NB22-10: How do Behavioral Approaches to Increase Savings Compare? Evidence from Multiple Interventions in the U.S. Army

In recent years, employers have developed a number of approaches to encourage retirement savings among their employees, and academic researchers have studied their varying effects. Research has documented that information provision, choice simplification, social messaging, active-choice frameworks, and automatic enrollment all increase retirement savings. However, gauging the relative efficacy of these approaches has been challenging because the existing research relies on different populations over a long period and very few studies have incorporated costs into their analyses. Moreover, few analyses have explored the potential heterogeneous effects of these different policy approaches with respect to individual characteristics including gender, race and ethnicity, education level, and family structure. Our preliminary analysis suggests that active choice programs are the most cost-effective method to generate new program participation and contributions for small, median, and large firms, while automatic enrollment is more cost-effective for very large firms.
The final study will complete three analyses. First, we leverage quasi-experimental and experimental variation within a single institutional setting to examine the relative efficacy of four leading policy approaches designed to increase retirement savings: behaviorally informed emails (e.g., action steps), simplification of enrollment via the provision of target contribution rates, active choice enrollment programs (i.e., requirement to explicitly state a choice by a deadline), and automatic enrollment programs. Second, we analyze the demographic heterogeneity of these effects in a workforce that is younger, less educated, and more racially and ethnically diverse than is found in other settings in which retirement savings is studied. Finally, we complement these analyses with administrative and estimated measures of program costs. These data enable us to estimate the relative cost-effectiveness of each program (e.g., the dollars spent per new enrollment or the dollars spent per new dollar of contributions).