Skip to page content
Office of the Secretary
Bookmark and Share

U.S. Department of Labor Futurework
  Trends and Challenges for Work in the 21st Century
---division bar --

What Do Nonunions Do?
What Should We Do About Them?

Daphne Gottlieb Taras, University of Calgary and Bruce E. Kaufman, Georgia State University
Task Force Working Paper #WP14
Prepared for the May 25-26, 1999, conference
“Symposium on Changing Employment Relations and New Institutions of Representation”

September 1, 1999

IV. Policy Implications

We believe public policy should promote two goals. The first is to permit companies to implement programs for employee involvement and participation and to allow them considerable discretion with respect to the role and operation of employee representation committees therein. The second goal we subscribe to is that public policy should fully protect the right of employees to join unions and collectively bargain and that employer practices of a coercive or punitive nature that infringe on this right should be prohibited. We hold this position given widespread evidence that some companies engage in exploitative, opportunistic and/or inequitable practices vis a vis treatment of their employees (Friedman, Hurd, Oswald, and Seeber 1994) and that trade unions and collective bargaining are an important and socially beneficial means employees have to rectify these conditions. But without strong legal protections, workers too often are prevented by employer acts of anti-union discrimination from obtaining independent representation.

Thus the challenge for public policy is to provide as much latitude as possible for nonunion companies to use employee representation committees as part of a legitimate cluster of progressive human resources practices, but at the same time prevent companies from using them as illegitimate tools of union avoidance. We recognize that well-run, successful NERPs tend to substantially reduce employee interest in independent representation, and are useful to employers as a union avoidance device, but this practice seems largely benign and even beneficial to the extent the employer provides wages and conditions of work that meet or exceed what a union can deliver. The practices we wish to prevent are the use of “sham” employee committees which employers hastily put in place to short-circuit union organizing drives and that have no greater purpose than short-run union avoidance and protection of the employer’s dominant position.

The policy issue is whether to allow nonunion representation, which is a form of clandestine activity by many American employers, to be practiced in the sunlight, as it is in other countries including Canada, Japan, Germany, Great Britain and Australia.

As we have argued elsewhere (Kaufman 1999; Taras 1998), both policy goals of latitude in employee involvement and representational practice and easy exercise of the union option can be accomplished by a two-pronged change in the NLRA. The first is to narrow the definition of a labor organization in Section 2(5) so that it only applies to independent employee organizations established for purposes of collective bargaining. The problem, according to Estreicher, is that the current 2(5) definition of labor organization “applies even when employees are not seeking to organize an independent union, and do not have the slightest interest in doing so.” Its “capture basin” must be limited by either changing “dealing with” to mean “bargaining” or some form of purposeful bilateralism, or tampering with the topics permitted (which we personally do not favor because of artificial limits to workers’ ability to pursue their own agenda). The Canadian solution is a possibility: a labor organization should mean a union, or at the very least, a collective entity whose purpose includes regulation of relations through collective bargaining. Even the Labor Policy Associate’s April 1999 official complaint that American law violates the NAFTA labor accords contains a discussion of solutions which move away from TEAM’s obsessive focus on 8(a)(2), towards other solutions including a change in Section 2(5), with specific attention directed to the Canadian solution (LPA 1999). This change effectively exempts from the coverage of the NLRA all nonunion employee committees that are company-created and operated for EI purposes.

At the same time, Section 8(a)(2) should remain unchanged so bona fide agencies of collective bargaining remain free of employer interference and domination. Such a provision is essential in order to prevent so-called “rat unions” (which really represent management interests) from being certified as the exclusive bargaining agent for employees, and being protected from raids by legitimate unions. Section 8(a)(2) is a clear, precise, and cogent directive that prohibits an unfair labor practice and allows the Labor Board to exercise its power to prevent the intent of the NLRA from being subverted by any overly-zealous companies. It ought not be tampered with in any way.

Appendix 2 provides a detailed examination of both the American treatment of nonunion representation and the different Canadian treatment. Note that without exception, every Canadian jurisdiction has the equivalent to a Section 8(a)(2), which is used to prevent management-dominated unions from achieving labor board certification as sole bargaining agent for employees. Canadian nonunion systems are lawful, not by virtue of tampering with the clear language of Section 8(a)(2), but rather, by defining labor organizations more narrowly than is the case in Section 2(5).

The other prong of the legislative change effort should be to both strengthen the penalties against employers for acts of anti-union discrimination and streamline the representation election process. Thus, financial penalties for employee discharge and discrimination for union activity should be substantially increased and the NLRB should be given expanded authority to seek immediate injunctive relief to remedy illegal employer acts. Likewise, NLRB administrative procedures should be streamlined in order to expedite the holding of representation elections so that elapsed time from petition to election is reduced from a median of six weeks to, say, four weeks. Finally, a new unfair labor practice provision should be written into the law that declares it illegal for an employer to create or establish any type of employee representation committee or plan once a union has filed for a representation election. All of these revisions parallel in broad outline Canadian law. The key is to ensure that dissatisfied employees in a nonunion system can rapidly and effectively unionize. We realize that there is no appetite for the kind of reform that would move American labor law in whole cloth toward the Canadian system of expedited elections with no employer campaigning (Taras 1997a). But any move to make it easier for employers to have nonunion representation must be matched by changes that increase the union threat. There are sound empirical reasons: nonunion systems work better in the presence of the union threat.

The animating idea behind this proposal is that if employees have a relatively unrestricted, low-cost means to obtain union representation then nonunion companies are effectively constrained to form and operate employee representation committees only in ways that promote mutual gain. Should the programs promote only management’s interests, or be operated in a manner that is unfair or otherwise unsatisfactory, employees can readily voice their unhappiness and replace the company’s representation plan with an independent union. The existence of a credible union threat effect thus serves as an effective competitive check on the use and purpose of NERPs. This check is then augmented, in our proposal, by an explicit ban on creation of employee committees during an organizing drive—the time “low road” employers are most likely to form an employee committee for illegitimate purposes of union avoidance. At the same time as law provides employees relatively free access to independent representation, our proposal also frees nonunion employers to establish and operate whatever form of employee representation council or committee (if any) they desire and to discuss with these groups as wide or narrow a range of topics as deemed appropriate. Those employers that are interested in long-run, constructive “high-involvement” employment practices are thus given maximum opportunity to use employee representation groups as part of their EI programs.

These revisions to the NLRA are superior, we believe, to those in two other recent proposals. The first is the recommendations contained in the Final Report of the Commission on the Future of Worker-Management Relations (“Dunlop Commission”). These recommendations include the following (Commission on the Future of Worker-Management Relations, 1994b):

  • The broad definition of a labor organization in Section 2(5) should be maintained.
  • The language of Section 8(a)(2) should also be maintained in order to prevent the re-emergence of management dominated company unions,” but a qualifying statement should be appended that permits nonunion employee representation groups to deal with employers over terms and conditions of employment as long as these discussions are incidental to issues related to productivity and quality.
  • The financial penalties for employer unfair labor practices should be strengthened, the time between petition and conduct representation elections should be shortened, and the NLRB should be given greater authority to issue injunctive relief in cases of employer acts of anti-union discrimination.

The second reform proposal is the TEAM Act legislation approved by both houses of Congress in 1996 but vetoed by President Clinton (Maryott 1997). It proposes the following changes in the NLRA:

  • The Section 2(5) definition of a labor organization should be maintained.
  • Section 8(a)(2) should be modified so that employers and employees can address matters of mutual interest, including terms and conditions of employment.
  • The prohibition of employer domination of labor organizations should be maintained for employee groups that seek certification as exclusive bargaining agents or to enter into collective bargaining.
  • The union representation election process, penalties for unfair labor practices, and NLRB administrative procedures should remain unchanged.

Relative to the recommendations advanced in this paper, it is apparent that both the Dunlop Commission proposal and the TEAM Act legislation are one-sided and unbalanced with respect to promoting competition and free choice in employee representation. The Dunlop Commission’s proposals are one-sided because they strengthen the protections given to workers to obtain independent union representation but then, having established conditions for fair and effective competition between union and nonunion representational forms, fail to go the next step and remove the tight constraints imposed by Sections 2(5) and 8(a)(2) on nonunion employers. The net effect is to promote union representation while continuing to restrict nonunion representation. The TEAM proposal is also one-sided but in the opposite direction. The legislation largely frees nonunion companies to form and operate whatever type of employee representation plan is desired, but it does nothing to strengthen the NLRA’s protection of the right to organize. The net effect is to allow employers to establish dominated labor organizations without at the same time creating the conditions (i.e., low cost, relatively unobstructed access to independent representation) necessary to insure that companies operate these groups only for mutual gain. Furthermore, TEAM legislation leaves untouched the root cause of the problem with the NLRA—the overly expansive definition of a labor organization in Section 2(5).

A labor act which sanctions only unions as the lawful vehicle by which two or more employees can discuss issues of central concern to them, the terms and conditions of their employment defeats the workers’ voice and forces a choice between unions or silence. To ban employees from being able to meet and deal with management about matters at the heart of the employment relationship, even those employees who do not wish union representation, is a draconian measure. The paradox is that employees are able to talk only about matters that mean the most to management, productivity, but are forbidden from speaking about the matters that mean the most to employees. We believe that the NLRA today is misguided in persisting to legitimize the notion of collective rights only by allowing one institutional form of collective action, unions, to prevail. To forbid employees from enjoying the benefits of collective representation because it might harm unions is to embrace a position that the worse off the law can make the nonunion worksite, the more it can help unions. This position we find unpalatable. Raudabaugh poses the interesting question “does participative management threaten employee free choice or merely make the union’s job of selling the benefits of representation more difficult?” If the answer to the former is yes, we must find solutions to preserve free choice. If the answer to the latter is yes, then unions must develop strategies to make themselves more relevant as they compete with alternative forms of representation.

We turn our attention to how consensus on reform might be developed.

The National Academy of Arbitrators, a large association of skilled union-management labor arbitrators who act as third-party neutrals in disputes, in the past decade was confronted with what likely was its most controversial issue. Arbitrators were being asked to take cases in response to the burgeoning growth in employment (nonunion) arbitration. Some nonunion companies require employees to sign agreements that they will take employment disputes to arbitration, relinquishing their statutory rights to appear before the courts or other adjudicative tribunals. After bitter fighting about whether member arbitrators should be permitted to hear nonunion cases, (and in so doing, “sell out” the organized labor movement which for decades was the mainstay of the profession) the NAA finally struck a committee that produced a document entitled “Guidelines on Arbitration of Statutory Claims Under Employer-Promulgated Systems,” (known as the Protocol). The Protocol had this to say about arbitrators who participate in nonunion arbitration schemes:

Members should recognize that in adjudicating a statutory claim they are in some respects acting as substitutes for a court rather than serving as the final step of a grievance procedure under a collective bargaining agreement. Arbitrators in statutory discrimination cases are confronted with an array of proposed procedures of varying degrees of formality. This will present the sometimes challenging question of whether procedures might be so lacking in fundamental due process that an employee claimant could not receive a fair hearing. The purpose of these guidelines is to provide an outline of practical, procedural, and evidentiary questions of application that the arbitrator might encounter in deciding whether to hear these cases and, if so, how they might be resolved...

Arbitrators should be aware that the power to withdraw from a case in the face of policies, rules, or procedures that are manifestly unfair or contrary to fundamental due process carries considerable moral suasion. . .

The Protocol then lists the many considerations that ensure fair and reasonable standards of due process in some detail -- for example, that the parties have adequate rights of representation, that the hearing location is fair, that the compensation arrangement does not lead to bias, that the arbitrator is granted full remedial authority, that the hearing respects and safeguards substantive statutory rights of the parties, and so on. An arbitrator may decline to hear a particular case. Even more extreme, groups of arbitrators may come together to denounce an employer-promulgated system as failing to meet any due process standards (and the most recent example is the NAA’s 1999 amicus brief against Hooter’s).

Perhaps a comparable step forward in drafting new legislation balancing the removal of restrictions against nonunion plans with increasing union organizing protections is to bring together a committee consisting of members of the NAA, the Federal Mediation and Conciliation Service, the Industrial Relations Research Association, the Human Resources Division of the Academy of Management, members of the labor law bar, National Labor Relations Board, designates of the AFL-CIO, and senior practitioners within companies most interested in pursuing high involvement systems.

Our proposal combines in broad outline the recommendations of the Dunlop Commission and the provisions of the TEAM Act and, in so doing, achieves a compromise solution to reform of the NLRA that serves the interests of all parties to the employment relationship. We argue that Section 2(5) should be modified to permit workers to meet and deal with management on matters of direct interest to their employment relationship in tandem with the insertion of appropriate provisions compelling greater speed in election periods, restricting management action during union organizing, and providing strong, rapid and effective remedies against the commission of unfair labor practices.

Previous Section Next Section