Imagine a market that looks like this. You're thinking of buying something—a house, insurance, shares of stock, a prescription drug, or a medical procedure. There is an entity (call it the "anti-seller") that considers it in its economic interest that you not buy the item. The anti-seller has a great deal of money and all of your working hours to persuade you not to buy. The anti-seller hires lawyers and consultants to run a campaign to block the sale. The consultants require your supervisor at work to meet with you daily in one-on-one meetings to tell you why you shouldn't buy. The anti-seller makes and screens a movie starring your co-workers in which they explain why it's a bad idea to buy.1 Although the law prohibits direct threats, the message underlying anti-seller's advertising campaign is that your employer may go out of business or be forced to lay you off if you buy the item. Meanwhile, the seller from whom you propose to buy has to work hard even to figure out who you are and where to reach you to speak with you. Unlike the anti-seller, the seller cannot speak with you during your working time, cannot email you at work, and can't even leave a leaflet on the windshield of your car in the parking lot at work. If union representation is sold in a market, as Professor Bodie suggests,2 that is what the market looks like.3
Is this market a properly functioning one? The one thing that we can say about that market is that it is unlike any other market for the sale of goods or services that exists in America today.4 The usual problem with information is with sellers who have very good access to consumers and must be forced to provide accurate and material information in the market. In the union market, however, the seller has poor access to the consumer. Moreover, the problem is not just that workers get too little information or that unions as sellers mislead; it is that workers get a huge amount of information, some of it misleading, from an entity who is not the seller but who has an incentive to do everything possible to persuade them not to buy. In addition, unlike in most other markets, the anti-seller has the consumers as a captive audience all day each workday and an unparalleled ability to intimidate the buyer by indirectly threatening his or her job. Unless we accept the dubious contention that only unions are likely to exaggerate or mislead workers about unionization, any genuinely effective regulation the market must deal both with the seller and the anti-seller when it comes to both the quantity and the quality of the information.
It is an article of faith among lawyers and economists that information is good. Perfect information is the gold standard in economic analysis. The free market of ideas is the conventional justification for free speech. A duty to provide information has great normative force. It's like freedom or literacy or the benefits of play for young children—you can hardly find anyone who will argue against it, and the halo surrounding the concept makes it hard to realistically access whether or when it really is as good as we think. That is particularly true when it comes to important decisions. Employee decisions whether to join a union and management decisions whether or how strenuously to oppose unionization are enormously significant by almost any measure. Many studies show that unionization results in pay increases and a net transfer of wealth from shareholders to employees. Unionization typically reduces the ability of the firm arbitrarily to fire employees. Unionization usually increases compliance with workplace safety and health laws. Given the importance of the decision and the normative appeal of information, it seems obvious to suggest law should force full disclosure of information about it.
The law regulating union elections does a terrible job of ensuring full and fair disclosure of information to any of the people—employees, managers, senior managers, customers, and shareholders—whose lives or wealth may be affected by the unionization of a group of employees. Indeed, the law of union elections finds few defenders at almost any point of the ideological spectrum (not even among those who make their living off its Byzantine complexity). And, as Professor Bodie also points out, economics would suggest that the more workers will benefit (at the expense of management or shareholders) by unionization, the greater the incentive of management to oppose it. The hard question is what to do about information. Whole bodies of law exist to force disclosure of relevant information, including in securities, real estate, insurance, banking, pharmaceutical and food labeling, and the informed consent concept in medicine. What do these areas of law suggest about how a disclosure obligation would work in practice, and on whom the obligation would rest?
In most other areas of law in which a consumer purchases goods or services, the law obligates the seller to disclose. Labor law, however, is more complicated. The provision of information involves two components: access and accuracy. The law gives the employer much better access to communicate with employees. Unions cannot go on company property; employees cannot use working time to communicate with each other and cannot distribute literature at any time in working areas. Employers can prohibit unions and employees from using company email systems. The union cannot even get the names and contact information of workers until it already enjoys substantial support among workers.5 When it comes to the substance of speech, although law treats the seller and the anti-seller equally—there are few limits on truthfulness and none at all on its fairness, and short of promises and threats, anything goes6—the equality is rather like that enjoyed by the poor of Paris.7 When the employer and union each render a prediction about whether unionization will result in the employer going out of business or laying off workers, the employer's prediction has a whole different weight.
The metaphor of the decision to choose a union as being a decision to purchase services in the market lends itself to thinking that the duty to disclose should rest on the union, because that is where law usually places disclosure obligations. If a duty of increased disclosure fell on the union, how would the union discharge that duty without equal access to employees? Many areas of law that require the seller to disclose allow the buyer either to rescind a transaction or to sue for damages if there is inadequate disclosure. How would such a rule be applied to unions? Could employees who opposed the union have a union certification set aside? Would there be an action for damages if the workers later concluded that they made an unwise choice in voting for a union? If part of the information problem is the inequality in access between employer and union, and the remedy is to impose a duty on employers too, should workers be able to set aside the election if the union lost? Or bring an action for damages against the employer that provided inaccurate information?
Depending on whether one thinks the main problem with the market for union services is inequality of access or the problematic content of information provided, one might propose different legal reforms. Any law aiming to reduce the quantity of information provided by firms is not realistic. First, such a law would require an amendment to Section 8(c) of the National Labor Relations Act ("NLRA"),8 and that is not likely to emerge from Congress until Democrats get both the White House and a filibuster-proof majority of the Senate. Second, even if such legislation were enacted, any significant restriction on employer speech is likely to encounter a stiff First Amendment challenge in the Supreme Court. Reform is therefore more likely to come, if it comes at all, in the form of enhancing union access—as by legislatively overruling Lechmere9 and NLRB v. United Steelworkers (Nutone & Avondale).10
But what about the content of the speech? There is no First Amendment right to mislead, so one could imagine a legislative change to Midland National Life Insurance Co.11 and maybe even a narrowing of the right to make menacing "predictions" that are currently allowed under Gissel.12 Absent equal access, however, no one would expect that changes to the content of union election speech would make much difference. Is it realistic to think that courts would uphold, even if Congress enacted, some sort of "fairness doctrine" requiring employers to give unions equal time if they choose to run anti-union campaigns? The fact that the Court struck down the application of the fairness doctrine to newspapers does not bode well for such a rule in union elections.13
One difficulty in designing appropriate legal rules to address whatever information problems exist is that the reasons for the information problems are different than the usual ones. Inadequate disclosure by unions, unlike inadequate disclosure by sellers of other goods and services, is not simply a function of the seller's self-interest. If there is any truth to the data showing that unionization increases the wealth of workers at the expense of shareholders, most unions would be happy to provide more truthful information to workers about what unions do for their members (they might also wish to provide misleadingly positive information too, of course). But even if unions were required to provide only material and accurate information, they would encounter two problems in doing so. One is that law restricts their access and the second is a lack the resources. The information problem could be partially addressed by lowering the barriers to access. The law might say that any form of speech the employer chooses to engage in should be made available to the union: captive audience speeches, one-on-one meetings at the workplace, email messages, films, etc. This would be a radical change in labor law, and I doubt it will happen until Democrats get a filibuster-proof majority in the Senate and occupy the White House. Even then, one wonders what it would take to get such a legal change through the inevitable legal challenges in federal courts. But assume that happened. Would unions take advantage of the opportunity for equal time?
That raises the question of resources. Unions are nonprofits. Even the wealthiest union has less money than most large firms. Moreover, labor law limits the use of union dues for organizing. Unlike every other form of nonprofit organization, and unlike their corporate adversaries, unions are required by law (the duty of fair representation) to provide services to existing members and limited by law in spending revenues to sell their services to prospective buyers (i.e., to organize). Under some circumstances, employees represented by a union but who refuse to pay full dues can refuse to fund organizing, which requires unions to segregate their assets and devote toward organizing only the money raised from members who have not exercised their right to pay agency fees.14 Employer associations are vigorously pushing for even greater restrictions on the ability of unions to raise and spend money to organize new workers.15 So the trend in the law is toward reducing, rather than increasing, union resources for information disclosure.
Having long ago despaired of any significant legal change in the area of campaign speech, unions have decided that the only solution to one-sided and misleading employer speech is to shorten the selection process through card-check and eliminate misleading and covertly threatening speech through neutrality. Card check and neutrality are essentially prophylactic: if you can't get equal speech, then just have less speech. I suspect that the National Right to Work Organization will seize on Professor Bodie's article as vindication of their view that card-check and neutrality are bad because they silence speech.
There is a very real risk that lawyers and courts will attend only to half of Professor Bodie's argument. They may accept the contention that information is important and that card-check and neutrality and the Board's proposed new rule for expedited twenty-eight-day election campaigns16 are bad because they limit information in the union election process. But they may overlook the substantial information problems that already exist in the process. While Professor Bodie does an admirable job of explaining why information matters, the process will not be significantly improved if an argument for more information is taken as an argument to protect the status quo of misleading and one-sided information.
Preferred citation: Catherine L. Fisk, Thoughts on Treating Union Representation Processes as a Market in Need of Legally Required Disclosure of Information, 94 Va. L. Rev. In Brief 1 (2008), http://www.virginialawreview.org/inbrief/2008/03/24/fisk.pdf.
[Back to this issue]