Time-Varying Equilibrium Rates of Unemployment: An Analysis with Australian Data

Article excerpt

Abstract

We explore a new approach to understanding the evolution of the unemployment rate in Australia. Specifically, we use gross worker flows data to study the consequences of assuming that there is no unique equilibrium rate of unemployment but rather a continuum of stochastic equilibrium rates which reflect the movement of the unemployment entry and exit rates over time. The stochastic equilibrium unemployment rate and the observed unemployment rate are very closely related and we explore the reasons why this is so. We then examine the short-run dynamics of the entry and exit rates (specifically, the impulse response functions) and the impact of shocks to the entry and exit rates on the unemployment rate. We find that shocks to the entry rate have been more important than shocks to the exit rate in bringing about variations in the unemployment rate over our sample period. Finally, we present a new way to disentangle the effects of the business cycle from the effects of structural shifts on the (equilibrium) unemployment rate. It would appear that there was a once and for all downward shift in the equilibrium rate(s) of unemployment in Australia in the early 1990s, which likely reflects the introduction of a more generous system of disability pension benefits.

(ProQuest: ... denotes formulae omitted.)

1. Introduction

In this paper we present a new approach to understanding the evolution of the unemployment rate in Australia. Specifically, we use gross worker flows data for Australia to explore the consequences of assuming that there is no unique equilibrium rate of unemployment, but instead a continuum of stochastic equilibrium rates which reflect the movement of the entry and exit rates over time. In so doing we are following up on ideas to be found in Burgess and Turon (2005) and in Hall (2004 and 2005a) and which are implicit in Beveridge Curve analysis. Besides taking the notion of a 'stochastic equilibrium' seriously, our contribution is to show that it provides a new way to empirically disentangle the effects of the business cycle from the effects of structural shifts on the (equilibrium) unemployment rate. We conjecture (i) that there is a stable trade-off curve relating unemployment entry and exit rates as they vary over the cycle and, (ii) that it is by examining shifts in this curve that we have the best prospect of obtaining a time series for an equilibrium rate of unemployment which is purged of cyclical influences. Using quarterly data for Australia for 1979:3 through to 2007:3, time series techniques are used to explore the properties of the entry and exit rates, and to evaluate both the implied time path of the stochastic equilibrium rate of unemployment and its persistence. We find that the entry rate is highly counter cyclical and that shocks to the entry rate have been more important than shocks to the exit rate in bringing about variations in the unemployment rate over our sample period. We also find evidence of a structural break in the relationship between the entry and exit rates to unemployment in Australia from the early 1990s, which likely reflects a change in social security arrangements. This change implies a significant reduction in the equilibrium unemployment rate and also a more rapid speed of adjustment of the actual rate to the equilibrium rate.

The rest of the paper is organised as follows. In the following section we give a very brief outline of traditional models of the equilibrium unemployment rate and identify our point of departure. Key elements of the data and our model are presented in sections 3 and 4. The model is extended to examine in further detail the relationship between the exit and entry rates to unemployment, and possible shifts in this relationship. In particular, the negative relationship between entry and exit rates over the business cycle (the en-ex schedule) and its implications for the cyclical behaviour of the unemployment rate is highlighted. …