What Do Australian Unions Do?*

Abstract

WHAT DO AUSTRALIAN UNIONS DO ? Paul Miller and Charles Mulvey University of Western Australia and University of Western Australia and University of California at Berkeley I INTRODUCTION In this paper we survey the literature on trade union activity in Australia. The survey is primarily concerned with empirical research which estimates the impact of unions on economic variables. However, the empirical literature must be grounded in a body of theory and it is to that issue that we first turn our attention. In the century which preceded the 1980s the economic theory of trade unions was developed only sporadically, the most notable advances being due to Marshall [see Marshall (1920)], Dunlop [e.g. Dunlop (1950)] and Cartter (1959). By the end of the 1970s, however, the body of theory was still rudimentary and lacked a general framework at a time when theory in labour economics generally had become sophisticated and coherent. An empirical literature, mainly estimating union relative wage effects, had emerged but made little reference to theory. The 1980s witnessed a rapid development of the theory of trade unions and an associated empirical literature. It would be convenient for us if a single, all encompassing, theory of unionism had emerged so that our survey of the empirical literature could be contained within a single model. However, the theoretical advances of the 1980s addressed a number of issues with two main, complementary, avenues providing the most important advances. The first main theoretical advance concerns the distinction between the so-called 'right-to-manage 1 [RTM] model of trade unions and the 'efficient bargaining' [EB] model [McDonald and Solow (1981), Nickell (1982), Oswald (1985) Farber (1986), and Manning (1992)]. The analysis of the RTM and EB models of unions is mainly concerned with the microeconomics of collective bargaining - whether bargaining between unions and employers is about both wages and employment or only wages, and therefore whether the wage settlement is on or off the demand curve. While the interpretation of the empirical research must take account of the possibility that the wage/employment combination is off the demand curve, the hypotheses investigated in the empirical literature which is the focus of this survey relate primarily to the exit/voice model. Accordingly, we discuss the theoretical basis of only the latter model, and then only very briefly, and concentrate for the rest on the empirical work.