|
|
|
Job creation and destruction and labour market segmentation
Job creation and destruction and labour market segmentation in Australia
Labour
market segmentation and job creation and destruction in Europe and Australia
Job creation and destruction and labour market segmentation in Australia
This research is part of the large ARC Discovery Project DP0449654 "The dynamics of job creation and job destruction in Australia" obtained by Bill Mitchell. That project aims to conduct the most thorough Australian analysis to date of the processes of job creation and job destruction (JC&D) with a particular focus on the processes that drive full-time and part-time employment by gender, age and industry sector. The work that Bill Mitchell and Joan Muysken have been doing on cyclicality, non-linearity and regime shifts has helped establish the framework for the work covered by DP0449654. In part, our earlier work was the proof of concept for the successful DP0449654 proposal.
We aim to develop new time series measures of JC&D using, in part, gross flows data, which will then be used to test various hypotheses including the role of small firms in the job reallocation process and the impact of real exchange rate changes. Further, one of the major aims of DP0449654 is to develop new economic theory linking asymmetric investment behaviour to JC&D processes as a means of achieving a better understanding of the dynamics of unemployment. In the DP0449654 application, it is explicitly stated that Joan Muysken will assist in the analysis of gross labour market flows and the formalisation of the asymmetric investment model.
Specifically, we will aim to:
(a) Extend our previous work
on Australian Bureau of Statistics (ABS) Gross Flows data to develop new time
series of JC&D. Mitchell and Muysken (2003a) developed a method to reconcile
sample with population and flows with stocks on a monthly and quarterly basis to
overcome known problems with this data. These techniques will be applied to
unpublished data for age-gender and part-time/full-time employment breakdowns.
The resulting time series will be used to explore several key unresolved
hypotheses in the literature concerning the cyclical behaviour of JC&D
processes with an emphasis on how JC and JD evolve during recession episodes.
The analysis will provide a detailed account of the way in which age-gender
groups are affected by the business cycle and cyclical behaviour of full-time
and part-time employment.
(b) Extend economic theory by developing a new
theoretical work following the sketch in Mitchell and Muysken (2003b). The aim
is to enhance our understanding of the JC&D processes in an economy subject
to demand asymmetries. Mitchell and Muysken (2003b) proposed a simple model
where demand asymmetries drive employment and unemployment asymmetries.
Investment asymmetry is based on irreversibilities driven by endemic uncertainty.
Firms' investment behaviour varies according the state of the business cycle.
The labour market is characterised by high and low skill segments with differing
degrees of flexibility in hours and persons adjustment.
In developing and
testing new theories about the links between product markets and labour markets,
the outcomes will extend our understanding of asymmetries and causal
interactions in the Australian economy. The linking of a complex dynamic model
of asymmetric investment behaviour to asymmetries in the labour market flows is
highly innovative, original work, and provide the basis for more coherent and
empirically sustainable economic policy.
While the above is at the center of the
Project, the application also mentions somewhat more peripherically several
important issues which are neglected in the existing empirical literature on JC&D.
Inadequately addressed areas include: (a) the dynamics of full-time and
part-time employment and related gender issues; (b) the rapid expansion of
service sector employment over the last twenty years (almost all studies
concentrate on manufacturing); (c) the breakdown between private and public
sector employment dynamics; (d) the interaction of JC&D processes and
unemployment.
Three data sets - Industry employment data from the ABS Survey of
Employment and Earnings (SEE), matched records Household employment data from
the ABS Labour Force Survey (LFS) which provide estimates of gross labour market
flows, and Establishment level data from the ABS Business Longitudinal Survey (BLS)
- will be creatively employed to address all five of these unexplored yet
significant issues. First, the research will determine the extent to which the
JC&D are substitution processes (full-time jobs for part-time jobs) rather
than processes that change total employment levels - Mitchell and Muysken
(2003a) indicate the plausibility of the substitution hypothesis. This is
particularly important for an understanding the employment outcomes for
different demographic groups (for example, older males, teenagers, married women)
where the incidence of part-time employment is high. Second, the Project will
broaden the analysis of JC&D to include the service sector. A focus on
manufacturing alone may overstate the degree of flux in the labour market. An
interesting observation in this respect is that the incidence of male part-time
employment has roughly doubled in most sectors of the economy (less in
construction) in the period 1984 - 2003, whereas the incidence of female
part-time employment increased by about 20 percent ponts. Thirdly, the research
will determine the extent to which the JC&D processes are influenced by
employment choices made by the public sector relative to market driven dynamics
- this is also relevant because traditonally non-profit services now become
increasingly privatised over the last decade. Fourth, comparing the dynamics of
JC&D with those in unemployment is highly innovative, in particular when the
implications for long-term unemployment are elaborated, and the movements in and
out of the labour force.
There has also been a recent debate concerning the
quality of jobs. The research will present findings on the sorts of jobs that
are created which will inform this debate. More generally, the findings will
provide new insights into JC&D processes at levels of disaggregation not
previously studied (for example, full-time versus part-time jobs; demographic
outcomes; public/private variations; service sector variations). The relevance
of this lies in the importance of these labour market dynamics to the economic
and social well-being and security of the Australian workforce and their
families. Policies designed to enhance the fluidity of the labour market and
minimise the costs of adjustments will be informed by the research findings.
Finally, unemployment remains the most compelling unsolved labour market problem
in Australia. The research findings will establish links between job turnover
and unemployment and trace it to demand and supply dynamics in the broader
economy. This intelligence will aid policy developments to build on our efforts
to date in combating unemployment.
Labour market segmentation and job creation and destruction in Europe and Australia
The recently funded ARC Discovery Project DP0449654 will
conduct the most thorough Australian analysis to date of the processes of job
creation and job destruction (JC&D) with a particular focus on the processes
that drive full-time and part-time employment by gender, age and industry
sector. The method employed in that study is macroeconomic in nature, focussing
on highly disaggregated industry sectors using time series data.
The
collaborative research project proposed here will complement and extend that
research by using survey based data. It will focus on labour market segmentation
and the quality of jobs, in order to analyse the prevalence of dual labour
markets. The assumption of dual labour markets underpins research in this area (see
for example Mitchell and Muysken, 2003a). The project will also consider
developments in two European labour markets, thus providing an international
comparative context for the Australian analysis.
Research into JC&D has
tended to focus on the macroeconomic level. The proposed research is innovative
since it will analyse these processes at the level of the individual. The
research findings will thus provide new insights by relating JC&D to the
characteristics of the persons and jobs involved. The relevance of this lies in
the importance of the labour market dynamics to the economic and social
well-being and security of the workforce and their families. Policies designed
to enhance the fluidity of the labour market and minimise the costs of
adjustments will also be investigated, with a comparison of Australian and
European experiences.
The proposed project concentrates on household survey data
for Australia and two European countries. The Australian source is the Household,
Income and Labour Dynamics in Australia (HILDA) survey, which will cover three
years of data, from 2001 to 2003. Comparable surveys exist from the mid-1980s
for Germany (Sozio-oekonomische Panel, GSOEP) and for the Netherlands
(Organisatie voor Strategisch Arbeidsmarktonderzoek, Vraagpanel, OSA and
Sociaal-economisch Panel, SEP). These surveys provide a valuable data source for
examining both labour market flows and jobs quality, and these variables can be
connected with many potential explanatory factors available in the same data
sets.
The proposed project consists of three components, which build on the
experience of the participants:
(1) The development of a stock-flow model of the
labour market, which fits into the framework on the asymmetric behaviour of
unemployment over the business cycle developed in Mitchell and Muysken (2003b).
This will provide a model for analysing both the dynamics of full-time and
part-time employment and related gender issues, and the interaction of JC&D
processes with unemployment and movements into and out of the labour force.
Moreover, it will provide a theoretical background in terms of flow behaviour
for the influential Lego-approach developed in Gregory (1991) for the Australian
labour market, in which changes in stocks are analysed, distinguishing between
full-time and part-time and gender.
(2) Panel data analysis using the
HILDA-survey data set, to analyse the incidence of labour market segmentation.
This will enable the investigation of the correspondence between part-time
employment and casual employment, and an analysis of the secondary labour market
characteristics of part-time employment. Other characteristics associated with
secondary labour market groups will also be explored (Borland, Gregory and
Sheehan, 2001; Cowling, Mitchell and Watts, 2003). The results will be
reproduced for the GSOEP and OSA/SEP survey data for Germany and the
Netherlands, respectively, which are available from the mid-1980s.
(3) A
comparative analysis of the policy issues focussing on labour market
segmentation and the incidence of casual labour, using experiences from
Australia, Germany and the Netherlands. ….
Back to
