Explaining differences in job retention between alien and nonalien workers after an in-company training

Abstract

This study focuses on the job retention of unemployed workers, after they attended a subsidized in company training programme in Flanders. Using a new large scale dataset of the Flemish Labour Exchange, we look for differences in the probability of employment between aliens and nonaliens during the 36 months following their on the job training. We further investigate whether differences persist after controlling for several socio-economic characteristics and labour market related variables. Estimating a modified probit model we find that, on average, being an alien lowers the probability of employment after training by approximately 15%. This effect reduces to 10% when controlling for other variables. The effect of education on a trainee's employment chances differs for aliens and nonaliens. Aliens have a markedly lower return on investment in education. Good language skills and a longer in company training period also increase employment probability, but more so for aliens. Other control variables do not significantly improve the explanatory power of the model. The main conclusion is that even after a tailored on the job training, aliens still lag behind nonaliens in terms of employment success.

Acknowledgements

Financial support by the National Bank of Belgium (grant no. 3067) and by the University of Antwerp (BOF ID 2007 2467) is gratefully acknowledged. We also thank Prof. Joep Konings (KUL) and all seminar participants at the Department of Applied Economics for their helpful comments, and Bart Delbeke for their research assistance.

Notes

¹ An example of this is the partial or full exoneration of employee contributions for 2–3 years that employees can benefit from when hiring someone for the first time, like the trainee after finishing the IVT.

² The language schooling is very intensive, task-based and commercially oriented. The IVT-participant learns exactly what he needs to know to perform well in the respective company.

³ This is without taking into account possible crowding-out effects.

⁴ This is a summary measure of four scores on reading, writing, speaking and comprehension, each with a maximum of five.

⁵ Here, only the EU-17 is considered as European countries in order to also assess the employment success of Eastern European workers.

⁶ Since 1 January 2003, all employers are required by law to declare all employees to the Social Security Service (RSZ) electronically through DIMONA. The latter is thus a data bank that contains accurate and up-to-date information on all wage earners at any given time (Bourdeaud’hui and Vanderhaeghe, 2008).

⁷ Results are available on request.

⁸ A simple Chow test shows that the effects of the independent variables are significantly different for the two subgroups.

⁹ Studies have shown that contact with higher achieving peers has a positive influence on students’ educational achievement (Zimmer and Toma, 2000; Hanushek et al., 2003).