Boomerang effect
A norm message that backfires for people already doing well.
What it means
The boomerang effect is the backfire that occurs when a descriptive-norm message tells above-average performers what others typically do, inadvertently pulling them back toward the worse average. The mechanism is that descriptive norms act as a magnet toward the norm in both directions: telling heavy users that others use less nudges them down, but telling efficient users the same average signals that they are doing more than necessary and may relax. The classic energy study showed that low consumers increased their usage after learning the neighborhood average, undoing the intended saving among the very people who were already behaving well. The documented remedy is to pair the descriptive norm with an injunctive signal of approval — a simple smiley face or message of social approval for the desirable behavior — which keeps good performers motivated while still nudging poor performers toward the average. It matters for the design of social-comparison feedback in energy, health, finance, and tax compliance, where a norm message that ignores this asymmetry can quietly harm the best-behaved segment.
How it works
The mechanism is comparison, not persuasion. A descriptive norm hands each person a benchmark and quietly invites them to close the gap between their own behavior and it. Someone worse than the average sees room to improve; someone better than it has been shown, in effect, that they are overshooting. Because a bare average carries no verdict on whether high or low is the good direction, it pulls both tails toward the middle. Adding an injunctive cue — a marker of approval or disapproval — repairs this by separating what is typical from what is right. The strong performer now learns that being below the average is not merely unusual but praised, which supplies a reason to hold position rather than drift back toward the pack.
What the evidence shows
The 2007 field experiment is the founding demonstration: door hangers reporting neighborhood consumption, with a hand-drawn smiley or frown as the injunctive layer. The descriptive-only message raised usage among already-frugal homes, and the emoticon removed that rebound. How general the effect is turns out to be a harder question. Allcott's analysis of roughly 600,000 households on Opower's Home Energy Reports found the average report cut use about 2 percent, with the largest savings among heavy users — but no meaningful boomerang among light users, plausibly because those reports already carried injunctive smileys and action tips. In college drinking, Prince and colleagues found no boomerang among light drinkers across four samples. Reviewers now describe the evidence as inconsistent rather than settled.
Where it shows up
The vulnerable ingredient is structural: any feedback that broadcasts a single central tendency to a mixed audience puts its best-behaved segment at risk. Utility bills comparing a household to its neighbors, water-use reports during droughts, personalized drinking feedback on campuses, retirement-savings dashboards, and tax reminders quoting a compliance rate all share this shape. The danger is largest when the desirable direction is the minority one, so the average sits below the exemplary user. It shrinks when good behavior is already common, because then the average itself points the right way. This is why careful designers avoid publishing a lone number and instead show a target, a distribution, or an explicit approval marker beside the comparison.
Related but distinct
The boomerang is easily confused with psychological reactance, but the two run on different engines. Reactance is a reaction against perceived control: a heavy-handed "you must conserve" provokes defiance and the opposite behavior. The norm boomerang needs no sense of pressure at all — it emerges from a neutral, even friendly, statement of what others do, because that statement doubles as a benchmark the good performer is already beating. The remedies differ too. Reactance is eased by softening the coercive tone; the boomerang is eased by adding an evaluative signal that tells the over-performer their behavior is valued. Treat one as the other and you fix the wrong thing — softening language where you should be adding approval, or the reverse.
Examples
Telling low energy users their neighbors use more nudges them up — adding a smiley face to efficient users keeps them efficient.
A university tells students that most of their peers drink less than they imagine. The lightest drinkers hear an average higher than their own habit, and drift up towards it.
A tax letter saying 'nine in ten people pay on time' can nudge the always-punctual towards the deadline; a line thanking them for paying early keeps them early.
A clinic mails each doctor a chart of their antibiotic-prescribing rate against the practice average. A physician who already prescribes sparingly, told they sit well below their peers, reads the average as the expected rate and edges their prescribing upward.
A pension dashboard tells a member that most colleagues save a modest slice of salary. An employee already putting away far more takes the typical figure as the expected level and quietly trims their rate.
First described in Schultz, Nolan, Cialdini, Goldstein & Griskevicius (2007).
Key references
- Kuang, J., Delea, M. G., Thulin, E., et al. (2020). Do descriptive norms messaging interventions backfire? Protocol for a systematic review of the boomerang effect. Systematic Reviews, 9, 267. doi.org/10.1186/s13643-020-01533-0
- Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J., & Griskevicius, V. (2018). The constructive, destructive, and reconstructive power of social norms: Reprise. Perspectives on Psychological Science, 13(2), 249-254. doi.org/10.1177/1745691617693325
- Prince, M. A., Reid, A., Carey, K. B., & Neighbors, C. (2014). Effects of normative feedback for drinkers who consume less than the norm: Dodging the boomerang. Psychology of Addictive Behaviors, 28(2), 538-544. doi.org/10.1037/a0036402
- Allcott, H. (2011). Social norms and energy conservation. Journal of Public Economics, 95(9-10), 1082-1095. doi.org/10.1016/j.jpubeco.2011.03.003
- Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J., & Griskevicius, V. (2007). The constructive, destructive, and reconstructive power of social norms. Psychological Science, 18(5), 429-434. doi.org/10.1111/j.1467-9280.2007.01917.x