When I started writing popular articles about economics, one of my major motivations was to demonstrate to my undergraduate students that it didn’t require advanced degrees in economics for them to competently evaluate economic policy issues and claims. They could get a long way simply by using careful, logical thinking, and consistently-applied basic economic principles, because those tools reveal most of the errors. I thought of Frédéric Bastiat and Henry Hazlitt as inspirations.
Over 30 years later, applying that approach a multitude of times has convinced me even more strongly of the value of that focus. And despite my early fears that I would run out of issues to write about, that has not happened. It seems there is a virtually inexhaustible supply of poorly thought out policies and proposals.
Project Labor Agreements (PLAs) provide a good example, and the possibility of over $1.5 trillion in new infrastructure spending on the horizon means that there are plenty of zeros at stake in getting our understanding right.
A PLA is a pre-hire collective bargaining agreement which establishes the terms and conditions of employment for a specific construction project. A government-mandated PLA is an agreement typically drafted by unions with no input from non-union contractors that governments require both union and nonunion contractors to follow prior to accepting bids on a public construction project. PLAs establish the terms and conditions of employment that will be required for all construction workers employed on a specific project. Typical PLA terms include union representation and mandatory membership or union dues for all workers (including those who work for non-union employers). PLA terms also mandate following union work classifications and rules, including hiring all workers from union hiring halls, hiring all apprentices from union apprenticeship programs, and contributing to union benefit and multi-employer pension plans (that few, if any, non-union members will get a penny from, in the long run).
Unions claim that PLAs are justified because they generate lower project costs, while increasing safety and quality, “leveling the playing field” for competing bidders, buying labor peace, and guaranteeing that projects are completed on time. Defenders put plenty of resources into reiterating those claims and into “studies” that back PLA’s superiority by “friends” that have never met a union assertion they didn’t like.
Relying on such claims, the federal government has permitted, and even encouraged, state and local governments to require PLAs for projects utilizing federal money. (See President Obama’s Executive Order 13502, which encourages federal agencies building federal construction projects exceeding $25 million to mandate PLAs on a case-by-case basis, and allows state and local governments to mandate PLAs on federally-assisted projects.) Some Democrat-controlled state and local governments have enacted mandatory PLAs when their money is involved.
The Case Against PLAs
There is a great deal of evidence that contradict these claims of efficiency. For example, a May 5, 1998 GAO investigation could not document any cost efficiencies or quality improvements from PLAs. Diana Furchtgott-Roth concluded that a PLA “drives out small businesses from competing for these projects; raises their cost to the taxpayers; and funnels a larger stream of union dues from taxpayers’ pockets to union treasuries.” Wharton professor Herbert Northrup wrote in the Journal of Labor Research, that “restraints imposed by government directed PLAs are political decisions which have little or no economic rationale, nor can they be defended on the grounds of labor peace, enhanced safety, or other reasonable criteria.” A May 2017 study by Beacon Hill Institute, which went to great efforts to assemble as sophisticated a data set and statistical analysis as possible, because of the welter of unproven claims and the ease of biasing results by statistical sleights-of-hand by those determined to do so, found government-mandated PLAs raised Ohio school project costs by 13 percent, which was consistent with earlier studies measuring the impact of PLA mandates and school construction conducted by think tanks and government entities in California, New Jersey, New York, Connecticut, and Massachusetts.
While discussions of PLAs can get very complicated and well into the weeds of “lies, damned lies and statistics,” asking questions suggested by simple logic can provide plenty of reasons to question union claims.
If PLAs save money and are superior in many other ways, as unions claim, what would be gained by government entities requiring them or banning alternatives? If PLAs are superior, competitive contracts would go to firms using a PLA, anyway. The same logic applies to giving preferential treatment to bidders offering PLAs voluntarily or excluding other costly regulations if a PLA is used on a project.
Yet giving PLAs such advantages are common. My favorite example was California Senate Bill 922 in 2011, barring local governments from banning PLAs. It was reinforced with AB 436 and SB 790, which exempted local governments from some fees if they required PLAs and required public utilities to pay into a union-controlled fund promoting such agreements. When such special advantages are given to PLAs, how can unions claim that mandating them provides “a level playing field”?
Why is there no other area in which PLA-style contracts are used in a comparable manner, if they are such a superior approach? In fact, why are such contracts generally prohibited by the National Labor Relations Act, but with a special exemption for construction PLAs. Further, when PLA mandates and advantages should, following proponents’ logic, be unnecessary for users to win contracts, their continued use makes one suspect that the superiority claims are false.
They Don't Save Money
If PLAs are superior ways of organizing the relevant type of projects, why have projects originally bid under PLA agreements, and then rebid without a PLA, seen more competition and substantially lower bid costs as a result? Why would non-union competitors object so strenuously to being required to use PLAs? Wouldn’t their costs be lower and their workforce more efficient, increasing their ability to win bids? For that matter, why are the vast majority of construction industry workers in non-union organizations, if using union organization and rules was generally superior? And where are the studies that show union organization of any industry lowers costs?
Why, when government-mandated PLAs have been banned more generally (e.g., when George Bush’s Executive Order 12818 banned government-mandated PLAs between 2001 and 2008, and now in a total of 24 states) or in a jurisdiction, has there been no jump in costs, decrease in quality, increase in delays, or increase in labor unrest, if those supposedly justify PLAs? Because they are rationalizations, not rationales.
PLAs Aren't More "Inclusive"
Given that PLAs tout special minority apprenticeship and training programs to demonstrate their outreach efforts and garner minority support, why do major black, minority, and women’s groups in construction oppose them? Largely because there are fewer minorities in unions than in the non-union sector, so letting unions allocate workers to job sites harms most of them. Those trained in non-union apprenticeship programs (which unions have also long tried to hamstring) are not eligible, either. Further, requiring journeymen and journeymen wages for most tasks, and reducing helper jobs, crowds out minority workers, who are, on average, less skilled. That also limits their ability to get experience and learn their way to higher future incomes. The National Association of Women Business Owners has opposed PLAs, as has the United States Pan-Asian American Chamber of Commerce. The National Black Chamber of Commerce has called PLAs “a license to discriminate against black workers.”
Why do PLAs get defended as a tool in lessening labor unrest? Non-union workers and contractors don’t strike. Unions strike. So it makes more sense to say that PLAs reduce union unrest. But that means PLAs are rewarding unions to pay them off not to disrupt projects, and penalizing those who would not strike and who therefore represent the solution to the problem of unrest. Even some courts have seen this as the extortion it is.
PLAs Aren't Representative of All Workers
Why don’t negotiations in creating PLAs include non-union contractors? It is not about providing a “level playing field,” as unions claim. It is about mandating the unions’ playing field for those who would not otherwise choose it, and getting union dues from workers and employers unions they could not convince to be members voluntarily. Similarly, why must non-union contractors pay into union health and retirement trust funds under PLAs, even though they contribute into their own health and retirement funds, and few, if any, of their employees will get vested to be eligible for any union benefits?
Why do PLA-endorsing “studies” ignore the many far more reputable studies in the other direction? So they can be presented as authoritative, whether they really are or not. Why do they misrepresent what other studies have found? So they can use strawman arguments to appear defensible when they are not.
And even when PLA proponents torture the data to make it look like PLAs are superior in some cases, what does it prove? It only proves that it might be possible they are better in some cases. But that provides no logical support for mandating PLAs, unions’ most strongly-backed policy choice, since PLA organizations would outcompete other forms of organization where they are superior, anyway. There is no reason to require PLAs, whenever PLA organization is superior; there is also no reason to eliminate the possibility that PLAs would be outcompeted, when they are in fact less efficient.
Given the support unions have provided for so many left-leaning politicians, it is hardly surprising that they back PLA “sweetheart deals” for unions on public construction contracts, particularly since very few Americans understand what is involved. But the arguments made for such policies are Swiss-cheesed with logical errors and internal inconsistencies, as well as questionable statistics. And until supporters can come up with better answers to the questions asked here, and others, there is no reason to give their claims credence.