Using Italian data, we estimate an option value model to quantify the effectof financial incentives on retirement choices. As far as we know, this isthe first empirical study to estimate the conditional multiple-years modelput forward by Stock and Wise (1990). This implies that we account fordynamic self-selection bias. We also present an extended version of thismodel in which the marginal value of leisure is random.The models yield plausible estimates of the preference parameters. Dynamicself-selection results in a considerable downward bias in the estimate of themarginal utility of leisure. We perform a simulation study to gauge theeffects of a dramatic pension reform. Underestimation of the value of leisuretranslates into sizeable over-prediction of the impact of reform. For thefemale sample, the model is able to predict almost perfectly the age-specifichazard rates. For the male sample, we obtain a good fit. Results for malesshould, however, be interpreted with caution since we are not able to fullycorrect for dynamic self-selection bias.
retirement option value model dynamic self-selection unobserved preference heterogeneity