VII. ALTERNATIVE FORMS OF EMPLOYEE
REPRESENTATION OR INVOLVEMENT

A Competitive Market for Workplace
Representation Services

Samuel Estreicher
New York University School of Law

Abstract

     Our labor laws as currently structured do a poor job of establishing a competitive market for representation at the workplace. A "one size fits all" model, only nonprofit, democratic membership organizations may vie for representational authority. Once a union obtains such authority, it is very difficult for represented employees to vote out the incumbent and bring in alternative representatives. This is in part because the law allows only a limited window period for decertification petitions and such petitions must be supported by at least 30 percent of the unit. Collective action of this type is exceedingly difficult to effect. This article proposes a new approach to help bolster competitive forces in the marketplace for representation services.

Introduction

     Our current labor law system is based on the premise that one size fits all, that if workers want collective representation it has to take the form of a nonprofit membership organization adhering to formal internal democratic rules, and that once a union has been voted in it should be difficult to vote the union out or change bargaining agents. Workers thus must decide on this fairly important economic decision on the basis of limited information and facing the prospect that, if they make the wrong decision, they will for all practical purposes be stuck with that decision. Responsiveness to employee wishes is thought to be ensured by the democratic rights safeguarded by the Labor-Management Reporting and Disclosure Act of 1959 (LMRDA, or Landrum-Griffin Act), despite decades of research indicating that workers generally do not attend union meetings and have considerable difficulty organizing themselves to promote internal opposition to union leaders. Moreover, under current law, the right to participate in internal union affairs belongs only to union members willing to subject themselves to union discipline.

     This paper explores whether introducing an element of competition in the market for workplace representation services would enhance the responsiveness of bargaining agents to their represented employees and, if so, what form should such a change take.

Agency Costs of Unionism and Exit-Voice Paradigm

Sources of Agency Costs

     All organizations pose issues of "agency costs" (i.e., the costs that inhere when principals must rely on agents to promote their interests). In common law agencies, these costs are kept to a tolerable minimum because, unless an irrevocable agency has been created, the principal can always revoke the agent's authority if he is no longer satisfied with the agent's representation.

     In the labor relations context, however, the problem of agency costs is magnified because the law creates a political agency that binds the principal (the employees of the bargaining unit) to the agent (the exclusive bargaining agent) on the basis of majority rule and restricts the ability of even the majority to revoke the agent's authority. The problem is further complicated by the fact that the bargaining agent is also a private membership organization beholden in principle only to its members and to the directives of the (inter)national organization of which it is a part.

"Exit" and "Voice" Strategies

     The question for the legal regime is whether its rules do a good job of fully minimizing these agency costs so that, without sacrifice of organizational efficiency, unions are as responsive as they can be to the employees of the bargaining unit on whose behalf they exercise exclusive bargaining authority. Whether this is so requires an assessment--to borrow terminology from Hirschman's influential work (1970)--of the effectiveness of the "exit" and "voice" options available to unit employees.

     "Exit." Bargaining-unit employees in the real world have limited opportunities to police their bargaining agent. The basic premise of labor law is that workers cannot readily quit their jobs, either because they are rooted in their communities, have made investments in firm-specific skills, or receive wages and benefits that are keyed to length of service. For these workers, exit from the firm is not a realistic option. (And even if an employee were leave to leave his job, he simply would be replaced by a new employee; the size of the bargaining unit would not change, and hence there would be no impact on the bargaining agent.)

     Exit from the bargaining agency is no less practicable. Workers can certainly resign their union membership, and the Supreme Court held in the Pattern Makers case that they can do so at any time, even in the midst of an active strike, and irrespective of limits on resignation set forth in the union's governing documents. But resignation from the union requires relinquishing all rights to participation in union decisions, including votes on the officers of the union, levels of dues, strike authorization, and contract ratification (Hyde 1984). More important, from the standpoint of influencing union behavior, resignation outside of "right-to-work" states does not entail any significant diminution of dues income to the union. Under the conventional union security clause, the former union member is still a unit employee who must continue to fund his share of the union's collective bargaining and contract administration functions.

     Employees can, in theory, seek to decertify the bargaining agent. This route is a difficult one, for employees have to be able to overcome the collective action problem that inheres whenever individuals must organize others to accomplish a change the benefits of which cannot be confined to the individuals undertaking the effort. It is made even more difficult by labor law rules designed to promote the stability of bargaining relationships. Thus, under the National Labor Relations Board's (NLRB) "contract bar" rules, when confronted with a facially valid collective bargaining agreement for a term of years, the agency will entertain a petition for decertification only (1) once every three years, (2) if supported by at least 30 percent of the employees in the unit, and (3) if filed within a limited "window" period no earlier than 90 days and no later than 60 days before contract expiration. For quite understandable reasons, the employer cannot provide any assistance to the employee petition.

     Even after all of these hurdles are overcome, a timely and properly supported petition has been filed, and a majority of the employees vote to decertify, the typical end result is not a new representative but no collective representation at all. This is because, since the 1954 merger of the AFL and CIO and the absorption of the Teamsters into the Federation, there is no rival unionism to speak of in this country.

     "Voice." Unions know ex ante that they have very little to fear from the ability of employees to exit their jobs or exit the bargaining agency. Hirschman teaches that, where exit is unavailable, voice becomes all the more important, and this is where the LMRDA comes in.

     One problem with the LMRDA is that participational rights are available only to union members. Bargaining-unit employees who do not want to expose themselves to union discipline have no rights under the LMRDA.

     The larger problem with the LMRDA is the belief that workers in fact have a strong interest in participating in internal governance decisions. Bargaining-unit employees certainly have an interest in economic decisions directly affecting them, such as the size of the dues they will have to pay, whether they accept the employer's final offer, and whether they wish to go on strike. It is doubtful that workers generally care about who holds office in the union, to what uses dues monies are put, or the nature of relations between local and parent organizations.

     Many of us (especially in the academy) instinctively resist this conclusion, maintaining that union members should care; perhaps when polled, they even say they do care. But from the standpoint of what union members actually do--their preferences as revealed by their willingness to attend union meetings, run for office, support rival candidates, and the like--it is clear that these matters do not seem to be worth the time and energy effective voice requires.

     Internal union matters simply are not "salient" to the vast majority of union members, in much the same way that local school board elections are not salient in most communities. The people who get involved are those who have some professional or business interest in the school's affairs; the vast majority of parents and taxpayers are passive. So, too, with internal union affairs: incumbent officers and their staff have a keen interest and occasionally a challenger will emerge, but the vast majority will not get involved.

     We may bemoan these facts, but they often represent a rational response on the part of union members. People rationally ration their time and energy. Participation is costly. To paraphrase Oscar Wilde, "The problem with democracy [he was discussing socialism] is that it requires too many evenings out." Union members have to be willing to get informed and to use up part of their spare time to attend meetings, and if they do attend they face meetings controlled by a political elite--incumbent officers and their staff--who are far more knowledgeable about union affairs and parliamentary procedure, and have more time to devote to these affairs.

     We certainly can point to some examples of contested union elections, but they are few and far between and often the product of unusual circumstances.

     We do know that strong autocratic union leaders have had considerable staying power. Jimmy Hoffa, Sr. was widely popular among truckers, despite a management style that did not brook dissension and shadowy ties with organized crime, because he delivered rich contracts for the membership. A more recent example is Gus Bevona's stewardship of Local 32B-32J of the Service Employees International Union. Despite a salary in excess of $400,000 a year--the highest in the labor movement--a penthouse apartment, and other luxuries funded with union dues, dissidents were unable to dislodge Bevona in open elections because, like Hoffa, the janitors and doormen liked the contracts he negotiated. Even the much-applauded Walter Reuther of the Auto Workers ran his union as "a one-party state whose challengers, when they dared to raise their heads, might just as readily find themselves in trusteeship as campaigning for election" (Fraser 1998:36).

     It should not be surprising that most rank and file union members care more about their wages and working conditions than they do about their union's regard for procedural niceties. A union is a limited-purpose organization that, when it works as it should, advances the membership's economic goals; it is not for most members a vehicle for realizing their noneconomic, spiritual needs. They may even be acting out of an intuitive sense that too much democracy will undermine the union's raison d'etre (Fraser 1998).

Costs of the Current System

     How one responds to this imperfect framework of "exit" and "voice" opportunities depends on (1) whether it has overall negative effects on the principal objective of the labor relations system--maintaining an effective collective representation option for workers desiring such representation--and (2) whether alternative frameworks are possible that will do a better job of reducing the agency costs of unionism without undermining the effectiveness of the bargaining agency.

Effect on Employee Attitudes toward Unionism

     Unions in 2002 represented well under 10 percent of workers in private employment. There are several reasons for this decline in density. All observers agree that at least a good part of the cause is a change in employee attitudes toward unionization (Farber 1990; Farber and Krueger 1993). Survey data suggest that Americans continue to have a fairly low opinion of union leaders, and the recent work of Freeman and Rogers on the attitudes of nonunion workers suggests an unrealized demand for forms of collective representation not presently supplied by AFL-CIO affiliates (Freeman and Rogers 1998; see also Estreicher 1996).

Unresponsive union leadership. To the extent that nonunion workers form unfavorable attitudes toward unions because of perceptions of unresponsive leadership, union corruption, or simply Rabelasian excess of the Gus Bevona variety, we should be open to new institutional arrangements that might do a better job of reducing the agency costs of unions.

     Unions are indeed responsive bargaining agents in many places, but fair-minded observers cannot dispute that in at least an equal number of places they act as agents largely unresponsive to their principals, because they know bargaining-unit employees have no realistic exit strategy and the voice option is too costly to invoke in a manner to challenge effectively the incumbents' hold on office.

Union corruption and excess. Union corruption and excess admittedly stem from many causes. One contributing factor is a system that confines unions to the form of nonprofit membership organizations. The reasons why unions take this form are obscure; it may well be due to tax considerations or, possibly, the requirements of labor's antitrust exemption under the Clayton Act, which is limited to labor organizations "instituted for the purposes of mutual help, and not having capital stock or conducted for profit" (Clayton Act, §6).

     Entrepreneurs in the for-profit world have the same desires as the Gus Bevonas of the world to capture for themselves as much of the surplus from their activity as they can. Where such entrepreneurs are in business for themselves, they are naturally attentive to whether their profits are deployed in ways that generate the highest returns. Where they operate in a corporate form, however, their interests as managers diverge from the interest of the ownership. To a considerable (though imperfect) extent, managers are compelled to hew closely to shareholder interests because of the market for corporate control, principally the threat of hostile takeovers. Such constraints on these agency costs are largely absent in the union context (Schwab 1992).

     Unfortunately, union leaders such as Bevona have no strong incentive to minimize union expenditures because, again, workers are effectively locked into the bargaining agency and, despite the statutory guarantee of elections over dues increases, voice effects are as diminished here as they are with respect to contests for union office. Such leaders, moreover, are legally prevented from plowing returns from their entrepreneurial leadership into profit-generating activities. Instead, to the extent dues revenue exceeds expenditures on collective bargaining and contract administration functions, the surplus is spent on expensive office buildings, lavish apartments, "no-show" jobs for relatives and friends, and the like.

Fear of exposure to union discipline. A third determinant of negative worker attitudes toward unionism is the fear many workers have of union discipline. Nonunion workers may want the economic returns that strong unionism can bring but are unwilling to expose themselves to the disciplinary authority of unions. It is true, as discussed above, the NLRA has been interpreted to give workers a right to resign from the union at any time--and hence free themselves from any exposure to discipline--but this freedom can be purchased only at the cost of relinquishing any right to participate in strike and contract ratification decisions that will continue to affect the nonmember in the bargaining unit. Nonunion workers ex ante may not want to enter a system that puts them to such a choice--a message that employers often emphasize during organizing campaigns. To the extent unions maintain disciplinary authority not as a function of what they believe to be essential to effective bargaining power but rather as an incident of their form as a membership organization, this feature of unionism bears reexamination.

Absence of a Marketplace for Representational Services

     A second major cost of the current system is that it produces a limited supply of bargaining agency service providers, and those that are produced are exclusively of the nonprofit membership form capable of compliance with LMRDA requirements.

     The limited supply of service providers is due not only to the AFL-CIO's no-raiding pact, but also to aspects of the present legal regime. For-profit organizations will not vie for exclusive bargaining authority, because they are not entitled to protection from the antitrust laws when they attempt to cartelize labor markets. Even organizations assuming a nonprofit form are constrained by legal requirements. Although, in theory, individuals can be labor organizations under the NLRA and Railway Labor Act, the LMRDA applies to all labor organizations seeking authority to represent workers covered by the NLRA and Railway Labor Act (LMRDA §3[j]).

     The upshot is that if you want to go into the business of providing collective representation services, you have to take on a nonprofit form and a membership structure conforming to LMRDA rules. Moreover, you must strip your organization of any other labor-market activities that regulatory agencies may deem productive of potential conflicts of interest (St. John's Hospital 1982; Sierra Vista Hospital 1979), and as the professional teachers and nurses associations have learned you must eliminate, for all practical purposes, any programs for supervisors or would-be supervisors. In effect, the market for representational services is distorted because regulations erect high barriers to entry, effectively insulating labor organizations of the traditional variety from competition.

The Proposal

Indifference as to the Form of Labor Organization

     One mistake of the current regime is worrying too much about the form that permissible labor organizations may take. As long as workers are provided low-cost opportunities to cast secret-ballot votes on the economic issues most directly of concern to them, the law should be indifferent as to the form, the internal structure of labor organizations. I would propose that existing laws be amended to allow any individual or organization free of employer domination or support to vie for exclusive bargaining status.

     To implement this part of the proposal (1) the Clayton Act would be amended to make clear that for-profit enterprise is immune from the antitrust laws insofar as they limit their anticompetitive activity to restraints on labor market competition and (2) the LMRDA's requirements would be permissive, binding only for those organizations that voluntarily assume those requirements.

     Under this proposed regime, individuals, for-profit companies (whether they issue stock or not), nonprofit associations with limited membership rights under state law, nonprofit associations with ancillary employment referral and other for-profit labor market functions, and traditional nonprofit membership organizations voluntarily conforming to the LMRDA all could compete for authority to act as exclusive bargaining agents.

     The NLRA and National Mediation Board would continue to exercise their authority to determine the appropriate unit for collective bargaining purposes and would hold elections under current rules to determine whether workers in an appropriate unit desired collective representation. Laws restricting violence, fraud, association with criminal enterprise, and the like would continue unimpaired. In addition, state law would be available to enforce union constitutions or other contractual undertakings.

Separating Participational Rights from Membership

     Under the proposed regime, participational rights in critical economic decisions directly affecting the welfare of bargaining-unit employees will be divorced from membership in the labor organizations. Some providers of representational services will not be membership organizations at all, or membership will be restricted to business associates, relatives, or shareholders. Workers' rights will be a function of their involvement in a political agency--exclusive bargaining status under the labor law.

     What bargaining agencies charge for their services will be determined in the first instance by the service providers, and workers in bargaining units when they decide whether they wish to be collectively represented will at the same time vote on whether they approve of the service provider's proposed dues assessment. In essence, service providers will vie for employee support not only on the basis of their record as bargaining agents at other employers but also the projected cost of representation--a projection that will be binding on the service provider for a fixed period.

     What bargaining agencies choose to do with the revenues they receive from dues will be their business. The revenues can be used to line the pockets of the service providers, to invest in organizing other units, to fund political activities for the benefit of represented workers, or for any other lawful activity.

Mandatory Participational Rights

     The NLRA and Railway Labor Act would also be amended to provide that all bargaining-unit employees subject to an exclusive bargaining agency would have statutory rights, whether they are union members or not, to vote in secret ballot on the following critical economic decisions: (1) whether to have a collective representation, who it should be, and whether to approve the dues proposed to be assessed by that representative; (2) whether to reauthorize the bargaining agency within a defined period of time, say, two or three years; (3) whether to approve or disapprove of the employer's final offer; (4) whether to authorize a strike; and (5) whether to ratify the proposed contract.

     With the possible exception of the initial vote on collective representation, all of the votes on the other critical economic decisions, including reauthorization of the bargaining agency, would require no showing of interest, no effort on the part of the affected employees to organize their colleagues or attend meetings as a precondition to exercise of the franchise. It would be the job of the NLRB or National Mediation Board to hold these elections or certify the parties themselves to conduct these elections if they conform to statutory guarantees as to secrecy of the votes and integrity of the ballot-counting process.

Related Features

Single-purpose bargaining agencies. It would be lawful under the proposed regime for organization to seek authority only on a particular issue, say, wages or pensions. Such organizations would be restricting some of their bargaining leverage, by reducing the scope for tradeoffs, and would be competing with organizations claiming that they are more effective bargaining agencies because they can provide the full range of bargaining and contract administration services.

Rival bids. For a truly competitive market for representational services to take hold, the NLRB and the Mediation Board should relax their requirements for bids by rival organizations. Such organizations certainly cannot be given automatic access to the ballot because any additional choice complicates the decision-making process for unit employees; rivals also may be acting strategically, seeking a presence on the ballot simply to preclude a majority vote for their rival. One approach would be that for both the initial organization and reauthorization contests, a rival organization should be permitted on the ballot on a 10 percent showing of interest. This is a considerably lower threshold than under current law, but one high enough to curb gamesmanship and unnecessarily complicated ballots.

Referral hall unions. Under the proposal, current law would not change on the ability of unions to act as "members only" organizations providing nonexclusive hiring hall services to signatory employers; the proposed new regime is triggered only with the establishment of an exclusive bargaining agency. Political agency concerns are implicated only when unions act as exclusive bargaining agencies as where to obtain agreements providing that their hiring halls be the exclusive source of labor.

     The rules governing bargaining agencies for casual labor is an area where current law is entirely deficient. I would propose here that signatory employers maintain a registry of workers who have been referred to work on their projects during a defined period, say, three years. These workers would be considered the bargaining unit for the mandatory participational rights envisioned by the proposal. Where unions do not provide hiring hall services as such but act as the exclusive bargaining agent for workers who are hired on a project-only basis, as is the case in some sectors of the entertainment industry, the law would be amended to allow prehire agreements, and a similar registry of workers performing work for signatory employers over a defined period would be maintained.

Multiemployer Pension and Welfare Benefit Plans

     Under current arrangements, unions enjoy a certain entrenchment effect because service credit for pension and welfare benefits accrues only for work among signatory employers; workers know that any change in the bargaining agency could undermine their built-up benefits. Yet, in theory, even today workers can decertify such bargaining agencies.

     The proposal outlined here should reduce somewhat (though not eliminate) this entrenchment effect; however, rival organizations would still have to be able to persuade bargaining-unit employees that they would be able to ensure continuity of benefit levels after a change in bargaining agency.

Participational Rights and Bargaining Structure

     As is true today, bargaining could occur permissively at a different level than the bargaining unit. Even where single-facility units are found to be appropriate, labor organizations and employers could agree to bargain on a region-wide, employer-wide, or multiemployer basis. The participational rights envisioned by the proposal, however, would be exercised at the level of the bargaining unit. Thus, for example, even where bargaining occurs on a company-wide basis, the unit employees would cast secret-ballot votes on whether their unit authorizes the strike or ratifies the contract.

Optional Features

Lowering the barriers to initial organization: "easy in, easy out." It would be entirely consistent with (though not required by) the proposal outlined here to lower the barriers to initial organization by, say, providing for bargaining authority elections on the basis of authorization cards evidencing supermajority support in the unit, much like some of the provincial Canadian laws. Under current law, one of the strongest arguments for allowing contested elections is that an employee vote for collective representation is not easy to rescind and has long-term consequences. Giving the employer the opportunity to present the arguments against collective representation or representation by the petitioning organization provides critical information that employees making such a decision should have. In essence, we have a "hard in, hard out" regime for union certification.

     Under my proposal, the exit option is significantly bolstered because bargaining agencies will be subject to periodic secret-ballot reauthorization votes without requiring a prior showing of interest in decertification. If exit is enhanced, the barriers to entry can be correspondingly relaxed.

Internet posting of collective agreements, dues structures, and internal discipline policies. Another point to consider would be a legal regime for all exclusive bargaining agencies, and all organizations seeking such authority, to post at a designated place on the Internet all collective agreements negotiated by the organization in the particular industry, as well as a clear statement of the organization dues structure and policy on seeking court-imposed fines. Information of this type would enhance the operation of a competitive market for representational services because it would facilitate an informed vote.

     In addition, consideration should be given to the NLRB and the National Mediation Board providing on their Internet home pages the dates of scheduled initial certification and reauthorization elections. This would facilitate the emergence of rival bids.

References

Clayton Act. 1914. 38 Stat. 730, 15 U.S.C. §§ 12 et seq.

Estreicher, Samuel. 1996. "Freedom of Contract and Labor Law Reform: Opening Up the Possibilities for Value-Added Unionism." New York University Law Review, Vol. 71, no. 3 (June), pp. 8'-49.

Farber, Henry S., and Alan B. Krueger. 1993R. "Union Membership in the United States: The Decline Continues." In Bruce E. Kaufman and Morris M. Kleiner, eds., Employee Representation: Alternatives and Future Directions. Madison, WI: Industrial Relations Research Association, pp. 105-34.

Farber, Henry S. 1990. "The Decline of Unionization in the United States: What Can be Learned from Recent Experience?" Journal of Labor Economics, Vol. 8 (January), pp. S75-S105.

Fraser, Steve. 1998. "Is Democracy Good for Unions?" Dissent (summer), pp. 33-39.

Freeman, Richard B., and Joel Rogers. 1998. "What Do Workers Want? Voice, Representation, and Power in the American Workplace." In Samuel Estreicher, ed., Employee Representation in the Emerging Workplace: Alternatives/Supplements to Collective Bargaining. Proceedings of New York University 50th Annual Conference On Labor. Cambridge, MA: Kluwer Law, pp. 1-31.

Hirschman, Albert O. 1970. Exit, Voice and Loyalty: Responses to Declines in Firms, Organizations, and States. Cambridge, MA: Harvard University Press.

Hyde, Alan. 1984. "Democracy in Collective Bargaining." Yale Law Journal, Vol. 93 (April), pp. 793-856.

Labor-Management Reporting and Disclosure Act (LMRDA, or Landrum-Griffin Act). 1959. 73 Stat. 519, as amended, 29 U.S.C. §§ 401-531.

National Labor Relations Act (NLRA). 1935. 49 Stat. 449, as amended, 29 U.S.C. §§ 151-69.

Pattern Makers' League of North America v. NLRB. 1985. 473 U.S. 95.

Railway Labor Act. 1926. 44 Stat. 577, as amended, 45 U.S.C. §§ 151-88.

Schwab, Stewart J. "Union Raids, Union Democracy, and the Market for Corporate Control." University of Illinois Law Review, Volume 1992, No. 2, pp. 367-416.

Sierra Vista Hospital. 1979. 241 N.L.R.B. 631.

St. John's Hospital & Health Center. 1982. 264 N.L.R.B. 990.