Health shocks, disability and work

This paper focuses on the relation between the onset of a disability and employment outcomes. We develop an event-history model that includes accidents as a measure for health shocks and we estimate the model using data from the British National Child Development Study (NCDS). We show that experiencing such a health shock increases the likelihood of an onset of a disability by around 172%. However, accidents are relatively rare events and therefore the larger part of observed disability rates result from gradual deteriorations in health. The absence of a direct effect of accidents on employment outcomes, allows instrumenting the onset of disabilities by the occurrences of accidents. We find that onset of a disability at age 25 causally reduces the employment rate at age 40 with around 14 percentage points. The effect is stronger for males and forlow educated workers. Our results indicate that about two-third of the association between disability and employment can be explained by the causal effect of the onset of a disability on employment. This fraction is higher for men and for lower educated workers.