How Davis-Bacon Rates Are Determined

A Davis-Bacon wage determination is not a survey result printed as found. It is the product of a definition, a data-collection program and a set of rules about which data count, all in 29 CFR part 1, which the Department of Labor rewrote in a final rule published August 23, 2023 and effective October 23, 2023. This page describes the method as the regulations now state it. The Davis-Bacon basics page covers coverage and the contractor’s duties; how to look up a wage determination covers reading the result on SAM.gov.

The definition: majority, then 30 percent, then average

Section 1.2 defines the prevailing wage for a classification on similar projects in an area during the survey period in three steps. If one wage is paid to a majority, more than 50 percent, of the workers in the classification, that wage prevails. If no single wage reaches a majority, the wage paid to the greatest number prevails, provided that number is at least 30 percent of those employed. Only if no wage reaches 30 percent is the prevailing wage a weighted average of the wages paid, weighted by the number employed at each. The 2023 rule restored the 30 percent step, which the Department had used from 1935 until a 1982 rule removed it in favor of a majority-or-average test. The rule’s preamble notes that the most frequent rate is often a collectively bargained one, so a determination can list union-scale rates for some classifications and survey averages for others. Fringe benefit data are collected in the same surveys and listed separately on the determination.

Where the data come from

Section 1.3 directs the Administrator of the Wage and Hour Division to run a continuing program of collecting wage information and lists what may be considered: statements of wages paid on projects (with the contractors, locations, costs, dates, types of construction, and the number of workers in each classification at each rate), signed collective bargaining agreements, rates set by state and local officials under their own prevailing wage laws, data agencies submit through conformance actions, state highway department information for federal-aid highway projects, and anything else pertinent. In practice the data come from the Davis-Bacon wage survey, described in Fact Sheet 81: the Wage and Hour Division sets the survey’s geography, construction type and time frame, identifies active projects from commercial construction reports, notifies the contractors and interested parties, holds briefings, and collects wage data on form WD-10 from private as well as public projects. Data from federally funded projects are excluded from building and residential surveys unless there would otherwise be too little, but are used for heavy and highway surveys (1.3(d)).

Under 1.7, the area is normally the county. When a county yields too little current data, the Division may look to surrounding counties, then to comparable counties or groups of counties in the state, then statewide, and only then to projects completed more than a year before the survey. Variable rates that a collective bargaining agreement or written policy treats as functionally equivalent may be counted as one rate (1.3(e)).

Adopting state and local rates

Two provisions added in 2023 let the Division publish rates it did not survey itself. Under 1.3(f) it may list a classification that is regularly requested through conformances but lacks survey data, with a rate that bears a reasonable relationship to the others on the determination. Under 1.3(g) and (h) it may adopt a rate determined by a state or local government for its own public works, with or without modification, when the state’s process is open to all interested parties, separates the basic hourly rate from fringe benefits, classifies workers in a way the construction industry recognizes, and uses criteria substantially similar to the federal ones. The state prevailing wage laws page shows which states have such laws, as the Department of Labor tabulates them.

Keeping rates current between surveys

A survey rate stands until the next survey, and the 2023 preamble says such rates had fallen out of date. Section 1.6(c)(1), as revised in 2023, allows the Division to adjust non-collectively-bargained rates on general wage determinations using the Bureau of Labor Statistics’ Employment Cost Index, no more often than every three years and no sooner than three years after a rate’s publication. Rates taken from collective bargaining agreements are updated when the agreement changes. A revised determination applies to a new contract according to the timing rules in 1.6(c)(2), and to certain long-running contracts at their anniversaries or when options are exercised.

What the survey is not

The Davis-Bacon survey is a count of workers at each wage on construction projects of one type in one area, not a sample survey of employers. It is therefore not the same measure as the Bureau of Labor Statistics’ Occupational Employment and Wage Statistics survey that supplies the construction trade wages on this site’s occupation pages and state profiles: OEWS reports a percentile distribution across all projects and employers in a state or metropolitan area, where a determination reports the single wage that prevails under the definition above in a county for one construction type. A reader who wants to know what electricians earn in a state should use the OEWS table; a contractor bidding a covered project uses the determination, and this site mirrors none.