Date of Award

1989

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Economics

First Advisor

William J. Moore

Abstract

The theoretical foundation of this model is the economic theory of regulation as developed by Stigler; in addition, modifications of the economic or special interest theory implied by the Public Choice school of thought are employed. A synthesis of these two theories of regulation, then, provides the framework upon which to develop a general model of the causes and consequences of laws regulating the bargaining rights of state employees. Topic One is devoted exclusively to the determinants of state-wide bargaining rights laws. The data are pooled over two years, and state laws are classified into one of three possible categories: bargaining prohibition (or nonexistence of a law), mandatory meet and confer, and mandatory bargaining. The dependent variable, then, is ordinal. Each observation is a discrete realization of the underlying, unobservable variable, sentiment toward public unionism. Estimation proceeds by the technique of McKelvey and Zavonia (1976). The independent variables are divided into two categories: economic/demographic, and political. Inclusion of variables in the former category is motivated by the Chicago School's interest group theory, while the Public Choice School and Economics of Legislatures School suggest variables proxying characteristics of the political process. Topic Two, employing the basic model developed in Topic One, estimates the determinants and effects of bargaining laws simultaneously. The dependent variable in the "effects" equation is union density. Single equation estimation of either process is assumed to suffer from simultaneous equations bias, the consequences of which are biased and inconsistent parameter estimates. An econometric technique developed by Heckman (1978) is employed in order to account for the estimation of an ordinally measured endogenous variable within a simultaneous equations system. Topic Three applies the simultaneously estimated general model developed in Topic Two to three separate employee groups: teachers, firefighters, and police officers. By doing this, comparisons may be made among individual sub-groups, and the applicability of the general model tested.

Pages

252

Share

COinS