Behavioral Economics Workshop: Saurabh Bhargava

Saurabh Bhargava (Carnegie Mellon)

Save(d) by Design
Joint with Lynn Conell-Price, Rick Mason, Shlomo Benartzi

Abstract: Despite the successes of automatic enrollment (AE) in raising participation, a significant share of 401(k) enrollees remain at-risk for retirement insecurity. We assert that this risk may be shaped by an employee’s initial decision to either confirm enrollment at the default rate or to personalize enrollment at an adjusted rate and that this decision itself may be heavily influenced by non-economic features (e.g., the use of visual elements like color, the standardization of language, the salience of previously disclosed plan details) of the (digital) enrollment experience. In collaboration with a major financial provider, we test whether design affects enrollment decisions by randomizing several thousand potential enrollees across 500 plans to digital interfaces that vary only in their “psychological design.” The field studies, supplemented by lab experiments with hypothetical enrollment decisions and a survey of plan administrators, yield five main findings. Our primary contribution is to document the large influence of psychological design on enrollment. The enhanced design we test increased the share of personalized enrollment by 0.09 (0.60 baseline) and full match take-up by 0.11 (0.58 baseline) and raised average contributions by 62 basis points (5.41 percent baseline)—predictively equivalent to expanding the typical plan match by 62%. Second, enhanced design not only increased average savings but amplified employee sensitivity to the generosity of the plan match, offering evidence for a novel complementarity between economic and psychological design. Third, we show that the marginal enrollee who personalizes enrollment due to design behaves similarly to their inframarginal counterparts in substantively increasing their contribution relative to the default. This implies that plans could meaningfully increase savings by heightening employee exposure to personalized enrollment. Fourth, lab evidence suggests that design does not influence behavior by shifting beliefs or preferences for savings, as might be predicted by standard economic models of enrollment, pointing instead to the potential role of non-standard factors such as employee confusion or cue-based heuristics. Finally, surveys reveal misunderstanding of the influence of psychological design among those administering plans, highlighting the counterintuitive nature of our results and the practical value of expanding notions of fiduciary duty to encompass the previously underexamined economic influence of digital design.