s provocative paper argues that agency shop provisions in state public sector collective bargaining legislation, provisions that have recently been found to be constitutional by the Supreme Court in the Abood case (Abood v. Detroit Board of Education, 97 S.Ct. 1782 (1977), may increase the costs of public education. Their underlying analytical framework is that of a state which confers monopoly power on a public sector union and a union whose sole concern is to win monopoly wage gains for its members. West and Staaf claim that the costs of these wage gains may far exceed the benefits of the legislation, the resulting reduction in negotiation costs, that are perceived by legislatures and the judiciary. In sum, they conclude that agency shop provisions in public education are undesirable because the provisions aid and encourage unions and the latter have primarily negative effects on the educational process.I must confess to being unsympathetic to their conclusions and less than totally satisfied with their paper, for it presents little empirical support for the propositions they set forth. Allow me to offer four examples here.First, one might get the impression from their paper that agency shop provisions in the public sector are fairly prevalent in the U.S. and are uniform across states. In fact, nothing could be further from the truth. As of January 1978, only 16 states explicitly provided for any form of union security clause, other than voluntary dues check-off, in their public sector collective bargaining legislation and not all of these contained agency shop provisions. Moreover, 31 states explicitly prohibited union security clauses for public school teachers (see Hanslowe et al., 1978).The variation in legislative provisions across states provides one with the means to test one of their fundamental propositions, namely that union power is abetted by union security clauses. Presumably this could be done by modifying the models used in previous studies of public sector wage determination to allow the effect of unions to vary with the form of union security clause present in the state and then estimating the modified models (for example, see Ehrenberg and Goldstein, 1975). Of course in doing so, an Public Choice 36: 641-645 (1981) 0048-5829/81/0363-0641 $00.75.