Organized Labor Concedes: Employer Violations Rare in Secret Ballot Elections

Report Jobs and Labor

Organized Labor Concedes: Employer Violations Rare in Secret Ballot Elections

June 15, 2009 17 min read Download Report
James Sherk
James Sherk
Research Fellow, Labor Economics
As research fellow in labor economics at The Heritage Foundation, James Sherk researched ways to promote competition and mobility.

Organized labor argues that Congress should effectively take away workers' right to vote in secret ballot elections because employers allegedly intimidate workers in the run-up to elections by firing and threatening to fire pro-union workers. However, a recently released study commissioned by two union-funded organizations, American Rights at Work and the Economic Policy Institute, shows that employers rarely break the law during organizing campaigns. Research by the study's author--Kate Bronfenbrenner, a former union organizer--also shows that employers and unions agree on contracts within two years in most newly organized workplaces and that ineffective union organizing campaigns--not management opposition--explains the labor movement's failure to organize more workers. The labor movement's own figures rebut its case for replacing secret ballots with publicly signed union cards. Organized labor's figures also show why unions want to employ card check in the first place: Unions win 100 percent of card-check campaigns in the public sector. Taking away workers' rights to a secret ballot guarantees workers will have union representation, whether they want it or not.

Killing the Secret Ballot to Prevent "Intimidation"

Organized labor's highest legislative priority is the misnamed Employee Free Choice Act (EFCA). EFCA would replace traditional secret-ballot elections, where workers decide on joining a union in the privacy of a voting booth, with "card check"--where workers publicly sign union cards, often in full view of union organizers. EFCA also authorizes the government to write and impose the terms of collective-bargaining agreements on workers and businesses at newly organized companies.

Labor unions publicly argue for congressional passage of EFCA by maintaining that the current organizing system is broken. They claim that most American workers want to join a union but employer campaigns of "coercion, intimidation, and retaliation" deter workers from voting for the union representation they desire.[1] Unions allege that during pre-election campaigns employees are routinely "harassed, intimidated, spied on, threatened and fired."[2] They argue that this creates a climate of fear that terrifies workers into voting against the union in the private voting booth. Labor unions contend that Congress should eliminate secret ballot elections in order to protect workers from this alleged intimidation.

This argument makes little sense, of course. Secret ballots specifically protect privacy so that voters can express their views without fear of retaliation. Union membership has fallen not because of widespread intimidation, but because unions do not fit into the modern economy well and most non-union workers simply do not want to organize: A recent Rasmussen poll found that only 9 percent of non-union workers want to join a union.[3]

Misinformation on Employer Misconduct

Several studies analyzing data from the National Labor Relations Board (NLRB) have shown that the union movement heavily inflates charges of employer misconduct. These studies find that employers illegally fire workers in at most 3 to 4 percent of organizing campaigns.[4] These numbers make sense: Employers who want to defeat a union campaign must persuade their workers that they deserve their trust. Threatening workers demonstrates supervisors' untrustworthiness and can make workers more likely to turn to the union for protection. Unions win more elections if employers harass and discipline union activists than when they do not.[5]

Now a new study commissioned by the labor movement confirms the results of these other studies. The labor movement's own figures show that employers rarely threaten to fire workers who want to join unions.

Bronfenbrenner Study

This recent study is "No Holds Barred: The Intensification of Employer Opposition to Organizing," by Kate Bronfenbrenner, a former union organizer and now a professor at Cornell University.[6] The Economic Policy Institute, a union-funded think tank and American Rights at Work, a union-backed organization established for the purpose of advocating the passage of EFCA, jointly released Bronfenbrenner's study. It has the labor movement's full endorsement.

Bronfenbrenner analyzed two data sets to determine how frequently employers break the law during union campaigns. First, Bronfenbrenner re-analyzed data from a previously published survey of lead union organizers in 1,004 organizing campaigns between 1999 and 2003. She asked the lead organizer of each campaign whether employers used various tactics to defeat the organizing drives.[7] Second, Bronfenbrenner requested data under the Freedom of Information Act on any Unfair Labor Practices (ULP) charges filed with the NLRB during those campaigns and their ultimate disposition.[8]

Organizers' Impressions Not Based on Facts

The study relies heavily on the self-reported survey of union organizers--data that suffers from many biases. Most important, union organizers do not directly observe what happens in the workplace. They only hear second- and third-hand reports. They do not know if an employer laid off an employee because he supported the union or because of chronic tardiness. They do not know if an employer "interrogated" employees or had a conversation in which an employee mentioned his views on unionizing. They do not know if employers "threatened" workers or simply explained that in collective bargaining everything is on the table and workers' compensation can go both up and down--a fact that union organizers avoid mentioning, and a statement that is permissible within the boundaries of the National Labor Relations Act.

Additionally, union organizers suffer from their own cognitive biases. Surveying union organizers several years after an election campaign amounts to asking them to remember what made it difficult for them to make a sale. People also have a natural tendency to take credit for success and deny responsibility for failure, and union organizers will want to lay any blame on management, not their own shortcomings. Organizers view management as the opposition and naturally accept most criticisms, baseless or not, of a company's management--just as Republicans and Democrats easily accept criticism of the opposing party's candidate during presidential election campaigns.

Organizers also greatly exaggerate workers' interests in unionizing. While only 9 percent of non-union workers want to join a union, 47 percent of union members believe that most workers want one.[9] Organizers more naturally believe that widespread management opposition explains their failure to organize more workplaces. These biases incline union organizers to believe employer misconduct occurs much more frequently than it does. Union organizers' self-recollections of employer misconduct are highly unreliable.

NLRB Figures Are Reliable

Unions will virtually always file ULP charges in response to illegal employer behavior during a campaign. ULP charges make effective campaign propaganda to demonstrate that workers should not trust their employers.[10] The National Labor Relations Board conducts a thorough and impartial investigation of ULP charges and provides much more reliable data on employer misconduct than surveying union organizers. Unions rely on NLRB data for other figures, such as the number of employer back-pay awards ordered by the NLRB.[11]

Bronfenbrenner's analysis of ULP filings indicates that unions withdraw, or the NLRB dismisses, 50 percent of union allegations of threats, 57 percent of union harassment allegations, and 68 percent of claims of firings for supporting the union.[12] This underscores the unreliability of union organizers' assessments of employer wrongdoing. In many cases where organizers allege illegal behavior, the employer did nothing wrong whatsoever.

Bronfenbrenner claims that unions do not file charges in elections they expect to win because "employers can use the ULP charges to indefinitely delay or block the election."[13] She argues that ULP charges thus understate employer misconduct. This misrepresents the law. Fear of delaying a successful election would discourage unions from filing charges and skew the NLRB data--if the law allowed employers to delay elections. It does not. Unions decide whether or not the election goes forward while the NLRB investigates ULP charges. The union has the option of putting the election on hold--called a "blocking charge"--while the NLRB investigates the allegation. If the union expects to win the election, it submits a request to proceed and the election occurs as scheduled.[14]

Contrary to Bronfenbrenner's claims, the law does not give unions any reason to hesitate to file ULP charges.[15] Her figures also show that unions file ULP charges frequently--in two of five election campaigns.[16] The best and most objective available data for determining the extent of employer adherence to or violation of labor laws comes from the NLRB.

Union Figures Show: Employer Illegalities Rare

The Bronfenbrenner study highlighted the unreliable results of the survey of union organizers. These organizers reported that management frequently violated the law, threatening workers in 69 percent of election campaigns, firing workers in 34 percent of campaigns, and harassing workers in 41 percent of elections.[17] The Economic Policy Institute and American Rights at Work have highlighted these figures as further evidence of widespread management intimidation.

However, the much more reliable data from the National Labor Relations Board rebuts these claims. The NLRB data show that only a small minority of employers ever break the law. Employers fire union supporters in just 6 percent of elections, threaten workers in 7 percent of elections, and harass workers in just 2 percent of elections.[18]

Chart 1 shows the full results of Bronfenbrenner's study from both the survey of union organizers and the NLRB data.[19] Impartial data from NLRB investigations show that most employers obey the law. While bad apples do exist, organized labor's own analysis of NLRB data shows that the overwhelming majority of employers obey the law and respect their employees' rights during union campaigns. Only by citing completely unreliable data can the labor movement contend that widespread employer misconduct occurs.

Persuasion Campaigns

Employer intimidation rarely occurs in organizing campaigns because workers vote in privacy. In order to win an election against union formation, employers must persuade workers that they do not need union representation. Intimidating workers is counterproductive, causing them to turn to the union for protection from hostile managers. This is why most organizing campaigns become contests of persuasion, not intimidation.

Unions start with the advantage. They visit workers in their homes and try to convince them to support the union months before the employer learns of the campaign. In one-on-one and small group meetings, union organizers use sales techniques, such as SPIN selling, to convince workers that they have a problem that creating a union will solve.


Unions train their organizers to move employees through four emotional states: situation, problem, implication, and need payoff (SPIN).[20] Unions identify key issues with which an employee is dissatisfied--such as wages, work schedule, or amount of respect received at work--and attempt to convince workers that a union will solve these problems. Former United Food and Commercial Workers organizing director Rian Wathen explains this process:

They track what stage of those four emotions that employee is at and write down a list of questions that they need to ask them the next time in order to move them down that emotional scale to get them to that point where you need to have them... what you're trying to do is to move people down that emotional scale: ...It's about getting them to understand that signing that card, or voting for the union as it is under the current law, is what will get them what they need to solve whatever problem they have.[21]

Organizers also use workers within the plant to build support for the union. They use member volunteer organizers alongside paid organizers, establish representation committees in companies, and employ group pressure tactics, such as solidarity days, rallies, and marches. These group actions create a bandwagon effect and persuade undecided workers to join their peers in support of the union.

Employers start out at a disadvantage in these campaigns because they usually do not learn about the organizing drive until after the union files for an election. Once this occurs they typically have only one to two months--compared to the unlimited amount of time beforehand that the union has already had to build support--to counter the union campaign and persuade workers that they do not need to organize.

The law prohibits employers from threatening retaliation if employees unionize, interrogating workers about whether they support the union, promising to improve working conditions if workers vote down the union, or conducting surveillance to determine which workers want to unionize. These acts are illegal, and--as Bronfenbrenner's analysis of the NLRB data she collected shows--the overwhelming majority of employers obey the law.

The law does allow employers to express their opinion and explain the facts to the employees. After employers learn of the organizing drive, they hold staff meetings throughout the campaignto make their case. These staff meetings--derided by unions as "captive audience" meetings--are the principal means by which employers communicate their side to employees. These meetings cover different topics and focus on giving workers factual information about the union that the organizers avoided during their sales pitches.[22] As workers learn uncomfortable facts that the organizers failed to mention, their support for organizing drops.

Employers explain that the law does not require them to agree to concessions and that in collective bargaining employee benefits can go up or down. They explain that in order to obtain concessions important to the union, such as automatic dues deductions, unions will agree to accept lower wages or benefits. Employers explain that workers lose both their pay and benefits for the duration of a strike. They show that union organizers portrayed only the potential positive aspects to unionization, and none of the risks for the workers.[23]

Employers also show the workers the downsides of joining the particular union seeking representation. They hand out copies of the union's internal constitutions and by-laws--documents that typically give the union, not union members, all the rights and power. If the union has a track record of corruption or ineffectiveness, employers will bring this to the workers' attention.[24] Employers win secret-ballot elections by educating workers about what unionizing entails, not by firing workers who want to unionize.

Unions Object to Employee Education

Unions object to education campaigns because they are effective. Bronfenbrenner's survey of union organizers did not just cover illegal firings and threats. It also covered widespread methods of educating workers about the risks of joining a union: staff meetings, distributing educational leaflets and letters, and distributing mock pay stubs with deducted union dues.[25] Bronfenbrenner's study treated these educational activities as equivalent to firings and threats to close the plant.

When unions complain about employer resistance they do not just object to employers firing or threatening to fire workers who support a union. They also object to legitimate efforts to provide workers information before they vote on whether or not to unionize. Card-check methods and snap elections are intended to prevent employers from having the opportunity to educate workers before they vote--not to prevent the rare instances of employer misconduct.

Ineffective Organizing Explains Failures. Bronfenbrenner's recent study highlighted the role of employer campaigns in weakening union support, but ignored her own research examining the effectiveness of union election tactics.[26] Her research finds that unions have a variety of organizing tactics at their disposal that are effective in persuading workers to join unions. These tactics include:

  • Staffing the campaign. Using at least one organizer for every 100 workers, and using female and minority organizers in minority-dominated workplaces.
  • Researching the target. Researching the company for vulnerabilities before the start of the campaign.
  • Creating organizing committees. Creating organizing committees with at least 10 percent of the eligible workers on the committee and one-on-one communication between committee members and their peers in the workplace and at private meetings in workers' homes.
  • Using volunteers. Using volunteer union members from other organized units of the same company to speak to workers.
  • Making personal visits. Making union visits with a majority of workers at the firm to a worker's home in order to solicit support.
  • Focusing on issues that resonate. Focusing on issues that resonate in the workplace and community.
  • Using internal pressure tactics. Solidarity days, job actions, rallies, and marches on the boss for recognition.[27]
  • Using external pressure tactics. Developing corporate campaigns that damage the company's reputation, using either paid or free media.

Bronfenbrenner finds that when unions use these tactics they fully offset the effects of employer campaigns:

[A]lthough employer opposition and election environment all have a significant impact on election outcome, the number of comprehensive organizing tactics has as much impact as employer opposition and more impact than election environment.[28]

However, despite the availability of these effective tactics, most union organizers choose not to use them. Adopting these tactics is expensive and requires significant effort on the part of the union organizers. Consequently, they are rarely used and most unions run significantly less effective organizing campaigns than they could. Unions do not need to take away secret ballots to win more elections. They can do it by running intelligent, multi-faceted campaigns with methods they already know how to use. Bronfenbrenner's own research finds that unions are failing to organize workers because they are running ineffective organizing campaigns:

[I]t is too easy to simply blame employer opposition and the organizing environment. American unions themselves must shoulder a good portion of the responsibility for their organizing failures. Although our results demonstrate that even in the most difficult contexts, unions can dramatically increase their organizing success when they run more multifaceted strategic campaigns, the majority of unions organizing today still run weak, ineffectual campaigns that fail to build their strength for the long haul.[29]

The union movement's lead researcher has found that unions could fully offset the effect of employer education campaigns by conducting more intelligent organizing campaigns themselves. That admission undercuts the labor movement's stated rationale for EFCA.

First Contracts Regularly Reached

EFCA has two main provisions: replacing secret ballots with card check and giving government officials the power to write and impose contracts on workers and employers. Card check has attracted the most media attention, but imposed contracts radically depart from seven decades of labor law that states neither employees nor employers may be forced to make a concession. Organized labor contends that imposed contracts are necessary because employers negotiate in bad faith, intending never to agree to a contract but running out the clock and hoping workers will vote to decertify the union.

These concerns lack substance. The law requires employers to negotiate in good faith, but labor law has never expected employers to voluntarily agree to contracts out of sheer generosity of spirit. The law gives workers the right to organize and engage in collective action--such as work slowdowns and strikes--to counteract the employer's bargaining power. If employers engaged in widespread bad-faith bargaining, unions could use their economic clout to force them to make reasonable concessions.

In her recent study Bronfenbrenner argues that bad-faith bargaining occurs frequently, reporting that only 63 percent of newly unionized workers negotiate a first union contract with their employer within two years and only 70 percent negotiate a contract within three years.[30] This appears to provide evidence of widespread misconduct. However, Bronfenbrenner neglected to mention that her own research shows that this figure dramatically understates the true first contract rate. Commenting on a blog post about contract rates, Bronfenbrenner wrote:

[O]ne area that unions have been making improvements in is first contracts. That's because the unions that are successful in organizing are running the kinds of campaigns that would make them more likely to win first contracts, and because we have more non-NLRB campaigns where the first contract rate is much higher, and because there has been an in organizing in the non-profit sector [sic], where first contract rates are also higher. So first contract rates according to my data are consistently averaging 68% for NLRB campaigns, above 90% for non-NLRB campaigns (and have always averaged above 90% in the public sector). That means that...the rate in the private sector not getting contracts within two years of the election is more likely to average closer to 15%.[31]

Organized labor's top researcher, Bronfenbrenner, has found that newly organized workers negotiate first contracts within two years at 85 percent of workplaces. The labor movement's own numbers show that employers do not attempt to delay negotiating contracts. However, in "No Holds Barred," Bronfenbrenner chose to highlight a lower number based on a partial sample of newly organized workplaces. Why would a serious researcher do that? Three years ago Bronfenbrenner explained that:

The problem within the labor movement is that there are some who want the bad news number--that it is getting harder and harder to organize, even when it is based on a flawed source, because they believe it will help in the argument for labor law reform.[32]

Bronfenbrenner has now done exactly what she formerly criticized her colleagues in the labor movement for doing. The union movement knows that employers rarely delay first contract negotiations, but claims otherwise in order to advance the public case for EFCA.

An Offer Workers Cannot Refuse

Organized labor's own analysis shows that employer misconduct rarely occurs. Union analysis of NLRB data shows that only a small minority of employers threaten, harass, or fire pro-union workers. Union figures show that unions negotiate first contracts relatively quickly in most newly organized workplaces. They also show that while employer education campaigns undercut union support, the labor movement could counteract this by using more effective organizing tactics themselves. The union movement knows that its public arguments for eliminating secret ballots and allowing the government to impose first contracts do not stand up to scrutiny. So why does the labor movement want Congress to pass EFCA?

Bronfenbrenner's research reveals the answer. Five states in her sample of public-sector unionizing elections allow card-check organizing for government employees.[33] In those states, unions had a 100 percent win rate in card-check organizing campaigns. Unions organized every workplace they targeted.

Not even the polls commissioned by the AFL-CIO suggest that every non-union worker wants to unionize. But unions organized every government employee in every workplace they targeted with card check. Taking away workers' right to a secret ballot exposes them to peer pressure, harassment, and threats. It allows unions to come back to hold-outs again and again to press them to change their minds, often in the presence of a union organizer, before hearing the other side from their employer. Card check allows unions to organize workers, whether they want union representation or not.


Organized labor contends that most workers prefer to join a union, but that employers intimidate workers into voting against them in secret-ballot elections. Unions propose to solve this problem by effectively eliminating secret-ballot elections and having the government impose contracts on newly organized workers with the Employee Free Choice Act. However the union movement's own data demonstrate most workers do not want to unionize.

Recent analysis of NLRB data conducted by the union movement's top researcher and released by two prominent union-backed organizations shows that employers rarely break the law. This same research has found that unions could counteract employer education campaigns by using more effective organizing tactics and that employers promptly negotiate first contracts at most newly organized firms. The labor movement's own figures rebut its public case for EFCA. This research does reveal why unions actually want card check. In states that allow card check for public-sector workers, unions win 100 percent of card-check campaigns. Card check ensures the membership of millions of new dues-paying union members--whether or not those workers want to unionize.

James Sherk is Bradley Fellow in Labor Policy in the Center for Data Analysis at The Heritage Foundation.

[1]Kate Bronfenbrenner, "No Holds Barred: The Intensification of Employer Opposition to Organizing," Economic Policy Institute Briefing Paper No. 235, May 20, 2009, p. 4, at (June 9, 2009).

[2]Nancy Schiffer, Associate General Counsel, AFL-CIO, testimony before the Education and Labor Committee, U.S. House of Representatives, February 8, 2007, at
(June 9, 2009).

[3]Rasmussen Reports, "Only 9% of Non-Union Workers Want to Join a Union," March 16, 2009. Sample of 1,000 adults conducted March 13-14, 2009, with a margin of error of + or - 3 percent.

[4]J. Justin Wilson, "An Analysis of Current NLRB Data on Unlawful Terminations During Union Organization Campaigns, 2007 to 2008," The Center for Union Facts, February 26, 2009, finds that illegal firings occur in a maximum of 3.75 percent of organizing campaigns. John-Paul Ferguson, "The Eyes of the Needles: A Sequential Model of Union Organizing Drives, 1999-2004," Industrial and Labor Relations Review, Vol. 62, No. 1 (October 2008), Tables 1 and 2. Employer violations of sections 8(a)(1) and 8(a)(3) of the National Labor Relations Act occurred 914 times during 22,382 election campaigns between 1999 and 2004.

[5]Bronfenbrenner, "No Holds Barred," Table 3. Unions win 55 percent of elections when employers harass and discipline union activists and only 49 percent of the elections where they do not. This difference is not statistically significant, however.


[7]562 organizers responded to the survey--a response rate of 56 percent.

[8]Bronfenbrenner obtained NLRB data on 98 percent of the campaigns selected.

[9]Rasmussen Reports, "Only 9% of Non-Union Workers Want to Join a Union."

[10]Phone interview conducted May 26, 2009, with Rian Wathen, former organizing director, United Food and Commercial Workers Local 700 in Indianapolis. Wathen is now a regional president of Labor Relations Services, Inc.

[11]Testimony of Nancy Schiffer. Schiffer misrepresents the reason for back-pay awards: Most of the time, back pay is awarded when the NLRB requires employers to provide restitution after they unilaterally change working conditions, not because of discrimination against union employees.

[12]Bronfenbrenner, "No Holds Barred," Table 7.

[13]Ibid., p. 7.

[14]Patrick Hardin and John Higgens, Jr., eds., The Developing Labor Law, 4th ed., Vol. 1 (Arlington, Va.: BNA Books, 2001), Chap. 10.II.B, p. 519.

[15]It is unclear why Bronfenbrenner, a former union organizer and veteran of the Labor movement, was unaware that unions can file a motion to proceed with the election when ULP charges are pending.

[16]Bronfenbrenner, "No Holds Barred," p. 15. This is roughly twice the rate of ULP filings found by John-Paul Ferguson in his analysis of NLRB data. Ferguson, "The Eyes of the Needles."

[17]Ibid., Figure A.


[19]An employer violation is a ULP charge in which the NLRB either upheld the allegation or the employer settled it.

[20]Homer L. Deakins, Jr. and Rian Wathen, "Consequences of the Employee Free Choice Act: Union and Management Perspectives," Heritage Foundation Lecture No. 1119, April 27, 2009, at http://www.heritage


[22]Phone interview conducted May 26, 2009, with Rian Wathen.



[25]Bronfenbrenner, "No Holds Barred," Table 3. Note that data on these activities are not available from the NLRB because they are legal, protected by employers' First Amendment right to free speech.

[26]Kate Bronfenbrenner and Rob Hickey, "Changing to Organize: A National Assessment of Union Organizing Strategies," in Rebuilding Labor: Organizing and Organizers in the New Union Movement, Ruth Milkman and Kim Voss, eds. (Ithaca, N.Y.: ILR Press, 2004), at http://digitalcommons
(June 10, 2009).

[27]Ibid., Appendix 1.1.

[28]Ibid., Chapter 1, p. 53.

[29]Ibid, p. 54.

[30]Bronfenbrenner, "No Holds Barred," Figure B.

[31]Kate Bronfenbrenner, comment on "Stop the Presses: Employer Resistance Rises," Working Life, July 9, 2006, at http://www.
(June 10, 2009).


[33]Those states in her sample are New York, California, New Jersey, Illinois, and Washington. Other states permit card-check organizing for government employees, but were not included in her sample of election campaigns.


James Sherk
James Sherk

Research Fellow, Labor Economics