Using Participant Data to Improve Target Date Fund Allocations

Abstract

Economic theory says that participants in 401(k) plans should gradually rebalance their portfolios away from stocks and toward less risky bonds as they approach retirement.  Conventional target date funds attempt to do so by automatically rebalancing the household’s portfolio periodically, but they take account of only one aspect of the individual: his expected retirement age.  This paper investigates whether plan providers could improve on this “one-size-fits-all” approach by making use of information that is known to the employer, namely each employee’s income, 401(k) balance, and saving rate.  Using a stochastic dynamic optimization model, incorporating both labor- and financial-market risk, it calculates the compensation a household following an optimal portfolio allocation would require for adopting three alternatives: a typical, a “one-size-fits-all,” or a “semi-personalized” portfolio allocation.