RDP 9404: Wage Dispersion and Labour Market Institutions: A Cross Country Study 1. Introduction

The present trend of labour market reform towards an enterprise focus and away from occupational ‘awards’ is a belated recognition of the disaster that the Australian industrial relations system has inflicted on us (Blandy, 1993, p. 10).

…it should be appreciated that the push for labour flexibility is part of an economically and socially conservative agenda that includes an attack on trade union power and a reduction in the working conditions that are currently enshrined in Australia's industrial award system (Burgess and McDonald, 1989, p. 29).

Australian discussion of wage determination is parochial. There is an almost hypnotic fascination with the arbitral system and its trappings, with the result that appropriate perspective on wage fixing and economic problems is lost… this navel-gazing leads to the perpetuation of too many myths… (Withers, 1986, p. 243).

As the first two quotes above demonstrate, labour market reform – perhaps more than any other issue of Australian economic policy – generates much passionate rhetoric. The current labour market reform agenda in Australia has focused on issues concerning labour market flexibility, typically a poorly defined concept. On the basis that centralised wage setting institutions are too “rigid”, policies have evolved to permit wage determination at the enterprise or firm level. The changes have been based on the belief that centralised systems do not permit sufficient relative wage flexibility. The resulting distorted price signals in the labour market contribute to supply and demand mismatches, impeding productivity growth and reducing the potential for improved economic prosperity in general.

In order to illustrate and analyse these issues, this paper examines data from fourteen OECD countries which encompass a wide spectrum of wage setting institutions. We present quantitative measures of labour market flexibility based on wage dispersion across sectors, and determine whether a high degree of wage dispersion is necessarily linked with decentralised wage setting institutions. Given that a flexible labour market should be characterised by wages which adjust to conditions within the market, we test if such flexible relationships exist. The relationship among the deviation of sectoral wages from the mean wage and similarly, deviations of productivity, prices and growth across countries over a similar time horizon is estimated to determine if the expected positive relationship exists between these variables. The linkage between sectoral deviations of wages and employment growth is also analysed. In addition, we examine the relationships between our measure of flexibility and various indicators of macroeconomic performance.

We do not find a strong systematic relationship between our measure of wage dispersion and the degree of centralisation of wage setting institutions. Low wage dispersion can be associated with centralised wage setting institutions, but is not necessarily so. Insofar as we can compare absolute measures of wage dispersion across countries, we observe that Australia recorded a relatively high degree of dispersion in the 1980s compared to other countries despite having a centralised wage setting system. As expected, we generally find positive relationships among the deviations of wages across sectors and the deviations of sectoral productivity levels and prices. Relative to most other OECD countries examined, deviations of Australian wages are found be strongly related to deviations in productivity and prices across sectors. This suggests that, contrary to much popular perception, Australia's wage setting system has been flexible by international standards, at least according to this method of analysis.

The rest of the paper is organised as follows. In Section 2 we discuss the institutional background to the current policy debate in Australia. The existing literature relating labour market institutions and economic performance is reviewed in Section 3. In Section 4 we present a model which demonstrates the cost of relative wage rigidity, while in Section 5 we derive a structural model relating sectoral wages, productivity and prices. Our measures of wage dispersion are presented in Section 6 along with the results from our tests of labour market flexibility across countries. Section 7 concludes.