Productivity Enhancement: Project Labor Agreements

1. Background

What is a PLA?
How are today’s PLAs different?

Old school PLAs
Stop-loss PLAs
Market-share PLAs
But why the controversy?
What do we know about the effects of PLAs?
PLAs and bidding
PLAs effect on bid price
PLAs and human resource outcomes: compensation, strikes, safety and minority employment
Conclusions

What is a PLA?

Project labor agreements are primarily agreements, so we need to know what is being considered and agreed upon and by whom. PLAs are project- specific, collectively-bargained labor agreements regarding wages, benefits, hours of work and other terms and conditions of employment. On the one side of the agreement is a collection of construction unions perhaps under the leadership of a local construction labor council or some other form of multicraft organization. On the other side of the agreement is usually a project or construction manager representing the interests of the construction user. This contrasts with typical collectively bargained labor agreements in construction where separate craft unions bargain with their corresponding contractor associations about wages and working conditions. Traditional collective bargaining has no specific construction project in mind, and no one at the table controls upcoming work. In PLA bargaining, unions bargain as a group with someone who controls upcoming work.

In typical construction collective bargaining, the electricians might look over their shoulders to see the outcome of the plumbers’ negotiations, and the laborers are going to keep in mind what the carpenters are getting. But there is no formal structure or binding agreement in traditional, craft-separated collective bargaining to ensure that the various contracts signed in a local area by the various crafts and contractor groups will have similar holidays, similar hours of work, similar drug testing provisions, etc. or even similar contract expiration dates. A PLA provides the legal structure whereby everyone can (if they so choose) get on the same page regarding all of the issues.

The fact that through the project manager the construction user is on the other side of the table also makes PLAs different. In traditional collective bargaining in construction, contractors are on the other side of the table. Users have something to bargain with that contractors do not have. Users have the work: they have the project under consideration. Individual contractors have to bid to win work. Contractors as a group have a higher prospect of someone in their group winning the project, but if the economy turns sour, chances of getting the job diminish. As long as the project goes forward, the construction user has the work, and on large projects that work could last for years. Through traditional collective bargaining, users bring something of value to the table, something worth bargaining over.

With PLAs, construction users can (and often do) bargain their control of work in exchange for union concessions relative to the existing set of local labor agreements. Rarely do these concessions involve lower wages and benefits.More commonly, in an effort to harmonize the terms and conditions of work across trades, some trades have to make concessions to mirror terms and conditions in another trade’s contract. The fact that the user has the work and is willing to provide it in exchange for such concessions may motivate a trade’s willingness to compromise on working conditions. Sometimes a user may convince all the trades to make an across the board concession in exchange for the job. In one case, a bridge contractor signed a PLA with the various relevant trades for long term work on a major bridge reconstruction project in exchange for altering all the unions’ overtime provisions, so the project could proceed without overtime pay in off hours to avoid backing up traffic. Under traditional collective bargaining with no specific consideration to a specific project, such a concession would not make much sense to any union and to obtain this concession across all unions would be impossible. A PLA made it happen.

In one sense, all PLAs are across-the-board concessionary contracts because, universally, all PLAs have no-strike clauses in effect through the entire duration of the project. For long-lasting projects, these no-strike clauses are meaningful because inevitably in a two or three year period, one or more traditional union contracts will expire, leading to the possibility of a negotiation stalemate and a strike. PLAs take the user’s work off the traditional collective bargaining table and insulate it from strikes. This can be very important to the user who has a vital completion date. So the construction user comes to the PLA bargaining table ready to exchange work for harmonized working conditions, occasional project-tailored terms and conditions, and a guaranteed uninterrupted labor supply through the duration of the project. Only PLAs can get all of this done with multiple craft unions, multiple contractor associations and differing contract expiration dates. In short, PLAs bring new players to the table and thus create the possibility of bargaining to new win-win solutions.

What is in a PLA for unions besides various possible concessions? In a word: work. PLA projects tend to be large and long-lasting. In private sector PLAs, the work is what the unions bargain for, and that is what they get because private sector PLAs typically restrict bidding to union contractors. On public sector work, restricting bidders to union contractors usually violates public procurement rules. Nonunion contractors are allowed to bid on public PLA jobs. Nonetheless, when working on a covered project, all contractors (including nonunion contractors) agree to abide by the terms of the PLA as well as any provisions of local agreements that are specifically referred to in the PLA or not limited by the PLA. The means of assuring this compliance by all contractors is a letter of assent the PLA requires.

The following letter of assent comes from a Missouri PLA and is typical: Pursuant to Article II, Section 1, Paragraph 3, of the above-referenced Agreement, the undersigned contractor hereby agrees that it will be bound by and comply with all terms and conditions of said Project Labor Agreement, and any amendment thereto for this Project only. This Letter of Assent will remain in effect for the duration of the Agreement, and any extensions, after which this understanding will automatically terminate, except as provided in Article II, Section 6 [concerning repairs and rework] of the Agreement.

As a practical matter this means that all contractors usually agree to use union referral mechanisms (e.g. hiring halls), pay union scale, contribute to jointly administered (i.e. union sector) benefit programs and, in general, operate as union contractors while on a project—whether or not they are usually union contractors. Sometimes PLAs have key worker provisions that allow nonunion contractors to use a limited number of key nonunion workers. Occasionally, nonunion workers are permitted to apply to the project manager for work rather than go through the union hall. But the basic point is this: through PLAs, unions exchange concessions for work. If the PLA cannot deliver at least most of the work, the construction user has nothing to bargain with.

There are two players not at the PLA bargaining table—the union contractor and the nonunion contractor, both of whom might end up working on a public PLA project. From the perspective of traditional collective bargaining, PLAs are a topsy-turvy world. Usually the union agrees with the contractor, and then the contractor goes out and finds the work. Under a PLA, the unions, as a group, go out and find the work. Wages and benefits are set. Then, on private jobs, union contractors bid for the project and, on public jobs, all contractors willing to abide by the terms of the PLA bid on the project. Union contractors get a level playing field, but that is all.

The other absent player is the nonunion contractor willing to pay the PLA wage rates and abide by the terms and conditions of the PLA. These participating nonunion contractors stand on the sidelines along with the union contractors until the project is let out for bid. Technically, PLAs are prehire agreements because the terms and conditions of work are agreed upon prior to the hiring of workers. But, effectively, PLAs are usually also prebid agreements because the terms and conditions are set prior to any bidding on the project.

And, of course, there is one absent non-player— contractors unwilling to bid on the project because of the terms and conditions of the PLA. These, typically nonunion contractors, may not be able to compete with the higher labor productivity called forth by the PLA wages. They may not wish to expose their key workers to union workers. They may not wish to have their non-key workers go through the hiring hall to get work. They may philosophically object to PLAs. They may have other reasons for not participating. In any case, nonunion contractors’ nonparticipation may lower the number of contractors who bid on a PLA project. Alternatively, the presence of a PLA may attract contractors who otherwise might not bid on the project. The effect of PLAs on the number of bidders is an open empirical question that chapter four addresses.

Because PLAs set wages and benefits close to or at the local union rates, PLAs probably encourage contractors to shift towards capital intensive and high skill construction strategies. PLAs may also alter the composition of contractors shifting towards more heavily capitalized firms. Some public entities, restricted in their ability to pre-qualify contractors by public procurement regulations, may be attracted to PLAs, in part, due to the way PLAs probably sort through potential bidders shifting the mix towards more established, capital intensive and skill oriented contractors.

Thus, PLAs are first of all agreements where unions, as a group, bargain for work from construction users in exchange for concessions on strikes and working conditions. Until the PLA is signed, contractors sit on the sideline. Once signed, union contractors know that even their nonunion competitors will have to pay the same wages and benefits. Nonunion contractors may be excluded entirely from private projects but on public works they are still players. Some, however, will withdraw not wanting to agree to the terms of the PLA. Both union and nonunion high-wage/high-skill contractors are likely to be attracted. Whether ultimately PLAs discourage more bidders than they attract is an empirical issue, but some public construction users may be partially attracted to PLAs based on what type of contractor is attracted and what type of contractor is repelled by PLAs.

How are today’s PLAs different?

Old-School PLAs

From the first major use of PLAs to around 1980, PLAs were generally restricted to a particular and relatively unusual type of construction project— the large, long-lasting, typically complex and often rural construction project. Construction users bringing these projects to market faced three problems. First, if the project was rural (such as a hydroelectric dam located where the water was or a coal-fired power plant located where the coal was), the size of the project was likely to overwhelm the capacity of the local construction industry and labor market. By having a PLA, the construction user could create regular and known wages and working conditions needed to attract workers from far away.

Second, if the project was specialized and complex (such as a nuclear facility), the skill requirements of the job might overwhelm the local labor market even in a non-isolated area. A PLA would provide ready access to distant union workers again by establishing appropriate wages and conditions and by invoking the union system of using skilled traveling workers.

Third, if the project was long-lasting (say three or more years), and schedule and completion were important to the user, a no-strike provision in a PLA would insulate the project from labor/management conflict during the bargaining between local craft unions and their corresponding contractor organizations. Whatever work stoppage or lockout might occur through the normal operations of collective bargaining would not affect a PLA project. In short, bargaining impasse would not interrupt the PLA project.

So PLAs for many years were a specialized and relatively rare construction contract designed to obtain a ready and qualified supply of labor to large, complex and long-lasting projects.

Stop-Loss PLAs

In the 1980s, PLAs took on a new role. The downturn in construction in the 1980s was very sharp. Price competition (as opposed to quality or scheduling competition) is most intense when an economy slows and customers are more price-conscious and less concerned about timeliness or even quality. This environment favored nonunion contractors. But in order to keep some of the union sector’s biggest and best industrial customers and stop the loss of jobs, PLAs were written that contained wage and benefit concessions. American manufacturers facing severe overseas competition on both price and quality terms needed quality infrastructure built at the lowest price possible. PLAs became a way of delivering quality work at low prices to demanding customers. These PLA based wage cuts were partially offset by the promise of steady work for an extended period of time during a period when construction work was anything but steady. The PLAs in the 1980s traded lower wages for longer work. Thus, it was possible, in part, because the agreement was with a user who had work to exchange for concessions in wages and conditions.

Market-Share PLAs

In the 1990s, however, the construction economy improved, leading to a decade long boom that has recently slowed but not collapsed. Union workers were working; local union unemployment rates were low, and the attractiveness of trading hourly wages for more assured work faded. But PLAs did not fade. In fact, they proliferated primarily in areas where construction unions were relatively strong but even in areas where union coverage was low. And the new PLAs were often used on more modest projects, such as schools and court houses, and cover renovations as well as new construction.

Two economic conditions (other reasons will be discussed below) converged to lead to the proliferation of PLAs. First, construction labor markets were becoming increasingly tight. Not only was unemployment down, but also apprenticeship training was down. As the nonunion sector proliferated in the 1980s, union apprenticeship programs reduced their enrollments or even in a few instances shut down. The nonunion sector did not fill the gap, in part, because they were happily harvesting union-trained workers in need of jobs, and because the nonunion sector had not been able to find a viable alternative to collective bargaining to finance apprenticeship training. So construction users were hungry for available and qualified craft construction workers. The Business Roundtable, a group of large construction users, stated in an analysis of skill shortages in construction, “The union sector has always excelled in craft training through the joint labor/management apprenticeship programs…the open shop, as a whole, has not supported formal craft training to the extent necessary.” ³

Second, while the construction economy had recovered and construction union membership was growing, the union share of the construction labor market was either still declining or merely stabilizing, depending on the area. PLAs emerged as a new key instrument for both providing users with an uninterrupted supply of qualified workers and in helping unions to stabilize or expand their share of the construction market.

But why the controversy?

Old-school PLAs were used with little controversy in both the private and public sectors throughout the postwar period—a period during which much of the construction sector was highly unionized.With strong unions, there was a great desire on the part of construction users and contractors to avoid labor disputes and to gain the best economic deal possible relative to local agreements. The climate changed, however, when union market share dropped and construction users and the nonunion sector became better organized.⁴ In the new environment, with large nonunion contractors able to compete for all types of work in virtually every state and with the growing strength of a nonunion contractors’ association, Associated Builders and Contractors (ABC), challenges to PLAs became more common. In the past decade, all branches and levels of government have been dragged into the PLA debate.⁵ It is probably not an exaggeration to say that ABC has challenged nearly every large public sector PLA that has been proposed during the past ten or twelve years.

Two state court cases To give two examples of state court decisions, in the consolidated case of New York State Chapter, Associate General Contractors v. New York State Thruway Authority (666 NE 2d 185, 151 LRRM 2891, N.Y. Court of Appeals, March 28,1996) the New York Court of Appeals upheld the use of a PLA on the renovation of the Tappan Zee Bridge, but overturned the one attached to the construction of dormitories at the Roswell Park Cancer Institute. In Associated Builders and Contractors of Rhode Island v. Department of Administration (787 A2d 1179, 170 LRRM 2054, R.I. Supreme Court, January 4, 2002) the Rhode Island State Supreme Court overturned a PLA for a new sports facility at the University of Rhode Island. In the former case, the court held that New York law does not prohibit nor absolutely permit PLAs but does require that there be an adequate reason to apply a PLA to a project and further requires that sufficient analysis be done to determine whether a PLA advances the purposes of the state’s competitive bidding statute. For the Tappan Zee Bridge, the Thruway Authority had determined that the need for quick completion and labor peace supported the use of a PLA. The authority also found that it would save over $6 million by using a project agreement (as opposed to operating under local contracts). However, in the dormitory case, the state agency had already begun the project without a PLA. Later, it attached one to the project without doing any serious analysis of the benefits. The court voided that PLA stating that the agency had failed to “consider the goals of the competitive bidding statute.” The facts of the Rhode Island case are somewhat similar to those of the New York dormitory case. The University of Rhode Island had already begun construction of a $73 million basketball and ice hockey facility.Work on the project involved 34 separate bid packages. Six bids had been awarded with no mention of a PLA. But in the fall of 2000, more than one year into the project, a PLA was signed. Immediately thereafter, fourteen additional packages went out to bid requiring adherence to the new agreement. The Rhode Island Supreme Court found that the PLA violated state law. The court wrote (170 LRRM at 2060): [We] are of the opinion that an awarding authority may include a PLA as a bid specification in a public contract, but the awarding authority may do so only after it has established that (1) the size and complexity of the project are such that a PLA supports the goals and objectives of the state purchases act, and (2) the record demonstrates that the awarding authority has conducted an objective, reasoned study using reviewable criteria in determining that the adoption of a PLA helps achieve the goals of the state purchases act. Since the sports facilities were nearly complete, the court let the project go forward and did not award any damages to the plaintiffs.

However, not all challenges have resulted in the outcome sought by PLA opponents. A watershed event was the 1993 United States Supreme Court decision in the so-called Boston Harbor case.⁶ Although the case dealt with the narrow question of whether local public sector PLAs should be preempted by the National Labor Relations Act, the unanimous court decision allowing a Massachusetts water resources board to go ahead with its PLA bolstered the efforts of proponents to seek agreements on a wide range of public projects.

Viewing market-share PLAs as a threat to their members’ market position, the ABC and its state affiliates have mounted intensive national and local campaigns to oppose the use of PLAs. This effort has included numerous court cases, media campaigns and lobbying efforts.⁷ Most of the legal action since Boston Harbor has concerned bidding statutes and ordinances and if PLAs, since they place conditions on successful bidders and arguably limit the number of bidders, violate either the letter or the spirit of such laws. Court decisions have been mixed.⁸ In a number of cases, state courts have refused to overturn PLAs, while in other cases they have found that a particular PLA did violate a bidding statute.

The situation at the federal level, however, is different. One of President George W. Bush’s first actions in office was to reverse altogether a Clinton administration’s policy encouraging PLAs. On February 21, 2001, the President issued Executive Order 13208 prohibiting the federal government or a construction manager acting on its behalf from placing in its bid specifications any language that denotes the following:

1. Require or prohibit bidders, offerors, contractors, or subcontractors to enter into or adhere to agreements with one or more labor organizations on the same or related projects
   
   
2. Otherwise discriminate against bidders, offerors, contractors or subcontractors for becoming or refusing to become or remain signatories or otherwise to adhere to agreements with one or more labor organizations, on the same or related construction projects

The President amended the order on April 6, 2001 to exempt agreements that had already been entered into. And Executive Order 13208 allows successful bidders to enter into PLAs voluntarily, but it prohibits the mandatory acceptance of a PLA as a condition of bidding. The result is that PLAs are not currently being applied to most federally funded projects. This has not, however, slowed their use in the private sector nor on public projects that use only state or local funds. It is not possible to determine precisely how many PLAs are in effect at any time, nor how many are public sector and how many are private sector. However, based on findings in previous research, it is likely that at least three-quarters of PLAs are private sector.⁹ Therefore, Executive Order 13208 may have only a small effect on the overall use of such agreements. Nevertheless, market-share PLAs are controversial because they involve a struggle between union contractors, high-wage nonunion contractors and lowwage nonunion contractors over market share in the public sector.

What do we know about the effects of PLAs?

The controversy over PLAs has spurred research on the effects of PLAs on a variety of issues, including the number of bidders on a project, labor costs and final bid price. Unfortunately, much of the research is of low quality and has originated from organizations or individuals with a clear prior position. This research typically relies on anecdotes and spurious comparisons. For example, ABC’s Union Only Project Agreements: The Public Record of Poor Performance discusses eighteen projects on which there were cost overruns. Of these, six are described as union only projects but are not PLAs. No attempt is made to compare a sample of PLA and non-PLA projects.¹⁰

Some of the research, however, is a bit more sophisticated. Two important topics that have been examined by researchers are the effects of PLAs on the number of bidders on a project and the ultimate effect of a PLA on project cost.

PLAs and bidding

The research on bidding can be divided into three categories: studies that compare the number of bidders on PLA and non-PLA projects, those that look at the union/nonunion mix of contractors on PLA projects and those, based on survey research, that gauge the likelihood of nonunion contractors bidding on PLA projects.

The Empire State Chapter of ABC, in studying construction at the Roswell Park Cancer Institute in New York concluded that packages put out to bid without a PLA stipulation received 21% more bids than projects with a PLA attached.¹¹ Andrews, the General Accounting Office (GAO); and Opfer, Son and Gambatese all examined participation by nonunion contractors on PLAs.¹² Andrews studied the Boston Harbor project and found that nonunion participation was lower than reported by the construction manager. He also found that less than half of the nonunion contractors were supplying construction services, with the remainder involved in material supply or professional services. A study of a project run by the South Nevada Water Authority, Opfer, Son and Gambetese concluded that between 16% and 33% of contractors were nonunion and one percent to 27% of the volume work was done by nonunion contractors. The authors interviewed representatives of two nonunion firms that had worked on the SNWA project but indicated that they would not work on PLA projects again. Among the problems cited by the firms were jurisdictional disputes among unions, poor performance by union workers and obligations to support union sector benefits funds. The GAO’s study found that 86 of 286 contracts on the Idaho National Engineering Laboratory were awarded to nonunion contractors, despite eight of eleven nonunion contractors telling the GAO that they would not bid on the project because of the PLA provisions.

All of the studies cited above have problems. For example, the ABC study failed to account for differences in the types work covered and not covered by PLAs at the Roswell facility, and Andrews’s sample is much too small to produce valid, statistically significant results. However, a more important question is the relationship between the number of bidders and project cost. In two studies in New York State, Carr found that project costs fall between 3.2% and 3.8% for each additional bidder.¹³ However, Carr’s statistics show that his model accounts for only 11% of the variance in project costs, suggesting that a number of possibly critical variables are not included in his analysis. If important variables are excluded, effects may incorrectly be attributed to the number of bidders that when, in fact, other causes are at play.

PLAs affect on bid price

One stream of research simply looks at the direct effects of PLAs on bid price regardless of the number of bidders. Research conducted by the Beacon Hill Institute (BHI) at Suffolk University in Boston has been widely reported. In 2003, BHI conducted two studies of school construction projects in the Boston area. In 2004, it replicated its research in Connecticut. In all of the studies, BHI reported substantial cost premiums associated with PLAs. In the original Boston study, the researchers found that PLAs increased school construction costs by 17.3% or about $31.74 per square foot. A follow-up study on a larger sample pegged the estimate at 14% or $18.83 per square foot. The Connecticut study estimated that PLAs added about thirty dollars per square foot to costs.¹⁴

More detail resides in later sections; however, in brief, the BHI team did an insufficient job at controlling for variables that affect construction costs. Hence, much of what was attributed of the presence of a PLA is actually explained by other variables, such as project location (e.g. the inner city) and building amenities (heating systems, swimming pools, etc.).

PLAs and human resource outcomes: compensation, strikes, safety and minority employment

Two studies examine the impact of PLAs on wages. In the GAO paper on the INEL project, researchers found that wages on the project were 17% to 21% higher than the Davis-Bacon prevailing wage rates for the area. In a 1997 article, Lyons argued that the executive memorandum issued by President Clinton to encourage the use of PLAs on federal construction projects would raise federal construction costs between 2.3% and 7.2%.¹⁵ In the GAO piece, however, most of the difference was accounted for by the travel allowances included in the agreement, and the critical problem with Lyons’s calculation is that he used the national average construction wage as a proxy for the Davis- Bacon rate.

Several studies have addressed the complaint by nonunion contractors that PLAs force them to pay into the union sector benefits funds while maintaining their own pension and health care plans.¹⁶ Lund and Oswald point out, however, that this argument may be more theoretical than actual, since many nonunion workers lack any benefit coverage. ¹⁷ Either their employers do not offer coverage, or the short tenure of nonunion workers precludes their participation in benefits’ programs. It is also the case that participation would be governed by the PLA and could vary from agreement to agreement (see, for example, the Toyota agreement discussed in Chapter Five).

A central feature of PLAs is the inclusion of a no-strike/no-lockout clause. In research done by Johnston-Dodds in California, 26 of 59 reviewed PLAs contained blanket no-strike provisions, while the remaining 33 allowed strikes only in the event of contractor delinquency in payments to joint funds.¹⁸ PLA proponents champion such provisions as an important element in raising certainty on construction projects.

Opponents discount such provisions on several grounds. First, they note that no-strike provisions have been violated (though proponents counter that dispute settlement procedures have been highly effective in quickly resolving problems). Second, PLA opponents point to the generally low strike rates in construction today. And, finally, they note that such disruptions are rare on nonunion worksites.

Available research on safety is, for most part, restricted to two case studies: work done by Dunlop on the Boston Harbor project and Opfer, Son and Gambatese’s work on the SNWA project.¹⁹ Dunlop found that lost time incident rate on the Boston Harbor Project was 4.1 while the national average for heavy construction was 6.2. Further, the lost workday incident rate was 134.7 for Boston Harbor versus a national heavy construction rate of 150.4. Opfer, Son and Gambatese, however, found contrary evidence when examining the SNWA project.

Finally, the research on minority (including female) employment is also sketchy and primarily anecdotal. PLAs have been opposed by a number of minority contractor associations. However, membership in such associations is likely dominated by nonunion firms. In addition, ABC argues that the emphasis placed on minority employment by PLA proponents is designed to “deflect criticism of unionized construction emanating from minority and women’s groups.”²⁰ Johnston-Dodds provides perhaps the most interesting description of a minority employment program in her description of the Port of Oakland, California PLA.²¹ The agreement included a small/local business utilization program and a local hiring program, which provided for set-asides and targets for minority contractor and worker participation. The PLA also called for a social justice committee to oversee implementation of the minority hiring provisions. The social justice components of the PLA were supported by a contribution of up to $1.15 per hour for all work done under the PLA. Although some difficulties were mentioned in meeting some of the PLA’s goals, the report does not contain an analysis of the overall effectiveness of the program.

Conclusions

A PLA is an agreement between a multicraft set of labor unions and a construction user represented by the project manager or some other agent qualified to sign a labor agreement. Bringing new parties to the table—a user who controls work and a combination of unions who can collectively harmonize their local labor agreements—creates new bargaining possibilities, and new win-win solutions become possible. PLAs fall into three historical categories.

Old School PLAs were dominant from WWII to around 1980. They were large, long-lasting, often technical or rural projects that needed to draw workers from long distances and proceed uninterrupted by strikes in an environment with widespread unionization. PLAs set the wages, conditions, traveling arrangements and no-strike clauses that made these goals possible.

Stop-Loss PLAs emerged in the 1980s in response to stagnation in the construction labor market and loss of work to the nonunion sector. These concessionary PLAs granted primarily to large industrial owners discounted local union wages and benefits to preserve work. Neither PLA was particularly controversial for its time except for those union members who objected to the concessions embedded in Stop-Loss PLAs.

Modern Market-Share PLAs are applied to a wide range of private and public projects attracting owners based on new win-win possibilities associated with a new bargaining table. Market-Share PLAs are controversial because these contracts serve as weapons in the struggle between union and some nonunion contractors (those who cannot or will not compete for PLA work) over market share.

While most PLAs are on private work, the controversy over PLAs is focused on public work: if a private owner wishes to sign a PLA, there is no public policy that would stop the owner doing so. Consequently, the debate is over whether PLAs are good for the public sector. Thus far, most of the debate has been on whether PLAs raise public construction costs. Analytically, this is a delicate argument to make because most Market-Share PLAs exist where unions are strong and public works require prevailing wages and those wages (and benefits) tend to correspond to the wages and benefits required by PLAs. So the argument must be that PLAs restrict bidders, thus reducing competition and raising prices. The problem with this argument is one need only about half a dozen bidders to get the full effect of bidding competition on prices. Furthermore, research to date only looks at whether nonunion contractors are discouraged and not whether union or high wage nonunion contractors are attracted by PLAs. In short, we do not know whether or to what extent PLAs discourage bidding. Nonetheless, some research has argued that PLAs raise total costs on prevailing wage jobs by around 15%. This is not only a surprising result because it cannot be derived from increased wages, but also because labor costs as a percent of total costs typically is around 30% in construction.

Readers should not be dismayed at the preliminary, incomplete, and often inadequate results of research on PLAs. This field of research is young, and from the heat of current controversy there may yet emerge information. Some of the problems with prior work simply reflect the inherent difficulties with this type or research (e.g. getting adequate data, comparing very different projects). In other cases, results are compromised by low quality research, including poor statistical modeling. Perhaps the most disheartening weakness is that some studies simply attempt to support a previously held position, with findings merely leading to a foregone conclusion. Nonetheless, this research literature will mature, become more sophisticated and solve some of its methodological problems, and thoughtful conclusions will drive out preconceived notions. This study is an attempt to contribute to that maturation process.

Back to main content menu