In this paper we investigate whether unemployment traps exist and are significant in Belgium. In order to assess them we use panel data sample selecion models. Specifically, we estimate a parametric random effects models composed by a wage equation and a selection equation by maximum likelihood techniques. Replacement rates for the all the individuals in the sample are computed in different ways exploiting the predicted wages resulting from the estimation of the wage equation corrected for sample selectivity. In doing this we use the data extracted from the waves 19931997 of the Panel Study of Belgium household and has been led separately on (unbalanced) samples of men and women. The results highlight the importance of unemployment traps for women moving into employment.