First-generation immigrant transfers and mobility intentions: longitudinal evidence from France

ABSTRACT

Migrants who intend to stay permanently in the host country are less likely to be attached to their origin country, meaning that a temporary migration should increase the likelihood of sending financial transfers to the origin country. We test the relevance of this hypothesis using a unique dataset on the integration of first-generation migrants living in France and who were interviewed three times from 2010 to 2013. Family transfers sent to the origin country are frequent, with an incidence of donors ranging between 40% and 50%, and the probability of remittances is higher among temporary migrants. The return–transfer relationship is not affected by the intention of migrants to bring other family members with them to France and sending money abroad is more frequently observed when migrants express their intention to consolidate their families in France. These findings suggest that remittances help immigrants prepare for an expected return while maintaining relationships with family members remaining in their origin country.

KEYWORDS:

• Remittances
• return plans
• immigrants
• France

Acknowledgements

I would like to thank the guest three editors of this special issue and two anonymous reviewers for their very helpful remarks and suggestions, as well as participants at the International Workshop on ‘Intergenerational Transfers and Immigrant Population’ held at the University of Bologna, Italy.

Databases are available for use in research from the French Data Archives for Social Sciences (Réseau Quetelet).

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. Its design was similar to other international surveys of this type, such as the Longitudinal Survey of Immigrants to Australia, the Longitudinal Survey of Immigrants to Canada or the New Immigrant Survey in the United States (migrants are tracked over time), but the French sample is not restricted to migrants who have just arrived in the host country.

2. Migrants had to come from third countries, i.e. outside the European Economic Area and Switzerland (see Régnard and Domergue 2011).

3. There is no information on the household in the origin country. These time-varying omitted factors are likely to influence both remittance decisions and return projects.

4. We construct an unknown category for household income, as around one migrant out of five refused to give an answer to this question. This incidence varies little by region of origin.

5. Using data on migrants aged 45–60 in France, de Coulon and Wolff (2010) study the determinants of the going back and forth intention at the time of retirement. In France, Wolff (2015) shows that the permanent return and shared residence intentions have a similar positive influence on the decision of remitting money to the origin country.

6. Another empirical strategy would be to estimate a nested model with a first equation explaining factors that make migrants remit or not and a second model explaining what distinguishes a regular remitter from an occasional one once the migrant has decided to remit.

7. Time-invariant observable characteristics (like gender) are dropped from the fixed effect regression since their values are always zero after demeaning.

8. For a variable $x_{it}$ observed during T periods, the within transformation is such that $x_{it}^W=x_{it}-x_{i\bullet }$ with $x_{i\bullet }=\left({\sum }_{r=1}^T{x}_{it}\right)/T$. The Helmert transformation is $x_{it}^H=\sqrt{(T-t)/(T-t+1)}\times \left(x_{it}-\left({\sum }_{n=r+1}^T{x}_{in}\right)/(T-t)\right)$.

9. In such a setting, transfers could be sent particularly to cover the expected migration costs of future entries.