Who can predict their own Demise? : accuracy of longevity expectations by education and cognition

Biased longevity expectations will lead to suboptimal decisions regarding saving, retirement, annuitization and health, with consequences for wellbeing in old age. Systematic differences in the accuracy of longevity expectations may partly explain heterogeneity in economic behaviour by education and cognitive functioning. Analysis of eight waves of the US Health and Retirement Study reveals that individuals with lower levels of education and cognitive functioning report survival probabilities that are less accurate in predicting their in-sample mortality. There is little evidence that the gradients in the veracity of expectations are due to the less educated and cognitively able responding less to changes in objective mortality risks. However, high school dropouts and the least cognitively able report survival probabilities that are less stable and display greater un explained variability. These disadvantaged groups appear to be less confident in their longevity beliefs, which is justified given that their expectations are less accurate.