The Microeconometrics Seminar: Every Tuesday at 12:15 pm.
Time: 12:15 pm – 1:30 pm
Date: 9 th of April 2019
Place: Room 3001
Saurabh BHARGAVA (Carnegie Mellon University) – “Save(d) by Design”

Abstract: While discussions on retirement preparedness usually cite challenges associated with insufficient 401(k) access and participation, we present new administrative evidence from hundreds of automatic enrollment (AE) plans suggesting that low preparedness among US employees extends to a large share of actual enrollees and, further, that such preparedness greatly varies across employers. To explore whether these patterns can be attributed to variation in the “psychological design” of initial enrollment across AE plans—namely, non-economic features of the online interface from which employees can confirm, personalize or decline enrollment and the intensity with which employers encourage use of this interface—we administer a series of large-scale field experiments across 500 401(k) AE plans. The field studies, supplemented by hypothetical choice experiments, a survey of hundreds of plan administrators, and heterogeneity analyses across plans, yield four main findings. First, modest changes to the interface design significantly raise the likelihood of personalized enrollment (+0.09 from a 0.60 baseline) and full match take-up (+0.11 from 0.58) as well as the average contribution rate (+0.62 from a 5.41 baseline)—equivalent to that predicted from a 62% increase in the plan match. Second, the savings behavior of the marginal employee who personalized enrollment or visited the online interface, resembles that of infra-marginal counterparts. This implies that the considerable naturalistic variation in the design of enrollment across plans may help explain much of the variation in employee preparedness and that simple modifications to design could sharply reduce the risk of retirement poverty. Third, lab evidence supports the possibility that enrollment decisions are governed by a heuristic choice process in which non-economic cues play a meaningful role. Finally, surveys pointing to the inability of those overseeing and administering enrollment to forecast the influence of specific design elements or the overall influence of design relative to matching incentives suggest the value of expanding our understanding of fiduciary responsibility to include the large, and largely unrecognized, influence of digital design on employee well-being.

Organizers:

Laurent DAVEZIES (Laboratoire de Microéconométrie-CREST)
Benoît SCHMUTZ (Laboratoire de Microéconométrie-CREST)
Arne UHLENDORFF (Laboratoire de Microéconométrie-CREST)
Sponsors:
CREST