A joint U.S.-Canadian conference convened last year to explore and gain insights into nonunion forms of representation. Held in Banff, Alberta, the conference sought to promote a dialogue on NUR, and address worker issues, legal and consulting strategies, public policy, union perspectives, and management concerns around the issue. The conference was organized by Professors Bruce E. Kaufman of Georgia State University and Daphne G. Taras of the University of Calgary, with financial support provided by the Government of Canada Human Resources Development’s Labour-Management Partnership Program, Imperial Oil, Industrial Relations Counselors, Inc., the John M. Olin Institute for Employment Practice and Policy, and the Industrial Relations Research Group and Faculty of Management at the University of Calgary. Although the proceedings are slated to be published at a later date, we believe that the topic of nonunion forms of employee representation is of sufficient importance to report here briefly on some of the conference discussion.
Giving Employees Voice
The concepts of “exit” and “voice” have been applied by Richard B. Freeman of Harvard University to the labor market and employer-employee relations. Freeman said that if workers are unhappy on the job they have the option of quitting, but the presence of a union gives them an alternative, a “collective voice” through which they can try to change the conditions that are making them unhappy.1 Labor leaders and the advocates of unionism therefore contend that the decline in unionization and collective bargaining has left American workers without voice.
A major reason why workers lack voice is that the National Labor Relations Act bars nonunion forms of representation (NUR), which were quite extensive prior to 1935, when the act was passed. Actually, employees can have voice, albeit outside the areas covered by collective bargaining, through such devices as alternative dispute resolution procedures, committees dealing with specific problems of the workplace, e.g, safety and health, and employee involvement programs. Some of these programs provide individual rather than collective voice. Employee involvement (EI), for example, does not include representation, but gives workers a say in decisions that affect their own work. Even so, it has been subject to charges of running afoul of the NLRA ban on nonunion representation.
This conundrum whereby we decry the disappearance of voice in the workplace while at the same time declaring illegal attempts to provide it, except in a unionized setting, is peculiar to the United States. In Canada there is no such problem; nonunion forms of employee representation are legal in Canada and exist alongside unions.
The Employee Representation Plan
What became the predominant form of non-union employee representation was a fusion of Canadian and U.S. ideas. Early this century, William Lyon Mackenzie King, active as a labor conciliator in Canada, was sympathetic to union calls for justice and better conditions for workers, but wary of unions themselves. He favored labor-management cooperation, which required that workers be collectively represented in their dealings with management, but he believed that those representatives should come from within the companies’ workforces.
Papers by Professor Bruce E. Kaufman of Georgia State University and Professor Daniel Nelson of the University of Akron show how King’s ideas took hold in the United States. King was afforded the opportunity to put his ideas into action in the United States following a violent strike in 1914 by the United Mine Workers at the Colorado Fuel and Iron Company, in which women and children died. Dubbed the Ludlow Massacre, the incident brought disrepute to John D. Rockefeller, Jr., the major CFI shareholder, and influenced him to ask King how CFI might foster cooperation between labor and management. King formulated a plan granting individual workers the right to appeal grievances and establishing a joint industrial council (JIC), to which the workers would elect representatives. The plan was designed to change the behavior of company managers as much as that of the workers. Adopted by CFI in 1915 and approved by an overwhelming vote of the workforce, the JIC plan bore tangible results: working conditions, housing, and safety all improved; wages became more equitable; and supervisors lost their power to terminate employees. As a result of its success at CFI, the employee representation plan (ERP), as it became known, spread throughout North America, becoming the most widely practiced nonunion representation plan.
Kaufman sees the 1920s ERP plan as part of a new “high-performance” human resource management paradigm. Recognizing that there were conflicting interests between employers and employees, progressive managers saw the ERP as a means of settling those differences in a fair and equitable manner. Companies’ commitment to ERPs was unrelated to concerns about union organization; indeed, many of those companies with such plans, including Standard Oil of New Jersey, AT&T, and Du Pont, faced no threat from outside unions, not even at the height of organizing efforts in the 1930s.
The Legislative Assault on ERPs
According to Nelson, it was their imperviousness to unionization that was responsible for the aggressive union campaign against employee representation plans, which culminated in Section 8(a)(2) of the National Labor Relations Act. Because there was little likelihood that employers with ERPs or their workers would voluntarily embrace outside unions, organizing campaigns were not likely to be cost-effective. Legislation banning ERPs was a comparatively inexpensive alternative.
Well-established, legitimate employee representation plans also fell victim to imitators with other motives. As union organizing exploded in the 1930s, a number of companies set up crude copies of ERPs, often simply to ward off unionization. (In fact, these jerry-built bodies frequently became halfway houses on the road to unionism.) Organized labor, however, used them to charge that all nonunion representation was a fraud and simply another big-business device to thwart unionism.
These sham ERPs inflamed public opinion. Just as Kaufman points out, the best approach to preventing such possible abuse was not to ban nonunion plans, but to create conditions for effective employee free choice. The National Labor Relations Act (Wagner Act) was supposed to do just that, by establishing the procedure for secret-ballot representation elections by workers. The only problem is that according to interpretations by the National Labor Relations Board and courts, sections 8(a)(2), which banned employer domination of a labor organization, and 2(5), which very broadly defined a labor organization, thereby excluded nonunion representation plans, whether workers wanted them or not. (See “Sections of the Wagner Act Inhibiting Nonunion Employee Representation.”) The NLRB then proceeded to root out nonunion representation plans wherever they were encountered.
Why was the law written so as to enable the elimination of nonunion forms of representation, even though evidence showed that in many instances employees preferred them? Professor Samuel Estreicher of New York University Law School argues that Senator Robert F. Wagner, the NLRA’s author, was fearful that employers could use ERPs to twist employee preferences against outside representation. Kaufman, however, claims that Wagner was less interested in promoting employee voice than in carrying out the New Deal economic program. Behind the idea that union organization was good lay a purchasing-power theory of depressions, which required that wages be raised to spur demand for goods and services; Wagner believed that unions would be more effective than ERPs in pushing up wages. Whatever the motive, the result was that the trade union movement was able to destroy the employee representation plan through legislation, something it could never do by winning over members of those plans.
Local Independent Unions
Not all ERPs, however, disappeared or were absorbed by outside unions. A number broke free and established themselves as independent local unions (ILUs), representing the employees of a particular plant or company but not affiliated with a national union. Nevertheless, membership in local independent unions dropped from 2.5 million in 1935 to 0.4 million by 1941. According to Professor Sanford M. Jacoby of the University of California-Los Angeles, the ability of the surviving ILUs to withstand both NLRB scrutiny and raiding attempts by national unions encouraged their growth after World War II. They were also aided by the Taft-Hartley Act provision barring the NLRB from discriminating between affiliated and unaffiliated unions.
The ILUs tended to be conservative, which helped them in smaller companies and outside big cities and, since they rarely went on strike, made them more acceptable to employers. In urban areas, they held the loyalty of members by servicing them better than did business agents who had to deal with numerous affiliated locals. ILU members also wanted to preserve compensation premiums that might otherwise be eliminated by national unions. Finally, ILUs drew support from groups who felt that national unions were insensitive to their needs and concerns. Black workers at Swift, for example, much preferred the ILU to the AFL’s Meatcutters Union, which had a long history of discrimination, and the large female workforce of the Bell System was turned off by the aggressive, take-charge attitudes of national union organizers.
After the late 1950s, however, ILUs went into a decline from which they never recovered. Jacoby attributes that decline to a change in management attitudes. Indeed, as the number of easily organizable nonunion enterprises steadily shrank, ILUs were raided by national unions. Some ILUs played on management fears of national union affiliation and won gains that raised labor costs above those at plants with national unions. Management recognized how costly this strategy was, and turned away from the ILUs.
Other companies sought to provide voice by setting up employee committees with which to discuss workplace issues but steer clear of collective bargaining. That approach came to a halt, however, after the Supreme Court, in its 1959 Cabot-Carbon decision, ruled that an employee committee that does not bargain with management nevertheless is a labor organization, as defined in Section 2(5), and governed by Section 8(a)(2), which forbids management domination. Prudent employer conduct was to abandon employee committees, and management in nonunion establishments turned to the behavioral sciences, focusing employee influence primarily on the work group.
Nonunion Employee Representation: A Different Story in Canada
Professor Daphne G. Taras of the University of Calgary has examined Canadian experience with nonunion employee representation. In 1919 a national industrial conference unanimously adopted a resolution approving the formation of “joint industrial councils as a means of furthering greater cooperation between employer and employees.” NUR was prevalent in the 1920s, particularly at branch plants of U.S. corporations. The industrial council also appealed to Canadian employers that were adopting American business philosophies of professional centralized approaches to personnel matters.
The eruption of industrial unionism in the United States in the 1930s spilled over to Canada, with major growth coming during World War II. Collective bargaining legislation-similar to the Wagner Act in the United States-was introduced in 1944, but with no ban on nonunion forms of representation, which might have impeded worker-management cooperation in many operations. Some Canadians, moreover, preferred “independent company unions” to foreign- (American-) controlled national ones. The Trades and Labor Congress (the AFL-affiliated body in Canada) did not push to ban ERPs, fearing that might give organizing momentum to its bitter rival, the CIO-affiliated Canadian Congress of Labor.
Because Canadian law defines labor organizations narrowly, Canadian firms freely operate formal nonunion plans that include many of the elements specifically targeted for eradication in the United States: for example, worker election of representatives, meeting on company time and premises, management representation in the plan, discussion of wages and working conditions, and drawing up documents similar to collective bargaining agreements. Such agreements, however, are not a shield against union organizing, and unions can raid nonunion plans at any time. Today, as a result, a strong Canadian union movement coexists with a relatively healthy environment for nonunion representation, while the U.S. union movement has weakened considerably in the presence of legal restrictions on nonunion representation. (See sidebar for views on one of the most enduring JICs in Canada.)
Evaluation of NUR: Findings from Three Disciplines
The Banff conference presented assessments of NUR from three disciplines.
In the late 1980s, faced with lagging productivity and stiff global competition, American companies adopted computerized technology, which forced changes in work organization. They adopted mechanisms that empowered workers to participate, at the work-unit level, in problem solving and decisions about such issues as work processes, quality, safety, and customer service. These innovations helped improve American manufacturing efficiency and competitiveness.
According to Estreicher, employers had been able to live with Section 8(a)(2) because its premises fit the more traditional Tayloristic workplace, with its narrowly defined tasks and hierarchical decision making. The proliferation of work teams-with their committed, “smart” production workers participating in problem solving, troubleshooting, and decision making-render inappropriate the unilateral styles contemplated by 8(a)(b). Yet these work innovations put management at potential risk, as occurred in the NLRB’s Electromation decision, which found such committees to violate the labor-law ban on company-dominated labor organizations.
The case for employee committees has typically been made on either political grounds-democracy is good-or grounds for distributional equity-the need to overcome employees’ weak bargaining power. Professor David Levine of the University of California at Berkeley posited an additional argument: economic efficiency. Safety committees, mandated by law in ten states, show evidence of being very effective for both employers and employees. In Oregon between 1990 and 1993 such committees contributed to a $1.5 billion decrease in workplace injury costs. In a world with completely free markets, such employee committees would appear wherever they are efficient mechanisms for employee voice. In reality, market failures affect workplace issues like safety, training, or other HR matters and can prevent the establishment of such committees where they are most needed. Levine believes that incentives targeted at the establishment of employee committees in these areas are likely to prove more efficient than broad legislative mandates.
Professor Tove Hammer of Cornell University pointed out that, according to need-motive-value theories, people are motivated by curiosity and need for challenge, autonomy, and mastery, which can be satisfied through work designed to allow psychological growth and development. In the United States, collective bargaining by unions to decide wages, conditions of employment, and dispute settlement-done with only indirect worker participation-addresses some needs but largely ignores workers’ psychological needs. On the other hand, recent innovations in nonunion forms of representation, such as employee involvement, job redesign, and team-based production systems, which involve direct worker participation, have as their goals better job performance and increased job satisfaction, rather than employee power. She contends that joint committees, adaptations of the works council model, or representation on corporate boards of directors would provide vehicles for employee voice in decisions about terms and conditions of employment.
The Current Legal Dilemma
The essence of employee involvement initiatives is sharing with employees authority and responsibility, otherwise vested in company owners, to make decisions concerning the process of production and distribution of rewards and costs. This sharing may be legislatively mandated (as in Germany), won through economic pressure or legal certification (collective bargaining), or voluntarily bestowed by the employer.
In case studies of eight U.S. companies, Professors Kaufman, David Lewin (UCLA), John Fossum (University of Minnesota), and Anil Verna (University of Toronto) found a significant gap between what is permissible under a strict reading of the law and what is done in some of the companies’ EI programs. If the Electromation decision were applied, most would have to discard or substantially modify significant portions of their programs. Even so, many managers reported that efforts to comply with the law had led to counterproductive compromises: To avoid a charge of “dealing with” employees, some employers had ruled all employment-related issues “off limits.” Other companies, to avoid bilateral interaction, had either completely delegated authority to the employees or had limited a committee’s role to communication and information exchange, with no role in decision making. The law forces companies to focus EI on the things of interest to management-productivity and quality-but to avoid talking about things of primary concern to employees-wages, hours, and working conditions. In the researchers’ opinion, the real losers are the employees.
After a review of 17 NLRB cases, a survey of 111 work teams from six large companies, and interviews with six management attorneys, Professor Michael H. LeRoy of the University of Illinois supported this perspective. He concludes that the consensus mode for managing teams probably makes many of them unlawful. Since they do provide employees with a moderately high degree of workplace democracy, LeRoy advocates modifying Section 8(a)(2) to allow these teams to handle a wider range of subjects (e.g., health and safety) and to permit legitimate consensus decision making among managers and employees.
Congress had attempted to remedy the EI dilemma by passing the Teamwork for Employees and Managers Act of 1996 (TEAM). This legislation would have amended 8(a)(2) to permit employers to establish and support committees of supervisors and employees that would “address matters of mutual interest, including, but not limited to, issues of quality, productivity, efficiency, and safety and health…” but, under pressure from organized labor, President Clinton vetoed it. Employers were left to walk a wobbly line between doing what seems in the best interest of both the business and its employees and adhering strictly to the letter of a law that seems dubious at best.
The U.S.: Any Future for Nonunion Employee Representation?
The United States finds itself in a strange situation. Everyone proclaims the advantages of employee empowerment and employee-management cooperation, yet the labor law declares illegal many attempts at such programs. There is evidence that employees want some voice as to their working conditions, but the only way for them to have it is through unions, which 90 percent in the private sector have not embraced. So while a company may have its employees work in teams, if it allows team members a say in how they should be remunerated it becomes guilty of an unfair labor practice.
One approach that many companies are following is simply to ignore the NLRA. An employer is at jeopardy only when a union files an unfair labor practice charge. With fewer than one of ten private sector workers unionized, and membership continuing to decline in a most robust economy, the chances of a union organizing attempt that could prompt an unfair labor practice charge are very slim. Consequently, employers are willing to forge ahead with employee empowerment.
Disregard of the law, however, is not a posture that companies like to assume. Employer strategy has been to fight for the TEAM Act, but that has met with a presidential veto. Some labor lawyers, moreover, are dubious about how much TEAM would achieve and believe that the better course would be to try to amend the NLRA. Even here, however, there is disagreement, with some wanting to concentrate on 8(a)(2), and others saying that changing 2(5), the very broad definition of a labor organization, would be sufficient.
No change in the law, however, is likely to bring back the employee representation plan. It was very successful in its time, and had the Wagner Act not made it illegal, it would have survived and given many more American employees today a say in the terms and conditions of their work. Employers, however, have moved beyond the ERP, and organized labor would use all its power to prevent its return.
Despite the legal situation, American companies have been empowering their workforces, which, combined with new technology and organizational systems, has made a significant contribution to higher rates of productivity in the manufacturing sector and to the restoration of U.S. competitiveness. If the law were amended to allow further empowerment, we might do still better.
If the conference produced no solutions, it did clearly show how ridiculous the American situation has become. The Canadian example shows that employee voice would pose no threat to organized labor, since Canada allows nonunion representation and has a much higher degree of unionization. One way to undercut organized labor’s opposition to allowing employees to have greater voice in the workplace might be to combine any change in the law with greater protection for legitimate organizing activities. After all, the aim is the same-ensuring that employees have the freedom to choose whatever form of representation they desire in order to achieve greater voice in the workplace.
Richard B. Freeman, “Individual Mobility and Worker Voice in the Labor Market,” American Economic Review, May 1976.