How Much to Save? Decision Costs and Retirement Plan Participation

How to cite

Patterson, Richard; Skimmyhorn, William; Goldin, Jacob; Homonoff, Tatiana. 2019. "How Much to Save? Decision Costs and Retirement Plan Participation." AEA RCT Registry. AEARCTR-0004588. https://doi.org/10.1257/rct.4588-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Richard Patterson* , William Skimmyhorn , Jacob Goldin , Tatiana Homonoff

* Primary investigator

Registry ID
AEARCTR-0004588
DOI
https://doi.org/10.1257/rct.4588-1.0
First registered
Last updated
Status
Completed
Trial dates
2016-01-27 – 2019-08-15
Countries
United States of America ()
JEL codes
D14, D91, G41
Registry record
http://www.socialscienceregistry.org/trials/4588

Abstract

Deciding how much to save for retirement can be complicated. Drawing on a field experiment conducted with the Department of Defense, we study whether such complexity depresses participation in an employer-sponsored retirement saving plan. We find that simplifying one dimension of the enrollment decision, by highlighting a potential rate at which non-participants might contribute, increases participation in the plan. Similar communications that did not include a highlighted rate yielded smaller effects. The results highlight how reducing complexity on the intensive margin of a decision (how much to contribute) can spill over into extensive margin behavior (whether to contribute at all).

Intervention

In this study, we investigate whether the complexity of the retirement savings decision depresses participation in retirement savings plans. We analyze the results of a field experiment conducted by the U.S. Department of Defense (DOD). The experiment involved the roughly 300,000 active duty U.S. Army servicemembers who were not enrolled in the Thrift Savings Plan(TSP), the defined contribution portion of the retirement plan the U.S. government offers to its employees. The DOD randomly assigned a subset of these individuals to receive a one-time email that provided information about how to enroll in the TSP and encouraged them to join. For some randomly assigned individuals, the email also attempted to reduce the complexity of the savings decision by highlighting a specific rate (i.e., 1%, 2%, ...8%) at which the individual could choose to contribute. A final group of individuals were randomly assigned to a control group that did not receive any email.

Experimental design

In January 2016, the Department of Defense (DOD) randomly assigned all active-duty Armed Service servicemembers who had not contributed to a TSP account (Including approximately 300,000 Army servicemembers) to be assigned to one of the following 10 treatment groups:
1) Control group (no intervention)
2) Receive a one-time email that provided information about how to enroll in the TSP and encouraged them to join.
3-10) Receive a one-time email that provided information about how to enroll in the TSP and encouraged them to join and the following the message: "Many servicemembers like you start by contributing at least X% of their basic pay into a Traditional or Roth TSP account,", where X% varied between 1%, 2,%, 3%, 4%, 5%, 6%, 7%, and 8% for treatment groups 3-10.

Design details

Randomization method
Last 2 digits of SSN
Randomization unit
Individual
Clusters
291,552 Army Servicemembers
Observations
291,552 Army Servicemembers
Arms
291,552 Army Servicemembers
Primary outcomes
(1) Participation in Thrift Savings Plan (TSP) (2) Percent of income contributed (3) Total savings (4) Indicator for contributing specific rates (i.e. 1, 2,...,8%)
Secondary outcomes
Types of TSP funds

Cited by

No citation data available for this registration yet.

Keywords