July 19, 2011
Testimony of Dean Baker Before the NLRB.
Hearing on "The Risk of Dismissal for Union Organizing and the Need to Modify the Process"
I am here to briefly summarize the findings of two papers by my colleague Dr. John Schmitt, on the risk of dismissal in the course of a unionization drive. Dr. Schmitt, together with Ben Zipperer, used data on reinstatements from the National Labor Relations Board (NLRB) to update prior work on the probability of a worker being dismissed in the course of a unionization election.
The two papers used a methodology that was originally developed by Harvard Law School professor Paul Weiler in a 1983 paper to estimate the probability of a pro-union worker being fired in the course of a unionization drive. Weiler took the number of cases where workers were reinstated by the NLRB for wrongful dismissal. He divided this number by the number of workers who voted for union representation in an NLRB sponsored election. He reviewed this data over the prior three decades finding that there had been a sharp upturn in the probability that a worker would be dismissed in the course of an organizing campaign over this period. By the end of the period, he put the odds that a pro-union worker would be dismissed at close to 1 in 20.
Professor Weiler’s work was criticized in a 1991 paper by economist Robert LaLonde and law professor Bernard Meltzer, both from the University of Chicago.  LaLonde and Meltzer argued that Weiler’s methodology overstated the risk of a pro-union worker being fired in an organizing campaign primarily because it assumed that all the reinstatements were related to dismissal in an organizing drive. LaLonde and Meltzer argued that many dismissals took place for other reasons and that only around half of the dismissals were associated with organizing drives. Re-examining the NLRB data, they concluded that 1 in 63 pro-union workers were dismissed in the course of an organizing drive.
In their analysis, Schmitt and Zipperer followed the more conservative methodology of the LaLonde and Meltzer paper. Schmitt and Zipperer found that in the period from 2000 to 2005 the probability that a pro-union worker would be dismissed in the course of an organizing drive was 1.9 percent. This was up sharply from a 1.2 percent probability in the late 90s, but still below the 2.7 peak rate in the years 1981-1985.
The paper also examines factors that could lead this estimate could be either too high or too low. The most obvious reason that it could be too low is that unions are increasingly turning to majority sign-up (also known as “card check”) as an alternative path toward recognition. This means that many pro-union workers will never actually vote in an NLRB certified election. The implication is that both the Weiler and the Lalonde and Meltzer methodology would understate the number of union supporters since they only count votes for unions in NLRB elections, not the number of people who indicated their support for a union through the majority sign-up procedure.
To adjust for this understatement of the number of workers supporting unions in organizing drives, Schmitt and Zipperer used two independent data sets to get an estimate of the total number of people organized in the years 1998-2003. They then compare this number to the number of people who joined unions through the NLRB election process. This ratio was approximately 1.3 to 1. Schmitt and Zipperer therefore adjusted upward the number of workers who were union supporters in by a factor of 1.3 to determine the likelihood of a union supporter being fired in an organizing drive. With this adjustment, the percentage of pro-union workers fired in the course of an organizing drive falls to 1.0 in the years 1996-2000 and 1.4 in the years 2000-2004.
There are two other factors that could affect these ratios that were not taken into account in the calculations in Schmitt and Zipperer. First, it is possible that the percentage of reinstatements attributable to cases where workers had been wrongly discharged in an organizing drive has changed since LaLonde and Meltzer examined the issue in the late 1980s. Their research led them to assume that half of reinstatements involved dismissals for taking part in an organizing drive. If the causes of the wrongful dismissals overturned by the NLRB have changed since the late 1980s, then this assumption will be misstating the likelihood of being dismissed.
Another factor that can lead this calculation to seriously understate the likelihood of dismissal is the possibility that not all cases of wrongful dismissal may be brought to the NLRB. Furthermore, there is no guarantee that the NLRB rules correctly in every case.
In the first instance, there are undoubtedly many workers who are dismissed in the course of an organizing effort who do not bring their case to the NLRB. The potential gains from winning an NLRB case are relatively limited. If a worker did not place much value on being re-instated in their job, they may just not bother with carrying through an NLRB process. It is also likely that in many cases employers voluntarily offer re-instatement if they think there is a strong likelihood that they will lose in an NLRB hearing.
Both of these situations would not be picked up in the NLRB data used in Schmitt and Zipperer’s analysis. If a case is never brought to the NLRB, then there is no possibility of an NLRB ordered re-instatement. Furthermore, if a company preempts the NLRB by voluntarily offering re-instatement, then there would also be no record in the NLRB data of the firing. For these reasons, it is likely that the Schmitt and Zipperer calculations understate the probability of being fired for organizing activity.
If we accept the study’s range of estimates for the most recent period, 1.4 percent to 1.9 percent, there is still a question of how we should think about such numbers. In both cases, the overwhelming majority of pro-union workers do not get fired, and therefore the risk of dismissal may be seen as a minor factor affecting union organizing drives.
However, dismissals are not likely to be evenly spread among union supporters. Presumably when an employer violates the law to dismiss a worker for supporting a union, he would focus on dismissing those workers who are actively organizing for union representation. If one in ten union supporters is as actively involved in organizing, and this group is generally the target for dismissals, then the probability that a union organizer will get fired in the course of an organizing drive is between 14 and 19 percent. This is likely to act as a substantial deterrent to getting involved in an organizing drive. Many workers who might otherwise opt to organize in support of a union may choose to remain in the background rather than risk being fired.
The second paper looks at the percentage of organizing drives in which at least one worker was fired. The paper finds that pro-union workers were fired in 26 percent of union election campaigns over the period 2001-2007 (most recent available data). The 26 percent rate is up from about 16 percent in the last half of the 1990s. The share of elections in 2001-2007 with an illegal firing was almost as high as the historical peak for such activity --31 percent during the period 1981-1985.
In short, the calculations in the Schmitt and Zipperer papers suggest that there is a high probability that union organizers will get fired in the course of an organizing drive. The fear of an unlawful discharge could be a major obstacle to workers’ freedom to join a union if they choose.
 Schmitt, John and Ben Zipperer, 2009. “Dropping the Ax: Illegal Firings During Union Election Campaigns, 1951-2007. Washington, DC: Center for Economic and Policy Research and Schmitt, John and Ben Zipperer, 2007. “Dropping the Ax: Illegal Firings During Union Election Campaigns.” Washington, DC: Center for Economic and Policy Research.
 Weiler, Paul. 1983. “Promises to Keep: Securing Workers’ Rights to Self-Organization Under the NLRA,” Harvard Law Review, vol. 96., no. 8 (June), pp. 1769-1827.
 LaLonde, Robert J. and Bernard D. Meltzer. 1991. “Hard Times for Unions: Another Look at the Significance of Employer Illegalities,” University of Chicago Law Review, vol. 58, pp. 953-1014.