Does your history matter? A survival approach to labour market persistence in the UK labour market
|Day:||Thu 4 Jul|
The issue of labour market dependence has received particular concerns from policy makers partly because it would determine whether short-run government labour market interventions would persist in the long-run. This study tries to revisit the evidence on ‘true’ labour market dependence and ‘scarring’ effects in the UK labour market for those individuals aged 16-65, by utilizing the British Household Panel Survey (BHPS) Waves 1-18. Adopting the discrete-time survival analysis method, we estimate three different forms of labour market dependence, i.e. the duration, occurrence, and lagged-duration dependence. After controlling for the unobserved heterogeneity, we do find that individual’s labour market histories do matter for his or her future labour market status. More specifically, we find that the occurrence dependence variable is not scarring in the case of labour market transitions from unemployment and inactivity status; yet, it is the lagged-duration dependence that is scarring. This finding suggests that having a one-time long spell of previous unemployment (or inactivity) in the past is worse than having multiple short spells of being in-and-out of the unemployment (or inactivity) state. Moreover, we also find evidence of negative duration dependence. Having controlled for the duration dependence, the impacts from occurrence and lagged-duration dependence are relatively small compared to the effects from duration dependence. Thus, this study concludes that as much as the past labour market histories matter, it is the current labour market status that would give more impacts to one’s future labour market status.