To the depths of the sunk cost: Experiments revisiting the elusive effect
Beknazar-Yuzbashev and colleagues note that the sunk cost effect, endlessly discussed in practice and in teaching, remains difficult to demonstrate cleanly, and argue that the usual test design is part of the reason. They ran two pre-registered experiments. In the first, more than eighteen hundred people paid one of three prices to enter a task and were then free to spend as long on it as they liked; those who had paid more stayed longer, a moderate effect amounting to a little over a minute on a fourteen-minute task. The second was a field experiment on more than eleven thousand videos, varying how long viewers had to wait before an advertisement could be skipped. Overall engagement did not change reliably, because a sizeable number of people simply left before the video began when the wait was longer. The conclusion is even-handed: the effect is real in a controlled setting and hard to exploit in a policy or product setting.
Beknazar-Yuzbashev and colleagues note that the sunk cost effect, endlessly discussed in practice and in teaching, remains difficult to demonstrate cleanly, and argue that the usual test design is part of the reason. They ran two pre-registered experiments. In the first, more than eighteen hundred people paid one of three prices to enter a task and were then free to spend as long on it as they liked; those who had paid more stayed longer, a moderate effect amounting to a little over a minute on a fourteen-minute task. The second was a field experiment on more than eleven thousand videos, varying how long viewers had to wait before an advertisement could be skipped. Overall engagement did not change reliably, because a sizeable number of people simply left before the video began when the wait was longer. The conclusion is even-handed: the effect is real in a controlled setting and hard to exploit in a policy or product setting.
Written from the abstract, OpenAlex.
What our sources record
Reproduced as it was written when this source was checked, figures and all.
241, 107352. Pre-registered. Online real-effort task, N = 1,806, a $2 price variation on a fourteen-minute task producing 0.12 SD, about 1.3 extra minutes. YouTube field experiment across 11,328 videos varying pre-roll ad length before skippability: overall effect on engagement not significant, with 5.2 percentage points more abandonment before the video began in the extended condition. Abstract only.
Where we use it
- Why You Finish the Series You Stopped Enjoying, Why We Act
Reliability of this reference
The link below was matched against the publisher's record on title, first author and year, and all three agree.
The source
https://doi.org/10.1016/j.jebo.2025.107352
DOI: 10.1016/j.jebo.2025.107352