Statement of the U.S. Chamber of Commerce
ON: “EMPLOYEE FREE CHOICE ACT – UNION CERTIFICATION”
TO: SENATE SUBCOMMITTEE ON LABOR, HEALTH AND HUMAN SERVICES, AND EDUCATION OF THE COMMITTEE ON APPROPRIATIONS
BY: CHARLES I. COHEN
DATE: JULY 16, 2004
Statement of Charles I. Cohen Senior Partner, Morgan, Lewis & Bockius LLP on behalf of the United States Chamber of Commerce before the Subcommittee on Labor, Health and Human Services, and Education of the Senate Committee on Appropriations
Chairman Specter and Members of the Subcommittee, I am pleased and honored to be here today. Thank you for your kind invitation.
By way of introduction, I was appointed by President Clinton, confirmed by the Senate, and served as a Member of the National Labor Relations Board from March 1994 until my term expired in August 1996. Before becoming a Member of the Board, I worked for the NLRB in various capacities from 1971 to 1979 and as a labor lawyer representing management in private practice from 1979 to 1994. Since leaving the Board in 1996, I have returned to private practice and am a Senior Partner in the law firm of Morgan, Lewis & Bockius LLP. I am a member of the Labor Relations Committee of the U.S. Chamber of Commerce, and Chair of its NLRB subcommittee, and am testifying today on behalf of the U.S. Chamber of Commerce.
The National Labor Relations Act was enacted in 1935 and has been substantially amended only twice—once in 1947 and once in 1959. The Act establishes a system of industrial democracy which is similar in many respects to our system of political democracy. At the heart of the Act is the secret ballot election process administered by the National Labor Relations Board. In order to understand how recent trends in organizing are diluting this central feature of the Act, some background is necessary.
The NLRB’s Secret Ballot Election Process
If a group of employees in an appropriate collective bargaining unit wish to select a union to represent them, the Board will hold a secret ballot election based on a petition supported by at least thirty percent of employees in the unit. The Board administers the election by bringing portable voting booths, ballots, and a ballot box to the workplace. The election process occurs outside the presence of any supervisors or managerial representatives of the employer. No campaigning of any kind may occur in the voting area. The only people who are allowed in the voting area are the NLRB agent, the employees who are voting, and certain designated employee observers.
The ultimate question of union representation is determined by majority rule, based on the number of valid votes cast rather than the number of employees in the unit. If a majority of votes are cast in favor of the union, the Board will certify the union as the exclusive bargaining representative of all employees in the collective bargaining unit. Once a union is certified by the Board, it becomes the exclusive representative of all of the unit employees, whether or not they voted for the union. The employer is obligated to bargain with the union in good faith with respect to all matters relating to wages, hours, and working conditions of the bargaining unit employees.
The Board is empowered to prosecute employers who engage in conduct that interferes with employee free choice in the election process, and may order a new election if such employer interference with the election process has occurred. The Board will also order the employer to remedy such unfair labor practices, for example by ordering the employer to reinstate and compensate an employee who was unlawfully discharged during the election campaign. In extreme cases, the Board may even order an employer to bargain with the union without a new election, if the Board finds that its traditional remedies would not be sufficient to ensure a fair rerun election and if there is a showing that a majority of employees at one point desired union representation. The Supreme Court affirmed the Board’s power to issue this extraordinary remedy in NLRB v. Gissel Packing Co., 395 U.S. 575 (1969). When issuing a Gissel bargaining order, the Board will determine whether majority support for the union existed by checking authorization cards signed by employees during the organizing process.
As the Board and the Supreme Court have acknowledged, the use of authorization cards to determine majority support is the method of last resort. A secret ballot election is the “most satisfactory—indeed the preferred—method of ascertaining whether a union has majority support.” Gissel Packing, 395 U.S. at 602. Unions likewise prefer an NLRB secret ballot election, at least when they are faced with a potential loss of majority support. In Levitz Furniture Co. of the Pacific, 333 NLRB 717 (2001), the United Food and Commercial Workers, supported by the AFL-CIO as amicus curiae, took the position that “Board elections are the preferred means of establishing whether a union has the support of a majority of the employees in a bargaining unit.” Id. at 719 (emphasis added). The Board agreed with the unions’ position in Levitz. See id. at 725 (“We agree with the General Counsel and the unions that Board elections are the preferred means of testing employees’ support.”).
Although authorization cards may adequately reflect employee sentiment when the election process has been impeded, the Board and the Court in Gissel recognized that cards are “admittedly inferior to the election process.” Gissel Packing, 395 U.S. at 602. Other federal courts of appeal have expressed the same view:
• “[I]t is beyond dispute that secret election is a more accurate reflection of the employees’ true desires than a check of authorization cards collected at the behest of a union organizer.” NLRB v. Flomatic Corp., 347 F.2d 74, 78 (2d Cir. 1965).
• “It would be difficult to imagine a more unreliable method of ascertaining the real wishes of employees than a ‘card check,’ unless it were an employer’s request for an open show of hands. The one is no more reliable than the other…. Overwhelming majorities of cards may indicate the probable outcome of an election, but it is no more than an indication, and close card majorities prove nothing.” NLRB v. S. S. Logan Packing Co., 386 F.2d 562, 565 (4th Cir. 1967).
• “The conflicting testimony in this case demonstrates that authorization cards are often a hazardous basis upon which to ground a union majority.” J. P. Stevens & Co. v. NLRB, 441 F.2d 514, 522 (5th Cir. 1971).
• “An election is the preferred method of determining the choice by employees of a collective bargaining representative.” United Services for the Handicapped v. NLRB, 678 F.2d 661, 664 (6th Cir. 1982).
• “Although the union in this case had a card majority, by itself this has little significance. Workers sometimes sign union authorization cards not because they intend to vote for the union in the election but to avoid offending the person who asks them to sign, often a fellow worker, or simply to get the person off their back, since signing commits the worker to nothing (except that if enough workers sign, the employer may decide to recognize the union without an election).” NLRB v. Village IX, Inc., 723 F.2d 1360, 1371 (7th Cir. 1983).
• “Freedom of choice is ‘a matter at the very center of our national labor relations policy,’ … and a secret election is the preferred method of gauging choice.” Avecor, Inc. v. NLRB, 931 F.2d 924, 934 (D.C. Cir. 1991) (citations omitted).
Having recognized in Gissel that a secret ballot election is the superior method for determining whether a union has majority support, the Supreme Court in Linden Lumber v. NLRB, 419 U.S. 301 (1974), held that an employer may lawfully refuse to recognize a union based on authorization cards and insist on a Board-supervised secret ballot election. The only exceptions to an employer’s right to insist on an election are when the employer, as in the Gissel situation, has engaged in unfair labor practices which impair the electoral process or when the employer has agreed to recognize the union based on a check of authorization cards. Thus, an employer can agree to forgo a secret ballot election and abide by the less reliable card check method of determining union representation.
The Increasing Use of Neutrality/Card Check Agreements in Organizing Campaigns
One of the highest priorities of unions today is to obtain agreements from employers which would allow the union to become the exclusive bargaining representative of a group of employees without ever seeking an NLRB-supervised election. These agreements, which are often referred to as “neutrality” or “card check” agreements, come in a variety of forms. In some cases, the agreement simply calls for the employer to recognize the union if it produces signed authorization cards from a majority of employees. In many cases, the agreement includes other provisions which are designed to facilitate the union’s organizing campaign, such as:
• An agreement to provide the union with a list of the names and addresses of employees in the agreed-upon unit;
• An agreement to allow the union access to the employer’s facilities to distribute literature and meet with employees;
• Limitations or a “gag order” on employer communications to employees about the union;
• An agreement to start contract negotiations for the newly-organized unit within a specified (and short) time frame, and to submit open issues to binding interest arbitration if no agreement is reached within that time frame; and
• An agreement to extend coverage of the neutrality/card check agreement to companies affiliated with the employer.
Whatever form the agreement may take, the basic goal is the same: to establish a procedure that allows the union to be recognized without the involvement or sanction of the National Labor Relations Board. Neutrality and card check agreements therefore present a direct threat to the jurisdiction of the Board and its crown jewel, the secret ballot election process. I have written two law review articles discussing this trend. See Charles I. Cohen, Neutrality Agreements: Will the NLRB Sanction Its Own Obsolescence?, The Labor Lawyer (Fall, 2000); Charles I. Cohen and Jonathan C. Fritts, The Developing Law of Neutrality Agreements, Labor Law Journal (Winter, 2003).
The motivating force behind neutrality/card check agreements is the steady decline in union membership among the private sector workforce in the United States. Unions today represent only about 8% of the private sector workforce, about half of the rate 20 years ago. See U.S. Dep’t of Labor, Bureau of Labor Statistics, Union Members in 2003 (Jan. 21, 2004), available at http://www.bls.gov/news.release/pdf/union2.pdf. There are many explanations for this precipitous decline: the globalization of the economy and the intense competition that comes with it, the increasing regulation of the workplace through federal legislation rather than collective bargaining, and the changing culture of the American workplace. While unions may not disagree with these explanations to varying degrees, they claim that the NLRB’s election process is also to blame. Unions argue that the NLRB’s election process is slow and ineffective, and therefore an alternative process is needed—namely, neutrality/card check agreements.
I believe there are two basic problems with this argument. First, it is not supported by the facts. The NLRB’s election process is efficient and fair, as demonstrated by hard statistics. Second, neutrality/card check agreements limit employee free choice and are generally the product of damaging leverage exerted by the union against the employer.
The NLRB’s Election Process Is Efficient and Fair
The standard union criticisms of the NLRB’s election process are more rhetorical than factual. Unions argue that the NLRB’s election process is slow and allows employers to exert undue influence over employees during the pre-election period. Both of these arguments are not supported by the facts.
The NLRB’s election process is not slow. In fiscal year 2003, 92.5% of all initial representation elections were conducted within 56 days of the filing of the petition. Memorandum GC-04-01, Summary of Operations (Fiscal Year 2003), at p. 5 (December 5, 2003), available at http://www.nlrb.gov/nlrb/shared_files/gcmemo/gcmemo/gc04-01.pdf?useShared=/nlrb/about/reports/gcmemo/default.asp. During that same time period, the median time to proceed to an election from the filing of a petition was 40 days. Id. Based on my experience over the past 30 years, these statistics demonstrate that the Board’s election process has become even more efficient over time.
Unions are currently winning over 50% of NLRB secret ballot elections involving new organizing. This is the category of elections that unions are seeking to replace with neutrality/card check agreements, and it is also the same category of elections that would be replaced by the Miller-Kennedy bill. If anything, unions’ win rate in representation elections is on the rise. The NLRB’s most recent election report shows that unions won 58.9% of all elections involving new organizing. See NLRB Election Report; 6-Months Summary – April 2003 through September 2003 and Cases Closed September 2003, at p. 19 (March 26, 2004). This figure is about the same as it was 40 years ago. In 1965, unions won 61.8% of elections in RC cases (cases which typically involve initial organizing efforts, as opposed to decertification elections or employer petitions). See Thirtieth Annual Report of the National Labor Relations Board, at p. 198 (1965). After 1965, unions’ election win rate declined before rising back to the level where it is today:
• In 1975, unions won 50.4% of elections in RC cases. See Fortieth Annual Report of the National Labor Relations Board, at p. 233 (1975).
• In 1985, unions won 48% of elections in RC cases. See Fiftieth Annual Report of the National Labor Relations Board, at p. 176 (1985).
• In 1995, unions won 50.9% of elections in RC cases. See Sixtieth Annual Report of the National Labor Relations Board, at p. 153 (1995).
These statistics undermine any argument that the NLRB’s election process unduly favors employers, or that the recent decline in union membership among the private sector workforce is attributable to inherent flaws the NLRB’s election process. Unions are winning NLRB elections at the same or higher rate now than they have in almost 40 years. To be sure, there are “horror stories” of employers who abuse the system and commit egregious unfair labor practices in order to prevail in an election. In such cases, the law provides remedies for the employer’s unlawful behavior, including Gissel bargaining orders. But these situations are the exception rather than the norm. In the overwhelming majority of cases where employees choose not to be represented by a union, they do so based on the information that is presented by both sides during the campaign process.
Unions attempt to portray the Board’s secret ballot election process as fundamentally unfair (except when unions are faced with a challenge to their majority status) by making unfavorable comparisons between Board elections and a typical political election in the United States. In doing so, unions frequently ignore several important facts about the NLRB election process:
• The union controls whether and when an election petition will be filed. Imagine if the challenger in a political election controlled the timing of the election.
• The union largely controls the definition of the bargaining unit in which the election will occur, because the union need only demonstrate that the petitioned-for unit is an appropriate bargaining unit. Imagine if the challenger in a political election had almost irreversible discretion to gerrymander the voting district to its maximum advantage.
• The union usually has obtained signed authorization cards from a majority of employees at the time the petition is filed. Thus, the union already knows the voters and has conducted a straw poll before the employer is even aware that an election will be held. Imagine if the challenger in a political election could campaign and poll the electorate without the incumbent’s knowledge, wait until the polls show that the challenger has majority support, and then give the incumbent less than 60 days’ notice of the election.
• Even though the union already knows the voters well by the time the election petition is filed, the employer must give the union a list of all of the voters’ names and home addresses after the petition is filed. The union, but not the employer, is then permitted to visit the employees at home to campaign for their vote.
• The union, unlike the employer, can make campaign promises to the employees to induce them to vote for the union.
• The union, like the employer, may designate an observer to be present in the voting area for the duration of the election, in order to check every voter and make sure that no irregularities occur.
These facts illustrate that, far from being unfair to unions, the NLRB’s election process offers unions many unique advantages.
Problems with Neutrality/Card Check Agreements
The fundamental right protected by the National Labor Relations Act is the right of employees to choose freely whether to be represented by a union. 29 U.S.C. § 157. Neutrality/card check agreements limit employee free choice by restraining employer free speech. Section 8(c) of the Act protects the right of employers to engage in free speech concerning union representation, as long as the employer’s speech does not contain a threat of reprisal or a promise of benefit. 29 U.S.C. § 158(c). Unions, through neutrality/card check agreements, seek to restrain lawful employer speech by prohibiting the employer from providing employees with any information that is unfavorable to the union during the organizing campaign. Such restrictions or “gag orders” on lawful employer speech limit employee free choice by limiting the information upon which employees make their decision.
A second problem with neutrality/card check agreements is the method by which they are negotiated. In my experience, neutrality/card check agreements are almost always the product of external leverage by unions, rather than an internal groundswell from unrepresented employees. The leverage applied by the union can come from a variety of sources. In many cases, the union has leverage because it represents employees at some of the employer’s locations. The union may be able to use leverage it has in negotiations for employees in an existing bargaining unit, in order to win a neutrality/card check agreement that will facilitate organizing at other locations. Bargaining over a neutrality/card check agreement, however, has little or nothing to do with the employees in the existing bargaining unit, and it detracts from the negotiation of the core issues at hand—wages, hours, and working conditions for the employees the union already represents.
In other cases, the union exerts pressure on the employer through political or regulatory channels. For example, if the employer needs regulatory approval in order to begin operating at a certain location, the union may use its political influence to force the employer to enter into a neutrality/card check agreement for employees who will be working at that location. Political or regulatory pressure may be coupled with other forms of public relations pressure in order to exert additional leverage on the employer. In general, this combination of political, regulatory, public relations and other forms of non-conventional pressure has become known as a “corporate campaign,” and it is this type of conduct—rather than employee free choice—that has produced these agreements.
Thus, when a union succeeds in obtaining a neutrality/card check agreement, it generally does so by exerting pressure on the company through forces beyond the group of employees sought to be organized. The pressure comes from bargaining units at other locations, and/or it comes from politicians, regulators, customers, investors, and the public at large. It is a strategy of “bargaining to organize,” meaning that the target of the campaign is the employer rather than the employees the union is seeking to organize.
The strategy of “bargaining to organize” stands in stark contrast to the model of organizing under the National Labor Relations Act. Under the Act, the pressure to organize comes from within—it starts with the employees themselves. If a sufficient number of employees (30%) desire union representation, they may petition the NLRB to hold a secret ballot election. If a majority vote in favor of union representation, the NLRB certifies the union as the employees’ exclusive representative and the collective bargaining process begins at that point. At all times, the focus is on the employees, rather than on the employer or the union.
There is no cause for abandoning the secret ballot election process which the Board has administered for 7 decades. The Act’s system of industrial democracy has withstood the test of time because its focus is on the true beneficiaries of the Act—the employees. In my view, the Miller-Kennedy bill is not sound public policy because it would deprive employees of the fundamental right to determine the important question of union representation by casting their vote in a Board-supervised secret ballot election. Indeed, that it would be unwise public policy to abandon government-supervised secret ballot elections in favor of mandatory card check appears to me to be a self-evident proposition. It likewise would eviscerate the proud tradition of industrial democracy that has been the hallmark of the NLRB for nearly seven decades.
I am aware that Senator Lindsey Graham (R-SC) has introduced quite opposite legislation, S. 2637, which would require that union representation for currently unrepresented groups of employees be determined by a secret ballot election. Without the increasing use of corporate campaigns and neutrality/card check agreements over the last decade—a trend that has eroded employee free choice and reflects a shift in focus from organizing employees to organizing employers—such legislation would not be needed. But, in light of this trend, such legislation, in my view, is necessary to protect the interests of the employees the Act is intended to benefit, by ensuring that their right to vote is not compromised by agreements that are the product of external pressure on their employer.
The Miller-Kennedy Bill’s Interest Arbitration Provisions
In addition to mandating recognition by card check rather than a secret ballot election, the Miller-Kennedy bill would eviscerate another fundamental tenet of U.S. labor law: voluntary agreement. As the Supreme Court held in H. K. Porter v. NLRB, 397 U.S. 99 (1970), the Act is founded on the notion that the parties, not the government, should determine the applicable terms and conditions of employment:
The object of this Act was not to allow governmental regulation of the terms and conditions of employment, but rather to ensure that employer and their employees could work together to establish mutually satisfactory conditions. The basic theme of the Act was that through collective bargaining the passions, arguments, and struggles of prior years would be channeled into constructive, open discussions leading, it was hoped, to mutual agreement. But it was recognized from the beginning that agreement might in some cases be impossible, and it was never intended that the Government would in such cases step in, become a party to the negotiations and impose its own views of a desirable settlement.
Id. at 103-04 (emphasis added). The Miller-Kennedy bill would destroy this bedrock principle of the Act by mandating that, if the parties are not able to reach agreement on a first contract within a 120-day period, the terms of the contract will be set by an arbitration panel designated by the Federal Mediation and Conciliation Service. As with the abandonment of the secret ballot election, I believe this interest arbitration requirement is unwise public policy. The employer is the entity which must run the business, remain competitive, and pay the employees each week. The union has the opportunity to influence the employer’s thinking by engaging in economic warfare. But, the actual agreement is forged in the crucible of what the business can sustain. I firmly believe that our present system has it right for employers and employees covered by the NLRA and that the employer must retain the power to determine whether the terms of the agreement are acceptable to it.
Conclusion
This concludes my prepared oral testimony. I look forward to discussing my comments in more detail during the question and answer period, but before that, I would again like to thank the Subcommittee for inviting me here today, and for its attention to these very important developments regarding labor law in the 21st century.
