The decision of when to claim Social Security Administration retirement benefits is a highly individual decision, incorporating family situations, income and consumption needs, and substantial uncertainty around individual longevity and health status. Beyond economic issues of when to optimally claim, individual differences in psychological preferences can also affect the claiming decision. While different strands of research have investigated the impact of heterogeneity on the claiming decision, as well as the impact of information displays and interventions on claiming intentions, there has been little work pulling together both measurement of individual differences and impacts of interventions in the same experiment. In this research, we test 1) the effectiveness of 13 different interventions meant to help consumers better reason about claiming age, 2) the role that a variety of individual differences play in claiming age intentions, and 3) interactions between interventions and individual differences on claiming age intentions. Per a preregistered design, we recruited 4,500 pre-screened participants, and then analyzed intended claiming age as a linear function of the indicator variables, one for each treatment arm. We also tested each individual difference measure and 23 theoretically relevant preregistered interactions. We replicate these results in a follow on study and, in so doing, better model the claiming decision and thus generate better predictive models of which retirees will decide to claim earlier or later, as well as suggest which interventions may be most effective for targeted populations.

AI, Big Data, and Behavioral Science Workshop Series

• PK Kannan (U of Maryland): October 4, 2019
  
• Suzanne Shu (UCLA): November 18, 2019
  
• Mandy Hu (CUHK): January 10, 2020