A Duration Analysis of the Time Taken to Find the First Job for Newly Arrived Migrants in Australia
Date
2006-07
Authors
Thapa, Prem J
Gørgens, Tue
Journal Title
Journal ISSN
Volume Title
Publisher
Centre for Economic Policy Research (CEPR), Research School of Social Sciences, The Australian National University
Abstract
This paper extends the traditional static focus of research on the labour market assimilation of migrants in Australia by analyzing the dynamics of job search and actual time taken to find the first job after arrival in Australia. The Longitudinal Survey of Immigrants to Australia (LSIA) covers two cohorts of recent migrants to Australia that differ considerably in immigration selection criteria and other policy settings, as well as in the macroeconomic employment conditions at their time of arrival in Australia. This gives rise to very different early labour market outcomes for the migrants in these two cohorts; and this paper looks at one specific aspect of this differing outcome – the time taken to find the first job after arrival in Australia.
¶
We analyze the inter-cohort differences in migrant characteristics and job search behaviour, and explicitly model the duration of the time taken to find the first job in Australia for the sub-sample of migrants who were the Principal Applicants in their visa applications. Using a conventional proportional hazards model framework, estimation results are presented for a Cox model specification and for a parameterized version of the baseline hazard which permits a formal test of equivalent hazards faced by the migrants in the two cohorts of LSIA.
¶
We find that the hazard rates of time to first job are determined by pre-immigrant characteristics, such as education and qualifications, recent work history in source country, the Australian visa categories, English language proficiency, and whether the migrants had previously visited Australia. The parametric model results show that both the underlying baseline hazards and the proportional hazard coefficients on most variables of interest differ significantly between Cohort 1 (arrived in Australia in 1993 through 1995) and Cohort 2 (arrived in Australia in 1999-2000).
¶
We then use the parametric model results to simulate the full distribution of the time taken to find the first job for these two migrant cohorts. We present alternative decompositions of the inter-cohort gap into effects due to differences in the observable characteristics of the migrants, and residual effects which can be attributed to different selection criteria and other policy settings and macro economic conditions that are applicable for the two time periods.
¶
The decomposition results suggest that the more favourable outcomes for Cohort 2 migrants are predominantly due to the setting of the Cohort 2 time period. This effect is most pronounced for the sample of female Principal Applicants. Cohort 1 migrants would also have experienced much quicker exits to a first job had they arrived under the macroeconomic and immigration policy setting of Cohort 2. Although the advantage derived from the Cohort 2 setting is not itself further decomposed, our analyses does indicated an important role for changes in migrating selection criteria to affect initial labour market outcomes of migrants after arrival in Australia.
Description
Keywords
immigration, employment, visa categories, proportional hazards, simulation
Citation
Source
Type
Working/Technical Paper
Book Title
Entity type
Access Statement
Open Access
License Rights
DOI
Restricted until
Downloads
File
Description