315 Dece mber 15, 1983 TRIMMING THE HUGE COST .OF DAWS-BACON c INTRODUCTION The Davis-Bacon Act, enacted in 1931, requires contractors to pay the llprevailingll wage on federally funded construction projects. The purpose of the Act was to protect local construc ti on labor from the competition of itinerant contractors and their lower paid, often minority, employees.
The Davis-Bacon Act has generated controversy since it became law. Critics argue that it raises wage levels artificially boosting the cost of federal co nstruction projects, and discrimi nates against nonunion and minority workers. There is also considerable dispute concerning the method used to determine the prevailing wage Nearly every academic study has concluded that the Act's primary effect is to rai se'the wage and employment levels of union workers to.the detriment of nonunion and minority workers.
It should be no surprise, therefore, that organized labor has been the Act's biggest supporter.
By driving up the price of federal construction, Davis-Bacon currently is costing the taxpayer more than 1 billion annua1ly.l This is an unnecessary cost and a particularly heavy burden at a time of huge federal deficits.
As such, Davis-Bacon at long last should be repealed.
Administration's objective of reduc ing government spending while eliminating burdensome regulations. Most important, repeal would generate thousands of new employment opportunities for the jobless This would be consistent with the Reagan Co'ngressional Budget Office Modifying the Davis-Bac o n Act: Implications for the Labor Market and the Federal Budget July 1983, p. 29. 2 BACKGROUND The Davis-Bacon Act was the first federal legislation man dating minimum wages for nonfederal workers contractors of federally funded construction projects to p a y wage rates and fringe benefits at least equal to those prevailing on similar projects in private industry contracts with a value of over $2,000 applies to federal construction, its prevailing wage provisions have been added to nearly 60 other laws deali n g with federally financed and federally assisted construction The Act requires The law applies to all And while the Act itself The prevailing wage rates are determined by the Department of Labor (DOL) based upon its assessment of the compens.ation paid to equivalent workers in the area. Davis-Bacon, in other words is a variant 0.f the minimum wage.
The purpose of the Act was to shield local contractors and construction workers from the competition of outside contractors paying lower wage rates in order to win federal government con tracts. Or as Representative Robert Bacon (R-NY) put it during the debates in 1931, "A practice has been growing up in carrying out the building where certain itinerant, irresponsible contrac tors, with itinerant, cheap, bootleg labor have been going around throughout the country picking off a contract here and a contract there. If Wage Determinations A 1979 study by the General Accounting Office (GAO) concluded that "after nearly 50 years, the Department of Labor has not develop e d an effective program'to issue and maintain current and accurate wage determinations; it may be impractical to ever do so.113 Much of the problem involves a precise definition of the term "prevailing wage.If The .DOL'S determination is based upon the rat e paid to 1) the majority of workers in the area of the proposed construction project, adjusted for the appropriate job classification; or (2 if there is no majority, then the rste paid to the greatest number of workers providing it is greater than 30 perc ent of those employed; or (3) if no such rates exist then the average wage rate used.4 74, Congressional Record, 6510, February 28, 1931.
U..S. Comptroller General, The Davis-Bacon Act Should Be Repealed (HRD 79-18) (Washington, D.C General Accounting Office, April 27, 1979).
In May 1982, the DOL published new regulations for the administration of Davis-Bacon. Among the revisions was the elimination of the so-called 30 percent rule, and effective June 28, 1983, this step has been omitted.
Ho wever, the ruling may be appealed in the courts and the rule could be reinstated. 3 1 Unless the average wage rate is used, however, the determina tion may be nowhere near the prevailing wage in any normal sense of the term. Indeed, the provisions of the A ct have often been based on as few as 30 percent of workers within a job classifica tion paid to workers in the area.5 toward using the generally higher union wage scales, rather than the typical construction wage, since union members are more likely to b e paid a standard rate, while nonunion rates tend to vary considerably between workers. In addition, the determina tions are based on both field surveys and the voluntary submis sion of wage data. The latter method tends to encourage a union rate bias, sin c e data from collective bargaining agreements are easily available, while data for nonunion workers are more diffi cult to compile. In fact, one analysis by GAO found that 57 percent of determinations were based on collective bargaining agreements.6 Not on l y was the department's methodology flawed in this way, contended the GAO study, but the DOL in many instances deleted, added, and changed the wage data received without adequate reason or rationale. Ir7 This may be quite different from the average wage ra t e There is also an inherent bias Since the use of union negotiated wages as the base for the prevailing wage tends to push wage rates on federal projects above market levels, federal construction costs are raised.8 Open shop (nonunion) contractors are for ced to raise their bids on federal construction projects and pay above market wages.
Several other weaknesses in the way the Act is administered also lead to artificially high wage determinations and project costs. Worker classifications and the appropriat e geographic area for wage comparisons, for instance, result in inflated determinations. The GAO found that many of the wage rates prescribed by Labor were not based on similar construction work.Ir9 Davis-Bacon rates are supposed to reflect wage rates pre v ailing in the private sector, but the DOL usually includes in its calcu lations labor costs from other federally funded projects. This tends to raise the rates still further, leading to excessive costs 8 In some areas, there is not enough construction to provide the information needed for an accurate determination of the pre vailing wage. In such cases, the DOL may consider wage rates paid GAO, op. cit., pp. 52-53 Ibid. 9 p. 43.
Ibid., p. iii.
In 1981, unionized construction workers earned an .average of $2,700 more year than their nonunion counterparts. Calculated from CBO, op. cit p. 16.
GAO, op. cit., p.
50. This may occasionally occur because there is no comparable construction in the private sector, e.g on interstate high ways returned to the state and local level or the private sector anyway One could easily argue, however, that most such building should be 4 on the nearest similar construction. But this can lead to an up ward bias in the prevailing wage determination. One recent study of rural co n struction projects revealed that only 28 percent of the federal projects ended up being built by local contractors compared with 47 percent of the private projects.1 So the Act's original aim of protecting local employment is actually undermined in the ca se of many rural projects.
The cumulative effect of 'these factors affecting wage determinations is that the cost of federal projects is raised relative to nonfederal. The net result, according to the DOL was that Davis-Bacon rates in 1982 were more than 5 percent above average rates, resulting In additional costs of.$568 million.11 Use of Labor The Davis-Bacon Act affects the way labor is used. For example, the DOL often does not recognize semiskilled or helper categories in its wage determinations. If a w orker undertakes a particular job, then he must be paid the full Davis-Bacon rate for that job classification, regardless of his skill level. When a general laborer is employed to paint on a public project, for instance, he must be paid a skilled painter' s wage; if he then does some carpentry, he must be paid a skilled carpenter's wage.
This provision discriminates against nonunion contractors who employ less specialized, more adaptable workers. Nonunion firms also tend to make greater use of apprentices a nd helpers and forcing them to pay journeymen's wages to relatively inex perienced workers raises costs and limits their ability to undertake federal projects. The DOL has estimated that allowing the unlimited use of helpers on public projects would have saved 480 million in fiscal year 1982.12 Compliance Excessive prevailing wage determinations and the misalloca tion of labor are not the only costs imposed by Davis-Bacoi.
Compliance costs raise the expense of federal projects to con tractors and hence to taxpayers. Contractors must submit weekly payroll information, such as hours worked and wages paid, for each employee on a Davis-Bacon project. These requirements are particularly burdensome'for small contractors who lack the staff to comply easily costs a mounted to 100 million in 1982.13 There are, of course The DOL has estimated that these compliance lo Cited in CBO, op. cit., p. 26 l1 U.S. Department of Labor, Final Regulatory Impact and Regulatory Flexibil ity Analysis of Davis-Bacon Related Regulation s (1982), cited in CBO op. cit p. 26 l2 Ibid l3 Ibid. 5 additonal costs to the taxpayer arising from the federal bureau cracy needed to administer the program.
In sum, DOL estimated that in 1982 Davis-Bacon inflated wage costs by 570 million of $100 millio n and $480 million in other costs by its restrictive use of helpers resources and inflate the cost of federal construction by over $1 billion, the taxpayer gains nothing In addition, it led to compliance costs Since the Act does little more than misalloca t e THE ECONOMIC EFFECTS Job Losses In addition to the direct budget costs of Davis-Bacon, the Act harms the economy. Perhaps most damaging is its depressing impact on construction employment levels. By setting an artifi cially high wage floor, Davis-Bacon i ncreases the cost of each employee. The result: the number of job openings available for any given federal outlay is reduced outlays rise to take account of the labor costs imposed by the Act, federal construction employment may not fall, but the wind fal l gains to these construction workers merely reduce the funds and jobs available to workers in other ind~str1es.l If federal construction Barriers to Low-Skilled Workers Davis-Bacon favors union workers. High prevailing wage determinations, for instance, g i ve contractors the incentive to hire only the most skilled workers, since less skilled workers do not warrant the high wages mandated by the Act unskilled workers may be willing to accept lower wages to gain entrance to the construction industry and devel op a skill, they are precluded from doing so by the raw economics of the Davis Bacon Act While young Open-shop contractors awarded a federal contract must'raise the wage levels of workers on the project to Davis-Bacon levels.
This can create morale problem s when employees on the contract are paid more than others doing similar work in the private sector. Further discontent may arise when wage rates are returned to market levels on completion of the project survey found that 23 percent of the open-shop cont r actors ques tioned reported that working on Davis-Bacon projects would be disruptive to their normal routine and so they would be unlikely One I I i While not reducing employment in the construction industry in this case Davis-Bacon may raise the industry ' s unemployment rate since higher wage rates will increase l.abor supply, i.e., those seeking work, in the con struction industry. In addition, by raising costs to the federal govern ment, resources and jobs will be drained from the private sector. 6 to bi d on a federal contract. Of those who had undertaken Davis-Bacon construction, 20 percent said they would not want to do so again.
Discrimination A government imposed wage rate above the market level creates a surplus of job seekers, making discrimination easier. By undermining competitive forces in this way, Davis-Bacon allows employers to discriminate by race (or for any other reas o n without suffering a financial penalty. Without Davis-Bacon minority workers could offer their services at a lower wage, and employers who discriminated would then have higher labor costs and consequently lower profits. Thus, a free market would make dis crimination more difficult.
Davis-Bacon also eliminates opportunities for low-skilled minority workers, because trainee and helper restrictions deny these workers employment experience and on-the-job training.
Consider the case of the House of Umoja's Boy stown Construc tion Project in Philadelphia.16 In 1981-1982, the House of Umoja, a neighborhood group, undertook a construction and rehabil itation project to house delinquent black youth consisted of two phases: Phase A was partly funded from govern ment sources, and therefore subject to Davis-Bacon requirements.
The contractor did not hire any of the Umoja youth for his part of the project. His reason: local youth were generally not productive enough to merit payment at the high Davis-Bacon scale the Act consisting of ex-offenders and young local workers monitored by experts project, and it was completed at about 40 percent of the cost of Phase A, even though the construction was almost identical.
Phase B allowed disadvantaged minority youths to acquire the basic skills needed for a future in the construction industry An additional three phases of the Umoja Boystown are being planned, using 700,000 of public funds, but the House of Umoja staff complain that Davis-Bacon will restrict the opportunities for young people As construction director Thomas Massaro puts it, IfFor the proud and proven young men of the Umoja Construction Co. to sit by once again as outsiders earn big dollars on their block would be a tragedy. Ir1 The project Phase B, however, was pr ivately financed and not subject to The work was performed by the Umoja Construction Company Umoja supplied about two-thirds of the man-hours on the l5 l6 Cited in CBO, op. cit p. 31.
See Memo from project director Thomas Massaro to Robert L. Woodson, "House of Umoja Boystown Construction Project Winning and Losing on Davis-Bacon,"
November 11, 1982 l7 Ibid p; 5 7 The original supporters of the Act knew well that the law would reduce opportunities for blacks. George Mason University economist Walter Willia ms noted in recent testimony before the Senate Subcommittee on Labor that South Africa imposes restric tions very similar to Davis-Bacon White racist unions there support laws like the Davis Bacon Act for the expressed purpose of protecting white labor fr o m low wage black competition. Over there they call these laws '!standard rate." Their stated intent is one thing; yours is another. The effect is roughly the same in both places.18 Effect on Inflation Some critics of Davis-Bacon argue that the Act is infl a tionary, since higher wages on construction projects raise costs and push up prices prices of public construction and related goods, leaving taxpayers with less money to spend on other goods prices of thqse other goods to ease itself inflationary, though it changes relative prices. Davis Bacon, however, does contribute to inflation when its direct consequences, such as an increase in the deficit or higher unem ployment, encourage the government to adopt the easy money policies that do cause inf1ati0n.l Hi g her federal construction costs do push up But this causes the So Davis-Bacon is not of Competition The assumption underlying Davis-Bacon is that competition is The view that competition is destructive to economic harmful to jobs and income and that protec tionist measures are beneficial improvement is, of course, erroneous. Proponents of Davis-Bacon misunderstand the role of competition in improving productivity and income levels.
In addition, there should be no reason why the local con struction industry s hould be Ilprotectedll from competition Any more than other occupations in a given locality. If defenders of Davis-Bacon are to be consistent in their arguments, they should be against competition among suppliers or any firms servic ing the industry-indee d, all competition.
Stability Supporters of the Davis-Bacon Act invariably fall back on the contention that construction is more unstable and insecure than l8 Testimony of Professor Walter Williams in "Hearings Before the Subcommittee on Labor of the Commi ttee on Labor and Human Resources, U.S. Senate,"
April 29, 1980, p. 2
44. See Dwight R. Lee, The Inflationary Impact of Labor Unions (College Station, Texas: Center for Education and Research in Free Enterprise Research Monograph Series, No. 5 l9 8 other industries. Cyclical and seasonal factors tend to raise unemployment in the construction industry workers are paid higher hourly wages than most other workers these higher wages tend to offset the seasonal risk.
Davis-Bacon actually adds to the risk of u nemployment by pricing some workers out of a job But since construction Moreover Productivity Defenders of Davis-Bacon maintain that it has a positive effect on productivity by inducing contractors to improve worker skills to warrant the prevailing wage. T his is a variant of the theory that minimum wage laws Ifshock" employers into greater efficiency, and it is refuted by University of Dallas economist W. H. Hutt It is just not true that prospects of. adversity stimu lated managerial and technological imag ination, enter prise, and effort more than the prospects of prosperity.
If it were true, it would be wise for governments to impose burdens on any sector of the economy they wished to foster-taxing industry to give it a jolt and thereby cause it to flouris h!20 Any wage floor, including Davis-Bacon, simply encourages firms to change their mix of inputs higher quality labor for their existing workforce or substitute capital equipment for labor. This will raise the productivity of labor, but reduce the produc t ivity of other inputs. Labor produc tivity is not synonymous with efficiency vention that r.equires profit-maximizing firms to alter their production mix means higher total costs and reduced efficiency.21 They'might substitute Any government inter POSSIBL E REFORMS Repeal of the Davis-Bacon Act would be the ideal reform. If repeal is not now possible, at least the most damaging prohions of the Act should be removed. The Labor Department, for example in May 1982, issued regulations to significantly improve a d minis tration of the Act. These changes include 1) replacing the 30 percent rule" with guidelines that would set the prevailing wage based on a majority, or, in the absence of a majority, the average pay rate; (2) changing the lflocalityll rules so that h igh wage rates from an urban area would less likely be applied to a 2o 21 Cited in ibid p. 303.
For a fuller discussion of the productivity debate, see Joseph D. Reid Jr Labor Uni0n.s in the American Economy: An Analytical Survey,"
Journal of Labor Research, Summer 1982, pp. 277-294, and Michael R.
Metzger and Robert S. Goldfarb DO Davis-Bacon Minimum Wages Raise Product Quality?" Journal of Labor Research, Summer 1983, pp; 265-272. 9 suburban or rural region (the opposite would als o be true 3 excluding federal projects from all wage determinations 4) allowing two llhelpersll for every three journeymen on a con struction site, thus giving employers greater flexibility in hiring unskilled and semiskilled workers and thereby reducing l abor costs; and (5) relaxing the reporting requirements for fiims Though these proposed changes are a step in the right direc tion, 'legal challenges led the courts to strike down two of the regulations. Fortunately, the "30 percent ruletr has been elimi n ated from the wage determination procedure. This bias long has favored union wages In addition to regulatory reform of the Act, legislative changes could alleviate some of the worst features.23 threshold, which has remained constant since 1935, could be r aised in line with current construction costs. This would reduce the number of projects subject to Davis-Bacon requirements.
In addition, certain types of construction, such as military projects and low-income housing, could be exempted from the Act.
The threshold for many housing projects is based on unit size not value So the Davis-Bacon trigger for federally assisted housing could be raised from eight units to, say, twenty. Partial deregulation of this could prove to be successful, because it would rem o ve some of the worst effects of the law, while-arousing less political opposition than outright repeal vailing wage on a range of wages paid to workers in the private sector. Also, wage data from other federal projects could be excluded in the calculation of the prevailing wage, as could the application of metropolitan rates to rural areas. Though these latter reforms have been adopted through the regulatory process they could be strengthened in place legislatively.
Helpers and apprentices could be given m ore recognitiqn in The 2,000 Further legislative changes could include basing the pre the wage determination process. While the ideal would be to I I I allow unlimited use of helpers and apprentices, the maximum ratio of helpers to journeymen at least cou ld be eased. This would provide many new employment opportunities for minority and unskilled workers.
Finally, compliance costs could be reduced dramatically by allowing firms to submit compliance statements, rather than weekly payroll records, and to repo rt on a less frequent basis 22 23 The provision would not apply to highway and heavy construction projects where there is little nonfederal construction.
Two bills incorporating many useful changes are S. 1172, introduced by Senator Don Nickles (R-OK and H.R. 3846, introduced by Congressmen Charles Stenholm (D-TX) and Arland Stangeland (R-MN I 10 Such a change would only require a statement at the beginning and end of a contract,.though this would require dealing with the Copeland Anti-Kickback Act C0NCLU S .ION The Davis-Bacon Act is justified on neither efficiency nor equity grounds floor on federal construction contracts; it increases unemployment in the construction industry; it boosts costs to the taxpayer and it prices lower-skilled workers-most notabl y nonunion and minority workers-out of the market.
The net effect of the Act is to transfer resources from tax payers, consumers, nonunion workers, and the unemployed to union ized construction workers. Not only does Davis-Bacon lower total economic output by misallocating resources it redirects income to workers typically paid above average wages this inequity is quite simple: Unions and union contractors have a great deal of political clout It distorts the labor market by imposing a wage The reason for T h e government should undertake policies that raise produc tivity and wages for all Americans by providing incentives to improve capital accumulation and innovation as well as education and training. A law that raises real wages by providing a wage floor ca n never benefit all workers. The Davis-Bacon Act and many other government regulations serve only to.benefit special interest groups at everyone else's expense.
Peter G. Germanis Schultz Fellow