We test the signalling hypothesis using detailed flow-based employer–employee data from Denmark. Our main findings are: the signalling value is proportional to the proportion of employees dismissed from the workplace and the signal is uniform across occupation groups.

We investigate whether agency employment is a bridge into regular employment for immigrants using the timing-of-events approach. We provide evidence of large positive in-treatment effects for all immigrants. Post-treatment effects are fairly high for male non-western and Eastern European immigrants.