Adaptation or recovery after health shocks? Evidence using subjective and objective health measures.

In this paper, we analyse the effect of an onset of a health shock on subjective survival probability and compare it with objective survival probability and self-reported health measures. In particular, we are interested in whether expectations of people respond to health shocks and whether these follow the evolution of objective life expectations and self-reported health measures over time. Using longitudinal data from the Health and Retirement Study, we estimate fixed effects models of adaptation for the objective and subjective survival probabilities and for some self-reported health measures. The results show that after cancer diagnosis, conditional on surviving, both the objective and subjective longevity and self-reported health measures drift back to the before diagnosis trajectories. For stroke and heart attack, in spite of their persistent negative effect on survival, subjective life expectations and self-reported health measures seem to indicate only a transient effect of the health shock. The differences between the objective and subjective measures are in line with the concept of adaptation. We discuss the policy implications of our results.