At the same time as union membership has declined in the U.S. private sector it has actually risen in the public sector (Freeman 1986, 1988). In light of this divergence it is interesting to examine the effect of changing unionization on the growth of wage dispersion within the public and private sectors. An evaluation of the effect of unions on inequality in the public sector is particularly compelling because shifts in public sector unionism have occurred for largely exogenous reasons, associated with changes in legal barriers to unions in the public sector, rather than for potentially endogenous reasons, such as shifts in demand that may have also contributed to rising wage inequality. Moreover, an analysis of union wage effects in the public and private sectors presents an opportunity to ask whether unions act differently in the two sectors — in particular, whether unions exert a greater “flattening” effect across skill groups in a non-competitive versus a competitive environment.


In modelling the effects of unions on wage outcomes it is useful to begin by assuming that workers can be categorized into homogeneous skill categories. Let wni(c) represent the log wage that individual i in skill category с would earn in the nonunion sector, and let w“(c) represent the log wage for the same individual if he or she worked in a unionized job. Assume that
where w”(c) and wu(c) are the mean nonunion and union wages for individuals in skill group c. Homogeneity within skill groups is reflected in the assumption that Е[е”|попишоп] = E[€“|union] = 0. (In other words, all of the differences in pay between union and nonunion workers in a given skill group are attributable to the effect of unions). The union-nonunion wage gap for skill group с will be denoted by
In addition to affecting the mean level of wages, unions may also affect the dispersion of wages within skill category.

Tags: , ,