Davis-Bacon Act: Process Changes Could Address Vulnerability to Use of
Inaccurate Data in Setting Prevailing Wage Rates (Testimony, 06/20/96,
GAO/T-HEHS-96-166).
GAO discussed the vulnerabilities in the Department of Labor's
prevailing wage determination process under the Davis-Bacon Act. GAO
noted that: (1) Labor sets prevailing wage rates for numerous job
classifications in about 3,000 geographic areas; (2) Labor's wage
determination process depends on employers' and third parties' voluntary
participation in a survey that reports wage and fringe benefits paid for
similar jobs on comparable construction projects in a given area; (3)
due to limited resources, Labor concentrates on those geographical areas
most in need of wage rate revisions; (4) Labor wage determinations can
be appealed by any interested party; (5) process weaknesses include
limited data verification, limited computer capabilities to detect
erroneous data, and the lack of awareness of the appeals process; (6)
erroneous data could result in setting the wage rate too low so that
construction workers are underpaid or setting the wage rate too high so
that the government incurs excessive construction costs; (7) Labor
initiatives to improve its rate-setting process include having employers
verify certain third-party data, informing survey respondents of the
serious consequences of willfully falsifying wage data, and proposing a
long-term strategy to review the entire wage determination process; and
(8) Labor should immediately improve its verification of employer wage
data and make the appeals process more effective.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: T-HEHS-96-166
TITLE: Davis-Bacon Act: Process Changes Could Address
Vulnerability to Use of Inaccurate Data in Setting
Prevailing
Wage Rates
DATE: 06/20/96
SUBJECT: Wage surveys
Minimum wage rates
Data collection operations
Construction contracts
Labor law
Data integrity
Appellate procedure
Construction costs
Fringe benefits
Labor statistics
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Before the Subcommittee on Oversight and Investigations and the
Subcommittee on Workforce Protections, Committee on Economic and
Educational Opportunities, House of Representatives
For Release on Delivery
Expected at 10:00 a.m.
Thursday, June 20, 1996
DAVIS-BACON ACT - PROCESS CHANGES
COULD ADDRESS VULNERABILITY TO USE
OF INACCURATE DATA IN SETTING
PREVAILING WAGE RATES
Statement of Carlotta C. Joyner, Director
Education and Employment Issues
Health, Education, and Human Services Division
GAO/T-HEHS-96-166
GAO/HEHS-96-166T
(205328)
Abbreviations
=============================================================== ABBREV
ESA - Employment Standards Administration
WD-10 - Labor's form WD-10; WD should be small caps.
WHD - Wage and Hour Division
DAVIS-BACON ACT: PROCESS CHANGES
COULD ADDRESS VULNERABILITY TO USE
OF INACCURATE DATA IN SETTING
PREVAILING WAGE RATES
============================================================ Chapter 0
Mr. Chairmen and Members of the Subcommittees:
We are pleased to be here today to discuss the results of recent GAO
work on the vulnerabilities in Labor's prevailing wage determination
process under the Davis-Bacon Act.\1 The Congressional Budget Office
estimates that for fiscal year 1996, the federal government will
contract $42 billion in construction projects throughout the nation.
Given the magnitude of these expenditures, inaccurate wage
determinations could lead either to excessive government construction
costs or to large numbers of workers receiving wages and fringe
benefits that are lower than required by the law.
In early 1995, allegations of the use of inaccurate and fraudulent
wage data to determine the prevailing wages paid on federally funded
construction projects in the Oklahoma City area precipitated a
criminal investigation by the Department of Justice that is still
ongoing. Because of your concerns that Labor's processes may allow
the use of data that could result in inaccurate wage determinations,
my comments today will address three questions:
-- What process does Labor use to collect data and determine
prevailing wages?
-- Are there weaknesses in that process that could allow the use of
fraudulent or otherwise inaccurate data?
-- To what extent is Labor addressing any identified weaknesses in
the process?
My remarks are based on the GAO report completed for your
Subcommittees on May 31, 1995.\2 In summary, we found that Labor's
wage determination process is based on voluntary participation by
employers and third parties in surveys that report wage and fringe
benefits data on construction projects. Using this information,
Labor sets prevailing wage rates for job classifications in four
different types of construction in about 3,000 counties or groups of
counties. Any interested party can appeal the final wage
determinations. This process does contain weaknesses that could
permit the use of fraudulent or inaccurate data in the setting of
prevailing wage rates. First, verification of wage and fringe
benefits data is largely limited to telephone contacts. Second,
limited computer capabilities hinder Labor's ability to detect
erroneous data. Third, a lack of awareness about the appeals process
may limit its effectiveness. If these weaknesses allow erroneous
data to be used, the result may be in either of two directions: if
the wage rate is set too low, construction workers may be paid less
than the amount to which they are entitled; if the rate is set too
high, the government may pay excessive construction costs. Labor has
developed some short- and long-term initiatives to improve the
prevailing wage determination process. In addition, it has agreed to
implement recommendations we made in our report, which will increase
confidence that the wage rates are based on accurate data.
--------------------
\1 Labor's regulations define a prevailing wage as the wage paid to
the majority (more than 50 percent) of the workers in the job
classification on similar projects in the area during the period in
question. If the same wage is not paid to a majority of those
employed in the classification, the prevailing wage will be the
average of the wages paid, weighted by the total employed in the
classification.
\2 See Davis-Bacon Act: Process Changes Could Raise Confidence That
Wage Rates Are Based on Accurate Data (GAO/HEHS-96-130, May 31,
1996). Our study focused on whether Labor's wage determination
process was vulnerable to the use of inaccurate data. We did not
attempt to determine the extent to which Labor's prevailing wage
determinations have actually been based on inaccurate data.
BACKGROUND
---------------------------------------------------------- Chapter 0:1
The Davis-Bacon Act requires workers on federal construction projects
valued in excess of $2,000 to be paid, at a minimum, wages and fringe
benefits that the Secretary of Labor determines to be prevailing for
corresponding classes of workers in the locality where the contract
is to be performed. The act covers every contract to which the
United States or the District of Columbia is a party for
construction, alteration, or repair of public buildings or public
works.
Labor's Wage and Hour Division (WHD), within Labor's Employment
Standards Administration (ESA), has responsibility for administering
the Davis-Bacon Act. Approximately 50 staff in the Washington, D.C.,
headquarters and in six regional offices are involved in the wage
determination process. Two other Labor offices are sometimes
involved in the administration of Davis-Bacon: Labor's
Administrative Review Board hears appeals of prevailing wage
determinations, and the Office of the Solicitor provides legal advice
and assistance to Labor personnel relative to the act and represents
WHD in Davis-Bacon wage determination cases before the Administrative
Review Board.
In previous reviews of the Davis-Bacon Act,\3 we have expressed
concerns about the accuracy of Labor's wage determinations. Labor
has made regulatory changes that addressed some of our specific
concerns about the process used to determine prevailing wages.
Technological changes have also improved Labor's ability to
administer the wage determination process. Despite these changes,
however, we reported in 1994 that data verification problems still
existed.
--------------------
\3 The Davis-Bacon Act Should Be Repealed (GAO/HRD-79-18, Apr. 27,
1979) and Davis-Bacon Act (GAO/HEHS-94-95R, Feb. 7, 1994). We also
discussed changes to the Davis-Bacon Act in a more recent report,
Addressing the Deficit: Budgetary Implications of Selected GAO Work
for Fiscal Year 1996 (GAO/OCG-95-2, Mar. 15, 1995), and identified
reforming or repealing the Davis-Bacon Act as an option that the
Congress might wish to consider.
LABOR'S WAGE DETERMINATION
PROCESS BASED ON VOLUNTARY
SURVEY PARTICIPATION
---------------------------------------------------------- Chapter 0:2
In setting prevailing wages, Labor's task is to determine and issue
prevailing wage rates in a wide range of job classifications in each
of four types of construction (building, residential, heavy, and
highway) in more than 3,000 counties or groups of counties. It also
needs to update these wage determinations frequently enough that they
continue to represent the prevailing wages.
Labor's process for determining the wage rates is based primarily on
a survey of the wages and fringe benefits paid to workers in similar
job classifications on comparable construction projects in the
particular area.\4 This information is submitted voluntarily by
employers and third parties. Labor encourages the submission of wage
information from all employers and third parties, including employee
unions and industry associations that are not directly involved with
the surveyed projects.
Although an individual wage survey typically covers only one kind of
construction, most surveys gather information on projects in more
than one county. In fiscal year 1995, Labor completed 104 survey
efforts resulting in wage determinations for over 400 counties.\5
The wage determination process consists of four basic stages:
planning and scheduling surveys, conducting the surveys, clarifying
and analyzing respondents' wage data, and issuing the wage
determinations. In addition, any employer or interested party who
wishes to contest or appeal Labor's final wage determination can do
so.
Given the large number of prevailing wage determinations and Labor's
limited resources, Labor develops an annual plan to identify those
geographic areas or counties for which wage determinations are most
in need of revision. For each area designated for survey, Labor
identifies the counties for which the wage determination should be
conducted and determines what construction projects will be surveyed.
The work of conducting the surveys and clarifying and analyzing the
data is done by about 30 staff distributed among six regional
offices. The survey is distributed to the participant population,
which includes the general contractor for each construction project
identified as comparable and within the survey's geographic area. In
surveying the general contractors, Labor requests information on
subcontractors to solicit their participation. Labor also surveys
interested third parties, such as local unions and construction
industry associations that are located or active in the survey area.
Once the data submissions are returned, the analysts review and
analyze the returned survey forms. They follow up with the employer
or third parties to clarify any information that seems inaccurate or
confusing. The analysts then use this information to create
computer-generated recommended prevailing wages for key construction
job classifications. The recommended prevailing wages are reviewed
and approved by Labor's National Office in Washington, D.C. Labor
publishes the final wage determinations in printed reports and on its
electronic bulletin board.
The opportunity to appeal a final wage determination is available to
any interested party at any time after the determination is issued.
For example, appeals could come from contractors, contractor
associations, construction workers, labor unions, or federal, state,
or local agencies. Appeals may take the form of informal inquiries
resolved at the regional office level or formal requests for
reconsideration that are reviewed at the regional office or the
National Office and may be appealed to the Administrative Review
Board for adjudication.
--------------------
\4 For those counties where the type of construction is represented
by a collective bargaining agreement, prevailing wage rates are
typically revised through the use of newly negotiated agreements,
rather than through a survey.
\5 Given that prevailing wages are set for multiple job
classifications in about 12,000 combinations of construction type and
geographical areas, most wage determinations are not changed
annually. Labor estimated that the average age of a wage
determination for a county dominated by nonunion wage rates in fiscal
year 1995 was about 7 years.
WEAKNESSES IN LABOR'S WAGE
DETERMINATION PROCESS COULD
LEAD TO INACCURATE PREVAILING
WAGE RATES
---------------------------------------------------------- Chapter 0:3
Labor's wage determination process contains weaknesses that could
permit the use of fraudulent or inaccurate data for setting
prevailing wage rates. These weaknesses include limitations in the
degree to which Labor verifies the accuracy of the survey data it
receives, limited computer capability to review wage data before
calculating prevailing wage rates, and an appeals process that may
not be well publicized to make it accessible to all interested
parties. Wage determinations based on erroneous data could result in
wages and fringe benefits paid to workers that are higher or lower
than the actual prevailing rates.
VERIFICATION OF WAGE DATA
LARGELY LIMITED TO TELEPHONE
CONTACTS
-------------------------------------------------------- Chapter 0:3.1
Labor's regional staff rely primarily on telephone responses from
employers or third parties to verify the information received on
Labor's WD-10 wage reporting forms. Regional office staff told us
that most of the verification--clarifications concerning accuracy,
appropriateness, or inclusion--was done by telephone. Labor's
procedures also do not require and Labor staff rarely request
supporting documentation--for example, payroll records--to supplement
the information on the forms submitted by employers. Labor officials
and staff told us that if an employer insists that the wages reported
are accurate, the wage analyst generally accepts that statement.
It is because of resource constraints, according to Labor
headquarters officials, that verification is limited to telephone
contacts without on-site inspections or reviews of employer payroll
records to verify wage survey data. In recent years, Labor has
reduced the number of staff allocated to Davis-Bacon wage-setting
activities. For example, the number of staff in Labor's regional
offices assigned to the Davis-Bacon wage determination process--who
have primary responsibility for the wage survey process--decreased
from a total of 36 staff in fiscal year 1992 to 27 staff in fiscal
year 1995. Labor officials in one region also told us that staff had
only received two work-related training courses in the last 6 years.
Labor's regional staff told us that the staff decline has challenged
their ability to collect and review wage survey data for accuracy and
consistency.
LIMITED COMPUTER
CAPABILITIES HINDER
DETECTION OF ERRONEOUS DATA
-------------------------------------------------------- Chapter 0:3.2
Labor's administration of the Davis-Bacon wage determination process
is also hampered by limited computer capabilities. Labor officials
reported a lack of both computer software and hardware that could
assist wage analysts in their reviews. Instead, they said that
analysts must depend on past experience and eyeballing the wage data
for accuracy and consistency. For example, Labor offices do not have
computer software that could detect grossly inaccurate data reported
in Labor's surveys. Regional staff reported only one computer edit
feature in the current system that could eliminate duplicate entry of
data received in the wage surveys. As a result, several review
functions that could be performed by computers are conducted by
visual reviews by one or more wage analysts or supervisory wage
analysts in Labor's regional offices.
Labor's ability to review wage survey data is also hindered by a lack
of up-to-date computer hardware. For example, in the Atlanta and
Philadelphia regional offices, most of the computer hardware is
outdated. In these offices, because of the computers' limited memory
and storage capabilities, Labor staff told us that they are unable to
store historical data on prior wage determinations that would allow
wage analysts to compare current with prior recommendations for wage
determinations in a given locality.
These limitations could be significant given the large number of
survey forms received and the frequency of errors on the WD-10
reporting forms. In fiscal year 1995, Labor received wage data on
about 75,000 WD-10 wage reporting forms; these were from over 37,000
employers and third parties, some of whom provided information on
multiple construction projects. Labor staff reported that
submissions with some form of data error were quite common. The
frequency of errors could be caused in part by employer confusion in
completing the wage reporting forms. Depending on the employer's
size and level of automation, completing the WD-10 reporting forms
could be somewhat difficult and time consuming. For example,
employers must conduct so-called peak week calculations where they
must not only compute the hourly wages paid to each worker who was
employed on the particular project in a certain job classification
but also do so for the time period when the most workers were
employed in each particular job classification. We were told that
this can be especially difficult for many smaller, nonunion
employers.
Although Labor staff reported that wage surveys with data errors are
fairly common, agency officials believe that it is very unlikely that
erroneous wage data went undetected and were used in the prevailing
wage determination. They said that a key responsibility of Labor's
wage analysts is to closely scrutinize the WD-10 wage reporting forms
and contact employers as necessary for clarification. Labor
officials contended that, over time, this interaction with employers
and third parties permitted Labor staff to develop considerable
knowledge of and expertise in the construction industry in their
geographic areas and to easily detect wage survey data that are
inaccurate, incomplete, or inconsistent.
LACK OF AWARENESS OF THE
APPEALS PROCESS MAY LIMIT
ITS EFFECTIVENESS
-------------------------------------------------------- Chapter 0:3.3
Labor's appeals process could provide an important safeguard against
reliance on inaccurate data in that it allows any interested party to
question the validity of the determinations. But our review suggests
that this mechanism is not understood well enough to serve its
purpose.
The appeals process has both informal and formal aspects. An
interested party may make informal inquiries by telephone or in
writing for quick resolution of questions about wage determinations.
Labor reported that most inquiries on its wage determinations are
informal and are generally resolved quickly over the telephone at the
regional offices. If an informal inquiry is not resolved to the
satisfaction of the interested party, he or she may submit a formal
request for reconsideration to either the regional or National
Office. A formal request for reconsideration of a wage determination
must be in writing and accompanied by a full statement of the
interested party's views and any supporting wage data or other
pertinent information. A successful request for reconsideration
typically results in Labor modifying an existing determination or
conducting a new wage survey. An interested party may appeal an
unsuccessful request--that is, one in which he or she is dissatisfied
with the decision of the WHD Administrator--to Labor's Administrative
Review Board for adjudication. Labor officials said it is extremely
rare for anyone to appeal formal requests for reconsideration of a
determination to the Board, reporting that there had been only one
such case in the last 5 years.
The infrequency of formal appeals to the Board can be interpreted in
more than one way. Labor officials interpreted this record to mean
that there is little question about the accuracy and fairness of the
prevailing wage determinations issued. Alternatively, this could
reflect interested parties' lack of awareness of their rights and the
difficulty they face in collecting the evidence necessary to sustain
a case. Representatives of construction unions and industry trade
associations told us that employers were generally unaware of their
rights to appeal Labor's final wage determinations. Officials with a
state Labor Department also told us that, even if an interested party
wanted to appeal a wage determination to the National Office and the
Administrative Review Board, the effort it takes to independently
verify wage data submissions could discourage such an action. They
reported that it took a state investigation team a full month to
gather information to support the need for Labor to reconsider some
wage determinations--and that involved investigating and verifying
the information for only three construction projects. A private
employer or organization wishing to appeal a determination might
experience similar difficulties.
CONSEQUENCES OF WAGE
DETERMINATIONS BASED ON
ERRONEOUS DATA
-------------------------------------------------------- Chapter 0:3.4
Wage determinations based on erroneous data could result in workers
being paid higher or lower wages and fringe benefits than those
prevailing on federal construction projects. Higher wages and fringe
benefits would lead to increased government construction costs. On
the other hand, lower wages and fringe benefits would result in
construction workers being paid less than is required by law.
Although they considered it unlikely, Labor officials acknowledged
that, in general, there could be an incentive for third parties,
particularly union contractors, to report higher wages than those
being paid on a particular construction project. By reporting higher
wages, they could influence the prevailing wages in a local area
toward the typically higher union rate.
The use of inaccurate data could also lead to lower wages for
construction workers on federal projects than would otherwise be
prevailing. Labor officials acknowledged that an employer in a
largely nonunion area who had been paying lower than average wages
would have an incentive to "chisel" or report wages and fringe
benefits levels somewhat lower than what he or she was actually
paying, in an attempt to lower the Davis-Bacon rate. However,
officials also said that it is much more likely for some employers to
report data selectively in an effort to lower the prevailing wage
rate. For example, a contractor may only submit data on those
projects where the wages paid were relatively low, ignoring projects
where a somewhat higher wage was paid.
In addition, the wages required under the Davis-Bacon Act have
implications for construction projects other than those specifically
covered by the act. Industry association members and officials told
us that in several parts of the country, employers, especially
nonunion contractors, paid wages on their private projects below the
prevailing wage levels specified by the Davis-Bacon Act in their
areas.\6 These officials told us that this differential sometimes
proved problematic for contractors in retaining their skilled labor
force. An official of an employer association told us, for example,
that an employer who successfully bid on a Davis-Bacon contract but
who typically paid wages below the prevailing rate would be required
to pay the workers employed on the new project at the higher
Davis-Bacon wage rates. Depending on the local labor market
conditions, when the project was completed, these workers typically
received their pre-Davis-Bacon, lower wages and fringe benefits on
any future work. In such cases, some employees became disgruntled,
believing that they were being cheated, and may have suffered lower
morale that sometimes led to increased staff turnover. Depending on
local labor market conditions, if the employer did not bid on the
Davis-Bacon project, he or she could still be affected if the
employer's skilled workers quit to search for work on the new, higher
wage federally funded project.
--------------------
\6 In at least some parts of the country, nonunion wages determine
the vast majority of prevailing wage determinations. For example, in
Labor's Atlanta region, nonunion wage rates dominated 72 percent of
all existing wage determinations. In these determinations,
prevailing wages were determined by a weighted average of the
submitted wage data.
LABOR ACKNOWLEDGES WEAKNESSES
OF ITS WAGE DETERMINATION
PROCESS
---------------------------------------------------------- Chapter 0:4
Labor has acknowledged the weaknesses of its current wage
determination process and it has proposed both short- and long-term
initiatives to improve the accuracy of the data used to make
prevailing wage determinations.
One recent change improves the verification process for data
submitted by third parties. In August 1995, Labor began requiring
its wage analysts to conduct telephone verifications with the
employer on all third-party data that appear to be inaccurate or
confusing. In addition, the new policy requires analysts to verify
with the employers at least a 10-percent sample of third-party data
that appear to be accurate.
Labor has also proposed a change that would specifically inform
survey respondents of the possible serious consequences of providing
false data, since it is a crime under federal law to knowingly submit
false data to the government or use the U.S. mail for fraudulent
purposes. In February 1996, Labor solicited comments in the Federal
Register on its proposal to place a statement on the WD-10 survey
reporting form that respondents could be prosecuted if they willfully
falsify data in the Davis-Bacon wage surveys. The comment period for
this proposal ended in May 1996, and the proposed regulation has now
been sent to the Office of Management and Budget.
Labor has also proposed a long-term strategy to review the entire
Davis-Bacon wage determination process. In late 1995, Labor
established an ongoing task group to identify various strategies for
improving the process it uses to determine prevailing wages. These
continuing discussions have led to the identification of various
weaknesses in the wage determination process and steps Labor might
take to address them. In its fiscal year 1997 budget request, Labor
asked for about $4 million to develop, evaluate, and implement
alternative reliable methodologies or procedures that will yield
accurate and timely wage determinations at reasonable cost.
Approaches that it is considering include alternatives such as use of
other existing databases to extrapolate wage data instead of
collecting its own survey data. Labor anticipates making a general
decision on the overall direction of its strategy for improving its
wage determination process by late 1996.
In our report, we expressed our concern that until Labor completes
development and implementation of long-term initiatives to improve
its processes, it would continue to issue new wage determinations and
enforce compliance with existing ones that may be based on fraudulent
or inaccurate data. Therefore, we recommended that, while it
continues its more long-term evaluation and improvement of the
overall wage determination process, it move ahead immediately to
improve its verification of wage data submitted by employers. We
also recommended that it make the appeals process a more effective
internal control to guard against the use of fraudulent or inaccurate
data. Specifically, we recommended that Labor improve the
accessibility of the appeals process by informing employers, unions,
and other interested parties about the process--about their right to
request information and about procedures for initiating an appeal.
In its response to our draft report, Labor agreed to implement these
recommendations.
CONCLUDING REMARKS
---------------------------------------------------------- Chapter 0:5
Our review confirmed that vulnerabilities exist in Labor's current
wage determination process that could result in wage determinations
based on fraudulent or otherwise inaccurate data. Although we did
not determine the extent to which Labor is using inaccurate data in
its wage calculations nor the consequences, in terms of wages paid,
of such use, we believe that these vulnerabilities are serious and
warrant correction. We believe that the process changes we
recommended address those vulnerabilities and, if implemented in a
timely manner, could increase confidence that the wage rates are
based on accurate data. Specifically, Labor needs to move ahead
immediately to improve its verification of wage data submitted by
employers.
We recognize, however, that the wage determinations could be flawed
for other reasons. For example, other problems with the survey
design and implementation, such as the identification of projects to
survey or the response rates obtained, could affect the validity of
the determinations. In addition, untimely updating of the wage rates
decreases confidence in their appropriateness. Nevertheless, using
only accurate data in the wage determination process is, in our view,
a minimum requirement for ultimately issuing appropriate wage
determinations.
-------------------------------------------------------- Chapter 0:5.1
Mr. Chairmen, that concludes my prepared statement. At this time, I
will be happy to answer any questions you or other members of the
Subcommittees may have.
CONTRIBUTORS
---------------------------------------------------------- Chapter 0:6
For information on this testimony, please call Charles A. Jeszeck,
Assistant Director, at (202) 512-7036; or Linda W. Stokes,
Evaluator-in-Charge, at (202) 512-7040.
*** End of document. ***