The Endowment Effect
Four items. Each priced twice. Once you own it. Once you do not. Same item, same person, same fair market price as the reference. The slider runs 0-100 where 50 is fair market. WIZ measures the gap that 40 years of experimental data say will appear the moment ownership is assigned.
“The current state has a privileged status. Moves away from it are evaluated as gains or losses. Losses loom roughly twice as large as equivalent gains.”Tversky & Kahneman (1991) QJE vol 106
Kahneman Knetsch & Thaler (1990) JPE vol 98 handed half a Cornell classroom a university-logo coffee mug. The owner half quoted a median selling price of $7.12. The non-owner half quoted a median buying price of $3.12. A 2.28-to-1 ratio on objects neither group had owned for more than ninety seconds, assigned by row number. The Coase theorem predicts equal-allocation trade regardless of initial assignment. Almost no trades occurred.
Knetsch & Sinden (1984) QJE vol 99 ran the first clean empirical demonstration on raffle tickets with coin-flip assignment: owners demanded $5.18 to sell, buyers offered $1.28 to buy, a 4-to-1 ratio. Carmon & Ariely (2000) JCR vol 27 took the paradigm to Duke Final Four basketball tickets: owners demanded $2400 median, buyers offered $170, a 14-to-1 ratio. Strahilevitz & Loewenstein (1998) JCR vol 25 showed the gap grows with ownership duration.
Mechanism. Tversky & Kahneman (1991) QJE vol 106 formalized the reference-dependent loss-aversion explanation: the current state is the reference, losses are coded twice as painful as equal-magnitude gains, selling is a loss, buying is a foregone gain. The same dollar amount on the two sides of the transaction is psychologically asymmetric. Beggan (1992) JPSP vol 62 mere-ownership effect and Belk (1988) JCR vol 15 extended-self compound the asymmetry.
Meta-analytic status. Horowitz & McConnell (2002) JEEM vol 44 pooled 45 studies and reported a median WTA/WTP ratio of 2.6 on ordinary private goods and above 5 on public goods. Sayman & Öncüler (2005) Decision Analysis vol 2 replicated across 154 study-comparisons. The effect is one of the most robustly replicated findings in behavioral economics.
The exercise is not a test of greed or attachment. The exercise measures how much your valuation of the same object depends on which side of the trade you happen to be on. The literature answer is: usually a lot more than the textbook says, and a lot more than you would predict for yourself before the role assignment.