Reliability of Federal Procurement Data (30-DEC-03, GAO-04-295R).
Reliable information is critical to informed decision making and
to oversight of the procurement system. The Federal Procurement
Data System (FPDS) has been the federal government's central
database of information on federal procurement actions since
1978. Congress and executive branch agencies rely on FPDS to
assess the impact that governmentwide acquisition policies and
processes are having on the system generally, as well as with
respect to specific geographical areas, markets, and
socio-economic goals. Yet despite the importance of the data, we
continue to find that FPDS data are inaccurate and incomplete.
Although we have not fully assessed the extent of reporting
errors, we have found sufficient problems to warrant concern
about the current reliability of FPDS information.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-04-295R
ACCNO: A09065
TITLE: Reliability of Federal Procurement Data
DATE: 12/30/2003
SUBJECT: Data bases
Data collection
Data integrity
Federal procurement
Management information systems
Procurement records
Reporting requirements
Procurement policy
GSA Federal Procurement Data System
GSA Federal Procurement Data System-Next
Generation
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO Product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
******************************************************************
GAO-04-295R
United States General Accounting Office Washington, DC 20548
December 30, 2003
The Honorable Joshua B. Bolten
Director, Office of Management and Budget
Subject: Reliability of Federal Procurement Data
Dear Mr. Bolten:
The purpose of this letter is to convey our serious and continuing
concerns with the reliability of the data contained in the Federal
Procurement Data System (FPDS) and to recommend steps to help improve data
reliability as the successor system, FPDS-Next Generation (FPDS-NG), is
implemented.
Reliable information is critical to informed decision making and to
oversight of the procurement system. FPDS has been the federal
government's central database of information on federal procurement
actions since 1978. Congress and executive branch agencies rely on FPDS to
assess the impact that governmentwide acquisition policies and processes
are having on the system generally, as well as with respect to specific
geographical areas, markets, and socio-economic goals.
Yet despite the importance of the data, we continue to find that FPDS data
are inaccurate and incomplete. Although we have not fully assessed the
extent of reporting errors, we have found sufficient problems to warrant
concern about the current reliability of FPDS information. We are pleased
that the administration is moving forward with plans to modernize that
system through FPDS-NG.
The examples following reflect a few of the many instances in which we
either have been unable to assess fully the implementation of procurement
programs or have had to limit our reliance on FPDS data:
o In our 2003 review1 of the simplified acquisition procedures test
program, we found gross errors in the FPDS data. For example, use of the
test program by the reported three largest users in fiscal year 2001 was
either overstated or understated by millions of dollars. We found similar
inaccuracies with the Department of Defense's data system,2 which feeds
information into FPDS.
1 U.S. General Accounting Office, Contract Management: No Reliable Data to
Measure Benefits of the Simplified Acquisition Test Program, GAO-03-1068
(Washington D.C.: Sept. 30, 2003).
2 The Department of Defense uses the Defense Contract Action Data System
(DCADS) to supply data to FPDS.
Page 1 GAO-04-295R Reliability of Federal Procurement Data
Because of the unreliability of the data, we could not determine the
extent to which executive branch agencies used the test program or assess
the benefits realized.
o In our 2001 review3 of the HUBZone program, we found that the value of
contracts awarded to HUBZone firms could have been hundreds of millions of
dollars different than reported. The FPDS categorized 1,034 contracts
worth $325 million as HUBZone awards even though the Small Business
Administration (SBA) had not certified the awardees of those contracts as
HUBZone firms. Conversely, FPDS data did not report as HUBZone awards
1,712 contracts worth $589 million with firms that may have been
SBAcertified at time of contract award. Although the Federal Procurement
Data Center (FPDC) caught some of these inconsistencies through edit
checks, FPDC officials said that they had neither the knowledge to correct
the data nor the authority to require agencies to correct them.
o In our 2003 review of task and delivery orders, 4 we found multiple
orders reported as single transactions. For instance, an order reported as
$11,443,000 in FPDS should have been reported as 87 separate transactions
at or over $25,000 and one cumulative transaction for contracts below
$25,000.
As a result of numerous errors in these and other data elements of FPDS,
we could not determine the extent to which agencies had implemented
regulations effectively, nor could we assess the impact of acquisition
policies governmentwide.
Our concerns about the reliability of FPDS data are not new. In 1980, we
reported that FPDS failed to alert users that the system contained known
errors. In 1994, we reported that the FPDC, which manages FPDS, did not
have standards detailing the appropriate levels of accuracy and
completeness of FPDS data.5
Our work and that of agency inspectors general indicates that many of the
errors in FPDS are due to data entry mistakes by agency contracting
personnel. In some cases, those processing the data did not have a
complete understanding of the information requirements. In other cases,
there was inadequate verification of the information entered into agency
feeder systems, which, in turn, was transferred into FPDS. When we
discussed these problems with agency officials during the course of our
reviews, they cited a lack of training, high personnel turnover, the
complexity of the agency systems, and frequent changes to FPDS data entry
requirements as reasons for the errors.
FPDS-NG became operational effective October 1, 2003. As we understand the
design of the system, FPDS-NG should, if implemented effectively,
eliminate many of
3 U.S. General Accounting Office, Small Business: HUBZone Program Suffers
from Reporting and Implementation Difficulties, GAO-02-57 (Washington
D.C.: Oct. 26, 2001).
4 U.S. General Accounting Office, Contract Management: Civilian Agency
Compliance with Revised
Task and Deliver Order Regulations, GAO-03-983 (Washington D.C.: Aug. 29,
2003).
5 U.S. General Accounting Office, The Federal Procurement Data
System-Making It Work Better,
GAO/ PSAD-80-33 (Washington D.C.: Apr. 18, 1980); and U.S. General
Accounting Office, FPDS
Improvements, GAO/AIMD-94-178R (Washington, D.C.; Aug. 19, 1994).
Page 2 GAO-04-295R Reliability of Federal Procurement Data
the sources of the errors in the current FPDS. At the heart of the new
system are data entry methods that rely less on manual inputs and more on
electronic "machineto-machine" approaches. For example, most agencies are
expected to have computerized contract writing systems that will allow for
direct submission of data to FPDS. Reliability of data is expected to
improve because agency submissions to FPDS-NG will be based on data
already in the contract writing systems, reducing or eliminating separate
data entry requirements. The system provides for immediate data
verification to detect errors. If errors are detected, agency procurement
officials will have the opportunity to correct them immediately while the
information is still readily available.
Information in FPDS-NG can only be as reliable as the information agencies
enter through their own systems. In the long term, data reliability should
improve as agencies fund and implement electronic contract writing
systems. But not all agencies have done so. In the short term, as the
transition to FPDS-NG occurs, agencies that do not have contract writing
systems capable of interfacing directly with FPDS-NG need to review their
procurement data feeder systems and take steps to improve the reliability
of the information contained in those systems.
RECOMMENDATIONS
We recommend that the Director of OMB:
o Ensure that agencies allocate the resources necessary to implement
contract writing systems capable of electronic transfer of information to
FPDS-NG,
o Require agencies that have not yet implemented electronic contract
writing systems to report regularly to OMB on their plans to ensure
reliability of the information reported to FPDS-NG.
o Request that major agencies, in consultation with GSA, conduct regular
reviews of their procedures for collecting and reporting information to
FPDS-NG. Agencies should conduct such reviews annually until a
satisfactory level of reliability is achieved, and periodically
thereafter.
AGENCY COMMENTS
We provided a draft of this letter to the Office of Management and Budget
for comment. An OMB official commented orally that OMB generally agreed
with the recommendations. We also sought comments from the General
Services Administration, which funds and manages the Federal Procurement
Data Center. An official from GSA told us by email that GSA concurred with
the recommendations and hoped they would lead to more accurate reporting
of procurement information to FPDS-NG.
Page 3 GAO-04-295R Reliability of Federal Procurement Data
SCOPE AND METHODOLOGY
The information in this letter is based on previous GAO reviews and
additional work conducted from June to November 2003, in accordance with
generally accepted government auditing standards. In conducting that work,
we reviewed relevant GAO workpapers and GAO and inspector general reports.
We also held discussions with officials at OMB and GSA and reviewed
relevant documents concerning FPDS and FPDS-Next Generation. We did not
conduct an in-depth review of either FPDS or FPDS-NG.
This letter contains a recommendation to the Director of OMB. The head of
a federal agency is required under 31 U.S.C. 720 to submit a written
statement of actions taken on our recommendations to the Senate Committee
on Governmental Affairs and the House Committee on Government Reform not
later than 60 days after the date of this letter and to the Senate and
House Committees on Appropriations with the agency's first request for
appropriations made more than 60 days after the date of this letter.
We are sending copies of this letter to the Chairs and Ranking Members of
the Senate Governmental Affairs Committee, the House Government Reform
Committee, and other interested congressional committees, as well as to
the Administrator of General Services. We will provide copies to others
upon request. This letter also will be available on GAO's home page at
http://www.gao.gov. Please contact me at (202) 5128214 or Hilary Sullivan
at (214) 777-5652 if you or your staff has questions. Major contributors
to this letter were Robert Ackley, Thomas Barger, William Bricking, John
Clary, Barbara Johnson, Ronald Salo, and Robert Swierczek.
Sincerely yours,
William T. Woods
Director
Acquisition and Sourcing Management
(120261)
Page 4 GAO-04-295R Reliability of Federal Procurement Data
*** End of document. ***