Loss Aversion

Often, roughly, on average, but not as a law. The figure of 2.25 came from 25 students in 1992. A 2024 meta-analysis of 607 estimates put the average near two, while tests limited to risky choices found 1.31 and a re-analysis found almost no asymmetry when gains and losses were offered on equal terms.

What the report found

1. The asymmetry is real and widely found. People turn down fair bets, owners ask far more to give up a mug than buyers will pay for it, and the classic prospect theory problems replicated in 19 countries. Home sellers, investors and teachers all behave as the idea predicts.
2. The famous number is narrower than it sounds. The often-quoted coefficient of 2.25 was fitted to the choices of 25 graduate students in 1992. Kahneman himself later gave the ratio as a range of 1.5 to 2.5.
3. Its size depends on how it is measured. A meta-analysis of 607 estimates put the average near two. Another, limited to risky choices between gambles, put it at 1.31, and a re-analysis found almost no asymmetry when gains and losses were offered on equal terms.
4. Researchers disagree about how general it is. One side argues that inertia, not loss aversion, explains many of the classic results. The other finds loss aversion across five samples of more than 17,000 people, smaller with experience and larger with age.
5. Use it with care. Framing a cost as a loss changed behaviour in field studies of shopping bags and teacher pay. In health messages, loss framing has not beaten gain framing, and people asked to put their own money at risk mostly declined.

What is inside

1. The idea and where it came from
2. The original studies, explained
3. What happened when others tested it
4. Where researchers disagree
5. What we can say with confidence

Plus a timeline of the idea, a summary of the original studies, a table of field tests and a graded list of claims.

Who it is for

For anyone who has heard that losses hurt twice as much, and anyone who designs with that idea.

How we work

Evidence Reviews follow one idea from its first studies to the latest reviews. Every claim is graded Established, Emerging or Untested. Where researchers disagree, as they do on how general loss aversion is and how large, both sides are set out before we say where we stand.

Cite this report

The Why We Project (2026) Loss Aversion: Do losses really hurt twice as much? Evidence Review No. 02. Dubai: Why We Tech Ltd. Available at: thewhyweproject.com