The endowment effect is the tendency for people to value a good more highly when they own it than when they do not. In behavioral economics, it commonly appears as a gap between the minimum compensation an owner demands to surrender an item and the maximum amount a non-owner will pay to acquire it. Experiments also examine whether randomly assigning ownership makes people more likely to retain an item rather than exchange it for an alternative. The effect concerns the influence of possession or perceived ownership on valuation; it does not mean that every difference between buying and selling prices is a behavioral bias.
Definition and measurement
Two monetary measures are central to the study of the endowment effect:
- Willingness to pay (WTP): the maximum amount someone will pay to acquire a good.
- Willingness to accept (WTA): the minimum compensation someone requires to give up a good they possess.
For example, a participant randomly given a mug might refuse to sell it for less than 8 monetary units, while a participant without a mug might offer no more than 4 units to buy the same item. If comparable participants are randomly assigned to these roles, a systematic difference provides evidence that ownership or the decision context affects valuation. These figures are illustrative, not a general estimate of the effect.
Researchers also use exchange experiments, which do not require participants to state prices. Participants receive one of two goods, such as a mug or a chocolate bar, and can exchange it for the other. If randomly assigned owners disproportionately retain whichever good they receive, relative to an appropriate comparison group, the result is consistent with an endowment effect.
Neither measure is conclusive in isolation. A buying–selling price gap can reflect differences in wealth, incentives, information, or the interpretation of the task. Reluctance to exchange can reflect transaction costs, uncertainty about the alternative, or a preference not to act. Experimental design therefore matters to both the size and interpretation of the observed effect.
Intellectual history
Richard Thaler introduced the term “endowment effect” in a 1980 article on consumer choice. It described situations in which giving up something already held seemed more costly than forgoing an equivalent acquisition.
The concept became closely associated with prospect theory, developed by Daniel Kahneman and Amos Tversky in 1979. Prospect theory evaluates outcomes as gains and losses relative to a reference point, rather than solely by final wealth. Its principle of loss aversion—that losses can weigh more heavily than comparable gains—provided an influential explanation for why owners might demand more to relinquish a good than non-owners would pay to obtain it.
Jack Knetsch’s exchange experiments and the experimental work of Kahneman, Knetsch, and Thaler helped establish the phenomenon as a central topic in behavioral economics. Their 1990 study used markets for goods such as mugs to investigate whether valuation differences persisted when participants had opportunities to trade.
Later research examined how trading experience, instructions, elicitation procedures, expectations, and perceived ownership alter the effect. This work made the endowment effect both an influential behavioral finding and a subject of debate about what experimental results establish.
Main explanations
Reference dependence and loss aversion
The best-known explanation treats ownership as establishing a reference point. For an owner, selling an item entails losing it; for a buyer, obtaining it entails a gain. If losing the item has a greater psychological impact than acquiring it, the owner’s minimum selling price can exceed the buyer’s maximum purchase price.
A schematic representation of reference-dependent valuation is:
where is an outcome, is the reference point, and assigns value to gains and losses. A common illustrative value function is:
The parameter represents greater sensitivity to losses. This is a simplified illustration, not a complete model of a transaction: buying and selling involve changes in both money and goods, and predictions depend on how those changes are evaluated.
Reference points need not coincide with legal possession. Expectations about future ownership, customary entitlements, or anticipated consumption can also influence whether an outcome is experienced as a gain or a loss.
Psychological ownership and attachment
Possession can change how people perceive an object. Handling, using, customizing, or associating an item with personal identity may strengthen attachment or make its desirable features more salient. These mechanisms can increase valuation without being identical to loss aversion.
Psychological ownership is the feeling that something is “mine,” which can differ from formal property rights. A person can feel attached to a borrowed or customized object without legally owning it, while a trader may legally own inventory without feeling personally attached to it.
Long-standing attachment also differs from the narrow experimental claim that ownership alone changes valuation. A used personal object may provide memories or other benefits that a visually identical replacement cannot supply.
Attention, reasoning, and transaction framing
Buyers and sellers may focus on different considerations. Sellers can concentrate on reasons to keep an item, while buyers concentrate on the expenditure required to acquire it. The sequence in which people consider these arguments may affect their resulting valuations.
Trading roles can also convey unintended information. A participant may interpret a gift as a recommendation or suspect that an offered exchange benefits the experimenter. These interpretations complicate attempts to attribute all reluctance to trade to ownership itself.
Relationship to standard economic theory
The endowment effect is often contrasted with models in which preferences over final consumption bundles remain stable regardless of initial possession. Under those models, ownership can change what a person can afford without necessarily changing how the person ranks the goods themselves.
Nevertheless, standard theory does not require WTA and WTP to be equal in every setting. An income effect can create a difference because owning a valuable item changes a person’s economic position. The availability of substitutes also matters: compensation demanded for losing an irreplaceable amenity may greatly exceed willingness to pay to acquire it, even without loss aversion.
In a simplified setting with stable preferences and utility that is linear in money, acquiring and relinquishing the same good can imply equal monetary valuations. More general preferences need not produce that equality. Consequently, a WTA–WTP gap is evidence about valuation, but it is not by itself proof of an ownership-induced preference change.
The endowment effect is also related to, but distinct from:
- Status quo bias: favoring the existing state of affairs, whether or not ownership is involved.
- Sunk-cost effects: allowing irrecoverable past expenditures to influence current decisions.
- Opportunity-cost neglect: failing to account fully for alternatives forgone by keeping or acquiring something.
- Sentimental value: valuing an object because of personal history or associations.
These mechanisms can coexist in the same decision.
Experimental evidence and methodological debate
Random assignment and real consequences
Random assignment helps distinguish the effect of ownership from pre-existing preferences. If participants choose which goods to own, owners may simply be people who already value those goods most highly.
Experiments with actual payments and exchanges reduce some problems associated with hypothetical questions. Researchers also use procedures designed to make truthful reporting advantageous, drawing on incentive compatibility. However, participants must understand those procedures. A formally well-designed mechanism may still produce misleading measurements if its incentives are misunderstood.
Useful experimental controls include checking comprehension, standardizing information about the goods, avoiding negotiation when measuring reservation prices, and separating reluctance to exchange from inconvenience.
Experience and procedures
Some field and laboratory research has found that market experience attenuates endowment-like behavior. John List’s work on traders is an influential example. Such findings suggest that familiarity with exchange and viewing goods as inventory rather than personal possessions can affect observed behavior.
Charles Plott and Kathryn Zeiler showed that, under their experimental procedures, a WTA–WTP gap could disappear after extensive controls for possible misconceptions. Related work examined how features of exchange experiments affect reluctance to trade.
These results challenge the interpretation of particular findings as direct measures of loss aversion. They do not establish that ownership never affects valuation. Conversely, finding a valuation gap under one protocol does not establish a universal psychological constant.
Variation across settings
The size and persistence of the effect depend on the good, the population, the assignment procedure, and how preferences are measured. Goods intended for personal use may be evaluated differently from goods acquired for resale. Expectations and the credibility of the ownership assignment can also matter.
There is therefore no single endowment-effect multiplier that reliably converts a buyer’s valuation into a seller’s valuation across settings.
Economic applications and implications
Trade and initial allocation
If ownership changes reservation values, mutually acceptable trade may be less frequent than a model with ownership-independent preferences predicts. Initial allocation can then influence final allocation even when participants have opportunities to exchange.
This has implications for the Coase theorem. In its familiar formulation, clearly defined rights and sufficiently low transaction costs allow bargaining to achieve an efficient allocation under specified assumptions. If assigning rights changes valuations, allocation may depend on more than bargaining costs or purchasing power. This qualifies applications of the theorem that assume stable, ownership-independent valuations; it does not mean that every failure to trade is inefficient.
Consumer markets
The concept is used to analyze product trials, customization, return arrangements, and resale decisions. Such settings may create perceived ownership before a purchase becomes final or may strengthen attachment afterward.
These practices also affect information, convenience, risk, and switching costs. Their commercial effects therefore cannot automatically be attributed to the endowment effect alone.
Environmental valuation and public policy
In cost–benefit analysis, the distinction between paying for an improvement and accepting compensation for a loss can substantially affect measured benefits or damages. The appropriate question depends partly on the rights and baseline conditions being evaluated.
For environmental goods without close substitutes, large WTA–WTP differences may arise for several reasons, including standard preference effects, perceived entitlements, and reference dependence. Treating every such difference as an endowment bias would obscure these distinctions.
Welfare interpretation
The endowment effect creates a problem for welfare economics: if valuation changes with ownership, there may be no single ownership-independent figure that unambiguously represents a person’s welfare.
An owner’s refusal to sell cannot simply be dismissed as mistaken, nor must a non-owner’s lower willingness to pay automatically be treated as the correct valuation. Welfare assessments depend on the explanation for the difference, the relevant rights and expectations, and the normative framework used.