Why project labor agreements are not in the public interest.

• Author: Tuerck, David G.
• Position: Report

Project labor agreements (PLAs) are agreements between owners of construction projects and construction unions, under which firms retained to work on a construction project must enter into collective bargaining with the unions, hire workers through union hiring halls, and pay union wages and benefits. Contractors must operate, in effect, as union contractors, whether they ordinarily use union labor or not. Workers must usually pay union dues whether they belong to a union or not.

PLAs do not preclude nonunion contractors from bidding on construction projects. But they authorize the unions to negotiate the wages and work rules under which a contractor (whether it uses union labor or not) must operate. For that and other reasons, nonunion contractors generally oppose PLAs. The construction unions favor PLAs, and with increasing vigor.

PLAs are common on major projects in both the private and the public sector. Disney World and the Trans-Alaska Pipeline, both private projects, were conducted under PLAs. The discussion here is focused on public projects, however. Both private and public owners are under pressure, politically, to enter into PLAs for major projects and, in negotiating a PLA, to agree to terms favorable to the unions. But public owners are governmental entities made up of elected officials or persons appointed by elected officials, who must answer to the unions, along with other powerful voting blocs, through the political process. Private owners are accountable, instead, to their shareholders. School boards and state transportation authorities do not have to make a profit on their projects. Builders of theme parks and pipelines do.

Thus, private owners can decide on whether to use a PLA on the basis of whether it reduces construction costs and thus increases profitability (that is, assuming the choice is freely made without the threat of union retaliation should the decision go against a PLA). Public owners, on the other hand, are constrained only by competitive bidding laws, the effectiveness of which is arguably compromised by the very decision to enter into a PLA. Elected officials have an incentive, in contracting for projects, to weigh political concerns against cost effectiveness. A decision by public owners to enter into a PLA is a signal that they are willing to subordinate the interests of the general public to those of the unions. Private owners do not have the luxury, or the incentive, to subordinate the interests of their shareholders in a similar manner.

By one estimate, PLAs add 12-18 percent to the cost of public projects. PLAs were banned from federally funded construction projects throughout the administration of President George W. Bush. Had they not been banned and had they been applied to major federal construction projects conducted in the final year of the Bush administration, they would, by the same estimate, have increased construction costs by $1.6 to $2.6 billion in that year (Tuerck, Glassman, and Bachman 2009).

Unions in Decline

Despite the clout they continue to exercise in the political arena, the power of unions to set wages and to control work rules is declining. Over the period 1983-2008 (the period over which comparable data are available), there was a decline in the percentage of workers who belonged to unions and a decline in the wage "premium" earned by union members. The fraction of all U.S. workers who belonged to unions fell from 20.1 percent in 1983 to 12.4 percent in 2008 (U.S. BLS 2009a). In 1983, union workers earned 38.2 percent more than nonunion workers. By 2008, this wage premium had shrunk to 28.2 percent (BLS 2009b).

Among union members, construction workers experienced a similar trend. According to one report, 87.1 percent of construction workers belonged to unions in 1947 (Baskin 1998). The percentage was 27.5 percent in 1983 and 15.6 percent in 2008 (BLS 2009c). The wage premium earned by union construction workers fell in tandem with that earned by all union workers. In 1983, union construction workers earned 74 percent more per week than nonunion construction workers. In 2008, they earned 51.8 percent more (BLS 2009d).

The decline in the wage premium for construction workers has been particularly steep. In 1983, construction workers, both union and nonunion, earned 14.2 percent more per week than all workers, both union and nonunion. In 2008, they earned 1.4 percent less. The decline in relative earnings has been steep also among union members. In 1983, union construction workers earned 31.2 percent more per week than all union workers. In 2008, they earned 14.5 percent more.

The increasing energy with which the construction unions and some elected officials encourage PLAs on public projects can be seen as an effort to shore up the construction union wage premium against further decline. The question, however, is whether there is a legitimate public policy reason for slowing or reversing this decline. The construction unions, as shown below, seldom couch their pleas for PLAs in terms of their own interests. But, political considerations aside, an elected official might see these trends as arguing for policies that would strengthen the unions, to the end of protecting workers' wages from erosion brought about by declining union influence.

The idea that it is possible to help construction workers in general by helping union construction workers in particular is, however, contradicted by the evidence. First, the decline in the union wage premium does not necessarily reflect a general decline in wages. The reason that the union wage premium has declined is that nonunion wages have risen faster than union wages, for both construction workers and all workers. It could well be that some workers have abandoned the unions because wages are growing baster for nonunion workers. Moreover, and as shown below, the very process by which the construction unions keep union wages higher than nonunion wages consists in part of using PLAs to exclude nonunion contractors and workers from public projects.

Finally, the construction industry is far too fractionalized for even the biggest construction finns to be able to suppress wages or impose onerous work rules. According to the U.S. Census Bureau (2007), there were 725,101 construction establishments in the United States in 2007. There were, on average, 10 workers per establishment. According to the U.S. Small Business Administration (2009), 86.1 percent of construction firms have fewer than 500 workers.

Thus, it is not possible to rationalize a decision to use a PLA as an action that will improve the lot of the average construction worker. PLAs benefit only the minority of construction workers who belong to unions.

The History of PLAs

PLAs date back at least to 1938, when work began on the Shasta Dam in California. In those days, when strikes against public projects were commonplace, a contractor taking on a project of this size would see a PLA as a method of maintaining labor peace. In view of the tumultuous labor conditions at the time, the general contractor under the Shasta Dam PLA readily agreed to hire only members of the recognized unions and to pay union scale, The project was completed without incident, apparently as a result of this arrangement (Johnston-Dodds 2001).

Although the construction unions remained strong for decades thereafter, the short-term nature of most construction work posed an obstacle to union organization efforts. Thus, in 1959, a sympathetic Congress passed the Landrum-Griffith Act (U.S. Labor-Management Reporting and Disclosure Act of 1959), which made it possible for unions to represent workers even when only a minority of workers actually belonged to unions. By passing this act, Congress opened the door to "pre-hire" agreements, such as PLAs, that owners could sign with the unions before any workers were hired.

Another milestone was reached when the U.S. Supreme Court upheld a 1987 PLA entered into by the Massachusetts Water Resources Authority, which was conducting a court-ordered cleanup of the Boston Harbor. The MWRA had required its contractors, whether union or not, to accept the terms of the PLA. The court decision (Building Trades v. ABC 1993) constituted a serious rebuke to efforts by nonunion contractors to mount a legal challenge to PLAs.

The Boston Harbor decision prompted the construction unions to press more aggressively for PLAs on public projects. Boston's "Big Dig," which became the most expensive (and probably the most controversial) public works project in U.S. history, was conducted under a PLA.

By the time that the Court approved the Boston Harbor PLA, however, the decline of the construction unions was in full bloom. Indeed, the decisions to use PLAs on the MWRA project and the Big Dig can be seen as made in order to protect Boston area construction unions from further decline.

By the 1980s, a PLA was no longer necessary in order to get a contractor to take on a major public construction project, as it had been as with the Shasta Dam project. Instead, PLAs had become arrangements into which public owners would enter out of deference to the unions, whose political significance had come to outweigh their ability to pose a threat to labor peace. As one union official (Northrup and Alario 1998: 21) put it, the purpose of PLAs now was "to fight the growing nonunion element throughout the country."

This growing nonunion element was not the result of any organized attempt by nonunion contractors to diminish union power. It was the result, rather, of shifting living patterns and of technological progress.

One author who has analyzed these changes points out that the post-World War II housing boom shifted a large portion of the construction business from the cities, where the unions are strong, to the suburbs, where they are not. Over the years, "standardized factory-made sub-assemblies, materials, and fasteners, ... specialized tools and techniques...