them.  In January 1990 and February 1991, CEA issued the
working group's two sets of recommendations for improvement.  In
releasing the recommendations, CEA noted that, although the
recommendations addressed well-known problems with those statistics
that most affected budget and monetary policymaking, the
recommendations were limited to those that were most cost-effective
and could be executed in the short term. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

We identified 38 recommendations in ESI that sought to address
well-known problems in economic statistics.  Because of budget and
time constraints, the ESI working group did not intend these
recommendations to be comprehensive but focused on actions that were
feasible in the near term. 

The federal agencies responsible for implementing the recommendations
made plans to implement most of them.  During fiscal years 1990
through 1994, the agencies requested more than $95 million and
received about $50 million, slightly more than one-half of the funds
they requested, to carry out their plans.\4

Agency officials told us that their plans to address the
recommendations were in various phases of completion, with most of
the plans still reported to be in progress as of May 1994.  In some
cases, agencies considered other efforts undertaken apart from ESI to
be responsive to the ESI recommendations, and thus made no new plans. 
Agencies generally cited budget limitations as reasons for not
completing plans to implement the ESI recommendations. 

Agency officials also said that they made other additional
improvements during the 4 years with existing funding.  According to
these officials, further efforts were needed after fiscal year 1994
because ESI was never intended to address all the problems in
economic statistics and because many of the plans for the ESI
recommendations were not completed.  The agencies requested $38
million for additional improvements in fiscal year 1995 and, as of
December 31, 1994, $18 million in appropriations had been approved. 
Agency officials' comments varied on what they believed was needed to
make further improvements in economic statistics. 


--------------------
\4 The Federal Reserve's funding is not included in this report since
it does not receive appropriated funds from Congress. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :3

To accomplish our first objective of identifying the ESI
recommendations and describing their impetus, we reviewed the two
sets of ESI recommendations, hearing records, and pertinent
literature on economic statistics and interviewed users and producers
of economic statistics and economic statistics experts.  Because the
format and content of the two sets of ESI recommendations differed,
we identified the full list of ESI recommendations as those made in
either or both sets of recommendations.  In some cases, this required
our judgment that recommendations in the two sets of recommendations
were sufficiently similar to be considered the same.  Our combination
of the ESI recommendations and the resulting listing of 38
recommendations are provided in appendix II. 

We interviewed agency program and budget officials and reviewed
agency planning and budget documents that they provided to accomplish
our other objectives to describe the agencies' plans, funding,\5
actions to implement other improvements the agencies made to economic
statistics, and fiscal year 1995 budget requests for additional
economic statistical improvements.  We did not assess whether the
agencies' plans fully addressed all the elements of the ESI
recommendations, nor did we independently verify the budget data or
data provided by the agencies on the specific aspects of their
actions.  Because several agencies shared responsibility for
implementing a number of recommendations (see app.  II), our
reporting that an agency had completed its planned work in response
to an ESI recommendation does not necessarily mean that the
recommendation is fully implemented.  Also, the recommendations are
not necessarily equal in terms of complexity and importance.  Thus,
an exact count of how many recommendations are completed or pending
would not be meaningful and is not presented.  We also interviewed
senior policy officials at Census, BEA, BLS, the Federal Reserve,
OMB, NSF, NASS, and CEA to obtain their views on ESI and other
efforts to improve federal economic statistics. 

We did our work in Washington, D.C., between July 1993 and February
1995 in accordance with generally accepted government auditing
standards.  We obtained comments on a draft of this report from the
Department of Commerce, the Federal Reserve, OMB, CEA, BLS, NASS, and
NSF.  Written comments by BLS and NASS are discussed in the letter
and presented in greater detail in appendixes XII and XIII.  Comments
by Commerce, OMB, CEA, the Federal Reserve, and NSF are discussed in
the letter. 


--------------------
\5 We obtained budget documents to determine additional resources the
agencies requested and received to implement the ESI recommendations
for fiscal years 1990 through 1994.  Because our objective was to
determine and report new funding for economic statistics programs, we
did not report any reprogrammed funding--resources that were shifted
within an appropriated fund account for a purpose different than that
contemplated at the time of appropriation.  Several agencies'
officials said they reprogrammed existing resources to fund ESI
efforts. 


   IDENTIFICATION AND CONTEXT OF
   ESI
------------------------------------------------------------ Letter :4

We identified 38 recommendations in the 1990 and 1991 ESI
announcements and categorized them into 7 areas:  the national income
and product accounts, service sector, price measurement, labor
market, poverty and income, international transactions, and
systemwide statistics recommendations.  The working group that
developed these recommendations included major federal producers and
users of federal economic statistics and sought to address well-known
problems in statistics that affected public policy.  The most serious
problem noted by the working group was in measuring output and
quality improvements to goods and services in a rapidly changing
economy.  At the same time, the working group sought to focus the
recommendations on areas that it considered feasible for short-term
implementation.  Because of budget and time constraints, the
recommendations were not intended to be comprehensive.  (See app.  I
for a description of the overall context in which the ESI
recommendations were developed.  Apps.  III through IX also discuss
context for each of the 38 ESI recommendations; the appendixes are
based on our categories of the 38 recommendations.)


   AGENCIES MADE PLANS AND
   REQUESTED AND RECEIVED AT LEAST
   PARTIAL FUNDING FOR MOST ESI
   RECOMMENDATIONS
------------------------------------------------------------ Letter :5

The federal agencies responsible for implementing the ESI
recommendations made plans to implement most of them.  The agencies
requested more than $95 million and received about $50 million,
slightly more than one-half of the funding they requested, to carry
out their plans.  For example, Census officials said that Census
planned to increase coverage of the service sector by beginning a new
annual survey of communication services; Census requested about $7
million and received more than $1 million for its plan to address
this recommendation.  For a few recommendations, the agencies did not
make plans because the agencies said they were already carrying out
efforts they believed satisfied the intent of the recommendations. 
For example, BLS officials told us that BLS did not make a plan
specifically intended to respond to an ESI recommendation--to
reconcile and reduce discrepancies in estimates of employment from
two different surveys--because this effort was integral to its
overall efforts to improve the two surveys.  BLS received $35 million
(71 percent of the funds appropriated for ESI) for one systemwide,
two-price measurement, and four labor market statistics
recommendations.\6 (Apps.  III through IX describe the agencies'
plans to carry out each recommendation and the funding requested and
received to carry out each recommendation from fiscal year 1990 to
1994.)


--------------------
\6 According to BLS officials, BLS would have needed more than $57
million for fiscal years 1990 through 1994 to fully implement the
recommendations as originally envisioned. 


   AGENCIES' PLANS TO ADDRESS ESI
   RECOMMENDATIONS WERE STILL IN
   PROGRESS
------------------------------------------------------------ Letter :6

As of May 1994, the agencies' plans were in various phases of
completion.  The agencies' officials said that the majority of the
agencies' plans for implementing the ESI recommendations were still
in progress as of that date.  For example, BLS officials said that
partial funding limited actions on BLS' plan to improve coverage of
service industries in price indexes.  Agency officials said that a
few actions they planned for the ESI recommendations were completed
by May 1994.  For example, Census officials said that Census
completed its plan to undertake an annual investment survey. 
Agencies also reported no progress on a few of the plans.  Officials
of these agencies generally cited budgetary and other resource
constraints as the reason for not taking or completing action.  For
example, Census officials told us that since they did not receive the
$5 million they requested, they were not able to begin Census' plan
to expand corporate financial data by collecting quarterly financial
data for the service sector and by expanding coverage of small
corporations.  There were also instances where an agency completed
its plan without receiving all the funding it requested.  For
example, Census officials said that they were able to complete the
substance of Census' plans to increase coverage of the service
sector, although it received $1 million of the $7 million it
requested.  (Apps.  III through IX describe the progress made on the
plans for each recommendation as of May 1994.)


   AGENCIES MADE OTHER
   IMPROVEMENTS TO ECONOMIC
   STATISTICS
------------------------------------------------------------ Letter :7

While ESI enabled agencies to receive additional funding for specific
improvements in economic statistics, agency officials reported that,
as part of their regular ongoing efforts to adapt to changes in the
economy and in source data, other improvements were made with
existing funding.  These improvements were made in addition to the
ESI improvements from fiscal years 1990 through 1994.  For example,
officials at Census said that Census improved the quality of economic
statistics by expanding existing surveys, creating new surveys,
expanding data collection, increasing coordination efforts, and
developing computer processing projects.  (App.  X describes other
major improvements in federal statistics that agencies reported.)


   AGENCIES REQUESTED FUNDING FOR
   FURTHER IMPROVEMENTS IN FISCAL
   YEAR 1995
------------------------------------------------------------ Letter :8

According to agency officials, because ESI was never intended to
address all the problems in economic statistics and because many of
the plans for the ESI recommendations were not completed, further
efforts were needed to improve the quality of economic statistics. 
The agencies requested an additional $38 million for such
improvements in fiscal year 1995.  As of December 31, 1994, $18
million in appropriations had been approved for the fiscal year. 
(App.  XI provides information on funding requested and received for
planned fiscal year 1995 economic statistical improvements.)

Senior policy officials expressed their views of what is still needed
to address problems in economic statistics.  OMB officials said that
the administration is committed to improving economic statistics and
that the improvements proposed by BEA, Census, and BLS will result in
overall improvements to their statistics.  According to a BEA
official, although there are further methodological improvements BEA
can make, adequate funding will be required to fill gaps in source
data.  BLS officials said that additional resources are the number
one requirement to improve BLS' statistics.  According to the
Commissioner, BLS needs to improve service sector data, provide more
information on work practices, as well as be more user friendly in
data collection and dissemination in the future.  Census officials
noted that the future of economic statistics is also dependent on
effective leadership, not only to oversee current operations, but
also to consider how the statistical system may need to change. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :9

The Department of Commerce, the Federal Reserve, OMB, CEA, BLS, NASS,
and NSF provided comments on a draft of this report.  Commerce's
written comments incorporated comments from BEA and Census.  All of
Census' and most of BEA's comments were suggestions for technical
clarifications and corrections, and we have incorporated these
suggestions where appropriate.  BEA said that our report was a useful
overview of progress under ESI.  BEA noted that the report reinforces
the conclusion of BEA's recent mid-decade review of its economic
accounts, which was that, despite recent progress, much work remains
toward improving the economic accounts and economic statistics in
general.  In addition, BEA said that it made progress in virtually
all the methodological issues addressed in ESI but was unable to make
progress in areas where improvements were dependent on the
development of new surveys or the modifications of existing surveys
to fill gaps in coverage, if those improvements and modifications
were not funded.  BEA indicated several places in the report where it
believed this point needed to be clarified.  We have made several
revisions to the report to explain that BEA reported it made progress
in improving methodologies but was unable to develop or revise
surveys due to lack of funding. 

BLS, in its written comments, said that policymakers, other data
users, and the statistical agencies themselves have identified many
areas where improvements are needed in statistical programs.  BLS
also said that ESI had made significant progress in addressing some
of these problem areas, given the resources made available to it. 
BLS said, however, that our draft report appeared to imply that the
BLS initiatives as originally envisioned were nearly fully funded. 
According to BLS, constraints caused them to curtail planned
improvements and thus BLS' subsequent budget requests were
significantly lower than they would have been to fund the planned
improvements.  We have revised the report to explain that BLS
received substantially less funding than it believed necessary to
implement the ESI recommendations fully.  Senior BLS officials also
suggested technical corrections to the report that we have
incorporated where appropriate.  (See app.  XII for comments from
BLS.)

NASS' written comments stated that NASS had benefited from ESI,
particularly from the recommendation that supported NASS' efforts to
improve the coverage of its list of farm and farm business operators. 
NASS noted, however, that although the coverage of the list had
improved, complete coverage of the farm operator universe would not
be achieved until NASS and Census undertook a mutually cooperative
list building effort.  According to NASS, current laws prevented it
from receiving the mailing list used by Census for its Agricultural
Census or the list of farms identified in the Census.  NASS said that
the Department of Agriculture strongly supported a draft legislative
proposal to permit list sharing among federal statistical agencies
for statistical purposes only.  As we noted in the report, the
development of data-sharing legislation was one of the ESI
recommendations.  OMB drafted a data-sharing proposal that was
circulated among federal departments for comments; this proposal had
not been introduced in Congress as of May 19, 1995.  (See app.  XIII
for comments from NASS.)

The Chief Statistician and a senior economist in OMB's Office of
Statstical Policy said they generally agreed with the information as
presented.  However, they said that, in discussing funding for ESI,
the report should have noted those instances where agencies
reprogrammed funds in order to implement an initiative.  We did not
discuss reprogramming in the report because our objective was to
track new funds specifically appropriated for economic statistics
improvements.  We did add a footnote to the letter explaining that
agencies did reprogram funds to implement some initiatives that did
not receive new funding.  The officials also offered suggestions for
several technical changes, which we have incorporated where
appropriate. 

CEA provided oral comments on the draft report.  A senior economist
at CEA said that the report should have noted BEA's strategic review
of the National Income and Product Account (NIPA), known as the
mid-decade review.  This review is intended to evaluate and to
develop a plan to maintain and improve the performance of BEA's
economic accounts.  The official said that this review was a good
example of an agency working with its data users to develop a
framework for continuous efforts aimed at improving economic
statistics.  We acknowledge that this BEA effort is a significant
data improvement initiative.  However, we did not include it in our
discussion of agency initiatives because it was released after we had
completed our review of the ESI recommendations in May 1994. 

Officials at the Federal Reserve and NSF said that they had no
substantive comments on the information presented in the draft
report.  The Federal Reserve's Assistant Director of the Research and
Statistics Division offered suggestions for technical clarifications,
which we have incorporated where appropriate.  The Director of NSF's
Methodology, Measurement and Statistics Program indicated that NSF
agreed with the information presented in the draft report concerning
the Center for Survey Methods. 


---------------------------------------------------------- Letter :9.1

We are sending copies of this report to interested congressional
committees, the Secretary of Commerce as well as the Directors of BEA
and Census, the Secretary of Labor and the Commissioner of BLS, the
Chairman of the Board of Governors of the Federal Reserve System, the
Acting Chair of CEA, the Director of OMB, the Chairperson of NSF, the
Secretary of Agriculture, the Director of NASS, and other interested
parties.  Copies will also be made available to others on request. 

Major contributors to this report are listed in appendix XIV.  If you
have any questions about this report, please contact me on (202)
512-7824. 

Sincerely yours,

L.  Nye Stevens
Director, Federal Management
 and Workforce Issues


BACKGROUND ON THE ECONOMICS
STATISTICS INITIATIVE
=========================================================== Appendix I

In recognition of the long-standing problems with federal economic
statistics, the Bush administration in 1989 began a multiyear effort
to improve the nation's most important economic statistics.  A
working group, chaired by the Chairperson of the Council of Economic
Advisers (CEA), released two sets of recommendations in 1990 and 1991
that were aimed at improving the statistics that had the greatest
importance to policymaking as well as the greatest problems.  Even
though the efforts consisted of two sets of recommendations and
spanned multiple years, these efforts are referred to collectively as
the Economics Statistics Initiative (ESI). 


   EXPERTS NOTED PROBLEMS WITH
   FEDERAL ECONOMIC STATISTICS
--------------------------------------------------------- Appendix I:1

Many public and private sector experts have noted problems with
federal economic statistics.  Problems have been attributed to
underfunding, changes in program philosophy, or poor management that
occurred during the 1980s.  Janet Norwood, a senior fellow of the
Urban Institute and the former Commissioner of the Department of
Labor's Bureau of Labor Statistics (BLS), for example, observed that
budget cutbacks in federal statistics were tantamount to mortgaging
the future. 

The former Executive Director of the Council of Professional
Associations on Federal Statistics, now the Chief of the Office of
Management and Budget's (OMB) Office of Statistical Policy, said that
a fundamental philosophical shift that emphasized reducing business'
reporting burdens and the extent of data collected occurred along
with reduced government spending on data collection during most of
the 1980s.  The budget cuts in the 1980s had the effect of curtailing
government efforts to improve economic statistics when their role and
importance were greater than ever. 

In a frequently cited assessment of the federal statistical system,
the Office of Technology Assessment (OTA)\1 concluded that in 1988
the decentralized federal statistical system lacked a coherent
strategy and leadership from OMB.\2 As a result of these and other
problems, key questions about the U.S.  economy, such as how fast it
grew or how its growth affected incomes and their distribution, were
very difficult to answer. 

In the private sector, organizations representing individuals,
businesses, and other organizations that rely upon federal statistics
also have noted problems with economic statistics.  The American
Economic Association (AEA), the National Bureau of Economic Research
(NBER), and the National Association of Business Economists (NABE)
have sponsored studies that concluded that the federal statistical
system critically needed improvement.  The Chairperson of AEA's
Committee on the Quality of Economic Statistics wrote in 1988 that
federal statistics were not adequate to inform decisionmakers for
understanding and choosing among policies and, without improvement,
would become increasingly inadequate to meet future policy needs.\3
He cited examples of problems such as decisionmakers' inability to
understand factors affecting national productivity, outdated U.S. 
international statistics that may fail to reflect global economics,
and serious deficiencies in savings and investment data.  Similarly,
NBER's 1989 Conference on Research on Income and Wealth focused on
problems with economic statistics, particularly problems with
measuring international transactions.\4 Noting a decline in federal
budget support for data collection and maintenance and a growing
concern among professional economists about economic statistics, NBER
authors concluded that international economic statistics were in need
of critical assessment and, in some cases, significant overhaul. 
Finally, NABE's special committee on statistics raised similar
concerns.\5 The committee reported in 1988 and testified in
congressional hearings in 1990 that the government was investing
insufficient resources to improve and maintain the quality of
economic statistics to enable policymakers to develop appropriate
public policy based on accurate and timely economic statistics and
for businesses to plan adequately for the future. 


--------------------
\1 Statistical Needs for a Changing U.S.  Economy, Background Paper,
U.S.  Office of Technology Assessment (Sept.  1989). 

\2 The federal statistical system is a designation of government
agencies that collect, process, analyze, and use quantitative data. 
More than 70 agencies are classified by OMB as conducting statistical
activities because they have annual budgets of at least $500,000 for
such activities. 

\3 The State of U.S.  Economic Statistics:  Current and Prospective
Quality, Policy Needs, and Resources.  F.  Thomas Juster,
Chairperson, AEA Committee on the Quality of Economic Statistics. 
Paper prepared for the May 1988 Conference on Research on Income and
Wealth, pp.  2, 3-5. 

\4 International Economic Transactions:  Issues in Measurement and
Empirical Research, proceedings from the November 1989 Conference on
Research on Income and Wealth.  Edited by Peter Hooper and J.  David
Richardson, University of Chicago Press, 1989, p.  1. 

\5 Improving the Quality of Economic Statistics, statement by James
F.  Smith, President of the National Association of Business
Economics, before the U.S.  Congress, Joint Economic Committee, Mar. 
29, 1990. 


   BUSH ADMINISTRATION ASSESSED
   PROBLEMS IN FEDERAL ECONOMIC
   STATISTICS
--------------------------------------------------------- Appendix I:2

Responding to these concerns, President Bush, as a supplement to his
first State of the Union message in February 1989, proposed an
increase of $16 million in federal spending to improve the quality
and accuracy of business and economic statistics.  He specifically
noted the need to improve service sector and foreign trade
statistics. 

To that end, a working group was established in April 1989 to assess
problems within the federal economic statistical system and recommend
needed improvements.  This group, chaired by the CEA Chairperson
Michael J.  Boskin, included representatives of the major federal
government producers and users of federal economic statistics.\6

To accomplish its mission, the working group surveyed statistical
agencies' existing plans and priorities; gathered suggestions for
further improvements from the agencies and community of users in the
administration, Congress, and private sector; and developed
recommendations for improvements.  These recommendations concentrated
in three agencies--BEA, BLS, and Census.\7 The working group
acknowledged that due to budget and time constraints, the multiyear
recommendations were interim and not comprehensive.  In addition, it
noted that, while the recommendations addressed well-known problems
in areas that affected public policy, it kept its recommended
improvements to those that were most cost-effective and could be
executed in the short term.  For example, the recommendations did not
address organizational issues or the long-run needs of the system. 

On January 25, 1990, CEA publicly released the working group's first
set of recommendations.  These recommendations were discussed in
congressional hearings later that year.  The recommendations were
organized into three areas of economic statistics:  (1) productivity,
output, and prices; (2) investment, saving, and national wealth; and
(3) employment, income, and poverty. 

In February 1991, CEA released the working group's second set of
recommendations and announced that the Bush administration would seek
an increase of $230 million from fiscal years 1992 through 1996. 
This figure represented the amount of additional funding the agencies
believed they would need to implement the recommendations.  Also, in
this announcement the working group officially named its efforts
ESI.\8 As time passed, and budget constraints had an impact on this
proposal, this amount decreased as parts of the original ESI proposal
were never implemented due to lack of funding. 

The second set of recommendations was organized by the working group
into seven, rather than three, areas.  These seven areas are (1)
improving the national and international economic accounts, (2)
increasing the coverage of the service sector, (3) separating quality
and inflation changes in price data, (4) improving establishment
payroll and household surveys, (5) tracking changes across
industries, (6) preparing for future statistical workforce needs, and
(7) sharing of statistical data.\9


--------------------
\6 The members of the working group included officials from the
following agencies:  the U.S.  Trade Representative, OMB, the
President's Office of Policy Development, the U.S.  Department of the
Treasury; the U.S.  Department of Commerce, including its Bureau of
Economic Analysis (BEA) and the Bureau of the Census; the U.S. 
Department of Labor, BLS; the Board of Governors of the Federal
Reserve System (Federal Reserve); the Department of Housing and Urban
Development; the Department of Agriculture; and CEA. 

\7 Other responsible agencies were the Federal Reserve, OMB, the
National Science Foundation (NSF), and the National Agricultural
Statistics Service (NASS), within the U.S.  Department of
Agriculture. 

\8 The initiative is also referred to as the "Boskin Initiative."

\9 These 7 areas are different from the 7 areas we used to categorize
the 38 ESI recommendations. 


ECONOMICS STATISTICS INITIATIVE
RECOMMENDATIONS
========================================================== Appendix II

   Dollars in thousands

   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

   Note:  GAO's 7 areas and the 38
   ESI recommendations are our
   categorizations of the
   Economics Statistics Initiative
   recommendations that were in
   the 1990 and 1991 CEA releases.

   (See figure in printed
   edition.)

\a The 1990 recommendations are quoted directly from a prepared
statement for Michael J.  Boskin, Chairperson of the Council of
Economic Advisers at a hearing before the Joint Economic Committee,
U.S.  Congress, March 1, 1990. 

\b The 1991 recommendations are paraphrased by GAO from a February
14, 1991, Council of Economic Advisers announcement. 

\c Information is recorded for each responsible agency for
recommendations with multiple responsible agencies. 

\d BEA combined these three recommendations into one budget request
to increase funding to stop the deterioration in the quality of the
national economic accounts. 

\e Federal Reserve funds are not included here since it does not
receive appropriated funds from Congress. 

\f BEA combined these two recommendations into one budget request to
improve balance of payments and international investment data. 

\g Census combined these four recommendations into one budget request
to improve foreign trade statistics. 

\h Total request includes $14,047,000 in reinstated requests, which
means an agency requested funding more than once when the funding was
not received in prior years. 


THE NATIONAL INCOME AND PRODUCT
ACCOUNTS STATISTICS
RECOMMENDATIONS
========================================================= Appendix III

The National Income and Product Accounts (NIPA) are a detailed
description of the overall U.S.  economy.  NIPA depicts in dollar
terms the volume, composition, and use of the nation's output of
goods and services.  These core national economic statistical
measures are used to analyze past and current economic performance
and also to forecast economic developments. 

In recommending areas for improvement, the working group gave
priority to statistical measures they viewed as having the greatest
policy importance and recommended reallocating and focusing resources
on core national economic statistical measures, which are used in
determining fiscal and monetary policies.  NIPA statistics
recommendations were, in most part, designed to address weaknesses in
the core statistical measures used to quantify the economic position
of the United States.  Emphasis was also placed on modernizing NIPA
to follow the System of National Accounts (SNA), a set of
international guidelines for economic accounts.  SNA could serve as a
framework for modernizing the U.S.  accounts, provide information in
a more integrated format than policymakers have now, and make the
U.S.  accounts more comparable with the accounts of other countries. 


   INDIRECT ESTIMATION METHODS
------------------------------------------------------- Appendix III:1

Indirect estimation methods are procedures used to construct a
statistical measure from data that are not direct observations of the
activity being measured, or when data needed to implement the
conceptually correct, or preferred, measure are not available.\1


--------------------
\1 For example, the preferred method for adjusting an industry's
gross output, or value-added, for inflation is the double deflation
method.  With double deflation, an industry's gross output and its
intermediate inputs are deflated separately and constant-dollar gross
product is derived as the difference between deflated gross output
and deflated intermediate inputs.  However, industry specific data
are sometimes not available to estimate real gross product by double
deflation, and output must be estimated indirectly using physical
indicators.  See the glossary for definitions and descriptions of
technical terms and phrases that are used throughout this appendix. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:1.1

To construct NIPA statistical measures, the U.S.  Department of
Commerce's Bureau of Economic Analysis (BEA) often relies upon data
that have been collected by other government agencies for other
purposes.  As a result of data limitations, BEA must sometimes use
indirect methods of estimation.  For example, in some industries,
output (the statistical measure of what an industry produces) is
estimated by measures of labor inputs (total hours used by an
industry to produce its goods and services).  In other industries,
especially those with rapid introduction of goods and services,
output is estimated by dividing net sales by a price index.  In the
1990 Economic Report of the President, CEA said that price indexes
that appropriately adjust for quality changes can be quite important
in measuring output.  However, CEA noted the difficulty in separating
pure price increases from those arising from improvements in product
quality or service for industries with rapid rates of innovation.\2

According to BEA, appropriate methods to adjust output for inflation
have not been developed for all industries.\3 BEA reports that
long-term research on the definition of output of several industries,
most notably financial services, is required to improve estimation
methods. 


--------------------
\2 Council of Economic Advisers, Economic Report of the President
(Washington, D.C.:  U.S.  Government Printing Office, 1990) pp. 
281-285. 

\3 Frank de Leeuw, Michael Mohr, and Robert P.  Parker, "Gross
Product by Industry, 1977-88:  A Progress Report on Improving the
Estimates," Survey of Current Business, Vol.  71 (January 1991), pp. 
23-37. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:1.2

In 1990, the working group recommended the exploration of alternative
methods for estimating constant-dollar output.  In 1991, the working
group specifically recommended that BEA use indirect estimation
methods to close data gaps.  These methods were to address price
measurement of high tech goods and certain services such as banking
and improve deflation of purchases by state and local governments.\4


--------------------
\4 According to a BEA official who was at CEA at that time, the
working group was also concerned with measures of constant-dollar
output in other specific industries where double deflation, the
preferred measure of output, was not used.  It was in the context of
these concerns and budgetary restrictions that the working group
recommended methodological corrections using indirect estimation
methods to fix these and other data problems and gaps in NIPA. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:1.3

BEA planned to develop improved measures of constant-dollar, or real,
output--especially of service industries.  Current procedures would
be either replaced with direct estimates of output via double
deflation or modified to incorporate allowances for productivity
change developed from indirect evidence.  BEA also planned to develop
improved measures of constant- and current-dollar estimates for
various components of gross domestic product (GDP),\5 including high
tech goods, banking services, state and local government purchases,
services, and the nonprofit sector. 


--------------------
\5 In December 1991, the Department of Commerce began to use GDP as
the primary measure of economic performance.  While gross national
product (GNP) measures output of U.S.  labor and property regardless
of location, GDP measures output of all labor and property located in
the United States.  According to the Department of Commerce, for the
United States, there is little difference between the dollar levels
of GDP and GNP.  Since the change occurred between the two sets of
Economics Statistics Initiative (ESI) recommendations, we use both
GNP and GDP in this report. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:1.4

For all three of the recommendations designed to stop the
deterioration in the quality of NIPA, which included the indirect
estimation methods, input-output tables, and construction-methodology
recommendations, BEA requested more than $3 million and actually
received nearly $2 million total for fiscal years 1990 through
1994.\6


--------------------
\6 BEA officials provided combined budget information for the three
recommendations. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:1.5

BEA officials noted that BEA completed an overhaul of the methodology
used to estimate annual GDP by industry and presented revised
estimates.  The double deflation estimation method was extended from
25 to 51 of the 64 industries in GDP.  They said that extensive
improvements were made in the weighting of detailed prices and in the
selection of suitable price indexes for deflation of inputs and gross
output to constant dollars. 

BEA also introduced several alternative price indexes and quantity
(constant-dollar) indexes that can be used to determine quarterly
constant-dollar estimates of GDP.  These indexes also allow the
construction of several alternative measures of constant-dollar
output, which allow for changes over time in relative prices, such as
those developed for computers.  The alternative measures provide an
improved basis for analyzing long-term growth and business cycles. 

BEA officials noted progress in several other areas using existing
data and alternative or new estimation methods.  BEA reworked and
improved measures of consumer expenditures on several services and on
output of service industries.  Improved estimates of consumer
expenditures on child care, brokerage commissions, and motor vehicle
leasing were incorporated in the comprehensive 1991 GNP revision. 

BEA conducted further research on developing an improved definition
of banking services that was presented to professional groups for
review and discussion.\7 Several improvements were introduced in the
price indexes for computers, including a separate index for imports. 
BEA initiated research on prices for medical imaging technology,
dynamic access memory chips, semiconductors, and aircraft.  Work was
begun on a nonprofit-institution sector account that produced
research papers.  The form and direction of BEA's research work was
discussed in the early 1990s at a joint BEA-National Academy of
Sciences conference. 


--------------------
\7 This was an effort for which funding was requested under the data
gaps recommendation. 


   INPUT-OUTPUT TABLES
------------------------------------------------------- Appendix III:2

Input-output tables are the basic source of information about the
interrelationships between industries.  These tables show what each
of 537 industries purchases from other industries.  The tables trace
the flow of goods and services among industries in the production
process and show the value added by each industry and the detailed
commodity composition of national output. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:2.1

According to the National Association of Business Economists' (NABE)
Statistics Committee,\8 input-output tables are very important in
determining productivity and in measuring the interrelationships
between various industries in the United States.  Requiring large
amounts of data for numerous industries, these tables are very
difficult and time-consuming to calculate and produce.  At the time
of the NABE report in March 1991, the 1977 input-output tables were
published 7 years after the economic census year; 1982 input-output
tables were published 9 years after the census year.  Since the
economic censuses are conducted every 5 years, this means that the
data in benchmark tables could be as much as 12 to 15 years old by
the time the new benchmark tables were issued. 


--------------------
\8 Improving Federal Statistics:  Hearings Before the Joint Economic
Committee of Congress of the United States.  102nd Cong., 1st Sess.,
pp.  67-127 (1991) (statement by Martin Fleming, Chairperson of the
National Association of Business Economists' Statistics Committee). 
The statement reflects the views of NABE's Statistics Committee. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:2.2

In 1990, the working group recommended the expedited compilation of
input-output data.  In 1991, the working group specifically
recommended a 2-year reduction in the time to produce the benchmark
and annual input-output tables. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:2.3

BEA set out a long-term goal to produce its benchmark input-output
tables within 5 years of the economic census, rather than the 7 to 9
years it had previously taken, and reduce from 5 to 3 years the time
to produce its less detailed annual input-output tables.  For
example, to produce its tables in these shorter time frames and
reduce the size of annual GNP revisions, BEA devised a set of
procedures that would capture the most important parts of the 1987
Economic Census data.  These procedures would shorten the normal
time-consuming process of assembling a wide variety of other data for
developing components not based on economic census data.  To shorten
the preparation time for the 1987 input-output tables, BEA planned to
distribute the total economic census construction output among 5
industries rather than analyzing and separately estimating
approximately 50 construction industries. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:2.4

For all three of the recommendations designed to stop the
deterioration in the quality of NIPA, which included the input-output
tables, indirect estimation methods, and the construction-methodology
recommendations, BEA requested more than $3 million and actually
received nearly $2 million total for fiscal years 1990 through
1994.\9


--------------------
\9 BEA officials provided combined budget information for the three
recommendations. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:2.5

BEA officials said that the 1987 benchmark input-output tables were
released in April 1994, and that they have begun planning for work on
the 1992 benchmark input-output tables.  BEA hopes to achieve the
long-term goal of making the input-output tables available within 5
years of the economic census year and within 1 year of the release of
all economic census data.  The 1987 input-output tables were
available within 7 years of the economic census year, 2 years earlier
than the release of the 1982 input-output tables.  Officials said
that BEA has developed abbreviated procedures to produce the 1992
input-output tables.  With the assistance of Census to expedite the
availability of the 1992 Economic Census data, BEA hopes to have the
1992 input-output tables available within 5 years of the economic
census year. 


   CONSTRUCTION - METHODOLOGY
------------------------------------------------------- Appendix III:3

Construction methodology refers to the procedures used to measure the
economic activity on construction of residential and nonresidential
structures. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:3.1

According to the NABE Statistics Committee, serious problems in the
measurement of construction have resulted in data that are
inconsistent from one survey to another.  NABE also reported that
special studies of the industry by Census indicated substantial
differences in construction spending from those reported in NIPA. 
These data are used to measure productivity in the construction
industry. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:3.2

In 1990, the working group recommended the completion of ongoing
methodological and data collection improvements; these were to be
incorporated in the upcoming GNP revision. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:3.3

BEA planned a major effort to eliminate shortfalls in nonresidential
construction estimates by measuring nonresidential construction
prices directly with a statistical survey conducted by Census.  The
initial plan called for research to establish the most feasible
method to obtain data on construction prices and to prepare a price
index suitable for the deflation of current-dollar construction
expenditures into constant-dollar estimates for GDP.  In addition to
improvements in the constant- dollar estimates, BEA planned to
improve the current-dollar estimates by filling a gap in measuring
the value of additions, alterations, and repairs in the private
nonresidential and state and local government construction estimates
as part of BEA's 1991 GNP revision.  These current-dollar
improvements for construction expenditures were to begin
retroactively with 1973 estimates and were to be incorporated into
the input-output tables. 

BEA also planned to provide a basis for improving its estimates of
U.S.  construction prices by exploring alternative methodologies. 
One method was to utilize Census survey data on multifamily housing
construction prices.  Another method was to combine information on
the U.S.  economy with a nonresidential price index prepared by
Statistics Canada.\10 These efforts were to provide improved price
indexes for making constant-dollar estimates for use in the 1991 GNP
revision. 


--------------------
\10 Statistics Canada is the single government agency that collects,
analyzes, and publishes Canadian statistical information on
population, commerce, and a broad range of other activities. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:3.4

For all three of the recommendations designed to stop the
deterioration in the quality of NIPA, which included the
construction-methodology, indirect estimation methods, and
input-output tables recommendations, BEA requested more than $3
million and actually received nearly $2 million total for fiscal
years 1990 through 1994.\11


--------------------
\11 BEA officials provided combined budget information for the three
recommendations. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:3.5

According to BEA officials, funding was not sufficient to support the
most important construction improvements:  the research, design, and
implementation of a new direct survey of construction prices by
Census.  Working within the funding provided, BEA was able, mainly
through methodological innovations and "fixes" that used existing
data, to develop a new procedure for benchmarking construction
expenditures that was incorporated as planned in the 1991 GNP
revision.  BEA officials noted that the new procedure improved the
estimates of nonresidential construction and raised them by nearly 25
percent.  In addition, BEA officials noted that a new index of
multifamily housing construction prices was developed and implemented
in the 1991 GNP revision.  This index accounts for quality changes by
adjusting for the characteristics of housing units.  BEA officials
said, however, that limited funding curtailed the methodological
efforts in estimating U.S.  construction prices, and the 1991 GNP
revision target was not met.  BEA and Census studied the use of
Canadian data on nonresidential prices in preparing a price index for
the U.S.  economy.  BEA also initiated work with Census to develop a
direct survey as the basis for a nonresidential construction price
index, but the project was never completed because BEA and Census
lacked the funding to conduct the direct survey. 


   SYSTEM OF NATIONAL ACCOUNTS
------------------------------------------------------- Appendix III:4

SNA is the international standard for the compilation of a nation's
economic accounts and statistical measures.  The revised SNA adopted
by the United Nations in 1993 followed a decade-long revision that
updated the 1968 SNA, clarified the standard, and harmonized it more
completely with other sets of international guidelines.  The revised
SNA envisions a comprehensive recording of the stocks and the flows
in the U.S.  economy.  For each sector that makes up the economy
(business, government, and households), SNA calls for an opening
balance sheet and a closing balance sheet--the stocks of assets and
liabilities of the economy.  And for each sector, current and capital
accounts are designed to record all the transactions and other
changes that explain the differences between the balance sheets--the
flows of the economy that measure national income and production. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:4.1

According to BEA officials, the SNA revision provided an opportunity
for the United States to modernize and extend its national accounts
and to meet the needs for domestic and cross-country analysis and
policy formulation.  The adoption of SNA would also allow the United
States to take advantage of the substantial body of expertise vested
in the revised SNA, which is based on a worldwide review of 20 years
of progress in economic accounting by more than 65 national accounts
experts.  Such a move to an internationally oriented economic
accounting system would allow comparisons that cannot be made with
the system presently used in the United States.  For example, if the
United States used SNA, the U.S.  savings rates, inflation, and
economic growth could be compared with those of other countries;
government capital expenditures could be broken out to provide
information for fiscal policymaking; and satellite accounts in areas
such as health, education, and the environment, would supplement the
traditional NIPA. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:4.2

The working group viewed the measurement of national investment and
saving in NIPA as conceptually flawed, inaccurate, and
internationally inconsistent.  In making the recommendation in 1990
to revise NIPA to be consistent with the major components of SNA, the
working group said that the adoption of SNA would provide a better
analytical base for budget, trade, tax, and regulatory policies. 
Modernizing and extending NIPA to follow SNA was again recommended by
the working group in 1991 with two specific features:  (1) an
integrated set of current and capital accounts that include both
financial and nonfinancial transactions and (2) satellite accounts. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:4.3

To adopt SNA's standard in the United States by the mid-1990s, BEA
developed a multistage plan to

  evaluate SNA's standard and develop general strategies for its
     adoption;

  familiarize the staff and users of the economic accounts with SNA;

  develop satellite accounts and make them available as they are
     developed;

  develop provisional integrated current and capital accounts,
     including balance sheets for each sector and the nation; and

  carry out the move to SNA in the mid-1990s through a comprehensive
     revision of the U.S.  accounts. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:4.4

BEA requested nearly $6 million and actually received slightly more
than $1 million total for fiscal years 1990 through 1994 to implement
the SNA recommendation. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:4.5

BEA officials noted that this recommendation received significantly
less than full funding.  They said that without such funding, the
adoption of the SNA standard by the mid-1990s would be impossible. 
BEA officials noted that with the funding provided, BEA modified its
plan toward an incremental approach by developing provisional
integrated current and capital accounts one sector at a time and
reducing the number of satellite accounts under preparation. 
However, BEA made progress toward revising its NIPA to be consistent
with SNA. 

  A special SNA staff was assembled and evaluation was begun on how
     to apply SNA within the context of the U.S.  economy and
     statistical system. 

  Training of BEA staff in the national economic accounting and in
     the specifics of SNA was started.  BEA began consultations with
     users, such as NABE, and data suppliers, such as Census and the
     Energy Information Administration. 

  BEA published its satellite accounting framework for natural
     resources and prototype estimates for mineral resources in the
     April 1994 issue of Survey of Current Business. 

  BEA changed its treatment of several types of transactions in the
     comprehensive NIPA revision released at the end of 1991 that
     made the NIPA more consistent with SNA.  These changes included
     the featuring of GDP instead of GNP, the classification of
     receipts for certain services as government sales instead of
     personal nontaxes, the classification of bad debt losses as
     financial transactions, and the recognition of capital gains
     distributions of regulated investment companies as dividends. 

  BEA published its research and development satellite account in the
     November 1994 issue of Survey of Current Business. 

  The development of a government sector account has proceeded.  The
     Committee on National Statistics held a workshop in November
     1993, sponsored by BEA and Census, which discussed the issues
     involved in modernizing NIPA government sector account and
     developing an SNA-based government sector account.  In addition,
     BEA and the Federal Reserve presented a prototype government
     sector account at the International Association for Research in
     Income and Wealth Conference in August 1994. 


   INFLATION ADJUSTMENTS
------------------------------------------------------- Appendix III:5

Inflation adjustments are procedures that are used to capture the
pure increase in prices apart from increases due to changes in
quality and quantity.  BEA officials said that this inflation
adjustments recommendation specifically referred to the removal of
nominal capital gains from income originating from property. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:5.1

BEA officials said that BEA makes adjustment to profits by excluding
from operating profits the portion of companies' reported profits
associated with holding gains on inventories and the understatement
of depreciation costs by historical cost accounting.  According to a
BEA official who was at CEA at the time of the development of ESI,
CEA Chairperson Michael Boskin essentially wanted BEA to develop a
procedure to separate the inflation component from the real component
for other types of property income, such as interest, rent, and
dividends.  Further, the former CEA official said academic
researchers and tax analysts had attempted to make such adjustments,
using various estimates of anticipated inflation, with little
agreement or success. 

According to the NABE Statistics Committee, NIPA treats physical and
financial assets differently in determining their value over time. 
While various sources of income are adjusted for various accounting
abnormalities, methodologies have not been developed for some income
sources, such as rental and interest income. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:5.2

The working group said in 1990 that NIPA should report real, as well
as nominal, incomes.  Noting that current statistical measures of
interest, rents, and saving in both the government and private
sectors did not adjust appropriately for inflation, the working group
recommended the addition of a supplementary series to NIPA that would
separate the real and inflation components of the return to capital. 
Such an adjustment would make the income calculations comparable with
the corporate profits series in NIPA. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:5.3

BEA officials said that there had not been a great deal of success in
this area by other researchers.  In addition, they said that BEA's
efforts to move toward SNA should result in improvements in inflation
adjustments.  As a result, they did not feel BEA should devote its
resources to this recommendation, except for those inflation
adjustments that had already been made for profits. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:5.4

No funds were requested for this recommendation. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:5.5

BEA officials said no improvements were made. 


   BUSINESS ESTABLISHMENT DATA -
   CENSUS AND BEA
------------------------------------------------------- Appendix III:6

Statistical agencies survey business establishments to obtain data
about their business practices and results.  The agencies use
business establishment lists, which are directories of business
establishments, to select business establishments to be surveyed. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:6.1

Several statistical agencies compile lists of businesses for specific
uses.  Since these lists originate for different purposes, these
lists may capture different private entities, or classify the same
entities differently.  For instance, a group classified as a
"business" by one agency may be classified as a group of
self-employed individuals by another statistical agency.  Coverage of
the business population can be more comprehensive and consistent if
the various statistical agencies share their business lists and the
data from the establishments on the lists.  A law that authorizes
such data sharing is the Foreign Direct Investment and International
Financial Data Improvements Act of 1990 (P.L.  101-533), which
permits sharing between BEA and Census and between BEA and BLS of
data needed for improving statistics on foreign direct investment. 
However, confidentiality restrictions limit how certain business
lists and data can be shared among agencies.  One example of a
confidentiality restriction involves data provided by the Internal
Revenue Service (IRS).  Although Census and BEA may each request tax
return information from IRS, the Internal Revenue Code prohibits
these agencies from then sharing with each other the data they
receive if it identifies a particular taxpayer. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:6.2

In 1990, the working group recognized that concerns about
confidentiality prevented statistical agencies from sharing business
establishment lists.  Nonetheless, it noted that BLS was moving to
make its list available to other agencies.  As an interim measure,
the working group said that Census should explore ways to share its
establishment data with BEA for use in improving NIPA. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:6.3

According to BEA officials, neither BEA nor Census made plans for a
broad sharing of data under this recommendation.  BEA officials said
they and Census, however, made plans to share Census' establishment
data and BEA's enterprise data on foreign-owned firms; and Census
planned to provide BEA with establishment lists of firms engaged in
international services transactions.  In addition, BEA and BLS made
plans to link BEA enterprise data to BLS' establishment data from its
business list.  The sharing of these data was authorized by the
Foreign Direct Investment and International Financial Data
Improvements Act. 

Census officials said that, in addition to linking Census and BEA
data, Census planned to improve its coding of new business
establishments.  Census planned to obtain standard industrial
classification codes from the Social Security Administration (SSA)
for new establishments entering SSA's data system. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:6.4

Census requested $400,000 total for fiscal years 1990 through 1994
for this recommendation, but received no funding.\12


--------------------
\12 According to BEA officials, BEA (and Census through reimbursable
agreements) received ESI funding to share Census' establishment data
and BEA's enterprise data on foreign-owned firms.  Census under
another reimbursable agreement also provided BEA with establishment
lists of firms engaged in international service transactions.  These
projects were authorized by the Foreign Direct Investment and
International Financial Data Improvements Act of 1990.  According to
BLS officials, BLS received funds for its foreign direct investment
program beginning in fiscal year 1991; these funds were not ESI
funds.  See also the business establishment lists - Census and BEA
recommendation under the international investment and capital flows
recommendation (app.  VIII). 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:6.5

BEA officials said that BEA, BLS, and Census successfully completed
the "link" projects covering 1987 data for Census and 1989 and 1990
data for BLS with the results being published in the summer of 1992. 
BLS subsequently linked this establishment-level file to
establishments in its occupational employment statistics survey to
develop estimates of employment in foreign-owned U.S. 
establishments.  These results were published in October 1993. 
Results from linking BEA's 1991 enterprise file to BLS' establishment
file were published in 1994.  Building on experience gained in the
initial project, BEA and Census then linked their respective
enterprise and establishment data for the years 1988 through 1991. 
The results, published in 1993 and 1994, presented a greatly expanded
level of detail for manufacturing industries.  The data link projects
were to provide more detailed and precise information about the
industrial activities of U.S.  affiliates than was previously
available.  Census officials said the lack of funding prevented them
from obtaining codes from SSA.\13


--------------------
\13 Commerce officials said they broadened the scope of the working
group's recommendation and that Census participated in planning
efforts that included sharing business establishment data.  BEA and
BLS officials said that BEA, BLS, and Census were active participants
with Congress in designing and drafting the Foreign Direct Investment
and International Financial Data Improvements Act of 1990, a major
piece of data-sharing legislation.  (See app.  IX.)


   CONSTRUCTION - COVERAGE
------------------------------------------------------- Appendix III:7

Coverage measures a survey's ability to obtain information from all
units in a specific population.  Undercoverage occurs when a survey
fails to represent all units in the survey.  Construction coverage is
the ability of a survey to obtain information about residential and
nonresidential structures in a survey. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:7.1

According to the NABE Statistics Committee, surveys missed
substantial portions of construction activity in residential
remodeling and nonresidential reconstruction.  According to Census
officials, construction activity resists easy measurement.  The
industry is characterized by a heterogeneous product, producers who
are widespread and small-scale, and complex financial and working
arrangements.  In particular, the coverage of nonresidential
improvements and alterations (reconstruction) and construction done
at industrial sites is incomplete.  The development of nonresidential
price indexes, needed for the measurement of real investment and
productivity, is difficult since each new project is unique in some
way.  Census officials said that nonresidential price indexes have
been recognized as inadequate for 30 years. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:7.2

In 1991, the working group noted that private nonresidential
construction activity was understated by as much as 25 percent a
year, primarily in the areas of construction of industrial plants and
commercial reconstruction, such as renovation and improvement of
existing structures.  The working group recommended the improvement
of the coverage and accuracy of construction statistical measures. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:7.3

Census planned to improve its data on the values of nonresidential
construction and industrial construction, and develop a
nonresidential price index. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:7.4

Census requested about $4 million total for fiscal years 1990 through
1994 for this recommendation, but received no funding. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:7.5

Census officials said that Census developed new methods to improve
nonresidential construction coverage and price indexes, but
implementation of those methods requires future funding.\14


--------------------
\14 According to BEA officials, the methodological fixes BEA used to
attempt to eliminate the shortfalls in nonresidential construction on
the 1991 GNP revision do not alleviate the need for improved survey
coverage of both prices and current-dollar spending. 


   INVESTMENT AND SAVING
------------------------------------------------------- Appendix III:8

According to The MIT Dictionary of Modern Economics,\15

investment is a term "used to describe the flow of expenditures
devoted to increasing or maintaining the real capital stock [the sum
of goods which are used as inputs for further production]... 
Investment is the flow of expenditures devoted to projects producing
goods which are not intended for immediate consumption.  These
investment projects may take the form of adding to physical and human
capital as well as inventories."\16

According to the Encyclopedia of Economics,\17 "saving is the process
of withholding current income for future use and results in the
accumulation of tangible and financial assets.  The amounts so
accumulated over past periods are referred to as savings....On a net
basis, individuals save when personal income after taxes exceeds
personal outlays, businesses save through retained profits, and
governments save when current receipts exceed current expenditures. 
Gross saving includes...depreciation allowances covering the wear and
tear on real assets for future replacement....The channels through
which gross saving flows into gross investment, including financial
intermediaries are traced in the flow-of-funds accounts."


--------------------
\15 David W.  Pearce, ed., The MIT Dictionary of Modern Economics,
3rd ed.  (Great Britain:  The Macmillan Press, 1986), pp.  51, 55,
216, and 378.  Note that these are conceptual definitions, and the
recommendation addresses the issue of how these terms are to be
defined in statistical practice. 

\16 The MIT Dictionary of Modern Economics also defines capital stock
as the sum of goods that are used as inputs for further production. 

\17 Sally S.  Ronk, "Saving," in Encyclopedia of Economics, Douglas
Greenwald, ed., (New York:  McGraw-Hill, 1982), p.  837. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:8.1

According to the NABE Statistics Committee, statistical measures of
investment and saving in the United States do not allow policymakers
to address the question as to whether there is enough saving to
support the type of investment the nation needs to improve America's
competitiveness.  Estimates of saving and investment are not
internationally comparable and may be misleading.  For example, the
NABE Statistics Committee stated that statistical measures of
household saving come from two sources.  In NIPA as published by BEA,
saving is calculated by subtracting spending from income (a current
accounting convention method).  In the flow-of-funds estimates as
published by the Federal Reserve, saving is calculated from changes
in household assets and liabilities (a capital accounting convention
method).  At the 1987 National Bureau of Economic Research's
Conference on the Measurement of Saving, Investment, and Wealth, the
differences between the measurement of saving in NIPA and the
flow-of-funds estimates were examined.\18 Possible measurement errors
were noted for both methods. 


--------------------
\18 Robert E.  Lipsey and Helen Stone Tice, eds., The Measurement of
Saving, Investment, and Wealth, Vol.  52 (Chicago:  The University of
Chicago Press, 1989), pp.  1-3, 11-12. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:8.2

In 1990, the working group noted the significant differences
researchers found in statistical measures of saving and investment,
especially between NIPA and the flow-of-funds statistical measures. 
It said that the Federal Reserve and the statistical agencies should
work to improve the measures and, to the extent possible, reconcile
the differences between the various measures of saving. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:8.3

The Federal Reserve began its data improvement effort for the flow of
funds in 1991 and expanded the program in 1992.  All told, 17
specific data improvement projects were planned: 

  household sector improvements;

  state and local government sector improvements;

  noncorporate business sector improvements;

  pension sector benchmarks;

  property/casualty insurance sector improvements;

  finance company sector improvements, including review of the
     treatment of mortgage companies;

  credit union sector improvements;

  expansion of the sector for issuers of asset-backed securities;

  detailed accounting of closed-end funds;

  review of treatment of mortgage transactions;

  review of treatment of commercial paper transactions;

  review of treatment of corporate equity estimates;

  review of treatment of repurchase agreements;

  review of treatment of foreign transactions;

  improvements to the methodology for the monthly debt aggregate;

  balance sheet enhancements; and

  exploration of moving toward market valuation of assets and
     liabilities. 

In addition, the Federal Reserve planned to begin work on a
public-use source manual to document sources and methods. 

BEA and Federal Reserve officials said that the best way to reconcile
the measures of investment and saving would be to improve underlying
investment and saving data.  BEA officials said that they interpreted
this recommendation to be the plans for improvements they made for
NIPA (for an example, see the construction-methodology recommendation
above) and its international accounts (for an example, see the
international investment and capital flows recommendation in app. 
VIII).  BEA officials also said that as BEA moved to the SNA concept
of investment and saving and SNA fully integrated flows and stocks,
BEA's saving and investment data would both be consistent with other
countries and reconciled to changes in stocks (see the SNA
recommendation above). 

Census, the agency that collects some of the underlying data for the
measurement of saving and investment, planned to improve its coverage
of the manufacturers shipments, inventories, and orders survey. 
Monthly estimates of the shipments of goods from manufacturers are
the bases of BEA's current quarterly estimates of business investment
in producers' durable equipment.  Monthly estimates of manufacturers'
inventories are the bases of BEA's current quarterly estimates of the
manufacturing component of the change in the business inventory
component of gross private domestic investment.\19

No plans were made to reconcile directly the differences between the
various measures of investment and saving because the agencies wanted
to improve the underlying data used in the measurements. 


--------------------
\19 BEA officials noted that in terms of source data, Census' Annual
Capital Expenditures Survey (ACES) is the most important improvement
related to investment estimates. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:8.4

Census requested $900,000 and actually received $638,000 total for
fiscal years 1990 through 1994 for this recommendation. 

Since Congress does not appropriate funds for the Federal Reserve, we
did not request specific funding information about what the Federal
Reserve spent on this recommendation.  The Federal Reserve's Board of
Governors funded all improvements for this recommendation.  Officials
said additional staff were hired and funds were made available to
acquire additional data and to conduct research. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:8.5

Federal Reserve officials said 11 of the 17 data improvement projects
have been completed: 

  pension sector benchmarks;

  property/casualty insurance sector;

  finance company sector improvements, including review of the
     treatment of mortgage companies;

  credit union sector improvements;

  issuers of asset-backed securities;

  closed-end funds;

  mortgage transactions;

  commercial paper transactions;

  corporate equity estimates;

  foreign transactions; and

  monthly debt aggregate methodology changes. 

The remaining improvement projects are ongoing.  The Federal Reserve
also documented the differences in concepts of personal saving
between its flow-of-funds estimates and NIPA.\20

BEA officials said that progress on this recommendation is reported
under NIPA and the international transactions statistics categories
of recommendations.  For example, BEA officials said that under the
construction-methodology recommendation a 25 percent upward revision
was made in nonresidential construction, which resulted in an
improvement in the measure of investment.  Another example cited by
BEA was an improvement in international data that improved the
Federal Reserve's financial measures of investment and saving that is
reported under the international investment and capital flows
recommendation (see app.  VIII). 

Census officials said Census improved the measurement of investment
by improving coverage of the manufacturers shipments, inventories,
and orders survey. 


--------------------
\20 Board of Governors of the Federal Reserve System, Guide to the
Flow of Funds Accounts (Washington, D.C.:  Federal Reserve System,
1993). 


   FLOW OF FUNDS
------------------------------------------------------- Appendix III:9

Flow-of-funds accounts, also referred to as the capital finance
accounts of the United States, show the role of financial
institutions and instruments in transforming saving into investment
and the changes in assets and liabilities that result from this
transformation.  The Federal Reserve prepares and publishes
flow-of-funds accounts data.  The flow-of-funds accounts detail how
the current investment in tangible and financial assets contribute to
a buildup of a stock of assets for each sector of the economy and to
the creation of national wealth.  The flow-of-funds accounts can be
viewed as combining data on the flows of saving and tangible
investment published in NIPA with additional details on borrowing and
lending for specific economic sectors. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix III:9.1

As of the late 1980s, detailed information about the construction of
the flow-of-funds accounts was not available in a form that was
accessible to users outside the Federal Reserve.  This problem was
noted by CEA Chairperson Michael Boskin in a letter sent to the
Federal Reserve prior to release of ESI.  The Federal Reserve had
previously planned to upgrade its documentation and subsequently
incorporated that work into its broader initiative to improve its
flow-of-funds accounts. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix III:9.2

The working group, in 1990, recommended improvement in the
collection, coverage, and processing procedures for the financial
data in the Federal Reserve's flow-of-funds accounts. 


      AGENCY PLANS
----------------------------------------------------- Appendix III:9.3

As part of its overall data improvement program for the flow of funds
as discussed under the investment and saving recommendation, the
Federal Reserve planned to publish detailed documentation of how the
flow of funds is estimated--the total sources of funds flowing to the
economic sector and the sector's uses of funds.  This would enable
researchers to identify where and how the estimates were derived. 
The Federal Reserve also planned to improve its internal processing
by upgrading its software and seasonal adjustment procedures. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix III:9.4

Because Congress does not appropriate funds for the Federal Reserve,
we did not request specific funding information about what the
Federal Reserve spent on this recommendation.  The Federal Reserve's
Board of Governors funded all improvements for this recommendation. 
Officials said additional staff were hired and funds were made
available to acquire additional data and to conduct necessary
research. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix III:9.5

The Guide to the Flow of Funds Accounts, the documentation of how the
flow of funds is estimated, was published in June 1993.  Federal
Reserve officials said that improvements in internal processing were
implemented also. 


   ANNUAL INVESTMENT SURVEY
------------------------------------------------------ Appendix III:10

According to the NABE Statistics Committee, an annual investment
survey would be a program that integrated, consolidated, and
developed quarterly, annual, and periodic investment surveys.  The
survey program would provide detailed information by type of asset
that is needed for the nonresidential fixed investment component of
NIPA.  The program would improve the consistency of capital
expenditure estimates across all investment data collection programs
and would consolidate the capital expenditure questions from several
different periodic surveys. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix III:10.1

According to the NABE Statistics Committee, the data used to estimate
business investment expenditures have been inconsistent and
incomplete.  For example, estimates of equipment investment were
constructed from producers' shipments, rather than measures of
investors' purchases.  A new annual survey program was recognized as
a way to consolidate questions contained in several periodic surveys,
improve the consistency of capital expenditure estimates across all
investment data collection programs, and provide more detail by type
of asset than was being collected. 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix III:10.2

In 1990, the working group recommended the redirection of funds from
an outdated plant and equipment survey to undertake an annual
investment survey that had been proposed by Census.  The plant and
equipment survey was incomplete in coverage, subject to nonresponse
bias, and was no longer used in NIPA. 


      AGENCY PLANS
---------------------------------------------------- Appendix III:10.3

Census planned three phases to institute its new annual capital
expenditure survey (ACES).  Census first planned to consult with
Statistics Canada and other experts at a conference on the planned
survey, and then conduct a pilot survey.  Next, Census planned to
send its survey to 4,400 companies with 5 or more employees.  Early
respondents were to be contacted to identify problems in completing
the survey.  Also, in this second phase, survey forms were to be
mailed to 11,000 firms and these data were to be aggregated and
published.  The third phase was to fully implement ACES with 45,000
firms. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix III:10.4

Census requested and actually received $2 million total for fiscal
years 1990 through 1994 for this recommendation.\21


--------------------
\21 Census officials said that as of May 1994 the plant and equipment
survey was still being conducted on a limited scope because Census
found it necessary to continue the survey until ACES is fully
implemented.  Therefore, the funds for the plant and equipment survey
had not been redirected to ACES.  In subsequent comments, Census
officials said that the plant and equipment survey was conducted on a
limited scope until December 1994.  Effective January 1995, funds for
the plant and equipment survey were redirected to ACES. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix III:10.5

Census said it held a 1-day conference of experts and consulted with
Statistics Canada.  Based on these consultations, Census developed a
pilot survey for 1992 and conducted personal and telephone interviews
with companies to discuss their ability to report.  The surveys were
mailed to 4,400 companies in 1992 and to 11,000 companies in 1993. 
In the surveys, companies reported investment in industries in which
they operate.  Census aggregated the data from the latest mailing
that had an 81- percent response rate.  These data, which were
released in July 1994, were used to estimate total capital
expenditures and expenditures for each industry. 

Census found that collecting equipment investment information from
large companies imposed extensive burden on those companies. 
Therefore, Census will collect detailed information annually on types
of equipment and structures on a rotating basis once every 5 years to
reduce the burden on any particular company. 


   DATA GAPS
------------------------------------------------------ Appendix III:11

Data gaps refer to a situation where insufficient information is
available to derive an accurate observation of evidence based on
data.  Data gaps may result from a lack of up-to-date estimates, or
revised data, or a data source may not be available that would
provide more conclusive information. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix III:11.1

Although BEA publishes NIPA and the international economic accounts,
it does not directly collect most of the data used to produce the
accounts.  BEA obtains statistical measures with data relevant to the
accounts from other statistical agencies.  According to a BEA
official, much of the data are not in the form needed by BEA.  Some
of the data are not collected, thereby creating data gaps in the raw
source data BEA uses to produce NIPA and the international economic
accounts. 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix III:11.2

In 1991, the working group recommended a central integrating role for
economic accounts by providing BEA with funds to support selected
improvements in the source data it obtains from other agencies.  The
funds were to be used to secure additional information from
administrative records held by government agencies, to support new
statistical surveys, or extensions to existing surveys, and to carry
out research to determine feasible and cost-effective ways of closing
gaps in source data. 


      AGENCY PLANS
---------------------------------------------------- Appendix III:11.3

BEA officials said they planned to improve data in 17 categories by
reducing data gaps for

  retirement and other benefit payments, which result from changes in
     payroll and income tax rules;

  corporate expenses and income, which result from changes in tax law
     that no longer require separation of this information on tax
     returns;

  foreign income of U.S.  corporations, which result from rapid
     growth in size of these corporations and an increase in
     diversity of their income and expense items;

  unregulated activities, which result from industries not supplying
     statistical or administrative data because they are no longer
     regulated;

  nonprofit institutions, which result from a lack of source data
     from industries such as health and education;

  merchandise trade and international trade in services, which result
     from missing source data;

  health services, which result from infrequent data collection;

  manufacturing, which result from a lack of data on shipments by
     detailed product classifications across all manufacturing
     industries;

  no-employee firms, which result from undercoverage of firms without
     employees;

  direct investment firms, which result from a lack of cross-checking
     between sources of information and the changing universes of
     U.S.  affiliates of foreign parent firms and foreign affiliates
     of U.S.  parent firms;

  transactions involving the U.S.  government, which result from slow
     data reporting and lack of data on grants-in-aid and military
     transactions;

  economic classifications, which result from emerging industries not
     being tracked because they are not separately identified by the
     standard industrial classification (SIC) system;

  bank output, which result from changes in the quality of services
     provided to bank customers;

  international portfolio investment, which result from the increased
     volume of security transactions flowing outside of traditional
     financial channels;

  international trade in services, which result from the increased
     number of firms becoming international corporations;

  unilateral transfers, which result from the increase in personal
     remittances paid to foreign residents by U.S.  residents; and

  data exchange and reconciliations, which result from the desire to
     improve the accuracy and speed of reporting between major
     trading partners. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix III:11.4

BEA requested about $5 million total for fiscal years 1990 through
1994 for this recommendation, but received no funding.\22


--------------------
\22 BEA officials said they also included a request for funding for
the SIC recommendation under this request for the data gaps
recommendation. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix III:11.5

BEA officials said they made no progress implementing 11 of the 17
categories of data gaps because they received no funding.  For six
categories, however, BEA requested and received funds under other ESI
recommendations.  Progress for the economic classifications data gap
is reported under the SIC recommendation (see app.  IX); bank output
is reported under the indirect estimation recommendation (see
elsewhere in this app.); and international trade in services is
reported under the trade in services recommendation (see app.  VIII). 
Progress for the international portfolio investment, unilateral
transfers, and data exchange and reconciliations categories of data
gaps are reported in the international investment and capital flows
recommendation (see app.  VIII). 


   SUMMARY
------------------------------------------------------ Appendix III:12

A summary of the funding requested and actually received by the
statistical agencies for the working group's 11 NIPA statistics
recommendations is shown in table III.1.  Census and BEA requested
nearly $21 million total for 10 recommendations for fiscal years 1990
through 1994.\23 They received nearly $6 million for 6 of the NIPA
statistics recommendations in fiscal years 1990 through 1994. 



                                                                      Table III.1
                                                        
                                                            Funding for the NIPA Statistics
                                                         Recommendations for Fiscal Years 1990-
                                                                          1994

                                                                 (Dollars in thousands)


                      Requeste           Requeste
Recommendation               d   Actual         d   Actual    Requested     Actual   Requested     Actual   Requested     Actual   Requested     Actual
--------------------  --------  -------  --------  -------  -----------  ---------  ----------  ---------  ----------  ---------  ----------  ---------
Indirect estimation          0        0    $1,900    $ 400      $ 1,300    $ 1,300           0          0           0          0     $ 3,200     $1,700
 methods\a

 Input-output
 tables\a

 Construction
 methodology\a
System of national           0        0     1,600      700        1,700        500       1,200          0       1,200          0       5,700      1,200
 accounts
Inflation                    0        0         0        0            0          0           0          0           0          0           0          0
 adjustments
Business                     0        0         0        0          200          0         200          0           0          0         400          0
 establishment data
 Census and BEA
Construction                 0        0         0        0        1,300          0       1,300          0       1,300          0       3,900          0
 coverage
Investment and               0        0       900      638            0          0           0          0           0          0         900        638
 saving\b
Flow of funds\b
Annual investment            0        0     2,000    2,000            0          0           0          0           0          0       2,000      2,000
 survey
Data gaps                    0        0         0        0            0          0       3,900          0         800          0       4,700          0
=======================================================================================================================================================
Total                        0        0    $6,400   $3,738       $4,500     $1,800      $6,600          0      $3,300          0   $20,800\c     $5,538
-------------------------------------------------------------------------------------------------------------------------------------------------------
\a BEA combined these three recommendations into one budget request
to increase funding to stop the deterioration in the quality of the
national economic accounts. 

\b Federal Reserve budget information on this recommendation was not
included because the Federal Reserve does not receive any funds from
Congress and internally funded the recommendation. 

\c Total request includes $5,200,000 in reinstated requests, that is,
funding that was not received in prior years. 

Sources:  BEA, Census. 


--------------------
\23 The Federal Reserve provided funding for two recommendations. 
Congress does not appropriate funds for the Federal Reserve;
therefore, the funds used for these initiatives are not included in
table III.1. 


SERVICE SECTOR STATISTICS
RECOMMENDATIONS
========================================================== Appendix IV

The service sector--the part of the economy that produces products
with an intangible component, like assistance or information--has
grown in the last three decades.  The service sector in 1960
accounted for more than three-fifths of all employment and about
two-fifths of the gross domestic product (GDP); in 1990, it accounted
for three-fourths of employment and one-half of GDP.  In the view of
the ESI working group and other statistics users, the statistical
measures used to track this growth are outdated. 

The working group was concerned about the inability of the
statistical agencies to keep up with the changing nature of the
economy.  Because of this concern, the working group created a
subgroup on service sector statistics to develop a comprehensive work
plan to improve service sector statistics.  The work plan included
the expansion of basic data collection as well as other
recommendations that we have classified in other appendixes.  The
working group's recommendations added to and provided direction for
improvement efforts that had been identified by various agencies in
1984.  The working group's recommendations also called for an
acceleration of the implementation of these 1984 service sector
improvement efforts. 


   SERVICE SECTOR SURVEYS
-------------------------------------------------------- Appendix IV:1

Service sector surveys are detailed studies that collect information
from the part of the economy that produces intangible products, like
assistance or information.  All sectors, except those classified as
agriculture, construction, manufacturing, and mining, are included in
these surveys.\1


--------------------
\1 See the glossary for definitions and descriptions of technical
terms and phrases that are used throughout this appendix. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IV:1.1

The Committee on National Statistics of the National Research Council
(CNSTAT) reported that because of the deregulation of industries such
as transportation, the collection and publication of data by the
agencies regulating these industries generally has been reduced or
discontinued.\2 Notably, CNSTAT reported that the data needed to
measure the effects of deregulation were being affected.  For
example, the series on financial data for radio and television
stations was discontinued by the Federal Communications Commission. 
In addition, businesses entered the formerly regulated markets, and
in some instances, businesses expanded their range of offered
services as regulations were lifted or reduced.  For example, banking
institutions now offer mutual funds. 


--------------------
\2 Sol D.  Helfand, Vito Natrella, and Alan E.  Pisarski, Statistics
for Transportation, Communication, and Finance and Insurance:  Data
Availability and Needs, Committee on National Statistics, Commission
on Behavioral and Social Sciences and Education, National Research
Council, report of the National Academy of Sciences (Washington,
D.C.:  1984), pp.  xi, 3. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IV:1.2

In 1990, the working group stated that better statistical measures
for the service sector ultimately required better raw data.  In
addition, the working group recommended that the statistical agencies
accelerate their plans for long-term improvements so that industries
whose measurement problems were the most significant would be
addressed sooner.  In 1991, the working group recommended that Census
specifically increase the detail and coverage of the service sector
in its service annual survey and the periodic census of service
industries.  These improvements would permit more complete coverage
of the service sector and provide a more detailed picture of the
composition of the service industries. 


      AGENCY PLANS
------------------------------------------------------ Appendix IV:1.3

Following the improvement effort of various agencies that was
developed in 1984, Census had introduced in fiscal year 1986 an
annual survey coverage of the for-hire trucking and public
warehousing industries; it planned to begin a new annual survey of
communication services.  Census also planned to draw additional
samples from the 1987 Economic Censuses.  On the basis of the 1987
Standard Industrial Classification (SIC) system--a federally designed
system classifying establishments by industry--samples of the
financial, insurance, and real estate industries were to be used to
expand the service annual survey. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IV:1.4

Census requested about $7 million and actually received more than $1
million total for fiscal years 1990 through 1994. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IV:1.5

Census began reporting information from its annual survey of
communication services of telephone communications, telegraph and
other message communications, radio and television broadcasting
stations and networks, cable and other pay television services, and
other communication services, such as radar station operations.  The
first set of survey data was published in March 1992.  Census
increased the sample size for the 1991 Service Annual Survey from
20,000 to 30,000 firms.  This increase allowed Census to publish data
on industries and organizations in personal, business, amusements and
recreation, social, health, and other professional services that were
previously covered only in the economic censuses.  The number of
service industries whose data were published in the 1991 Service
Annual Survey increased from 90 in 1990 to over 130.  Census also
expanded service sector data by providing expense data in addition to
revenue data for tax-exempt firms in selected service industries and
detailed information on receipts for six published industry groups. 


   PURCHASED SERVICES
-------------------------------------------------------- Appendix IV:2

Purchased services are assistance, information, and other intangible
products that businesses and organizations buy from businesses and
governments.  Examples of purchased services include when a book
publisher contracts with a cleaning service for janitorial work
performed in its warehouses, or when a manufacturer contracts for
computer services it needs, but formerly provided itself. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IV:2.1

According to the National Association of Business Economists' (NABE)
Statistics Committee, weak industrial activity reported in the
mid-1980s may have reflected in part the statistical agencies'
inability to trace the new firms being formed.\3 Many services and
functions that had previously been handled within one company were
contracted out to new small service firms when manufacturing firms
downsized.  Employees who lost their jobs were often hired by the new
service firms.  In many cases the same employees continued to perform
the same work.  As a result, the proportion of employment classified
in manufacturing was reduced.  To maintain coverage of economic
activity, a statistical system needs to quickly identify and track
these new firms.  Such tracking is a difficult and costly process
because new firms are constantly opening and closing. 

A Census official said that Census collects data every 5 years on
purchased services.  The official also said that Census needs data on
an annual basis for developing value-added statistical measures on
manufacturing industries.  The official said that such measures may
be overstated by as much as 30 percent because of limited data. 

BEA officials said that one of the most serious problems in tracking
new firms is the lack of an up-to-date list of firms with no paid
employees.  Small businesses are particularly difficult to track. 


--------------------
\3 Improving Federal Statistics:  Hearings Before the Joint Economic
Committee of Congress of the United States.  102nd Cong., 1st Sess.,
pp.  67-127 (1991) (statement by Martin Fleming, Chairperson of the
National Association of Business Economists' Statistics Committee). 
The statement reflects the views of the NABE's Statistics Committee. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IV:2.2

In 1991, the working group recognized that changing trends, which
resulted from purchasing legal, accounting, and other services by
manufacturing industries, were not accurately measured by existing
economic surveys.  To improve the tracking of the sources of economic
growth and structural change by industry, the working group
recommended the provision of needed statistical measures. 


      AGENCY PLANS
------------------------------------------------------ Appendix IV:2.3

To meet this recommendation, Census planned to improve the
measurement of services purchased by manufacturing industries with an
annual collection of data on purchased services, such as management
consulting, engineering, and legal assistance.  Census proposed a
3-year plan.  In the first year of its plan, Census was to complete
the sample design.  In the second year of its plan, Census was to
pretest questionnaires with manufacturing firms to determine the (1)
most important purchased services and (2) best level to survey in a
manufacturing firm to obtain data on purchased services.  In the
third year of its plan, Census was to introduce the data-collection
instrument on purchased services as part of its annual survey of
manufacturers. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IV:2.4

Census requested more than $3 million total for fiscal years 1990
through 1994 for this recommendation, but received no funding. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IV:2.5

Census officials said they made no progress implementing the
recommendation because they received no funding.\4


--------------------
\4 Census officials noted that the 1992 Economic Censuses collected
additional information on purchased services.  We also reported in
1993 that more components of the service sector were included for the
first time in the 1992 Economic Censuses.  See Federal Data
Collection:  Status of 1992 Agriculture and Economic Censuses and
Future Challenges (GAO/GGD-93-152BR, Sept.  1993). 


   CORPORATE FINANCIAL DATA
-------------------------------------------------------- Appendix IV:3

Corporate financial data are the economic accounts of an institution
that make up the balance sheet and income statement.  Data in these
reports include assets, liabilities, and stockholders' equity, income
(or loss) from operations, and the breakdown of income before and
after taxes. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IV:3.1

The Quarterly Financial Report (QFR) program is the primary source of
information about the financial position of U.S.  corporations. 
However, the QFR program does not cover corporations operating in the
service sector and inadequately covers small corporations operating
in retail and wholesale trade, manufacturing, and mining.  According
to Census officials, the GDP estimates and the flow-of-funds accounts
would benefit significantly by expanding the scope of the QFR program
to include services and small corporations. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IV:3.2

In 1991, the working group recognized that the comprehensive
statistical measure of the financial performance of the service
sector and the small business community was either deteriorating or
nonexistent.  It recommended greater precision in estimates by
industry and more comprehensive balance sheet and financial ratio
data by asset size. 


      AGENCY PLANS
------------------------------------------------------ Appendix IV:3.3

Census planned to expand corporate financial data by collecting
quarterly financial data for the service sector and by expanding
coverage of small corporations.  These corporate financial data
estimates, which are included in Census' Quarterly Financial Report
Data, are used in the calculation of GDP and flow-of-funds accounts. 
Census already had collected corporate profit data for manufacturing,
mining, and retail and wholesale trade industries but planned to
expand data to service sector industry areas, such as business
computer services and managerial consulting, because business
services account for one-third of corporate profits in the service
area. 

Census proposed a 3-year plan to expand corporate profit data for
service sector industries.  In the first 2 years of its plan, Census
was to complete the survey design, select samples of business
services, and develop a questionnaire.  In the third year of its
plan, Census was to administer the questionnaire to business service
sector firms. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IV:3.4

Census requested nearly $5 million total for fiscal years 1990
through 1994 for this recommendation, but received no funding. 
Specifically, for the small corporations improvements, Census
requested slightly more than $2 million.  For the other corporate
financial data improvements, Census requested slightly less than $3
million. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IV:3.5

Because of the lack of funding, Census officials said that they made
no progress in implementing this recommendation. 


   SUMMARY
-------------------------------------------------------- Appendix IV:4

A summary of the funding requested and actually received by Census
for the working group's three service sector statistics
recommendations is shown in table IV.1.  Census requested more than
$15 million total for the three recommendations for fiscal years 1990
through 1994.  Census received more than $1 million for only the
service sector surveys recommendation in fiscal years 1990 through
1994. 



                                                                       Table IV.1
                                                        
                                                           Funding for Census' Service Sector
                                                         Statistics Recommendations for Fiscal
                                                                    Years 1990-1994

                                                                 (Dollars in thousands)


                      Requeste           Requeste
Recommendation               d   Actual         d   Actual    Requested     Actual   Requested     Actual   Requested     Actual   Requested     Actual
--------------------  --------  -------  --------  -------  -----------  ---------  ----------  ---------  ----------  ---------  ----------  ---------
Service sector               0        0    $1,050        0       $1,950     $1,400      $2,099          0      $2,017          0     $ 7,116     $1,400
 surveys
Purchased services           0        0       523        0        1,023          0       1,023          0         923          0       3,492          0
Corporate financial          0        0       868        0        1,668          0       1,626          0         808          0       4,971          0
 data
=======================================================================================================================================================
Total                        0        0    $2,441        0       $4,641     $1,400      $4,748          0      $3,748          0   $15,579\a     $1,400
-------------------------------------------------------------------------------------------------------------------------------------------------------
\a Total request includes $6,547,000 in reinstated requests, that is,
funding that was not received in prior years. 

Source:  Census. 


PRICE MEASUREMENT STATISTICS
RECOMMENDATIONS
=========================================================== Appendix V

Price measurements are indexes that trace the relative changes in the
price of an individual good or service, or a market basket of goods
and services over time.  Price indexes are used to measure
inflation.\1

The working group initially focused on problems in measuring price
changes in the service sector because the measure of real output and
productivity is affected by statistical measures of inflation.  For
example, if the measured change in prices overstated actual
inflation, then the statistical measure of productivity would be
understated.  In the next year, the working group recognized that the
problems of separating quality changes from "pure" price changes were
also difficult for physical goods and recommended additional research
in nonservice price measures. 


--------------------
\1 See the glossary for definitions and descriptions of technical
terms and phrases that are used throughout this appendix. 


   SERVICE PRICES
--------------------------------------------------------- Appendix V:1

Service prices are measures of price changes in the service
sector--the part of the economy composed of businesses that sell
assistance and expertise rather than tangible goods.  Because
services are intangible, measuring their output and breaking down
changes in spending on services into changes in output and changes in
prices has been difficult.  Service prices are measured in three
price indexes. 

  The Consumer Price Index (CPI) is a BLS program that measures
     average price change for two population groups--all urban
     consumers and urban wage earners and clerical workers--using a
     specified market basket representing all goods and services
     purchased for everyday living. 

  The Producer Price Index (PPI) is a BLS program that measures
     average changes in prices received by domestic producers for
     their output (crude materials, semifinished goods, finished
     goods, and some services), relative to prices received in the
     base year.  It is an industry-based survey that provides monthly
     price indexes for virtually all agriculture, mining, and
     manufacturing industries and selected service industries. 

  The International Price Program (IPP) measures price changes of
     commodities traded between the United States and the rest of the
     world.  IPP produces the U.S.  Export Price Index and the U.S. 
     Import Price Index.  These indexes are intended to reflect price
     trends of foreign-produced goods entering the United States or
     domestic products leaving this country or its territories. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------- Appendix V:1.1

The absence of reliable price indexes for services affects the basic
measurements of inflation and productivity.  According to the
National Association of Business Economists' (NABE) Statistics
Committee,\2 the impact on inflation of changes in the service sector
is relatively unknown but is perceived to be large and not easily
controlled by anti-inflation policies.  Also, the lack of services
price data limits the ability to develop reliable indexes of
constant-dollar output of service sector industries.  (See indirect
estimation methods recommendation in app.  III.) Since statistical
measures of output are used to calculate productivity, the
reliability of this measure is also questioned by NABE.  Available
statistical measures suggest that much of what appears to be a
slowdown in U.S.  productivity is attributed to relatively low
productivity growth in service sector industries.  But, because of
measurement error, NABE concludes that the reasons for the slowdown
are not clear. 

According to BLS officials, IPP had no infrastructure to support
service indexes until BLS received a small appropriation in the
mid-1980s.  BLS said that it had to devote a disproportionate amount
of its resources to establishing staffs for sampling, estimation, and
data processing, in addition to the staff for planning, data
collection, and analysis.  Following this initial funding for
services, work was begun on developing a few indexes in the
transportation area. 


--------------------
\2 Improving Federal Statistics:  Hearings Before the Joint Economic
Committee of Congress of the United States.  102nd Cong., 1st Sess.,
pp.  67-127 (1991) (statement by Martin Fleming, Chairperson of the
National Association of Business Economists' Statistics Committee). 
The statement reflects the views of the NABE's Statistics Committee. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------- Appendix V:1.2

In 1990, the subgroup on service sector statistics reported on BLS'
major research initiative to improve quality adjustments in measuring
price change over time.  The subgroup said that BLS' initiative
included accelerating research efforts on improving quality
adjustments in the CPI; conducting experimental work and research on
computers, semiconductors, hospitals, television, and retail trade
for PPI; expanding coverage for real estate and beginning coverage of
the health insurance industry for PPI; and developing indexes for
water and air transportation, and electricity for IPP. 

In 1990, the working group recommended that BLS accelerate its plans
to expand and improve its CPI, PPI, and IPP to measure service prices
more accurately.  Later in 1991, the working group recommended that
BLS conduct research to develop accurate, replicable, and
standardized statistical measures of output for the service sector. 
These measures were to permit the introduction of new service sector
price indexes into PPI and IPP. 


      AGENCY PLANS
------------------------------------------------------- Appendix V:1.3

BLS planned to add health services and most of commercial real estate
to its PPI.  Also, for IPP, BLS planned to complete coverage that
began in the mid-1980s of most transportation industries as well as
telephone message services, both imported and exported.  No plans
were made for expanding or improving service prices in the CPI. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------- Appendix V:1.4

BLS requested more than $7 million and actually received less than $6
million total for fiscal years 1990 through 1994.\3

Specifically for PPI improvements, BLS requested about $4 million and
received slightly more than $3 million.  For IPP improvements, BLS
requested more than $3 million and received less than $3 million. 
For CPI, BLS did not request or receive funding.  BLS requested
$100,000 for productivity research, which officials said was related
to this recommendation. 


--------------------
\3 According to BLS officials, their original funding request was
reduced in fiscal years after 1992.  They said that their subsequent
budget requests were intended to fund implementation of the
recommendation at this reduced level, rather than fund the
recommendation as originally envisioned. 


      PROGRESS REPORTED
------------------------------------------------------- Appendix V:1.5

BLS expanded the coverage of its PPI by adding the hospitals index in
January 1993 and doctors of medicine index in January 1994.  BLS
officials said that the medical laboratories index was published in
August 1994; the nursing homes index will be published in 1995; and
the operators of nonresidential buildings index and the real estate
agents and managers index will be published in 1996.  The additional
planned service sector industries in medical services and commercial
real estate were dropped because of subsequent budget reductions. 
BLS officials said that resources do exist, however, to develop a
limited number of additional indexes for such industries as legal,
accounting, and auditing services and insurance carriers. 

According to BLS officials, a series of budget reductions following
the recommendation's initial funding resulted in eliminating any
development of additional new indexes in IPP.  Publication of the
service indexes in IPP is limited to existing coverage in
transportation.  This includes import indexes for air passenger, air
freight, liner freight, oil tanker freight, and electrical energy. 
It also includes export indexes to air passenger services. 


   SEPARATION OF QUALITY AND
   INFLATION CHANGES
--------------------------------------------------------- Appendix V:2

The accurate measurement of inflation depends on the characteristics
of a good or service staying the same over time.  According to BLS,
measurement problems occur when the quality of the item and its price
change at the same time, or when an item is discontinued.  In these
circumstances, the price increase reflects not only an increase in
price but a change in the item's ability to meet the consumer's (or
producer's) needs. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------- Appendix V:2.1

BLS attempts to determine how a change in quality affects the price
for a number of products and makes adjustments to capture such
changes.  But BLS reports these changes cannot be easily made and are
especially difficult as more items appear in the marketplace.  BLS
also reports that substitutions for discontinued items are
particularly difficult to make. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------- Appendix V:2.2

In 1991, the working group recommended additional funding for BLS to
conduct research to determine the best way to separate the quality
change from the inflation component of price changes.  The working
group was concerned that measurement difficulties could result in an
overstatement of the rate of inflation. 


      AGENCY PLANS
------------------------------------------------------- Appendix V:2.3

For the CPI improvements, BLS planned to develop improved quality
adjustment procedures in consumer electronics, shelter, and apparel. 
For PPI improvements, BLS planned to accelerate resampling, which is
necessary to better maintain the currency of PPIs for industries that
introduce products more rapidly, and to develop procedures for 10
rapidly changing producer goods.  For PPI, BLS also planned to
undertake a very limited effort to develop additional statistical
methods of adjusting for quality change. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------- Appendix V:2.4

BLS requested more than $3 million and actually received more than $2
million total for fiscal years 1990 through 1994.  For the CPI
improvements, BLS requested $450,000, but received no funding. 
Specifically for the PPI improvements, BLS requested more than $3
million and received more than $2 million. 


      PROGRESS REPORTED
------------------------------------------------------- Appendix V:2.5

No improvements were made in the CPI.  BLS officials said that
acceleration of resampling for PPI was completed by 1993 for
approximately eight industries.  The resampling period for those
industries was reduced from the average of every 7 years to every 2
years.  BLS officials also said that the methods to statistically
adjust PPI for quality changes were in place on a very limited basis. 


   SUMMARY
--------------------------------------------------------- Appendix V:3

A summary of the funding requested and actually received by BLS for
the working group's two price measurement statistics recommendations
is shown in table V.1.  BLS requested nearly $11 million total and
received nearly $8 million for the two recommendations for fiscal
years 1990 through 1994. 



                                                                       Table V.1
                                                        
                                                           Funding for BLS' Price Measurement
                                                         Statistics Recommendations for Fiscal
                                                                    Years 1990-1994

                                                                 (Dollars in thousands)


                      Requeste           Requeste
Recommendation               d   Actual         d   Actual    Requested     Actual   Requested     Actual   Requested     Actual   Requested     Actual
--------------------  --------  -------  --------  -------  -----------  ---------  ----------  ---------  ----------  ---------  ----------  ---------
Service prices               0        0      $550        0       $2,250     $2,198      $2,769     $1,815      $1,672     $1,672      $7,241     $5,685
Separation of                0        0         0        0        1,790        871         970        719         719        719       3,479      2,309
 quality and
 inflation changes
=======================================================================================================================================================
Total                        0        0      $550        0       $4,040     $3,069      $3,739     $2,534      $2,391     $2,391     $10,720     $7,994
-------------------------------------------------------------------------------------------------------------------------------------------------------
Source:  BLS. 


LABOR MARKET STATISTICS
RECOMMENDATIONS
========================================================== Appendix VI

Labor market statistics are measures that provide information about
persons in the population, their employment status (employed,
unemployed, or not in the labor force), compensation received for
employment, costs to employers, and other measures of employment by
characteristics of the employed persons and their employers. 

The issues addressed by the working group in the labor market
statistics recommendations involved updating statistical measures to
keep up with a changing economy.  Many of these issues were
identified in 1979 by the National Commission on Employment and
Unemployment Statistics,\1 but due to budget constraints, BLS was not
able to implement its plans to address these issues in the 1980s.  In
1990, the working group gave priority to efforts to improve
statistical measures in wage trends such as the Employment Cost Index
(ECI) and to improve and modernize the Current Employment Statistics
(CES) program--a nationwide monthly business establishment survey,
which collects information on employment, workers' hours, and
earnings from the payroll records of employers--and the Current
Population Survey (CPS)--a nationwide monthly household survey, which
collects information on labor force, employment, and unemployment
experience of the nation's population.  In 1991, the working group
recommended that BLS complete specific recommendations to improve the
coverage and accuracy of CES, automate data collection for CPS, and
reconcile measurement differences between CPS and CES. 


--------------------
\1 National Commission on Employment and Unemployment Statistics,
Counting the Labor Force (Washington D.C.:  U.S.  Government Printing
Office, 1979). 


   EMPLOYMENT COST INDEX
-------------------------------------------------------- Appendix VI:1

ECI is a quarterly index measuring changes in total
compensation--wages, salaries, and employers' costs for employee
benefits.  ECI tracks changes in labor costs that are free from
influences of employment shifts among occupations and industries.\2


--------------------
\2 See the glossary for definitions and descriptions of technical
terms and phrases that are used throughout this appendix. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix VI:1.1

According to the National Association of Business Economists' (NABE)
Statistics Committee,\3 some sectors, such as the federal government,
agriculture, and private household, were not covered in ECI in the
late 1980s.  Neither was the index seasonally adjusted with
statistical procedures to eliminate the effects of events and other
practices that affect short-term trends in data.  Also, NABE said
that the lack of detail in ECI makes it difficult to estimate labor
costs in important industries, such as health care. 


--------------------
\3 Improving Federal Statistics:  Hearings Before the Joint Economic
Committee of Congress of the United States.  102nd Cong., 1st Sess.,
pp.  67-127 (1991) (statement by Martin Fleming, Chairperson of the
National Association of Business Economists' Statistics Committee). 
The statement reflects the views of NABE's Statistics Committee. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix VI:1.2

In 1990, the working group recommended that BLS expand its ECI to
cover additional sectors of the economy and to provide special
studies of important industries, such as health care.  The working
group also recommended that BLS seasonally adjust the index. 


      AGENCY PLANS
------------------------------------------------------ Appendix VI:1.3

BLS specifically planned to expand and publish separate statistical
measures of change in employer labor costs for nursing and personal
care facilities.  BLS planned to seasonally adjust the index.  BLS
did not make plans for the special studies as recommended by the
working group. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix VI:1.4

BLS requested and actually received less than $2 million total for
fiscal years 1990 through 1994 for this recommendation. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix VI:1.5

BLS expanded the sample for the nursing and personal care industry by
about 200 facilities and began publishing the data in 1992.  In
January 1991, BLS began publishing seasonally adjusted ECI data. 


   COVERAGE OF PAYROLL EMPLOYMENT
   ESTIMATES
-------------------------------------------------------- Appendix VI:2

Payroll employment estimates, as measured by CES, represent the
number and characteristics of employees on business payrolls for the
pay period including the 12th of every month.  These estimates are
published at detailed geographic and industry levels.  Coverage of
payroll employment estimates is the current level of published
industry detail that is available.  Publication coverage is limited
by the ability of the employment survey to develop sufficiently
accurate estimates at detailed industry levels. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix VI:2.1

Since it began in 1915, the CES program has focused on the employment
and earnings trends of manufacturing industries.  Despite the
continuing growth of the service sector, the majority of detailed
published data are for the manufacturing sector.  As of 1990, only 25
percent of the country's private sector employment was in the
manufacturing sector, yet detail for 55 percent of the industries in
that sector was published.  At the same time, while 75 percent of the
private sector employment was in the service sector, detail for only
32 percent of the industries in that sector was published. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix VI:2.2

The working group recognized that, despite the shift from
manufacturing industries to service sector industries, there was
substantially more detailed coverage of employment, hours, and
earnings in manufacturing industries than in service industries.  In
1991, the working group recommended that BLS expand the publication
of detailed industries in the service sector by adding 110 service
sector industries to its payroll survey. 


      AGENCY PLANS
------------------------------------------------------ Appendix VI:2.3

BLS planned to increase the number of service sector industries
published by 110 so that the proportion of industry detail published
for the service sector was equal to the proportion published for the
manufacturing sector.  This included over 650 new employment, hours,
and earnings monthly data series.  BLS selected the new industries
based on economic significance and included industries that had not
experienced large measurement revisions in the past.  This expansion
represented a 50-percent increase in the published service sector
detail available each month. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix VI:2.4

BLS requested nearly $9 million and actually received more than $7
million total for fiscal years 1990 through 1994.\4


--------------------
\4 According to BLS officials, their original funding request was
reduced in fiscal years after 1992.  They said that their subsequent
budget requests were intended to fund implementation of the
recommendation at this reduced level, rather than fund the
recommendation as originally envisioned. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix VI:2.5

According to BLS officials, because funding was reduced, BLS was only
able to publish one-half of the intended number of additional service
sector industries.  BLS published 10 additional industries in 1991,
18 in 1992, 18 in 1993, and the final 9 in May 1994 for a total of 55
new industries and over 300 new employment, hours, and earnings data
series. 


   ACCURACY OF PAYROLL EMPLOYMENT
   ESTIMATES
-------------------------------------------------------- Appendix VI:3

Accuracy of payroll employment estimates is the ability of the
"preliminary" monthly employment estimate, which is based on the
initial returned sample reports, to be a precise and exact measure of
the "final" estimate, which is produced 2 months later, after all
sample reports have been received and processed. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix VI:3.1

Payroll employment estimates, which BLS publishes from CES program
data, are a major indicator of economic activity in the United
States.  According to BLS officials, sizeable differences between the
preliminary and final estimates--revisions--were caused by the
significant lag time under mail data collection, which resulted in
the preliminary monthly estimates usually being based on only 50
percent of the sample.  According to BLS officials, in recent years,
revisions to payroll employment data of over 100,000 employed persons
have proven unacceptable to users and clearly limited the usefulness
of the data for policymaking purposes.  Between 1981 and 1990, almost
every year had more than 2 months revisions to payroll employment
data greater than 100,000 employed persons. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix VI:3.2

To improve the accuracy of BLS' preliminary monthly estimates of
payroll employment, the working group focused on the adoption of
automated data collection techniques.  These techniques would reduce
reporting time lags and thereby achieve substantial reductions in the
size of the monthly revisions.  The working group noted that
policymakers could then be provided with timely and more accurate
information on the direction of the economy. 


      AGENCY PLANS
------------------------------------------------------ Appendix VI:3.3

To decrease by two-thirds the number of monthly revisions to payroll
employment data over 100,000 employed persons, BLS planned to
introduce automated data collection techniques.  To get the
employment data faster, BLS planned to have all large employers with
more than 50 employees and a sample of smaller employers provide
information to BLS via a computer-assisted telephone interview (CATI)
system and an automated Touch-Tone data entry system. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix VI:3.4

BLS requested slightly more than $13 million and actually received
about $14 million total for fiscal years 1990 through 1994. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix VI:3.5

In 1992, BLS created two new regional data collection centers to
support automated data collection.  BLS officials said that the
conversion to automated data collection is progressing on schedule
and will be fully implemented during fiscal year 1996.  Due to fiscal
year 1993 funding reductions, the scope of the program was
temporarily reduced to a sample of only late-reporting employers with
more than 50 employees, with no employers with fewer than 50
employees being included.  This reduced by nearly one-half the number
of employers included in the effort from the planned 160,000
employers to 90,000.  However, in fiscal year 1994, BLS received
sufficient additional funding to return to the original full scope of
the program. 

According to BLS, this recommendation is proving highly successful
with the size of monthly revisions in payroll employment currently
being reduced by about 40 percent.  In addition, between 1991 and May
1994, there was only one monthly revision of payroll employment that
was greater than 100,000 employed persons--a 96-percent reduction
compared to the average for the preceding 10-year period. 


   AUTOMATED DATA COLLECTION FOR
   CURRENT POPULATION SURVEY
-------------------------------------------------------- Appendix VI:4

Automated data collection for CPS involves interviewers gathering
information with a computer, using an automated survey instrument
rather than a printed questionnaire, as well as using improved data
processing procedures. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix VI:4.1

The CPS questionnaire had last been changed in 1967; since then rapid
changes in the U.S.  economy had generated demand for improved
measures of labor force activity.  Although the National Commission
on Employment and Unemployment Statistics made recommendations in
1979 for improving statistical measures, no major changes were made
immediately.  Later, in 1986, Census and BLS convened an interagency
group that developed an extensive plan to overhaul CPS during the
1990s, taking the above into account.  Among other improvements, the
plan included redesigning the questionnaire to take into account
cognitive interview techniques and using automated collection
procedures to reduce measurement error and to improve data quality. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix VI:4.2

In 1991, the working group noted that CPS information was collected
via printed questionnaires, and the working group recommended the use
of automated data collection techniques.  This recommendation aimed
to improve employment estimates by permitting the administration of a
more comprehensive questionnaire, introducing more efficient editing
and data verification procedures during the interview, and reducing
coding errors. 


      AGENCY PLANS
------------------------------------------------------ Appendix VI:4.3

BLS, along with Census, planned to redesign the CPS questionnaire for
use by field staff on laptop and centralized computers and to better
measure aspects of the population and improve the longitudinal
capability of the survey.  BLS' plan included purchasing laptop
computers, designing a central control system for the integrated
computer-assisted interviewing network, and studying the effects of
automated data collection. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix VI:4.4

BLS requested $5 million and actually received more than $3 million
total for fiscal years 1990 through 1994.\5


--------------------
\5 According to BLS officials, their original funding request was
reduced in fiscal years after 1992.  They said that their subsequent
budget requests were intended to fund implementation of the
recommendation at this reduced level, rather than fund the
recommendation as originally envisioned. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix VI:4.5

Because of limited funding, BLS officials said they were only able to
buy laptop computers and provide limited staff training.  From July
1992 through December 1993, BLS and Census conducted a test of the
new questionnaire on 12,000 households, with data collection via
laptop computers.  The test was conducted concurrently with the
regular paper-and-pencil version of CPS to assess the differences
between the two methodologies.  BLS said in November 1993 that, based
on the results of this test, the changes in the questionnaire and the
use of automated data collection techniques may result in an increase
of 0.5 percentage point in the unemployment rate from prior
estimates.  Beginning in January 1994, BLS began using the redesigned
questionnaire as the basis for its estimates.  Concurrently, it also
began a test, which ran through May 1994, using printed
questionnaires in 12,000 homes to continue monitoring the
differences. 


   RECONCILIATION OF EMPLOYMENT
   ESTIMATES
-------------------------------------------------------- Appendix VI:5

Reconciliation of employment estimates is the process of making
employment data as conceptually consistent as possible between CPS
and CES. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix VI:5.1

Data from CPS and CES differ from each other because of differences
in definitions and coverage, sources of information, methods of
collection, and estimating procedures.  The two surveys supplement
each other, with each survey providing significant types of labor
force information that the other is not designed to supply. 
According to NABE, the magnitude of the differences and,
occasionally, the direction of movement, have implied that the
problems of reconciling differences in employment estimates extend
beyond conceptual differences. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix VI:5.2

The working group recommended that BLS continue its efforts to
reconcile and reduce discrepancies between the employment series
arising from the CPS and the CES surveys. 


      AGENCY PLANS
------------------------------------------------------ Appendix VI:5.3

BLS did not make specific plans for this recommendation.  Officials
did not regard this as a recommendation that required a separate
plan, but said that it was integral to their overall efforts to
improve CES and CPS. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix VI:5.4

BLS did not request funding for this recommendation. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix VI:5.5

BLS officials said their efforts to improve CES and CPS should reduce
the discrepancy in employment estimates between the two by about
two-thirds.  BLS officials said the discrepancy will be more fully
understood as data become available from the redesigned CPS
questionnaire. 


   SUMMARY
-------------------------------------------------------- Appendix VI:6

A summary of the funding requested and actually received by BLS for
the working group's five labor market statistics recommendations is
shown in table VI.1.  BLS requested about $29 million total for four
recommendations for fiscal years 1990 through 1994.  BLS received
slightly more than $26 million for four of the labor market
recommendations in fiscal years 1990 through 1994. 



                                                                       Table VI.1
                                                        
                                                        Funding for BLS' Labor Market Statistics
                                                         Recommendations for Fiscal Years 1990-
                                                                          1994

                                                                 (Dollars in thousands)


                      Requeste           Requeste
Recommendation               d   Actual         d   Actual    Requested     Actual   Requested     Actual   Requested     Actual   Requested     Actual
--------------------  --------  -------  --------  -------  -----------  ---------  ----------  ---------  ----------  ---------  ----------  ---------
Employment cost              0        0      $350     $350         $450       $450        $450       $450        $450       $450      $1,700     $1,700
 index
Coverage of payroll          0        0       600      600        4,260      2,770       2,150      2,150       1,938      1,938       8,948      7,458
 employment
 estimates
Accuracy of payroll          0        0       500      500        4,400      4,400       4,850      3,650       3,544      5,244      13,294     13,794
 employment
 estimates
Automated data               0        0         0        0        3,000      2,000       1,000      1,000       1,000        510       5,000      3,510
 collection for
 Current Population
 Survey
Reconciliation of            0        0         0        0            0          0           0          0           0          0           0          0
 employment
 estimates
=======================================================================================================================================================
Total                        0        0    $1,450   $1,450      $12,110     $9,620      $8,450     $7,250      $6,932     $8,142     $28,942    $26,462
-------------------------------------------------------------------------------------------------------------------------------------------------------
Source:  BLS. 


INCOME AND POVERTY STATISTICS
RECOMMENDATIONS
========================================================= Appendix VII

Income and poverty statistical measures are used together to measure
the economic well-being of families.  A family's total money
income--the sum of money wages and salaries, net income from
self-employment, and income other than earnings--is compared with
poverty thresholds--income levels that signify that all members of
the family are poor when the family's total money income equals or
falls below these levels. 

The working group said that statistical measures of real family
income and poverty drive political debates and influence decisions
about social and welfare policies.  Yet, the poverty thresholds are
based on data collected in the 1950s and 1960s.  Although most major
statistical data series are revised every 5 years to reflect current
prices, consumption, and production patterns, the concept of the
official poverty thresholds has not been significantly revised since
its development in the mid-1960s.  The working group also noted that
the official statistical measures of real family income and poverty
use a version of the Consumer Price Index (CPI), which does not use
the rental equivalence approach to measuring shelter services for the
years before 1983, when the CPI officially converted to this
methodology.  In 1990, the working group charged a subgroup on income
and poverty measurement to develop a process and research agenda that
would review current and alternative measures of income and poverty. 


   POVERTY THRESHOLDS
------------------------------------------------------- Appendix VII:1

Poverty thresholds are income cutoff levels that vary by family
characteristics, such as number of children, age of householder, and
family size.  For example, the poverty threshold was $11,631 for a
three-person family with one child in 1993 and $11,642 for one adult
with two children.  Families with incomes equal to or lower than the
thresholds, which are updated annually for inflation, are considered
poor.\1


--------------------
\1 See the glossary for definitions and descriptions of technical
terms and phrases that are used throughout this appendix. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix VII:1.1

Poverty thresholds had their origins in 1955 when the average family
spent one-third of its income on food.  To determine poverty
thresholds for families of three or more persons, the U.S. 
Department of Agriculture's lowest cost food plan was multiplied by
three.  Experts\2 view food consumption data that underlie the
thresholds as outdated.  According to Patricia Ruggles, other
aspects, such as family size categories and price indexes used to
annually update the thresholds, are also questionable. 


--------------------
\2 Patricia Ruggles, Drawing the Line:  Alternative Poverty Measures
and Their Implications for Public Policy (Washington, D.C.:  The
Urban Institute Press, 1990), pp.  2-3. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix VII:1.2

In 1990, the working group said that the statistical measures of
poverty may not be well suited to measure poverty in the 1980s or
1990s.  The working group recommended that research be initiated to
develop a new benchmark estimate of poverty appropriate to prices,
consumption patterns, and family composition in the 1990s. 


      AGENCY PLANS
----------------------------------------------------- Appendix VII:1.3

In response to the working group's 1990 recommendation, the subgroup
on income and poverty measurement developed a plan to address this
recommendation.  It was presented to the ESI working group in 1990. 
The poverty thresholds recommendation, however, was not included
among the working group's recommendations in 1991, and Census did not
develop a plan to address the recommendation. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix VII:1.4

Census did not request funding for this recommendation.  Census
officials, however, said that the plan developed by the subgroup on
poverty and income measurement would have cost over $10 million for
fiscal years 1992 through 1995.\3


--------------------
\3 Funding of $600,000 was appropriated to BLS to fund a National
Academy of Sciences (NAS) panel to study the measurement of poverty;
$400,000 was then forwarded from BLS to Census who contracted with
NAS to conduct the study.  Census officials told us that the
objectives of the NAS study panel were different from those outlined
by the subgroup.  The NAS panel was also tasked to assess the
conceptualization and measurement issues to develop a methodology to
determine standards for minimum welfare benefits, for which $400,000
was appropriated to the U.S.  Department of Health and Human
Services.  The total contract was $800,000 for both studies on
poverty measurement and the development of methodologies for
standards for minimum welfare benefits.  A report from the panel was
due in 1994; it was released April 30, 1995.  See Constance F.  Citro
and Robert T.  Michael, eds., Measuring Poverty:  A New Approach,
Committee on National Statistics, National Research Council
(Washington, D.C.:  National Academy Press, 1995). 


      PROGRESS REPORTED
----------------------------------------------------- Appendix VII:1.5

Census officials told us that no changes or improvements were made. 
They said, however, that Census staff have begun limited research
studies on experimental poverty thresholds. 


   EXPERIMENTAL ESTIMATES OF
   INCOME AND POVERTY
------------------------------------------------------- Appendix VII:2

Experimental estimates of income and poverty are statistical measures
other than those officially used to determine poverty status.  This
recommendation specifically addresses the inclusion of noncash
benefits--such as public housing, food stamps, and
government-provided health care--as income, and annual adjustments of
income and poverty levels with CPI-U-X1.\4


--------------------
\4 CPI-U-X1 is an experimental CPI that BLS developed for researchers
who wish to make historical comparisons with the current CPI, which
uses the rental equivalence approach to measuring shelter services. 
Prior to 1983, the measurement of homeowner costs in CPI included
changes in the asset value of homes.  The rental equivalence approach
isolates the consumption from the investment aspects of
homeownership. 


      IMPETUS FOR RECOMMENDATION
----------------------------------------------------- Appendix VII:2.1

The determination of poverty is based solely on money income. 
Poverty determination does not reflect the fact that many low-income
persons receive noncash benefits.  And, until 1989, it did not adjust
the thresholds with CPI-U-X1 for the years before CPI-U-X1's official
use in 1983.  According to the National Association of Business
Economists' (NABE) Statistics Committee,\5 the absence of these
adjustments, as well as other problems, make the measurement of
poverty subject to significant error. 


--------------------
\5 Improving Federal Statistics:  Hearings Before the Joint Economic
Committee of Congress of the United States.  102nd Cong., 1st Sess.,
pp.  67-127 (1991) (statement by Martin Fleming, Chairperson of the
National Association of Business Economists' Statistics Committee). 
The statement reflects the views of NABE's Statistics Committee. 


      WORKING GROUP'S
      RECOMMENDATIONS
----------------------------------------------------- Appendix VII:2.2

In 1990, the working group recommended that experimental estimates of
real family income and poverty, which use CPI-U-X1 consistently
across all years, be published alongside the official estimates. 
These experimental estimates were to be published until the poverty
thresholds were revised.  (See poverty thresholds recommendation
above.) The working group also recommended the continued publication
of experimental estimates of real income and poverty that include
noncash benefits.  These were to be published in the same report as
the official estimates. 


      AGENCY PLANS
----------------------------------------------------- Appendix VII:2.3

Census did not make any plans because it was already publishing the
experimental estimates.  In 1989, Census began publishing poverty
estimates and trend data of estimates of family income that use
CPI-U-X1.  Census continues to publish these estimates of family
income and poverty that are adjusted with CPI-U-X1.  Also, Census is
publishing income and poverty estimates with noncash benefits, as
directed by the U.S.  Senate in 1980.\6


--------------------
\6 Department of State, Justice, and Commerce, The Judiciary and
Related Agencies Appropriation Bill, 1981.  U.S.  Senate, 96th
Congress, 2nd Session.  (Sept.  16, 1980), pp.  33-34. 


      FUNDING REQUESTED AND
      RECEIVED
----------------------------------------------------- Appendix VII:2.4

Census did not request funds for this recommendation since it was
already publishing experimental estimates of income and poverty. 


      PROGRESS REPORTED
----------------------------------------------------- Appendix VII:2.5

Census officials said that Census was publishing its experimental
estimates before the working group made its recommendations. 
Although the experimental estimates of family income, which are
adjusted with CPI-U-X1, have been published alongside the official
family income estimates, the experimental family income and poverty
estimates that include noncash benefits and the experimental poverty
estimates that are adjusted with CPI-U-X1, have been published
separately from the official estimates of poverty. 


   SUMMARY
------------------------------------------------------- Appendix VII:3

As shown in table VII.1, Census did not request funds for either the
income or poverty statistics recommendations. 



                                                                      Table VII.1
                                                        
                                                         Funding for Census' Income and Poverty
                                                         Statistics Recommendations for Fiscal
                                                                    Years 1990-1994

                                                                 (Dollars in thousands)


                      Requeste           Requeste
Recommendation               d   Actual         d   Actual    Requested     Actual   Requested     Actual   Requested     Actual   Requested     Actual
--------------------  --------  -------  --------  -------  -----------  ---------  ----------  ---------  ----------  ---------  ----------  ---------
Poverty thresholds\a         0        0         0        0            0          0           0          0           0          0           0          0
Experimental                 0        0         0        0            0          0           0          0           0          0           0          0
 estimates of income
 and poverty
=======================================================================================================================================================
Total                        0        0         0        0            0          0           0          0           0          0           0          0
-------------------------------------------------------------------------------------------------------------------------------------------------------
\a This recommendation was dropped following a working group meeting
held after July 1990.  A portion of the work envisioned by a subgroup
on income and poverty was funded separately through a $600,000
appropriation to BLS.  BLS then passed $400,00 to Census who then
contracted with the National Academy of Sciences to organize a study
panel on the measurement of poverty. 

Source:  Census. 


INTERNATIONAL TRANSACTIONS
STATISTICS RECOMMENDATIONS
======================================================== Appendix VIII

International transactions statistics are measures that provide
information about the economic transactions between the residents of
different nations.  For example, the U.S.  current account reports
the transactions of goods, services, investment income, and
unilateral transfers between residents of the United States and
residents of the rest of the world during a particular period. 

The working group placed the need for improving international
transactions statistics in a similar context as improvements needed
for service sector statistics and the National Income and Product
Accounts (NIPA).  The working group noted in 1990 that U.S. 
merchandise trade statistical measures had improved, and Census had
resumed seasonal adjustment of its monthly trade data.  According to
the working group, however, additional information was required as
international trade gained importance in the economy.  The working
group made several recommendations to address issues in the
international economic accounts and to make improvements in import
and export data. 


   TRADE IN SERVICES
------------------------------------------------------ Appendix VIII:1

Trade in services refers to economic transactions in transportation,
travel, and other business, personal, and government services between
the residents of the United States and the residents of the rest of
the world. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix VIII:1.1

According to F.  Thomas Juster, Chairperson of the American Economic
Association Committee on the Quality of Economic Statistics, data on
international transactions in services were inadequate in 1988.\1 In
1989, the U.S.  Office of Technology Assessment (OTA) reported that
services produced by a firm's foreign subsidiaries are particularly
difficult to value, especially as more firms conduct more of their
business internationally.\2


--------------------
\1 F.  Thomas Juster, "The State of U.S.  Economic Statistics: 
Current and Prospective Quality, Policy Needs, and Resources," (Paper
prepared for the 50th anniversary conference, Conference on Research
in Income and Wealth, Washington, D.C.:  May 12-14, 1988). 

\2 Statistical Needs for a Changing U.S.  Economy, Background Paper,
U.S.  Office of Technology Assessment, Pub.  No.  OTA-BP-E-58 (Sept. 
1989), p.  20. 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix VIII:1.2

In 1990, the working group noted the improvements that had been made
in estimates of trade in services and recommended the acceleration of
additional improvements.  The working group stated that additional
accuracy and detail on trade in services would be valuable for trade
policy discussions.  Following the work plan developed by the
subgroup on service-sector statistics, in 1991, the working group
specifically recommended that the Bureau of Economic Analysis (BEA)
undertake surveys of bank and nonbank financial institutions'
noninterest service income and improve its survey of international
trade in other services. 


      AGENCY PLANS
---------------------------------------------------- Appendix VIII:1.3

BEA planned to (1) develop and conduct a new survey of financial
services to estimate noninterest income for U.S.  and foreign banks
and nonbank financial institutions, (2) strengthen methodologies to
estimate several types of services for which survey data are
incomplete or not available, and (3) in conjunction with Census,
develop monthly estimates of trade in goods and services. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix VIII:1.4

For both of the recommendations to improve balance of payments and
international investment data, which included the trade in services
and international investment and capital flows recommendations, BEA
requested more than $7 million and actually received slightly more
than $3 million total for fiscal years 1990 through 1994.\3


--------------------
\3 BEA officials provided combined budget information for the two
recommendations. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix VIII:1.5

BEA officials said that because they did not get the requested
funding, BEA was unable to conduct a new survey of financial services
to estimate noninterest income for U.S.  and foreign banks and
nonbank financial institutions.  BEA noted, however, that a draft of
a questionnaire for a financial services survey is under review by
data users and financial services respondents.  According to BEA
officials, the only improvement BEA actually attempted in the area of
noninterest income financial services was asking nonbank financial
service firms to report information such as underwriting and
commissions fees on existing surveys.  BEA officials said this effort
was not very successful.\4

BEA, however, has made other methodological improvements for
estimating some services.  BEA reported that it

  began using Mexico's and Canada's data on travel payments and
     receipts,

  introduced estimates for international cruise transactions in the
     North American market and for collecting information on
     interline settlements (i.e., the amounts foreign airlines pay to
     U.S.  airlines when passengers purchasing tickets on foreign
     airlines must transfer to a U.S.  airline to reach their final
     destination),

  improved its methodology for estimating U.S.  rail carriers'
     revenues for transporting foreign-owned goods shipped in transit
     through the United States to be reexported to their final
     destination,

  showed the quarterly balance on goods and services as a separate
     line item, and

  added several international services categories to the annual
     services survey and increased the coverage of the annual and
     benchmark surveys of selected services transactions with
     unaffiliated foreigners by lowering the threshold for mandatory
     reporting.\5

BEA also developed monthly estimates of international services that
were included in a new BEA and Census monthly release on goods and
services. 


--------------------
\4 In subsequent comments, BEA officials said that the reprogramming
of existing activities permitted BEA to develop a benchmark survey of
financial services performed by bank and nonbank institutions.  The
survey covered receipts and payments in 10 separate types of
financial services, the results of which will be incorporated into
the accounts when they become available.  The results will permit BEA
to improve its estimates of noninterest income received and paid by
bank and nonbank institutions.  BEA is now in the process of
designing an annual follow-on survey to the benchmark survey. 

\5 For the agency plan for this improvement, see data gaps
recommendation in appendix III. 


   INTERNATIONAL INVESTMENT AND
   CAPITAL FLOWS
------------------------------------------------------ Appendix VIII:2

International investment is purchasing the ownership of assets-- such
as securities; businesses; titles to land, buildings, or equipment;
and bank deposits--in a foreign country by a private individual,
business, institution, or government.  It also may be thought of as
the export of capital from one country to another.  International
capital flows occur when residents in one country (the capital
exporters) extend loans to, or purchase the title to assets from the
residents of another country (the capital importers).  Such flows of
capital compose the U.S.  capital account.  The capital and current
accounts together constitute the U.S.  balance of payments, which is
the statistical summary of all the country's international
transactions.\6


--------------------
\6 See the glossary for definitions and descriptions of technical
terms and phrases that are used throughout this appendix. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix VIII:2.1

According to the National Academy of Sciences (NAS), of all the U.S. 
international economic transactions, international investment and
capital flows statistical measures are most subject to errors and
gaps.  NAS notes that although the United States collects much
detailed information on its capital flows with the rest of the world,
the "explosion in direct and portfolio investments\7 across U.S. 
national boundaries in the 1980s outpaced improvements in the
statistical system that monitors them."\8

According to the National Association of Business Economists' (NABE)
Statistics Committee,\9 the system's inability to adequately measure
the rapid pace of financial innovations and increasing integration of
world capital markets has resulted in large discrepancies in the U.S. 
balance of payments.  This discrepancy occurs when the balance of the
current account is not offset by the balance of the capital account. 
For example, the NABE committee noted that the statistical
discrepancy between the current and capital account was $50 billion
for the first 6 months of 1990. 


--------------------
\7 Portfolio investments are investments in financial markets; direct
investments are investments in foreign businesses where the investor
has a controlling interest.  For more detailed definitions, see the
glossary. 

\8 Anne Y.  Kester, ed., Behind the Numbers:  U.S.  Trade in the
World Economy, Panel on Foreign Trade Statistics, Committee on
National Statistics, National Research Council (Washington, D.C.: 
1992), p.  156. 

\9 Improving Federal Statistics:  Hearings Before the Joint Economic
Committee of Congress of the United States.  102nd Cong., 1st Sess.,
pp.  67-127 (1991) (statement by Martin Fleming, Chairperson of the
National Association of Business Economists' Statistics Committee). 
The statement reflects the views of NABE's Statistics Committee. 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix VIII:2.2

In 1990, the working group recommended using market values or
replacement cost rather than historical cost to estimate direct
investments.  As well as being an important conceptual improvement,
the working group said that the change was necessary for consistency
with other parts of the national accounts.  The working group noted
that efforts should be made to improve compliance with reporting
requirements for BEA's direct investment survey.  The working group
also recommended that problems with the measurement of international
portfolio investment and other capital flows be addressed. 

In 1991, the working group again recognized the difficulty that the
U.S.  statistical system was having in keeping up with the volume of
transactions in the integrated world capital markets.  The working
group recommended the improvement of coverage of capital flows and
investment income.  Such improvements would help reduce the large
statistical discrepancy in the international payments accounts.\10


--------------------
\10 Unlike the 1990 recommendations, the 1991 recommendations did not
address the problems of direct investment valuation or compliance
with the direct investment survey.  However, BEA's plans for
implementing the international investment and capital flows
recommendation addressed both of these problems. 


      AGENCY PLANS
---------------------------------------------------- Appendix VIII:2.3

BEA planned several improvements to the data on international
investment and capital flows.  BEA's plans included (1) expanding and
improving the data collection system on foreign direct investment in
the United States, which included efforts to link data with Census
(see business establishment lists - Census and BEA recommendation in
app.  III); (2) improving methodologies for estimating direct
investment capital flows from firms that do not report or are exempt
from reporting; (3) changing from estimating direct investment on a
historical-cost basis to a current replacement-cost and market value
basis as part of its revision of the international economic accounts;
(4) updating and improving its methodologies for estimating several
types of capital-related income flows for portfolio investment; and
(5) working with the Department of the Treasury to conduct a new
benchmark survey of U.S.  portfolio investment abroad. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix VIII:2.4

For both of the recommendations to improve balance of payments and
international investment data, which included the international
investment and capital flows and trade in services recommendations,
BEA requested more than $7 million and actually received slightly
more than $3 million total for fiscal years 1990 through 1994.\11


--------------------
\11 BEA officials provided combined budget information for the two
recommendations. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix VIII:2.5

Despite the lack of full funding, BEA officials said that they made
numerous and significant improvements in areas where there were
severe gaps in coverage.  The following are the improvements BEA
officials said the agency made in balance of payments and
international investment data.  BEA officials said that BEA

  improved coverage of U.S.  holdings in foreign banks by using
     banking data from the United Kingdom, Canada, Netherlands,
     Germany, France, Italy, U.S.-owned banks in the Caribbean, and
     the Bank for International Settlements; \12

  with funding to implement provisions specified in the Foreign
     Direct Investment and International Financial Data Improvements
     Act of 1990 (P.L.  101-533), linked its data on foreign-owned
     companies with Census and BLS data on U.S.  plants and expanded
     the number of industries for foreign-owned companies from 135 to
     800; (also, see business establishment lists - Census and BEA
     recommendation in app.  III.)

  introduced improved estimates of personal remittances by
     immigrants;\13

  improved estimation methodologies for portfolio investment income;

  improved its procedures for following up on late respondents to the
     foreign direct investment surveys, which helped improve data
     quality by restoring the compliance rate for large firms to the
     previous levels after declines in 1987 and 1988;

  revalued the U.S.  international investment position from a
     historical-cost to both a market-value and a replacement-cost
     basis;

  estimated earnings on a replacement-cost basis through the
     introduction of a capital consumption adjustment and removal of
     capital gains;

  improved estimates of portfolio capital flows to capture data on
     transactions between U.S.  pension and mutual funds and foreign
     residents;

  provided funding for and assisted in the design of Treasury's
     benchmark survey of U.S.  portfolio investment abroad for March
     1994;\14

  developed new methodologies for estimating some components of
     portfolio investment income, which corrected much of the
     underestimation of these components that previously existed;
     also, reported that until the outbound portfolio investment
     benchmark survey is completed, country and area investment
     position estimates will continue to contain significant
     errors;\15 and

  expedited the completion of Treasury's benchmark survey of foreign
     portfolio investment in the United States. 


--------------------
\12 For the agency plan for this improvement, see data gaps
recommendation in appendix III. 

\13 For the agency plan for this improvement, see data gaps
recommendation in appendix III. 

\14 BEA officials said that the benchmark survey of portfolio
investment abroad was not begun in 1993 because of the lack of
funding and, in part, was delayed to comply with respondents'
requests that it be conducted 3 months after the end of the year. 

\15 For the agency plan for this improvement, see data gaps
recommendation in appendix III. 


   RECONCILIATION OF IMPORT AND
   EXPORT DATA
------------------------------------------------------ Appendix VIII:3

Reconciliation of import and export data is the process of
identifying and quantifying the reasons for discrepancies between the
bilateral trade statistics published by the United States and its
trading partners. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix VIII:3.1

According to Census officials, as a result of tariffs and quotas,
import data are collected more carefully than export data.  The
officials noted, however, that the accuracy of import data may
diminish as the United States moves toward free trade with other
nations. 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix VIII:3.2

Recognizing the effort to reconcile Canadian import data with U.S. 
export data, which it found to yield cost savings and an improvement
in accuracy, the working group recommended the extension of efforts
to reconcile import and export data with Mexico, the European
Community, South Korea, and Japan. 


      AGENCY PLANS
---------------------------------------------------- Appendix VIII:3.3

On the basis of the success of reconciling international trade data
with Canada, Census proposed doing similar reconciliations with other
countries, including Mexico, the European Community, South Korea,
Australia, and Japan. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix VIII:3.4

For all four recommendations for improvements to foreign trade
statistics, which included the reconciliation of import and export
data, automation of import and export data, merchandise exports
model, and access to trade data recommendations, Census requested
more than $1 million total for fiscal years 1990 through 1994, but no
funds were received.\16


--------------------
\16 Census officials provided combined budget information for the
four recommendations. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix VIII:3.5

Census officials said that they have completed reconciliations on
1989 and 1990 data with Japan, 1989 and 1991 data with South Korea,
1989 data with the European Community, and 1991 data with Australia. 
Census is still working on initial reconciliations with Mexico and
China and is continuing its reconciliation efforts with Japan, Korea,
Australia, and the European Union for 1992 data. 


   AUTOMATION OF EXPORT AND IMPORT
   DATA
------------------------------------------------------ Appendix VIII:4

Automation of export and import data involves gathering data
electronically, which allows flexibility for data users without an
increase in respondent burden. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix VIII:4.1

The automation of import data began in 1983 when the U.S.  Customs
Service, Department of the Treasury, began the implementation of its
Automated Commercial System (ACS).  ACS allows importers to submit
information, including statistical information, electronically to
Customs.  Census began receiving export data on magnetic tape from
exporters in the early 1970s.  By 1989, 70 percent of import
transactions and 15 percent of export data were received
electronically.  Census believed that the automation of import and
export transactions helped it process import and export data more
efficiently.  To further increase efficiency, Census wanted to
increase the percentage of import and export transactions received
electronically. 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix VIII:4.2

In 1990, the working group noted that increased automation and other
improvements reduced the monthly carryover of unprocessed data from
44 percent of imports in 1985 to less than 1 percent in 1988.  This
reduction allowed the resumption of seasonal adjustment of the trade
data.  The working group recommended the continuation of efforts to
increase automation of export and import data collection. 


      AGENCY PLANS
---------------------------------------------------- Appendix VIII:4.3

In 1990, plans were already in place for automating export and import
data.  Census had an ongoing automated program for export data, and
Customs had an ongoing automated program for import data.  Census
continued to encourage and work with exporters to bring them into
Census' Automated Export Reporting Program. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix VIII:4.4

For all four recommendations for improvements to foreign trade
statistics, which included the automation of export and import data,
reconciliation of import and export data, merchandise exports model,
and access to trade data recommendations, Census requested more than
$1 million total for fiscal years 1990 through 1994,\17 but no funds
were received.\18


--------------------
\17 Census officials provided combined budget information for the
four recommendations. 

\18 Funding for Customs' ongoing automated program is not included
because Customs does not have a budget for statistical activities,
and therefore did not request ESI funding for this recommendation. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix VIII:4.5

Customs is currently developing, with Census participation, an
automated export system which, if implemented, will replace Census'
automated and manual collection systems.\19


--------------------
\19 According to Census, over 95 percent of imports are now captured
via direct automated submissions into Customs' Automated Commercial
System.  Over 55 percent of export transactions are collected through
automated means; approximately 30 percent are collected through an
automated data exchange with Canada, and another 25 percent are
collected directly from large automated exporters and freight
forwarders. 


   MERCHANDISE EXPORTS MODEL
------------------------------------------------------ Appendix VIII:5

The merchandise exports model is the design and manipulation of a
mathematical representation that simulates the export of merchandise
so that the effect of various international changes can be studied
and forecast. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix VIII:5.1

For years, U.S.  exports to other countries were undercounted.  In a
1988 study, the Federal Reserve Bank of St.  Louis estimated that
U.S.  merchandise exports had been undercounted throughout the period
1960 through 1986.\20 Census officials said that, based on analyses
of total international databases maintained by the United Nations and
Statistics Canada, all exports appear to be understated to some
extent.  Census has attributed the undercount mainly to exporters
failing to properly file export documents with Customs.  Customs does
not strictly enforce requirements that exporters submit documents
accurately describing the type, value, and destination of the goods
that are exported.  Consequently, exporters have little incentive to
report their shipments accurately, or at all. 


--------------------
\20 Mack Ott, "Have U.S.  Exports Been Larger Than Reported?" Federal
Reserve Bank of St.  Louis, (Sept./Oct.  1988), p.  3. 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix VIII:5.2

In 1991, the working group noted that reasonably good estimates of
imports are available because of the revenue-generating aspect of
merchandise imports.  Merchandise exports, however, may have
systematic understatement of value, in part, because exporters want
to reduce import duties for their customers.  The working group
recommended the development of a model to adjust for such
understatements of exports, as well as other improvements in the
measurement of exports. 


      AGENCY PLANS
---------------------------------------------------- Appendix VIII:5.3

According to a Census official, the plan for this recommendation had
two aspects.  The first was to develop a database of exporters that
would allow Census to identify and contact exporters who consistently
submitted inaccurate export declarations.  Census would construct the
database by matching the employer identification number that appears
on the export declaration with the list of employers that Census uses
for its economic censuses.  Census officials said they thought the
database could be used to sample and survey exporters in lieu of
collecting information from export declarations, if a decision were
to be made to switch to surveys for export data collection. 

The second component of this plan was to continue to conduct a series
of port audits that Census could use to estimate the extent of the
export undercount as well as the undervaluation of exports.  The
audits would consist of checking a sample of export declarations
against actual shipments.  Census planned to do audits at seaports
and land ports on the Mexican border. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix VIII:5.4

For all four recommendations for improvements to foreign trade
statistics, which included the merchandise exports model,
reconciliation of import and export data, automation of import and
export data, and access to trade recommendations, Census requested
more than $1 million total for fiscal years 1990 through 1994, but no
funds were received.\21


--------------------
\21 Census officials provided combined budget information for the
four recommendations. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix VIII:5.5

According to Census officials, the exporter database is completed;
however, it needs to be updated periodically.  A Census official said
that exporters who had consistently submitted inaccurate data now
have an average of 20 percent fewer mistakes on their declarations. 
The database has also allowed Census to produce more detailed
breakouts of export data. 

According to Census, no port audits were done after 1989 because of
lack of funding. 


   ACCESS TO TRADE DATA
------------------------------------------------------ Appendix VIII:6

Access to trade data is providing to users a means to obtain
information about the economic transactions between the residents of
different nations. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix VIII:6.1

Congress established the National Trade Data Bank (NTDB) in the
Omnibus Trade and Competitiveness Act of 1988.  The impetus for the
Act was in part a concern that trade and foreign economic data were
available from many sources, but were not easily available to
prospective users.  NTDB was to be the focus of foreign economic and
trade data held by the federal government.  NTDB consolidates trade
information from 15 agencies.  The Office of Business Analysis, U.S. 
Department of Commerce, developed NTDB.  It sells information from
NTDB to businesses and trade analysts.\22


--------------------
\22 We reported in 1991 that NTDB provides considerable information
to many users; however, we also recommended that access to trade data
could be further improved by adding more databases to NTDB.  See
Trade and Economic Data:  Many Federal Agencies Collect and
Disseminate Information (GAO/NSIAD-91-173, May 1, 1991). 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix VIII:6.2

In 1990, the working group recommended increasing the ease of access
to trade data. 


      AGENCY PLANS
---------------------------------------------------- Appendix VIII:6.3

Census planned to continue its program of improving its data
products, such as releasing data on CD-ROM. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix VIII:6.4

For all four recommendations for improvements to foreign trade
statistics, which included the access to trade data, reconciliation
of import and export data, automation of import and export data, and
merchandise exports model recommendations, Census requested more than
$1 million total for fiscal years 1990 through 1994, but no funds
were received.\23


--------------------
\23 Census officials provided combined budget information for the
four recommendations. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix VIII:6.5

Census expanded the availability of its data products on CD-ROM,
which was introduced in 1989, and through electronic bulletin boards. 


   INTERNATIONAL GUIDELINES FOR
   ECONOMIC ACCOUNTS
------------------------------------------------------ Appendix VIII:7

International guidelines for economic accounts are standards
developed by the International Monetary Fund (IMF) for balance of
payments accounting. 


      IMPETUS FOR RECOMMENDATION
---------------------------------------------------- Appendix VIII:7.1

IMF planned to revise its manual for balance of payments accounts.\24
To maintain their consistency with the accounts of other nations, the
U.S.  balance of payments accounts needed to be revised to
incorporate major changes included in the new IMF manual. 


--------------------
\24 International Monetary Fund, Balance of Payments Manual, 5th ed. 
(Washington, D.C.:  1993). 


      WORKING GROUP'S
      RECOMMENDATIONS
---------------------------------------------------- Appendix VIII:7.2

In 1991, the working group recommended the modernization and
extension of the international economic accounts to follow IMF's new
guidelines, which were to be introduced in 1993 and 1994.  The
principal features of the new IMF guidelines were (1) an integrated
set of current and capital accounts, including balance sheets, and
(2) new detail in several policy-oriented and analytically
significant areas. 


      AGENCY PLANS
---------------------------------------------------- Appendix VIII:7.3

BEA planned to move toward making U.S.  balance of payments accounts
consistent with the guidelines in IMF's recently published fifth
edition of the balance of payments manual over the next several
years.  Initially, its work would involve evaluating the guidelines
and then developing plans to adapt the guidelines to the U.S. 
international transactions system.  To achieve consistency with the
revised IMF manual, BEA's plan noted that the United States would
need to introduce a fully integrated system of stocks and flows and a
far more detailed set of accounts for travel, transportation,
construction, insurance, finance, other business services, and other
personal services. 


      FUNDING REQUESTED AND
      RECEIVED
---------------------------------------------------- Appendix VIII:7.4

BEA requested $1 million total for fiscal years 1990 through 1994 for
this recommendation, but received no funding. 


      PROGRESS REPORTED
---------------------------------------------------- Appendix VIII:7.5

Because BEA received no funding for this recommendation, it has not
yet fully conformed the U.S.  international economic accounts to the
guidelines in the fifth edition of the IMF balance of payments manual
issued in 1993.  Even so, in June 1992, BEA implemented several
methodological and presentational changes in its U.S.  international
economic accounts that were consistent with both the fourth and fifth
editions of the IMF's manual.  No progress was made on developing new
source data needed to fully conform with the fifth edition of the
IMF's manual.  BEA removed capital gains and losses and restated
various categories of services on a gross basis.  According to BEA,
these improvements increase consistency between the U.S.  national
and international accounts and between the U.S.  international
accounts and the accounts of other nations. 


   SUMMARY
------------------------------------------------------ Appendix VIII:8

A summary of the funding requested and actually received by BEA and
Census for the working group's seven international transactions
statistics recommendations is shown in table VIII.1.  These agencies
requested $10 million total for the seven recommendations for fiscal
years 1990 through 1994.  BEA received slightly more than $3 million
for the trade in services recommendation and the international
investment and capital flows recommendation for fiscal years 1990
through 1994; Census received no funds. 



                                                                      Table VIII.1
                                                        
                                                         Funding for International Transactions
                                                         Statistics Recommendations for Fiscal
                                                                    Years 1990-1994

                                                                 (Dollars in thousands)


                      Requeste           Requeste
                             d   Actual         d   Actual    Requested     Actual   Requested     Actual   Requested     Actual   Requested     Actual
--------------------  --------  -------  --------  -------  -----------  ---------  ----------  ---------  ----------  ---------  ----------  ---------
Trade in services\a          0        0    $3,500   $2,600       $1,800       $500      $1,300          0      $1,000          0      $7,600     $3,100

 International
 investment and
 capital flows\a
Reconciliation of            0        0       500        0          400          0         500          0           0          0       1,400          0
 import and export
 data\b

 Automation of
 import and export
 data\b

 Merchandise exports
 model\b

 Access to trade
 data\b
International                0        0         0        0          500          0         500          0           0          0       1,000          0
 guidelines for
 economic accounts
=======================================================================================================================================================
Total                        0        0    $4,000   $2,600       $2,700       $500      $2,300          0      $1,000          0   $10,000\c     $3,100
-------------------------------------------------------------------------------------------------------------------------------------------------------
\a BEA combined these two recommendations into one budget request to
improve balance of payments and international investment data. 

\b Census combined these four recommendations into one budget request
to improve foreign trade statistics. 

\c Total request includes $2,300,000 in reinstated requests, that is,
funding that was not received in prior years. 

Sources:  BEA, Census. 


SYSTEMWIDE STATISTICS
RECOMMENDATIONS
========================================================== Appendix IX

Not all statistical measurement problems can be solved solely by one
statistical agency.  Sharing and coordinating statistical data among
statistical agencies is needed when many economic sectors are linked
in an expanding economy.  The systemwide statistics recommendations
are improvements that require statistical agencies to seek
cooperative solutions and reach outside their specific agencies. 

Although some professional groups, such as the American Economic
Association (AEA) and the National Association of Business Economists
(NABE), as well as the U.S.  Office of Technology Assessment (OTA),
have identified problems associated with the decentralized U.S. 
statistical system, the working group did not directly address the
organization of the current federal statistical system or the
long-run future needs of the statistical system.\1 Instead, the
working group identified areas that could be addressed by multiple
agencies in a timely manner, including linking survey data with
administrative data, eliminating data duplication, cooperating with
academic researchers, and creating an academic program to train top
quality professionals in the field. 


--------------------
\1 A subgroup was formed to examine alternative organizational
structures for the federal statistical system.  However, this issue
was not included among the recommendations. 


   SURVEY OF INCOME AND PROGRAM
   PARTICIPATION
-------------------------------------------------------- Appendix IX:1

The Survey of Income and Program Participation (SIPP) provides data
that can be used to study socioeconomic characteristics of the
population.  It also provides comprehensive information on the
economic resources of the U.S.  population and how public transfer
and tax programs affect the financial circumstances of members of the
population.  SIPP is intended to provide policymakers with a means to
measure the efficiency of government tax and transfer programs, for
estimating future program costs and coverage, and for assessing the
effects of proposed policy changes.\2


--------------------
\2 See the glossary for definitions and descriptions of technical
terms and phrases that are used throughout this appendix. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IX:1.1

According to the NABE Statistics Committee,\3 administrative records
could be used to increase SIPP's sampling efficiency, validate survey
data, and supplement SIPP data with administrative data for items
that are difficult to obtain through a survey. 


--------------------
\3 Improving Federal Statistics:  Hearings Before the Joint Economic
Committee of Congress of the United States.  102nd Cong., 1st Sess.,
pp.  67-127 (1991) (statement by Martin Fleming, Chairperson of the
National Association of Business Economists' Statistics Committee). 
The statement reflects the views of NABE's Statistics Committee. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IX:1.2

The working group recognized that microeconomic data needed to be
supplemented with administrative data to develop a better
understanding of the dynamics of the behavior of individuals.  In
1990, the working group stated that the value of the research
information in SIPP would be greatly increased if it were carefully
linked with administrative records.  Recognizing the importance of
confidentiality, the working group recommended that all avenues be
explored to provide such a linkage. 


      AGENCY PLANS
------------------------------------------------------ Appendix IX:1.3

Census did not make plans for this recommendation since this was an
ongoing effort when it was announced in 1990. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IX:1.4

No funds were requested since Census was already linking SIPP with
administrative data. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IX:1.5

Census officials said that SIPP data is being linked with (1) IRS tax
records, (2) food stamps administrative data, (3) Aid to Families
with Dependent Children data, (4) BLS employment and wage data, (5)
social security administrative data, (6) civil service retirement
data, (7) Veteran's Administration data, and (8) Pell grant records. 
Another Census effort, known as the Milwaukee cognitive study, is an
experimental survey to replicate SIPP.  It links the experimental
survey's responses with administrative data to assess the
experimental survey's quality.  Comparing the two sets of data has
allowed Census to determine which survey provides the most accurate
response. 

Census officials said that they safeguard confidentiality through
data sharing agreements.  Documents of the agreements between Census
and SSA indicate that administrative data are to be available only to
persons allowed by Census to have access to confidential data. 


   STANDARD INDUSTRIAL
   CLASSIFICATION
-------------------------------------------------------- Appendix IX:2

The standard industrial classification (SIC) system is a federally
designed system classifying establishments by industry. 
Establishments are allocated into categories according to their
principal economic activity. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IX:2.1

OTA reported that the output of emerging industries, which experience
extraordinary growth, cannot be tracked because they are not
separately identified by the SIC system.\4 Instead, their growth is
lumped into broad categories that obscure the source of the change. 
Even though the SIC system was revised in 1987, OTA reported that
problems still exist.  For example, one category combines all sizes
of computers, from minicomputers to mainframes, while output of
established industries such as footwear is broken down into several
categories.  Although attempts to retain consistency over time are
important, OTA reported that additional detail for large and growing
industries seems warranted. 


--------------------
\4 Statistical Needs for a Changing U.S.  Economy, Background Paper,
U.S.  Office of Technology Assessment, Pub.  No.  OTA-BP-E-58 (Sept. 
1989), p.  13. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IX:2.2

Recognizing rapid changes in industries and growth of new industries,
in 1991, the working group recommended funding to ensure that the SIC
system keeps track of emerging industries and to develop innovative
classification methods to keep up with the rapid changes occurring
across all industries. 


      AGENCY PLANS
------------------------------------------------------ Appendix IX:2.3

Census officials said that they planned to begin research on methods
of identifying new industries, the reasons behind companies switching
industry codes, and structural changes affecting the SIC system. 
Census also planned to hold an international conference to address
the issue of a future SIC system. 

BEA officials said that they, in conjunction with other federal
agencies, made plans for this recommendation following Census'
international conference, which served as the impetus for an
OMB-directed mandate to restructure the SIC system.  BEA, as the lead
agency for OMB's Economic Classification Policy Committee (ECPC),
which was chartered in 1993, planned with BLS and Census to develop
and implement a new SIC system.  The North American Industry
Classification System (NAICS), which will provide a common
production-oriented industry classification system for the United
States, Canada, and Mexico, was planned to replace the current SIC
system in the late-1990s. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IX:2.4

Census requested and received $100,000 total for fiscal years 1990
through 1994 for this recommendation.  BEA officials said that BEA
requested funding for this recommendation under the data gaps
recommendation (see app.  III). 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IX:2.5

Census began research on methods of identifying new industries,
reasons behind companies switching industry codes, and structural
changes affecting the SIC system.  In fiscal year 1991, Census also
held an international conference to address the issue of a future SIC
system.  Census officials said they conducted research on alternative
classification systems and their use, continued with existing SIC
system research, and implemented research results for the 1992
Economic Censuses. 

BEA officials said that ECPC prepared and circulated issue papers on
various aspects of economic classification systems.  ECPC also
carried out an independent evaluation of U.S.  industries to assess
whether industries in the current SIC system met the conditions for
inclusion in the new NAICS.  In July 1994, ECPC requested comments on
various aspects of developing and implementing NAICS. 


   FARM LISTS
-------------------------------------------------------- Appendix IX:3

The census of agriculture collects information on farms through
mailed questionnaires.  To conduct the census, a list of names and
addresses of agricultural operations is compiled before the
questionnaires are mailed.\5


--------------------
\5 For more information on how the census of agriculture is
conducted, see Federal Data Collection:  Status of 1992 Agriculture
and Economic Censuses and Future Challenges (GAO/GGD-93-152BR, Sept. 
23, 1993). 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IX:3.1

Three agencies in different departments maintain lists of farming
operations.  The U.S.  Department of Agriculture's National
Agricultural Statistics Service (NASS) maintains a farm list and an
area agricultural sampling frame for the United States.  According to
NASS, the quality of a farm list can be measured in currency,
completeness, lack of duplication, and coverage.  In the early 1990s,
NASS sampling frame covered 80 percent of all land in farms.  Census
also maintains a listing of farms in the United States.  Census uses
its list, NASS' list, and the list of farm tax filers from the U.S. 
Department of the Treasury's IRS to compile the list used to conduct
Census' quinquennial census of agriculture. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IX:3.2

In 1991, the working group noted that only about 90 percent of actual
farms are included in the census of agriculture; small farms are
undercounted and other sizes are overcounted.  The working group also
noted that a large number of nonfarm establishments are also included
in the census of agriculture.  The working group recommended support
to NASS for a more complete and accurate farm list for the 1992
Census of Agriculture. 


      AGENCY PLANS
------------------------------------------------------ Appendix IX:3.3

In fiscal year 1991, NASS planned to enhance its list of farm
operators, including small farms, by improving the coverage and
quality of its list for commodity estimating programs.  In fiscal
year 1992, NASS planned to update and improve its list and area
agricultural frames to meet new data needs.  These needs included
environmental concerns relating to water quality and food safety,
agricultural labor, and specialty commodities.  NASS also planned to
enhance its list for compilation of Census' farm list used in the
census of agriculture.  As part of a cyclical improvement plan,
designed to match the 5-year census cycle, NASS planned in fiscal
year 1993 to maximize coverage for cropland, livestock inventories,
and specialty commodities.  In fiscal year 1994, NASS planned to
increase coverage for specific commodities by geographic area. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IX:3.4

NASS requested about $5 million and actually received slightly more
than $2 million total for fiscal years 1990 through 1994 for this
recommendation. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IX:3.5

NASS officials said that their efforts improved its farm list by
increasing coverage for the number of farm operations.  Also,
duplicate operations were removed from the list.  The officials said
that more complete and accurate lists increased sampling efficiency,
which improves data quality and lowers survey costs.  In the fall of
1991, NASS provided its improved farm list to Census for the
development of the 1992 Census of Agriculture list.  In the spring of
1992, NASS provided updates, additions, and deletions to the list
previously provided to Census.  NASS officials said that fiscal year
1993 efforts resulted in obtaining list information for large and
specialized operations, which are expected to remain in business over
the long term. 


   BUSINESS ESTABLISHMENT
   LISTS-CENSUS AND BLS
-------------------------------------------------------- Appendix IX:4

Business establishment lists are directories of businesses maintained
by statistical agencies, which are used when the agencies survey
establishments to obtain information about their business practices. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IX:4.1

For their survey programs, Census and BLS maintain separate lists of
business establishments.  In 1987, the Working Group on the Quality
of Economic Statistics to the Economic Policy Council recommended
that legislation be submitted to permit Census to disclose business
identification and classification information to specified
statistical agencies.  The 1987 working group also recommended that
BLS serve as the agency in charge of compiling a common business
establishment list.  The 1987 working group said that coverage of the
lists, as well as the quality and comparability of statistics, would
improve if legislative and administrative barriers to sharing lists
were removed.  In 1989, OTA reported that these efforts were stymied
by a lack of legislation to amend the confidentiality law allowing
BLS access to Census data.  BLS officials said that the failure to
establish a common list was not only due to a lack of legislation,
but because Census determined it needed to maintain its own separate
list.  Census officials said that to move ahead with data-list
sharing, legislative and regulatory changes affecting both Census and
BLS, and IRS for tax data, are needed.  In their view, sharing is a
critical first step in developing a common list.  This is because BLS
and Census lists need to be compared and evaluated to determine their
strengths and weaknesses relative to what may be different agency
needs. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IX:4.2

Although the uses are not entirely identical, the working group noted
that the maintenance of two separate establishment lists by Census
and BLS results in substantial duplication of effort.  According to
the working group, the first step toward reconciling the two lists
was to take place in 1991, when it recommended new research to expand
these efforts to other sectors.  The working group also recommended
that the differences in the information requirements imposed by
Census and BLS be addressed. 


      AGENCY PLANS
------------------------------------------------------ Appendix IX:4.3

Census officials said that while full data sharing would require a
change in legislation, they pursued more opportunities for using BLS
information. 

BLS officials said they planned to improve its Business Establishment
List (BEL).  BLS officials said that an improved BEL would generally
improve BLS' survey programs and thereby result in improved BLS data. 
For this recommendation, BLS planned to obtain better information on
small employers and improve BLS' codes of business establishments. 
BLS' plan also called for conducting long-term longitudinal studies
to track trends, such as comparison of employment growth rates for
small and large employers. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IX:4.4

BLS requested nearly $2 million total for fiscal years 1990 through
1994 for this recommendation, but received $500,000.\6


--------------------
\6 According to BLS officials, their original funding request was
reduced in fiscal years after 1992.  They said that their subsequent
budget requests were intended to fund implementation of the
recommendation at this reduced level, rather than fund the
recommendation as originally envisioned. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IX:4.5

Census officials said that beginning in 1991 it began using BLS codes
to classify previously unclassified and partially classified
manufacturing records.\7 BLS officials said that it made selected
technical improvements, which included improved coding of selected
establishments, installation of exportable computer systems, and
purchase of equipment. 


--------------------
\7 Census officials said that they participated in an interagency
task force that recommended the removal of legal barriers to sharing
business information.  (See elsewhere in this app.)


   COOPERATION
-------------------------------------------------------- Appendix IX:5

With respect to improving the quality and relevance of statistical
measures, this recommendation proposes the cooperation, or the action
of statistical agencies--Census, BEA, and BLS--and academic users, to
work together, and through joint efforts develop and expand upon
concepts and definitions of statistical measures. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IX:5.1

According to F.  Thomas Juster, Chairperson of the AEA Committee on
the Quality of Economic Statistics,\8 the quality and relevance of
statistical measures depends on an active research and development
program.  In lieu of funding such efforts in the statistical
agencies, Juster reported that the development of strong linkages
between the agencies and academic users could possibly be the most
important factor in improving statistical measures. 


--------------------
\8 F.  Thomas Juster, "The State of U.S.  Economic Statistics: 
Current and Prospective Quality, Policy Needs, and Resources," (Paper
prepared for the 50th anniversary conference, Conference on Research
in Income and Wealth, Washington, D.C.:  May 12-14, 1988). 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IX:5.2

In 1990, the working group reported that better measurement concepts
and definitions were potentially some of the most important
improvements that could be made and recommended that Census, BEA, and
BLS expand, accelerate, and coordinate their efforts in improving
measurement concepts and definitions.  To achieve this outcome, the
working group recommended that statistical establishments increase
cooperation with academic researchers. 


      AGENCY PLANS
------------------------------------------------------ Appendix IX:5.3

Census officials said they planned to continue to promote the
availability of the Longitudinal Research Database (LRD) to qualified
researchers through their Center for Economic Studies.\9


--------------------
\9 BLS officials said that they participated in an interagency task
force that established a framework, which was not formally part of
ESI, to take action to implement this recommendation.  (See elsewhere
in this app.)


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IX:5.4

No funds were requested for this recommendation.\10


--------------------
\10 BEA officials said that this recommendation did not require
funding. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IX:5.5

In addition to promoting the availability of LRD, Census' Center for
Economic Studies expanded the scope of LRD and developed plans for
the Research Data Center (RDC).  RDC, which opened in January 1994 at
Census' Regional Office in Boston, provides access to Census'
economic microdata to qualified researchers.  In addition,
researchers played an active role in the international conference on
the SIC system and continue to be active participants in the Census
Advisory Committee of Professional Associations.\11


--------------------
\11 Officials from the statistical agencies and OMB noted the
creation of the Joint Program in Survey Methodology at the University
of Maryland meets some aspects of the cooperation recommendation in
that cooperation with academic researchers traditionally occurs with
such a program.  (For progress reported, see center for survey
methods recommendation below.) BEA officials stated that cooperation
between BEA, BLS, and Census was increased in a number of ways that
are reported elsewhere in this report.  (For example, see trade in
services recommendation in app.  VIII.)


   MANDATORY VS.  VOLUNTARY
   SURVEYS
-------------------------------------------------------- Appendix IX:6

In mandatory surveys, respondents are obligated or authoritatively
ordered to complete a survey questionnaire.  In voluntary surveys,
respondents are not obligated to complete a survey questionnaire;
they have a free choice to answer a survey or not. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IX:6.1

According to the NABE Statistics Committee, the tradition of a low
response rate for voluntary surveys has resulted in the consideration
of mandatory surveys.  NABE said, however, that compliance costs may
be high for businesses that do not have staff available to respond to
mandatory surveys.  According to Census officials, all of their
censuses and virtually all annual surveys are mandatory, as well as a
few of lesser frequency.  However, none of the important monthly
economic indicator series are mandatory.  According to OMB officials,
a large number of the federal surveys with the highest response rates
are voluntary surveys, in part, because the voluntary nature forces
the agency to be sensitive to the needs of the respondent and to
allocate resources to advance notification and follow up with
respondents. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IX:6.2

In 1990, the working group recommended, where necessary, mandatory
surveys for service industries.  The working group also reported that
methods and technologies for reducing cost and respondent burden
should be actively pursued; however, it recognized that the cost of
collection and paperwork would increase.  A comparison study of
mandatory and voluntary surveys that examined accuracy and response
rates was recommended.  To increase compliance and quality of data,
the significance and urgency of the current statistical problems were
to be stressed to business leaders and to small business trade
associations. 


      AGENCY PLANS
------------------------------------------------------ Appendix IX:6.3

In 1987 and 1988, Census planned and implemented activities to
explore the effect of mandatory versus voluntary response conditions
in censuses and surveys.  Census activities included a study and
focus-group discussions of placing a greater emphasis on the
mandatory nature of census responses.  Survey activities included
nonresponse tests in selected annual and monthly manufacturing
surveys. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IX:6.4

No funds were requested for this recommendation. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IX:6.5

Although Census had already conducted a study when the recommendation
was announced in 1990, Census officials said some progress was made. 
For example, according to Census officials, when the mandatory
response requirement was highlighted on mailing envelopes and other
materials for the 1992 Economic Censuses, the response rate rose to
an historically high level (7 percentage points above the 1987 rate). 
Census officials also stated that initial tests in manufacturing
surveys suggest that mandatory responses can substantially increase
response rates, improve reported data quality, and minimize program
costs. 


   DATA DUPLICATION
-------------------------------------------------------- Appendix IX:7

Data duplication involves the collection of the same information from
one source, or respondent, by more than one statistical agency. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IX:7.1

In 1987, the Working Group on the Quality of Economic Statistics to
the Economic Policy Council, a predecessor to the Economics
Statistics Initiative Working Group, recommended that agencies share
business establishment lists.  OTA later reported that such efforts
were stymied by a lack of legislation that would allow one agency to
access another agency's data.\12


--------------------
\12 As noted under the business establishment lists - Census and BLS
recommendation, BLS officials said that the failure to establish a
common business establishment list was not only due to a lack of
legislation, but because Census determined it was necessary for it to
maintain its own separate list.  Census officials said that to move
ahead with data-list sharing, legislative and regulatory changes
affecting both Census and BLS, and IRS for tax data, are needed.  In
their view, sharing is a critical first step in developing a common
list.  This is because BLS and Census lists need to be compared and
evaluated to determine their strengths and weaknesses relative to
what may be different agency needs.

In Management Reform:  GAO's Comments on the National Performance
Review's Recommendations (GAO/OCG-94-1), we agreed with the need for
better data sharing and cooperation among federal statistical
agencies and endorsed the National Performance Review's
recommendations that the administration propose legislation that
would encourage better data sharing between statistical agencies and
would bring about better data sharing of business data. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IX:7.2

In 1990 and 1991, the working group recommended eliminating
unnecessary duplication, but avoiding the loss of unique and
important alternative data.  In 1991, they recommended the
preparation of legislation that would provide a standardized
mechanism for the limited sharing of confidential information solely
for statistical purposes. 


      AGENCY PLANS
------------------------------------------------------ Appendix IX:7.3

With officials of the major statistical agencies, OMB planned to
prepare a legislative proposal that coordinated policy governing the
exchange of confidential data to be exclusively used for statistical
purposes. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IX:7.4

No funds were requested for this recommendation. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IX:7.5

OMB officials said that during the Bush administration legislation
was prepared by OMB that promoted data sharing while maintaining
confidentiality of respondents.  That legislation, however, was not
submitted to Congress.  OMB continues to revise and formulate
data-sharing legislation. 


   CENTER FOR SURVEY METHODS
-------------------------------------------------------- Appendix IX:8

The center for survey methods (Center) was envisioned to be an
institution that would provide training for persons pursuing
graduate-level education in survey methods and statistical science. 


      IMPETUS FOR RECOMMENDATION
------------------------------------------------------ Appendix IX:8.1

According to F.  Thomas Juster, Chairperson of the AEA Committee on
the Quality of Economic Statistics, the academic community has
diminished its ability to provide analysis to support improvements in
government statistics and to provide appropriately trained staff to
the government statistical system because of the academic community's
deficient interest and involvement in empirically based public
policy. 


      WORKING GROUP'S
      RECOMMENDATIONS
------------------------------------------------------ Appendix IX:8.2

In 1991, the working group reported that the federal statistical
system had difficulty in recruiting and retaining sufficient numbers
of high quality statisticians.  The working group noted that the
difficulty was due not only to the pay gap, which exists between the
private and public sectors, but also because universities have tended
to de-emphasize training in the field of survey statistics.  In 1991,
the working group recommended that a Center be created at a
greater-Washington, D.C., area university to improve the talents and
skills of the existing federal statistical workforce and attract
highly qualified individuals into the field of specialization. 


      AGENCY PLANS
------------------------------------------------------ Appendix IX:8.3

The National Science Foundation (NSF) planned to award a contract for
the establishment of the Center that would provide (1) advanced
training and research in survey methods in the social, behavioral,
economic, and statistical sciences and (2) graduate-level education
and training for current and future employees of federal statistical
agencies and for other individuals who will employ survey methods
more effectively in survey practice and in the conduct of research at
academic and other institutions. 


      FUNDING REQUESTED AND
      RECEIVED
------------------------------------------------------ Appendix IX:8.4

NSF requested and received slightly more than $2 million total for
fiscal years 1990 through 1994 for this recommendation.\13


--------------------
\13 Following an open, competitive process, NSF awarded a 5-year,
$4.1 million cooperative agreement, starting in fiscal year 1993, to
a consortium including the University of Maryland at College Park;
the University of Michigan; and Westat, Inc., a private survey
organization in Rockville, MD. 


      PROGRESS REPORTED
------------------------------------------------------ Appendix IX:8.5

NSF officials reported that the Joint Program in Survey Methodology
was established in January 1993 on the campus of the University of
Maryland. 


   SUMMARY
-------------------------------------------------------- Appendix IX:9

A summary of funding requested and actually received by Census, BLS,
NASS, NSF, and OMB for the working group's eight systemwide
statistics recommendations is shown in table IX.1.  The agencies
requested about $9 million and received about $5 million total for
fiscal years 1990 through 1994 for four of the recommendations. 
Funding was not requested for four of the recommendations. 



                                                                       Table IX.1
                                                        
                                                           Funding for Systemwide Statistics
                                                         Recommendations for Fiscal Years 1990-
                                                                          1994

                                                                 (Dollars in thousands)


                      Requeste           Requeste
Recommendation               d   Actual         d   Actual    Requested     Actual   Requested     Actual   Requested     Actual   Requested     Actual
--------------------  --------  -------  --------  -------  -----------  ---------  ----------  ---------  ----------  ---------  ----------  ---------
Survey of income and         0        0         0        0            0          0           0          0           0          0           0          0
 program
 participation
Standard industrial          0        0       375      375        (275)      (275)           0          0           0          0         100        100
 classification
Farm lists                   0        0     2,400    1,200        2,400      1,195           0      (145)           0          0       4,800      2,250
Business                     0        0         0        0        1,400        500         500          0           0          0       1,900        500
 establishment lists
 -Census and BLS
Cooperation                  0        0         0        0            0          0           0          0           0          0           0          0
Mandatory vs.                0        0         0        0            0          0           0          0           0          0           0          0
 voluntary surveys
Data duplication             0        0         0        0            0          0           0          0           0          0           0          0
Center for survey            0        0         0        0          400        400         700        700       1,000      1,000       2,100      2,100
 methods\a
=======================================================================================================================================================
Total                        0        0    $2,775   $1,575       $3,925     $1,820      $1,200       $555      $1,000     $1,000      $8,900     $4,950
-------------------------------------------------------------------------------------------------------------------------------------------------------
\a NSF provides funding to the Center through a cooperative
agreement. 

Sources:  Census, BLS, NASS, OMB, NSF. 


OTHER IMPROVEMENTS TO ECONOMIC
STATISTICS MADE FROM 1990 THROUGH
1994
=========================================================== Appendix X

In addition to the efforts carried out through ESI, Census, BEA, BLS,
the Federal Reserve, and NASS made other improvements to their
economic statistics programs with existing funding between 1990 and
1994.  The following discussion of these efforts includes those that
the agencies reported as ongoing, as well as those that the agencies
reported as completed. 


   CENSUS FOCUSED ON DEVELOPING
   NEW SURVEYS AND MAKING
   TECHNICAL IMPROVEMENTS
--------------------------------------------------------- Appendix X:1

Census officials said they had difficulty differentiating between
activities that were a direct result of the working group
recommendations versus activities previously envisioned.  At the time
of the working group's recommendations, Census had already turned its
attention to improving the quality of economic statistics. 

According to Census officials, during this time the agency focused on
implementing a wide range of improvements, other than those of ESI,
that directly improved the quality of economic statistics.  These
other improvements include expanding the economic censuses, creating
new surveys, expanding data collection, increasing coordination
efforts, and developing computer processing projects.  For example,
Census officials said these improvements allowed the expansion of the
1992 Economic Censuses to cover 98 percent of gross domestic product
(GDP) and earlier release of economic censuses data.\1

To expand the areas covered by Census' surveys, Census planned a
number of new surveys, including three transportation surveys.  In
addition, Census implemented a new annual survey of the communication
sector, a commodity flow survey that measures the transportation of
goods (funded by the Department of Transportation), and new surveys
on manufacturing technology. 

According to Census officials, Census also expanded the amount of
data it was collecting, such as data on nonresidential construction,
women-owned corporations, purchased services, and company-level
information on types of health care plans currently offered for
employees covered and total costs.  Census also began publishing
information from its Quarterly Financial Report program for four
additional industry groups and created a historical country database,
which provided access to bilateral trade information on a monthly
basis from 1985 to the present. 

Census reported that it also (1) entered into a reimbursable
agreement with BEA in order to speed up publication of the
input-output tables; (2) sponsored the Exporter Education Program to
reduce reporting errors; (3) increased interagency cooperation with
the Customs Service, such as providing Census staff information for
the development of the Automated Export System; (4) opened a Puerto
Rican foreign statistics branch to improve the quality and coverage
of U.S.-Puerto Rican trade statistics; (5) implemented a joint
Census/BEA monthly trade release incorporating trade both in goods
and services beginning with January 1994 statistics; (6) conducted a
1992 Affiliation Survey to provide updated industry codes for the
1992 service sector expansion; (7) replaced personally enumerated
area samples with administrative record information; and (8) updated
the 10-year-old sample for the building permits survey. 

According to Census officials, the computer processing projects
undertaken by Census included:  improving its capability for
electronic data collection through electronic data interchange;
implementing a series of computer processing-related foreign trade
statistics improvements; implementing the automation of the data
collection operation for the value of new construction put-in-place
series; and using telephone data entry instruments to collect data
for the manufacturers' shipments, inventories, and orders survey from
more than 1,000 companies. 

Finally, Census officials reiterated that, for several of the ESI
recommendations, they received either no or only partial funding.  As
a result, Census carried out these activities through a reprogramming
of funds or through reimbursable financing.  Among these activities
were the creation of an exporter database for all exporters that
operated during the 1987 Economic Censuses; the participation in a
pilot of Custom's Automated Export System to automate and improve the
quality of export data received by the Customs Service and Census;
the introduction of three monthly import and export CD-ROM releases;
the use of BLS classification codes for unclassified business
establishments; work associated with developing a restructured
economic classification system; and the reconciliation of trade data
with major trading partners. 


--------------------
\1 In September 1993, we reported on this expansion in Federal Data
Collection:  Status of 1992 Agriculture and Economic Censuses and
Future Challenges (GAO/GGD-93-152BR, Sept.  23, 1993). 


   BEA FOCUSED ON MAKING TECHNICAL
   IMPROVEMENTS AND DATA
   DISSEMINATION
--------------------------------------------------------- Appendix X:2

BEA officials said that BEA focused on several low-cost, high-payoff
technical improvements between 1990 and 1994 to improve the quality
of its statistics.  BEA officials noted that none of BEA's
initiatives under ESI were fully funded in any given year and that
they had difficulty in differentiating between activities carried out
under ESI funding and those carried out under normal funding.  BEA
introduced GDP as the featured measure of U.S.  production (rather
than GNP) to begin bringing U.S.  accounts in line with international
guidelines.  In addition, BEA redefined net exports in both NIPA and
the balance of payments accounts in order to identify international
factor incomes separately from services transactions.  BEA also noted
that it undertook its ninth comprehensive (benchmark) revision of
NIPA, making improvements in several areas, including interest paid
by real estate partnerships, tax return misreporting adjustments, and
imputed rental value of farm dwellings. 

BEA officials also noted technical improvements in its regional
accounts, including enhanced procedures that allowed BEA to publish
data on wages and salaries and county employment 12 months after the
close of the reference year (down from 16 months for total personal
income), the comprehensive revision of state income as well as
product estimates, and the introduction of improved methodologies for
journey to work flows, and the incorporation of improved IRS data on
partnerships and proprietors. 

Officials from BEA reported that they are standardizing the
time-series database management system used to prepare NIPA and
introducing a modernized computer network system.  These changes will
improve the review, analysis, and final table preparation. 

Finally, in the area of data dissemination, BEA officials noted that
BEA pioneered the use of CD-ROMs for disseminating its large regional
database, and it participated in both Census' and Commerce's Economic
Statistics Administration's bulletin boards. 


   BLS FOCUSED ON MODIFYING
   SURVEYS AND MAKING TECHNICAL
   IMPROVEMENTS
--------------------------------------------------------- Appendix X:3

BLS officials said during this time BLS focused on modifying existing
surveys to obtain better measures of occupations, wages, and
benefits.  BLS' efforts focused as well on several technical
improvements that BLS officials believe will enhance the adequacy of
BLS' data. 

According to BLS officials, BLS combined its several wage surveys
into the Occupational Compensation Survey Program in response to the
Federal Pay Comparability Act of 1990.  The new survey expands BLS'
coverage of professional and administrative occupations and
industries.  BLS also is in the process of combining its Employment
Cost Index (ECI) survey with its employee benefits program to obtain
a better measure of employee benefits and the costs of these
benefits.  BLS also revised its annual survey of occupational
injuries and illnesses to better help employers learn how to avoid
workplace accidents. 

Technical improvements that BLS reported carrying out included (1)
incorporating a population undercount from the 1990 Decennial Census
into CPS; (2) correcting for an employment overcount as reported by
businesses in BLS' Employment and Wages Program (known as the ES-202
program), as well as enhancing its level of detail by requiring
businesses to report all their different work sites; and (3)
developing a computer system that will decrease the labor resources
required for the ES-202 program.  BLS also automated its Current
Employment Statistics (CES) program, as well as expanded the
publication to include seasonally adjusted employment data at the
state level to better serve its geographic data users.  Finally, BLS
has been sharing standard industrial classification (SIC) system
codes with Census to improve the usefulness and accuracy of Census'
economic censuses and surveys.  BLS officials also noted that, under
the auspices of the Foreign Direct Investment and International
Financial Data Improvements Act, it undertook a project with BEA. 
BLS obtained access to BEA enterprise data on foreign direct
investment in the United States.  It then linked these data to
establishment data from BLS' business list.  BLS published its first
link results--1989 data for all industry divisions--in the summer of
1992. 


   FEDERAL RESERVE BOARD FOCUSED
   ON SOFTWARE IMPROVEMENTS
--------------------------------------------------------- Appendix X:4

Officials from the Federal Reserve said that they are revamping the
computer software for the flow-of-funds accounts.  The modernization
of this software will facilitate staff input into the production
process, taking full advantage of the capabilities of the computer
network system at the Federal Reserve and allowing for more
flexibility and clarity in the published tables. 


   NASS FOCUSED ON IMPROVING FARM
   LISTS
--------------------------------------------------------- Appendix X:5

NASS officials said during this period they focused on making
additional improvements to their farm lists.  However, they noted
these activities were carried out under ESI funding as well as
funding from the Department of Agriculture, and they could not
differentiate between which activities were carried out under which
type of funding.  (For a discussion of NASS' activities under ESI,
see the initiative on farm lists in app.  IX.)


   COOPERATIVE EFFORTS CARRIED OUT
   BY THE AGENCIES
--------------------------------------------------------- Appendix X:6

According to Census officials, on June 3, 1993, Census, BEA, and BLS
established a joint task force with the mutual goal of reducing
respondent burden on the business community.  Specifically, the task
force, called the Statistics 2000 Task Force, sought to identify
opportunities to (1) reduce the reporting burden on business, (2)
eliminate or reduce duplicative data requests, (3) facilitate and
simplify reporting by businesses, (4) reduce agency costs, and (5)
improve the efficiency of federal business statistics.  The task
force identified 34 specific opportunities that focused on improving
data sharing among agencies, improving business registers, expanding
the use of administrative record information, and facilitating
reporting and reducing respondent burden.  Of these 34 opportunities,
the task force determined that the enactment of broad-based
data-sharing legislation was the single, most important activity
required to implement many of the other opportunities identified by
the task force.  Data-sharing legislation was also recommended by the
ESI working group.  (The status of the data-sharing legislation is
discussed in the ESI recommendation on data duplication in app.  IX.)

To date, the task force has focused on 13 of the activities,
specifically those that the task force believes offer initial
benefits in fiscal years 1994 or 1995, or those that are already high
priorities for one or more of the agencies. 


FISCAL YEAR 1995 FUNDS REQUESTED
AND RECEIVED FOR ECONOMIC
STATISTICAL IMPROVEMENTS
========================================================== Appendix XI

BEA, Census, BLS, and NASS requested more than $38 million and
received more than $18 million in additional funding in fiscal year
1995 for economic statistical improvements.\1


--------------------
\1 The Federal Reserve is not included here as it does not receive
appropriated funds from Congress.  In addition, NSF did not request
additional funding for economic statistics improvements.  For fiscal
year 1995, the federal agencies ceased using the phrase "Economics
Statistics Initiatives" in reference to their requests for additional
funding to improve economic statistics. 


   DEPARTMENT OF COMMERCE
   REQUESTED $16.7 MILLION AND
   RECEIVED $1.3 MILLION
-------------------------------------------------------- Appendix XI:1

The Department of Commerce, in which both BEA and Census operate,
requested a total of $16.7 million in budget increases for fiscal
year 1995 to improve the coverage, timeliness, and accuracy of the
statistics for which it is responsible.  Several of the requested
improvements are continuations of those included in ESI.  Commerce
received $1.3 million to continue the efforts begun under the SIC
recommendation and to reengineer the data-gathering process. 


      BEA REQUESTED $8.1 MILLION
------------------------------------------------------ Appendix XI:1.1

With BEA's requested increase of $8.1 million, the Office of the
Undersecretary for Economic Affairs proposed a plan it is calling the
Economic Information Infrastructure Initiative (EI\3 ).  This plan
proposed to improve the nation's economic information infrastructure
in four fundamental ways:  (1) strengthen existing statistical
programs by making statistics more accurate and dependable, (2)
launch new initiatives that reflect new concerns, (3) reengineer the
data-gathering process, and (4) make information a competitive asset. 
As shown in table XI.1, BEA has five initiatives that address the
first three components of this plan.  BEA did not request any funds
to address the fourth component of the plan for fiscal year 1995. 



                          Table XI.1
           
              BEA's Fiscal Year 1995 Funding for
              Economic Statistical Improvements

                    (Dollars in thousands)


                                              Amount  Amount
                                              reques  receiv
Fiscal year 1995 improvements                    ted      ed
--------------------------------------------  ------  ------
Strengthen existing programs by making
 statistics more accurate and dependable
Strengthen underlying source data               $400      $0
Expand coverage of international trade           800       0
Launch new initiatives that reflect new
 concerns
Improve the international comparability of     2,118       0
 economic data
Incorporate environmental performance in       1,471       0
 economic measurements
Reengineer the data-gathering process
Develop a BEA information system for the       3,272     334
 year 2000
============================================================
Total                                         $8,061    $334
------------------------------------------------------------
Source:  BEA. 

BEA's two initiatives, which address the first component of the plan,
are designed to improve NIPA and the measurements of such areas as
services and international trade.  BEA officials said these
initiatives include (1) speeding up the preparation of input-output
tables, (2) obtaining a better understanding of the relationship
between quality and price changes, (3) gathering an updated picture
of U.S.  investment abroad, and (4) filling gaps in the coverage of
international trade in services, especially financial services.  In
addition, they said components of these two initiatives are related
to what BEA wanted to do under ESI but was unable to as a result of
limited funding.  For a discussion of the input-output tables
recommendation, see appendix III.  For a discussion of the
international investment and capital flows recommendation, see
appendix VIII. 

BEA's two initiatives for the second component of the plan reflect
the growing interest in today's concerns, such as environmental
accounting and international competitiveness.  These initiatives
include introducing the new international standards for the SNA and
balance of payments accounts, revamping the reporting of
international capital flows, and developing satellite accounts that
bring environmental quality into economic measurements.  BEA
officials said the initiative to incorporate environmental
performance in economic measurements expands the earlier progress
made under ESI with satellite accounts.  For a discussion of the SNA
recommendation, which includes satellite accounts, see appendix III. 

Finally, BEA had one initiative that addresses the third component of
its plan.  The new information system will reengineer the data
collection process, which will move BEA's 1970's mainframe-oriented
data processing system to an integrated microcomputer network. 

The final initiative, which aims to make information a competitive
asset by giving widespread access to data using electronic
technologies, is planned in future years, but no funds were requested
for this initiative in fiscal year 1995. 


      BEA RECEIVED $0.3 MILLION
------------------------------------------------------ Appendix XI:1.2

BEA received $334,000 for its initiatives under EI\3 .  BEA said that
these funds will be applied to reengineer the data collection process
initiative.  BEA is also attempting to obtain additional departmental
funding to support this initiative.  Additional funds were
appropriated to BEA for an independent study of developing satellite
accounts that bring environmental quality into economic measurements;
these funds cannot be used to support BEA's work in this area. 


      CENSUS REQUESTED $8.6
      MILLION
------------------------------------------------------ Appendix XI:1.3

With its request for $8.6 million, Census proposed a plan to improve
and expand coverage of the service sector, construction, and the
environment.  One improvement also focuses on jointly restructuring
the existing industry classification system with Canada and Mexico to
create a North American Industry Classification System and
establishing a common commodity classification system.  As shown in
table XI.2, there are four improvements in this plan. 



                          Table XI.2
           
             Census' Fiscal Year 1995 Funding for
              Economic Statistical Improvements

                    (Dollars in thousands)


                                              Amount  Amount
                                              reques  receiv
Fiscal year 1995 improvements                    ted      ed
--------------------------------------------  ------  ------
Improve and expand service statistics         $3,833      $0
Improve coverage and quality of construction   1,332       0
 data
Expand coverage of the Survey of Pollution     1,056       0
 Abatement to include environmental spending
 by selected nonmanufacturing businesses
Restructure the SIC system and create a        2,423   1,000
 commodity classification system for the
 United States
============================================================
Total                                         $8,644  $1,000
------------------------------------------------------------
Source:  Census. 

Census' first improvement for service statistics would (1) provide
annual information on services purchased by manufacturers, (2)
provide quarterly estimates of corporate income and profits for the
service sector by expanding the scope of the Quarterly Financial
Report program, and (3) expand coverage of annual service surveys to
include more transportation industries and provide detailed annual
measures of industry output and expenditures for business automotive
rental and leasing, amusement and recreation, and health service
industries.  Census officials said that these initiatives build on
prior efforts begun under ESI.  For a discussion of the purchased
services recommendation and the service sector surveys
recommendation, see appendix IV. 

Census' second improvement for the coverage and quality of
construction data would improve the coverage of nonresidential
additions and alterations, as well as the quality of industrial plant
construction data.  This improvement corrects serious deficiencies in
the measurement of new construction and would provide for improved
measures of nonresidential reconstruction.  Again, Census officials
said this initiative continues earlier efforts under ESI.  For a
discussion of the construction - coverage recommendation, see
appendix III. 

Census' third improvement would expand the Pollution Abatement
Survey, which provides operations cost information, to include
nonmanufacturing industries.  This improvement would provide a more
comprehensive measure of environmental spending by businesses. 

Finally, Census' fourth improvement to restructure the SIC system and
create a commodity classification system would provide for the
development and implementation of a restructured industry
classification system to serve the needs of the 21st century.  A
commodity classification system would be for use by all government
agencies.  Census officials said this improvement builds on efforts
begun under ESI.  For a discussion of the SIC recommendation, see
appendix IX. 


      CENSUS RECEIVED $1 MILLION
------------------------------------------------------ Appendix XI:1.4

Census received $1 million to restructure the SIC system and create a
commodity classification system.  This will continue efforts begun
under ESI. 


   BLS RECEIVED NEARLY FULL
   FUNDING
-------------------------------------------------------- Appendix XI:2

BLS requested more than $17 million and received nearly $17 million
to carry out three of four improvements, which, according to BLS
officials, are the highest priorities of BLS as well as the
Department of Labor for 1995.  These improvements would address the
overall quality of BLS' data and help BLS respond to the current
needs of its users.  BLS was appropriated funding for increasing the
CPS sample size for two states, although it had not requested funding
to do this.  (See table XI.3.)



                          Table XI.3
           
              BLS' Fiscal Year 1995 Funding for
              Economic Statistical Improvements

                    (Dollars in thousands)


                                              Amount  Amount
                                              reques  receiv
Fiscal year 1995 improvements                    ted      ed
--------------------------------------------  ------  ------
Revise the CPI                                $5,140  $5,134
Improve the quality of ES-202 program data     4,977   4,977
Increase BLS' ability to provide timely data   1,000       0
 for emerging policy issues for CPS
 supplements
Provide for an increase to begin a new         6,036   6,036
 national longitudinal survey cohort of
 youth
Increase CPS sample size for two states            0     750
============================================================
Total                                         $17,15  $16,89
                                                   3       7
------------------------------------------------------------
Source:  BLS. 


      BLS REQUESTED $17.2 MILLION
------------------------------------------------------ Appendix XI:2.1

As shown in table XI.3, BLS requested a little over $17 million for
four improvements.  According to BLS officials, the first improvement
to revise the CPI is something BLS has been wanting to do for several
years but was unable to accomplish because of funding constraints.\2
For this improvement, BLS plans to revise the market basket of goods
and services used to calculate the CPI. 

According to BLS officials, the second improvement aims to address
the quality of the ES-202 program, which provides employment and wage
data for workers covered by unemployment insurance laws.  BLS
officials said that, because the ES-202 program covers the universe
of establishments from which it obtains its business establishment
list and draws its survey samples, improving the ES-202 program will
naturally improve the adequacy of its other surveys.  Officials said
the $5 million for this improvement includes about $3 million in
additional funding to cover additional workload needs and about $2
million for further improvements to the ES-202 program's business
establishment list.  The business establishment list improvements
were included in ESI; BLS was able to make some improvements in the
list despite receiving limited additional funding for fiscal years
1990 through 1994.  For a discussion of the business establishment
lists - Census and BLS recommendation, see appendix IX. 

The next two improvements--to increase BLS' ability to provide timely
data for emerging policy issues for Current Population Survey (CPS)
supplements and to begin a new national longitudinal survey cohort of
youth--are related, according to BLS officials.  They said that both
improvements would help BLS better measure issues occurring in the
economy. 


--------------------
\2 A recommendation to revise CPI was also included in the
administration's recent report, From Red Tape to Results:  Creating a
Government That Works Better and Costs Less, report of the National
Performance Review, Vice President Albert Gore, Sept.  7, 1993. 


      BLS RECEIVED $17 MILLION
------------------------------------------------------ Appendix XI:2.2

As shown in table XI.3, BLS received either full or nearly full
funding for three of its four improvements.  The improvement to
provide timely data for emerging policy issues for the CPS supplement
received no funding.  However, BLS received $750,000 to increase CPS
sample sizes for two states so that CPS data could be used directly,
rather than through estimation methods, to determine employment and
unemployment statistics for these states; BLS did not request this
funding. 


   NASS REQUESTED $4.4 MILLION BUT
   DID NOT RECEIVE FUNDING
-------------------------------------------------------- Appendix XI:3

As shown in table XI.4, NASS requested a total of $4.4 million and
received no funding for two economic statistical improvements.  NASS
officials said they will continue to seek these funds and add them
into the 1996 budget request. 



                          Table XI.4
           
              NASS' Fiscal Year 1995 Funding for
              Economic Statistical Improvements

                    (Dollars in thousands)


                                              Amount  Amount
                                              reques  receiv
Fiscal year 1995 improvements                    ted      ed
--------------------------------------------  ------  ------
Reduce mainframe processing costs             $1,100      $0
Expand coverage of chemical use and crops in   3,300       0
 pesticide data program
============================================================
Total                                         $4,400      $0
------------------------------------------------------------
Source:  NASS. 

NASS officials said these improvements continue the work they have
been doing outside the ESI framework (see discussion in app.  X). 
The first improvement would reduce mainframe processing costs by
distributing NASS' farm list maintenance systems and survey
processing systems to local area networks and state statistical
offices.  This improvement also has the potential for substantial
cost savings to the government and would make NASS' farm list frame
system more responsive with the addition of a database system.  This
would also reduce respondent burden and improve data quality. 

NASS' second improvement to expand the coverage of the Pesticide Data
Program would help NASS build new lists and rebuild old lists of
specialty commodity producers.  For this improvement, NASS planned to
add chemical use information for 13 additional states and 10
additional crops. 




(See figure in printed edition.)Appendix XII
COMMENTS FROM THE BUREAU OF LABOR
STATISTICS
========================================================== Appendix XI




(See figure in printed edition.)Appendix XIII
COMMENTS FROM THE NATIONAL
AGRICULTURAL STATISTICS SERVICE
========================================================== Appendix XI



(See figure in printed edition.)


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix XIV

Neil T.  Asaba
Paul M.  Aussendorf
Richard S.  Krashevski
Edward J.  Laughlin
Tamara A.  Lumpkin
James M.  McDermott
Victoria E.  Miller
Patrick R.  Mullen
Pamela R.  Pavord
A.  Elizabeth Powell
Lori Rectanus
Kathleen K.  Scholl
Loren Yager


GLOSSARY
=========================================================== Appendix 0

Readers interested in the exact authoritative definitions, which are
often governed by internationally agreed upon guidelines, should
consult documentation on concepts, data sources, and methods
published by BEA, Census, BLS, the Federal Reserve, OMB, NSF, and
NASS, as well as similar documents published by the United Nations,
the Organization for Economic Cooperation and Development, IMF, the
World Bank, and the Commission of the European Communities.  These
definitions, which are largely based on that documentation, are
intended to provide assistance to the general reader. 


      ANNUAL SURVEY OF
      COMMUNICATION SERVICES
------------------------------------------------------- Appendix 0:0.1

Survey of revenue and expenses for the communication services
industries.  Data are used in NIPA, input-output tables, and to
develop other economic measures. 


      ANNUAL CAPITAL EXPENDITURES
      SURVEY (ACES)
------------------------------------------------------- Appendix 0:0.2

Collects broad-based data on business spending for new plants and
equipment.  The survey data serve as a benchmark for other capital
expenditure statistics programs. 


      AREA AGRICULTURAL FRAMES
------------------------------------------------------- Appendix 0:0.3

Area-based sampling frame for agricultural commodities classified by
land use, such as agricultural-urban or intensively cultivated.  (See
Sampling Frames.)


      ASSET-BACKED SECURITIES
      (ABS)
------------------------------------------------------- Appendix 0:0.4

As used in the flow-of-funds accounts, a sector similar to federally
related mortgage pools, in that the issuers are not a group of
institutions but a set of contractual arrangements.  ABS represent
claims against loan assets that have been pooled to serve as
collateral for issues of securities.  The principal assets of ABS
sector are mortgages and mortgage-backed securities, consumer credit
(automobile and credit card), business loans, and trade receivables. 
(See Securitization.)


      AUTOMATED DATA COLLECTION
      TECHNIQUES
------------------------------------------------------- Appendix 0:0.5

Techniques, including the application of computer-assisted
interviewing techniques, which allow for more effective interviewing,
increased timeliness of reporting, and enhanced data analysis
capabilities. 


      AUTOMATED EXPORT REPORTING
      PROGRAM
------------------------------------------------------- Appendix 0:0.6

The process of electronically filing export data directly with
Census. 


      AUTOMATED TOUCH-TONE DATA
      ENTRY
------------------------------------------------------- Appendix 0:0.7

Technique for conducting surveys using a Touch-Tone telephone. 
Respondents follow voice prompts and enter their answers by pressing
the telephone keypad. 


      AUTOMATION OF EXPORT AND
      IMPORT DATA COLLECTION
------------------------------------------------------- Appendix 0:0.8

Gathering data electronically, which allows flexibility for
respondents without an increase in respondent burden. 


      BALANCE OF PAYMENTS (BOP)
------------------------------------------------------- Appendix 0:0.9

In the international economic accounts, BOP reports the economic
transactions of the residents of a nation with the residents of the
rest of the world, during a particular point in time, highlighting
international trade flows and the international payment mechanisms. 
BOP accounts for all of the wealth flowing in and out of a country. 


      BALANCE SHEET
------------------------------------------------------ Appendix 0:0.10

Also referred to as a statement of condition or statement of
financial position, a balance sheet is a statement that reports the
type and amount of assets, liabilities, and owners' equity on a given
date. 


      BALANCE SHEET ENHANCEMENTS
------------------------------------------------------ Appendix 0:0.11

A process or technique that improves a balance sheet.  An improvement
might be a more frequent account of information, such as a quarterly
versus annually published statement.  Another enhancement might be
the incorporation of additional available information, or data that
have been revised based on better information. 


      BENCHMARK
------------------------------------------------------ Appendix 0:0.12

Anything that serves as a standard against which other things are
compared and measured; a reference point. 


      BENCHMARK PROCESS
------------------------------------------------------ Appendix 0:0.13

The systematic method of deriving a benchmark.  (See Benchmark.)


      BUSINESS ESTABLISHMENT LISTS
------------------------------------------------------ Appendix 0:0.14

Directories of businesses maintained by statistical agencies that are
used when the agencies survey establishments to obtain data about
their business practices. 


      CAPITAL ACCOUNT
------------------------------------------------------ Appendix 0:0.15

Measures the flow of capital that consist of changes in the U.S. 
assets abroad and foreign assets in the United States.  Financial
assets include international claims payable in money, such as loans,
bank deposits, drafts, acceptances, notes, government and private
debt and equity securities, and intercompany accounts.  (For a more
complete definition see The Balance of Payments of the United States: 
Concepts, Data Sources, and Estimating Procedures, U.S.  Department
of Commerce [DOC], BEA (Washington, D.C.:  U.S.  Government Printing
Office [GPO], May 1990), p.6.)


      CAPITAL ASSET
------------------------------------------------------ Appendix 0:0.16

An asset needed to create a product (or service) normally acquired
with the intention of being held at least 1 year rather than being
resold.  Generally, this includes fixed assets, such as land,
buildings, and equipment. 


      CAPITAL EXPENDITURE
------------------------------------------------------ Appendix 0:0.17

Outlay of money to acquire or improve capital assets such as
buildings and machinery.  (See Capital Asset.)


      CAPITAL FLOWS
------------------------------------------------------ Appendix 0:0.18

The selling and purchasing of capital assets from one country to
another.  (See Balance of Payments.)


      CAPITAL GAINS
------------------------------------------------------ Appendix 0:0.19

Appreciation in the value of an asset.  The amount by which the
asset's selling price exceeds the asset's purchase price.  For income
tax purposes, a capital gain is the gain from the sale or exchange of
a capital asset. 


      CAPITAL LOSSES
------------------------------------------------------ Appendix 0:0.20

Depreciation in the value of an asset.  The amount by which the
proceeds from the sale of a capital asset are less than the cost of
acquiring it.  For income tax purposes, a capital loss is the loss
from the sale or exchange of a capital asset.  (See Capital Gains.)


      CAPITAL RELATED INCOME FLOWS
------------------------------------------------------ Appendix 0:0.21

With respect to U.S.  direct investment abroad, consists of equity
capital flows, reinvested earnings, and intercompany debt flows of
U.S.  direct investors to both their incorporated and unincorporated
foreign affiliates.  (For a more complete definition see The Balance
of Payments of the United States:  Concepts, Data Sources, and
Estimating Procedures, DOC, BEA, (Washington, D.C.:  GPO, May 1990),
p.  90.) (See Direct Investment.)


      CLOSED-END FUNDS
------------------------------------------------------ Appendix 0:0.22

Type of fund that has a fixed number of shares, usually listed on a
major stock exchange.  It tends to have specialized portfolios of
stocks, bonds, convertibles, or combinations thereof, and may be
oriented toward income, capital gains, or a combination of these
objectives. 


      COGNITIVE INTERVIEW
      TECHNIQUES
------------------------------------------------------ Appendix 0:0.23

An intellectual process by which knowledge is gained about
perceptions or ideas and used to improve questionnaire designs prior
to their use in national surveys. 


      COMMERCIAL PAPER
------------------------------------------------------ Appendix 0:0.24

Short-term promissory notes used by businesses to obtain money,
usually for a specific need. 


      COMMODITY FLOW SURVEY
------------------------------------------------------ Appendix 0:0.25

Survey of commodity movements within the United States.  The data
provide periodic detailed measures of commodity movements by
establishments with paid employees and engaged in manufacturing,
mining, wholesale trade, or selected retail and services industries. 
(For a more complete definition see Economic Data Programs, DOC,
Economic and Statistics Administration, Census (Dec.  1993).)


      COMPENSATION
------------------------------------------------------ Appendix 0:0.26

Cash (wages and salaries) and noncash (employee benefits) payments
given to employees on the basis of the value of their work
contributions and performance. 


      COMPUTER-ASSISTED TELEPHONE
      INTERVIEW (CATI)
------------------------------------------------------ Appendix 0:0.27

A telephone interviewing technique for data collection.  Responses to
surveys are directly entered into the survey form on the computer. 


      CONFIDENTIAL
------------------------------------------------------ Appendix 0:0.28

Respondent is assured that dissemination of data is prohibited that
would allow public identification of the respondent or that would in
any way be harmful to the respondent. 


      CONSTANT DOLLAR
------------------------------------------------------ Appendix 0:0.29

A dollar measured in terms of prices of a base period, currently 1987
for most purposes, to remove the influence of inflation.  The
resulting constant dollar is the value that would exist if prices had
remained the same as in the base period.  (See Current Dollar.)


      CONSTANT-DOLLAR OUTPUT
------------------------------------------------------ Appendix 0:0.30

Output expressed in dollar amounts that remove the influence of
inflation. 


      CONSUMER PRICE INDEX (CPI)
------------------------------------------------------ Appendix 0:0.31

A program that measures average price change for two population
groups--all urban consumers (CPI-U) and urban wage earners and
clerical workers (CPI-W)--using a specified market basket
representing all goods and services purchased for everyday living. 
The CPI measures the same basket relative to a designated base
period.  The CPI is used to adjust for inflation. 


      CORPORATE EQUITY ESTIMATES
------------------------------------------------------ Appendix 0:0.32

Estimates of the claims to share in the net income and the assets of
a corporation, such as common stock. 


      CORPORATE FINANCIAL DATA
------------------------------------------------------ Appendix 0:0.33

The economic accounts of an institution that make up the balance
sheet and income statement.  Data in these reports include assets,
liabilities, and stockholders' equity, income (or loss) from
operations, and the breakdown of income before and after taxes. 


      COVERAGE
------------------------------------------------------ Appendix 0:0.34

A survey's ability to obtain information from all representative
units in a specific population.  Undercoverage occurs when a survey
fails to represent all units in the survey. 


      CPI-U-X1
------------------------------------------------------ Appendix 0:0.35

An experimental CPI that BLS developed for researchers who wish to
make historical comparisons with the current CPI, which uses the
rental equivalence approach to measuring shelter services.  Prior to
1983, the measurement of homeowner costs in the CPI included changes
in the asset value of homes.  The rental equivalence approach
isolates the consumption from the investment aspects of
homeownership. 


      CURRENT ACCOUNT
------------------------------------------------------ Appendix 0:0.36

Measures transactions in goods, services, and unilateral transfers
between countries.  "Goods" include tangible and visible commodities,
which may be raw materials or intermediate or final products. 
"Services" refers to an economic output of intangible commodities
that may be produced, transferred, and consumed at the same time. 
Unilateral transfers measures transfers of real resources of
financial assets to a country where the country receiving the
resources does not expect a payment of any kind in return.  (For a
more complete definition see The Balance of Payments of the United
States:  Concepts, Data Sources, and Estimating Procedures, DOC, BEA,
(Washington, D.C.:  GPO, May 1990), p.  6.)


      CURRENT COST ACCOUNTING
------------------------------------------------------ Appendix 0:0.37

(See Inflation Accounting.)


      CURRENT DOLLAR
------------------------------------------------------ Appendix 0:0.38

The dollar value of a good or service in terms of the period under
consideration, which reflects the then-prevailing prices of the good
or service.  (See Constant Dollar.)


      CURRENT EMPLOYMENT
      STATISTICS (CES)
------------------------------------------------------ Appendix 0:0.39

A monthly survey of 380,000 business establishments nationwide.  It
collects information on employment, workers' hours, and earnings from
the payroll records of employers.  This information is used to
produce a range of economic statistics. 


      CURRENT POPULATION SURVEY
      (CPS)
------------------------------------------------------ Appendix 0:0.40

A monthly survey of 60,000 households that provides a comprehensive
body of information on labor force, employment, and unemployment
experience of the nation's population, classified by age, sex, race,
and a variety of other characteristics. 


      DATA DUPLICATION
------------------------------------------------------ Appendix 0:0.41

The collection of the same information from one source, or
respondent, by more than one statistical agency. 


      DEFLATE
------------------------------------------------------ Appendix 0:0.42

To reduce a price level.  For example, to adjust a current-dollar
amount to its constant-dollar counterpart, or to remove the effects
of inflation. 


      DIRECT INVESTMENT
------------------------------------------------------ Appendix 0:0.43

Implies that a person in one country has a lasting interest in, and a
degree of influence over, the management of a business enterprise in
another country.  Generally, this type of investment involves a
long-term relationship and ownership of at least 10 percent of the
voting stock of, or an equivalent interest in, a business enterprise
located in another country.  (For a more complete definition see The
Balance of Payments of the United States:  Concepts, Data Sources,
and Estimating Procedures, DOC, BEA (Washington, D.C.:  GPO, May
1990), p.  84.)


      DOUBLE DEFLATION
------------------------------------------------------ Appendix 0:0.44

A method for deriving the constant-dollar estimates of the "value
added" or contribution an industry makes to the gross domestic
product.  It is the difference between constant-dollar gross output
and constant-dollar intermediate inputs.  It allows for changes over
time in the relationships between gross output and inputs. 
Generally, it is a preferred method among indirect estimation
methods; however, it is not the preferred method for three
industries--private households, federal general government, and state
and local general governments.  (For a more complete definition see
Frank De Leeuw, Michael Mohr, and Robert Parker, "Gross Product by
Industry, 1977-88:  A Progress Report on Improving the Estimates,"
Survey of Current Business, Vol.  71, No.  1 (Jan.  1991), p.  30.)


      EMPLOYMENT COST INDEX (ECI)
------------------------------------------------------ Appendix 0:0.45

A quarterly index measuring changes in total compensation--wages,
salaries, and employers' costs for employee benefits.  ECI tracks
changes in labor costs that are free from influences of employment
shifts among occupations and industries. 


      EMPLOYMENT AND WAGES PROGRAM
------------------------------------------------------ Appendix 0:0.46

Commonly called the ES-202 program, this BLS program provides
employment and wage data for workers covered by state unemployment
insurance laws and for civilian workers covered by the unemployment
compensation for federal employees program.  The ES-202 program
provides wage and employment data for a virtual census of
nonagricultural employees and their wages.  In addition, about
one-half of all agricultural workers are included.  The ES-202
program provides the universe for almost all BLS establishment survey
samples. 


      ENVIRONMENTAL ACCOUNTING
------------------------------------------------------ Appendix 0:0.47

Monitoring and measuring the environmental conditions in a society
with respect to the economic growth in that society to determine if
there are associated costs.  For example, if the benefits of
increased income are associated with deteriorating health conditions,
by what amount is the income offset by the cost of poorer health
conditions? 


      ESTABLISHMENT
------------------------------------------------------ Appendix 0:0.48

As defined under the Standard Industrial Classification (SIC) system,
an economic unit that typically is based at a single physical
location, where a business is operated or where services or
industrial operations are conducted.  A distinct economic activity is
performed and information detailing the number of employees, their
wages and salaries, business transactions, sales and receipts of the
business, and other types of establishment data can be reported. 
Establishments are classified into a SIC industry on the basis of
their principal product or service.  (For a more complete definition
see 1987 Standard Industrial Classification Manual, Executive Office
of the President, OMB (Washington, D.C.:  1987), p.  12.)


      EXPERIMENTAL ESTIMATES
------------------------------------------------------ Appendix 0:0.49

An approximate calculation or judgment regarding the value, amount,
and size of an outcome derived from an experience-based study or
test. 


      FAMILY INCOME
------------------------------------------------------ Appendix 0:0.50

The sum of money wage and salary income, net income from
self-employment, and income other than earnings.  Total family income
is the sum of the income amounts received by all individuals in the
family. 


      FARM
------------------------------------------------------ Appendix 0:0.51

A farm is a place in which the land involved must be used for or
connected with agricultural operations, and it must be operated under
the day-to-day control of one individual or management (e.g.,
partnership or corporation).  As defined in the 1987 Census of
Agriculture, a farm is an operation that would normally have had
$1,000 or more in total value of sales of agricultural products
during the reference year. 


      FARM LISTS
------------------------------------------------------ Appendix 0:0.52

A mailing list of names and addresses of agricultural operations that
is compiled before census or survey questionnaires are mailed.  (See
Farm.)


      FINANCIAL INTERMEDIARIES
------------------------------------------------------ Appendix 0:0.53

Commercial bank, savings and loan, mutual savings bank, credit union,
or other institutions, such as a mutual fund or a life insurance
company that balances out the flow of funds between savings-surplus
units with those of savings-deficit units.  These institutions
redistribute savings into productive uses and at the same time
provide diversification of risk and liquidity to the individual
saver. 


      FLOW OF FUNDS
------------------------------------------------------ Appendix 0:0.54

Also referred to as the capital finance accounts of the United
States, accounts that show the role of financial institutions and
instruments in transforming saving into investment and the changes in
assets and liabilities, which result from this transformation.  The
flow-of-funds accounts detail how the current investment in tangible
and financial assets contribute to a buildup of a stock of assets for
each sector of the economy and to the creation of national wealth. 
The flow-of-funds accounts can be viewed as combining data on the
flows of saving and tangible investment published in NIPA with
additional details on borrowing and lending for specific economic
sectors. 


      GOVERNMENT SECTOR
------------------------------------------------------ Appendix 0:0.55

(1) Includes all levels of government--federal, state, and local. 
Income principally derives from direct and indirect taxes, and also
sales of goods and services.  (2) All economic activities--mostly
services--that are carried out directly by government agencies
largely for public benefit rather than for profit.  (3) The value
added to the economy by government agencies and other publicly owned
bodies, consisting primarily of services performed by government
employees. 


      GRANT-IN-AID TRANSACTIONS
------------------------------------------------------ Appendix 0:0.56

The transfer, through the use of U.S.  government financing, of real
resources or financial assets to foreigners under programs enacted by
Congress for the provision of military or nonmilitary foreign
assistance (grants) for which no repayment is expected. 


      GROSS
------------------------------------------------------ Appendix 0:0.57

An overall total, with no deductions, such as taxes or expenses. 
(See Net.)


      GROSS DOMESTIC PRODUCT (GDP)
------------------------------------------------------ Appendix 0:0.58

The value of all final goods and services produced within the borders
of the United States in a given period, whether produced by residents
or nonresidents.  (See Gross National Product.)


      GROSS NATIONAL PRODUCT (GNP)
------------------------------------------------------ Appendix 0:0.59

The total market value of all goods and services produced in a given
period by labor and property supplied by residents of a country,
regardless of where the labor and property are located.  GNP differs
from GDP primarily by including the excess of capital income that
residents earn from investments abroad less capital income that
nonresidents earn from domestic investment. 


      GNP REVISIONS
------------------------------------------------------ Appendix 0:0.60

Periodic estimates of GNP released as more detailed information
becomes available.  The first estimate of GNP is published near the
end of the second month after the end of a quarter.  A revised
estimate is released during the third month following the end of a
quarter.  The fourth quarter GNP estimate is delayed by an additional
month.  (For a more complete definition see "A Look at How BEA
Presents the NIPA's," Survey of Current Business, Vol.  73, No.  2
(Feb.  1993), p.  30.)


      HARMONIZED SYSTEM
------------------------------------------------------ Appendix 0:0.61

Adopted in the United States in 1989, it is an internationally
recognized goods classification system used by both importers and
exporters. 


      HOUSEHOLD SECTOR
------------------------------------------------------ Appendix 0:0.62

Individual households and nonprofit institutions, such as charitable
organizations, private foundations, schools, churches, labor unions,
and hospitals, whose financial assets and liabilities are reported on
a combined basis in the flow-of-funds accounts. 


      INDEX
------------------------------------------------------ Appendix 0:0.63

A statistical measure that places a current economic condition in the
context of a base period. 


      INDIRECT ESTIMATION METHODS
------------------------------------------------------ Appendix 0:0.64

Indirect estimation methods are procedures used to construct a
statistical measure from data that are not direct observations of the
activity being measured, or when data needed to implement the
conceptually correct, or preferred, measure are not available. 


      INFLATION
------------------------------------------------------ Appendix 0:0.65

A sustained upward movement of the price of goods and services. 


      INFLATION ACCOUNTING
------------------------------------------------------ Appendix 0:0.66

Techniques used to deal with the impact of inflation on accounts and
accounting procedures. 


      INFLATION ADJUSTMENTS
------------------------------------------------------ Appendix 0:0.67

Procedures used to capture the pure increase in prices apart from
changes in quality, discontinuation, and other changes in items
available for purchase. 


      INPUT-OUTPUT TABLES
------------------------------------------------------ Appendix 0:0.68

The part of the national accounts that trace the flow of goods and
services among industries in the production process and show the
value added by each industry and the detailed commodity composition
of national output. 


      INTERNATIONAL CAPITAL FLOWS
------------------------------------------------------ Appendix 0:0.69

Flows that occur when residents in one country (the capital
exporters) extend loans to, or purchase the title to, assets from the
residents of another country (the capital importers). 


      INTERNATIONAL ECONOMIC
      ACCOUNTS
------------------------------------------------------ Appendix 0:0.70

(See Balance of Payments.)


      INTERNATIONAL GUIDELINES FOR
      ECONOMIC ACCOUNTS
------------------------------------------------------ Appendix 0:0.71

Standards developed by the International Monetary Fund for balance of
payments accounting. 


      INTERNATIONAL INVESTMENT
------------------------------------------------------ Appendix 0:0.72

(1) Purchase of the ownership of assets--such as securities;
businesses; titles to land, buildings, or equipment; and bank
deposits--in a foreign country by a U.S.  private individual,
business, institution, or government.  (2) The export of capital from
one country to another. 


      INTERNATIONAL MONETARY FUND
      (IMF)
------------------------------------------------------ Appendix 0:0.73

A financial institution established in 1946 that strives to stabilize
worldwide currencies to maintain an orderly market in world trade. 
The establishment provides a framework in which governments can
consult and cooperate in determining the structure and functioning of
the international monetary system. 


      INTERNATIONAL PRICE PROGRAM
      (IPP)
------------------------------------------------------ Appendix 0:0.74

Program that measures price changes of commodities traded between the
United States and the rest of the world.  The IPP produces the U.S. 
export and import price indexes.  These indexes are intended to
reflect price trends of foreign-produced goods entering the United
States or domestic products leaving this country or its territories. 


      INVESTMENT
------------------------------------------------------ Appendix 0:0.75

According to The MIT Dictionary of Modern Economics, investment is
"used to describe the flow of expenditures devoted to increasing or
maintaining the real capital stock [the sum of goods which are used
as inputs for further production]...  Investment is the flow of
expenditures devoted to projects producing goods which are not
intended for immediate consumption.  These investment projects may
take the form of adding to physical and human capital as well as
inventories."\1


--------------------
\1 David W.  Pearce, ed., The MIT Dictionary of Modern Economics 3rd
ed.  (Great Britain:  The Macmillan Press, 1986), pp.  51, 55, and
216.  The MIT Dictionary of Modern Economics also defines capital
stock as the sum of goods that are used as inputs for further
production. 


      ISSUER
------------------------------------------------------ Appendix 0:0.76

Legal entity that has the power to issue and distribute a security. 
Issuers include corporations, municipalities, foreign and domestic
governments and their agencies, and investment trusts. 


      LABOR FORCE
------------------------------------------------------ Appendix 0:0.77

The number of noninstitutionalized persons over 16 years of age who
are employed. 


      LONGITUDINAL RESEARCH
      DATABASE (LRD)
------------------------------------------------------ Appendix 0:0.78

A longitudinal database that links establishment level data from
several censuses and surveys of manufacturers.  LRD contains data
that identify individual establishments, and detail manufacturing
inputs and manufactured products.  The database is used for ongoing
research on entrepreneurship, firms and product markets, labor
markets, growth and technical change, international issues, and
environmental regulation.  (See Longitudinal Study.)


      LONGITUDINAL STUDY
------------------------------------------------------ Appendix 0:0.79

A study or research observing the behavior or characteristics of a
specific group, or age-cohort, extending over a long period.  For
example, in a longitudinal panel survey the same people are
repeatedly interviewed over a period.  Interviewing may be carried
out once every few months, once a year, or once every 2 years over a
time span as short as 36 months or over a much longer period, such as
20 years. 


      MANDATORY SURVEYS
------------------------------------------------------ Appendix 0:0.80

Survey respondents are obligated or authoritatively ordered to
complete a survey questionnaire. 


      MANUFACTURING TECHNOLOGY
      SURVEYS
------------------------------------------------------ Appendix 0:0.81

Data on the presence, use, and planned use of advanced technologies
in the manufacturing sector. 


      MARKET BASKET
------------------------------------------------------ Appendix 0:0.82

A combination of goods and services used to track price changes in
the CPI.  The items in the market basket are statistically determined
to capture the price changes of goods and services purchased by
consumers. 


      MERCHANDISE EXPORTS MODEL
------------------------------------------------------ Appendix 0:0.83

The design and manipulation of a mathematical representation that
simulates the exportation of merchandise so the effect of various
international changes can be studied and forecast. 


      MILITARY TRANSACTIONS
------------------------------------------------------ Appendix 0:0.84

International transactions in which U.S.  government military
agencies participate. 


      MONTHLY DEBT AGGREGATE
------------------------------------------------------ Appendix 0:0.85

A measure of the total credit market debt of domestic nonfinancial
sectors.  A monitoring range has been established by the Federal
Reserve's Open Market Committee for the debt aggregate.  This range
applies to growth, from the fourth quarter of the previous year to
the fourth quarter of the current year, in the average amount of debt
outstanding. 


      MUTUAL FUNDS
------------------------------------------------------ Appendix 0:0.86

Investment companies that purchase financial assets, often with
specific investment objectives such as providing income to the
investor.  The value of the individual investor's share in a mutual
fund is determined by the market value of the underlying securities. 


      NATIONAL ECONOMIC ACCOUNTS
------------------------------------------------------ Appendix 0:0.87

Accounts that provide a coherent and comprehensive picture of the
nation's economy.  These accounts describe (1) the output of the
economy, its size, composition, and use and (2) the economic process
or mechanism by which this output is produced and distributed.  The
accounts include the national income and product accounts, capital
finance and balance sheet accounting, and input-output accounting. 
Each account describes some aspects of the structure, workings, and
performance of the economy. 


      NATIONAL INCOME AND PRODUCT
      ACCOUNTS (NIPA)
------------------------------------------------------ Appendix 0:0.88

Detailed descriptions of the overall U.S.  economy.  The accounts
depict in dollar terms the volume, composition, and use of the
nation's output of goods and services.  These core national economic
statistical measures are used to analyze past and current economic
performance and also to forecast economic developments. 


      NET
------------------------------------------------------ Appendix 0:0.89

Generally, the figure remaining after all relevant deductions have
been made from the gross amount.  For example, net sales are equal to
gross sales minus discounts, returns, and allowances; net profit is
gross profit less operating (sales, general, and administrative)
expenses; and net worth is assets (worth) less liabilities. 


      NONCASH BENEFITS
------------------------------------------------------ Appendix 0:0.90

Payments in a form other than money, which are received as public
benefits and as employer- or union-provided benefits to employees. 
Examples of public benefits include food stamps, public housing, and
government-provided health care.  Examples of noncash benefits to
employees include pension plans and group health insurance. 


      NONCORPORATE BUSINESS SECTOR
------------------------------------------------------ Appendix 0:0.91

As used in the flow-of-funds accounts, nonfarm partnerships
(businesses that file IRS form 1065); nonfarm sole proprietorships
(businesses that file IRS schedule C); tax-exempt cooperatives such
as rural electrification, rural telephone, and farmer marketing
cooperatives; and individuals who receive rental income and report
the income on IRS schedule E. 


      NONINTEREST INCOME
------------------------------------------------------ Appendix 0:0.92

Activities generating fee income in the banking industry, such as
mortgage servicing and refinancing fees, service charges on deposits
and in securities trading gains, fiduciary activities, and foreign
exchange gains and fees. 


      NONRESIDENTIAL STRUCTURES
------------------------------------------------------ Appendix 0:0.93

New construction (which includes hotels and motels, as well as mining
exploration, shafts, and wells), brokers' commissions on sale of
structures, and net purchases of used structures by business and
nonprofit institutions from government.  (For a more complete
definition see National Income and Product Accounts of the United
States, Vol.  2, 1959-88, D.C., Economics and Statistics
Administration, BEA, (Washington, D.C., Sept.  1992), p.  M-6.)


      OCCUPATIONAL COMPENSATION
      SURVEY PROGRAM
------------------------------------------------------ Appendix 0:0.94

Survey resulting from BLS' restructure of its area wage survey
program to provide the data needed under the Federal Employees Pay
Comparability Act of 1990 (P.L.  101-509, Sec.  529).  Implementation
of this act requires surveying pay rates for nonfederal employees in
various localities across the country.  Detailed information is
provided on salary levels and distributions for the types of private
industry and state and local government jobs published in the survey. 


      OUTPUT
------------------------------------------------------ Appendix 0:0.95

The amount of work done or the amount produced by persons or
businesses over a specific period. 


      PAYROLL STATISTICS
------------------------------------------------------ Appendix 0:0.96

Data taken from payroll records of establishments.  These data
provide the principal source of state and metropolitan-area labor
statistics. 


      PESTICIDE DATA PROGRAM
------------------------------------------------------ Appendix 0:0.97

A program of the U.S.  Department of Agriculture that collects
objective, comprehensive data on chemical use and pesticide residues
in fresh fruits and vegetables.  The information allows the
Environmental Protection Agency to more accurately assess the
exposure of pesticides and dietary risks to consumers. 


      PORTFOLIO INVESTMENT INCOME
------------------------------------------------------ Appendix 0:0.98

Differs from direct investment wherein the latter involves management
decisions and long-term interest in a business enterprise located in
another country.  Equity ownership of 10 percent or more is
considered evidence that some amount of management influence exists. 
Whereas, portfolio investment offers the opportunity to obtain
short-term income or capital gains in financial markets and the
opportunity to shift funds between various types of financial market
instruments or between countries.  (For a more complete definition
see The Balance of Payments of the United States:  Concepts, Data
Sources, and Estimating Procedures, DOC, BEA (Washington, D.C.:  GPO,
May 1990), pp.  64 and 84.)


      POVERTY THRESHOLDS
------------------------------------------------------ Appendix 0:0.99

Income cutoff levels that vary by family characteristics, such as
number of children, age of householder, and family size.  For
example, the poverty threshold was $11,631 for a three-person family
with one child in 1993; $11,642 for an adult with two children. 
Families with incomes equal to or lower than the thresholds, which
are updated annually for inflation, are considered poor. 


      PRICE INDEX
----------------------------------------------------- Appendix 0:0.100

Measurements that trace the relative changes in the price of an
individual good or service, or a market basket of goods and services
over time.  A price index is used to measure the growth rate of the
price level from one point in time to another, which is referred to
as the inflation rate. 


      PRODUCER PRICE INDEX (PPI)
----------------------------------------------------- Appendix 0:0.101

Program that measures average changes in prices received by domestic
producers for their output (crude materials and semifinished goods,
finished goods, and some services), relative to prices received in
the base year.  It is an industry-based survey that provides monthly
price indexes for virtually all agriculture, mining, and
manufacturing industries, as well as selected service industries. 
PPI statistical measures are used by producers of commodities in the
manufacturing, agriculture, forestry, fishing, mining, gas and
electricity, and public utilities sectors for the purpose of
identifying average price changes for specific goods over some
period, generally monthly or annual.  PPI was formerly known as a
wholesale price index. 


      PURCHASED SERVICES
----------------------------------------------------- Appendix 0:0.102

Assistance, information, and other intangible products that
businesses and organizations buy from businesses and governments. 
Examples include when a book publisher contracts with a cleaning
service for janitorial work performed in its warehouses and when a
manufacturer contracts for computer services it needs, but formerly
provided itself. 


      REAL
----------------------------------------------------- Appendix 0:0.103

Refers to statistical measures that remove the influence of inflation
when used as a modifier with statistical measures such as price,
income, and output. 


      RECONCILIATION
----------------------------------------------------- Appendix 0:0.104

The process of making data conceptually consistent between two or
more surveys or data sets. 


      RENTAL EQUIVALENCE
----------------------------------------------------- Appendix 0:0.105

The cost of shelter a home provides its owner.  This approach is used
in the CPI to measure the daily consumption value of a home's
services, a place to fix meals, relax, entertain, garden, etc.,
rather than the investment value of the home as an asset.  Basically,
it is the amount of rent that the homeowner would pay if the home
were leased. 


      REPLACEMENT COST
----------------------------------------------------- Appendix 0:0.106

As used in NIPA, the cost of replacing an asset with another that
will render similar services.  The depreciation, or the difference
between the original cost and current replacement cost of a
depreciable asset, is determined and accounted for. 


      REPROGRAMMING
----------------------------------------------------- Appendix 0:0.107

Shifting funds to use them for different purposes other than those
contemplated at the time of appropriation. 


      REPURCHASE AGREEMENT
----------------------------------------------------- Appendix 0:0.108

Agreement between a seller and buyer, usually of U.S.  government
securities, whereby the seller agrees to repurchase the securities at
an agreed upon price and usually at a stated time.  Repos, also
called RPs or buybacks, are widely used both as a money market
investment vehicle and as an instrument of the Board of Governors of
the Federal Reserve System's monetary policy. 


      RESAMPLING
----------------------------------------------------- Appendix 0:0.109

To repetitively sample to observe patterns and average outcomes. 
(See Sample.)


      RESPONSE RATE
----------------------------------------------------- Appendix 0:0.110

Quantity or amount of someone's reply or a group's reply or answers
to a given set of questions. 


      SAMPLE
----------------------------------------------------- Appendix 0:0.111

A group or subgroup selected from a particular population in such a
way that every member of that population has a known probability of
falling into the sample. 


      SAMPLING FRAMES
----------------------------------------------------- Appendix 0:0.112

A researcher's definition of the population for research purposes. 
This sampling frame is used to select a representative sample of the
general population under study. 


      SATELLITE ACCOUNTS
----------------------------------------------------- Appendix 0:0.113

A more detailed account of an economic field reported in the United
Nations' System of National Accounts (SNA).  Economic fields detailed
in the accounts include housing, agriculture, health, research and
development, natural resources, and social protection.  (See System
of National Accounts.)


      SAVING
----------------------------------------------------- Appendix 0:0.114

According to the Encyclopedia of Economics,\2 "saving is the process
of withholding current income for future use and results in the
accumulation of tangible and financial assets.  The amounts so
accumulated over past periods are referred to as savings...On a net
basis, individuals save when personal income after taxes exceeds
personal outlays, business saves through retained profits, and
governments save when current receipts exceed current expenditures. 
Gross saving includes...depreciation allowances covering the wear and
tear on real assets for future replacement....The channels through
which gross saving flows into gross investment, including financial
intermediaries are traced in the flow-of-funds accounts."


--------------------
\2 Sally S.  Ronk, "Saving," in Encyclopedia of Economics, Douglas
Greenwald, ed., (New York:  McGraw-Hill, 1982), p.  837. 


      SEASONAL ADJUSTMENT
----------------------------------------------------- Appendix 0:0.115

Statistical procedures used to eliminate the effects of weather
conditions, industry production schedules, holiday buying periods,
and other practices that are regularly occurring effects that may
affect short-term trends in data.  For example, weather changes may
affect crop production during certain months of the year, and an
increase in retail sales usually occurs during the holiday season. 
These statistical procedures are used to prevent seasonal highs and
lows during the year from being interpreted as basic changes in the
course of the economy.  (For a more complete definition see "A Look
at How BEA Presents the NIPA's," Survey of Current Business, Vol. 
73, No.  2 (Feb.  1993), p.  32.)


      SECTOR
----------------------------------------------------- Appendix 0:0.116

Institutional units grouped together based on their principal
functions, behavior, and objectives.  Generally, there are three main
sectors--households, businesses (goods and services), and government. 
(For a more complete definition see System of National Accounts 1993,
Inter-Secretariat Working Group on National Accounts (Washington
D.C., 1993), p.  19.)


      SECURITIZATION
----------------------------------------------------- Appendix 0:0.117

The process of transforming otherwise illiquid (i.e., not readily
convertible into cash) financial assets into marketable capital
market instruments.  Financial institutions can bundle together a
portfolio of loans (such as mortgages), collect the interest and
principal payments, and then pay them out to a third party.  The
claims to these interest and principal payments can thus be sold to a
third party as a security, and the financial institution makes a
profit by servicing the securitized loans (collecting the interest
and principal payments and paying them out) and charging a fee. 


      SERVICE SECTOR
----------------------------------------------------- Appendix 0:0.118

The part of the economy that produces products with an intangible
component, like assistance or information.  It includes all
industries except those classified as agriculture, construction,
manufacturing, and mining. 


      STANDARD INDUSTRIAL
      CLASSIFICATION (SIC)
----------------------------------------------------- Appendix 0:0.119

A system designed for use to classify establishments by type of
activity; to aid in the collection, tabulation, presentation, and
analysis of data relating to establishments; and for encouraging
consistency and comparability in statistical data collected by
various agencies of the U.S.  government, state agencies, trade
associations, and private research organizations.  (See
Establishments.) (For a more complete definition see 1987 Standard
Industrial Classification Manual Executive Office of the President,
OMB, p.  11.)


      STATISTICAL DISCREPANCY
----------------------------------------------------- Appendix 0:0.120

The entry necessary to balance recorded credits and debits. 
Discrepancies between credits and debits arise when data are
collected from many different sources.  Contributing factors
offsetting a balance include inaccurate valuation, timing
differences, incomplete reporting, and errors from estimating
procedures.  (For a more complete definition see National Income and
Product Accounts of the United States, Vol.  2, 1959-88, DOC,
Economics and Statistics Administration, BEA (Washington, D.C., Sept. 
1992), p.  M-8.)


      SURVEY OF INCOME AND PROGRAM
      PARTICIPATION (SIPP)
----------------------------------------------------- Appendix 0:0.121

A survey that provides data used to study socioeconomic
characteristics of the population.  It also provides comprehensive
information on the economic resources of the U.S.  population and how
public transfer and tax programs affect the financial circumstances
of members of the population.  SIPP is intended to provide
policymakers with a means to measure the effectiveness and efficiency
of government tax and transfer programs, for estimating future
program costs and coverage, and for assessing the effects of proposed
policy changes.  The survey covers the civilian noninstitutionalized
resident population, and respondents are interviewed every 4 months
over a 2-1/2 year period. 


      SURVEY OF POLLUTION
      ABATEMENT COSTS AND
      EXPENDITURES
----------------------------------------------------- Appendix 0:0.122

Survey of pollution abatement capital expenditures, operating costs,
and costs recovered by private industry. 


      SYSTEM OF NATIONAL ACCOUNTS
      (SNA)
----------------------------------------------------- Appendix 0:0.123

The international standard for the compilation of a nation's economic
accounts and statistical measures.  The revised SNA adopted by the
United Nations in 1993 following a decade-long revision that updated
the 1968 SNA, clarified the standard, and harmonized it more
completely with other sets of international guidelines.  The revised
SNA envisions a comprehensive recording of the stocks and the flows
in the economy.  For each sector that makes up the economy (business,
government, and households), SNA calls for an opening balance sheet
and a closing balance sheet--the stocks of assets and liabilities of
the economy.  And for each sector, current and capital accounts are
designed to record all the transactions and other changes that
explain the differences between the balance sheets--the flows of the
economy. 


      TRANSFER PROGRAMS
----------------------------------------------------- Appendix 0:0.124

Government programs, such as the food stamps program, in which
payments are made to individuals from whom no economic activity is
required in return. 


      VOLUNTARY SURVEYS
----------------------------------------------------- Appendix 0:0.125

Survey respondents are not obligated to complete a survey
questionnaire; they have a free choice to answer a survey or not. 

RELATED GAO PRODUCTS

Economic Statistics:  Measurement Problems Can Affect the Budget and
Economic Policymaking (GAO/GGD-95-99, May 2, 1995). 

Poverty Measurement:  Adjusting for Geographic Cost-of-Living
Differences (GAO/GGD-95-64, Mar.  9, 1995). 

Gross Domestic Product:  No Evidence of Manipulation in First Quarter
1991 Estimates (GAO/GGD-93-58, Mar.  10, 1993). 

