Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We propose a novel method to detect and disentangle moderate and severe health shocks in a general population survey based on a data-driven classification of sickness absences and hospitalizations. Both types of shocks are widespread with an annual incidence of about 1.7%, which rises steeply with age. We estimate the effects of both shocks on labor market outcomes and find that severe shocks have more persistent effects on employment (7.5 percentage point reduction), labor income, and household net income. Moderate shocks have transitory effects on employment (2 percentage point reduction), but more long-lasting effects on work hours.