Mental Health and Employment: A Bounding Approach Using Panel Data*

The effect of mental health on employment is a key policy question, but reliable causal estimates are elusive. Exploiting panel data and extending recent techniques using selection observables to provide information along unobservables, we estimate that transitioning into poor leads 1.6% point reduction in the probability employment; approximately 10% raw gap. Selection almost entirely based time-invariant characteristics, rendering fixed effects unbiased this context, meaning researchers no longer have rely narrow local average treatment most health/work IV studies.