Chairman Hershman and Members of the Tax and Fiscal Policy Committee, thank you for inviting me to testify. My name is James Sherk. I am a Research Fellow in Labor Economics at The Heritage Foundation. The views I express in this testimony are my own, and should not be construed as representing any official position of The Heritage Foundation.
The Common Construction wage insulates construction unions from competition. It requires public bodies to frequently pay union wage scales on construction projects. This inflates the cost of construction labor with no benefit to taxpayers. Research shows such “prevailing wage” laws do not increase the quality of construction work performed. Nor do unionized construction firms have high enough productivity to offset the higher cost of their labor. This is why unions have lost market share throughout the construction industry over the past generation.
The Common Construction wage benefits unions at the expense of the public good.
Competition and Cartels
Competition forces businesses to keep prices down. Absent competitive pressures they would produce less and charge higher prices. While this would benefit producers it would impose even greater economic losses on the rest of society. Federal anti-trust law forbids price-fixing for exactly this reason.
In economic terms unions operate as labor cartels. They also attempt to restrict the supply of labor in industry so as to drive up its price, wages. Like business monopolies labor cartels reduce social welfare. Higher wages for union members come at the cost of reduced job opportunities and higher prices for non-union members. These economic losses outweigh the gains the union enjoys.[1] Unlike businesses unions have a special exemption from anti-trust law. Unions can legally collude to raise the price of their labor.
However, unions have lost control of most of the labor markets they formerly dominated. This has happened because union contracts make businesses less competitive. Non-union firms have lower costs and can change their business plans without taking time to bargain with a union. This enables them to charge lower prices and adapt more quickly to changing market conditions. Over time they expand and gain market share at the expense of their unionized competitors.[2]
This dynamic has driven the dramatic decline in U.S. union density since the 1970s.[3] Over the past generation unionized firms have steadily lost ground to more efficient non-union competitors. In 1977 unions represented 23 percent of private-sector workers. By 2014 that figure fell to 7 percent.[4]
Construction unions have fared particularly poorly during this period. Unionized construction firms usually charge more than non-union firms for the same work. As a result they often lose contract bids and unionized jobs decrease.
Over the past 37 years total employment in the construction industry nearly doubled, rising from 4.0 million to 7.0 million. Non-union firms account for all of this growth. Between 1977 and 2014 non-union construction employment increased by almost 3.5 million workers while unionized employment fell by a net of 500,000 workers. As a result union density in the construction industry collapsed. In the 1970s, collective bargaining agreements covered 38 percent of construction workers. Today they cover just 15 percent.[5] Most unionized firms cannot compete on price with their non-union competitors.
Prevailing Wage Laws Limit Competition
Unionized construction workers now work primarily on public projects; federal and state laws insulate unions from competition on these jobs. The federal Davis–Bacon Act and similar state “prevailing wage” laws require contractors on federal and state projects to pay pre-determined wage scales. In theory these scales are supposed to reflect prevailing market rates. In practice they are usually union wage scales.
The Inspector General and Government Accountability Office (GAO) have long criticized the Labor Department for using unscientific and error-ridden techniques to calculate Davis–Bacon rates.[6] As a result of these errors almost two-thirds of Davis–Bacon rates are union wage scales, despite unions representing just one-seventh of construction industry workers.[7] Detailed analysis shows that federal Davis–Bacon rates average 22 percent above actual market wages.[8]
Indiana and 31 other states have their own state-level prevailing wage laws. Indiana’s Common Construction Wage (CCW) requires contractors on state-funded projects to pay pre-set rates to all employees.[9] Before any project costing more than $350,000 can go forward, the entity funding it must request a prevailing wage determination. All contractors on that project must pay at least those rates to their employees.
Like the Davis–Bacon Act, CCW rates usually reflect union wages. Bureau of Labor Statistics data shows that unions represent 32 percent of all construction workers in Indiana.[10] However, in 2014 fully 65 percent of CCW determinations reflected union wage scales.[11]
These union compensation scales are substantially higher than those of non-union contractors. Table 1 shows the difference between adopted AFL-CIO and Associated Builders and Contractors CCW hourly compensation rates across several Indiana counties.[12] The Associated Builders and Contracts (ABC) is a non-union construction trade association. In every case the union rate significantly exceeds the ABC rate. For example, in Tippecanoe County the most recent prevailing wage determination for skilled electricians called for compensation of $29.95 an hour. An AFL-CIO determination adopted the same day called for paying $46.44 an hour—more than 50 percent more.[13]
Across these counties the average union prevailing wage requires paying 62 percent more for electricians, 67 percent more for carpenters, and 51 percent more for bricklayers than the non-union rate. The CCW causes the state and local governments to pay considerably more for construction work than most non-government contractors pay.[14]
Prevailing wage laws force non-union firms to adopt the same cost structure as their unionized competitors. This makes it much harder to undercut unionized firms on the basis of price. Prevailing wage laws have enabled unions to maintain a strong presence in government construction work even as competition has driven them from most private projects in the rest of the economy.
Higher Costs for Taxpayers
Unions unsurprisingly strongly support prevailing wage restrictions. They mean higher wages and more jobs for their members. However, prevailing wages also raise costs for taxpayers. Limiting competition makes public construction cost more.
Researchers have determined that the Davis–Bacon Act raises the cost of federal construction projects by nearly 10 percent.[15] States that have repealed their prevailing wage laws have also experienced large savings in construction costs.
For example, Ohio repealed its prevailing wage law for school construction in 1997. The same legislation required the Ohio Legislative Service Commission (OLSC) to subsequently analyze the fiscal consequences of the repeal. The OLSC found that prevailing wage repeal reduced the cost of school construction by 10.7 percent.[16] Over the five-year time period studied it saved Ohio taxpayers approximately $500 million.
Interestingly, the OLSC report broke down the savings by type of construction project. It found much smaller savings on new school construction (1.2 percent) than on school additions (19.9 percent) or alterations (10.7 percent). Spending on material and construction equipment typically takes up a greater proportion—and labor a smaller proportion—of the budget for new construction than for additions or alterations. Prevailing wage requirements have a greater effect on more labor-intensive projects.
The OLSC report also included an analysis of the experience of the Westlake City school district that bid out a project initially under the prevailing wage law, then again after the legislature repealed the law. The initial bid came in at $8.7 million. The subsequent non-prevailing wage bid came in at $8.2 million dollars. Prevailing wage repeal saved Westlake City taxpayers almost $500,000 and lowered total costs by 5.8 percent.
In Michigan federal courts temporarily suspended the states’ prevailing wage law between 1994 and 1997. A federal district court ruled that the Employee Retirement Income Security Act (ERISA) preempted Michigan’s extensive prevailing wage requirements. The Sixth Circuit Court of Appeals reversed that decision on appeal two and a half years later. Economic analysis found the temporary suspension of the prevailing wage law saved Michigan taxpayers $275 million a year.[17]
Other studies come to similar conclusions. In 2001 California legislators required developers constructing subsidized low-income housing to pay prevailing wage rates. Economists found this requirement increased construction costs between 9 percent and 37 percent. This hurt taxpayers and low-income families. The higher costs caused developers to build 3,100 low-income housing units a year.[18]
The Citizens Housing and Planning Council in New York City also examined the impact of proposals to impose prevailing wage requirements on low-income housing. It found that prevailing wages would raise total costs by 25 percent and add $400 a month to rental costs—a significant burden on low-income renters.[19]
Prevailing wage laws transfer wealth from taxpayers and the recipients of public services to construction union members.
Productivity Does Not Offset Higher Costs
To defend prevailing wage requirements unions often contend that taxpayers pay no more for prevailing wage projects because their members have higher productivity than non-union workers. They argue that this higher productivity entirely offsets the higher labor costs.
The collapse of union density in private construction projects demonstrates that this is not the case. Many construction unions do invest heavily in training programs. On average, however, unionized construction workers do not produce enough in additional value to offset the cost of their higher compensation scales. If they did they would win more bids. Unions support prevailing wage laws precisely because productivity differences do not offset differences between union and non-union compensation scales.[20]
No Quality Benefits
Unions also justify prevailing wage restrictions on the basis that union members produce higher quality construction work. They argue that taxpayers might pay more, but they also get more. However, mandating higher wages is an indirect and potentially counterproductive way to increase quality. If policymakers want to ensure quality they should specify quality standards and require contractors to meet them. This would prevent substandard operators from winning by submitting bids that cut corners.
Attempting to indirectly increase quality by mandating higher wages can backfire. Contractors could partially offset their higher wage bills by using less expensive—and lower quality—construction materials. They could also use lower quality but less labor-intensive construction techniques. Nothing in prevailing wage requirements compels contractors to produce higher quality buildings.[21]
Actual empirical evidence shows that prevailing wages do not lead to higher quality construction projects. As part of its evaluation the OLSC asked school districts twice about the quality of construction performed after the repeal of the prevailing wage law. The first survey found 1 percent of school districts said quality had decreased, while 35 percent said quality increased. The second survey, a year later, found 3 percent of districts reporting lower quality and 6 percent reporting higher quality.[22] On the whole, school districts found repealing prevailing wages had no detrimental effect on quality. As one school district responded when asked if it had:
No, the quality of construction and the finished product remain the same as projects done prior to the exemption taking effect. I believe this is a function of how well the specifications are written, the reputation of the company doing the work, the quality of the product used, and the amount of supervision of the project by the owner and the architect. We have seen no change order increase nor additional delays with projects after the exemption went into effect. Specifications on all projects included a completion date.[23]
The collapse of union density in non-government construction projects also demonstrates that higher compensation rates do not automatically translate into higher quality. Private-sector clients care as much about quality as public entities. If they saw offsetting quality benefits from hiring union contractors they would hire them more frequently. The fact that they do not corroborates the experience of Ohio’s school districts. Prevailing wage restrictions are ineffective at increasing construction quality.
Conclusion
Indiana’s Common Construction Wage makes it much harder for non-union contractors to compete against union contractors on the basis of lower labor costs. Last year two-thirds of CCW determinations were union rates, despite unions representing less than a third of construction workers in Indiana. The CCW requires non-union firms to adopt union compensation packages.
Unions enjoy being insulated from competitive pressures. However these higher costs get passed onto taxpayers in the form of more expensive construction projects. Contrary to union claims, higher wages do not produce offsetting productivity or quality benefits. The CCW causes Hoosiers to pay higher taxes and enjoy fewer public services.
Endnotes
[1] George Borjas, Labor Economics, 3rd ed. (Columbus, OH: McGraw-Hill, 2005), pp. 413–415.
[2] Barry Hirsch, “Sluggish Institutions in a Dynamic World: Can Unions and Industrial Competition Coexist?” Journal of Economic Perspectives, Vol. 22, No. 1 (Winter 2008), pp. 153–176.
[3] Ibid.
[4] Barry T. Hirsch and David A. Macpherson, “Union Membership and Coverage Database from the Current Population Survey,” Industrial Labor Relations Review, Vol. 56, No. 2 (January 2003), pp. 349–354, http://www.unionstats.gsu.edu/ (accessed March 27, 2015). These figures include both the workers who formally belong to unions and non-members covered by collective bargaining agreements.
[5] Ibid.
[6] U.S. Government Accountability Office, Davis–Bacon Act: Methodological Changes Needed to Improve Wage Survey, GAO–11–152, March 2011, http://www.gao.gov/new.items/d11152.pdf (accessed March 27, 2015); U.S. Department of Labor, Office of Inspector General, Concerns Persist with the Integrity of Davis–Bacon Act Prevailing Wage Determinations, Audit Report No. 04-04-003-04-420, March 30, 2004, pp. 12–13, http://www.oig.dol.gov/public/reports/oa/2004/04-04-003-04-420.pdf (accessed March 27, 2015); U.S. Department of Labor, Office of Inspector General, Inaccurate Data Were Frequently Used in Wage Determinations Made Under the Davis–Bacon Act, Audit Report No. 04-97-013-04-420, March 10, 1997, http://www.oig.dol.gov/public/reports/oa/pre_1998/04-97-013-04-420s.htm (accessed March 27, 2015); and U.S. General Accounting Office, Davis–Bacon Act: Labor Now Verifies Wage Data, but Verification Process Needs Improvement, HEHS-99-21, January 1999, http://www.gao.gov/archive/1999/he99021.pdf(accessed March 27, 2015).
[7] U.S. Government Accountability Office, Davis–Bacon Act: Methodological Changes Needed to Improve Wage Survey, p. 20.
[8] Sarah Glassman, Michael Head, David G. Tuerck, and Paul Bachman, “The Federal Davis–Bacon Act: he Prevailing Mismeasure of Wages,” Suffolk University, Beacon Hill Institute, February 2008, at http://www.beaconhill.org/BHIStudies/PrevWage08/DavisBaconPrevWage080207Final.pdf (accessed March 27, 2015).
[9] Indiana Code 5-16-7-1 et seq.
[10] Heritage Foundation calculations using data from the outgoing rotation groups of the 2012, 2013, and 2014 Current Population Survey. Figures included all workers who belong to a union or are represented by a union on the job.
[11] Heritage Foundation analysis of 2014 adopted common construction wage scales published by the Indiana Department of Labor, “Adopted Common Construction Wage Scales,” http://www.in.gov/dol/2840.htm (accessed March 27, 2015).
[12] These rates include both cash wages and non-cash benefits.
[13] Indiana Department of Labor, “Adopted Common Construction Wage Scales,” determination numbers 20140256 and 20140281, http://www.in.gov/dol/2840.htm (accessed March 27, 2015).
[14] The disparity is larger than these figures suggest. The Bureau of Labor Statistics data includes both construction workers employed on government projects subject to prevailing wage laws and those employed on private projects without prevailing wage mandates. Union density is much higher on prevailing wage projects than throughout the overall economy. Consequently, the proportion of unionized workers on private projects is smaller than the 32 percent statewide average.
[15] Sarah Glassman, Michael Head, David G. Tuerck, and Paul Bachman, “The Federal Davis–Bacon Act: The Prevailing Mismeasure of Wages,”
[16] Ohio Legislative Service Commission, “The Effects of the Exemption of School Construction from Ohio’s Prevailing Wage,” Staff Repot No. 149, May 20, 2002, http://www.lsc.ohio.gov/research/srr149.pdf (accessed March 27, 2015).
[17] Richard Vedder, “Michigan’s Prevailing Wage Law and Its Effects on Government Spending and Employment,” The Mackinac Center for Public Policy, Report No. S99-07, September 1999, https://www.mackinac.org/archives/1999/s1999-07.pdf (accessed March 27, 2015).
[18] Sarah Dunn, John Quigley, and Larry Rosenthal, “The Effect of Prevailing Wage Requirements on the Cost of Low-Income Housing,” Industrial and Labor Relations Review, Vol. 59, No. 1 (2005), pp. 141–157.
[19] Elizabeth Roistacher, Jerilyn Perine, and Harold Schultz, “Prevailing Wisdom: The Potential Impact of Prevailing Wages on Affordable Housing,” Citizens Housing and Planning Council, 2008, http://chpcny.org/wp-content/uploads/2011/02/Prevailing-Wisdom-web-version1.pdf (accessed March 27, 2015).
[20] One study put forward claiming offsetting productivity differences analyzed Census Bureau data on construction spending. It reported construction workers in states with prevailing wages laws had 14 percent to 33 percent greater productivity. See Peter Philips, “Indiana’s Common Construction Wage Law: An Economic Impact Analysis,” January 2015, http://www.isbctc.org/Uploads/UploadedFiles/docs/Philips_Indiana_Report_January_2015.pdf (accessed March 27, 2015). This study misinterpreted the Census data. It divided Census estimates of industry value added by the number of construction workers per state. However, the value-added figures take the sales price of construction projects and subtract from them the cost of materials used in construction. The remaining costs include higher wage rates imposed by prevailing wage laws. The study found only that per-worker construction costs are higher in prevailing wage states—it says little about whether unionized construction workers actually produce more.
[21] Michael Metzger and Robert Goldfarb, “Do Davis-Bacon Minimum Wages Raise Product Quality?” Journal of Labor Research, Vol. 4, No. 3 (1983), pp. 265–272.
[22] Ohio Legislative Service Commission, “The Effects of the Exemption of School Construction from Ohio’s Prevailing Wage,” Staff Repot No. 149, May 20, 2002, pp. 26–35, http://www.lsc.ohio.gov/research/srr149.pdf (accessed March 27, 2015).
[23] Ibid., pp. 30–31.
