We estimate dynamic panel data models of unemployment incidence for men usi
ng the British Household Panel Survey. Econometric issues concerning unobse
rved indvidual heterogeneity, genuine state dependence, and the initial con
ditions problem are addressed in detail. We find strong evidence of state d
ependence consistent with the scarring theory of unemployment-an individual
's previous unemployment experience has implications for his future labour
market experience. This suggests that policies reducing short run unemploym
ent incidence will have longer-run effects by reducing the equilibrium or n
atural rate of unemployment.