The Elasticity of Demand for Labour in Australia PHILIP E. T. LEWIS Centre for Labour Market Research and School of Economics and Marketing, University of Canberra GARRY MAC ACDONALD School of Economics and Finance, Curtin University of Technology The elasticity of demand for labour at the aggregate level is an important parameter for macroeconomic analysis. In particular, policy issues concerning the impact of wage falls on employment and unemployment hinge on the size of this parameter. It is argued in the present paper that previous work on the elasticity of demand for labour in Australia has been unsatisfactory in a number of ways. A new set of estimates is provided that are derived using a better methodology than before. I Introduction Theaggregatedemandforlabourisoneofthe mostimportantrelationshipsinmacroeconomics. In particular, the impact of labour costs on demand for labour, employment and unemploy- ment are of crucial importance. In Australia the response of labour demand to wage changes has received particular attention given the long history of centralised wage ®xing and, more recently, the Prices and Incomes Accord. In this contextcontrolorin¯uenceofwagescanbeseen as a possible policy instrument aecting employ- ment.Anunderstandingofthenatureofdemand for labour is necessary to examine the eects of policies such as minimum wages, employment subsidies and incomes policy. It will be argued that the aggregate demand functions and the derived elasticities of demand are little understood and have led to the wrong interpretation in much of the Australian litera- ture. The aims of the present paper are, ®rst, to restate the theory of the demand for labour in order to clear up misunderstandings that per- meate the Australian literature. The second aim is to critically assess the Australian literature on the demand for labour to date. The third aim is to provide new estimates of the relevant elastic- ities of demand for labour using the latest econometric methods. The paper is structured as follows. The next section reviews the evidence on the aggregate demand for labour. This is followed by a statement of theory of the demand for labour. Newestimatesoftherelevantelasticitiesarethen provided. Finally, the last section provides a conclusion and discussion of the results. II Background Most studies implicitly adopt a neoclassical framework for the formulation of the demand for labour schedule, and hence the aggregate demand for labour is assumed to be negatively related to real wages and positively related to output. Assuming a constant elasticity of substi- tution production function, the `demand for labour' has typically been expressed as a linear inlogarithmsfunction.Thedependentvariableis usually total employment or hours worked. The independent variables are either real wages, or some other measure such as real unit labour costs, and real gross domestic product. A time trend is also usually included as a proxy for Hicks-neutral technical progress. In practice, most studies take employment to be equal to 18 Ó 2002. The Economic Society of Australia. ISSN 0013±0249 THE ECONOMIC RECORD, VOL. 78, NO. 1, MARCH, 2002, 18±30