Prevailing Wage Laws: Public Interest
or Special Interest Legislation?
George C. Leef
People of the same trade seldom meet together, even for
merriment and diversion, but the conversation ends in a conspiracy
against the public, or in some contrivance to raise
prices. But though the law cannot hinder people of the same
trade from sometimes assembling together, it ought to do
nothing to facilitate such assemblies, much less to render
—Adam Smith (1776)
The public policy of the United States is broadly in favor of competition. Our antitrust laws are premised on the idea that in the absence of such legislation private interests would seek to create monopolies, fix prices, restrain trade, and stifle competition. Moreover, the federal government, as well as the states and municipalities, has laws mandating competitive bidding on government contracts to guard the public against “sweetheart deals” that squander tax dollars. Open competition, in fact, is usually the undoing of those conspiracies against the public that Adam Smith saw as so
Cato Journal, Vol. 30, No.1 (Winter 2010). Copyright © Cato Institute. All rights reserved.
George C. Leef is Director of Research at the John W. Pope Center for Higher
Education Policy and author of Free Choice for Workers: A History of the Right to
Work Movement. 20951_CATOpages.qxd 1/20/10 1:00 PM Page 137
One glaring exception to the general rule favoring competition is “prevailing wage” laws. Those laws mandate that on government construction projects, the labor component will not be subject to competitive bidding; rather, the wages paid to the various classes of
construction labor are set by government officials at rates determined to prevail in the job site’s locality—typically, prevailing union wages. Labor wages and benefits are thus removed from competition by operation of law. Are prevailing wage laws a reasonable deviation from our general rule in favor of competition? Do they actually reflect the public interest, with benefits that outweigh the costs? Or are they merely an instance of rent seeking by a politically potent interest group, using its influence to use the law to enforce a price fixing scheme? This article concludes that the latter of those questions captures the truth. Prevailing wage laws favor special interests by concentrating benefits
and dispersing costs. They ought to be repealed.
The Davis-Bacon Act
Enacted in 1931, the Davis-Bacon Act (40 U.S. Code § 276) is the federal prevailing wage statute, requiring that contractors on federal construction projects must pay at least the wages and benefits found to be “prevailing for the corresponding classes of labors and mechanics” employed. The statutory language does not specify how this determination is to be made, leaving that to the Secretary of Labor. The “prevailing” rates do not necessarily have to be those of construction unions, but often they are. The Labor Department’s method for determining “prevailing” rates has been widely criticized for giving far too little weight to lower, nonunion rates that many construction workers in the area receive. Glassman et al. (2008: 38), for example, conclude that the wage survey method in use “employs unrepresentative survey and measurement methods that produce wages estimates that are biased upward.” Precisely how the government sets the rates, however, is much less important than the fact that contractors are forbidden to pay workers less even if their workers would accept a lower rate. The devil here is not in the details, but in the concept. The genesis of the Davis-Bacon Act (named for its sponsors,
Pennsylvania Senator James Davis and New York Representative Robert Bacon, both Republicans) was the construction of a Veterans Bureau Hospital on Long Island in 1927. The contractor who won the bid was from Alabama and employed many black workers. Representative Bacon was bothered that construction workers in his district had lost out on the job and he introduced a bill that would prevent the use of migrant labor on federal projects. The bill did not pass in that session, but he kept reintroducing it. In 1930, with the nation’s economic depression deepening rapidly, Bacon’s bill found strong support in the House and Senate. The hearings and debate on the legislation revealed some ugly racial overtones with comments on how “cheap colored labor” was driving down wages of white workers (Bernstein: 1993). Proponents played on racial animosity to increase support, but the skin color of the construction workers competing with local firms was of little importance; the very fact of competition was the problem. The bill was passed and President Herbert Hoover signed it into law on March 31, 1931. His motives had nothing to do with the race of the workers competing for jobs. Rather, the bill appealed to him because it was consistent with his theory that high wages caused prosperity and conversely that competition led to wage reductions that were exacerbating the depression.
State and Local Prevailing Wage Laws
Today 31 states and the District of Columbia enforce prevailing wage laws. Most were enacted during the New Deal but some predate it (e.g., New York’s law, which was passed in 1897), and some were enacted after World War II, with Minnesota’s being the most recent (1973). Nine states have never had such laws (Georgia, Iowa, Mississippi, North Carolina, North Dakota, South Carolina, South Dakota, Vermont, and Virginia), and in nine others they have been repealed (Florida, Alabama, Utah, Arizona, Colorado, Idaho, New Hampshire, Kansas, and Louisiana). In Oklahoma, the prevailing wage law was held to violate the state’s constitution in 1995 on the grounds that it impermissibly delegated authority to federal officials (Oklahoma City v. State ex rel. Department of Labor, 918 P.2d 26) and has not been enforced since. Prior to that litigation, the Oklahoma Commissioner of Labor had uncovered widespread fraud in the submission of data on which prevailing wage determinations were made (Cook 2005: 219). In addition, many cities, particularly those where organized labor is politically strong, have prevailing wage laws.
The state prevailing wage laws vary greatly. In states where organized labor is relatively weak (e.g., Nebraska), the wage determinations are close to the competitive market rates. Conversely, in states where organized labor is powerful (e.g., New Jersey), the prevailing
wage law mandates rates that are close to or even exactly the same as construction union rates. There would appear to be a conflict between state and local prevailing wage laws, which suppress competition in bidding on construction work, and the pro-competition federal antitrust statutes. That conflict, however, is readily resolved. Only private conduct is subject to the antitrust laws. For example, price fixing by attorneys through a bar association was held to violate the Sherman Antitrust Act in Goldfarb v. Virginia State Bar (421 U.S. 773 ) since the minimum fee schedules for legal work at issue were not mandated by state law. If, however, a unit of government acts to suppress price competition, that is beyond the reach of the Sherman Act, as the Supreme Court held in Parker v. Brown, (317 U.S. 341 ). Prevailing wage statutes are therefore legally permissible and organized labor has used them to diminish competition.
The Effects of Prevailing Wage Laws: Higher Costs and
The most salient effect of prevailing wage laws is to raise the cost of public construction. They do so in several ways. First, by preventing competitive bidding on the labor costs of public projects, a greater outlay of tax dollars is required to pay the construction workers employed than would otherwise be necessary. Second, prevailing wage laws often interfere with efficient labor utilization because their enforcement mandates adherence to union work rules. Third, they impose additional compliance costs, including litigation, on contractors. And fourth, prevailing wage laws require additional administrative costs in determining what wage rates “prevail” and also adjudication and enforcement costs. Numerous studies have examined the impact of prevailing wage laws and found that they add significantly to the cost of government construction projects; conversely, when such laws are suspended or repealed, costs fall. In Michigan, a state with a particularly severe and expansive prevailing wage law, a federal district court ruled in 1994 that the law was in conflict with the federal Employee Retirement Income Security Act (Associated Builders and Contractors v. Perry, 869 F. Supp. 1239). That decision was overturned by the Sixth Circuit Court of Appeals in 1997 (Associated Builders and Contractors v. Perry, 115 F.3d 386), but the 30-month time period while the law was suspended allowed for a an excellent “apples to apples” comparison. Economist Richard Vedder (1999) studied the effects of having no prevailing wage law for that period. He observed that on various state construction projects that were bid under open competition after the suspension of the law, nonunion contractors substantially underbid union contractors. In one school renovation project, for example, the winning non-prevailing wage bid was 16 percent less than the lowest union bid (Vedder 1999: 14). Vedder concludes that in 1995 the state saved approximately $275 million in construction outlays because competitive bidding prevailed rather than the “prevailing
wage” law. A subsequent study in Michigan by Paul Kersey (2007) examined the difference between the “prevailing rates”—which in Michigan must be union rates—and the wages of workers in the same lines of work as determined by wage surveys conducted by the U.S. Bureau of Labor Statistics (BLS). He found that on average, the rates that
were mandated on state construction projects were 39 percent
higher than the median wages in the construction industry (Kersey
2007: 9). For example, the median hourly rate for carpenters in
Wayne County (the Detroit area) in 2005 according to the BLS,
adjusted to include fringe benefits, was $26.33, but the prevailing
rate under Michigan law was $41.37 (Kersey 2007: 24). If the state
had allowed competitive bidding, the costs for virtually every class of
construction labor would have been significantly lower. Kersey
(2007: 18) also estimates that if Michigan municipalities had not
imposed prevailing wage requirements, they would have saved $16
million that year.
Ohio is another state with a strong prevailing wage statute and, in
1997, the legislature voted to exempt school construction and renovation
work from it. Like Michigan, Ohio bases its prevailing wages
exclusively on union collective bargaining contracts, so one would expect that the suspension would have resulted in substantial savings.
In 2002, the state legislature ordered the Ohio Legislative Service
Commission to study the impact of the suspension on school construction
costs. The commission’s report concluded that by allowing
competitive bidding Ohio schools had saved $487.9 million, which
was 10.7 percent of construction spending (Ohio Legislative Service
Committee Senate Bill 102 Report: 22–25).
California went in the opposite direction in 2001, extending the
coverage of its prevailing wage law to include the construction of
state subsidized housing. In a study of 205 low-income housing projects
in the state, Dunn, Quigley, and Rosenthal (2005) concluded
that the imposition of prevailing wage requirements raised construction
costs by at least 9 percent and as much as 37 percent. Under the
authors’ mid-range estimate, imposing the prevailing wage led to a
decrease in production of low-income housing units on the order of
3,100 units per year. The authors point to the conflict between state
law and federal policy: “In this way, state regulation of construction
wages conflicts with the federal goal of increasing access to new
housing for California’s low-income households” (Dunn, Quigley,
and Rosenthal 2005: 155).
A recent analysis done for the Citizens Housing & Planning
Council in New York City on the prospective application of prevailing
wage to low-income housing construction similarly found that
extending the prevailing wage mandate would have a strongly
adverse impact on the production of housing units and also on the
largely minority labor force that has been engaged in that work.
Prompted by bills in both the New York State Legislature and the
New York City Council that would expand the reach of prevailing
wage mandates to low-income housing construction and renovation,
the study concludes that adding a prevailing wage requirement
would significantly raise costs without any offsetting benefit
(Roistacher, Perine, and Shultz 2008).
It notes the salient fact that in New York City there is a very wide
differential between union and nonunion construction wages. When
fringe benefits are included, the median union wage was 74 percent
higher than the median nonunion wage. Moreover, the wage differential
does not completely capture the added costs because there are
other less easily quantified cost increases involved. Finally, taking
account of “the cost of compliance, job reclassification and workforce
composition would further increase the gap between prevailing and
non-prevailing wage construction projects” (Roistacher, Perine, and
Shultz 2008: 4). Including all cost elements, the authors conclude
that if state or city government were to impose a prevailing wage
requirement, costs per unit would increase by 25 percent, driving the
cost of a typical unit from $250,000 to $312,000. Amortizing the
additional expense would call for higher rent—approximately $400
per month, leading the authors to comment, “To the extent that subsidized
housing programs are designed to reach households which
cannot afford market rate housing, an additional $400 per month is
significant” (Roistacher, Perine, and Shultz 2008: 7).
The authors of that study also considered a social cost of prevailing
wages—namely, the impact on minority firms and workers. They
found that there is continuing although subtle discrimination in favor
of white construction workers in New York unions and that the imposition
of prevailing wage requirements is a large hindrance to minority-owned
firms. “Few such firms,” they write, “have the back-office
capacity to comply with the complex reporting and oversight requirements
of prevailing wages” (Roistacher, Perine, and Shultz 2008: 15).
Consequently, they conclude, if New York were to mandate the payment
of prevailing wages on low-income housing projects, many
black and Latino workers who currently are employed on such projects
would be thrown out of work.
Studies on the impact of the Davis-Bacon Act on federal construction
have consistently found that it adds significantly to costs. Gould
and Bittlingmayer (1980) calculated that Davis-Bacon added
between $0.5 billion and $1 billion per year to the cost of federal construction—in
current dollars, between $1.3 billion and $2.6 billion.
The authors broke the costs down as follows: use of higher-paid
union rather than nonunion labor, $228 million to $513 million; the
fact that Davis-Bacon prevailing wage determinations often lead to
paying unionized workers even more than they usually receive on
non-Davis-Bacon projects, $69 million to $284 million; administrative
costs contractors must bear to comply with the statute, $190 million;
and the cost to taxpayers of supporting Department of Labor
personnel assigned to prevailing wage determination and enforcement,
$12 million (Gould and Bittlingmayer 1980: 68).
In a study that closely examined the impact of Davis-Bacon on
construction in rural areas, Fraundorf, Farrell, and Mason (1984)
estimated that the law increased costs by 26.1 percent. That study
compared federal construction projects subject to Davis-Bacon with
similar private projects that were done with fully competitive bidding,
interviewing the contractors to obtain a detailed picture of their
respective costs and practices. The authors explain that their findings
show Davis-Bacon to have a greater impact on costs than other studies,
because they looked not only at wage differentials but also at how
the prevailing wage mandate affects labor efficiency. They note that
the Department of Labor usually has no prevailing wage classification
for “general building mechanics” and explain the significance of
that, writing, “Contractors must then pay as if the appropriate skilled
craftsman, not the lower-wage semi-skilled worker, had done each
task. Also, flexibility is reduced as contractors avoid extensive record
keeping by confining general mechanics to one type of work. The
result is that total labor costs increase by more than the wage rates
for any one craft” (Fraundorf, Farrell, and Mason 1984: 145).
More recently, in 2000 the Congressional Budget Office estimated
the savings in federal outlays that would be occur if DavisBacon
were repealed at $10.5 billion from 2001 to 2010, stating,
“Repealing the Davis-Bacon Act would allow the federal government
to spend less on construction. In addition, it would probably increase
the opportunities for employment that federal projects would offer
to less skilled workers” (Congressional Budget Office 2000: 283).
The most recent study to calculate the impact of Davis-Bacon’s
elimination of wage competition (Glassman et al. 2008: 50) concludes
that costs on projects subject to it are increased by 9.9 percent.
In sum, preventing wage competition on federal construction
projects does exactly what one would expect—it increases the cost of
Assessing the impact of prevailing wage mandates on labor efficiency
is more difficult than measuring the difference between market
and “prevailing” rates, but nevertheless important. That point
was stressed in the study by Fraundorf, Farrell, and Mason (1984).
The federal law and the more restrictive state statutes impose strict
job classifications that are integral to construction unions. Nonunion
contractors generally are much more flexible in their use of labor,
employing more helpers and having workers do a wider array of jobs
on the site than unions permit. As Philip Abrams, a former president
of the nonunion construction trade association Associated Builders
and Contractors, testified before a U.S. Senate committee,
The problem with restrictive work practices is difficult to
quantify, but in general, if union wages have been determined
and published, any such work practices that are part of union
collective bargaining agreements are then made a requirement
in the name of “area practices” and are the responsibility of the
contractor. The requirement, basically, is that skilled people do
unskilled work and arbitrary rules like “tools of the trade”
determine the hourly wages that a working person receives.
For instance, in one particular project, the fact that laborers
were using hammers caused the Department of Labor and the
Department of Housing and Urban Development to conclude
that the workers should be paid as carpenters, even though
they were in fact doing laborers’ work.
The sum and substance is that you lower productivity on construction
sites by enforcing featherbedding union work practices,
not by allowing “helper” job categories, and by refusing,
for example, to let unskilled people nail up insulation or carry
pieces or steel or pull wire or unload plumbing fixtures.
The worst and most pernicious part of these Department of
Labor requirements is that they are unwritten. You are simply
expected to comply with these area practices which are
enforced by the contracting officer and DOL arbitrarily and
without any previous notice to the contractor [Testimony
before the Subcommittee on Housing and Urban Affairs,
Senate Committee on Banking, May 2, 1979: 323-4].
A crucial difference between union and nonunion construction is
the considerable use the latter makes of “helpers” on the worksite—
that is, unskilled or semi-skilled workers who assist the skilled tradesmen
by doing much of the “grunt work.” In the course of doing so,
the helper often learns the trade and thereby becomes a potential
competitor. As Herbert Northrup (1992: 422) explains, “By accepting
less remuneration than union journeymen, helpers restrain union
wage acceleration. The determination of construction unions to
eliminate such competition . . . has been at the heart of negative
union policy towards helpers for almost a century.” By hindering the
use of helpers, Davis-Bacon and the more pro-union state laws have
been a large part of that policy by minimizing one of the main efficiency
advantages of nonunion contractors.
Prevailing wage laws are meant to eliminate competition in one of
the main elements in construction—namely, the cost of employing
the necessary labor. There is no doubt that they accomplish that and,
in doing so, they add significantly to public construction costs.
Public Interest Arguments for Prevailing Wage Laws
By the 1970s, prevailing wage laws were starting to come under
attack as special interest legislation that harmed taxpayers and lowercost
competitors. Advocates of the laws responded by advancing several
public interest rationales for them: improved safety, higher
quality work, less “harmful competition,” increased earnings and
taxes, and improved worker training.
In a working paper published by the Department of Economics at
the University of Utah, which was funded by organized labor, Phillips
et al. (1995) argued that the repeal of prevailing wage laws would lead
to increases in injuries and fatalities in construction work. The authors
contended that in the absence of prevailing wage laws, construction
firms would tend to cut corners and employ more inexperienced
workers who would be prone to accidents. If all prevailing wage laws
were repealed, they estimated that there would be an additional
76,000 construction workplace injuries annually, with approximately
30,000 of them sufficiently serious to require time off work for the
individual to recover, and between 130 and 150 fatalities.
The argument that eliminating wage competition in bidding on
construction projects is an efficient way of lowering accident rates
has come under severe criticism. Responding to the Utah study,
Armand Thieblot (1996: 17) writes,
The theoretical construct of this claim is derived from two
premises: the rate of injuries decreases substantially as
length of service increases; and large, experienced employers
in construction have injury rates that are 80 percent
lower than small-to-medium size contractors. The first of
these claims is grossly misleading and the second is demonstrably
wrong. Thieblot calls the Utah study “complex, but flawed,” and shows
that its projections regarding workplace accidents—based on only
two of the many relevant occupational categories—are refuted by
actual data from the Occupational Safety and Health Administration.
OSHA data show that accident rates did not increase in states that
repealed their prevailing wage laws and also that accident rates are in
fact higher in prevailing wage states than in states that never had
such a law (Thieblot 1996:18–23).
The recent New York study on low-income housing also examined
the safety issue and concluded that prevailing wage requirements
do not affect the incidence of fatalities. It found that “in New
York City, fatal construction accidents occur among union and
nonunion workers (both residential and non-residential) in roughly
the same proportion as there are union and nonunion workers in the
city’s building construction labor force” (Roistacher, Perine, and
Schultz 2008: 10). Imposing a prevailing wage mandate on lowincome
housing construction would not, therefore, improve safety
in that sector.
It is not surprising that construction accident rates are scarcely
affected by prevailing wage laws, since in the construction industry,
union and nonunion firms are subject to the same safety regulations
and have the same incentives to minimize their worker’s compensation
insurance rates by maintaining a safe workplace. If construction
safety (governmental as well as private sector) were a matter of concern,
it would make far more sense to address this directly rather
than through the roundabout and ineffective method of imposing a
price floor on construction wage rates. Contractors with particularly
bad safety records might, for example, be precluded from bidding.
Higher-Quality Work Advocates of prevailing wage laws contend that the use of higher wage labor on projects is partially compensated for by higher-quality
work. For example, a paper published by the Fiscal Policy Institute
in New York asserts, “The prevailing wage concept stems from a concern
that unchecked competition among employers would lead to a
less-skilled and less-productive workforce and to shoddy construction
practices and unsafe public buildings and infrastructure” (Fiscal
Policy Institute 2007: 17). Allegedly, employing only workers at prevailing
wage rates minimizes poor work that will prove costly to the
public in the long run. Thus, to insist on competitive bidding with
regard to labor is to be penny wise and pound foolish.
That argument is no more persuasive than the safety argument. It
is difficult to quantify improvements in work quality that might be
associated with prevailing wage laws or to show that work quality
diminishes without such laws. However, what evidence there is does
not support the quality reduction hypothesis.
Ohio, after suspending its prevailing wage law with regard to
school construction, sought to ascertain whether the suspension had
led to a decrease in work quality. A study by the Ohio Legislative
Service Commission (2002: 11) found that 91 percent of the school
district officials surveyed stated that there was no change in the quality
of work done while open bidding was allowed; of the remaining 9
percent, only 3 percent said that quality had fallen while 6 percent
said it had improved.
The quality argument also is weak theoretically. Metzger and
Goldfarb (1983) argue that since contractors have several margins for
adjustment, compelling them to pay top dollar for labor may lead
them to economize elsewhere, thus leading to no quality improvements.
They write, “If one is concerned about the output quality, it
is not obvious why price floors on only one input out of several
should necessarily guarantee a quality improvement; even if output
quality improves, there is no guarantee that the improvement could
be achieved at a cost that is not excessive” (Metzger and Goldfarb
Moreover, there is no reason to think that quality of workmanship
necessarily rises or falls depending on the wage paid. An experienced
mason, for example, is not likely to do a better job if paid at the “prevailing”
wage than at the wage he usually receives. Finally, if quality
of work is an issue, it can better be addressed directly, with careful
monitoring of the work to see that it is done according to contract
specifications, than by assuming that quality falls unless workers are
paid the prevailing wage. It is instructive to note that in the private
sector, where quality of work and adherence to specifications is at
least if not more important than in government construction, given
the strong incentive of owners to avoid financial losses, there is no
trend toward voluntary adoption of prevailing wages in contracting.
To the contrary, in the private sector, the use nonunion firms has
been increasing for decades. In 1950 nonunion workers were only
some 16 percent of the workforce, but by 2004 that figure had
increased to almost 85 percent (Cook 2005: 639). If the prevailing
wage concept were an efficient means of realizing high-quality con-
20951_CATOpages.qxd 1/20/10 1:00 PM Page 148
struction work, we would expect to see it used in the private sector,
but we do not.
Less “Harmful Competition”
Prevailing wage advocates argue that the laws are beneficial
because they stop what they call “harmful competition” that “slashes
wages” and “undercuts standards.” Terminology like that is indicative
of special pleading rather than objective analysis, but the arguments
must nevertheless be examined.
The Fiscal Policy Institute paper referred to above is one that
attempts to show that prevailing wage laws only prevent “harmful”
competition. The authors write, “Prevailing wage requirements help
ensure that competition among contractors in bidding for construction
projects is channeled into areas of overall cost efficiency, high
productivity, and innovative methods and not unduly focused on
driving down wages and benefits” (Fiscal Policy Institute 2007: 23).
The authors of the Utah study voice the same ideas, saying,
“Prevailing wage laws are intended to get the government out of the
business of pulling down wages” and to “prevent the government
from hiring labor at below-standard rates” (Philips et al. 1995:
These contentions aim at getting people to believe that the ordinary
process of competition is harmful when it occurs in the market
for construction labor. Yet the arguments are badly flawed.
Consider first the idea that prevailing wage laws compel contractors
to focus more intently on “good” competition rather than economizing
on labor costs. There is a false dilemma built into that
argument—namely, that contractors will either compete on the basis
of labor costs or on the basis of efficiency and innovation. In fact,
contractors seeking business (and not just on government construction
projects) look at all the various ways they might gain an advantage
over their rivals. Contractors who are able to procure labor at
lower rates still have every incentive to maximize innovation and efficiency
to win bids.
Second, allowing contractors to employ workers who will accept
less than the “prevailing” wage does not “drive down” wages. It simply
means that construction workers who are usually paid at the
union scale will not always be employed on government projects. If
a nonunion carpenter is usually paid $30 per hour and his firm wins
the bid and he gets to work on the project, his wages have not been
“driven down.” It is true that in the absence of prevailing wage laws,
unionized firms will find it harder to win bids due to their high wage
rates and that may, in turn, put downward pressure on union rates
unless the firms can improve their efficiency—that is, focus more on
Third, the implicit assumption in this argument is that there is
something sacred about union rates, but union wage scales are no
more “standards” than are sticker prices on new cars or the fees that
lawyers would like to receive for their services. They are merely the
prices that the unions and their members want to receive. Allowing
others to offer to sell for less is no more socially harmful than allowing
car dealers to negotiate deals below sticker prices. In point of
fact, it is not unusual for unionized firms to “undercut” their own
rates when it is necessary to do so to win contracts.
Increased Earnings and Taxes Perhaps the most astounding argument in the Utah study is that any savings from repeal of prevailing wage laws is more than offset
by decreasing income tax receipts. Assuming a fixed amount of construction
work, it is true that income tax collections from workers
who previously benefited from prevailing wage laws will be slightly
lower if they work at lower rates of pay. (The authors do not consider
the possibility that savings from repeal of prevailing wage laws will
increase other government or private spending, thus leading to
higher tax collections elsewhere.) But that is not all that Philips and
his co-authors contend. They maintain that repeal of prevailing wage
laws lead to a decline in earnings across the entire construction
industry (Philips et al. 1995: 32).
Even if that argument were correct, it still would not lead to the
conclusion that prevailing wage laws are beneficial. The interest of
the citizens of a state lies in obtaining whatever goods and services
the government provides at the least cost, not in maximizing government
There is, however, no reason to accept the idea that repealing prevailing
wage laws drives down the earnings of workers in the segment
of the construction industry that is not subject to governmental stifling
of competition. The Utah paper claims that repeal of prevailing
wage laws leads to a “ripple effect lowering wages throughout the
local construction industry” (Philips et al. 1995: 16). No direct evidence
is provided to support that claim; nor do the authors give any
theory to explain how the market price of construction labor could be
driven lower by virtue of the ending of the prevailing wage bonus for
the subset of the construction industry engaged in government work.
Philips and his co-authors speak of “overheated bidding” for government
projects following the repeal of Utah’s statute in 1981 (a tendentious
phrase that seems just to connote increasing competition),
but they do not show that such competition did or could have any
impact on construction wages where competition was already unrestrained.
Improved Worker Training
Advocates of prevailing wage laws often argue that in their
absence, training of construction workers is inadequate. This argument
begins with the premise that employment in the construction
industry is less stable than in most industries and that workers are apt
to move from company to company as jobs come and go.
Consequently, it is said, employers have little incentive to invest in
worker training because they probably would not capture the benefit
of such training. The Utah study maintains that repeal of prevailing
wage laws have “significantly lowered formal, organized, and
quality training of construction workers. The effect is to lower training
rates by about 40 percent” (Philips et al. 1995: 51). Construction
unions, the argument continues, provide a reliable system of apprenticeships,
but that system is undermined by the repeal of prevailing
wage laws. Supposedly, therefore, the supply of well-trained workers
depends on union training, which in turn depends on preventing
competition in wage rates on government construction projects.
If that argument held true, it would point to a glaring market failure:
that business firms in the construction industry would be unable
to obtain the services of competent workers if it were not for a positive
externality stemming from an anti-competitive intervention. In
fact, however, there is no more a market failure in the training of
construction workers than in the training of workers for any other
occupation. Construction workers are trained in several ways other
than union apprenticeship programs. The Associated Builders and
Contractors trade group runs apprenticeship programs. Workers can
also learn their skills at community colleges and private technical
institutes. As noted earlier, nonunion contractors encourage on-thejob
training for unskilled “helpers.” The idea that a state’s construction
labor force would be undertrained in the absence of prevailing
wage laws flies in the face of experience. In states with no prevailing
wage law and few construction unions (North Carolina, for example),
the construction industry nevertheless competently builds the same
array of things as are built in heavily unionized states with prevailing
wage laws. Like the others, the training argument for prevailing wage
Morgan Reynolds (1982: 307) calls the arguments for DavisBacon
(and state prevailing wage laws) “high-sounding justifications”
that “are a thin disguise for a law that brings home the bacon for
organized labor.” It is difficult to disagree with that characterization.
The purpose and effect of prevailing wage laws is to eliminate
competition on labor costs on government construction projects.
Bidders may search for the least-cost combination of other factors,
but labor costs are fixed by decree. This suppression of competition
is a substantial benefit to a small segment of the population, chiefly
construction unions and workers, at the expense of the rest of society,
which must pay more than would otherwise be necessary for
projects subject to prevailing wage mandates.
Efforts by prevailing wage proponents to depict the laws as having
some social benefit fail. Fixing the price of labor does nothing to
increase safety, train new workers, promote quality or any other
desirable objective. Nor is there any social benefit in “protecting”
union wage standards and work rules from competitive pressure.
Prevailing wage laws are special interest legislation trying to masquerade
as wise public policy.
People prefer to minimize or eliminate competition in markets
where they sell, while enjoying the benefits of competition in markets
where they buy. Prevailing wage laws are one of the various
approaches organized labor uses to shut down competition in labor
markets. Adam Smith was correct: It is bad public policy for government
to assist any group of sellers in their desire to fix prices and stifle
competition. That is why all prevailing wage laws should be repealed.
Bernstein, D. (1993) “The Davis-Bacon Act: Let’s Bring Jim Crow to
an End.” Cato Institute Briefing Paper #17. Available at www.cato
20951_CATOpages.qxd 1/20/10 1:00 PM Page 152
Prevailing Wage Laws
Congressional Budget Office (2000) Budget Options 2000. Available
Cook, S. (2005) Freedom in the Workplace: The Untold Story of
Merit Shop Contracting’s Crusade against Compulsory Trade
Unionism. Washington: Regnery.
Dunn, S.; Quigley, J.; and Rosenthal, L. (2005) “The Effect of
Prevailing Wage Requirements on the Cost of Low-Income
Housing.” Industrial and Labor Relations Review 59 (1): 141–57.
Fiscal Policy Institute (2007) “Building Up New York, Tearing Down
Job Quality: Taxpayer Impact of Worsening Employment
Practices in New York City’s Construction Industry.” Available at
Fraundorf, M.; Farrell, J.; and Mason, R. (1984) “The Effect of the
Davis-Bacon Act on Construction Costs in Rural Areas.” Review
of Economics and Statistics 66 (1): 142–46.
Glassman, S.; Head, M.; Tuerck, D.; and Bachman, P. (2008) “The
Federal Davis-Bacon Act: The Prevailing Mismeasure of Wages.”
Beacon Hill Institute at Suffolk University. Available at www
Gould, J. P., and Bittlingmayer, G. (1980) The Economics of the
Davis-Bacon Act: An Analysis of Prevailing Wage Laws.
Washington: American Enterprise Institute.
Kersey, P. (2007) “The Effects of Michigan’s Prevailing Wage Law.”
Mackinac Center for Public Policy. Available at www.mackinac
Leef, G. C. (2005) Free Choice for Workers: A History of the Right
to Work Movement. Ottawa, Ill.: Jameson Books.
Metzger, M., and Goldfarb, R. (1983) “Do Davis-Bacon Minimum
Wages Raise Product Quality?” Journal of Labor Research 4 (3):
Northrup, H. (1992) “The Helper Controversy in the Construction
Industry.” Journal of Labor Research 13 (4): 421–35.
Ohio Legislative Service Commission (2002) “S.B. 102 Report: The
Effects of the Exemption of School Construction Projects from
Ohio’s Prevailing Wage Law.” Staff Research Report #149.
Philips, P.; Mangum, G.; Waitzman, N.; and Yeagle, A. (1995)
“Losing Ground: Lessons from the Repeal of Nine ‘Little DavisBacon
Acts.’” Department of Economics, University of Utah.
20951_CATOpages.qxd 1/20/10 1:00 PM Page 153
Available at www.faircontracting.org/NAFCnewsite/prevailing
Reynolds, M. (1982) “Understanding Political Pricing of Labor
Services: The Davis-Bacon Act.” Journal of Labor Research 3 (3):
Roistacher, E.; Perine, J.; and Schultz, H. (2008) “Prevailing
Wisdom: The Potential Impact of Prevailing Wages on Affordable
Housing.” Citizens Housing and Planning Council. Available at
Rothbard, M. N. (1963) America’s Great Depression. Kansas City:
Sheed and Ward.
Smith, A. ( 1937) The Wealth of Nations. New York: Random
House, The Modern Library.
Thieblot, A. (1996) “A New Evaluation of Impacts of Prevailing
Wage Law Repeal.” Journal of Labor Research 17 (2): 297–322.
U.S. Senate, Committee on Banking, Subcommittee on Housing and
Urban Affairs (1979). Hearings on the administration of the
Davis-Bacon Act (2 May).
Vedder, R. (1999) “Michigan’s Prevailing Wage Law and Its Effects
on Government Spending and Construction Employment.”
Mackinac Center for Public Policy. Available at www.mackinac
20951_CATOpages.qxd 1/20/10 1:00 PM Page 154