Deadline-Reward Experiment from "Overcoming Frictions in Retirement Savings with Microincentives"
How to cite
Bhargava, Saurabh; Conell-Price, Lynn. 2026. "Deadline-Reward Experiment from "Overcoming Frictions in Retirement Savings with Microincentives"." AEA RCT Registry. AEARCTR-0019298. https://doi.org/10.1257/rct.19298-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
Many employees contribute too little to their employer-sponsored retirement plans to obtain the full employer match, even when doing so would generate a large and immediate financial return. A broader research project examines whether this behavior reflects informational frictions—such as uncertainty about how much to save or misunderstanding of the employer match—or frictions associated with engaging in and completing the contribution decision. In a field experiment with administrative contribution outcomes, personalized savings guidance and match clarification changed employees’ reported beliefs but did not measurably increase contributions, whereas a small, same-day engagement reward increased actual contributions. Because that field intervention combined a monetary reward with a same-day eligibility window, it does not separately reveal whether the response required such a short deadline. This individually randomized online experiment provides supplementary mechanism evidence by comparing a common recommendation to save more with a hypothetical $20 Amazon gift-card offer for increasing a 401(k) contribution by the end of today or within two weeks. Employed U.S. adults imagine that they earn $50,000, currently contribute 2 percent of salary, and can change their contribution in a few minutes. The experiment tests whether a reward with a longer eligibility window increases intended action during that window and whether shortening the window to the end of today accelerates intended action into the present.
Intervention
Common scenario
All participants are asked to imagine that they are employed by ABC Company, earn $50,000 per year, and currently contribute 2 percent of salary to the ABC Company 401(k). Contributions are described as automatically deducted from pre-tax pay over the year and invested in a diversified retirement fund. Participants are told that they can review and change their contribution at any time through the ABC benefits portal and that logging in and making a change takes only a few minutes.
All participants then receive the same statement: “You receive an email from ABC Company recommending that you save more for retirement by increasing your contribution to the ABC 401(k).”
Randomized conditions
1. Baseline: Participants receive only the common recommendation. The condition does not mention a reward or the absence of a reward.
2. Reward by end of today: Participants additionally read, “To encourage you to take action, ABC is offering you a $20 Amazon gift card if you log in to the benefits portal and increase your contribution by the end of today. The gift card will be emailed after you submit the increase.”
3. Reward within two weeks: Participants additionally read, “To encourage you to take action, ABC is offering you a $20 Amazon gift card if you log in to the benefits portal and increase your contribution within two weeks from now. The gift card will be emailed after you submit the increase.”
The $20 gift card is hypothetical. Participants receive only the participation payment stated in the Prolific study listing.
Outcome question
On the same page as the randomized email, participants are asked, “What action do you think you would take?” Response options appear in the following fixed chronological order:
1. I would increase my contribution today.
2. I would increase my contribution in the near future—sometime between tomorrow and 14 days from now.
3. I would increase my contribution later—sometime between 15 days and six months from now.
4. I would not increase my contribution within the next six months.
Experimental design
This is an individually randomized, three-arm online vignette experiment. Every participant sees one condition and answers one required behavioral-intention question. The two reward conditions differ only in the eligibility window attached to an otherwise identical hypothetical $20 Amazon gift card.
Design details
Cited by
No citation data available for this registration yet.