Working Paper: CEPR ID: DP3028
Authors: Jrgen Hansen; Magnus Lfström
Abstract: This Paper analyses transitions into and out of three different labour market states: social assistance, unemployment and employment. We estimate a dynamic multinomial logit model, controlling for endogenous initial conditions and unobserved heterogeneity, using a large representative Swedish panel data set, LINDA, for the years 1990-6. The unadjusted data indicates that immigrants are more likely to receive both social assistance and unemployment compensation than natives are. Immigrants are less likely to remain employed in consecutive years than natives are and are more likely to stay on welfare and to receive unemployment insurance in any year, given participation in the previous year. The empirical results suggest that refugee immigrants display a greater degree of ?structural? state dependence than natives. Further, immigrants from non-refugee countries display a similar degree of ?structural? state dependence as natives. The high welfare participation rates among refugee immigrants seem to be due to the existence of a ?welfare trap?, while participation among natives and non-refugee immigrants is largely due to permanent unobserved characteristics. These results suggest that welfare reforms will have differential effects on refugee immigrants and natives
Keywords: immigration; labour market behaviour; state dependence; transition; welfare
JEL Codes: I30; I38; J15; J18; J61
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| previous participation in welfare (I38) | current participation in welfare (I38) |
| refugee immigrants (F22) | welfare trap (I38) |
| natives (J15) | unobserved characteristics influence participation (D13) |
| welfare reforms (I38) | differing impacts on refugee immigrants (F22) |
| previous states (P30) | current probabilities of being in a given state (C11) |