You asked for information about a 1997 Ohio law exempting school construction projects from the state's prevailing wage law. You also asked for information about a 2002 study of the impact of the exemption by the Ohio Legislative Services Commission that estimates that the exemption resulted in an aggregate savings of 10.7% in school construction costs from 1997 to 2001.

SUMMARY

Prevailing wage laws require contractors to pay their workers a specified minimum wage rate for work on public construction projects. The federal government and 32 states have such laws. States set different project cost thresholds and other criteria to determine when the laws apply. States also have different methods of determining prevailing wage rates.

Ohio's prevailing wage law has a relatively low project threshold compared to Connecticut and the Ohio law uses local union wages as its prevailing wage. Connecticut uses the federal rates, which are set based on rates paid to a majority of workers in a particular trade or job in a particular area or, if there is no majority rate, on the average pay rate.

Ohio exempted all school construction projects from its prevailing wage law in 1997. As part of the exemption, the legislature mandated that its Legislative Services Commission (LSC) report on the impact of the exemption on school construction costs, quality, and construction wages after five years. The LSC submitted its report in May 2002. The LSC estimated that the exemption had resulted aggregate savings of $487.9 million (10.7%) in school construction project costs between 1997 and 2001 without producing any discernable diminution in construction quality or negative impact on construction wages.

The data and analysis underpinning these conclusions was challenged by Ohio State University professor Herbert F. Weisberg, who published a paper in July 2002 contending that the LSC's savings estimate is not valid and that its conclusions about the impact of the exemption on construction quality and wages are based on faulty surveys.

OHIO AND CONNECTICUT PREVAILING WAGE LAWS

Ohio's prevailing wage law was enacted in 1931, the same year as the federal prevailing wage law (known as the Davis Bacon Act) and two years before Connecticut enacted its state prevailing wage law. Ohio's law applies to public construction projects costing more than $67,853 for new construction and $20,955 for remodeling. These thresholds are adjusted every even-numbered year for inflation up to a maximum adjustment of 3%. Connecticut's law applies to new construction projects costing more than $400,000 and remodeling projects costing more than $100,000.

Ohio's prevailing wages are the union wages for each trade or job as determined by the collective bargaining agreements that apply in the project's immediate locality. If there is no collective bargaining agreement that applies where the construction is occurring, then the collective bargain contract rates in the nearest locality are used. Contractors are also subject to the same work rules as apply under collective bargaining agreements used to set the wages.

Connecticut uses the prevailing wages set by the U.S. Department of Labor based on data from surveys completed by unions, contractors, and government agencies. Federal regulations require that the prevailing rate be either (1) the wage paid to a majority of the workers in a particular trade or job classification in the geographic area where the project is located or (2) if the same wage is not paid to a majority, the average of the wages paid in that classification or trade in the area.

In 1997, the Ohio legislature exempted “public improvements undertaken by, or under contract for, the board of education of any school district or the governing board of any educational service center” from the requirements of its prevailing wage law (Ohio Rev. Stats., § 4115.04 (3)). The exemption took effect on August 19, 1997. The exemption was part of a larger law that created a new Ohio School Facilities Commission and transferred responsibility for school construction financing from the State Board of Education to the new commission. The state also increased its financing for school projects.

Connecticut's prevailing wage law applies to school construction projects that exceed the applicable thresholds.

OHIO LEGISLATIVE SERVICE COMMISSION IMPACT STUDY

The 1997 Ohio law required the Legislative Budget Office of the state's Legislative Service Commission to monitor and study the effects of the prevailing wage exemption for school projects and, within five years after the exemption took effect, to report to the House and Senate leaders on:

1. the amount school districts and educational service centers saved from the exemption,

2. the impact of the exemption on school building quality,

3. the effect on the wages of construction employees working on school projects, and

4. other subjects as determined by the budget office.

The Legislative Service Commission (LSC) issued the required report on May 20, 2002 (S.B. 102 Report: The Effects of the Exemption of School Construction Projects From Ohio's Prevailing Wage Law).

Cost Savings

The LSC report estimates the aggregate savings on school construction from 1997 through 2001 at $487.9 million or 10.7%. It breaks this total down as follows:

The report highlights several difficulties in calculating construction savings attributable to the presence or absence of prevailing wage laws. Construction costs vary according to several factors, including size, location, specifications, and labor and material shortages or surpluses, and the overall state of the economy at the time of construction.

The LSC used two types of data to determine savings. One was a survey of contractors who bid on school projects that asked them to provide, among other information, their actual bid price and the hypothetical bid price they would have submitted if the prevailing wage law was in effect. The other was a regression analysis of school construction activity data from F.W. Dodge. (Dodge is a national information service for construction professionals. It collects data on public and private construction and measures the value of contracts awarded to private construction firms.) Although the report analyzes both the survey and the Dodge data, the savings estimates shown above appear to derive from a statistical analysis of the Dodge data and not from the surveys.

The LSC acknowledged several shortcomings with the survey results, including a low response rate and the failure of the hypothetical bids to take into account changes in the way contractors might have bid or organized the work if they had to pay a higher prevailing wage. The survey results indicated savings, if any, in the range of 5% to 10%, which is well under the 20% to 60% savings exemption proponents anticipated when the exemption was adopted. Estimated savings were slightly higher in rural than in urban counties and also decreased over time.

The LSC cautions that its savings estimates apply only “for the specific exemption of school construction in the Ohio economy between 1997 and 2001. The effect of an expanded exemption in a different economic environment may not necessarily be the same” (LSC, p.25).

Construction Quality Impact

The report's conclusion that the exemption does not appear to have decreased construction quality is based on two surveys of school superintendents about whether they perceive any differences in construction quality and whether they are satisfied with the results of the construction. The survey response rates were 31% for the first survey and 58% for the second. (Ohio has 611 school districts.)

The LSC report admits that it has no quantitative method for measuring the difference, if any, in school construction quality with and without the prevailing wage law. It also states “LSC was unable to measure, for example, the longevity or future maintenance requirements of the buildings being constructed by workers being paid less than prevailing wages” (LSC, p. 35, fn. 52).

Impact on Construction Wages

The LSC report concludes that the exemption had no “discernable negative effect” on construction industry employment levels or wages (LSC, p. 47). It bases this conclusion information from the U.S. Bureau of Labor Statistics reports on construction industry average weekly wages (both nominal and adjusted for inflation), average weekly hours, average hourly earnings, and employment levels.

The report notes that school construction in Ohio in 1997 was 5% of the total value of construction, 6.4% of the value of building construction, and 10.5% of the value of nonresidential construction. Because school construction is a relatively small part of the construction industry in the state, the LSC says, trends and effects in the rest of the construction industry in the state “may overwhelm any effects of the prevailing wage exemption” (LSC, p. 36).

WEISBERG RESPONSE TO LEGISLATIVE SERVICE COMMISSION STUDY

On July 8, 2002, Ohio State University Political Science Professor Herbert F. Weisberg published a paper questioning the statistical analysis underlying the LSC report's conclusions (“Analysis of Regression and Surveys in Ohio LSC Report on S.B. 102 on Claimed Cost Savings from Exempting School Construction from prevailing Wage Requirements”). According to his Ohio State faculty information page, one of Weisberg's academic specialties is research methods with a focus on survey research methods and scaling techniques.

Weisberg argues that the cost savings estimates in the LSC report are based on “flawed interpretations of statistical analysis.” He contends that LSC's own regression equations for the F.W. Dodge construction project data find that the prevailing wage is a statistically insignificant contributor to differences in overall construction costs before and after the exemption. Since it would not be possible for the large savings LSC estimates to be statistically insignificant, Weisberg concludes that the savings are not real and are the result of a flawed analysis. In addition, Weisberg says, the Dodge data show only estimated costs at the start of projects based on accepted bids, not actual construction costs. Final construction costs “can be considerably higher if the company lacks the expertise to keep the costs within the level of the bid.” Thus, Weisberg says, “the estimated cost savings are not relevant to actual project costs” (Weisberg, p. 1).

Weisberg also questions the validity of the LSC's survey data. He notes the low response rates to the quality surveys and points out, as the LSC also does in the footnote cited above, that “quality cannot be fully assessed over a short period of time, since the goal for the construction is to be of high enough quality to be useful for a long term” (Weisberg, p. 12). Regarding the contractor survey asking for hypothetical bids under prevailing wages, he says it is generally difficult for survey respondents to answer hypothetical questions. In addition, the response rate seems low and the survey gives nonunion contractors an incentive to overstate the prevailing wage price. The LSC report also states its concerns regarding both these issues (LSC, p. 18).

Finally, Weisberg criticizes LSC's use of the Bureau of Labor Statistics surveys on construction wages because they are large-scale national surveys and the LSC admits that the Ohio data in them does not reflect a representative sample of Ohio construction workers (Weisberg, p. 12; LSC, Appendix 4, p. 68).

JL:ro