Recursive preferences characterize the trade-offs between current and future consumption by summarizing the future with a single index, the certainty equivalent of next period’s utility. Recursive utility functions are built from two components. A risk aggregator encodes trade-offs across the outcomes of a static gamble and, hence, defines the certainty equivalent of future utility. A time aggregator encodes trade-offs between current consumption and the certainty equivalent of future utility. We suggest functional forms for time and risk aggregators with desirable properties for applications in economics and finance, such as the standard intertemporal consumption/portfolio problem, which we solve using dynamic programming.
KeywordsBellman equation Certainty equivalent Disappointment aversion Dynamic optimization Elasticity of intertemporal substitution Expected utility Impatience Infinite horizons Preferences Rational expectations Recursive preferences Risk aggregator Risk aversion Stochastic dynamic models Time aggregator Time preference Utility functions Weighted utility
JEL ClassificationsD4 D10
- Backus, D.K., Routledge, B.R. and Zin, S.E. 2004. Exotic preferences for macroeconomists. NBER macroeconomics annual 2004 , vol. 19 M. Gertler K. Rogoff. Cambridge, MA: MIT Press.Google Scholar
- Hansen, L.P., and T.J. Sargent. 2004. Misspecification in recursive macroeconomic theory. Manuscript: University of Chicago, January.Google Scholar
- Kreps, D.M. 1988. Notes on the theory of choice. Boulder: Westview Press.Google Scholar
- Routledge, B.R. and Zin, S.E. 2003. Generalized disappointment aversion and asset prices. Working Paper No. 10107. Cambridge, MA: NBER.Google Scholar
- Uzawa, H. 1968. Time preference, the consumption function, and optimum asset holdings. In Value, capital, and growth: Papers in honour of Sir John Hicks, ed. J.N. Wolfe. Chicago: Aldine.Google Scholar