Job Mobility in Australia: Theories, Evidence and Implications

Abstract

Recent theoretical developments postulate that much of the mobility in the labour market arises from imperfect information. As new information is acquired, either about alternative opportunities or about the productivity of the current job, a worker may quit or be laid off. The paper reviews some of these developments and evaluates the predictions they generate against data from the Australian Mobility Survey. It is found that the tenure, age and occupational effects on mobility are broadly consistent with these predictions. The implications of these findings are discussed with reference to the longstanding concern about the flexibility of the Australian labour market.In the traditional theory of competitive labour markets, job mobility is a consequence of changes in technology and demand. Such changes cause wages and marginal products to diverge and start a process of reallocation of labour so as to restore the equality and achieve equilibrium in the labour market.In this process, relative wages are the main signals for the economic agents. From this, the policy prescription that flexible wages are a necessary condition for labour market efficiency is deduced.Whether Australia's centralized wage determination system provides the necessary flexibility has been a long-standing issue, and evidence about the mobility of labour has been important in shaping the various views. Thus Blandy and Richardson (1982) argue that the labour market works as implied by the traditional theory but the adjustments are mainly of 'wage-like* forms. Likewise, Keating (1983) does not reject the traditional model, but concludes that even small changes in relative wages are sufficient to achieve a desired reallocation of labour. The perspective adopted in these studies is, however, incomplete. Much recent work has shown that in the absence of perfect information mobility persists even in equilibrium. Job changes take place as a consequence of workers and firms acquiring more information, either about alternative job opportunities or about the quality of an existing match. Thus many moves are not brought about by relative wage changes, and it is not necessarily the case that higher wages must accompany expanding employment.Indeed, Bell and Freeman (1985) found that wages move in the opposite direction to what would be expected to obtain the desired allocation of labour. This suggests that a supply factor, the productivity of the workforce, may be a more important determinant of wages than demand factors. Nor can

Labour Mobility in the Australian Regions*

“…Male mobility, expressed as the probability of change, varies inversely with the individual's length of job tenure. This result applies in all regions and at both points on the cycle confirming Stromback's (1988) view of tenure as the dominating influence on Australian labour mobility.…”

“…We do not deny the possibility that in estimates of (1) (11, < 0, but it seems more likely that a, 2 0. Stromback (1988Stromback ( , 1994. in aggregate analyses of Australian labour mobility, finds that job tenure has a strong and dominant effect on separations.…”

“…The result is that (rg should be negative in estimates of (1). Stromback (1988) raises the prospect of a convex relationship between mobility and tenure and for this reason we include the squared tenure term (Sf) in (1).…”

Labour Market Flows Data for Australia

“…Data are collected on the number, and The main application of job mobility data has been to test different theories of worker mobility between jobs. Stromback (1988) used published data on job mobility to examine a range of theories of worker mobility; and Kilpatrick ( I 993) has applied unpublished data on job mobility to test the importance of sectoralshift theories of worker mobility and unemployment.…”

Labour Absence and Turnover in Australia

“…Since 1975 the Bureau has regularly surveyed labour mobility which enables separation probabilities to be derived. Using these data, Stromback (1988) calculated that separation probabilities were 23 and 30 per cent for male and female employees respectively in Australia in 1985. Separations, however, include not only employees who quit their jobs, but also employees who were laid-off and people who had completed a fixed term contract.…”