(Note: A shorter version of this commentary appears as a May 2006 Viewpoint)

Are federal laws inadequate to protect workers who seek to certify a union? Do employers use the federally mandated 42 days between a union’s petitioning for recognition and the certifying election to illegally turn the will of a majority of its employees against unionization?

These questions are raised in a recent study titled "Undermining the Right to Organize: Employer Behavior During Union Representation Campaigns," conducted by a research group at the University of Illinois at Chicago. Funded by American Rights at Work, an organization aligned with labor union interests, the study finds the answers to these questions to be "yes."

Stay Engaged

Receive our weekly emails!

Researchers reached this conclusion after examining the low success rate of unions seeking to organize new workers in the Chicago area, as well as the efforts of employers to defeat them, in 2002. Mark Gaffney, president of the Michigan AFL-CIO, had cited this study as support for his contention that workers want unions, but aren’t given a real opportunity to choose such representation.

These are important questions. Unfortunately, the Chicago study sheds little light on the answers, as it suffers from several weaknesses and deficiencies. For starters, the interviews cited were either of anonymous workers or union organizers. No reference is made to interviewing employers. This diminishes the usefulness of the anecdotal evidence.

The use of data is also questionable. For example, the study claims that 30 percent of employers facing potential unionization of their workforce fired a worker engaged in pro-union activities. This could be true. But it doesn’t demonstrate that actual engagement in pro-union work is why these employees were fired, instead of some other, legally justifiable reason. Indeed, actual National Labor Relations Board cases arising in Chicago do not support the conclusion of the study. The unions did not even file an unfair labor practice complaint in 70 percent of the alleged cases. The study argues that this was because of the difficulty in proving such a claim, but another possibility is that many of the claims were unfounded.

Perhaps most importantly, the Chicago study fails to analyze alternative explanations as to why unions seemingly lose support between the time an election is called and when it occurs. Instead, the study simply reaches the predetermined conclusion that fault lies with employers and their misconduct.

One alternative explanation worth exploring is that the union never actually enjoyed the original support it claimed. Such a situation is plausible, given the structure of federal labor law.

The National Labor Relations Act establishes a two-step process for union certification. The first step is obtaining, through a petition or card drive, evidence suggesting that at least 30 percent of employees want to be represented by a union. If so, the second step is to hold a secret ballot certification election 42 days later. If a majority votes for union representation, the union is certified. Accordingly, a certifying election must be held even if the union did not initially have majority support.

According to the Chicago study, 91 percent of the unions enjoyed at least 50 percent support at the petition stage, and in "several cases," unions demonstrated more than 80 percent support. So, what happened at the election stage? Employer intimidation?

We can’t be sure because the Chicago study did not correlate petition support with actual election outcomes. We cannot tell the fate of the 9 percent of the unions that did not have majority support when they asked for an election, nor can we correlate the specific level of initial support with success in the certifying election. All we know is that about one-third of the petitions for an election resulted in successful certifications; about one-third were withdrawn pre-election for lack of support; and about one-third failed to achieve a majority vote. We can speculate, however, that a worker might sign a public petition at the request of a friend, but vote another way when protected by privacy.

Unfortunately, the Chicago study does not address this issue, but simply utilizes questionable anecdotal evidence to conclude that employer intimidation must be the culprit.

Comparing pre-election petitions with the results of secret ballot elections from anecdotal accounts is of limited value. The problem is akin to the unreliability of pre-election polling. The human experience is that it is much harder to tell an ideologically driven pollster whom you intend to vote for than to vote one’s conscience in a secret ballot. Likewise, it is much easier to vote "no" on a secret ballot then to tell a petition-pushing friend that you are not interested in union representation.

But the importance of the secret ballot in the union certification is shown by the value union members place on it. In June 2004, Zogby International and the Mackinac Center for Public Policy conducted a survey of union members to gauge their attitudes toward their union’s performance. According to the survey report, "majorities (ranging from 53 percent to 84 percent) prefer a secret-ballot process as the way for workers to decide whether to organize a union in their workplace; a strong majority (66 percent) think it should be illegal for a union and a company to agree in advance to bypass the secret-ballot union election when organizing a workplace." Unionized employees undoubtedly support the secret ballot for many reasons, including fear of reprisals from their employer. But it should not be forgotten that one reason may also be fear of union coercion during an organizing drive. Remember that not every unionized employee voted for a union.

United States law prohibits employers from interfering with or discriminating against employees seeking to organize a union. Stung by its lack of success in certifying elections, organized labor advocates the passage of federal legislation that would give a union the right to declare, on the basis of a petition-like drive and without a secret election, that the union enjoys majority support and must be certified. The Chicago study will undoubtedly be cited by labor in support of this measure.

Common sense counsels that we keep secret-ballot certification elections. The conclusion of the 2004 Mackinac Center report remains valid: "[Union members’ preference for private elections] suggests crafting practices and policies that guarantee workers’ ability to vote on union representation and respect the individual choices of workers to support a union or not." Anything else would undermine the fairness necessary for the right to organize.


Thomas W. Washburne is director of labor policy for the Mackinac Center for Public Policy, a research and educational institute headquartered in Midland, Mich. Permission to reprint in whole or in part is hereby granted, provided that the author and the Center are properly cited.