Where it sits
Useful in moderation, costly when it blocks learning, and worrying by its absence.
Tell it apart
| Aspect | Self-serving bias | FAE |
|---|---|---|
| Target | Yourself | Other people |
| Failure means | Circumstances | Character |
| Function | Protects esteem | Simplifies judgment |
| Together | A double standard | A double standard |
Both distort blame, in opposite directions and about different people. The FAE is about others: their failure means their character. Self-serving bias is about you: your failure means your circumstances. Run together, they produce the classic double standard — I was late because of traffic, you were late because you're careless.
What it doesn't mean
It's close to a default setting in healthy attribution, and runs mostly below awareness. Notably it shows up weaker in several clinical populations, and depression is associated with a reduced one. A missing self-serving bias is more often a warning sign than a virtue.
It's a perception effect, not a deception. People genuinely encode and recall the causal story that protects self-esteem — the runner who blames the course usually believes it.
What it really means
A 2022 study compared 1,320 marathon runners' official bib times with the times they reported themselves afterwards. The runners who had missed their personal best were significantly more likely to report a time faster than the one they had actually run, and the wider they missed by, the likelier the misreport. These were not people being evasive to a researcher — this is what they had concluded, sincerely, about their own race.
That's self-serving bias, and the name misleads slightly. It sounds like a description of someone shifty. It's closer to a standard setting on healthy attribution, running under the level where you could decide to switch it off.
The genuinely counterintuitive finding is what happens when it's absent. Research has found the bias attenuated in several clinical populations, and reduced self-serving attribution is associated with depression. Which reframes the whole thing: taking credit for your wins and finding external reasons for your losses isn't a character flaw that better people avoid. It's part of the machinery that keeps people attempting things after they fail.
The cost is specific rather than general. The bias is expensive exactly where learning depends on accurate causes — the trader who books every gain as skill and every loss as noise, the team whose hits are strategy and whose flops are the market. Not too much self-belief. Self-belief pointed at the one question where you needed the truth.
Where it comes from
as in self-serving — acting in one's own interest
The phrase implies deliberate advantage-taking. The effect is almost entirely unconscious.
- 1975
Dale Miller and Michael Ross publish the review that consolidates scattered attribution findings into a named bias, framing the debate over whether it is motivational (ego protection) or purely cognitive.
- 2022
A study of 1,320 marathon runners finds that those who missed their best finish time were significantly more likely to report a faster time than their bib actually recorded, and the wider they missed by, the likelier the misreport — the effect in the field, not the lab.
In real life
- When the pitch wins the account, it's because you prepared so well. When it loses, the client was never serious.
- A student gets an A and credits hard work; a C, and the exam was badly written.
- A runner who sets a personal best puts it down to training; a slow race was the heat, the hills, the shoes.
Questions
Q01Is self-serving bias always bad?
No, and this is the part that surprises people. It buffers self-esteem after setbacks, which keeps people trying. A salesperson who reads one lost deal as 'bad timing' keeps dialling; one who reads it as 'I'm not good at this' may stop. It turns costly when it blocks learning — the trader who books every gain as skill and every loss as market noise never fixes the strategy.
Q02How do I catch it in myself?
Write down your causal explanation before you know the outcome. If you predicted 'this pitch will land because the deck is strong' and it fails, and you then find yourself saying 'the client was disorganised', the explanation moved to protect you. Post-mortems asking 'what would I have had to do differently?' are also commonly recommended — though structured foresight exercises can push blame outward instead, so this one is advice rather than a tested result.
Q03Do groups do it too?
Yes — it scales up as group-serving bias. A team that ships a hit calls it product vision; the same team after a flop cites the market, the launch window, or a competitor. Earnings calls are a well-studied habitat for exactly this.
Sources
- Self-Serving Bias in Performance Goal Achievement Appraisals: Evidence From Long-Distance Runners — PMC.
- Attenuated self-serving bias in people with internet gaming disorder — PMC.
- Mezulis, Abramson, Hyde & Hankin, 'Is there a universal positivity bias in attributions?' (Psychological Bulletin, 2004).