Daniel Kemptner
This paper proposes a structural life-cycle model to analyze the relationship between health risks, early retirement and saving behavior for employees in Germany. I rely on the framework of a dynamic programming discrete choice model with a discretized saving decision. The model accounts for both forward looking behavior and unobserved heterogeneity which is specified seminonparametrically. Health and labor market risks are modelled as a joint stochastic process. This becomes relevant when simulating two counterfactual policy experiments: the first one changes pension benefits of early retirees and the second one introduces a saving subsidy encouraging precautionary savings. My results point to a trade-off between the hedging of health risks and an employee's incentive to remain in the labor force (?). Health-related poverty can be reduced at lower costs by increasing pension benefits than by a saving subsidy (?).
Keywords: Early Retirement, Saving Decision, Dynamic Programming, Discrete Choice, EM Algorithm
DIW-Link
Array