Loss aversion, documented by Kahneman and Tversky as a core component of Prospect Theory, is the empirical finding that the disutility of losing an amount x is roughly twice the utility of gaining the same amount. With a loss aversion coefficient λ ≈ 2.25, a person requires a potential gain of about $225 to accept a 50-50 gamble of losing $100.
Evidence and Magnitude
Accept 50-50 gamble iff: gain > λ × loss
Endowment effect: WTA/WTP ratio ≈ 2:1
Loss aversion manifests across domains: the endowment effect (people value objects they own about twice as much as identical objects they don't own), the status quo bias (preferring the current state because changes involve potential losses), and the disposition effect in finance (holding losing stocks too long and selling winners too soon).
Neural Basis and Debates
Neuroimaging studies suggest that losses activate the amygdala and anterior insula more strongly than equivalent gains activate reward regions. However, the universality of loss aversion has been debated: some studies find reduced or absent loss aversion for small stakes, experienced traders, or decisions involving goods rather than money. Recent meta-analyses suggest that λ is reliably above 1 but may be closer to 1.5–2.0 than the original estimate of 2.25.