RDP 9509: Australian Wage and Price Inflation: 1971–1994 1. Introduction

Australian wage and price inflation has varied widely over the past 25 years displaying a number of distinct inflationary periods. Figure 1 shows that following low inflation in the early 1970s, inflation rose substantially with the first oil price shock (OPEC 1) and successive wage shocks. Inflation rose again in the late 1970s and early 1980s with the minerals wage boom and second oil price shock (OPEC 2) before moderating through the 1980s. During the recession beginning in 1989/90 inflation declined to rates not seen since the beginning of the period.

The evidence of distinctly different inflationary periods is consistent with standard macroeconomic models where economies can experience any value of inflation in the steady state. This implies that in a statistical sense inflation can exhibit changes in its mean between periods. The changing means is the first complication encountered when estimating price and wage equations.

Attempts to estimate price and wage equations for Australia are also complicated by the large and persistent changes in the income shares of firms and labour over the past 25 years. In Figure 2 we see the wage shocks of the early 1970s led to large increases in labour's income share. The increased income share persisted well above its pre-shock levels until the late 1980s. If we assume in the long run that labour and firms receive constant income shares then the large change in income shares during the 1970s and 1980s may be characterised in one of two ways. This period may represent very slow price adjustment in response to wage shocks. Alternatively, the persistently high labour income share may reflect a temporary higher equilibrium. Difficulty arises with both interpretations. The former conflicts with casual observation of the speedy adjustment of firms to cost increases while the latter is difficult to support theoretically.

This paper estimates price and wage equations for Australia. In doing so we highlight a number of the methodological and empirical problems associated with their estimation and we offer solutions to the two problems identified above: that of changing income shares and the changing means of wage and price inflation.

The remainder of this paper is in three sections. The next section sets out an imperfect competition model and uses this model to explain the changed relationship between labour's income share and unemployment over the past 25 years. A simple two-equation imperfect competition model is then solved to determine the form of the long-run price equation. This model is used to highlight the identification problems associated with estimating price and wage equations. Section 3 estimates two price and wage systems using quarterly Australian data for the period March 1971 to September 1994. The first system displays a unique rate of inflation in the steady state and, therefore, is inconsistent with standard macroeconomic models. Applying a restriction to the wage equation, the system is re-estimated in a form which avoids this inconsistency. Section 4 concludes.