The assimilation of Australian immigrants: does occupation matter?

ABSTRACT

This paper utilises the occupational attainment approach to investigate immigrant labour market assimilation, complementing other assimilation approaches such as employability, earnings, skills-match and job satisfaction. Our results show that all immigrant groups suffer from initial occupational attainment disadvantage. Worryingly, no ‘catch-up’ over time is evident – even when disaggregated to reflect different cultures and backgrounds. Nor is there much evidence that the occupational status of younger arrivals matches those of Australian born residents, despite being immersed in local mores and institutions while undertaking schooling in Australia. Newer cohorts of immigrants (those who arrived between 2000 and 2014) are also more prone to suffering an occupational penalty. We recommend policymakers subsidise bridging courses to aid recognition of overseas-obtained qualifications and encourage immigrants to obtain local qualifications that can complement their overseas-obtained work experience. This will increase their ‘Australian-ready’ skill-set and occupational attainment in their new host country.

KEYWORDS:

• Occupational attainment
• immigration
• panel data
• Household Income and Labour Dynamics in Australia (HILDA) survey

JEL CLASSIFICATION:

• C23
• F22
• J15
• J24
• J62

Acknowledgments

This paper uses unit record data from the Household, Income and Labour Dynamics in Australia (HILDA) Survey. The HILDA Project was initiated and is funded by the Australian Government Department of Social Services (DSS) and is managed by the Melbourne Institute of Applied Economic and Social Research (Melbourne Institute). The findings and views reported in this paper, however, are those of the authors and should not be attributed to either DSS or the Melbourne Institute. We thank Nicholas Rohde for his comments on the econometric methodology of the revised version of this study and an anonymous referee for pointing out critical issues to address.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ While it would have been preferable to also investigate the immigrants’ last occupation in their home country, there is no currently available Australian dataset that provide this information.

² This is done in this paper and many previous studies by looking at years since arrival (YSA). It is a well-established fact that longer YSA is almost always associated with greater labour market assimilation.

³ A common problem with panel data is attrition. Panel attrition may bias the estimation results if the probability of leaving the sample is systematically related to labour market outcomes (Fertig and Schurer 2007). For the data employed in this study, Breunig, Hasan, and Salehin (2013) compare education levels between those who stay in the survey and those who drop out of the survey, and find that the differences are fairly small with likely minor implications. Further, there are claims that in a regression attrition is likely to affect intercept terms, but has relatively little impact on the slopes of key coefficients (Fitzgerald, Gottshalk, and Moffitt 1998).

⁴ Nevertheless these studies are not directly comparable as these authors employ an older coding system that has since been superseded.

⁵ These occupational groups are distinguished from each other on the basis of skill level and a broader application of skill specialisation. For more information on the concept of skill level and skill specialisation refer to the Australian Bureau of Statistics (2009).

⁶ Note that we are, however, not replicating the work of Chiswick and Miller (2008). They use occupational rankings to explain earnings and the returns to education. This study instead employs a plethora of personal and labour force characteristics (see Table 1 and Appendix Table A1 and Table A2) to explain immigrant occupational rank on arrival relative to the native-born.

⁷ Results of the aggregated immigrant sample are available on request.

⁸ A fixed effects approach is considered a more convincing estimation tool because it allows arbitrary correlation between the unobserved individual-specific effect and the explanatory variables. However, in this paper, some of our key explanatory variables are constant over time and thus we cannot use fixed effects to estimate their effect on the dependent variable. With Mundlak correction, a random effects model produces almost identical estimates to the fixed effects model (see Mavromaras et al. 2013). In our model, the standard errors are clustered at the level of the panel variable (i.e. employees) hence are robust to any arbitrary intragroup correlation.

⁹ Table 3 excludes the ESB results for parsimonious purposes. Except for marginal changes in magnitude, there is no tangible difference between the ESB results presented in Table 2 and the ESB results not presented in Table 3 but included in our regression. Full results are available on request.