The determinants of voting by the National Labor Relations Board on unfair labor practice cases: 1955–1975

This study has employed logit analysis in an attempt to determine important variables that influence the voting by the NLRB on unfair labor practice cases. Pooled regression results on the odds of voting prolabor were reported for the administrations of Eisenhower, Kennedy-Johnson and Nixon and for the entire period 1955–1975. A number of interesting findings are offered in this empirical investigation of the NLRB. First, it was found that if a Board member is reappointed to the Board, the greater will be the odds that the member will vote prolabor. This verifies the speculation that the Democrats preferred a solidly prolabor voting Board, and that the Republicans desired a strong prolabor voting minority on the Board. Secondly, political variables such as the party of the administration appointing the NLRB member and the member's own political party had an influence on the odds of voting prolabor. If the administration in power was Democratic or if the Board member was a Democrat, the greater were the odds that the member would vote prolabor. Even if a member of the Board professed to be an independent politically, which was the case for two members during the Eisenhower Administration, this variable was found to have a positive effect on prolabor voting for that particular administration and for the entire study period. Also, in the Eisenhower and Nixon administrations, being a former nonmember employee of the NLRB may have had some impact on the voting behavior. Finally, variables for economic conditions in the economy were introduced into the regression equations to test whether or not these variables had any effect on the voting behavior of the NLRB members. In almost all of the logit models reported, the unemployment rate had an influence on voting by Board members. The effect was found to be positive in the Eisenhower period and negative in the Kennedy-Johnson and Nixon periods. The effect for the entire period was positive. It is suggested that whether or not the influence of the unemployment rate on the voting on unfair labor practice cases was positive or negative depended on how the Board perceived the role of organized labor in the tradeoff between inflation and unemployment. If the members believe that organized labor's demands for higher wages caused accelerated inflation when the administration in power was simultaneously trying to reduce unemployment, then the odds of the member voting prolabor were reduced. Besides the unemployment rate, some other economic condition variables that were found to have a possible effect on NLRB voting during the Kennedy-Johnson administration were real GNP, strike days lost, and union membership in the United States. While admittedly this study is only a first attempt at modeling the voting behavior of the NLRB on important unfair labor practice decisions, we believe that we have captured some of the most important political and economic variables that effect the behavior of a quasi-judicial agency — the NLRB. Copyright Martinus Nijhoff Publishers 1981