Report No.: GAO/OCG-93-12TR Date: December 1992
_______________________________________________________________________________
Author: United States General Accounting Office
Office of the Comptroller General
Addressee: Transition Series
This file contains the text of a GAO report. Delineations within the text
indicating chapter titles, headings, and bullets are preserved. No attempt has
been made to display graphic images or pagination of the typeset report.
Footnotes appear in brackets at the reference point in the text. Underlined
text is indicated by underscore characters (_Introduction_). Superscript
characters are preceeded by a backslash (\a). Figures may be omitted or
replaced with tables. Tables may not resemble those in the printed version.
A printed copy of this report may be obtained from the GAO Documents
Distribution Facility by calling (202) 512-6000 or faxing your request
to (301) 258-4066 or writing to P.O. Box 6015, Gaithersburg, MD 20877.
_______________________________________________________________________________
CONTENTS
Commerce Issues
Focusing Commerce's Role
- Focus on Promoting Competitiveness
Investing in Infrastructure to Ensure Mission Effectiveness
- Significant Investments Needed
- Better Departmental Oversight Is Necessary
Improving Economic Statistics to Support Efforts to Revitalize the U.S.
Economy
- Wide Agreement That Economic Statistics Need Improvement
Planning a More Accurate and Less Costly 2000 Census
- Decline in Census Accuracy Despite Increased Costs
- Census Reform Leadership Needed Now
Related GAO Products
Transition Series
- Economics
- Management
- Program Areas
Office of the Comptroller General
Washington, DC 20548
December 1992
The Speaker of the House of Representatives
The Majority Leader of the Senate
In response to your request, this transition series report discusses major
policy, management, and program issues facing the Congress and the new
administration in the area of commerce. The issues include (1) focusing the
Department of Commerce's role on competitiveness issues, (2) investing in
infrastructure to ensure mission effectiveness, (3) improving economic
statistics to support efforts to revitalize the U.S. economy, and (4) planning
a more accurate and less costly 2000 Census.
The GAO products upon which this report is based are listed at the end of this
report.
We are also sending copies of this report to the President-elect, the
Republican leadership of the Congress, the appropriate congressional
committees, and the Secretary-designate of Commerce.
The Department of Commerce is a large, diverse agency responsible for numerous
programs that cut across core national issues, including trade, technology,
competitiveness, industry, environment, and economic activity. Commerce's
missions include fostering technology, stimulating and regulating
international trade, promoting commerce, analyzing social and economic
activity, and studying the environment and natural resources.
Four years ago, we identified several important areas needing action at
Commerce, including improving export promotion activities and controls,
addressing problems associated with the decennial census, and monitoring
progress on a new automated patent system. These are still relevant issues,
though Commerce has made some improvements in export-related activities.
This report discusses the need for Commerce to (1) focus its missions on
improving competitiveness and play a more significant role among the many
federal agencies that share responsibility for that goal; (2) invest in
infrastructure--which, according to Commerce's estimates, may require at least
$7.4 billion over 15 years, including $4.6 billion for modernizing the
National Weather Service; (3) improve economic statistics to overcome concerns
about their quality and coordination; and
4) change planning and decisionmaking for the 2000 Census to ensure that
accuracy will not continue declining or costs will not continue rising. Census
Bureau planning staff estimated that if the current approach to taking the
census is retained for 2000, the costs could rise to about $4.8 billion in
current dollars, from $2.6 billion in 1990.
FOCUSING COMMERCE'S ROLE
-------------------------------------------------------------------------------
Enhancing the federal government's ability to spur economic growth by
increasing competitiveness is one key to economic security. Commerce, with its
programs in technology, trade, statistics, telecommunications, and industrial
development, has a role in this effort. Its programs are at the heart of
federal efforts to foster both near-term and long-term improvements in the
productivity of the U.S. economy.
The current organization of the federal government may not maximize the
nation's capacity to improve productivity and competitiveness. A number of
congressional bills and private sector reports have offered proposals aimed at
improving the federal approach to this important issue. They propose
restructuring programs and agencies that deal with the major elements of
competitiveness, such as international trade and technology, and generally
suggest significant changes to Commerce's present missions or organizational
structure.
With the growing convergence of sentiment on this issue, both inside and
outside the government, there will likely be attempts to reorder the federal
approach to competitiveness. If it is determined that Commerce should play a
key role in enhancing and promoting competitiveness, the Secretary must be
ready to address two issues that impair Commerce's ability to achieve that
aim. First, Commerce will need to better focus its mission on improving and
promoting competitiveness. Second, Commerce will need to play a more
significant role among the various federal agencies that share
responsibilities for improving competitiveness.
===============================================================================
FOCUS ON PROMOTING COMPETITIVENESS
Commerce is in a unique position to promote competitiveness because it
combines many of the essential elements of government programs necessary to
meet the goal--statistical analysis, international trade, technology
promotion, telecommunications, and economic development. However, Commerce's
Inspector General (IG) has described Commerce as a loose collection of more
than 100 programs delivering services to about 1,000 customer bases. While
many programs are directly related to the goal of supporting commerce and
industry, many others are, at best, only distantly connected to this goal.
The majority of Commerce's limited resources are applied to tasks that do not
have a tangible impact on improving competitiveness. For example, in fiscal
year 1993, an estimated 59 percent of Commerce's budget and 37 percent of its
staff is to be allocated to the National Oceanic and Atmospheric
Administration (NOAA) for weather, oceanic research, and other
activities--which are only tangentially related to improving and promoting
competitiveness. In this environment, it is difficult to apply sufficient
resources and attention to all necessary activities and mission priorities.
If Commerce is to play a key role in enhancing and promoting competitiveness,
the Secretary will need to formulate and set forth a strategy that taps into
Commerce's inherent potential toward this end. At the outset of developing a
strategy, the Secretary will need to carefully examine current components of
Commerce and work with the Office of Management and Budget (OMB) and Congress
to consider spinning off or eliminating those components that do not fit,
while clarifying the roles and interconnections among those that do. Further,
because the programs in Commerce do not encompass all the essential elements
of improving competitiveness and productivity, the strategy should ensure that
Commerce's programs are consistent with other agencies' efforts.
_______________________________________________________________________________
Playing a More Significant Role in Improving Competitiveness
The Secretary will also need to address the fundamental problem that Commerce
lacks the prominence and resources to play a significant role in improving
competitiveness. While many federal departments share mission-related
functions and programs with other departments and agencies, Commerce likely
faces the most complex web of divided authorities. For example, Commerce
shares its missions with at least 71 federal departments, agencies, and
offices. Further, Commerce has traditionally taken a back seat to departments
that had more prominent status, greater resources, and higher national
priority in the Cold War world.
Now, in the post-Cold War era, the issues traditionally administered by
Commerce have gained new priority and could push Commerce to a potential
leadership position. Yet Commerce does not control a majority of the resources
devoted to missions it shares with other agencies. For example, export
promotion programs are distributed among 10 agencies. The U.S. Department of
Agriculture, not Commerce, receives about 74 percent of total funding for
these programs, although it accounts for only about 10 percent of U.S.
exports.
If the incoming administration and the Congress determine that Commerce is to
play a central role in improving competitiveness, then Commerce should occupy
a more central position in the system of federal programs meant to achieve
this end. To do so, the Secretary needs to work with the President, other
Secretaries, and the Congress either to (1) secure clearer lines of
coordinating authority over related programs in other agencies or (2) relocate
those related programs into Commerce.
INVESTING IN INFRASTRUCTURE TO ENSURE MISSION EFFECTIVENESS
-------------------------------------------------------------------------------
The ability of Commerce to effectively achieve its missions--especially those
related to enhancing national competitiveness and measuring and analyzing the
environment--are largely dependent on the state of Commerce's technical and
research facilities. However, in the past, inadequate attention to strategic
issues and long-term responsibilities have left Commerce with some critical
infrastructure needs requiring immediate attention, including long overdue
capital investments and a lack of needed technology.
Past funding problems, shortcomings in departmental oversight, and planning
inconsistencies have left Commerce in a position of needing to invest heavily
in its infrastructure. Commerce estimates that infrastructure currently in
disrepair--including federal laboratories, a fleet of ships, weather
satellites and radars, information systems, and other facilities and
equipment--will require investments of at least $7.4 billion over a 15-year
period. Regardless of whether they remain in Commerce or become candidates for
relocation elsewhere, these infrastructure needs merit prompt attention. For
example,
-- The National Institute of Standards and Technology (NIST) has requested
$607 million over 10 years to begin overdue renovation and replacement of
laboratories and other research facilities that are between 25 and 40
years old. NIST's mission to advance scientific and technological
progress is already hampered by obsolete technical and
environmental systems, unreliable power supplies at research
buildings, and major safety and systems capacity problems.
-- NOAA is at risk of not being able to effectively carry out current and
future ocean research activities. Several of NOAA's vessels are currently
inactive, and the remainder of the fleet is obsolete or deteriorating
rapidly. However, NOAA's ability to successfully fulfill its mission
depends on its having fully operational vessels or access to
available charter vessels. NOAA estimates it will require about
$1.5 billion over 15 years to modernize its fleet.
-- The National Weather Service (NWS) is to provide weather forecasting and
warning services for the protection of life and property. However, much of
its current equipment is outdated, expensive to maintain, and difficult
to repair. For these reasons, NWS has embarked upon a comprehensive
$4.6 billion modernization of equipment, facilities, and related
information systems. This modernization will help to ensure
NWS' ability to monitor critical weather data, forecast
potential weather disasters, and assess environmental
conditions.
-- The Patent and Trademark Office (PTO) is in the midst of a 14-year, $700
million effort to automate its patent system. Patent information is
essential because it is the largest single source of information on
emerging technologies.
===============================================================================
BETTER DEPARTMENTAL OVERSIGHT IS NECESSARY
Although the funding for some of the needed enhancement of infrastructure is
now programmed, Commerce will need to pay particular attention to how these
improvements are managed. Past oversight problems of procurement and system
development have left Commerce vulnerable to overruns, inefficiencies, and
suboptimal achievement of its goals. For example,
-- Commerce's accounting and financial management systems are incapable of
providing managers with reliable information necessary for long-term
planning. Commerce has characterized the systems as archaic, unable to
produce reliable information, inadequately controlled, and often not
used by managers. Both Commerce and OMB have identified these as
high-risk areas, meaning that they are particularly vulnerable
to fraud, waste, or mismanagement.
-- NOAA's Next Generation Geostationary Operational Environmental Satellite
system has encountered problems that have resulted in substantial cost
growth, schedule slippage, and potentially reduced satellite capability.
This put Commerce in the position of having to borrow a satellite
from the European satellite consortium to ensure that NWS will be
able to meet minimum mission requirements of protecting public
safety through the forecasting of severe storms and issuance
of warnings. Commerce reported to OMB that this is
currently a high-risk problem.
-- The NWS modernization program has exceeded its expected cost and is far
behind schedule. The initial cost estimate of nearly $2 billion has risen
to $4.6 billion, and the projected date for completing the planned
upgrades has slipped from 1994 to 1998. Contract-related deficiencies
have been identified by us, OMB, and Commerce's IG as one of the
major contributors to the delay and cost overrun. Satellite and
radar systems have been especially problematic in the past,
and close contractor oversight of information system
development is vital to preventing similar problems in
that phase of the project. The NWS modernization
effort has been reported as a high-risk area
several times and is still on the high-risk
list.
-- After investing nearly 10 years and $417 million, PTO is unable to ensure
that efforts to automate its patent system will meet expected benefits and
fulfill users' needs. Some progress has been made in automating paper
files and manual processes, but PTO's estimated completion dates and
development costs continue to change.
Commerce's scientific and technical facilities are vital to improving national
competitiveness or to protecting human life and property in the event of
severe weather conditions. The incoming Secretary will need to pay close
attention to Commerce's infrastructure to ensure that critical facilities,
scientific equipment, and information systems are upgraded as efficiently and
as timely as possible. This will require the Secretary to work with the
Congress and OMB to identify and build resources for those high-priority areas
demanding greater funding and staffing requirements. To prevent future
problems and ensure continued attention to infrastructure and other
organizational capacity needs, the Secretary will also need to (1) ensure
adequate departmental oversight and accountability for procurement and systems
development projects, (2) place a high priority on improving financial
management systems, and (3) provide continuing leadership in developing more
effective strategic analysis and planning systems for all Commerce units.
IMPROVING ECONOMIC STATISTICS TO SUPPORT EFFORTS TO REVITALIZE THE U.S.
ECONOMY
-------------------------------------------------------------------------------
Many economists and policymakers have voiced concerns about the quality of the
nation's economic statistics and their failure to keep pace with the changing
U.S. and world economy. Without early attention to this problem, the new
administration may find itself hampered by a lack of reliable and timely
economic data as it attempts to revitalize the U.S. economy.
===============================================================================
WIDE AGREEMENT THAT ECONOMIC STATISTICS NEED IMPROVEMENT
Concerns about the quality of economic statistics have been long-standing and
bipartisan. The Carter administration complained that the economic information
base appeared "to provide inadequate support for policy decisions." The
Council of Economic Advisers noted in 1990 that structural changes were
"making it increasingly difficult to track the course of the economy
accurately" to the point where "many analysts question the accuracy of
measurement of even the most basic variables, such as output and inflation."
For example, we noted in a 1989 report that there is a strong possibility that
exports are underreported in merchandise trade statistics. As a result, the
U.S. trade deficit may have been smaller than reported.
The National Academy of Sciences concluded in a recent book that the existing
economic statistics system does not provide a comprehensive picture of U.S.
involvement in the global economy and that the government needs to devote more
attention and resources to increasing the coverage and quality of
international economic statistics.
_______________________________________________________________________________
Decentralized Statistical System and Lack of Leadership Create Special
Demands
While responsibility for supervising the organization and coordination of the
federal statistical system lies with OMB, resources devoted to carrying out
these responsibilities have been greatly reduced. The lack of OMB coordination
of the statistical system puts added demands on Commerce and its two
statistical agencies, the Census Bureau and the Bureau of Economic Analysis
(BEA), which calculates the gross domestic product (GDP) and other economic
indicators. BEA, in particular, relies extensively on source data provided by
other government agencies in producing its statistical reports on the nation's
economy.
A November 1992 Congressional Research Service report on the coordination of
federal economic statistics characterizes the current situation as being like
"an opera without a conductor." Without a strong OMB role in coordination,
Commerce must play an active part not only in coordinating the activities of
the Census Bureau with BEA but also in negotiating governmentwide solutions to
governmentwide problems with economic statistics.
A quick review of the specific problems identified during the formulation of
the economic statistics initiatives is a good place for the new Secretary of
Commerce to become familiar with the issues facing federal economic statistics
programs. These budget initiatives called for real increases in the budgets of
BEA, the Census Bureau, and other economic statistical agencies. Although the
near-term solutions emphasized the need for additional resources, the
long-term plans were to examine organizational, methodological, and other
overall improvements. A search for systemic solutions seems appropriate given
the systemic problems, such as the lack of overall coordination, facing
federal economic statistical agencies.
_______________________________________________________________________________
Commerce Must Ensure Integrity and Credibility of Economic Data
Press reports in 1991 and 1992 have called into question the integrity and
credibility of economic statistics. We examined several allegations of
manipulation of GDP and personal income estimates for the first quarter of
fiscal year 1991 and found no evidence of the alleged manipulation. However,
the very existence of these allegations poses a credibility threat to the
federal statistical system. An October National Academy of Sciences' report
said, "It is essential that an agency maintain credibility for itself and for
its data and information. In particular, an agency must be perceived to be
free of political interference and policy advocacy."
PLANNING A MORE ACCURATE AND LESS COSTLY 2000 CENSUS
-------------------------------------------------------------------------------
Reforming the decennial census will be a very important task for the new
Secretary of Commerce. The results of the 1990 decennial census indicated that
the current approach appears to have exhausted its potential for counting the
population at an acceptable level of accuracy at reasonable cost.
Innovative approaches to the census are possible, but their adoption requires
prompt policy guidance and leadership from senior Commerce and Census Bureau
officials early in 1993. Without decisive leadership at the start of the new
administration, sufficient time will not be available to build the political
consensus and public support necessary for major census reforms and to test
and integrate the many interwoven procedures and technologies that constitute
an undertaking as big and complex as the decennial census.
===============================================================================
DECLINE IN CENSUS ACCURACY DESPITE INCREASED COSTS
In 1990, accuracy fell below that of the 1980 Census even as census costs
escalated significantly. A broad, overall measure of census accuracy--the net
undercount (the percentage of the population not counted)--had declined each
census since it was first measured in 1940. However, the official estimate of
the net undercount rose to 1.8 percent in 1990, from 1.2 percent in 1980. A
particularly disturbing component of this result was that the difference in
the net undercount between blacks and non-blacks increased in 1990 to its
highest level since 1940. Furthermore, the 1990 Census appears to have missed
substantially more persons than suggested by the official net undercount
estimate. Overall, we estimated that it missed a minimum of 9.7 million
persons--about 3.9 percent of the count--and contained a minimum of 14.1
million gross errors, including double counts and fictitious enumerations.
The decline in the accuracy of the census cannot be attributed to spending
less than was spent in 1980. The 1980 Census cost $1.1 billion over 10 years,
while the 1990 Census is estimated to have cost $2.6 billion. Even adjusting
for inflation and workload growth, the cost of the 1990 Census was still 25
percent higher than the 1980 Census. Census Bureau planning staff estimated
that if the current approach to taking the census is retained for 2000, the
costs could rise to about $4.8 billion in current dollars.
===============================================================================
CENSUS REFORM LEADERSHIP NEEDED NOW
Important decisions based on wide consultation and staff analysis are needed
by the fall of 1993 to guide planning for field tests of census methods and
procedures scheduled for 1995, to shape budget and operational planning for
the rest of the census cycle, and to guide discussions with state and local
governments and other interested groups. The administration must also seek
early and active consultation and cooperation with the Congress.
_______________________________________________________________________________
Early Decisions Needed on Content of Census Questionnaire
Public cooperation is essential to obtaining accurate census data at a
reasonable cost. For reasons not entirely under the Census Bureau's control,
the mail response rate has been declining. In the future, this trend may
continue because the environment for census-taking by the traditional mail-out
and mail-back method may worsen. Although the Census Bureau cannot expect to
solve the problems created by societal changes, it can reduce the effects of
these problems by streamlining the questionnaire, thus making it easier for
the public to cooperate with the census.
In 1992, the Census Bureau tested a simplified, more user-friendly short
form--sent to about five of six households--and found that changes in the
short form also can increase the level of cooperation with the census. For
example, the test results indicated that generally asking fewer questions can
improve the mail response rates. As an indication of the magnitude of
potential savings, the Census Bureau has estimated that a streamlined
questionnaire in 1990 asking only five questions could have saved as much as
half a billion dollars. The Census Bureau also should examine the need for the
census long form, which had 59 questions, and the size of the sample to which
it was sent--about 1 in 6 households. Not unexpectedly, the response rate in
1990 for the census long form was lower (60 percent) than that of the short
form (66 percent).
An early decision on census content is required for political and operational
reasons. A political consensus must be built to support a simplified census
questionnaire or to ensure continuing support from the Congress and the public
for a census with the traditional scope and content. An early decision also
will provide time to find necessary substitutes for any data lost by
streamlining the census questionnaire. A late decision either to add or
eliminate a large block of questions poses significant operational risks to
the census.
_______________________________________________________________________________
Early Decision Needed on Whether to Use Sampling
Even with a simplified questionnaire, the Census Bureau's test showed that it
can expect a large number of Americans will still not return their form by
mail. Statistical sampling of these nonrespondents would reduce the time and
labor-intensive fieldwork required to follow up missing questionnaires. We,
the National Academy of Sciences, and the Commerce IG have recommended that
the Census Bureau explore if and how sampling can be used to improve the
census. Census Bureau staff estimated that the cost of the 1990 census would
have been reduced by $460 million if the Census Bureau had sampled 10 percent
of the nonrespondents rather than collecting data on all of them. In addition
to these cost savings, sampling will reduce the total respondent burden and
even potentially increase the data quality.
Sampling for nonrespondents will require early consideration by the Census
Bureau and other key policymakers inside and outside Commerce. Legislation may
be needed to allow the Census Bureau to sample those housing units not
returning a form by mail. Because of its critical nature to census planning
and the need for operational testing, legislation to permit sampling, if
necessary, should be proposed to and considered by the Congress as
expeditiously as possible.
_______________________________________________________________________________
Early Decisions Needed to Improve Address List and Geographic Information
An accurate and complete address list and associated geographic information is
a cornerstone of a successful mail-out census. For the 1990 Census, the Census
Bureau spent over a half a billion dollars to develop the means to identify
the mailing address and physical location of about 100 million housing units.
But even such an enormous expenditure of resources did not guarantee an
accurate and complete address list. The Census Bureau estimated that it missed
about 3.5 million housing units in 1990 and that about 2 million were
erroneously counted. Furthermore, we estimated that the Census Bureau spent
over $300 million visiting or searching for vacant or nonexistent housing
units.
The Census Bureau has major opportunities to improve its address list for the
2000 Census, but it must begin soon the process of updating, upgrading, and
integrating its 1990 automated address control file and the associated
automated geographic systems. In particular, an early decision is needed on
the important role the U.S. Postal Service (USPS) might play in these
activities. We believe that letter carriers are in a much better position than
temporary, inexperienced census workers to identify housing units and
determine whether they are occupied. Moreover, recent USPS technological
advances in building an automated national address list reinforce the
practicality of USPS' increased involvement in updating the Census Bureau's
address list and geographic information systems. Early decisions are necessary
because cooperative arrangements between the Census Bureau and USPS will
require extensive lead times to negotiate, plan, test, and prepare.
