Labour Market Status and Migration Dynamics
In this empirical paper we assess how labour market transitions and out- and repeat migration of immigrants are interrelated. We estimate a multi-state multiple spell competing risks model with four states: employed, unemployed receiving benefits, out-of-the-labour market (no benefits) and abroad. We discuss one-step ahead transitions from all four states and the transition probability, including all intermediate transitions, from employment. Based on the estimated parameters we simulate the labour-migration dynamics for a synthetic cohort to derive relevant economic indicators, e.g. the probability of experiencing an unemployment spell. For the analysis we use data on recent labour immigrants to The Netherlands, which implies that all migrants are (self)-employed at the time of arrival. We find that many migrants leave the country after a period of no-income. Employment characteristics and the country of origin play an important role in explaining the dynamics. The microsimulations of synthetic cohorts reveal that many migrants experience unemployment spells, but ten years after arrival only a few are unemployed. They also indicate that the Credit Crunch will not only increase the unemployment among migrants but also departure from the country. An increase in the number of migrants from the EU accession countries will lead to more dynamics. We do not expect that the recent simplification of the entry of high income migrants will have a lasting effect, as many of those migrants leave fast.
|Keywords||competing risks, immigrant assimilation, labour market transitions, migration dynamics|
|JEL||Duration Analysis (jel C41), International Migration (jel F22), Geographic Labor Mobility; Immigrant Workers (jel J61)|
|Publisher||Erasmus School of Economics (ESE)|
|Series||Econometric Institute Research Papers|
|Journal||Report / Econometric Institute, Erasmus University Rotterdam|
Bijwaard, G.E. (2009). Labour Market Status and Migration Dynamics (No. EI 2009-25). Report / Econometric Institute, Erasmus University Rotterdam (pp. 1–48). Erasmus School of Economics (ESE). Retrieved from http://hdl.handle.net/1765/17016