Discussion Paper, Center for Mathematical Studies in Economics and Management Science 1571
This paper argues that, to be forward-looking in a logically consistent sense, a decision maker must take account of his overall well-being, not just his instantaneous utility, in all future periods. However, such a decision-maker is necessarily time inconsistent. The paper explores the relationship between how a decision-maker discounts well-being and how he discounts instantaneous utility. It also provides simple axiomatizations of preferences that exhibit forward-looking behavior, including quasi-hyperbolic discounting (Phelps and Pollack (1968) ; Laibson (1997) ). Finally, the paper provides a rigorous way to think about welfare criteria in models with time inconsistent agents.