DT N° 010-2025: On the Dynamics of Mental Health

This paper studies the dynamics of mental health over the life cycle and introduces a parsimonious statistical model suitable for structural economic applications. Using data from the Panel Study of Income Dynamics (PSID), we document new facts on mental health dynamics: mental health generally improves with age, although it has worsened in more recent cohorts. Recovery rates are high and increase with age, and individuals are likely to remain in good mental health, with transitions depending on duration in the current state. Mental health is strongly correlated with fixed labor productivity and with the presence of depression early in life, suggesting that ex-ante conditions play a key role in shaping its evolution. Inequality in mental health remains stable across age. We estimate the model using the Simulated Method of Moments and show it replicates key empirical patterns. We then incorporate the statistical model into a life-cycle framework with endogenous labor supply decisions, calibrated to match observed differences in labor supply -both at the extensive and intensive margins- by mental health status. We find large monetary and welfare losses from depression symptoms, with significant heterogeneity by ex-ante conditions.