« Back to Results

Memory, Perception, and Asset Prices

Paper Session

Sunday, Jan. 5, 2020 8:00 AM - 10:00 AM (PDT)

Manchester Grand Hyatt, Seaport H
Hosted By: American Finance Association
  • Chair: Cary Frydman, University of Southern California

A Retrieved-Context Theory of Financial Decisions

Jessica Wachter
,
University of Pennsylvania
Michael Kahana
,
University of Pennsylvania

Abstract

Studies of human memory indicate that features of an event evoke memories of prior associated contextual states, which in turn become associated with the current event's features. This mechanism allows the remote past to influence the present, even as agents gradually update their beliefs about their environment. We apply the context framework from the memory literature to four problems in asset pricing and portfolio choice: over-persistence of beliefs, providence of financial crises, price momentum, and the impact of fear on asset allocation. These examples suggest a recasting of neoclassical rational expectations in terms of beliefs as governed by principles of human memory.

Investor Memory

Katrin Gödker
,
Maastricht University
Peiran Jiao
,
Maastricht University
Paul Smeets
,
Maastricht University

Abstract

How does memory shape individuals' financial decisions? We find experimental evidence of a self-serving memory bias, which distorts beliefs and drives investment choices. Subjects who previously invested in a risky stock are more likely to remember positive investment outcomes and less likely to remember negative outcomes. In contrast, subjects who did not invest but merely observed the investment outcomes do not have this memory bias. Importantly, subjects do not adjust their behavior to account for the fallibility of their memory. After investing, they form overly optimistic beliefs and re-invest in the stock even when doing so reduces their expected return. The memory bias we document is relevant for understanding how people form expectations from experiences in financial markets and, more generally, for understanding household financial decision-making.

Biased Assessment of Comovement

Ben Matthies
,
University of Notre Dame

Abstract

I document a systematic bias in the assessment of comovement: individuals assess a moderate relationship between two variables regardless of the actual strength of the relationship between them. In a survey of finance professionals, participant-assessed betas of different financial and macroeconomic variables with the stock market exhibit compression towards moderate values. In an empirical setting, electricity futures exhibit moderate comovement with gas futures despite persistent heterogeneity in the comovement of gas and electricity in the spot market. Trading against this bias generates annualized excess returns of 7.3 percent and a annualized Sharpe ratio of 1.14. Finally, professional forecasters also exhibit this bias, leading to predictable errors in macroeconomic forecasts.
Discussant(s)
Alp Simsek
,
Massachusetts Institute of Technology
Colin Camerer
,
California Institute of Technology
Valentin Haddad
,
University of California-Los Angeles
JEL Classifications
  • G1 - General Financial Markets