2004 EMPLOYEE TRAINING IN AUSTRALIA S53 S53 THE ECONOMIC RECORD, VOL. 80, SPECIAL ISSUE, SEPTEMBER, 2004, S53–S64 © 2004. The Economic Society of Australia. ISSN 0013–0249 tion (1962; 1964) that decisively marks economic thought in this area. Becker emphasises the import- ance of training as a factor in accumulating human capital and how this can be translated into effect- ive growth in the value of the worker’s marginal productivity and, consequently, in their wage. 1 In particular, Becker argues that the impact of training on marginal productivity and wages depends on the nature of the training, that is, whether it is specific or general in nature. More recent papers have returned to Becker and extended his work to consider training outcomes in a range of imperfectly competive enviroments, finding that the theoretical distinction between general and specific training becomes obscured (Polachek & Siebert 1993; Stevens 1994; Leuven 2002). A dicho- tomy between general and specific training is also rarely observed empirically (Bishop 1997) and many authors have sought to explain the apparent sharing of general training costs between firms and employees * We thank the AWIRS95 sponsors and providers; they are not responsible for any of the findings or claims made in the paper. We also thank participants of the ESA 2003 meetings, Sue Richardson, Miles Goodwin and the referees. Almeida-Santos is grateful for funding from the Fundacao para a Ciencia e Tecnologia – Ministerio da Ciencia e Tecnologia (Portugal) and Mumford is grateful for financial support from the Leverhulme Foundation, and for her Visiting Fellowship at the Economics Programme, RSSS, Australian National University. Correspondence: Karen Mumford. Department of Eco- nomics and Related Studies, University of York, York YO10 5 DD, UK. Email: kam9@york.ac.uk Employee Training in Australia: Evidence from AWIRS* FILIPE ALMEIDA-SANTOS 1 Department of Economics and Related Studies, University of York, UK and Instituto Universitário de Desenvolvimento e Promoção Social, Universidade Católica Portuguesa, Viseu, Portugal KAREN A. MUMFORD Department of Economics and Related Studies, University of York, York, UK and National Institute for Labour Studies, Adelaide, Australia We use linked data for 13 991 employees and 1494 workplaces to analyse the incidence of employer-provided training in Australia. We find potential experience, current job tenure, low education levels, skilled vocational training and part-time or fixed-term employment status are all associated with a lower probability of recent training. In contrast to studies for other countries, we find no evidence of discrimination on the basis of demographic characteristics in the provision of this job-related training. Finally, and in support of recent non-competitive training models, higher levels of wage compression are found to be positively related to a greater incidence of employee training. I Introduction In this paper we explore the determinants of workers receiving employer-provided training in Australia. The potential for training to impact on productivity and wages has long been discussed in the economics literature, perhaps beginning in 1776 with Smith’s Wealth of Nations (Smith 1952, p. 49). In the first half of the last century Pigou (1912, p. 12) and Rosenstein-Rodan (1943) developed and expanded many of the issues related to the provision of training, however, it is Gary Becker’s contribu- 1 Alternatively, Spence (1973) argues that high pro- ductivity workers merely use education as a signalling mechanism to employers (see also Autor 2000).