Skip to main content
UNITED STATES DEPARTMENT OF LABOR

Framing flexible spending accounts: A large-scale field experiment on communicating the return on medical savings accounts (Leight & Wilson, 2020)

Review Guidelines

Causal Evidence Guidelines

Absence of conflict of interest.

Citation

Leight, J., & Wilson, N. (2020). Framing flexible spending accounts: A large-scale field experiment on communicating the return on medical savings accounts. Health Economics, 29(2), 195-208. https://doi.org/10.1002/hec.3965

Highlights

  • The study’s objective was to examine whether additional messaging about a Flexible Spending Account (FSA) increased employees’ usage of an FSA. 
  • The study used a randomized controlled trial to assign participants to the treatment or control group. The authors used data from anonymous administration employee records to compare the changes in FSA contribution by the type of messaging received.   
  • The study found no statistically significant relationships between the type of messaging and FSA contributions. 
  • This study receives a high evidence rating. This means we would be confident that the estimated effects were attributable to additional messaging about an FSA, and not to other factors. However, the study did not find any statistically significant effects.  

Features of the Study

Flexible Spending Accounts (FSA) allow employees to direct pretax earnings from their paychecks into an account to receive reimbursements for medical expenses. The study used a randomized controlled trial to compare FSA contributions in 2018 based upon the type of messaging that was provided to employees of the U.S General Services Administration (GSA) in Washington D.C. A total of 11,192 GSA employees were randomly assigned to either the control group or a treatment group that received one of three different types of messages to enroll in the agency’s FSA through the online newsletter GSA Today. Employees either received: 1) a basic message to contribute to the agency FSA; 2) an absolute savings message that informed employees that they could save $949 dollars by contributing to the FSA; or 3) a relative saving message that informed employees they could receive a 36.5% tax savings by contributing to the FSA. Employees in the control group did not receive any messages about contributing to the FSA. FSA contribution data were obtained from the GSA human resources administrative records. The administrative records were anonymous, so demographic data such as race, gender, and income were not available; however, study authors received data about how much employees contributed to the FSA the prior year in 2017. The study used statistical models to compare the changes in FSA contribution by the type of messaging received. 

Findings

Employer benefits receipt

  • The study found no statistically significant impacts between the type of messaging received and FSA contributions.  

Considerations for Interpreting the Findings

The quality of causal evidence presented in this report is high because it was based on a well-implemented randomized controlled trial. This means we would be confident that any estimated effects would be attributable to additional FSA messaging, and not to other factors. However, the study did not find statistically significant effects.  

Reviewed by CLEAR

May 2024

Topic Area