Behavioral Insights of Reducing Administrative Burden: Evidence from a Field Experiment on Voting Method Choice

Last registered on September 21, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Behavioral Insights of Reducing Administrative Burden: Evidence from a Field Experiment on Voting Method Choice
RCT ID
AEARCTR-0019301
Initial registration date
September 12, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
September 21, 2026, 9:05 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Name
Tommy Morgan
Affiliation
University of Tennessee, Knoxville
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Michael Kofoed
PI Affiliation
University of Tennessee
Contact Investigator
PI Name
Joshua Price
PI Affiliation
Brigham Young University
Contact Investigator

Additional Trial Information

Status
On going
Start date
2026-06-01
End date
2026-12-31
Keywords
Electoral, Governance
Additional Keywords
voter registration
JEL code(s)
C93, D72, D91
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
During the mid 2010s, counties in Utah began to switch to universal vote by mail. By 2016, 21 out of 29 of Utah’s counties sent all their residents mail in ballots that they could return by mail or via drop box. In 2019, the entire state had moved to universal vote by mail. In 2025, however, the Utah State Legislature enacted HB300 that required the state to switch the default option back to in-person voting after the 2028 election. In the meantime, every registered Utah voter needs to sign into a complicated state website, navigate the voter registration form, and (after re-registering) indicate if they would like to continue to receive a by­-mail ballot or stop their by-mail ballot and vote in person. If a voter does not complete the process, then their mail ballot will automatically be stopped, and the voter will need to vote in-person.

The legislation also requires that the county clerks provide an informational insert with every ballot to inform voters about the change. We presented county clerks in Utah with the opportunity to participate in a randomized controlled trial in which we would help design three different versions of the informational insert that would have varying administrative burden and choice friction. The status quo option included information that the voter needed to make an active choice before the 2029 election to continue receiving their ballot by mail and included the URL for the state website where they could indicate their preference themselves. A second option had the same informational text plus a QR code that led to the same website. The third option had the same informational text but included a check box choice that allows the voter to indicate their preference to the county clerk's office directly, where the clerk's office would do all the work of registering their choice. Unnecessary complexity in decision-making can deter individuals from making a utility maximizing choice that they even claim they want to make, so we theorized that voters who receive the check box inserts would register their ballot method decision at a higher rate than voters in the other two treatments.
External Link(s)

Registration Citation

Citation
Kofoed, Michael, Tommy Morgan and Joshua Price. 2026. "Behavioral Insights of Reducing Administrative Burden: Evidence from a Field Experiment on Voting Method Choice." AEA RCT Registry. September 21. https://doi.org/10.1257/rct.19301-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
By law, the county clerks need to provide an insert to each voter and will be administering the three potential versions. We proposed three different versions of the informational insert that would have varying administrative burden and choice friction. The status quo option includes information that the voter needed to make an active choice before the 2029 election to continue receiving their ballot by mail and included the url for the state website to logon and indicate their preference. A second option had the same informational text plus a QR code that the voter can scan to get to the voter portal. The third option had the same informational text but includes a checked box that allows the voter to indicate their preference. The voter can then return the check box with their ballot and the county clerk will do all the work for them in the state system. The third option essentially shifts the administrative burden back to the county clerk and removes the decision friction.
Intervention Start Date
2026-06-01
Intervention End Date
2026-12-31

Primary Outcomes

Primary Outcomes (end points)
1) Probability of making a choice
2) Probability of making a given choice
3) Probability of returning a ballot
Primary Outcomes (explanation)
1) Does eliminating the decision friction (sometimes called sludge) and shifting the administrative burden from the voter to the county clerk increase the probability that a voter makes an active choice about their voting method by any method?
2) Does eliminating the decision friction (sometimes called sludge) and shifting the administrative burden from the voter to the county clerk increase the probability that a voter chooses to keep their by-mail ballot or only voter in-person?
3) Does eliminating the decision friction (sometimes called sludge) and shifting the administrative burden from the voter to the county clerk increase the probability that a voter participates in the election and returns their ballot to the county clerk’s office?

Secondary Outcomes

Secondary Outcomes (end points)
All the same outcomes, evaluated for various voter subgroups.
Secondary Outcomes (explanation)
Do the differing treatment arms affect various voter subgroups differently? For example, voting precincts in more Democratic vs Republican areas, rural vs urban, higher vs lower income, or average voter age of precinct?

Experimental Design

Experimental Design
The Utah Lieutenant Governor’s Office (LGO) requested that the research team help design the informational insert to better understand how to use behavioral economics principles to overcome the administrative burden and help voters make an active choice for their voter method. We proposed three different versions of the informational insert that would have varying administrative burden and choice friction. The status quo option includes information that the voter needed to make an active choice before the 2029 election to continue receiving their ballot by mail and included the url for the state website to logon and indicate their preference. A second option had the same informational text plus a QR code that the voter can scan to get to the voter portal. The third option had the same informational text but includes a checked box that allows the voter to indicate their preference. The voter can then return the check box with their ballot and the county clerk will do all the work for them in the state system. The third option essentially shifts the administrative burden back to the county clerk and removes the decision friction.

The county clerks from five counties have agreed to randomize receipt of one of the three versions on the voting precinct level. By law, the county clerks need to provide an insert to each voter and will be administering the three potential versions. The county clerks will be conducting the randomization (with some input from the research team) but the research team will have no contact with voters not have any part in the processing of ballots or updating the voter method preferences. Following the election, the Lieutenant Governor's Office will transmit de-identified data on the choices voters made to the research team.
Experimental Design Details
Not available
Randomization Method
Computer (Stata)
Randomization Unit
Randomized within counties at the voting precinct level.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
374 Precincts across 5 Counties
Sample size: planned number of observations
255,544 registered voters
Sample size (or number of clusters) by treatment arms
127 precincts with a text link
125 precincts with a QR code
122 precincts with a check box
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
University of Tennessee, Knoxville - Institutional Review Board - Office of Research, Innovation, & Economic Development
IRB Approval Date
2026-07-17
IRB Approval Number
STUDY00000890
Analysis Plan

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information