Understanding probabilistic expectations – a behavioral approach

Abstract

I propose a simple heuristic to explain how people estimate the probability of an event that has binary outcomes, such as a boom vs. a recession. Central to this approach is a psychological trait called “base-rate neglect” — a tendency for people to ignore prior probabilities in favor of more salient diagnostic information. When making predictions, people’s decisions are experience-based, but are influenced by base-rate neglect. I construct an algorithm to replicate this behavior, and find that its predictions match key features of the Anxious Index in the Survey of Professional Forecasters and stock market expectations in the Surveys of Consumers. If investors use this heuristic to form expectations in a consumption-based asset pricing model, both the equity premium and its volatility will be higher than those in a rational expectations model.