Updating the Davis-Bacon and Related Acts Regulations

Federal RegisterAug 23, 2023

Ask Donna

What actually matters in this document.

Text

DEPARTMENT OF LABOR

Office of the Secretary

29 CFR Parts 1, 3, and 5

RIN 1235-AA40

Updating the Davis-Bacon and Related Acts Regulations

AGENCY:

Wage and Hour Division, Department of Labor.

ACTION:

Final rule.

SUMMARY:

In this final rule, the Department of Labor (Department or DOL) updates regulations issued under the Davis-Bacon and Related Acts. As the first comprehensive regulatory review in nearly 40 years, revisions to these regulations will promote compliance, provide appropriate and updated guidance, and enhance their usefulness in the modern economy.

DATES:

Effective date:

This final rule is effective on October 23, 2023.

Applicability date:

The provisions of this final rule regarding wage determination methodology and related part 1 provisions prescribing the content of wage determinations may be applied only to wage determination revisions completed by the Department on or after October 23, 2023. Except with regard to § 1.6(c)(2)(iii), the provisions of this final rule are applicable only to contracts entered into after October 23, 2023. Contracting agencies must apply the terms of § 1.6(c)(2)(iii) to existing contracts of the types addressed in that regulatory provision, without regard to the date a contract was entered into, if practicable and consistent with applicable law. For additional information, see the discussion of Applicability Date in section III.C. below.

FOR FURTHER INFORMATION CONTACT:

Amy DeBisschop, Director, Division of Regulations, Legislation, and Interpretation, Wage and Hour Division, U.S. Department of Labor, Room S-3502, 200 Constitution Avenue NW, Washington, DC 20210; telephone: (202) 693-0406 (this is not a toll-free number). Alternative formats are available upon request by calling 1-866-487-9243. If you are deaf, hard of hearing, or have a speech disability, please dial 7-1-1 to access telecommunications relay services.

Questions of interpretation or enforcement of the agency's existing regulations may be directed to the nearest WHD district office. Locate the nearest office by calling the WHD's toll-free help line at (866) 4US-WAGE ((866) 487-9243) between 8 a.m. and 5 p.m. in your local time zone, or log onto WHD's website at

https://www.dol.gov////offices

for a nationwide listing of WHD district and area offices.

SUPPLEMENTARY INFORMATION:

I. Executive Summary

In order to provide greater clarity and enhance their usefulness in the modern economy, on March 18, 2022, the Department published a notice of proposed rulemaking (NPRM), 87 FR 15698, proposing to update and modernize the regulations at 29 CFR parts 1, 3, and 5, which implement the Davis-Bacon Act and the Davis-Bacon Related Acts (collectively, the DBRA). The Davis-Bacon Act (DBA or Act), enacted in 1931, requires the payment of locally prevailing wages and fringe benefits on Federal contracts for construction.

See

40 U.S.C. 3142. The DBA applies to workers on contracts entered into by Federal agencies and the District of Columbia that are in excess of $2,000 and for the construction, alteration, or repair of public buildings or public works. Congress subsequently incorporated DBA prevailing wage requirements into numerous statutes (referred to as “Related Acts”) under which Federal agencies assist construction projects through grants, loans, loan guarantees, insurance, and other methods.

The Supreme Court has described the DBA as “a minimum wage law designed for the benefit of construction workers.”

United States

Binghamton Constr. Co.,

347 U.S. 171, 178 (1954). The Act's purpose is “to protect local wage standards by preventing contractors from basing their bids on wages lower than those prevailing in the area.”

Univs. Research Ass'n, Inc.

Coutu,

450 U.S. 754, 773 (1981) (quoting H. Comm. on Educ. & Lab., Legislative History of the Davis-Bacon Act, 87th Cong., 2d Sess., 1 (Comm. Print 1962)). By requiring the payment of minimum prevailing wages, Congress sought to “ensure that Government construction and federally assisted construction would not be conducted at the expense of depressing local wage standards.” Determination of Wage Rates Under the Davis-Bacon & Serv. Cont. Acts, 5 Op. O.L.C. 174, 176 (1981) (citation and internal quotation marks omitted).

Available at:

https://www.justice.gov/sites/default/files/olc/opinions/1981/06/31/op-olc-v005-p0174_0.pdf.

Congress has delegated authority to the Department to issue prevailing wage determinations and prescribe rules and regulations for contractors and subcontractors on DBA-covered construction projects.

See

40 U.S.C. secs. 3142, 3145. It has also directed the Department, through Reorganization Plan No. 14 of 1950, to “prescribe appropriate standards, regulations and procedures” to be observed by Federal agencies responsible for the administration of the Davis-Bacon and Related Acts. 15 FR 3173, 3176 effective May 24, 1950, reprinted as amended in 5 U.S.C. app. 1 and in 64 Stat. 1267. These regulations, which have been updated and revised periodically over time, are primarily located in parts 1, 3, and 5 of title 29 of the Code of Federal Regulations.

The DBA and the Related Acts apply to both prime contracts and subcontracts of any tier thereunder. In this final rule, as in the regulations themselves, where the terms “contracts” or “contractors” are used, they are intended to include reference to subcontracts and subcontractors of any tier.

The Department last engaged in a comprehensive revision of the regulations governing the DBA and the Related Acts in a 1981-1982 rulemaking.

Since that time, Congress has expanded the reach of the Davis-Bacon

labor standards

significantly, adding numerous Related Act statutes to which these regulations apply. The Davis-Bacon Act and now more than 70 active Related Acts

collectively apply to an estimated $217 billion in Federal and federally assisted construction spending per year and provide minimum wage rates for an estimated 1.2 million U.S. construction workers.

The Department expects these numbers to continue to grow as Federal and State governments seek to address the significant infrastructure needs of the country, including, in particular, the energy and transportation infrastructure necessary to mitigate climate change.

See

46 FR 41444 (1981 NPRM); 47 FR 23644 (1982 final rule); 48 FR 19532 (1983 revised final rule).

The term “Davis-Bacon” is used in this final rule as a shorthand reference for the Davis-Bacon and Related Acts.

In this final rule, the term “Davis-Bacon labor standards” means, as defined in § 5.2 of the final rule, “the requirements of the Davis-Bacon Act, the Contract Work Hours and Safety Standards Act (other than those relating to safety and health), the Copeland Act, and the prevailing wage provisions of the other statutes referenced in § 5.1, and the regulations in parts 1 and 3 of this subtitle and this part.”

The Department maintains a list of the Related Acts at

https://www.dol.gov/agencies/whd/government-contracts/

These estimates are discussed below in section V (Executive Order 12866, Regulatory Planning and Review et al.).

See

Executive Order 14008, “Tackling the Climate Crisis at Home and Abroad,” section 206 (Jan. 27, 2021), available at:

https://www.whitehouse.gov/briefing-room/presidential-actions/2021/01/27/executive-order-on-tackling-the-climate-crisis-at-home-and-abroad/

In addition to the expansion of the prevailing wage rate requirements of the DBA and the Related Acts, the Federal contracting system itself has undergone significant changes since the 1981-1982 rulemaking. Federal agencies have dramatically increased spending through interagency Federal schedules such as the Multiple Award Schedule (MAS). Contractors have increased their use of single-purpose entities, such as joint ventures and teaming agreements, in construction contracts with Federal, State and local governments. Federal procurement regulations have been overhauled and consolidated in the Federal Acquisition Regulation (FAR; 48 CFR chapter 1), which contains a subpart on the Davis-Bacon Act and related contract clauses.

See

48 CFR 22.400

et seq.

Court and agency administrative decisions have developed and clarified myriad aspects of the laws governing Federal procurement.

During the past 40 years, the Department's DBRA program also has continued to evolve. Where the program initially was focused on individual project-specific wage determinations, contracting agencies now incorporate the Department's general wage determinations for the construction type in the locality in which the construction project is to occur. The program also now uniformly uses wage surveys to develop general wage determinations, eliminating an earlier practice of developing wage determinations based solely on other evidence about the general level of unionization in the targeted area. In a 2006 decision, the Department's Administrative Review Board (ARB) identified several survey-related wage determination procedures as inconsistent with the 1982 final rule.

See Mistick Constr.,

ARB No. 04-051, 2006 WL 861357, at *5-7 (Mar. 31, 2006).

As a consequence of these developments, the use of averages of wage rates from survey responses has increasingly become the methodology used to issue new wage determinations—notwithstanding the Department's long-held interpretation that the DBA allows the use of such averages only as a methodology of last resort.

Decisions of the ARB from 1996 to the present are available on the Department's website at

https://www.dol.gov/agencies/arb/decisions

The Department has also received significant feedback from stakeholders and others since the last comprehensive rulemaking. In a 2011 report, the Government Accountability Office (GAO) reviewed the Department's wage survey and wage determination process and found that the Department was often behind schedule in completing wage surveys, leading to a backlog of wage determinations and the use of out-of-date wage determinations in some areas.

The report also identified dissatisfaction among regulated parties regarding the rigidity of the Department's county-based system for identifying prevailing rates,

and missing wage rates requiring an overuse of “conformances” for wage rates for specific job classifications.

A 2019 report from the Department's Office of the Inspector General (OIG) made similar findings regarding out-of-date wage determinations.

See

Gov't Accountability Office, GAO-11-152, “Davis-Bacon Act: Methodological Changes Needed to Improve Wage Survey” (2011) (2011 GAO Report), at 12-19, available at:

https://www.gao.gov/assets/gao-11-152.pdf

Id.

at 23-24.

Id.

at 32-33.

See

Department of Labor, Office of the Inspector General, “Better Strategies Are Needed to Improve the Timeliness and Accuracy of Davis-Bacon Act Prevailing Wage Rates” (2019) (2019 OIG Report), at 10, available at:

https://www.oversight.gov/sites/default/files/oig-reports/04-19-001--Davis%20Bacon.pdf

Ensuring that construction workers are paid the wages required under the DBRA also requires effective enforcement in addition to an efficient wage determination process. In the last decade, enforcement efforts at the Department have resulted in the recovery of more than $229 million in back wages for over 76,000 workers.

But the Department has also encountered significant enforcement challenges. Among the most critical of these is the omission of DBRA contract clauses from contracts that are clearly covered by the DBRA. In one recent case, a contracting agency agreed with the Department that a blanket purchase agreement (BPA) it had entered into with a contractor had mistakenly omitted the Davis-Bacon clauses and wage determination, but the omission still resulted in an 8-year delay before the workers were paid the wages they were owed.

The listed figures have been corrected from the NPRM. The updated figures reflect the sum of the annual enforcement statistics from 2010-2019 in Gov't Accountability Office, GAO-21-13, “Fair Labor Standards Act: Tracking Additional Complaint Data Could Improve DOL's Enforcement” (2020) (2020 GAO Report), at 39, available at:

https://www.gao.gov/assets/gao-21-13.pdf.

Through this rulemaking, the Department seeks to address a number of these outstanding challenges in the program while also providing greater clarity in the DBRA regulations and enhancing their usefulness in the modern economy. In the NPRM, the Department proposed to update and modernize the regulations implementing the DBRA at 29 CFR parts 1, 3, and 5. In some of these proposed revisions, the Department had determined that changes it made in the 1981-1982 rulemaking were mistaken or ultimately resulted in outcomes that are increasingly in tension with the DBA statute itself. In others, the Department sought to expand further on procedures that were introduced in that last major revision, or to propose new procedures that will increase efficiency of administration of the DBRA and enhance protections for covered construction workers. The Department invited comments on these proposed updates and received 40,938 timely comments after a 60-day comment period.

The comments were from a broad array of constituencies, including contractors, unions, employer and industry associations, worker advocacy groups, non-profit organizations, social scientists, law firms, think tanks, Members of Congress, a state attorney general, a state department of labor, and other interested members of the public. All timely received comments may be viewed on the

regulations.gov

website, docket ID WHD-2022-0001. Some of the comments the Department received were general statements of support or opposition, and the Department also received approximately 40,200 “campaign” comments sent in response to organized initiatives. Commenters expressed a wide variety of views on the merits of particular aspects of the Department's proposal; however, most commenters favored some, if not all, of the changes proposed in the NPRM. The Department has considered the timely submitted comments addressing the proposed changes.

The Department also received a number of comments that are beyond the scope of this rulemaking. These included requests that would require Congress to amend statutory language in the DBRA. For example, many commenters suggested a change to the $2,000 threshold for DBA and certain Related Acts to apply. Others suggested eliminating or changing the weekly certified payroll requirement that is expressly required by 40 U.S.C. sec 3145.

Other comments beyond the scope of the rulemaking included those that suggested significant new regulatory provisions or changes that were not proposed in the NPRM. Among these, for example, the Iron Workers International Union suggested the codification of the requirement to thoroughly investigate “area practice” issues that arise during the wage survey

process.

See Fry Bros. Corp.,

WAB No. 76-06, 1977 WL 24823, at *6 (June 14, 1977),

aff'd sub nom. Fry Bros. Corp.

Dep't of Hous. & Urb. Dev.,

614 F.2d 732, 732-33 (10th Cir. 1980). The Iron Workers also suggested creation of a new administrative process for issuing “right to sue” notices to workers to pursue rights of action authorized by 40 U.S.C. sec 3144(a)(2). As noted in the comment, such an initiative would be better proposed in a separate and subsequent notice-and-comment rulemaking.

The Department reviewed the comments submitted in particular for assertions by interested parties of their reliance on the existing regulations in a way that would be adversely affected by the proposed rule. Although many comments stated that the current regulations had been in place for many years, few specified that parties had relied on the regulations so as to raise questions about the fairness or reasonableness of amending them in the current rulemaking. Nonetheless, the Department considered whether the rule as a whole, as well as its individual proposed provisions, could plausibly implicate significant and legitimate reliance interests, and the Department has concluded that the proposed amendments to the regulations do not raise reliance interests that would outweigh the agency objectives discussed throughout this preamble.

The Department did not identify significant reliance interests among contractors or others in the existing part 1 regulations. The part 1 regulations involve the Department's methodology for determining the prevailing wage rates that are required on covered contracts. Some of the changes the Department proposed to this part may lead to higher required wage rates in places and lower wage rates in others, and the new periodic adjustments of certain non-collectively bargained wage rates will result in a smoother increase in such wage rates over time instead of longer periods of the same wage rates for an area followed by steeper increases after the publication of new survey rates. Similarly, the new language clarifying the procedure for incorporating prevailing wage rates into multiple award schedules and other similar contracts may result in more frequent updates to prevailing wage rates on such contracts when options are executed. These types of changes, however, should not be significantly different in their effect on contractors than the fluctuations in prevailing wage rates that already occur between wage surveys as a result of changes in local economies and shifts in regional labor markets. Even if the part 1 changes were to have significant effects on prevailing wage rates in certain local areas, any reliance interests of local contractors, governmental agencies, or workers on prior prevailing wage rates would be limited, given that the changes to the wage determination processes generally will not affect current contracts—which will continue to be governed by the wage determinations incorporated at the time of their award, with limited exceptions. Most of the revisions to part 1 will only apply to wage surveys that are finalized after the rule becomes effective, and thus they will generally apply only to contracts awarded after such new wage determinations are issued.

Contractors will therefore be able to factor any new wage rates into their bids on future contracts.

As explained in § 1.6(c), whenever a new wage determination is issued (either after the completion of a new wage survey or through the new periodic adjustment mechanism), that revision as a general matter does not and will not apply to contracts which have already been awarded, with three exceptions. These exceptions are explained in § 1.6(c)(2)(iii), and they include where a contract or order is changed to include substantial covered work that was not within the original scope of work, where an option is exercised, and also certain ongoing contracts that are not for specific construction, for which new wage determinations must be incorporated on an annual basis under § 1.6(c)(2)(iii)(B) of the final rule. The final rule instructs contracting agencies to apply the terms of § 1.6(c)(2)(iii) to all existing contracts, without regard to the date of contract award, if practicable and consistent with applicable law. The Department does not anticipate that the application of the amended wage determination methodologies in these situations will result in unfair harm to reliance interests in a manner sufficient to outweigh the benefits of the final rule implementation as planned.

See also

section III.C. (“Applicability Date”) below.

Many of the amendments to part 5 of the regulations are regulatory changes that codify the Department's current practices and interpretations of existing regulations. As a result, such changes do not, in practical terms, impose new obligations on contractors or contracting agencies. Other changes, such as the new anti-retaliation provision, provide new remedies to address conduct that already may subject contractors to potential debarment. Any reliance interest in the ability to carry out such conduct with lesser potential consequences is particularly weak. Regardless, these new amendments to part 5 will generally only apply to contracts that are awarded after the effective date of this final rule. Contractors entering into new contracts issued after the rule is published and becomes applicable will have notice of the regulatory changes and will be able to take the changes into consideration as they analyze internal controls and develop their bids or negotiate contract pricing.

ABC argued that the Department's denial of requests to extend the public comment period beyond the 60 days provided was arbitrary and capricious, and other commenters expressed disappointment that the comment period had not been extended. As explained in the Department's public response to the extension requests in regulations.gov, the Department concluded that the 60-day period provided the public with a meaningful opportunity to comment on the proposed rule. The Davis-Bacon and Related Acts' applicability is limited to Federal and federally assisted construction projects, and therefore applies to a defined group of stakeholders. Additionally, various elements of the proposed and final rules codify or clarify longstanding policies, practices, and interpretations. As a result, stakeholders were familiar with many of the issues addressed in the NPRM. The public had additional time to review the NPRM, which was available on the Department's website on March 11, 2022, seven days in advance of its publication in the

Federal Register

. The comprehensive nature and substance of the comments received—both in favor of and opposing the proposed rule—support the Department's view that the 60-day period was appropriate and sufficient. Finally, the Department and the Office of Management and Budget have participated in several meetings pursuant to E.O. 12866 at which stakeholders have had opportunities to elaborate on their public comments.

Finally, some commenters raised concerns about the administrative or paperwork burdens contractors might face while adjusting to, and under, the Department's final rule. The Department considered such concerns in its economic analyses and concluded that the paperwork burdens associated with the rule are limited and are outweighed by the benefits of the regulation.

Having considered all of the comments, the Department has decided to adopt the NPRM's proposed changes with some modifications. Significant issues raised in the comments are discussed in more detail below in section III (“Final Regulatory Revisions”), along with the Department's responses to those comments.

This final rule includes several elements targeted at increasing the amount of information available for wage determinations and speeding up the determination process. In particular, the final rule amends § 1.3 of the

regulations by outlining a new methodology to expressly give the Wage and Hour Division (WHD) Administrator authority and discretion to adopt State or local wage determinations as the Davis-Bacon prevailing wage where certain specified criteria are satisfied. Such a change will help improve the currentness and accuracy of wage determinations, as many States and localities conduct surveys more frequently than the Department and have relationships with stakeholders that may facilitate the process and foster more widespread participation. This revision will also increase efficiency and reduce confusion for the regulated community where projects are covered by both DBRA and local or State prevailing wage laws and contractors are already familiar with complying with the local or State prevailing wage requirement.

The Department also amends the definition of “prevailing wage” in § 1.2, and in § 1.7, the scope of data considered to identify the prevailing wage in a given area. To address the overuse of weighted average rates, the Department returns to the definition of “prevailing wage” in § 1.2 that it used from 1935 to 1983.

Currently, a wage rate may be identified as prevailing in the area only if it is paid to a majority of workers in a classification on the wage survey; otherwise, a weighted average is used. The Department returns instead to the “three-step” method that was in effect before 1983. Under that method (also known as the 30-percent rule), in the absence of a wage rate paid to a majority of workers in a particular classification, a wage rate will be considered prevailing if it is paid to at least 30 percent of such workers. The Department also returns to a prior policy on another change made during the 1981-1982 rulemaking related to the delineation of wage survey data submitted for “metropolitan” or “rural” counties in § 1.7(b). Through this change, the Department will more accurately reflect modern labor force realities, allow more wage rates to be determined at smaller levels of geographical aggregation, and will increase the sufficiency of data at the statewide level.

The 1981-1982 rulemaking went into effect on Apr. 29, 1983. 48 FR 19532.

Revisions to §§ 1.3 and 5.5 are aimed at reducing the need for the use of “conformances” where the Department has received insufficient data to publish a prevailing wage for a classification of worker—a process that currently is burdensome on contracting agencies, contractors, and the Department. This final rule codifies a new procedure through which the Department may identify (and list on the wage determination) wage and fringe benefit rates for certain classifications for which WHD received insufficient data through its wage survey program. The procedure will reduce the need for conformances of classifications for which conformances are now often required.

The Department also revises § 1.6(c)(1) to provide a mechanism to regularly update certain non-collectively bargained prevailing wage rates based on the Employment Cost Index (ECI) published by the Bureau of Labor Statistics (BLS).

The mechanism is intended to keep such rates more current between surveys so that they do not become out-of-date and fall behind prevailing rates in the area.

Available at:

https://www.bls.gov/news.release/eci.toc.htm

The Department also strengthens enforcement in several critical ways. The Department addresses the challenges caused by the omission of contract clauses. In a manner similar to its rule under Executive Order 11246 (Equal Employment Opportunity), the Department designates the DBRA contract clauses in § 5.5(a) and (b), and applicable wage determinations, as effective by “operation of law” notwithstanding their mistaken omission from a contract. This is an extension of the retroactive modification procedures that were put into effect in § 1.6 by the 1981-1982 rulemaking, and it will expedite enforcement efforts to ensure the timely payment of prevailing wages to all workers who are owed such wages under the relevant statutes.

In addition, the Department finalizes new anti-retaliation provisions in the Davis-Bacon contract clauses in new paragraphs at § 5.5(a)(11) (DBRA) and (b)(5) (Contract Work Hours and Safety Standards Act (CWHSSA)), and in a new section of part 5 at § 5.18. The language ensures that workers who raise concerns about payment practices or assist agencies or the Department in investigations are protected from termination or other adverse employment actions.

Finally, to reinforce the remedies available when violations are discovered, the Department clarifies and strengthens the cross-withholding procedure for recovering back wages by including new language in the withholding contract clauses at § 5.5(a)(2) (DBRA) and (b)(3) (CWHSSA) to clarify that cross-withholding may be accomplished on contracts held by agencies other than the agency that awarded the contract. The Department also creates a mechanism through which contractors will be required to consent to cross-withholding for back wages owed on contracts held by different but related legal entities in appropriate circumstances—if, for example, those entities are controlled by the same controlling shareholder or are joint venturers or partners on a Federal contract. The revisions also include a harmonization of the DBA and Related Act debarment standards.

II. Background

A. Statutory and Regulatory History

The Davis-Bacon Act, as enacted in 1931 and subsequently amended, requires the payment of minimum prevailing wages determined by the Department to laborers and mechanics working on Federal contracts in excess of $2,000 for the construction, alteration, or repair, including painting and decorating, of public buildings and public works.

See

40 U.S.C. 3141

et seq.

Congress has also included the Davis-Bacon requirements in numerous other laws, known as the Davis-Bacon Related Acts (the Related Acts and, collectively with the Davis-Bacon Act, the DBRA), which provide Federal assistance for construction projects through grants, loans, loan guarantees, insurance, and other methods. Congress intended the Davis-Bacon Act to “protect local wage standards by preventing contractors from basing their bids on wages lower than those prevailing in the area.”

Coutu,

450 U.S. at 773 (quoting H. Comm. on Educ. and Lab., Legis. History of the Davis-Bacon Act, 87th Cong., 2d Sess., 1 (Comm. Print 1962)).

The Copeland Act, enacted in 1934, added the requirement that contractors working on Davis-Bacon projects must submit weekly certified payrolls for work performed on the contract.

See

40 U.S.C. 3145. The Copeland Act also prohibits contractors from inducing any worker to give up any portion of the wages due to them on such projects.

See

18 U.S.C. 874. In 1962, Congress passed CWHSSA, which, as amended, requires an overtime payment of additional half-time for hours worked over 40 in the workweek by laborers and mechanics, including watchpersons and guards, on Federal contracts or federally assisted contracts containing Federal prevailing wage standards.

See

40 U.S.C. 3701

et seq.

As initially enacted, the DBA did not take into consideration the provision of fringe benefits to workers. In 1964, Congress expanded the Act to require the Department to include an analysis of fringe benefits as part of the wage determination process. The amendment

requires contractors and subcontractors to provide fringe benefits (such as vacation pay, sick leave, health insurance, and retirement benefits), or the cash equivalent thereof, to their workers at the level prevailing for the labor classification on projects of a similar character in the locality.

See

Act of July 2, 1964, Public Law 88-349, 78 Stat. 238.

Congress has delegated broad rulemaking authority under the DBRA to the Department. The DBA, as amended, contemplates regulatory and administrative action by the Department to determine the prevailing wages that must be paid and to “prescribe reasonable regulations” for contractors and subcontractors. 40 U.S.C. 3142(b); 40 U.S.C. 3145. Congress also, through Reorganization Plan No. 14 of 1950, directed the Department to “prescribe appropriate standards, regulations and procedures” to be observed by Federal agencies responsible for the administration of the Davis-Bacon and Related Acts. 15 FR 3176; 5 U.S.C. app. 1.

The Department promulgated its initial regulations implementing the Act in 1935 and has since periodically revised them.

See

U.S. Department of Labor, Regulations No. 503 (Sept. 30, 1935). In 1938, these initial regulations, which set forth the procedures for the Department to follow in determining prevailing wages, were included in part 1 of Title 29 of the new Code of Federal Regulations.

See

29 CFR 1.1

et seq.

(1938). The Department later added regulations to implement the payroll submission and anti-kickback provisions of the Copeland Act—first in part 2 and then relocated to part 3 of Title 29.

See

6 FR 1210 (Mar. 1, 1941); 7 FR 687 (Feb. 4, 1942); 29 CFR part 2 (1942); 29 CFR part 3 (1943). After the Reorganization Plan No. 14 of 1950, the Department issued regulations setting forth procedures for the administration and enforcement of the Davis-Bacon and Related Acts in a new part 5. 16 FR 4430 (May 12, 1951); 29 CFR part 5. The Department made significant revisions to the regulations in 1964, and again in the 1981-1982 rulemaking.

See

29 FR 13462 (Sept. 30, 1964); 46 FR 41444-70 (NPRM parts 1 and 5) (Aug. 14, 1981); 47 FR 23644-79 (final rule parts 1, 3, and 5) (May 28, 1982). The Department also proposed a significant revision of parts 1 and 5 of the regulations in 1979 and issued a final rule in 1981.

See

44 FR 77026 (Dec. 28, 1979) (NPRM Part 1); 44 FR 77080 (Dec. 28, 1979) (NPRM part 5); 46 FR 4306 (Jan. 16, 1981) (final rule part 1); 46 FR 4380 (Jan. 16, 1981) (final rule part 5). The 1981 final rules, however, were delayed and subsequently replaced by the 1981-1982 rulemaking. The 1982 final rule was delayed by litigation and re-published with amendments in 1983 and 1985. 48 FR 19532-53 (Apr. 29, 1983) (final rule parts 1 and 5); 50 FR 4506 (Jan. 31, 1985) (final rule §§ 1.3(d) and 1.7(b)).

While the Department has made periodic revisions to the regulations in recent years, such as to better protect the personal privacy of workers, 73 FR 77511 (Dec. 19, 2008); to remove references to the “Employment Standards Administration,” 82 FR 2225 (Jan. 9, 2017); and to adjust Federal civil money penalties, 81 FR 43450 (July 1, 2016), 83 FR 12 (Jan. 2, 2018), 84 FR 218 (Jan. 23, 2019), 87 FR 2328 (Jan. 14, 2022), 88 FR 2210 (Jan. 13, 2023), the Department has not engaged in a comprehensive review and revision since the 1981-1982 rulemaking.

B. Overview of the Davis-Bacon Program

WHD, an agency within the U.S. Department of Labor, administers the Davis-Bacon program for the Department. WHD carries out its responsibilities in partnership with the Federal agencies that enter into direct DBA-covered contracts for construction and/or administer Federal assistance to State and local governments and other funding recipients that is covered by the Related Acts. The State and local governmental agencies and authorities that receive covered financial assistance also have important responsibilities in administering Related Act program rules, as they manage programs through which covered funding flows or the agencies themselves directly enter into covered contracts for construction.

The DBRA program includes three basic components in which these government entities have responsibilities: (1) wage surveys and wage determinations; (2) contract formation and administration; and (3) enforcement and remedies.

1. Wage Surveys and Determinations

The DBA delegates to the Secretary of Labor the responsibility to determine the wage rates that are “prevailing” for each classification of covered laborers and mechanics on similar projects “in the civil subdivision of the State in which the work is to be performed.” 40 U.S.C. 3142(b). WHD carries out this responsibility for the Department through its wage survey program and derives the prevailing wage rates from survey information that responding contractors and other interested parties voluntarily provide. The program is carried out in accordance with the program regulations in part 1 of Title 29 of the Code of Federal Regulations,

see

29 CFR 1.1 through 1.7, and its procedures are described in guidance documents such as the “Davis-Bacon Construction Wage Determinations Manual of Operations” (1986) (Manual of Operations) and “Prevailing Wage Resource Book” (2015) (PWRB).

Although part 1 of the regulations provides the authority for WHD to create project-specific wage determinations, such project wage determinations, once more common, now are rarely employed. Instead, nearly all wage determinations are general wage determinations issued for general types of construction (building, residential, highway, and heavy) and applicable to a specific geographic area. General wage determinations can be incorporated into the vast majority of contracts and create uniform application of the DBRA for that area.

The Manual of Operations is a 1986 guidance document that is still used internally for reference within WHD. The PWRB is a 2015 document that is intended to provide practical information to contracting agencies and other interested parties, and is available at

https://www.dol.gov/agencies/whd/government-contracts/prevailing-wage-resource-book

2. Contract Formation and Administration

The Federal agencies that enter into DBA-covered contracts or administer Related Act programs have the initial responsibility to determine whether a contract is covered by the DBA or one of the Related Acts and identify the contract clauses and the applicable wage determinations that must be included in the contract.

See

29 CFR 1.6(b). In addition to the Department's regulations, this process is also guided by parallel regulations in part 22 of the FAR for those contracts that are subject to the FAR.

See

48 CFR part 22. Federal agencies also maintain their own regulations and guidance governing agency-specific aspects of the process.

See, e.g.,

48 CFR subpart 222.4 (Defense); 48 CFR subpart 622.4 (State); U.S. Department of Housing and Urban Development (HUD), HUD Handbook 1344.1, Federal Labor Standards Requirements in Housing and Urban Development Programs (2013).

Available at:

https://www.hud.gov/sites/dfiles/OCHCO/documents/Work-Schedule-Request.pdf.

Where contracting agencies or interested parties have questions about such matters as coverage under the DBRA or the applicability of the appropriate wage determination to a specific contract, they are directed to submit those questions to the Administrator of WHD (the Administrator) for resolution.

See

29 CFR 5.13. The Administrator responds to such questions and provides periodic guidance on other aspects of the DBRA program to contracting agencies and other interested parties, particularly through All Agency Memoranda

(AAMs) and ruling letters. In addition, the Department maintains a guidance document, the Field Operations Handbook (FOH), to provide guidance for the regulated community and for WHD investigators and staff on contract administration and enforcement policies.

The FOH reflects policies established through changes in legislation, regulations, significant court decisions, and the decisions and opinions of the WHD Administrator. It is not used as a device for establishing interpretive policy. Chapter 15 of the FOH covers the DBRA, including CWHSSA, and is available at

https://www.dol.gov/agencies/whd/field-operations-handbook/Chapter-15

During the administration of a DBRA-covered contract, contractors and subcontractors are required to provide certified payrolls to the contracting agency to demonstrate their compliance with the incorporated wage determinations on a weekly basis.

See generally

29 CFR part 3. Contracting agencies have the duty to ensure compliance by engaging in periodic audits or investigations of contracts, including examinations of payroll data and confidential interviews with workers.

See

29 CFR 5.6. Prime contractors have the responsibility for the compliance of all the subcontractors on a covered prime contract. 29 CFR 5.5(a)(6). WHD conducts investigations of covered contracts, which include determining if the DBRA contract clauses or appropriate wage determinations were mistakenly omitted from the contract.

See

29 CFR 1.6(f). If WHD determines that there was such an omission, it will request that the contracting agency either terminate and resolicit the contract or modify it to incorporate the required clauses or wage determinations retroactively.

Id.

3. Enforcement and Remedies

In addition to WHD, contracting agencies have enforcement authority under the DBRA. When a contracting agency's investigation reveals underpayments of wages of the DBA or one of the Related Acts, the Federal agency generally is required to provide a report of its investigation to WHD, and to seek to recover the underpayments from the contractor responsible.

See

29 CFR 5.6(a), 5.7. If violations identified by the contracting agency or by WHD through its own investigation are not promptly remedied, contracting agencies are required to suspend payment on the contract until sufficient funds are withheld to compensate the workers for the underpayments. 29 CFR 5.9. The DBRA contract clauses also provide for “cross-withholding” if sufficient funds are no longer available on the contract under which the violations took place. Under this procedure, funds may be withheld from any other covered Federal contract or federally assisted contract held by the same prime contractor in order to remedy the underpayments on the contract at issue.

See

29 CFR 5.5(a)(2), (b)(3). Contractors that violate the DBRA may also be subject to debarment from future Federal contracts and federally assisted contracts.

See

29 CFR 5.12.

Where WHD conducts an investigation and finds that violations have occurred, it will notify the affected prime contractor(s) and subcontractor(s) of the findings of the investigation—including any determination that workers are owed back wages and whether there is reasonable cause to believe the contractor may be subject to debarment.

See

29 CFR 5.11(b). Contractors can request a hearing regarding these findings through the Department's Office of Administrative Law Judges (OALJ) and may appeal any ruling by the OALJ to the Department's ARB.

Id.; see also

29 CFR parts 6 and 7 (OALJ and ARB rules of practice for Davis-Bacon proceedings). Decisions of the ARB are final agency actions that may be reviewable under the Administrative Procedure Act (APA) in Federal district court.

See

5 U.S.C. 702, 704.

In addition to reviewing liability determinations and debarment, the ARB, the Secretary (when exercising discretionary review), and the courts also have jurisdiction in certain circumstances to review general wage determinations. Judicial review, however, is strictly limited to any procedural irregularities, as there is no jurisdiction to review the substantive correctness of a wage determination under the DBA.

See Binghamton Constr. Co.,

347 U.S. at 177.

III. Final Regulatory Revisions

A. Legal Authority

The Davis-Bacon Act, as enacted in 1931 and subsequently amended, requires the payment of certain minimum “prevailing” wages determined by the Department to laborers and mechanics working on Federal contracts in excess of $2,000 for the construction, alteration, or repair, including painting and decorating, of public buildings and public works.

See

40 U.S.C. 3141

et seq.

The DBA authorizes the Secretary of Labor to develop a definition for the term “prevailing” wage and a methodology for setting it based on wages paid on similar projects in the civil subdivision of the State in which a covered project will occur.

See

40 U.S.C. 3142(b);

Bldg. & Constr. Trades' Dep't, AFL-CIO

Donovan,

712 F.2d 611, 616 (D.C. Cir. 1983).

The Secretary of Labor has the responsibility to “prescribe reasonable regulations” for contractors and subcontractors on covered projects. 40 U.S.C. 3145. The Secretary, through Reorganization Plan No. 14 of 1950, also has the responsibility to “prescribe appropriate standards, regulations and procedures” to be observed by Federal agencies responsible for the administration of the Davis-Bacon and Related Acts “[i]n order to assure coordination of administration and consistency of enforcement of the labor standards provisions” of the DBRA. 15 FR 3176; 5 U.S.C. app. 1.

The Secretary has delegated authority to promulgate these regulations to the Administrator and to the Deputy Administrator of the WHD if the Administrator position is vacant.

See

Secretary's Order No. 01-2014, 79 FR 77527 (Dec. 24, 2014); Secretary's Order No. 01-2017, 82 FR 6653 (Jan. 19, 2017).

B. Overview of the Final Rule

The Department finalizes its proposals to update and modernize the regulations at 29 CFR parts 1, 3, and 5, which implement the DBRA. The sections below address these regulatory revisions as adopted in the final rule.

1. 29 CFR Part 1

The procedures for determining the prevailing wage rates and fringe benefits applicable to laborers and mechanics engaged in construction activity covered by the Davis-Bacon and Related Acts are set forth in 29 CFR part 1. The regulations in this part also set forth the procedures for the application of such prevailing wage determinations to covered construction projects.

i. Section 1.1 Purpose and Scope

The Department proposed technical revisions to § 1.1 to update the statutory reference to the Davis-Bacon Act, now recodified at 40 U.S.C. 3141

et seq.

The Department also proposed to eliminate outdated references to the Deputy Under Secretary of Labor for Employment Standards at the Employment Standards Administration. The Employment Standards Administration was eliminated as part of an agency reorganization in 2009, and its authorities and responsibilities were devolved into its constituent components, including the WHD.

See

Secretary's Order No. 09-2009 (Nov. 6, 2009), 74 FR 58836 (Nov. 13, 2009), 82 FR 2221 (Jan. 9, 2017). The Department further proposed to revise § 1.1 to reflect the removal of Appendix A of part 1, as discussed below. The Department also proposed to add new paragraph (a)(1) to reference the WHD website (

https://

www.dol.gov/agencies/whd/government-contracts,

or its successor website) on which a listing of laws requiring the payment of wages at rates predetermined by the Secretary of Labor under the Davis-Bacon Act is currently found.

The Department received one comment in favor of this proposal. The United Association of Journeymen and Apprentices of the Plumbing and Pipe Fitting Industry of the United States & Canada (UA) commented in support of the proposal, noting that the current information was outdated. The final rule therefore adopts this change as proposed, with one technical edit to delete an unnecessary conjunction that is not intended to reflect a change in the substance of this section.

ii. Section 1.2 Definitions

(A) Prevailing Wage

Section 1.2 contains the definition of the term “prevailing wage.” The DBA and the Related Acts require laborers and mechanics on covered projects to be paid a prevailing wage as set by the Secretary of Labor, but the statutes do not define the term “prevailing.” The Department's regulatory definition of the term “prevailing wage” in 29 CFR 1.2 specifies the basic methodology with which the Department determines whether a certain wage rate is prevailing in a given geographic area. The Department uses this methodology to prepare wage determinations that are incorporated into DBRA-covered contracts to set minimum wage rates for each classification of covered workers on a project.

In the NPRM, the Department proposed to redefine the term “prevailing wage” in § 1.2 to return to the original methodology for determining whether a wage rate is prevailing. This original methodology has been referred to as the “three-step process.”

Since 1935, the Secretary has interpreted the word “prevailing” in the Davis-Bacon Act to be consistent with the common understanding of the term as meaning “predominant” or “most frequent.” From 1935 until the 1981-1982 rulemaking, the Department employed a three-step process to identify the most frequently used wage rate for each classification of workers in a locality.

See

Regulation 503 section 2 (1935); 47 FR 23644.

This process identified as prevailing: (1) any wage rate paid to a majority of workers; and, if there was none, then (2) the wage rate paid to the greatest number of workers, provided it was paid to at least 30 percent of workers, and, if there was none, then (3) the weighted average rate. The second step has been referred to as the “30-percent rule.”

Implemented Apr. 29, 1983.

See

48 FR 19532.

The three-step process relegated the average rate to a final, fallback method of determining the prevailing wage. In 1962 congressional testimony, Solicitor of Labor Charles Donahue explained the reasoning for this sequence in the determination: An average rate “does not reflect a true rate which is actually being paid by any group of contractors in the community being surveyed.” Instead, “it represents an artificial rate which we create ourselves, and which does not reflect that which a predominant amount of workers are paid.”

Administration of the Davis Bacon Act: Hearings before the Spec. Subcomm. of Lab. of the H. Comm. on Educ. & Lab.,

87th Cong. 811-12 (1962) (testimony of Charles Donahue, Solicitor of Labor).

In 1982, the Department published a final rule that amended the definition of “prevailing wage” by eliminating the second step in the three-step process—the 30-percent threshold.

See

47 FR 23644. The new process required only two steps: first identifying if there was a wage rate paid to more than 50 percent of workers, and then, if not, relying on a weighted average of all the wage rates paid.

Id.

at 23644-45.

In eliminating the 30-percent threshold, however, the Department did not change its underlying interpretation of the word “prevailing”—that it means “the most widely paid rate” must be the “definition of first choice” for the prevailing wage. 47 FR 23645. While the 1982 rule continued to allow the Department to use an average rate as a fallback, the Department rejected commenters' suggestions that the weighted average could be used in all cases.

See

47 FR 23644-45. As the Department explained, this was because the term “prevailing” contemplates that wage determinations mirror, to the extent possible, those rates “actually paid” to workers. 47 FR 23645.

This interpretation—that the definition of first choice for the term “prevailing wage” should be an actual wage rate that is most widely paid—has now been shared across administrations for over 85 years. In the intervening decades, Congress has amended and expanded the reach of the Act's prevailing wage requirements dozens of times without altering the term “prevailing” or the grant of broad authority to the Secretary of Labor to define it.

In addition, the question was also reviewed by the Office of Legal Counsel (OLC) at the Department of Justice, which independently reached the same conclusions: “prevailing wage” means the current and predominant actual rate paid, and an average rate should only be used as a last resort.

See

Determination of Wage Rates Under the Davis-Bacon & Serv. Cont. Acts, 5 Op. O.L.C. 174, 176-77 (1981).

See, e.g.,

Act of Mar. 23, 1941, ch. 26, 55 Stat. 53 (1941) (applying the Act to alternative contract types); CWHSSA of 1962, Public Law 87-581, 76 Stat. 357 (1962) (requiring payment of overtime on contracts covered by the Act); Act of July 2, 1964, Public Law 88-349, 78 Stat. 238 (1964) (extending the Act to cover fringe benefits); 29 CFR 5.1 (referencing 57 Related Acts into which Congress incorporated Davis-Bacon Act requirements between 1935 and 1978).

Available at:

https://www.justice.gov/sites/default/files/olc/opinions/1981/06/31/op-olc-v005-p0174_0.pdf

In the 1982 final rule, when the Department eliminated the 30-percent threshold, it anticipated that this change would increase the use of artificial average rates. 47 FR 23648-49. Nonetheless, the Department believed a change was preferable because the 30-percent threshold could in some cases not account for up to 70 percent of the remaining workers.

See

46 FR 41444. The Department also stated that it agreed with the concerns expressed by certain commenters that establishing a prevailing wage rate based on 30-percent of survey wage rates was “inflationary” and gave “undue weight to collectively bargained rates.” 47 FR 23644-45.

After reviewing the development of the Davis-Bacon Act program since the 1981-1982 rulemaking, the Department has concluded that eliminating the 30-percent threshold has ultimately resulted in an overuse of average rates. On paper, the weighted average remains the fallback method to be used only when there is no majority rate. In practice, though, it has become a central mechanism to set the prevailing wage rates included in Davis-Bacon wage determinations and covered contracts.

Prior to the 1982 rule change, the use of averages to set a prevailing wage rate was relatively rare. In a Ford Administration study of Davis-Bacon Act prevailing wage rates in commercial-type construction in 19 cities, none of the rates were based on averages because all of the wage rates were “negotiated” rates,

i.e.,

based on collective bargaining agreements (CBAs) that represented a predominant wage rate in the locality.

The Department

estimates that prior to the 1982 final rule, as low as 15 percent of classification rates across all wage determinations were based on averages. After the 1982 rule was implemented, the use of averages may have initially increased to approximately 26 percent of all wage determinations.

See

Robert S. Goldfarb & John F. Morrall, “An Analysis of Certain Aspects of the Administration of the Davis-Bacon Act,” Council on Wage and Price Stability (May 1976), reprinted in Bureau of Nat'l Affs.,

Construction Labor Report,

No. 1079, D-1, D-2 (1976).

See Oversight Hearing on the Davis-Bacon Act, Before the Subcomm. on Lab. Standards of the H. Comm. on Educ. & Lab.,

96th Cong. 58 (1979) (statement of Ray Marshall, Secretary of Labor) (discussing study of 1978 determinations showing only 24 percent of classification rates were based on the 30-percent rule); Jerome Staller, “Communications to the Editor,”

Policy Analysis,

Vol. 5, No. 3 (Summer 1979), pp. 397-98 (noting that 60 percent of determinations in the internal Department 1976 and 1978 studies were based on the 30-percent rule or the average-rate rule). The authors of the Council on Wage and Price Stability study, however, pointed out that the Department's figures were for rates that had been based on survey data, while 57 percent of rates in the mid-1970's were based solely on CBAs without the use of surveys (a practice that the Department no longer uses to determine new rates).

See

Robert S. Goldfarb & John F. Morrall II., “The Davis-Bacon Act: An Appraisal of Recent Studies,” 34 Indus. & Lab. Rel. Rev. 191, 199−200 & n.35 (1981). Thus, the actual percentage of annual classification determinations that were based on average rule before 1982 may have been as low as 15 percent, and the percent based on the average rule after 1982 would have been expected to be around 26 percent.

The Department's current use of weighted averages is now significantly higher than this 26 percent figure. To analyze the current use of weighted averages and the potential impacts of this rulemaking, the Department compiled data for select classifications for 19 recent wage surveys—nearly all of the completed surveys that WHD began in 2015 or later. The data show that the Department's reliance on average rates has increased significantly, and now accounts for 63 percent of the observed classification determinations in this recent time period.

See

below section V (Executive Order 12866, Regulatory Planning and Review et al.).

Such an overuse of weighted averages is inconsistent with the Department's longstanding interpretation of Congress's use of the word “prevailing” in the text of the Act—including the Department's statements in the preamble to the 1982 rule itself that the definition of first choice for the “prevailing” wage should be the most widely paid rate that is actually paid to workers in the relevant locality. If nearly two-thirds of rates that are now being published based on recent surveys are based on a weighted average, it is no longer fair to say that it is a fallback method of determining the prevailing wage.

The use of averages as the dominant methodology for issuing wage determinations is also in tension with the recognized purpose of the Act “to protect local wage standards by preventing contractors from basing their bids on wages lower than those prevailing in the area.”

Coutu,

450 U.S. at 773 (internal quotation marks and citation omitted). Using an average to determine the minimum wage rate on contracts allows a single low-wage contractor in the area to depress wage rates on Federal contracts below the higher rate that may be generally more prevalent in the community—by factoring into (and lowering) the calculation of the average that is used to set the minimum wage rates on local Federal contracts.

For example, the 2001 wage determination for electricians in Eddy County, New Mexico, was an average rate based on responses that included lower-paid workers that had been brought in from Texas by a Texas electrical contractor to work on a single job. As the ARB noted in reviewing a challenge to the wage determination, the result was that “contract labor from Texas, where wages reportedly are lower, effectively has determined the prevailing wage for electricians in this New Mexico county.”

New Mexico Nat'l Elec. Contractors Ass'n,

ARB No. 03-020, 2004 WL 1261216, at *8 (May 28, 2004).

To address the increasing tension between the current methodology and the purpose and definition of “prevailing,” the Department proposed in the NPRM to return to the original three-step process. The Department expects that re-introducing the 30-percent threshold will reduce the use of average rates roughly by half—from 63 percent to 31 percent. The data from the regulatory impact analysis included in section V suggests that returning to the three-step process will continue to result in 37 percent of prevailing wage rates based on the majority rule, with the balance of 32 percent based on the 30-percent threshold, and 31 percent based on the weighted average.

As part of its review of the wage determination definition and methodology, the Department also considered, but decided against, proposing to use the median wage rate as the “prevailing” rate. The median, like the average (mean), is a number that can be unrelated to the wage rate paid with the greatest frequency to employees working in the locality. Using either the median or the average as the primary method of determining the prevailing rate is not consistent with the Department's long-held interpretation of the meaning of the term “prevailing” in the Davis-Bacon Act.

See

47 FR 23645. The Department therefore proposed to return to the three-step process and the 30-percent threshold, and did not propose as alternatives the use of either the median or mean as the primary or sole methods for making wage determinations.

(1) Comments on the Definition of “Prevailing Wage”

The Department received many comments regarding the definition of the term “prevailing wage” and the proposed return to the three-step process and the 30-percent threshold. These included comments in favor of the proposal, comments in favor of keeping the current definition, comments suggesting that the Department abandon the “modal” methodology entirely and use only an average, and comments suggesting the Department should use data from sources other than its wage surveys before applying any specific methodology. Having reviewed and considered all the comments, the Department has decided that the best course is to adopt the re-definition of “prevailing wage” as proposed and return to the three-step process that was in effect from 1935 to 1983.

The Department continues to believe, as it has consistently for over 85 years, that the best methodology for determining the “prevailing wage” under the Davis-Bacon Act is one that uses a mathematical mode to determine “the most widely paid rate” as the “definition of first choice.” 47 FR 23645. The modal definition of prevailing as “the most widely paid rate” is the methodology that is most consistent with Congress's use of the word “prevailing” in the statutory text. Commenters in support of the Department's proposal cited to various dictionary definitions of the word “prevailing” that support this conclusion. The Construction Employers of America (CEA), for example, noted the definition of “prevailing” as “most frequent” or “generally current” and descriptive of “what is in general or wide circulation or use” from Webster's Third New International Dictionary (1976).

Accord

5 Op. O.L.C. at 175. The Department agrees that this and other similar dictionary definitions support the use of a modal methodology as the method of first choice.

Although the legislative history of the Act does not suggest that Congress understood there to be only one possible way of determining the prevailing wage,

there is no question that a modal methodology was within the common and ordinary public meaning of the term “prevailing” at the time. One

contemporaneous exchange from 1932 is particularly instructive. During an early debate over potential amendments to the Act, the Associated General Contractors (AGC) explained that union representatives believed the prevailing rate should always be a collectively bargained union wage, while the contractors, many members of Congress, and Federal contracting agencies believed it should be “the rate paid to the largest number in a particular locality at a given time”—in other words, the modal rate.

See, e.g.,

74 Cong. Rec. H6516 (daily ed. Feb 28, 1931) (statement of Rep. William Kopp) (noting that some might argue “the term `prevailing rate' has a vague and indefinite meaning,” but that this was not an obstacle because “the power will be given . . . to the Secretary of Labor to determine what the prevailing rates are”).

See

Regulation of Wages Paid to Employees by Contractors Awarded Government Building Contracts: Hearings before the Committee on Labor, House of Representatives, 72nd Cong., 1st Sess., on S. 3847 and H. R. 11865 (Apr. 28, 1932) at 34-35. The National Association of Manufacturers, similarly, argued that the prevailing wages should be “considered as that being paid to the largest number in the particular locality at a particular time.”

Id.

at 71-72.

See also

5 Op. O.L.C. at 175-76 (noting that this testimony leading up to the 1935 amendments “indicates a common understanding by spokesmen for labor and management, as well as individual legislators, that the `prevailing' wage was the wage paid to the largest number of workers in the relevant classification and locality”).

Several commenters on the Department's current proposal also argued that a modal methodology is generally more consistent with the purpose of Davis-Bacon Act. These commenters, including the National Black Worker Center, the International Union of Bricklayers and Allied Craftworkers, and others, argued that the use of a modal methodology results in a prevailing wage rate that is “actually paid” to workers in the area. These commenters said that average rates are less preferable because they are “artificial” and may not mirror any of the actual wage rates paid in the community. North America's Building Trade Union (NABTU), among others, asserted that “average rates paid to no one are not `prevailing[.]' ” Many unions and contractor associations, including the Washington State Building and Construction Trades Council (WA BCTC) and NABTU, noted that the use of wage rates that are actually paid to workers in the community is more likely to protect local construction firms from being underbid by unscrupulous low-wage contractors, which is the purpose of the Act.

Accordingly, commenters in favor of the proposal said averages should only be used as a fallback method when there is no clear rate prevailing in a given area.

See also

Staff of the H. Subcomm. on Lab., 88th Cong., Administration of the Davis-Bacon Act, Rep. of the Subcomm. on Lab. of the Comm. on Educ. & Lab. (Comm. Print 1963) (1963 House Subcommittee Report), at 7-8; 5 Op. O.L.C. at 177 (quoting the 1963 House Subcommittee Report).

A wide range of commenters that supported the proposal agreed with the Department that the use of an average—rather than a “modal” number identifying the most prevalent wage rate—is less preferable because the use of an average allows outlier wage rates paid to very few workers to influence the prevailing wage. The Leadership Conference on Civil and Human Rights (LCCHR), the National Women's Law Center, Oxfam America, and several other civil rights and worker advocacy organizations similarly commented that “reliance on weighted averages creates the potential for a single employer's rates that are exceptionally high or exceptionally low having outsize influence in determining the prevailing wage.” Commenters noted that this feature of averages makes the overuse of averages less consistent with the Act's purposes of limiting the depressive effect of low-wage contractors on the wage rates in the local community.

Commenters supportive of the Department's proposal also argued that this characteristic of average rates is particularly problematic for maintaining prevailing local construction standards where the use of an average results in a prevailing wage rate that is lower than a modal rate. As a Professor of Economics at the University of Utah commented, “[b]ecause the mean is sensitive to a long tail of lower wages compared to the mode, the mode is less likely to undercut local labor standards, including fringe benefits which underpin training and apprenticeship programs.” Conversely, the commenter noted, “the modal wage will deter market failures associated with short-run bidding practices that incentivize bidders to jettison all but the most necessary short-run costs of specific projects.”

In addition to determining that a modal methodology continues to be preferable, the Department proposed to return to the lower 30-percent threshold for using the mode, before falling back to the use of an average rate. Several commenters, including think tanks such as Americans for Prosperity and Institute for the American Worker (AFP-I4AW) and Competitive Enterprise Institute (CEI), opposed this proposal because they asserted that only a wage rate paid to a “majority” of workers fits the term “prevailing.” The National Federation of Independent Business (NFIB) asserted that 30 percent did not fall within the meaning of “prevailing” when Congress enacted the DBA in 1931 and the Department's initial regulation was “erroneous” at the time. CEI cited to a definition of “prevailing” as meaning “accepted, used, or practiced by most people.”

CEI asserted that the term “most people” used in that context “can only mean `a majority' ” and therefore that “30 percent is not `prevailing' under any meaningful sense of the term.”

The comment raising this language cited to an entry for “prevailing” in the online version of Merriam-Webster's dictionary. The Department was not able to find that language at the cited location, but was able to find it in an online version of a thesaurus from the same publisher.

See Prevailing,

Merriam-Webster's Thesaurus,

https://www.merriam-webster.com/thesaurus/prevailing

On the other hand, many commenters supported the Department's proposal and criticized the 1982 rule for seeming to conflate the dictionary definitions of “prevailing” with a “majority.” These commenters, including Mechanical Contractors Association of America (MCAA), National Electrical Contractors Association (NECA), and the UA, argued that the term “prevailing” is properly understood and defined as the most common or prevalent—which may be, but is not necessarily, a “majority.” If Congress had intended for the Department to determine only a “majority” wage, they argue, Congress would have explicitly stated as much in the statutory text. NECA and CEA noted that the interpretation of “prevailing” as not necessarily a majority was supported by the 1963 report of the House Subcommittee that examined the 30-percent threshold in depth before the passage of the 1964 amendments to the Act.

A joint comment from the Pennsylvania Attorney General and the Pennsylvania State Department of Labor and Industry (PAAG and PADLI) supported the reversion to the original definition, noting that it “aligns with the underlying interpretation of the word `prevailing' as the `most widely paid rate.’ ”

1963 House Subcommittee Report, at 8.

The Department agrees with these commenters that the 30-percent threshold is consistent with the meaning of the word “prevailing” because “prevailing” is not coextensive with “majority.” A statute is normally interpreted with reference to the ordinary public meaning of its terms “at the time of its enactment.”

Bostock

Clayton Cnty.,

140 S. Ct. 1731, 1738 (2020). Dictionaries from around the time of the 1935 amendments to the Act, when Congress revised the DBA to require the Secretary to predetermine prevailing wage rates, had definitions similar to the one cited in the 1981 OLC opinion.

See, e.g., Prevailing,

Merriam-Webster, Webster's Collegiate Dictionary (5th ed. 1936) (“Very generally current; most frequent; predominant” with synonyms of common, widespread,

extensive, and prevalent);

Prevailing,

Oxford English Dictionary, Vol. VIII (1933) at 1334 (“2. Predominant in extent or amount; most widely occurring or accepted; generally current”); 5 Op. O.L.C. at 175. When there are only two kinds being compared, the “most frequent” or “most widely occurring” of the two kinds will be a majority, and thus only a majority will be prevailing. But the same is not true when a variety of kinds are compared. In such circumstances, even if a majority will still necessarily be prevailing, it does not follow that anything less than a majority cannot be considered prevailing. Rather, as the 1963 House Subcommittee Report concluded, “ ‘prevailing’ means only a greater number. It need not be a majority.”

1963 House Subcommittee Report, at 8.

In opposing the proposal, AFP-I4AW noted that in the 1981-1982 rulemaking the Department had agreed with commenters that stated “a rate based on 30 percent does not comport with the definition of `prevailing[.]' ”

The Department did not provide further explanation of this argument in the 1982 final rule, but had stated in the 1981 NPRM that the 30-percent rule “ignores the rate paid to up to 70 percent of the workers.”

See

46 FR 41444. Several commenters that opposed the return to the 30-percent rule, including AFP-I4AW, Associated Builders and Contractors (ABC), and Clark Pacific, stated that they still found this reasoning persuasive.

47 FR 23644.

Similarly, CEI opposed the use of the 30-percent rule because it stated that the fact that other workers may earn less than the wage determined to be the prevailing wage is “highly significant,” because it indicates that the labor market is “more competitive in terms of wages.” Under this reasoning, however, only an average rate would be sufficient, because any modal (or median) rate would not include all of the wage rates paid. Using only an average is not consistent with the Department's long-held understanding of the meaning of the term “prevailing.”

See

47 at FR 23644-45. Neither the text nor the legislative history of the Act suggests that the term prevailing wage was intended to necessarily capture and reflect all of the wage rates that are paid in an area. Instead, the Department has understood the statute as better carried out with a methodology that seeks to determine which among those wage rates is prevailing.

The Department disagrees. As an initial matter, the characterization of the 30-percent threshold as “ignoring” rates is not unique to that specific threshold. Rather, it is a feature of any rule based on a mathematical “mode,” in which the only value that is ultimately used is the value of the number that appears most frequently. This is in contrast to using a mean (average), in which the values of all the numbers are averaged together, or a median, which uses only the midpoint value. Both the 30-percent threshold and the majority rule are modal rules in which the values of the non-prevailing wage rates do not factor into the final analysis. This feature of a modal analysis can be viewed as particularly helpful for avoiding an unwarranted downward or upward impact from outlier wage rates. As the International Association of Sheet Metal, Air, Rail and Transportation Workers and the Sheet Metal and Air Conditioning Contractors National Association (SMART and SMACNA) noted in a joint comment, “[w]hen using the mean, unusually low or high values distort the data; the mode, by contrast, eliminates from the analysis data that grossly deviate from what workers are actually paid and, therefore, would depress labor standards if included.”

Moreover, the characterization of the 30-percent threshold as ignoring up to 70 percent of wage rates distorts how the analysis is applied in practice. In the three-step process, the first step is to adopt the majority rate if there is one. Under both the proposed three-step process and the current majority-only rule, any wage rate that is paid to a majority of workers would be identified as prevailing. Under either method, the weighted average will be used whenever there is no wage rate that is paid to more than 30 percent of employees in the survey response. The difference between the current majority process and the three-step methodology is solely in how a wage rate is determined when there is no majority, but there is a significant plurality wage rate paid to between 30 and 50 percent of workers. In that circumstance, the current “majority” rule uses averages instead of the rate that is actually paid to that significant plurality of the survey population. This is true, for example, even where the same wage rate is paid to 45 percent of workers and no other rate is paid to as high a percentage of workers. In such circumstances, the Department believes that a wage rate paid to between 30 and 50 percent of workers—instead of an average rate that may be actually paid to few workers or none at all—is more of a “prevailing” wage rate.

As the OLC concluded in 1981, the use of an average instead of the 30-percent rule may be particularly inappropriate in circumstances where “there is a wide variation in rates of wages and a large minority of persons paid significantly lower wages; use of an average in such a case might result in a contract wage well below the actual wages paid a majority of employees.” 5 Op. O.L.C. at 177 n.3.

NABTU and other commenters in favor of the Department's proposed return to the 30-percent threshold noted that Congress specifically considered on numerous occasions whether to abolish the 30-percent rule and declined to do so.

Similarly, CEA commented that Congress's repeated expansion and amendment of the Act from 1935 to 1982 without changing or addressing the definition of prevailing wage should be interpreted as “persuasive evidence that the interpretation is the one intended by Congress.’ ”

CFTC

Schor,

478 U.S. 833, 846 (1986) (quoting

NLRB

Bell Aerospace Co.,

416 U.S. 267, 274-75 (1974) (footnotes omitted)). The Department agrees that this legislative acquiescence is significant. It may not necessarily mean that the 30-percent rule was the only interpretation that was intended by Congress—especially in light of the subsequent Congressional acquiescence to the imposition of the majority-only rule.

However, the expansion of the Act, particularly in 1964 after the extensive hearings regarding the 30-percent rule, suggests that Congress did not believe that the 30-percent rule was “erroneous” at the time of its enactment or otherwise believe that it “did not comport” with the definition of prevailing.

Cf.

5. Op. O.L.C. at 176 (noting Congress had acquiesced to the Department's interpretation of the term prevailing as embodied in its 1935 regulations).

See, e.g.,

Federal Construction Costs Reduction Act of 1977 (S. 1540, H.R. 6100); Davis-Bacon Act—Fringe Benefits (H.R. 404): Hearings Before the General Subcomm. on Labor of the H. Comm. on Educ. & Labor, 88th Cong. at 38-39, 125, 219, 225-230 (Mar. 1, 7, 12, 21, 22, and 26, 1963).

One individual commenter opposing the Department's proposal asserted that Congress's inaction in reimposing the 30-percent rule should be considered evidence that the 30-percent rule “contravenes, rather than is required by, the statutory text.” But given the wide discretion the courts have found the DBA affords to the Secretary of Labor, the Department does not believe that the acquiescence to the Department's decision to use one specific modal threshold can be understood as barring it from using another.

In addition to considering questions regarding Congressional acquiescence, the Department has also considered whether the length of time that the majority-only rule has been in place has led to reliance interests among regulated entities that would counsel against reversion to the three-step process. While some commenters referred to the length of time the rule had been in effect, their comments generally did not focus on related reliance interests. The Department does not believe that any potential reliance interests would be so significant as to outweigh the objectives of seeking to align the prevailing wage methodology better with the longstanding meaning of the term prevailing and of seeking to better protect workers against the depressive

effect on wage rates of low-wage contractors. The Department's illustrative study of the proposed methodology change, in section V.D. below, suggests that the change may lead to higher required prevailing wage rates in some places and lower wage rates in others. The magnitude and direction of changes, however, should not be significantly different in their effect on contractors than the fluctuations in prevailing wage rates that already occur between wage surveys as a result of changes in local economies and shifts in regional labor markets. Even if the part 1 changes were to have significant effects on wage rates in certain local areas, any reliance interests of local contractors, governmental agencies, or workers on prior wage rates would be minimal, given that the changes to the wage determination processes generally will not affect current contracts—which will continue to be governed by the wage determinations incorporated at the time of their award, with limited exceptions. Most of the revisions to part 1 will only apply to wage surveys that are finalized after the rule becomes effective, and thus they will generally apply only to contracts awarded after such new wage determinations are issued.

Contractors will therefore be able to factor any new wage rates into their bids or negotiations on future contracts.

See also

section III.C. (“Applicability Date”) below.

The Department received many comments in favor of and opposed to the use of the 30-percent threshold for other reasons. A number of commenters commented favorably on the use of 30 percent specifically as a reasonable modal threshold to choose. As the LCCHR, the National Women's Law Center, Oxfam America, and several other civil rights and worker advocacy organizations commented, the choice of the 30-percent threshold appropriately aligns the rate selected with the actual wages paid to “significant shares” of workers in a covered job classification. The Dakotas Mechanical Contractors Association (DMCA) and the Sheet Metal, Air Conditioning and Roofing Contractors Association stated that if 30 percent are paid the same rate, it is likely the prevailing rate for skilled workers in the area. The Center for American Progress Action Fund noted that the 30-percent rule is also followed by some states in the implementation of their own State prevailing wage programs.

Some commenters argued that a 50-percent threshold for using a modal rate is simply too high for many geographic areas. The DMCA, for example, noted that when there are multiple large construction projects going on in the Dakotas, many contractors travel from outside the area, and counting wage rates from these out-of-town contractors can make it difficult for the actual local rate to satisfy a 50-percent threshold.

See, e.g.,

Haw. Code R. section 12-22-2(b) (30-percent threshold in Hawaii); 820 Ill. Comp. Stat. 130/4 section 4(a) (30-percent threshold in Illinois). Wyoming uses a version of the three-step process in which the prevailing wage is a majority, or 30-percent, unless more than one wage rate reaches the 30-percent threshold, in which case a weighted average is used.

See https://dws.wyo.gov/wp-content/uploads/2022/04/Labor-Standards-2022-Prevailing-Wage-Rates.pdf

. Minnesota and California use modal methodologies, but do not have specific thresholds.

See

Minn. Stat. section 177.42; Cal. Lab. Code section 1773.9(b)(1).

Several commenters opposing the proposed reversion to the 30-percent rule asserted that a reversion to the 30-percent rule would result in rates that are less accurate or less likely to reflect the actual wage and fringe benefit rates in a locality, and therefore are inherently not “prevailing” under the meaning of the statute. ABC stated that a survey of its Federal contractor members showed that only 12.6 percent of its respondents stated that the reversion to the 30-percent rule would increase the accuracy of wage determinations. The Modular Building Institute (MBI) commented that a 30-percent threshold is too small a sample on which to base a prevailing wage. According to the Taxpayers Protection Alliance, returning to the 30-percent rule “invites cherry-picking rather than serious analysis.” On the other hand, several commenters in favor of the Department's proposal asserted, similar to the minority of respondents to ABC's survey, that returning to the 30-percent rule would increase the accuracy of wage determinations.

In making arguments about accuracy, most commenters for and against the proposal did not reference data or evidence to support their views. Commenters opposing the proposal that did cite data compared potential outcomes under the 30-percent threshold—or any modal determinations based on voluntary wage surveys—with average rates calculated by other sources or by reference to studies that found increases in total costs from the use of any prevailing wage at all. Commenters also argued that accuracy can be judged by the potential for the percentage of wage determinations based on CBAs to be higher than the union density in the local area.

The Department does not agree with these measurements of accuracy and instead understands these arguments as fundamentally about what the meaning of “prevailing” should be, or whether prevailing wage laws are good policy in the first place. While a comparison of costs in jurisdictions in which a State prevailing wage law applies with those where there is no such requirement may be helpful to understanding the cost impacts of prevailing wage requirements, that comparison is not helpful in understanding whether a certain prevailing wage methodology results in wage determinations that are “accurate” or not, because the point of the prevailing wage law is to eliminate the payment of substandard wage rates that may be paid in the absence of the law. For similar reasons, a comparison with average rates or union density does not reflect accuracy—rather it reflects different understandings of the term “prevailing.”

ABC and the National Association of Home Builders (NAHB) cited data from a 2010 GAO report and subsequent data showing that as of 2010, a union rate prevailed in 63 percent of all then-existing wage determinations; in 2018, a union rate prevailed in 48 percent of determinations; and in 2022, a union rate prevailed in 42 percent of determinations. The commenters contrasted these numbers with data from the BLS that shows union density currently at less than 20 percent of the construction labor market.

The Department also notes that, while the percentage of overall wage determinations based on collective bargaining rates nationwide has been higher than measures of union density in the construction industry generally, the percentage of wage determinations based on collectively bargained rates has significantly declined in recent years. NAHB and ABC pointed out that the 2011 GAO report stated that at the time 63 percent of published wage rates were union prevailing.

See

2011 GAO Report, at 20. ABC notes current statistics from the Department show 42 percent are based on collectively bargained rates.

AFP-I4AW asserted that it is arbitrary to choose 30 percent instead of one of the other “infinite percentages that might be chosen between 0 and 50 percent.” The Department disagrees with the premise that the 30-percent threshold is arbitrary and therefore

impermissible. As one commenter in favor of the proposal, the Iron Workers International Union (Iron Workers), stated, the “30 percent” rule can be seen as a “middle position” that the Department adopted in 1935. Among modal rates, the wage rate based on a 20 percent modal rate or even lower might also have been considered a reasonable interpretation of the term “prevailing wage,” rendering 30-percent a compromise among all of the different definitions being advanced at the time.

See Bldg. & Constr. Trades Dep't

Donovan,

553 F. Supp. 352, 354 (D.D.C. 1982) (“There is nothing intrinsically appropriate or inappropriate to the thirty percent rule or to any other figure as representing the `prevailing wage.' ”).

The fact that the Department could have chosen an even lower number, or no modal threshold at all, does not make the choice of 30 percent impermissible. The number is a familiar one that the Department used over five decades; as commenters noted, it represents at least a significant share of workers in a survey; and the Department has tested the potential outcome of returning to the number and found that it will alleviate concerns about overuse of average rates.

Cf. Ralph Knight, Inc.

Mantel,

135 F.2d 514, 518-19 (8th Cir. 1943) (holding the percentage threshold in an FLSA regulation was not arbitrary because it was reasonable).

The 1982

Donovan

district court decision enjoined several elements of the 1981-1982 rulemaking but upheld the Department's decision to eliminate the 30-percent threshold. In affirming the district court's decision on the 30-percent threshold, the D.C. Circuit stated that it affirmed “generally for the reasons stated in [the district court's] opinion.”

Donovan,

712 F.2d at 616.

The ABC and several other commenters criticized the Department for proposing to return to the 30-percent threshold without addressing concerns they have about the methodology of the wage survey program that produces the underlying numbers to which the three-step process would be applied. According to ABC and others, the Department should use more sophisticated representative sampling and statistical regression methods to come up with prevailing rates because of low response rates, low sufficiency thresholds and therefore small sample sizes, and response bias in the Department's voluntary Davis-Bacon wage survey program. ABC and the National Association of Home Builders (NAHB) referenced reports by the Department's OIG expressing concern about low response rates to WHD's wage surveys, including a 2019 report in which OIG calculated that as many as 53 percent of eligible contractors had not provided wage data on 7 surveys that were analyzed.

ABC and others argued that union contractors have a higher interest in responding to the wage surveys, and so the surveys tend to disproportionately reflect union rates and are therefore unreliable.

In a joint comment, a group of housing industry associations and entities stated that certain segments of the residential building industry have “no incentive to participate in a survey method that provides no direct benefit to their business.” Without making changes to the survey process to better account for non-union contractors, ABC argued, the Department should not be changing the threshold for identifying the prevailing wage. ABC stated that the survey process in its current form is “incapable of accurately determining whether a single rate is paid to 30% (or a majority) of local construction workers.”

See

OIG, U.S. Department of Labor, No. 04-19-001-15-001, “Better Strategies are Needed to Improve the Timeliness and Accuracy of Davis-Bacon Act Prevailing Wage Rates,” 8, 15 (2019). Available at

https://www.oig.dol.gov/public/reports/oa/2019/04-19-001-15-001.pdf

As evidence that the Department's Davis-Bacon wage surveys are statistically unrepresentative of the construction workforce, ABC asserted that average wages—both economywide and in specific occupations (construction or otherwise)—are consistently higher than median wages in the United States and most industrialized economies. For example, ABC points to BLS's May 2021 Occupational Employment and Wage Statistics (OEWS) survey showing that, nationally, average wages exceed median wages in 51 of 64 detailed construction occupations. ABC argues that the Department's surveys are unrepresentative because, in the wage determinations developed using the survey data and using the majority rule, the majority rate (which should be the same as the median) consistently exceeds wages calculated as survey averages.

ABC, NAHB, and other commenters stated that the Department should have considered using data from the BLS, which performs representative sampling on surveys with higher response rates and larger sample sizes and uses other more sophisticated regression methods, and therefore would be more accurate. According to ABC and an individual commenter, the use of BLS data would result in more timely wage determinations and decrease the costs of Federal construction, making more projects viable and increasing construction employment. ABC acknowledged that the Department has previously declined to use BLS data for DBA wage determinations for a number of reasons, including that BLS data does not have the same benefits information, data by county level, or by construction type. But ABC asserted that none of these reasons entirely foreclose the use of such data, and it cited the fact that the Department already uses BLS data for wage determinations under the Service Contract Act (SCA), which has similar statutory parameters, as well as the Foreign Labor Certification Program, and with some statistical modeling, for Federal employee pay under the Federal Employee Pay Comparability Act. ABC also argued that the Department's current use of larger county groupings to identify wage rates for counties with insufficient data and the proposal in the NPRM to remove the bar on cross-consideration of rural and metropolitan data both undercut the Department's arguments against using BLS data. NAHB and the Mortgage Bankers Association (MBA) also suggested that the Department should consider outsourcing the wage data collection process to third-party organizations they believe would be better equipped to collect greater quantities of data.

A joint comment from the National Asphalt Pavement Association, National Ready Mixed Concrete Association, and National Stone, Sand & Gravel Association (NAPA, NRMPCA, and NSSGA) suggested that reverting to the use of a 30-percent threshold is “unnecessary” because there are other ways to improve the survey process. They suggested using the certified payrolls that are submitted on DBRA projects to help identify prevailing wages.

They also suggested updating and standardizing classifications that are “outdated” and confusing where they differ across political subdivisions. AGC suggested that the Department should revise the wage survey process to allow contractors to report wage information by individual craft classifications in each county by construction type, instead of broken-down project-by-project.

The Department appreciates this suggestion, but notes that using certified payrolls instead of the wage survey process would result in prevailing rates based entirely on data from DBRA-covered projects. While such data could be helpful in certain circumstances in which there is not sufficient data from private sources, it could not be used instead of the wage survey process because the DBA contemplates a wider analysis of wage rates that includes those on wholly privately funded projects where such data is available.

See generally infra

section III.B.1.iii.(B) (“29 CFR 1.3(d)”).

Several commenters stated that even if the 30-percent rule had been permissible previously, the Department could not reasonably return to it because the construction labor market has changed and prevailing rates “rarely occur in the modern economy.” ABC noted that union density has declined in the construction labor market from 34 percent in 1981 to under 14 percent in recent years. The Association of Washington Housing Authorities (AWHA) stated that the increase in

reliance on weighted averages actually reflects reality in certain construction types where union participation is lacking. AWHA also stated that there is no need to return to the 30-percent rule because there is better labor market wage information now available than there was when the 30-percent rule was last in effect, with both proprietary and public databases now containing “up-to-date wage and salary information on thousands of job classifications at varying geographic levels.”

Finally, comments from ABC and a group of U.S. Senators asserted that the Department's reasoning for its proposal is contrary to the D.C. Circuit's decision in

Building & Construction Trades' Department

Donovan,

712 F.2d 611, 616-17 (D.C. Cir. 1983). In that decision, the Department's 1981-1982 rulemaking eliminating the 30-percent threshold had been challenged. The D.C. Circuit stated that the Department's new definition of “prevailing” as, first, the majority rate, and second, a weighted average, was “within a common and reasonable reading of the term” and “would not defeat the essential purpose of the statute, which was to ensure that federal wages reflected those generally paid in the area.”

Id.

at 616-17. ABC stated that this holding allowing the Department to eliminate the 30-percent threshold could not be squared with the Department's reasoning in the NPRM that the overuse of averages was inconsistent with the text and purpose of the Act.

See

87 FR 15704.

Considering these comments, the Department agrees with the commenters in favor of the proposal that the 30-percent threshold is a reasonable threshold that represents the best course for making wage determinations based on wage rates that are actually paid to workers in the relevant area. The Department also believes that returning to the use of the 30-percent threshold at the second step in the wage determination process is preferable for the same reasons that it is preferable to use a modal methodology at all instead of using averages or the median for all wage determinations. The mode is more consistent with the term “prevailing,” and it is in general more protective of prevailing wage rates against the depressive effect of low-wage contractors. Even when adopting the current majority threshold for modal wage determinations in 1982, the Department reiterated this long-held interpretation that the “most widely paid rate” should be the “definition of first choice” for the prevailing wage, and that wage determinations should “mirror, to the extent possible, those rates actually paid in appropriate labor markets.” 47 FR 23645.

The Department disagrees that the D.C. Circuit's

Donovan

decision precludes a return to the 30-percent threshold or prevents the Department from concluding that an overuse of averages is in tension with the Department's long-held interpretation of the Act. In

Donovan,

the court stated that the majority-only rule was “within a common and reasonable reading” of the term prevailing, and “would not defeat the essential purpose of the statute.” 712 F.2d at 616-17. The court did not, however, state or even suggest that the majority rule represented the only proper reading of the statute. To the contrary, the court stated that it was upholding the new rule because “the statute delegates to the Secretary, in the broadest terms imaginable, the authority to determine which wages are prevailing.” 712 F.2d at 616.

As the Department explained in the NPRM, there has been a significant increase in the use of weighted averages between 1983 and the present—from as low as 15 percent prior to the implementation of the current regulations to 63 percent in the Department's review of 19 recent surveys. Several commenters noted that this increase in the use of averages appears to be far beyond what was expected at the time the Department implemented the majority-only rule and at the time of the D.C. Circuit opinion. For example, the unions that opposed the 1981-1982 rulemaking in court argued that it could result in “a third or more” of wage rates based on weighted averages.

Donovan,

712 F.2d at 616. Now, nearly double that number—two thirds—of prevailing wage rates published from recent surveys have been based on weighted averages. These new circumstances represent a departure from the Department's longstanding interpretation of the Act. 5 Op. O.L.C. at 176-77.

The Department also disagrees with comments suggesting the Department can only justify its return to the 30-percent threshold by finding that the current majority rule is per se not allowed by the statute and suggesting furthermore that the

Donovan

decision bars the Department from reaching that conclusion. As noted, however, the decision in

Donovan

reflects that there can be more than one possible threshold for determining whether a wage rate is prevailing, and that the statute delegates the decision about methodology to the Secretary of Labor. 712 F.2d at 616. The Department has concluded that the original three-step process is preferable to the majority-only rule because it is more consistent with the meaning of the word “prevailing” and will be more protective against the depression of wage rates by low-wage contractors. Under these circumstances, the Department does not need to find that the current overuse of averages renders the majority-only rule effectively barred by the statute.

The Department also considered the comments critiquing the interface between the wage survey program and the Department's use of a modal methodology to determine prevailing wages and the use of the 30-percent modal threshold in particular. The Department does not believe it is necessary or preferable to abandon the current Davis-Bacon wage survey process, or to require by regulation that survey data be adjusted with regression or other similar statistical analyses. The process of adjusting survey data using weighting, imputation, or other representative sampling methods would require additional data regarding the universe of projects and classifications of workers—divided by construction type—that does not currently exist and would be overly burdensome and costly to obtain.

Moreover, other commenters on the rule specifically opposed the use of sampling or other similar methodologies because the decisions about the underlying assumptions used in the calculations or modeling would give the Department too much discretion that would be difficult for stakeholders to scrutinize. Finally, such sampling or other statistical methods could also significantly increase the likelihood that the wage rates the Department publishes would be akin to weighted averages and would not be wage rates that are actually paid to workers in the relevant areas. The Department declines to impose such requirements in this final rule.

Similarly, the 2019 OIG report noted WHD officials' concern that using statistical sampling during the clarification process instead of manual reviews of survey data might be less efficient and effective than current processes, and that “use of statistical sampling in lieu of comprehensive clarification would likely result in the publication of fewer, and less robust, wage determinations.” Report at 7, 43.

The Department also considered ABC's and others' arguments that it should entirely discontinue the Davis-Bacon wage surveys and instead use data from BLS surveys to determine prevailing wages in the first instance. As ABC recognized in its comment, the Department has explored this possibility on various occasions in the past at the recommendation of the GAO and others. For example, ABC cited a 2004 letter

from the Assistant Secretary for Employment Standards, to the Department's OIG, noting the actions the Department had taken to consider this option, including funding pilot surveys to determine the feasibility of collecting fringe benefit data as part of BLS's National Compensation Survey (NCS), and working with BLS to examine the extent to which the Occupational Employment and Wage Statistics (OEWS) survey might provide detailed construction industry wage rate information by locality and occupation.

Letter from Victoria A. Lipnic, Assistant Secretary for Employment Standards, to Elliot P. Lewis, Assistant Inspector General for Audit (Feb. 18, 2004). Available at:

https://www.oig.dol.gov/public/reports/oa/2004/04-04-003-04-420x.pdf.

The Department has repeatedly concluded that relying on BLS data sources to determine prevailing wages instead of continuing to conduct Davis-Bacon wage surveys is not preferable, and the Department again reaches this conclusion. No BLS survey publishes, at a county level, the wage data, fringe benefit data, data for sufficiently specific construction craft classifications, and data by construction type, that would align with the Department's interpretations of the statutory requirements to determine prevailing wages for “corresponding class[es]” of workers on “projects of a character similar” within “civil subdivisions of the State” in which the work is to be performed. 40 U.S.C. 3142(b).

The Department does not agree with ABC that the Department's current use of larger geographic groupings under certain conditions suggests that the Department should adopt BLS data that is compiled for areas larger than a county. The scope of consideration regulations at § 1.7 allow the Department to consider data from larger geographic areas only when there is insufficient wage survey data in a given county. This reflects the Department's long-established position that the county level is the appropriate level at which to determine prevailing wage rates where possible, and as such that the wholesale adoption of BLS data compiled for larger areas generally would not be appropriate. The Department also considered whether it would be possible to combine BLS surveys or use underlying BLS microdata instead of the Department's wage surveys but determined that the BLS's methodology does not allow such a procedure because, among other reasons, BLS does not collect data on a project-by-project basis and therefore does not capture circumstances in which employees may be paid different hourly rates for work based on the type of project. Finally, the Department's conclusion is bolstered by the widespread practice of states, many of which have adopted prevailing wage laws, that have likewise determined that wage surveys are an appropriate mechanism to set prevailing wages.

The BLS OEWS program produces employment and wage estimates for the nation as a whole, for individual states, for metropolitan areas delineated by the Office of Management and Budget (OMB), and nonmetropolitan areas, but it does not produce wage estimates at the county level, which is the default “civil subdivision” that the Department uses to determine prevailing wages.

See

Michael K. Lettau and Dee A. Zamora, BLS, “Wage estimates by job characteristic: NCS and OES program data” (2013). Available at:

https://doi.org/10.21916/mlr.2013.27.

Additionally, the data for metropolitan and nonmetropolitan areas do not allow for wage rates for occupations by industry. The NCS program provides measures of compensation trends and the incidence of employer-sponsored benefits, but only at the national and Census region levels. The BLS's Quarterly Census of Employment and Wages has data at the county level, but the data are not available by craft. Both the OEWS and NCS programs classify occupations based on job duties and responsibilities that apply nationwide in accordance with the Standard Occupational Classification system. WHD's survey program, on the other hand, has always considered local area practice in determining how work is classified for each occupation.

See, e.g.,

26 Me. Rev. Stat. Ann. section 1308 (requiring the Maine Bureau of Labor Standards to determine prevailing wages through a regularly conducted wage and benefits survey); Minn. R. section 5200.1020 (providing for annual surveys to calculate prevailing wages on covered highway and construction projects); Mont. Code Ann. section 18-2-414 (authorizing the Montana Commission of Labor and Industry to either perform a wage survey or adopt the rates set by the United States Department of Labor); Tex. Gov't Code Ann. section 2258.022 (setting the state prevailing wage either through wage surveys or by incorporating the rates set by the United States Department of Labor).

ABC is correct that the Department uses BLS data for wage determinations under the SCA, which has important statutory similarities with the DBA in that it requires payment of wages “in accordance with prevailing rates in the locality.” 41 U.S.C. 6703(1). There are several reasons, however, why the Department's decisions have been different under the SCA than under the DBA. The first is that the SCA does not contain the same statutory text as the DBA requiring prevailing wages to be based on “projects of a character similar.” 40 U.S.C. 3142(b). This distinction underscores the Department's need to survey DBA wage rates by construction type, a level of detail that does not exist in any BLS data source. In addition, the SCA contains an alternative mechanism that gives weight to collectively bargained rates by requiring them to govern certain successor contracts where the predecessor contract was covered by a CBA. 41 U.S.C. 6703(1).

Comparisons between the DBA and SCA can also be fraught because construction work is significantly different from most service work. As a Professor of Economics at the University of Utah commented on this rulemaking, the construction industry is based on a “craft classification” model—in which crafts are understood to be a collection of related skills that allow a craft worker to address a range of jobs as that worker goes from project to project, and which can only be supported with proper investment and skills training. Protecting craft classifications where they prevail was one of the core original purposes of the Davis-Bacon Act.

See

Charles Donahue, “The Davis-Bacon Act and the Walsh-Healey Public Contracts Act: A Comparison of Coverage and Minimum Wage Provisions,” 29 Law & Contemp. Probs. 488, 508 (1964) (noting the Department's deference to local craft organization in wage determinations because “[t]o do otherwise would destroy craft lines which the statute seeks to preserve”);

see also Donovan,

712 F.2d at 625 (noting that Congress was “quite clear” in 1935 that it was an “evasion of the Act” to break down craft classifications where they prevail). This industrial organization and the legislative history support the Department's stricter approach under the DBRA to protecting actual wage rates that prevail because when those rates are higher than the average wage, they are often higher because they are incorporating apprenticeship and other training costs that are critical for the maintenance of the craft organization of the local construction market.

It also explains why the Department does not agree with ABC's suggestions that the Davis-Bacon program should adopt the standardized national Standard Occupational Classification system for identifying construction worker classifications and also abandon the division of wage rates by “construction type,” so as to align all Davis-Bacon classifications with the format of BLS program data. Similarly, the differences between the SCA and the DBA and the industry sectors they cover, and the craft-protection focus of the DBA, also explain why the Department does not believe it is appropriate, as ABC suggests, to adopt a single nationwide

fringe benefit rate under the DBRA in the same way that it has under the SCA.

Notwithstanding these differences, under the SCA regulations, the Department also may publish prevailing collectively bargained rates rather than rely on BLS data.

See

29 CFR 4.51(b) (“Where a single rate is paid to a majority (50 percent or more) of the workers in a class of service employees engaged in similar work in a particular locality, that rate is determined to prevail.”).

ABC commented that the Department should be more flexible with how it analyzes the statutory requirements and find that the statute permits the use of averages or modal approximations derived from statistical modeling rather than revert to the three-step process and retain the current wage survey process. ABC and other commenters also suggested the use of BLS data would have other important benefits. ABC stated that directly using BLS data would improve the timeliness of wage determinations because BLS surveys are updated annually. ABC and the group of U.S. Senators stated that using BLS data would eliminate an impediment preventing small firms from bidding on Davis-Bacon contracts because it would eliminate the problem of missing classifications on wage determinations. The commenters said that such missing classifications can be an impediment for small firms because it is costly and complicated to request conformances. ABC suggested that the Department should consider transferring funding from WHD to BLS by contracting with BLS to provide data, with the additional funding to BLS going to address any ways in which BLS methods are deficient for DBRA purposes.

Having considered these arguments, the Department continues to believe that the best course of action is to adopt the proposed reversion from the majority rule to the three-step process as the methodology for making wage determinations. The Department agrees that it is important to continue seeking ways to improve contractor participation in its voluntary wage surveys, which will have the benefit of increasing sample sizes for wage determinations and making wage determinations possible for more classifications. The Department has initiated a process to revise the wage survey form (WD-10 form) that is used during wage surveys. In that process, it proposed a number of changes in order to decrease the burden on contractors of responding to the survey and lead to higher survey response rates.

See

87 FR 36152, 36152-53 (June 15, 2022). The Department, through that process, is also considering updates to the directory of classifications that is listed on the form, and to procedures to assist in capturing information about local area practice and industry changes in classifications over time. Thus, the Department does not believe NAPA, NRMPCA, and NSSGA's concerns about outdated classifications is a persuasive reason not to adopt the changes to the methodology of determining prevailing wages from survey data. Collecting more accurate data and returning to the 30-percent threshold are supplementary, not mutually exclusive, means to determining appropriate wage rates. The Department is therefore not only returning to the use of the 30-percent threshold in this final rule, but also will continue to promote greater participation in its surveys and take related steps, such as its revision to the WD-10 form outside this rulemaking, in order to increase the pool of data that is available to determine accurate prevailing wage rates.

While the Department appreciates AGC's suggestions regarding revising the wage survey process to allow contractors to report data for workers more generally instead of on a project-by-project basis, the Department notes that the statute discusses the determination of the prevailing wage on the basis of “projects of a similar character,” 40 U.S.C. 3142(b), and that project-by-project reporting promotes accuracy in the survey process because it more readily enables the Department to identify the number of workers that were paid each reported rate (and hence to properly calculate the prevailing wage) in a given area. A data submission consisting solely of the wages and fringe benefits paid generally to a particular classification, particularly if such a submission did not identify how many workers received each identified rate, would at a minimum create challenges and inefficiencies in determining the prevailing wage rate.

The Department also agrees with commenters that addressing timeliness issues and the overuse of conformances are important goals. The use of BLS data, however, could cause its own problems with missing classifications. BLS's OEWS program, for example, uses the Office of Management and Budget's (OMB) Standard Occupational Classification (SOC) system when publishing wage estimates. The SOC system does not include a number of individual classifications that the Department commonly uses to appropriately account for local area practice and the craft system. For example, the Department often issues separate wage rates for Plumbers, Pipefitters, and Steamfitters. The OEWS program only issues a single wage rate in a given locality under SOC code 47-2152 (“Plumbers, Pipefitters, and Steamfitters”). For this reason, the Department believes that ABC's proposal to directly use the SOC system would result in less accurate craft classifications. As discussed further below, in this rulemaking, the Department is adopting new methods of reducing the need for conformances and more frequently updating wage determinations, including through the limited use of BLS data where it can reasonably be used to estimate wage-rate increases in between voluntary surveys. The Department believes these changes, once implemented, will improve the wage determination program without making a significant departure from longstanding interpretations of the statutory text and purpose of the DBRA.

(2) Comments Regarding Costs of the 30-Percent Threshold

In proposing the return to the 30-percent threshold, the Department also considered the other explanations it provided in 1982 for eliminating the rule in the first place—in particular, the potential for a possible upward pressure on wages, contract costs, or prices. In the 1982 final rule, the Department summarized comments stating that the rule is “inflationary because it sometimes results in wage determination rates higher than the average.” 47 FR 23644. The Department did not explain exactly what the commenters meant by the term “inflationary.”

See id.

Later, the Department stated simply that it “agree[d] with the criticisms of the 30-percent rule,” without specifically referencing the wage-inflation concerns.

Id.

at 23645. Later still, in a discussion of the final regulatory impact and regulatory flexibility analysis, the Department estimated that eliminating the 30-percent rule could result in a cost savings of $120 million per year.

Id.

at 23648. The Department then stated that it was adopting the new rule “not only because it will result in substantial budgetary savings, but also because it is most consistent with the `prevailing wage' concept contemplated in the legislation.”

Id.

In the current rulemaking, many commenters opposing the Department's proposed reversion to the 30-percent threshold, including several housing industry associations and entities, referenced and restated the earlier concerns about an “inflationary effect.” ABC and the group of Senators referenced criticism of the 30-percent rule by the GAO in the 1960's and 1970's, including the 1979 report that urged the repeal of the Act as a whole and related congressional hearings in which the GAO referred to the 30-percent rule as resulting in “inflated wage rates.”

Several commenters

pointed to two studies finding that prevailing wages under the DBA increase costs to taxpayers. The NAHB pointed to a 2008 study by the Beacon Hill Institute, finding that Davis-Bacon wage determinations increase the cost of Federal construction by “nearly 10 percent,”

and a study by the Congressional Budget Office (CBO) that estimated a $12 billion reduction in Federal spending from 2019 through 2028 if DBA requirements were not applied to covered projects.

CEI, stating that no more recent data is available on the economic impact of the 30-percent rule, cited a 1983 CBO estimate that the DBA's requirements added 3.7 percent to the overall cost of Federal construction projects.

They also cited a later estimate from after implementation of the majority rule, estimating that DBA requirements added 3.4 percent to the cost of Federal construction projects.

Comparing these two studies, CEI claimed, shows the difference between the 30-percent threshold and the majority-only rule accounts for about 8 percent in the overall cost of complying with the Act (or, presented differently, about 0.3 percent in the total cost of Federal construction projects).

See Administration of the Davis-Bacon Act, Hearings before the Spec. Subcomm. on Lab. of the H. Comm. on Educ. & Lab.,

87th Cong. 283 (1962) (testimony of J.E. Welch, Deputy General Counsel,

General Accounting Office) (“Our experience indicates that the methods and procedures by which minimum wage requirements for Federal and federally assisted construction contracts are established and enforced under present law have not kept pace with the expansion and increased use of such requirements.”);

Oversight Hearing on the Davis-Bacon Act, Before the Subcomm. on Lab. Standards of the H. Comm. on Educ. & Lab.,

96th Cong. 4 (1979) (testimony of Comptroller General Elmer Staats) and 60-64 (testimony of Secretary of Labor Ray Marshall criticizing GAO methodology).

Paul Bachman, Michael Head, Sarah Glassman, & David G. Tuerck, Beacon Hill Inst., “The Federal Davis-Bacon Act: The Prevailing Mismeasure of Wages,” (2008). ABC cited this 2008 report, as well as a subsequent Beacon Hill report, which updated it.

See

William F. Burke & David G. Tuerck, Beacon Hill Inst., “The Federal Davis-Bacon Act: Mismeasuring the Prevailing Wage,” (2022).

CBO, “Repeal the Davis-Bacon Act,” Dec. 13, 2018,

https://www.cbo.gov/budget-options/54786.

CBO, “Modifying the Davis-Bacon Act: Implications for the Labor Market and the Federal Budget,” July 1983,

https://www.cbo.gov/sites/default/files/98th-congress-1983-1984/reports/doc12-entire_0.pdf

CBO, “Toll Funding of U.S. Highways,” Dec. 1985,

https://www.cbo.gov/sites/default/files/99th-congress-1985-1986/reports/1985_12_tollfinancing.pdf

Several commenters, in particular in the residential building industry, expressed general concern that higher labor costs could put some projects at risk of being financially infeasible. The NAHB stated that “relatively small price increases can have an immediate impact on low-to moderate-income homebuyers and renters who are more susceptible to being priced out of the market.” According to NAHB, there are already a number of other current difficulties with building housing that the proposed change does not address, including rising costs for materials, an increasingly transient and aging workforce, and the economic impact of COVID-19. The National Association of Housing and Redevelopment Officials (NAHRO) stated that Congress has underfunded affordable and public housing, and that because there is a limited amount of funding for such efforts, the number of units built will go down if costs go up. Because of this, the organization recommended that the HUD programs be excluded from the final rule.

Some commenters stated their opposition to the proposed reversion to the three-step process but appeared to misunderstand that the rule does not require, or always result in, the highest wage rate being identified as prevailing. AFP-I4AW, for example, stated that the 30-percent rule would “serve to inflate the wage determination by relying only on the highest wage earners in the locality.” This assumption is not correct. The 30-percent threshold does not distinguish between rates based on whether they are higher or lower. Rather, under the rule, the Department will determine that a wage rate is prevailing if that wage rate is earned by the most workers in a wage survey and if that number is also more than 30 percent of workers in the survey—whether that wage rate is higher or lower than any other wage rate in the survey, and whether it is collectively bargained or not. The Department's review of recent wage surveys suggests that the return to the 30-percent threshold will in some cases result in wage rates that are higher than the currently used average and in other cases lower rates.

See

section V.D.1.ii. This is consistent with the results of the 30-percent threshold when it was last in effect before the 1981-1982 rulemaking.

See

1979 GAO Report, at 53 (noting the data showed that under the 30-percent rule, where a lower hourly rate prevailed, the Department identified the lower rate as the prevailing rate).

In contrast to the commenters that opposed the proposal, many commenters that supported the proposal argued that the rule would not significantly increase project costs or increase inflation. Several of these commenters noted studies regarding the cost effects of prevailing wage regulations in general. For example, the III-FFC noted that the “economic consensus” today is that prevailing wage requirements generally do not raise total construction costs. III-FFC cited a literature review that analyzed the 19 peer-reviewed studies that have been published since 2000 about the impacts of prevailing wage regulations in public construction (together covering more than 22,000 public works projects).

See

Kevin Duncan & Russell Ormiston, “What Does the Research Tell Us About Prevailing Wage Laws,” 44 Lab. Stud. J. 139, 141-42 (2018). A significant majority of those peer-reviewed studies did not find evidence that prevailing wages affected overall construction costs. As III-FFC noted, a key driver of this outcome is that contractors on covered projects will tend to hire higher-skilled workers and utilize more capital equipment.

See

William Blankenau & Steven Cassou, “Industry Differences in the Elasticity of Substitution and Rate of Biased Technological Change between Skilled and Unskilled Labor,” 43 Applied Econ. 3129-42 (2011); Edward Balistreri, Christine McDaniel, & Eina V. Wong, “An Estimation of U.S. Industry-Level Capital-Labor Substitution Elasticities: Support for Cobb-Douglas,” 14 N. Am. J. of Econ. & Fin 343-56 (2003). Other commenters submitted similar research showing that prevailing wages are associated with higher productivity and that labor costs are only a small part of overall project costs in many segments of the construction industry, limiting the impact of any increased wage costs on overall project costs.

See

Frank Manzo & Kevin Duncan, Midwest Econ. Policy Inst., Examination of Minnesota's Prevailing Wage Law: Effects on Costs, Training, and Economic Development 4 (2018); Nooshin Mahalia, Econ. Policy Inst., Prevailing Wages and Government Contracting Costs 3-4 (2008).

Several of these commenters specifically criticized the Department's apparent reliance in 1982 on arguments that the 30-percent rule had an “inflationary effect.” These commenters noted that the concerns about an “inflationary effect” at the time were drawn from the same 1979 GAO report on which the opponents of the proposal now rely.

The Iron Workers, for

example, noted that in 1979 the Department had strongly criticized the GAO report's statistical methods. In 1979, the Department maintained that the GAO's conclusions lacked “statistical validity” because it was methodologically flawed and failed to consider important variables, such as productivity.

See

1979 GAO Report, at 15. However, in its 1982 rulemaking, the Department did not acknowledge other evidence undermining the GAO's conclusions, or the Department's own prior position that the 1979 GAO report could not be relied upon. Another commenter noted that the GAO itself had conceded that its sample size was insufficient for projecting results with statistical validity.

Id.

The GAO issued a report in 1979 urging Congress to repeal the Act because of “inflationary” concerns.

See

Gov't Accountability Office, HRD-79-18, “The Davis Bacon Act Should be Repealed” (1979) (1979 GAO Report). Available at:

https://www.gao.gov/assets/hrd-79-18.pdf.

The report argued that even using only weighted averages for prevailing rates would be inflationary because they could increase the minimum wage paid on contracts and therefore result in wages that were higher than they otherwise would be. The House Subcommittee on Labor Standards reviewed the report during oversight hearings in 1979, but Congress did not amend or repeal the Act, and instead continued to expand its reach.

See, e.g.,

Cranston-Gonzalez National Affordable Housing

Act, Public Law 101-625, Sec. 811(j)(6), 104 Stat. 4329 (1990); Energy Independence and Security Act of 2007, Public Law No, 110-140, Sec. 491(d), 121 Stat. 1651 (2007); American Recovery and Reinvestment Act, Public Law 111-5, Sec. 1606, 123 Stat. 303 (2009); Consolidated Appropriations Act of 2021, Public Law 116-260, Sec. 9006(b), 134 Stat. 1182 (2021).

The commenters supporting the Department's current proposed reversion to the 30-percent rule also noted that, whatever its persuasiveness at the time, the 1979 GAO report cannot be relied on now because of its outdated statistical methods and because of the existence of other, more contemporary, evidence undermining its conclusions. Commenters noted that the three main studies relied on by opponents of the 30-percent threshold, including the GAO report, the Department's 1981-1982 regulatory flexibility analysis, and the Beacon Hill studies, were all based on a “wage differential” calculation methodology that has been discredited by peer-reviewed scholarship published since the 1981-1982 rulemaking.

In a comment, two Professors of Economics argued that “the results of any study that measures the cost of prevailing wages based on [the wage differential method] should be interpreted with extreme caution and is not suitable as a basis of public policy decisions.” Commenters noted that more advanced statistical methods than those used by GAO have since established that in the construction industry, the substitution of lower-wage and lower-skilled workers for higher-paid and higher-skilled workers does not necessarily reduce project costs because the lower productivity of lower-skilled workers can offset incrementally higher wages paid to more-skilled workers.

That is why, they asserted, the preponderance of peer-reviewed studies conclude that prevailing wage laws as a whole have little or no effect on overall project costs.

Given the evidence for prevailing wage laws as a whole, the commenters expressed skepticism that the return to the 30-percent rule would have an effect on project costs.

See

Kevin Duncan & Russell Ormiston, “What Does the Research Tell Us About Prevailing Wage Laws,” 44 Lab. Stud. J. 139, 141-42 (2018). The Beacon Hill Report was not peer-reviewed.

Id.

at 141. The 2022 Beacon Hill Report uses the same methodology as the 2008 Beacon Hill Report.

See

Duncan & Ormiston,

supra

note 61, at 142-48 (collecting peer-reviewed studies).

The Department agrees with those commenters that found the 1979 GAO Report and the Department's 1981-1982 analysis unpersuasive. The Department does not believe that these analyses are reliable or accurate.

For example, the Department's 1981-1982 analysis did not consider labor market forces that could prevent contractors from lowering wage rates in the short run. The analysis also did not attempt to address productivity losses or other costs of setting a lower minimum wage, such as higher turnover and a reduced ability to recruit high-skilled workers. For these reasons, the Department does not believe that the analysis in the 1982 final rule implies that the current proposed reversion to the 30-percent rule would have a significant impact on contract costs. Moreover, even if the Department were to rely on this analysis as an accurate measure of impact, such purported cost savings (adjusted to 2019 dollars) would only amount to approximately two-tenths of a percent of total estimated covered contract costs.

The Department has not attempted to assess the relative accuracy of the $120 million estimate over the decades, which would be challenging given the dynamic nature of the construction industry and the relatively small impact of even $120 million in savings. The Department at the time acknowledged that its estimate had been heavily criticized by commenters and was only a “best guess”—in part because it could not foresee how close a correlation there would be between the wage rates that are actually paid on covered contracts and the wage determinations that set the Davis-Bacon minimum wages. 47 FR 23648.

The two CBO reports from 1983 and 1985 cited in a comment by CEI are not persuasive for the same reason. The 1983 CBO study projected that the elimination of the 30-percent rule would save an average of $112 million per year from 1984 to 1988.

Id.

at 36. That report, however, was based on the Department's own analysis in the 1981-1982 rulemaking,

id.

at xii, which was flawed as previously noted. The 1985 CBO report did not contain an independent analysis and simply cited to the 1983 report.

See

1985 CBO Report, at 16 n.2. Thus, the reports provide no additional helpful evidence and instead suffer from the same analytical problems as the Department's own 1981-1982 study and other simple wage-differential analyses.

The 1983 CBO study acknowledged these issues. It noted that the 1979 GAO study had been questioned because of inadequate sample sizes, the choice of projects covering small volumes of construction, and inappropriate assumptions.

See

1983 CBO Report, at 48. It also noted that “questions have been raised regarding the general approach of translating wage increases directly into cost increases.” Such an approach, the report notes, “may be incorrect . . . to the extent that workers at different wage levels may not be equally productive.”

Id.

at 48-49. The 2018 CBO projection that NAHB cites does not explain its methodology, but it estimates savings from eliminating the entire Davis-Bacon Act as amounting to only 0.8 percentage points in project costs associated with a reduction in wages and benefits.

See supra

note 57,

https://www.cbo.gov/budget-options/54786

After considering the available data, and assuming for the purposes of this discussion that costs are in fact a permissible consideration in defining the term “prevailing wage,” the Department is not persuaded that returning to the 30 percent threshold will cause a meaningful increase in Federal construction costs. Based on the Department's demonstration in the economic analysis of what the prevailing wage would be after applying the 30-percent threshold to a sample of recently published prevailing wage rates, the Department found no clear evidence of a systematic increase in the prevailing wage sufficient to affect prices across the economy. The illustrative analysis in section V.D. shows returning to the 30-percent rule will significantly reduce the reliance on the weighted average method to produce prevailing wage rates. Applying the 30-percent threshold, some prevailing wage determinations may increase and others may decrease, but the magnitude of these changes will, overall, be negligible. Even where wage determinations may increase, the Department is persuaded by recent peer-reviewed research, which generally has not found a significant effect from wage increases related to prevailing wage requirements on the total construction costs of public works projects.

For similar reasons, the Department is not persuaded that the reversion to the 30-percent threshold would have any impact on national inflation rates. Several commenters, including CEI and certain members of Congress, stated that the Department's proposal is ill-timed because of the current levels of

economy-wide inflation and the risks of a wage-price spiral. Returning to the 30-percent rule, CEI claimed, “would likely contribute to the pressures” that could create such a spiral. Although CEI referenced the 1983 CBO Report to support its argument that the 30-percent threshold would increase construction costs, CEI did not note the conclusion in that study that the DBA as a whole “seems to have no measurable effect on the overall rate of inflation.” 1983 CBO Report, at xii, 30-31.

One individual commenter asserted that the Department should be required to consider not only whether the 30-percent rule can alone cause inflation, but also whether the proposal, in combination with other regulatory and spending measures, would have an effect on inflation and what that effect would be. The commenter stated that the infusion of Federal infrastructure spending from the Infrastructure Investment and Jobs Act (IIJA), Public Law 117-58, will likely lead to substantial compensation premiums for construction workers. The commenter stated that such wage increases would occur “because a sudden increase in federal infrastructure spending does not necessarily lead to a commensurate increase in construction sector employment.”

The Department disagrees that this rule will substantially impact inflation. As noted, the Department's illustrative analysis in section V.D. suggests that the reimplementation of the 30-percent threshold will result in some prevailing wage determinations increasing and others decreasing, but the magnitude of these changes will, overall, be negligible. In addition, even if this rule leads to an increase in some required prevailing wage rates, it will not have an equal impact on actual wages paid to workers on DBRA-covered contracts, because some workers may already be earning above the new prevailing wage rate.

If wages for potentially affected workers were to increase, the Department does not believe that it would lead to inflation. Recent research shows that wage increases, particularly at the lower end of the distribution, do not cause significant economy-wide price increases.

For example, a 2015 Federal Reserve Board study found little evidence that changes in labor costs have had a material effect on price inflation in recent years.

Even in the recent period of increased inflation, there was little evidence that the inflation was caused by increases in wages. A study of producer price inflation and hourly earnings from December 2020 to November 2021 found that inflation and wage growth were uncorrelated across industries.

Additionally, as two Professors of Economics commented, “since prevailing wages are not associated with increased construction costs, there is no reason to assume that the policy causes inflation in the macroeconomy.”

See, e.g.,

J.P. Morgan, “Why Higher Wages Don't Always Lead to Inflation” (Feb. 7, 2018), available at:

https://www.jpmorgan.com/commercial-banking/insights/higher-wages-inflation

; Daniel MacDonald & Eric Nilsson, “The Effects of Increasing the Minimum Wage on Prices: Analyzing the Incidence of Policy Design and Context,” Upjohn Institute working paper; 16-260 (June 2016), available at

https://research.upjohn.org/up_workingpapers/260/

; Nguyen Viet Cuong, “Do Minimum Wage Increases Cause Inflation? Evidence from Vietnam,” ASEAN Economic Bulletin Vol. 28, No. 3 (2011), pp. 337-59, available at:

https://www.jstor.org/stable/41445397

; Magnus Jonsson & Stefan Palmqvist, “Do Higher Wages Cause Inflation?,” Sveriges Riksbank Working Paper Series 159 (Apr. 2004), available at:

http://archive.riksbank.se/Upload/WorkingPapers/WP_159.pdf

; Kenneth M. Emery & Chih-Ping Chang, “Do Wages Help Predict Inflation?,” Federal Reserve Bank of Dallas, Economic Review First Quarter 1996 (1996), available at:

https://www.dallasfed.org/~/media/documents/research/er/1996/er9601a.pdf

Ekaterina V. Peneva & Jeremy B. Rudd, “The Passthrough of Labor Costs to Price Inflation,” Federal Reserve Board (2015), available at:

https://www.federalreserve.gov/econres/feds/the-passthrough-of-labor-costs-to-price-inflation.htm.

Josh Bivens, “U.S. Workers Have Already Been Disempowered in the Name of Fighting Inflation,” Figure A, Economic Policy Institute (Jan. 2022), available at:

https://www.epi.org/blog/u-s-workers-have-already-been-disempowered-in-the-name-of-fighting-inflation-policymakers-should-not-make-it-even-worse-by-raising-interest-rates-too-aggressively/.

More importantly, DBRA-covered contracts make up a small share of overall economic output. Because federally-funded construction only makes up approximately 13 percent of total construction output and the number of potentially affected workers (1.2 million) is less than 1 percent of the total workforce, the Department does not believe that any wage increase associated with this rule would significantly increase prices or have any appreciable effect on the macroeconomy.

Federally funded construction as a share of total construction output can be calculated from the data in Table 3 ($216,700,000,000 ÷ $1,667,000,000,000 = 0.13). The estimate of 1.2 million potentially affected workers is calculated in section V.B.2.

In sum, the factual conclusions about “inflationary effects” underlying the 1982 elimination of the 30-percent rule are no longer supportable because they have been discounted over the past 40 years by more sophisticated analytical tools. Furthermore, the available evidence does not suggest that concerns about the 30-percent threshold increasing project costs or national inflation rates are justified.

The Department also considered the comments that express concern about whether the 30-percent threshold may affect certain sectors or areas, and the residential construction industry in particular, differently than the national economy as a whole. As they are for other types of construction, respectively, prevailing wage rates for DBRA-covered residential construction are based on WHD wage surveys of residential construction projects. Residential construction can be distinguished from other construction types in several important ways: it tends to be less capital- and skill- intensive and thus generally has fewer barriers to entry for firms as well as for workers, projects tend to be of smaller and shorter duration, workers tend to move more often between firms, and firms tend to provide less training.

Wages also tend to be lower in residential construction than in nonresidential construction types, and unionization rates have historically been lower. Because of lower unionization rates in the residential construction industry, where the methodology for determining prevailing rates is based on the mode (whether majority or 30-percent threshold), the rates that prevail are more likely to come from non-union wage rates than from higher, collectively bargained rates. As a result, in comparison to other construction types, it is less likely—not more likely—that the 30-percent threshold will result in increases in prevailing wage rates on residential construction projects. However, in the more limited circumstances in which residential construction rates may change from averages to rates based on CBAs, the increases in wage rates could be larger given the generally lower wage floors in the industry.

See

Russell Ormiston et al., “Rebuilding Residential Construction,” in Creating Good Jobs: An Industry-Based Strategy 75, 78-79 (Paul Osterman ed., 2020).

Although the transfer analysis presented in Section V.D.1 is simply illustrative and may not be representative of the impact of this rule, the results of this analysis reflect that only 5 percent of the residential fringe benefit rates analyzed were affected by the reversion to the 30-percent threshold, compared to 14 percent of building fringe rates, 19 percent of heavy fringe rates and 23 percent of highway fringe rates. In those limited circumstances where residential fringe rates were affected, however, they tended to increase more significantly given their largely nonunion baseline.

Moreover, even if implementation of the proposal were to lead in some areas to increased wages, and even assuming those increased wages resulted in increased project costs for federally financed residential construction, the

effects on overall housing prices or rents would not be significant. DBRA-covered construction makes up only a very small percentage of the total new construction in the residential construction market—only 1 percent as of July 2022.

And, annual new residential construction itself tends to be less than 1 percent of all available residential units.

Among the residential construction covered by the DBRA, many projects would be unaffected by the proposed reversion to the 30-percent threshold. The Department's illustrative analysis suggests that the proposal would only affect the methodology for approximately one-third of new wage determinations, and of those, some would result in decreases in the required wage rate, not an increase.

See

section V.D.1.ii.

The most reasonable conclusion is that any limited potential increase in some construction costs for such a small percentage of the residential market would not affect housing prices or rents generally.

According to the Census Bureau, the Seasonally Adjusted Annual Value of Private Residential Construction Put in Place, as of July 2022, was $920.4 billion; public residential construction was $9.3 billion.

https://www.census.gov/construction/c30/c30index.html

See

U.S. Census Bureau, National and State Housing Unit Estimates: 2010 to 2019,

https://www.census.gov/data/tables/time-series/demo/popest/2010s-total-housing-units.html

There are additional reasons why increasing labor costs do not have a one-to-one correlation with housing and rent prices. In recent decades, housing prices have significantly outpaced real construction costs.

See

Joseph Gyourko & Raven Molloy, “Regulation and Housing Supply,” (Nat'l Bureau of Econ. Rsch., Working Paper No. 20536, 2014),

https://www.nber.org/system/files/working_papers/w20536/w20536.pdf

. Gyourko and Molloy conclude that, as a general matter, labor and material costs do not appear to act as a major constraint on residential development, in comparison to land-use policy constraints.

In addition, the reversion to the 30-percent threshold will not result in any wage increases in the short-term. Any effect on wage increase will only occur after wage new residential construction-type surveys are initiated and completed, and then wage determinations based on those surveys are incorporated into new construction contracts.

The Department also considered the concerns commenters raised about the construction of publicly funded affordable housing in particular. In a comment, two Professors of Economics said that three studies have found that the application of prevailing wage laws in general may be correlated with increased project costs for affordable housing projects.

But, for two reasons, these studies are of limited value for forecasting the effects of reversion to the 30-percent rule. First, as noted, the Department's illustrative analysis of the effects of the 30-percent threshold, which included residential construction survey data, does not show a systematic increase in prevailing wage rates. Second, the peer-reviewed studies showing potential increased project costs on affordable housing projects do not compare different prevailing wage methodologies, but instead compare whether projects are either covered or not covered at all by prevailing wage requirements. Where studies compare the existence of prevailing wage requirements at all (as opposed to a simple change in wage determination methodologies), other factors can explain project cost increases.

See also

Duncan & Ormiston,

supra

n. 61, at 142-48 (discussing peer-reviewed studies).

For example, cost differences may be attributable in part to reductions in independent-contractor misclassification, failure to pay overtime, and other basic wage violations that are disincentivized because of the prevailing-wage requirement to submit certified payroll.

Id.

at 146.

The Department also considered the comments regarding the potential effects of economic conditions that may result from increased infrastructure spending. While it is true that increases in construction spending can lead to increases in construction wage rates in the short run,

this potential does not suggest the Department's proposal is unwarranted. Under the 30-percent threshold, as under the current majority rule or any other measure of prevailing wages, wage determinations will and should generally reflect increases in wage rates that result from separate policy decisions by Federal, State, or Local governments, or other macro-economic phenomena. The commenters did not suggest, and the Department did not identify, any specific mechanism through which the 30-percent threshold would interact with construction spending increases in a way that would materially affect the results of the Department's illustrative analyses or suggest outcomes other than those supported by the peer-reviewed literature. Finally, the prevailing wage methodology in this rule is not a short-term policy; it is intended to apply during timeframes when public infrastructure spending is lower, as well as those when it is higher, and during all phases of the construction industry business cycle.

One commenter suggested that increased infrastructure spending could lead to an increase in demand for construction workers, and that the supply of skilled workers might not be commensurate in the short term, which could lead to an increase in wage rates.

Finally, the Department disagrees with NAHB that the proposal should be withdrawn because, among other reasons, the proposal does not address certain challenges in the residential building industry, including “an increasingly transient and aging workforce, increased building costs resulting from supply shortages, and the economic impact of COVID-19, among other things.” NAHB explains, in addition, that the residential construction industry has been “suffering from a skilled labor shortage for many years.” The Department agrees with NAHB that these topics are important for policymakers to consider. NAHB does not explain why the methodology for determining the prevailing wage under the DBRA is relevant to addressing these challenges, or why a methodology other than the Department's proposed reversion to the three-step process would be more beneficial. However, to the extent that the 30-percent threshold could increase wage rates in some areas, as NAHB also asserts, such an outcome would be beneficial to the industry by attracting more workers to the construction labor market and allowing required prevailing wages to more often support the maintenance of apprenticeship and training costs that will contribute to the expansion of the skilled workforce.

In addition to all of these factual arguments about whether costs or inflation may increase, however, several unions and contractor associations argued that the Department should not be permitted as a legal matter to consider contract costs or other similar effects of any wage increases when it determines the proper prevailing wage methodology. The United Brotherhood of Carpenters and Joiners of America (UBC) and NABTU argued that the Department's apparent goal in 1981-1982 of reducing construction costs was not consistent with the purpose of the Act. NABTU stated that such a reliance on cost considerations was arbitrary and capricious under the Supreme Court's decision in

Motor Vehicle Manufacturers Ass'n of the United States

State Farm Mutual Automobile Insurance Co.,

463 U.S. 29, 43 (1983) (

State Farm

), because it relied on a factor (cost) that Congress had not intended to be considered. To the contrary, commenters noted, statements in the legislative history suggest that Congress's “chief concern” was “to maintain the wages of our workers and to increase them wherever possible.” 74 Cong. Rec. 6513 (1931) (remarks of Rep. Mead);

See also

1979 GAO Report, at 52 (describing the difference between CBA pay scales and non-union contractor pay practices).

On the other hand, commenters supporting a return to the 30-percent rule criticized the reasoning in 1982 that the 30-percent rule provided “undue weight” to collectively bargained rates. These commenters argued that this reasoning was a symptom of anti-union bias and had no basis in the statute. The Iron Workers quoted the 1962 congressional testimony of Solicitor of Labor Charles Donahue regarding the interface between the rule and union rates. As Solicitor Donahue pointed out, the 30-percent rule did not uniformly lead to the identification of union rates as prevailing, but, in any case, the question of whether union or non-union contractors are disadvantaged by the Department's prevailing wage determinations is not something that the Department should be properly taking into consideration in making its wage determinations.

In a related comment, two Professors of Economics noted that the potential for union rates being identified as the prevailing rate does not necessarily mean that project costs will increase. The comment cited several peer-reviewed studies that found no statistically significant cost difference between projects built with prevailing rates based on union rates and projects that were not.

Administration of the Davis Bacon Act: Hearings before the Spec. Subcomm. of Lab. of the H. Comm. on Educ. & Lab.,

87th Cong. 819-20 (1962) (statement and submission of Charles Donahue, Solicitor of Labor).

See

Lamek Onsarigo et al., “The Effect of Prevailing Wages on Building Costs, Bid Competition, and Bidder Behaviour: Evidence from Ohio School Construction,” 38 Constr. Mgmt. & Econ. 917 (2020); Kevin Duncan & Jeffrey Waddoups, “Unintended Consequences of Nevada's Ninety-Percent Prevailing Wage Rule,” 45 Lab. Stud. J. 166 (2020); Jaewhan Kim et al., “The Effect of Prevailing Wage Regulations on Contractor Bid Participation and Behavior: A Comparison of Palo Alto, California with Four Nearby Prevailing Wage Municipalities,” 51 Indus. Rels. 874 (2012).

The Department is no longer persuaded that the 30-percen

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.