Loss aversion
The tendency for a loss to weigh more than a gain of the same size, judged from a reference point such as what one already has or expected. Losing £100 is felt and avoided more than winning £100 is sought.
What drives it
Weighting losses heavily protects what a person already has, which helps where losses are hard to recover from. For identification it explains refusing fair bets, holding falling shares, and fighting harder to keep a benefit than to win a new one.
How it shows
Work
Fighting to keep a budget line that is no longer needed, turning down better pay because the pension scheme is worse, framing every change as what will be taken away. A pay cut provokes more anger than a pay rise brings pleasure.
Relationships
Arguments about what one partner will ‘give up’ in a move or in merging finances rather than what both gain. Holding on to a flat or a car rather than sell below the purchase price.
Self
Notice whether you keep a losing investment until it gets back to what you paid, whether a £20 parking fine sours a day more than a £20 windfall lifts it, and how a choice changes when the same facts are described as a loss.
What tells it apart
Sunk cost fallacy: Sunk cost is continuing because of what has already been spent. Loss aversion is extra weight on future losses over future gains, though both show up when someone won’t sell at a loss.
Anchoring: Anchoring is the pull of an arbitrary first number on an estimate. Loss aversion also needs a reference point, but the effect is the extra weight on falling below it.
Money vigilance: Money vigilance is a learned belief in caution and saving. Loss aversion is a weighting found in most people, better seen in how the same option is judged when framed as a loss or as a gain.
A test
Describe the same choice twice, once as what will be kept and once as what will be lost. A sharp change of mind between the two versions makes loss aversion, or framing more broadly, more likely.
Questions to carry
- When the person refused the change, were they talking about what they would gain or what they would lose?
- What am I holding on to only because selling it now would mean admitting a loss?
How strong the research is
mixed. A 2024 meta-analysis of 607 estimates (Brown, Imai, Vieider and Camerer) puts the average loss aversion coefficient close to 2, meaning a loss weighs about twice an equal gain in risky choices. Gal and Rucker (2018) argue that much of the classic evidence reflects inertia or a preference for the status quo, and other studies find it weakens or disappears at small stakes, so it is better treated as conditional than universal.
Where it comes from
Daniel Kahneman and Amos Tversky, 1979 (Prospect theory, Econometrica). Amos Tversky and Daniel Kahneman, 1991 (Loss aversion in riskless choice)