A new study that concludes California school construction projects that use project labor agreements cost from 13% to 15% more is being disputed by a number of data researchers and union trade councils.

“Measuring the Cost of Project Labor Agreements on School Construction in California,”
by Vince Vasquez, Dr. Dale Glaser and W. Erik Bruvold of the National University System Institute for Policy Research (NUSIPR), San Diego, was announced via an independent News Wire story that appeared on the ENR California website last month. The study was funded by the Associated Builders and Contractors, the national trade association representing merit shop contractors and subs.

Without getting into comparisons of some heavy-duty statistics and data analysis, let’s conclude a few things. First, the study was funded by an anti-PLA group (on its website, ABC said it “strongly opposes government-mandated project labor agreements on public construction projects”). Researchers are often paid to come to conclusions they were paid to come up with, unless their findings are mandated to be objective.

Second, the study says it examined 551 schools in 180 different districts, which is four times larger than any other study. What comes to mind here is that the researchers had to adjust (or didn’t) to so many variables when coming to any conclusion. Is a five-story, structural-steel campus project with an underground parking garage situated near an earthquake zone comparable to a one-story, wood-framed elementary classroom? Plus, PLAs require prevailing wage rates, which vary categorically across the state, either adding or reducing the cost of the project.

The study’s press release singled out PLA projects for the Los Angeles Unified School District, and although there have been some distinct contractual no-no’s reported over the years, the majority of its projects were completed on time and on budget. However, the NUSIPR concluded that inflation-adjusted costs in the LAUSD were $312 per square foot while schools built outside of the district, many in equally urban settings, were $221. Again, comparing a complicated school project that uses a PLA to keep costs, delays and change orders in control with maybe a renovation of a cafeteria is just not comparable.

The Los Angeles/Orange Counties Building and Construction Trades Council obviously sees PLAs differently. In providing the Exposition Metro Line Construction Authority with some PLA background for its upcoming contracts, the council says its PLA with the LAUSD covers $27 billion of bond funding for new construction and renovations. It said the average of 38% of the hours worked have been performed by tradesmen that reside within the jurisdiction of the LAUSD, with 67% of all yearly hours worked having been performed by residents of Los Angeles County.

Besides local hire provisions, the council also touted its requirement of utilizing Small Business Enterprises (from fiscal year 2003 through 2010, $3.7 billion, or 47%, of the total contracts awarded to SBEs) and had 905 residents from the local communities accepted into a state-approved apprenticeship program sponsored by the Building Trades Unions.

The NUSIPR study also quoted the findings of Dr. Dale Belman, a professor at the School of Human Resources and Labor Relations at Michigan State University, regarding his 2005 study, “Project Labor Agreements’ Effect on School Construction Costs in Massachusetts.” The NUSIPR concludes that Dr. Belman and his research colleagues prove that PLA schools cost more. Dr. Belman, however, says that’s not the case and the NUSIPR study’s data actually confirms his conclusion of the opposite.

“This is a very bad study,” said Dr. Belman soon after he sent a lengthy rebuttal letter to the NUSIPR. “After reading it thoroughly, it actually supports the point of view that PLAs do not affect costs.”

He suggests that perhaps the only way to judge the NUSIPR findings is to release the data publically in a peer-reviewed journal, such as Industrial Relations, The Industrial and Labor Relations Review or Economic Inquiry.

We’ll see how this plays out.