Reinforcement Learning and Savings Behavior






    Search for more papers by this author
    • Choi is at the Yale School of Management and NBER. Laibson is at Harvard University, Department of Economics, and NBER. Madrian is at the Harvard Kennedy School and NBER. Metrick is at the Yale School of Management and NBER. This paper is a significantly revised version of a draft that circulated under the title “Consumption-Wealth Comovement of the Wrong Sign.” We thank Hewitt Associates for their help in providing the data. We are particularly grateful to Lori Lucas, Jim McGhee, Scott Peterson, and Yan Xu, some of our many contacts at Hewitt. Outside of Hewitt, we have benefited from the comments of Thomas Knox; Matthew Shapiro; Bill Zame; and seminar participants at Berkeley, Harvard, the NBER Monetary Economics Program Meeting, UCLA, and RAND. We are also indebted to John Beshears, Lucia Chung, Ian Dew-Becker, Keith Ericson, Fuad Faridi, John Friedman, Chris Nosko, Parag Pathak, Nelson Uhan, and Kenneth Weinstein for their excellent research assistance. Financial support from the National Institute on Aging (grants Nos. R01-AG021650 and R01-AG16605) is gratefully acknowledged. Choi acknowledges financial support from a National Science Foundation Graduate Research Fellowship, National Institute on Aging Grant No. T32-AG00186, and the Mustard Seed Foundation. Laibson acknowledges financial support from the National Science Foundation.


We show that individual investors over-extrapolate from their personal experience when making savings decisions. Investors who experience particularly rewarding outcomes from 401(k) saving—a high average and/or low variance return—increase their 401(k) savings rate more than investors who have less rewarding experiences. This finding is not driven by aggregate time-series shocks, income effects, rational learning about investing skill, investor fixed effects, or time-varying investor-level heterogeneity that is correlated with portfolio allocations to stock, bond, and cash asset classes. We discuss implications for the equity premium puzzle and interventions aimed at improving household financial outcomes.