In this sample of persons with low educational level, the difference between those with and those
without professional certificates remains unclear and even not in the usual way. There is, however, a
weak effect suggesting that those with certificates are less likely to migrate than others. We may
hypothesise that, while all seek jobs in the same type of labour market, it is easier for the former to find
employment.
Marital status does not have a significant effect. The population is certainly too young for the
emergence of such a phenomenon. On the other hand, the parents’ job situation seems to have a great
impact on the decision to migrate, particularly with respect to two situations concerning the father. An
individual whose father is unemployed is more likely to migrate (reference: father employed) while an
individual whose father is retired is more likely to be immobile. Income may provide an explanation for
this phenomenon. When the father is unemployed, the son/daughter will logically wish to find a job and
leave the family as soon as possible. On the other hand, retired persons in France enjoy a fixed income
and are in a position to support their children.
Wage equation: general comments
The comparison of the factors influencing the wage of migrants and non-migrants reveals many
similarities but also some interesting differences. Globally, there is no major opposite effect, but it
seems that the wage of non-migrants is sensitive to more variables than that of migrants. As usual, men
earn more than women irrespective of whether they migrate or not; the oldest non-migrants earn more
than the youngest but the average is not at the same age: for the migrants, it is the oldest that are
distinguishable with higher wage while for the non-migrants, it is the youngest that are with lower
wage.
Wage and human capital: the specific effect of migration
According to the hypothesis of the independence of migration and wage, migration has no significant
effect on wage. The estimation of the wage equation on the entire sample, introducing migration and λ
(to measure the endogenous selection) shows that the situation is more complex. The direct impact of
migration is significantly negative while the endogenous selection effect is positive. It seems that, at
least at the outset, migration is not a profitable investment. Nevertheless, it should be remembered that
these persons migrated at the most 4 years ago, and it is therefore impossible, on the strength of these
results, to draw conclusions as to the migration effect over time. This effect is compensated by an
endogenous selection between migrants and non-migrants.