Pension Systems and the Allocation of Macroeconomic Risk

Working Paper: CEPR ID: DP5949

Authors: A. Lans Bovenberg; Harald Uhlig

Abstract: This paper explores the optimal risk sharing arrangement between generations in an overlapping generations model with endogenous growth. We allow for nonseparable preferences, paying particular attention to the risk aversion of the old as well as overall 'life-cycle' risk aversion. We provide a fairly tractable model, which can serve as a starting point to explore these issues in models with a larger number of periods of life, and show how it can be solved. We provide a general risk sharing condition, and discuss its implications. We explore the properties of the model quantitatively. Among the key findings are the following. First and for reasonable parameters, the old typically bear a larger burden of the risk in productivity surprises, if old-age risk-aversion is smaller than life risk aversion, and vice versa. Thus, it is not necessarily the case that the young ensure smooth consumption of the old. Second, consumption of the young and the old always move in the same direction, even for population growth shocks. This result is in contrast to the result of a fully-funded decentralized system without risk-sharing between generations. Third, persistent increases in longevity will lead to lower total consumption of the old (and thus certainly lower per-period consumption of the old) as well as the young as well as higher work effort of the young. The additional resources are instead used to increase growth and future output, resulting in higher consumption of future generations.

Keywords: overlapping generations; pension systems; risk sharing; social optimum

JEL Codes: E21; E61; E62; H21; H55; O40


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
risk aversion of the old (D81)consumption dynamics of the old (E21)
risk aversion of the young (D81)consumption dynamics of the old (E21)
old-age risk aversion (J26)consumption stability (E21)
population growth shocks (J11)consumption of the young (J13)
population growth shocks (J11)consumption of the old (E21)
increased longevity (D15)total consumption of the old (E20)
increased longevity (D15)work effort from the young (D29)

Back to index