Congress permits unions to bargain for workers who do not want such representation, and it compounds this violation of freedom of association by permitting unions to force workers they represent to pay union dues and fees as a condition of continued employment. So-called union security has given rise to a circus of legal disputes which consume human resources that otherwise could be devoted to honest work.
Example: Should unions be permitted to charge workers who are already forced to accept their representation services for organizing expenses involved in the unions’ attempts to capture hitherto union-free workers?
Section 8(a)3 of the National Labor Relations Act (NLRA) gives a union the power to force objecting workers to pay fees to the union as a condition of continued employment. Section 2, Eleventh of the Railway Labor Act (RLA) imposes the same coercion in the railroad and airline industries. In its Ellis decision (1984) the Supreme Court ruled that organized workers under the RLA who object to the payment of union dues may be forced to pay only for collective bargaining, contract administration, and grievance adjustment. In its Beck decision (1988) the Court extended those same protections to objecting workers subject to the NLRA. In Ellis the Court specifically ruled out the expenditures a union incurs when it attempts to organize union-free workers as a permissible charge against already-organized objecting workers. In Beck the Court stated that Section 8(a)3 of the NLRA and Section 2, Eleventh of the RLA are “in all material respects identical,” that they are “statutory equivalents,” and that “Congress intended the same language to have the same meaning in both statutes.”
Thus a reasonable person might conclude that under Section 8(a)3 already-organized objecting workers cannot be charged for a union’s new organizing expenses. Indeed, even William Gould, a former chairman of the National Labor Relations Board (NLRB) who was appointed by President Clinton, expressed that view.
However, the NLRB has defied the Supreme Court.
The NLRB examined the implications of the Beck decision for workers subject to the NLRA in California Saw (1995). It decided that expenses incurred by a union for activities outside an objector’s bargaining unit may be charged 1) if they are “germane to the union’s role in collective bargaining, contract administration and grievance adjustment;” and 2) if, following Lehnert (1991), a Supreme Court public-sector case, the charges “may ultimately inure to the benefit of the members of the local union by virtue of their membership in the parent organization.” The specific expenses at issue in California Saw were litigation costs incurred by a parent union. Organizing costs were not part of the case.
The NLRB then defied the Court’s Ellis ruling on organizing expenses in its Meijer decision (1999). It ruled that such charges, if they are for “organizing within the same competitive market as the bargaining unit employer,” are permissible under the California Saw standard. Such organizing, the Board reasoned, inured to the benefit of the objectors by making collective bargaining on their behalf for higher (nominal) wages more successful. The Board held that the Supreme Court’s ruling in Ellis was irrelevant because the intent of Congress, as stated in Section 1 of the NLRA, was to promote organizing. When Section 2, Eleventh was added to the RLA in 1951, the airline and railroad industries were already fully organized, so promoting organizing was not part of Congress’s intent. This argument is a non sequitur because, notwithstanding Section 1 of the NLRA, the specific intent of Congress with Section 8(a)3, like Section 2, Eleventh, was to capture free riders. Congress never linked union security to organizing. The Supreme Court has not yet taken up this question, but since the Ninth and Fourth Circuit Courts of Appeal have issued different opinions it may. However, in the new political environment it may not.
The NLRB rested its decision in Meijer on empirical evidence presented by economists’ testimony that when unions successfully organize hitherto union-free workers in a specific competitive market, there will be fewer union-free employers against whom unionized employers will have to compete. This increases the ability of those employers to pass union-imposed cost increases forward to consumers, and thus makes the employers less resistant to union demands for higher wages. Thus increased organizing by unions will raise the wages paid to already-organized workers.
The expert witnesses referred to several empirical studies, using 1960s and ’70s data and published in the ’70s and ’80s, which purport to show that a 10 percent increase in the number of workers who are unionized in a specific competitive market will on average increase nominal wages by 2 percent. Similar results were found in a 1992 study of the retail grocery industry commissioned by the United Food and Commercial Workers union, which had organized employees at Meijer, Inc., a retail grocery chain based in Michigan.
In 2004 John DiNardo and David S. Lee (DL) published an important paper that cast doubt on those studies, which were based on old household interviews as reported in the Current Population Survey. The DL study used enterprise-based data from 1984 to 1999, which track specific employers over time, to estimate the effects of new unionization on wages (and other outcomes). Moreover, the earlier studies did not adequately address the “selection bias” problem. It may be that instead of unionization leading to higher wages, unions try to organize profitable firms that are likely to pay higher wages in any case. DL controlled for selection bias by using a statistical technique called “regression-discontinuity.” They found that the effects of new unionization on wages were “centered around zero.”
Apart from statistical issues, there are other problems with Meijer. For example, the expert witnesses relied on a standard neoclassical comparative-static model of the labor market: Increased organizing leads to increased density (the percent of workers who are organized), and this leads to higher wages through increased bargaining power (decreased price elasticity of the demand for labor).
This ignores the competitive market process. Even if it were true that capturing hitherto union-free workers initially raises real wages for already-organized workers, that increase (if it is above the competitive market rate) will not endure. The neoclassical model describes a union reducing the alternatives available to customers and owners of capital. But captured customers and owners of capital are quite entrepreneurial about escaping. The plight of American car makers and car buyers who had been captured by the United Auto Workers (UAW) illustrates the point. American car buyers fled the UAW by buying from union-free producers both here and abroad. American car makers fled the UAW by setting up manufacturing plants in foreign jurisdictions that are less union-friendly. The Big Three and the big UAW have lost significant market share. The UAW has had to accept wage cuts to reduce the loss of its jobs.
The market process message is clear: Too much organizing leads to less job security and lower real wages.