From: Heverling, Angela Sent: Wednesday, January 06, 2010 11:13 AM To: Wagner, Dean Subject: FW: Docket No. C-6964 Representation Election Procedure Dean, I would include this in the substantive comments even though he didn't put it into an attachment. -----Original Message----- From: brodyiir@berkeley.edu [mailto:brodyiir@berkeley.edu] Sent: Monday, December 21, 2009 9:22 PM To: Web OLA Email Subject: Docket No. C-6964 Representation Election Procedure December 21, 2009 To the Members of the National Mediation Board: This letter is in support of the Board’s proposed rule change for determining majorities (Docket No. C-6964). Others will be addressing the legal and policy implications of this shift. Here I would like to examine the matter historically. Specialists in the field are aware that the Railway Labor Act of 1926 made no provision for the determination of bargaining agents. Its main thrust was against the prevalence of company unionism in the industry, which it did by prohibiting “interference, influence, or coercion” by either side in the selection of representatives by the other. Despite its undoubted unlawfulness—definitively so after Texas & New Orleans (1930)—company unionism continued unabated, although generally departing from the usual works-council model and taking the form of “associations” that bore a faint resemblance to real unions. It soon became clear that the law would be ineffective without some mechanism for demonstrating the preference of employees (conjoined with an employer duty to bargain). The idea of majority rule had already been broached by the impotent Railway Labor Board under the Transportation Act 1920 and by railway unions seeking recognition under the RLA, with little effect (it seems that a grand total of seven carriers ever consented to voluntary elections). Efforts to amend the law began in earnest at the advent of the New Deal, coinciding with a national debate, triggered by Section 7(a) of the National Industrial Recovery Act (1933), over the very same question of employee choice of bargaining agents. This debate had not yet ripened when Congress acted on the RLA in mid-1934, with the result that the amended Section 2 was loosely drawn and inordinately susceptible to the influence of one powerfully-placed individual. This was Joseph B. Eastman, the emergency Coordinator of the Railroads, who took the position that the abhorrent practices of company unionism not be countenanced for any labor organizations, hence—over strenuous trade-union objection—the prohibition of compulsory membership and employer deduction of dues, and also, in light of the premise of company unionism that every employee was a participant, that representation by lawful unions be based on a majority of eligible voters. This last was not actually written into the law, but the original NMB appointees, familiar with Mr. Eastman’s views, treated it as legislative intent, and that, in a nutshell, is how current NMB practice came into being. Is there anything in this history that argues for maintaining that policy today? The answer is no. It was a product of its time, a historically contingent event. And so, in its own way, was the NLRB’s contrasting practice. In a 1936 case, it was confronted by a contested election in which one union withdrew and pressured other workers not to vote. Despite the fact that only a minority voted, the NLRB certified the winning union, on the grounds that to do otherwise would be to perpetuate the inter-union strife that had prompted the election in the first place. That set the precedent for basing majorities on votes cast. As a kind of offset, as a guarantee against “unwilling” majorities, the NLRB a year later added a no slot to the ballot, so that, whereas under NMB practice, there were two categories—votes for a representative and votes not cast—under NLRB practice there were three—votes for, votes against, and votes not cast. Since elections were generally contested and participation rates high (88%, the NMB reported, for 1934-1945), this was for many years a distinction without a difference. Times have changed. While union-contested elections have dwindled, elections are still contested, only now by the employer. For the NLRB jurisdiction, this was mainly a product of the Taft-Hartley free-speech provision. Although the NLRB couldn’t have foreseen Taft-Hartley—another of those little tricks of history—the no vote introduced in 1937 corresponds today to a vote for the employer. The RLA of course has no free-speech provision—indeed, it still promises employees “complete independence” in the exercise of self-organization—but differences in language do not translate into differences in employer behavior. What is different is the diverging impact of NLRB and NMB electoral practice: against the headwinds of employer resistance, only railway and airline unions must win an absolute majority in representations elections. Let me draw one final lesson from history. The NMB—at least in the early years I have surveyed—never regarded its eligible-voter rule as sacrosanct. It routinely departed from that rule at the joint request of competing unions or where there was evidence of employer interference. This last was the occasion for the earliest challenge to the amended law, Virginia Railway Co. v. Railway Employees (1937), in which the carrier (who had unlawfully discouraged workers from voting) claimed that Section 2, Fourth, required that certification be based on a majority of eligible voters. In rejecting this claim, the Supreme Court roundly endorsed the Board’s flexibility, which it found not only legally permissible, but in keeping with the country’s democratic practices and necessary to fulfill the purposes of the law. The rule change now under review is in the spirit of Virginia Railway and of the Board’s tradition of flexibility. In the world as it is today, the rule it proposes is the only remedy that fulfills the purposes of the Railway Labor Act. Respectfully submitted, David Brody Professor Emeritus of History, University of California, Davis