The Welfare Economics of Default Options : A Theoretical and Empirical Analysis of 401 ( k ) Plans ∗

According to previous research, changing the default contribution rate for a 401(k) pension plan has a powerful effect on the distribution of contributions among relatively new employees. We study the welfare effects of such changes and identify the default rates that maximizes total surplus. Because one can rationalize certain types of default effects in conventional models with switching costs, we begin with that setting. We identify conditions under which the optimal default is zero (equivalently, no automatic enrollment), and demonstrate that the optimum often diverges from the rate that minimizes the frequency with which employees opt out of the default (contrary to some previous assertions). The method is implemented empirically for three real firms. We then identify a collection of non-standard behavioral considerations that arguably have implications for the choice of a default: (1) the default rate may serve as a psychological anchor, influencing the choices even of those who depart from the default; (2) adherence to the default may reflect inattentiveness; and (3) adherence to the default may reflect procrastination (arising from time-inconsistency). In each case, we study the welfare effects of default rates and derive optimal defaults by applying the framework for behavioral welfare analysis developed by Bernheim and Rangel [2009]. We also analyze the quantitative implications of each behavioral consideration for the same three firms. ∗We would like to thank seminar participants at the CESifo Venice Summer Institute Conference on Behavioural Welfare Economics for helpful comments. The first author has benefited immeasurably from numerous conversations with Antonio Rangel concerning the topic of behavioral welfare economics, which have spanned many years. We acknowledge financial support from the National Science Foundation through grant number SES-0752854.