Hyperbolic Discounting and the Standard Model

Experiments on time preference document numerous .ndings that seem to con- tradict the standard model of intertemporal choice. These .ndings are based on how subjects choose between delayed rewards. This paper shows that if subjects integrate such rewards with their consumption plans, and expect changes in future consump- tion, then except for violations of basic properties like transitivity, the standard model (with CRRA utility) can rationalize all the popular experimental .ndings: preference reversals, dynamic inconsistency, hyperbolic discounting, magnitude e¤ect, sign e¤ect, delay-speedup asymmetry etc. It is demonstrated formally that the standard model has no peculiar testable implications for subjects.preferences between delayed money rewards, which is the data of typical experiments. A testable implication of the model is derived in the richer setting with risky prospects as rewards.