Fuzzy Math and Household Finance: Theory and Evidence*

Recent work on intertemporal choice has varied the specification of every key aspect of modeling except the opportunity cost of consumption. We present evidence that consumers have present-biased perceptions of this parameter: they tend to underestimate the cost of short-term borrowing and the return to long-term saving. We develop a new theory that fits this evidence and is based on a more general cognitive tendency to underestimate exponential growth. The theory generates distinct, behavioral predictions about relationships between biased perceptions and portfolio choice, wealth levels, and the use and benefits of financial advice. These predictions are borne out in microdata