Summary: Traditional financial theory has little to say about hedging non-diversifiable risks. It assumes that the set of assets is given and focuses on the efficient sharing of these risks through exchange. This diversification strategy has no effect on macroeconomic shocks (such as an oil crisis) which affect all asset prices in a similar way. This paper focuses on the intertemporal smoothing of risk. Risks which cannot be diversified at a given point in time can be averaged over time in a way that reduces their impact on individual welfare. One hedging strategy for these risks is intergenerational risk sharing, which spreads the risks associated with a given stock of assets across generations with heterogeneous experiences. Another strategy involves asset accumulation in order to reduce fluctuations in consumption over time.
In standard financial models, it is usually argued that someone must bear the non-diversifiable risk. Such models implicitly assume away possibilities for intertemporal smoothing. At the other extreme, in an ideal Arrow-Debreu world, cross-sectional risk sharing and intertemporal smoothing are undertaken automatically if markets are complete and participation in those markets is complete. This paper considers the consequences of intertemporal smoothing for welfare and for positive issues such as asset pricing in a model with incomplete markets. In contrast to previous papers, the authors analyze how the risk arising from the dividend stream of long-lived assets is not eliminated by financial markets but can be eliminated by an intermediary.
The authors use a simple model with two assets, a risky asset in fixed supply and a safe asset that can be accumulated over time. They demonstrate that in market equilibrium the safe asset is not usually held but, in fact, is dominated by the risky asset. The authors argue that there is a serious form of market failure in the market equilibrium allocation. The standard definition of Pareto efficiency disguises the potential for achieving a substantial increase in welfare through intertemporal smoothing. They demonstrate that by accumulating reserves in the form of the safe asset and using them to "smooth" the returns to the risky asset, it is possible to increase the welfare of all but a negligible set of agents by a non-negligible and uniform amount. The authors then interpret intertemporal smoothing as the product of intermediation and suggest that the contrasting performance of the U.S. and German economies may be understood in terms of this intertemporal smoothing mechanism. They further demonstrate, however, that the mechanism is fragile and that competition from financial markets can lead to disintermediation, which causes the smoothing mechanism to unravel. The last section of the paper extends these arguments and applies the ideas to social security, occupational pensions and investment in housing.