What Is the Endowment Effect?
The endowment effect is a context-dependent change in valuation associated with possessing an object, holding an entitlement, or experiencing something as one's own. In its most familiar experimental form, owners demand more money to surrender a good than otherwise comparable nonowners are willing to pay to obtain it. Ownership can also reduce willingness to exchange one good for another, even when the initial allocation was random. The central scientific question is how the allocation of a good changes the process by which people construct, retrieve, and express its value. Kahneman, Knetsch, and Thaler, 1990; Morewedge and Giblin, 2015.
Imagine two people who are assigned identical ceramic mugs at different moments. The first person already owns a mug and specifies a minimum acceptable selling price. The second considers buying an identical mug and specifies a maximum buying price. A higher selling price may reflect how ownership changes attention, anticipated loss, expectations, or the meaning of the object. It may also arise from income constraints, question format, market beliefs, or misunderstanding of an experimental pricing rule. Identifying an endowment-related psychological mechanism therefore requires more than observing that one reported price exceeds another. Plott and Zeiler, 2005; Ericson and Fuster, 2014.
In psychological science the term describes an experimentally investigated valuation phenomenon, not a clinical symptom or a mental disorder. Its direction and size depend on what is owned, how possession is induced, whether trading is real, what price is requested, what alternatives exist, and what participants understand about the task. Experimental evidence supports ownership-sensitive valuation in many settings while also documenting strong boundary conditions and competing interpretations. Morewedge and Giblin, 2015; Tunçel and Hammitt, 2014.
The Three Main Ways Researchers Observe the Effect
Willingness to Accept and Willingness to Pay
Willingness to pay (WTP) is the greatest amount a person would pay to acquire a good under a specified transaction. Willingness to accept (WTA) is the least compensation the person would accept to relinquish a good or entitlement. Researchers compare their distributions under controlled conditions. The gap is often reported as WTA minus WTP or the ratio WTA/WTP. The ratio cannot be interpreted without knowing the valuation scale, the goods, the elicitation method, and whether the same individuals or comparable groups supplied the estimates. Horowitz and McConnell, 2002; Tunçel and Hammitt, 2014.
Suppose a selling group reports a median reservation price of $8 and a buying group reports $4. The observed median price ratio is 2 in that experiment. It is not a universal multiplier, a diagnostic threshold, or proof that possession caused every dollar of the difference. Such data warrant examination of random assignment, sampling variation, incentives, uncertainty about value, and whether the comparison holds under alternative elicitation rules. The difference between group medians must also be distinguished from the median of individual WTA/WTP ratios, which may be undefined when denominators are zero.
Reluctance to Exchange
In exchange experiments, researchers randomly assign one of two desirable objects and later allow a trade. Under a simple benchmark in which objects are equally attractive before assignment, an arbitrary endowment should not systematically determine which object is preferred. A reduced rate of exchange suggests that initial allocation matters. But the exact normative prediction depends on preferences, heterogeneity, information about alternatives, and actual costs of switching. It is not always mathematically correct to declare that exactly half of every sample must trade. Knetsch, 1989; Kahneman, Knetsch, and Thaler, 1990.
Mere Ownership and Preference
A person may evaluate an object more positively after it is labeled as theirs, even when no sale is contemplated. This response is often called a mere-ownership effect and is related to, but operationally distinct from, a monetary seller–buyer gap. Positive evaluation can occur through association with self, attention, perceived entitlement, familiarity, or changed memory. An increase in liking alone does not demonstrate a specific monetary valuation gap or establish that the person will refuse a favorable trade. Morewedge and Giblin, 2015.
Historical Development: From Consumer Choice to Experimental Markets
Early Behavioral-Economic Formulation
Richard Thaler's early behavioral-economics work described cases in which people value retaining what they possess differently from acquiring the same good, challenging a simple account in which ownership merely determines which person initially holds a fixed preference. Later experiments made this intuition operational by comparing prices and exchange behavior after goods were randomly allocated. Kahneman, Knetsch, and Thaler's influential studies used mugs, market procedures, and control markets involving induced-value tokens to ask whether ownership of consumption goods suppressed exchange even when markets offered participants experience. Kahneman, Knetsch, and Thaler, 1990; Kahneman, Knetsch, and Thaler, 1991.
In the induced-value condition, each token's redemption value is assigned by the experimenter. Because buyers and sellers are given explicit monetary incentives, trade predictions can be derived from the assigned values. The difference between token and mug markets matters: it suggests that the valuation of familiar consumption objects may involve processes beyond the mechanics of a trading institution. Nevertheless, the comparison alone does not identify whether the mechanism is loss aversion, attention, misunderstood instructions, or changing reference points.
Nonreversible Preferences and Experimental Exchange
Knetsch's work showed that arbitrary initial allocations could influence whether participants preferred to exchange an item, providing an experimental challenge to treating an indifference curve as independent of a starting endowment. That evidence helped establish exchange reluctance as a second measure alongside reservation prices. Repeated trading, alternative incentive procedures, and participants' market knowledge subsequently became important areas of study. Knetsch, 1989; List, 2003.
The Move Toward Competing Mechanisms
Research gradually expanded from a single loss-aversion explanation toward models of ownership, reference points, information search, focused attention, beliefs about market value, and task understanding. Plott and Zeiler demonstrated that extensive procedural controls could eliminate a mug valuation gap in their setting, while later work found persistent gaps in other settings and for lotteries. This history matters because the operational result and its causal explanation must be assessed separately. Plott and Zeiler, 2005; Isoni, Loomes, and Sugden, 2011.
The Normative Benchmark: When Should Ownership Matter?
Stable Preferences, Transactions, and the Economic Model
In a basic model of consumer choice, a person compares utility across bundles of goods and money. If a small everyday item has close substitutes, income effects are negligible, transaction costs are absent, quality is known, and preferences are stable, the maximum purchase price and minimum selling price are expected to be relatively close. A large, systematic difference in this carefully delimited setting motivates a psychological explanation. The expectation follows from assumptions about utility and constraints; it is not a free-standing law that WTA must always equal WTP. Ericson and Fuster, 2014.
For irreplaceable environmental goods, health changes, large wealth effects, missing substitutes, or irreversible choices, economic theory can permit important disparities even with internally consistent preferences. Asking someone what compensation they require to lose clean drinking water is mathematically and psychologically different from asking what they can afford to pay to obtain a small decorative item. A valid normative evaluation must specify the good, property rights, wealth constraints, credible substitutes, information and consequences of losing it. Horowitz and McConnell, 2002; Tunçel and Hammitt, 2014.
The Coase Theorem and Bargaining
The Coase theorem concerns efficient allocation under strongly specified assumptions about property rights, bargaining and transaction costs. Endowment experiments use that framework to test whether an initial allocation systematically affects trade. Ownership-linked reductions in volume challenge a particular frictionless trading prediction; they do not show that all real-world markets are inefficient or that legal entitlements have no welfare significance. Actual markets include search, liquidity, negotiating power, taxes, warranties, product knowledge, and social constraints. Kahneman, Knetsch, and Thaler, 1990.
Descriptive Models and Psychological Processes
A descriptive psychological model explains what participants actually attend to, anticipate, remember, and report. One model treats loss of a possessed object as a disfavored departure from a reference point. Another argues that sellers and buyers retrieve different attributes of the same object. A third emphasizes how instructions, price beliefs, and strategic incentives change reports. These mechanisms may produce similar observed gaps, so researchers distinguish them using interventions that independently manipulate ownership, reference points, consideration order, and information. Morewedge and Giblin, 2015.
Reference Dependence and Loss Aversion
Ownership as a Reference Point
Reference-dependent models evaluate outcomes relative to a psychologically active baseline. For a buyer, receiving a mug may be framed as a gain; for an owner, surrendering the mug may be framed as a loss. If losses receive greater weight than matched gains within the relevant dimension, the price required to give up the object can exceed the purchase price. The relation between the two prices is an implication of a specified model, and the model requires a definition of what counts as the reference point. Kahneman, Knetsch, and Thaler, 1991; Ericson and Fuster, 2014.
Ownership is one candidate reference point, but expectations can also set the benchmark. A person who expects to acquire a good may partly experience nonreceipt as a loss, while a current owner planning to sell may already incorporate the future sale into their baseline. Expectation-based reference points help explain why the strength of an endowment effect changes with anticipated exchanges and the perceived stability of possession. The observed gap should therefore be interpreted in relation to experimentally documented reference points instead of assuming that legal title alone determines psychological loss. Ericson and Fuster, 2014.
Endowment Effect and Loss Aversion Are Different Constructs
Loss aversion is a preference-related asymmetry in evaluation of losses and gains under a defined reference point. The endowment effect is an outcome pattern linking initial possession or ownership context to valuation or exchange. A WTA/WTP gap does not identify loss aversion uniquely, and loss aversion can appear in decisions that contain no owned consumer good. Similarly, ownership-related shifts in attention or identity can produce effects without requiring a universal loss-aversion coefficient. The distinction is essential in empirical interpretation. Morewedge et al., 2009; Morewedge and Giblin, 2015.
When Apparent Loss Aversion May Be Rational
The loss of an object can include consequences that are not present in an acquisition scenario: replacing it might take time, its quality is known while that of alternatives is uncertain, or it has acquired personal significance. If those factors genuinely change the object's usefulness, higher WTA can reflect changed preferences or omitted attributes rather than a judgment mistake. Recent theoretical work explicitly argues that some ownership-linked valuation changes are adaptive responses to new information and contextual attributes. Such an account is a theoretical interpretation whose scope must be tested against experimental evidence. Altman, 2026.
How Attention and Memory Construct Value
Buyers and Sellers Consider Different Foregone Benefits
When sellers contemplate giving up a good, they may concentrate on the uses, qualities, and pleasures relinquished. Buyers may concentrate on the money, other purchases, or opportunities they would sacrifice to obtain the same good. Carmon and Ariely proposed that the contrast in what is mentally forgone contributes to differing price estimates. The implication is that valuation depends partly on the decision question, not just on a hidden price stored permanently in memory. Carmon and Ariely, 2000.
Query Theory: The Order of Retrieved Arguments
Query theory proposes that a person's valuation is constructed through a sequence of memory searches. Sellers can begin by generating reasons to keep an object, and buyers by generating reasons not to buy it. Earlier queries influence which attributes are salient and can interfere with later retrieval. Johnson, Häubl, and Keinan experimentally investigated how the order of considerations contributes to the endowment effect. This approach offers testable predictions about the impact of reversing the usual retrieval sequence. Johnson, Häubl, and Keinan, 2007.
An experimental manipulation of consideration order has a different meaning from merely telling someone to be objective. It changes the proposed cognitive mediator: which attributes become accessible and when. Researchers can compare recalled arguments, response times, and valuation under assigned query orders. Evidence that retrieval order matters does not exclude reference dependence; the theories may describe different levels of explanation or interact in a given task.
The Attribute-Sampling Account
In an integrative review, Morewedge and Giblin argued that ownership and buying–selling frames affect sampling of value-relevant attributes. The psychological effect can then emerge because people consider different information, even when they assign similar weights to the information once considered. Such a framework links seller–buyer gaps, reluctance to exchange, and mere-ownership evaluations. It also explains why instructions that redirect attention may change the result. Its integrative status does not establish that the same mechanism explains every documented disparity. Morewedge and Giblin, 2015.
Ownership, Identity, and Psychological Possession
Legal Title, Physical Possession, and Psychological Ownership
Legal ownership, physical control, and the subjective sense that something is 'mine' can coincide, but they need not. A person can legally own an item without using or identifying with it; another can feel personally connected to a borrowed, customized, or shared object. Experiments that manipulate these dimensions separately help determine whether title, possession, expected use, and self-association contribute distinct effects. The endowment effect therefore belongs to psychological research on valuation, not to a legal definition of property. Morewedge and Giblin, 2015.
Mere Ownership Does Not Require a Long History
A brief assignment of an object can be sufficient to alter evaluation in some experimental settings. This result makes the effect theoretically important: a lengthy sentimental attachment is not a necessary condition. Yet the size and durability of the effect can be different for a randomly assigned mug, a family heirloom, a personally assembled product, a scarce collectible, or a right whose loss has irreversible consequences. A study demonstrating a short-term response to assignment cannot by itself determine how people value possessions after months or years. Kahneman, Knetsch, and Thaler, 1990; Ericson and Fuster, 2014.
Ownership Versus Anticipated Loss
Morewedge and colleagues experimentally challenged the claim that anticipated loss alone causes the endowment effect. Their findings emphasized the role of ownership-related valuation while varying whether a person faced acquisition or relinquishment. The result shows why researchers should measure more than reluctance to sell: changes in positive evaluation, attention, and perceived ownership can be relevant explanatory variables. Like other experiments, the finding concerns particular assigned roles and materials, and it does not settle the relative causal contribution of all proposed mechanisms. Morewedge et al., 2009.
What Classic Experiments Actually Demonstrate
Randomly Assigned Mugs
The best-known laboratory design randomly endows some participants with mugs while others may purchase them. Sellers and buyers state prices, sometimes through an incentive-compatible mechanism intended to make accurate valuation advantageous. With random assignment, initial group differences in stable tastes are reduced in expectation. Kahneman, Knetsch, and Thaler found substantial selling–buying differences and lower-than-predicted exchange for ordinary consumption goods, supporting the proposition that initial allocations affect observed behavior. Kahneman, Knetsch, and Thaler, 1990.
Crucially, randomly assigned mugs do not make all psychological processes random. Participants may interpret selling as relinquishing a gift, infer that the experimenter expects a certain market price, or react to the complexity of a pricing mechanism. A compelling test examines whether differences persist when such alternative explanations are measured or experimentally controlled.
Induced-Value Token Markets
Induced-value tokens are assigned redemption prices, making their monetary value explicit within the experimental institution. Classic work observed substantially more trading consistent with the theoretical benchmark in these markets than in consumer-good markets. This comparison helps distinguish a general inability to operate the trading mechanism from valuation processes connected to the good. It still leaves open the possibility that ownership of ordinary consumption goods creates uncertainty or beliefs not shared by tokens. Kahneman, Knetsch, and Thaler, 1990.
Exchange of Roughly Comparable Goods
When participants receive one attractive object and subsequently have an opportunity to exchange it for another, the initial assignment can predict which item they retain. A careful analysis compares preferences elicited before and after allocation, varies the order of presentation, measures transaction effort, and records whether participants understood that exchange was free. Evidence of low trading is most informative when the alternatives and decision conditions are clearly defined. Knetsch, 1989.
Measurement: Designing a Valid Test
Incentive Compatibility and the Becker–DeGroot–Marschak Procedure
Some studies use a random-price mechanism often called Becker–DeGroot–Marschak (BDM). Participants state a maximum buying price or minimum selling price, a transaction price is drawn independently, and the trade occurs according to a rule designed so truthful revelation is optimal under specified assumptions. The theoretical incentive properties depend on participants understanding the mechanism and on how the random draw and payment are implemented. An instruction labeled 'incentive compatible' does not establish that a participant understood it. Plott and Zeiler, 2005.
Training, practice rounds, actual payments, anonymity, neutral instructions, and comprehension checks can change results. Designs also differ in whether participants state open-ended prices, accept randomly drawn prices, select from lists, or respond to hypothetical questions. Each approach may elicit a somewhat different construct. Researchers should report the transaction rule, actual payment stakes, endowment conditions, exclusion criteria, and the exact statistical comparison used to evaluate a gap.
Between-Person and Within-Person Comparisons
Between-person experiments compare randomized groups of owners and nonowners. Within-person experiments can track a participant's valuations under multiple roles or time points, but carry order, demand, learning, and carryover risks. A participant cannot simultaneously occupy both ownership states under identical circumstances. Counterbalancing and temporal controls therefore matter when attributing within-person price changes to possession rather than task repetition.
What a Reported Gap Means Statistically
A positive mean or median gap is a descriptive result. A credible causal interpretation requires a design and an uncertainty estimate such as a confidence interval. Researchers may analyze the difference between group mean reservation prices, a robust difference between medians, the logarithm of a WTA/WTP ratio, or willingness to exchange. Each estimand answers a different question. Highly skewed WTA distributions can make means sensitive to extreme responses, while log-ratio statistics require positive and appropriately measured values. Tunçel and Hammitt, 2014; Wang, Yan, and Li, 2025.
Experimental Quality Checklist
A strong study states which component of ownership has been manipulated, how the control group is defined, whether participants can transact, which information is held constant, and the benchmark against which trading or valuations are evaluated. It tests comprehension, distinguishes pre-registered hypotheses from exploratory analyses, reports uncertainty and exclusions, and examines whether the same effect appears under alternative elicitation procedures. A robust effect in one procedure is meaningful evidence for that procedure; generalization requires replication across materially different settings.
Replication, Procedural Challenges, and Scientific Debate
Why the Plott and Zeiler Results Matter
Plott and Zeiler introduced multiple controls intended to reduce misconceptions about value elicitation, including extensive training and modified mechanisms. In their mug setting they did not observe the usual WTA/WTP gap, raising a direct question about interpreting the gap as an automatic consequence of loss aversion. The result is important evidence about procedural sensitivity and about the need to specify what participants know. It does not imply that all prior results were invalid or that ownership never changes value. Plott and Zeiler, 2005.
Why the Subsequent Comment Matters
Isoni, Loomes, and Sugden reported significant persistent WTA/WTP disparities for lotteries despite procedures associated with Plott and Zeiler's work, together with new evidence confirming the absence of a gap for mugs under those particular controls. This combination undermines both an unconditional 'gap always exists' claim and an unconditional 'training eliminates it' claim. The type of object, risk representation, and elicitation design matter. Isoni, Loomes, and Sugden, 2011.
Market Experience as a Moderator
List examined market participants and found that greater trading experience was associated with behavior nearer conventional exchange predictions. A later framed field experiment exogenously induced market experience and supported a causal role for experience in reducing the anomaly. This evidence makes it important to distinguish novices' valuation reports from decisions by people with sustained experience buying and selling similar goods. It does not establish that a single practice trial reliably removes every form of ownership-related valuation. List, 2003; List, 2011.
Why Replication Must Preserve the Target Question
An experiment can reproduce a positive seller–buyer gap while failing to reproduce the proposed underlying cognitive mediator. Conversely, a different pricing procedure can show a smaller gap because it changes what is being measured. Construct replication, procedural replication, and mechanism replication must be distinguished. Reviews summarize a heterogeneous literature and should be read as evaluations of families of tasks rather than as a vote on whether one universal effect 'exists.' Ericson and Fuster, 2014; Morewedge and Giblin, 2015.
What Meta-Analyses Show About Effect Size
The Type of Good Makes a Difference
Horowitz and McConnell reviewed WTA/WTP studies and found that ratios tended to be larger for nonmarket goods than for ordinary private commodities, with experiments involving monetary goods at the lower end. This variation is theoretically informative because the availability of close substitutes, the size of the entitlement, and perceived irreversibility differ dramatically across goods. A pooled number conceals these distinctions and should not be exported to every product category. Horowitz and McConnell, 2002.
A Later Meta-Analysis Identified Moderators
Tunçel and Hammitt revisited the disparity using meta-analytic comparisons of mean WTA/WTP ratios. They found smaller disparities for ordinary private goods and in studies with experienced participants, repeated valuations, and incentive-compatible mechanisms; more recent studies also tended to report smaller gaps. Their conclusions caution against treating a large early laboratory ratio as an immutable property of human psychology. They also show why method and sample composition belong inside any scientific summary of the phenomenon. Tunçel and Hammitt, 2014.
A Ratio Is Not the Same as an Individual Bias Score
Studies reporting a group WTA/WTP ratio do not establish the percentage of people who exhibit an individual endowment effect. A high group ratio may coexist with substantial overlap in individual valuations. Some participants may have no discernible ownership difference or may show the opposite pattern. Samples, confidence intervals, attrition, elicitation assumptions, and measurement reliability determine what prevalence claims are justified. Cross-study ratios are especially difficult to compare when participants value different kinds of outcomes.
Cross-Cultural and Environmental Boundary Conditions
Cultural Self-Construal and Social Meaning
Maddux and colleagues compared participants from different cultural contexts and reported smaller ownership-associated valuation gaps among East Asian than Western participants in their tested conditions, interpreting the pattern partly through cultural differences in self-related processes. Such results document context-dependent differences among the sampled populations. They do not support treating all members of a region as psychologically uniform, and they cannot establish the mechanism without further cross-cultural controls. Maddux et al., 2010.
Hunter-Gatherer Field Experiments
Apicella and colleagues studied Hadza populations living in different degrees of contact with modern market environments. Participants in more isolated regions did not display the tested endowment effect, whereas more market-exposed participants did. These observations complicate the claim that the measured exchange phenomenon must be universal or invariant across environments. The study also makes clear that human decision-making develops within specific material institutions and patterns of exchange. Apicella et al., 2014.
Practical Experience and Local Institutions
Market exposure can change knowledge of typical prices, comfort with bargaining, expectations about a sale, and the social interpretation of giving something away. Consequently, a national or cultural comparison needs to consider product familiarity, income constraints, the exchange institution, language of instructions, and sampling methods. Cultural labels alone rarely identify the causal pathway behind a valuation gap. List, 2003; Apicella et al., 2014.
When the Effect Weakens, Disappears, or Reverses
Goods Held for Resale
Ownership of an item for personal use differs from holding inventory that is expected to be sold. Traders, retailers, and experienced collectors may treat an object as exchange stock, making its sale congruent with their reference point. This distinction helps explain reduced gaps among practiced market participants. A participant who is assigned an object as a temporary trading instrument may also interpret possession differently from someone told the object is a personal gift. List, 2011; Ericson and Fuster, 2014.
Borrowing and Sharing May Produce Different Price Patterns
In rental and sharing arrangements, the transaction is temporary access rather than permanent transfer. Research on peer-to-peer sharing finds circumstances in which owners' minimum accepted rental prices can be lower than nonowners' willingness to pay for corresponding access. That reversed ordering underscores why an observed rental price difference should not be equated with the classic seller–buyer endowment effect for permanent relinquishment. Ross, Kim, and Meloy, 2026.
Negative Ownership Experiences
An object can impose storage costs, maintenance burdens, unpleasant reminders, or risk. In such cases its ownership may reduce rather than increase a person's evaluation. The relevant economic comparison includes these consequences. An experiment in which an owner is eager to dispose of a disliked object cannot be dismissed as an exception to a presumed universal preference for keeping possessions; it helps identify the boundary of the construct.
Time, Familiarity, and Expectations
The effect may change as people learn the item's quality, inspect substitutes, use the object, or anticipate a future sale. Time can increase attachment through use, but it can also reduce novelty and uncertainty. The direction is an empirical question. Research designs should measure ownership duration and expectation rather than assigning every change to mere passage of time. Ericson and Fuster, 2014.
Endowment Effect Versus Related Concepts
Status Quo Bias
Status quo bias describes favoring an existing state or default relative to alternatives. It can concern health plans, institutional rules, software settings, or other arrangements without individually owned objects. The endowment effect addresses how ownership context changes valuation or exchange. The two can coincide when keeping an endowed good is also maintaining the current state, but a change in default alone does not prove ownership-based valuation. Kahneman, Knetsch, and Thaler, 1991.
Sunk Cost Fallacy
A sunk cost is an expenditure already incurred that cannot be recovered. The sunk cost fallacy refers to allowing such past costs to distort choices that should depend on future costs and benefits. The endowment effect may occur after a free random allocation and therefore requires no investment history. A paid-for possession can involve both mechanisms, but they require different experimental contrasts.
Framing Effect
Framing refers to changes in choice induced by different presentations of relevant information. The buyer–seller role is a frame that may alter what participants notice, yet the canonical endowment question asks how possessing an object affects valuation. A framing experiment might manipulate wording while holding possession constant; an endowment experiment can manipulate possession while holding descriptions constant. Morewedge and Giblin, 2015.
Mere Exposure, Self-Association, and Attachment
Familiarity or personal association can increase liking, but liking and reservation price are distinct dependent variables. Emotional attachment can be a genuine component of utility, especially for irreplaceable possessions. Psychological ownership concerns perceived connection to an object; its contribution to WTA must be demonstrated with appropriate measurement. A mere-exposure manipulation should not be labeled an endowment effect unless the study addresses ownership-related valuation.
Statistical Error, Heuristic, and Clinical Symptom
The endowment effect is a behavioral regularity estimated from experimental comparisons, not an arithmetic fallacy in every individual response. Some decisions can be rational under the person's real preferences and constraints. It is not a diagnosis, a marker of impaired cognition, or evidence of compulsive hoarding. Psychological evaluation must avoid turning an experimentally defined valuation tendency into a clinical category.
Real-World Implications for Markets and Consumer Behavior
Selling Personal Belongings
People listing used goods can anchor on purchase history, memories of use, replacement concerns, or perceived ownership value. A price above current market offers may reflect attachment, incomplete information, or an informed reservation value. The practical question is whether the seller is optimizing for revenue, speed, sentimental retention, or replacement utility. Research on endowment effects suggests investigating these components separately rather than assuming a single universal overpricing mechanism. Morewedge and Giblin, 2015.
Returns, Trials, and Product Adoption
Giving a consumer temporary possession can change how a product is evaluated, but its effect depends on whether the person expects to keep it, return it, or compare it with alternatives. Product-trial policies also involve learning, return effort, budget, and consumer protection, so their commercial outcomes cannot be inferred from the endowment effect alone. An ethically sound trial communicates ownership conditions and final price clearly.
Negotiation and Property Rights
Differences in buyer and seller reservation prices can be genuine disagreements about compensation, entitlement, or the risks of giving something up. Negotiators may gain clarity by separating objective replacement costs, private-use value, and bargaining strategy. Compensation demanded for relinquishing a legal right is especially sensitive to whether alternatives exist and whether the right can be restored. Ericson and Fuster, 2014.
Health and Environmental Valuation
In contingent valuation, respondents may state different amounts to avoid losing an existing benefit and to obtain a comparable benefit they lack. These questions can involve different property rights, irreversibility, wealth constraints, uncertainty, or emotional meaning. A 2025 systematic review of healthcare WTA/WTP studies emphasizes the influence of question format and the challenge posed by extreme observations. Such discrepancies cannot all be allocated mechanically to an ownership bias. Wang, Yan, and Li, 2025; Tunçel and Hammitt, 2014.
Ownership and Valuation in Digital Environments
Digital Goods and Perceived Control
Digital purchases, subscriptions, virtual goods, saved workspaces, and personalized settings differ in transferability and legal rights. A user may experience control over a file or personalized interface without holding an unrestricted property right. Research-inspired hypotheses about digital endowment effects should operationalize what users believe they own, whether they can transfer or retain it, and what loss or switching would entail. The established literature on physical goods provides concepts and experimental designs; direct effect estimates should come from the specific digital setting. Morewedge and Giblin, 2015.
Human–AI Decision Support
AI recommendations may affect perceived market price, prompt users to compare substitutes, or supply arguments for keeping and selling an object. A useful study would randomize the recommendation format, ownership status, and timing, then assess transaction choices and calibrated estimates. Any observed benefit could involve information quality, attention, or trust; it would not independently establish a general law of AI-mediated endowment. Within the Artificial Era, these systems create new settings for human decision research while the causal claims remain anchored in experimental psychology and behavioral economics.
UX Design and Ethical Choice Architecture
Interfaces can make abandoning saved work, switching services, or removing an item from a cart feel like relinquishing something already possessed. Design choices should be evaluated for transparency, informed consent, and reversibility. Legitimate personalization increases utility; exploitative prompts can obscure the terms of ownership or invent scarcity. Measuring whether an interface changes valuation requires comparing the actual options and costs, not merely the presence of possessive language.
Can People Reduce an Endowment-Related Valuation Gap?
Separate Personal-Use Value From Market Value
A useful decision process begins by asking two different questions: what would a comparable replacement cost in the current market, and what would this particular item contribute to the owner's future welfare? Both values can matter, but treating them as identical produces confusion in negotiation. For a personal item, sentimental value may justify retaining it; for an item bought only for resale, market offers may provide the more relevant benchmark.
Make Counterfactual Choices Symmetric
A buyer can imagine already owning the item and being offered cash; a seller can imagine having cash and deciding whether to buy that same item today. Comparing both formulations can reveal differences in which features receive attention. It does not guarantee a universal debiasing effect, because imagined ownership differs from actual possession and may introduce new assumptions. Query-theory experiments support the narrower proposition that consideration order can causally affect valuations. Johnson, Häubl, and Keinan, 2007.
Explicitly Search for Arguments on Both Sides
A structured prompt can require a participant to generate reasons for keeping, relinquishing, acquiring, and declining an object before stating a price. This attempts to correct information-sampling asymmetry rather than telling people to suppress feelings. Transfer to consequential decisions depends on the context, quality of information, and follow-through. Laboratory support for attribute sampling should not be presented as proof that a simple checklist will eliminate real-world bargaining differences. Carmon and Ariely, 2000; Morewedge and Giblin, 2015.
Learn the Transaction Rule and Obtain Price Feedback
Practice can help participants understand a pricing task, and market experience can reduce some observed disparities. In actual markets, checking realized sale prices, documenting comparable quality, and distinguishing asking prices from completed transactions may improve valuation calibration. Studies show that procedural comprehension and market experience matter, but effects on one task do not imply generalized immunity to judgment biases. Plott and Zeiler, 2005; List, 2011.
Make Improvements Measurable
A valid intervention study predefines its criterion: narrower WTA/WTP gaps when economic assumptions warrant convergence, better prediction of actual resale prices, reduced avoidable exchange inefficiency, or greater consistency with stable stated goals. A reduction in the gap is not automatically an improvement if ownership genuinely adds personal value. Researchers should test durability, effects outside training tasks, and unintended consequences rather than counting every move toward a chosen price as better judgment.
Frequently Asked Questions
Why do I think something I own is worth more than the same thing in a store?
Possession can change the reference point, the attributes you remember, the value of control, and the experience you anticipate losing. Your personal-use value may also differ legitimately from the current market price. The experimentally established pattern depends on the object and valuation procedure; there is no fixed markup that ownership creates in every person. Morewedge and Giblin, 2015.
Is the endowment effect the same as loss aversion?
No. Endowment effect denotes a pattern of ownership-associated evaluation or exchange. Loss aversion is one theoretical mechanism concerning the weighting of losses and gains relative to a reference point. Ownership, memory search, price beliefs, and experimental procedure provide additional accounts. Morewedge et al., 2009.
Does the effect occur immediately after someone receives an object?
It can. Random-allocation experiments have observed endowment-related differences shortly after assignment, although the details of possession, instructions and decision environment determine whether a particular study obtains the effect. Kahneman, Knetsch, and Thaler, 1990.
Does a WTA greater than WTP always show irrationality?
No. Economic models can predict disparities when income effects, scarce substitutes, irreversible outcomes, or genuine changes in utility are present. Even for inexpensive goods, a gap alone does not isolate a psychological error. The normative claim requires identifying the assumptions and experimental controls. Ericson and Fuster, 2014.
What is the difference between the endowment effect and overpricing?
Overpricing is a judgment relative to a specified market or valuation benchmark. An endowment effect is a causal or statistical comparison associated with ownership status. A seller can rationally decline a market offer when retaining the item has sufficient private value; a high asking price alone is insufficient evidence of a bias.
Can repeated buying and selling reduce the effect?
Evidence from market experience research indicates that it can reduce ownership-related exchange disparities under particular conditions. A replicated reduction in one market should not be assumed for every object, institution, or psychologically important entitlement. List, 2003; List, 2011.
Is the endowment effect universal across cultures?
Research documents cultural and environmental variation, including cross-cultural laboratory comparisons and field experiments among Hadza participants. The effect's prevalence and magnitude must be estimated in the populations and tasks under consideration. Maddux et al., 2010; Apicella et al., 2014.
Does the endowment effect apply to housing, jobs, or digital accounts?
Ownership-related reference points and attention may be relevant in these domains, but housing, employment, and digital services involve legal rights, substantial switching costs, personal history, and uncertainty. Applying a mug experiment's numerical effect size to them would be unsupported. Each domain requires its own measurement design and evidence.
How can an experiment distinguish the effect from a misunderstanding of prices?
Researchers can use real transactions, training, comprehension checks, alternative elicitation mechanisms, and randomized assignment. Comparing results across procedures reveals whether a particular disparity is robust to changes in understanding and incentives. Plott and Zeiler, 2005; Isoni, Loomes, and Sugden, 2011.
Can the effect reverse in sharing markets?
Yes, the usual ordering of owner and nonowner valuations is not guaranteed for every transaction. Research on renting owned possessions through sharing platforms documents settings with lower owner WTA than renter WTP. Permanent sale and temporary access are different economic decisions. Ross, Kim, and Meloy, 2026.
Scientific Synthesis
The endowment effect provides a precise research question: how does an initial allocation, perceived possession, or ownership context alter the valuation of a good or the willingness to exchange it? The strongest empirical answer emerges from designs that separate ownership from elicitation artifacts, reference points from actual utility changes, and subjective value from external market prices. Classic experiments established substantial ownership-associated valuation patterns; later studies demonstrated that procedures, experience, type of good, culture, and expectations can strongly moderate the result. Kahneman, Knetsch, and Thaler, 1990; Tunçel and Hammitt, 2014; Ericson and Fuster, 2014.
Contemporary psychological explanation integrates reference-dependent preference, ownership and identity, information sampling, attention, memory retrieval, procedural comprehension, and market learning. The presence of an observed WTA/WTP gap identifies a pattern that calls for explanation; decisive tests of its cause require interventions and explicit normative assumptions. This approach makes the endowment effect useful for consumer psychology, behavioral economics, public valuation, law, and human–technology decision research without converting a conditional finding into a universal rule. Morewedge and Giblin, 2015.
Related Articles
Decision-Making Under Risk and Uncertainty: Models and Examples — the parent scientific framework for choices with risk, uncertainty, value, and competing utility models.
Prospect Theory: How Gains, Losses, and Reference Points Shape Choice — reference dependence and evaluation relative to gains and losses.
Loss Aversion: Definition, Evidence, and Scientific Debate — the conceptual mechanism often used to explain seller–buyer differences.
Framing Effect: How the Wording of Choices Changes Decisions — how descriptions and roles influence the evaluation of equivalent options.
Cognitive Debiasing Strategies: Which Interventions Work? — evidence and limitations of attempts to improve judgment.
AI as Authority: Trust, Expertise, Automation Bias, and Human Decision-Making — human reliance on technological recommendations and decision support.
References
Altman, M. (2026). Why the endowment effect reflects unbiased smart decision-making and can increase the willingness to trade. Review of Behavioral Economics. https://doi.org/10.1108/RBE-02-2025-1007.
Apicella, C. L., Azevedo, E. M., Christakis, N. A., and Fowler, J. H. (2014). Evolutionary origins of the endowment effect: Evidence from hunter-gatherers. American Economic Review, 104(6), 1793–1805. https://doi.org/10.1257/aer.104.6.1793.
Carmon, Z., and Ariely, D. (2000). Focusing on the forgone: How value can appear so different to buyers and sellers. Journal of Consumer Research, 27(3), 360–370. https://doi.org/10.1086/317590.
Ericson, K. M. M., and Fuster, A. (2014). The endowment effect. Annual Review of Economics, 6, 555–579. https://doi.org/10.1146/annurev-economics-080213-041320.
Horowitz, J. K., and McConnell, K. E. (2002). A review of WTA/WTP studies. Journal of Environmental Economics and Management, 44(3), 426–447. https://doi.org/10.1006/jeem.2001.1215.
Isoni, A., Loomes, G., and Sugden, R. (2011). The willingness to pay—willingness to accept gap, the 'endowment effect,' subject misconceptions, and experimental procedures for eliciting valuations: Comment. American Economic Review, 101(2), 991–1011. https://doi.org/10.1257/aer.101.2.991.
Johnson, E. J., Häubl, G., and Keinan, A. (2007). Aspects of endowment: A query theory of value construction. Journal of Experimental Psychology: Learning, Memory, and Cognition, 33(3), 461–474. https://doi.org/10.1037/0278-7393.33.3.461.
Kahneman, D., Knetsch, J. L., and Thaler, R. H. (1990). Experimental tests of the endowment effect and the Coase theorem. Journal of Political Economy, 98(6), 1325–1348. https://doi.org/10.1086/261737.
Kahneman, D., Knetsch, J. L., and Thaler, R. H. (1991). Anomalies: The endowment effect, loss aversion, and status quo bias. Journal of Economic Perspectives, 5(1), 193–206. https://doi.org/10.1257/jep.5.1.193.
Knetsch, J. L. (1989). The endowment effect and evidence of nonreversible indifference curves. American Economic Review, 79(5), 1277–1284. https://www.jstor.org/stable/1831454.
List, J. A. (2003). Does market experience eliminate market anomalies? Quarterly Journal of Economics, 118(1), 41–71. https://doi.org/10.1162/00335530360535144.
List, J. A. (2011). Does market experience eliminate market anomalies? The case of exogenous market experience. American Economic Review, 101(3), 313–317. https://doi.org/10.1257/aer.101.3.313.
Maddux, W. W., Yang, H., Falk, C., Adam, H., Adair, W., Endo, Y., Carmon, Z., and Heine, S. J. (2010). For whom is parting with possessions more painful? Cultural differences in the endowment effect. Psychological Science, 21(12), 1910–1917. https://doi.org/10.1177/0956797610388818.
Morewedge, C. K., and Giblin, C. E. (2015). Explanations of the endowment effect: An integrative review. Trends in Cognitive Sciences, 19(6), 339–348. https://doi.org/10.1016/j.tics.2015.04.004.
Morewedge, C. K., Shu, L. L., Gilbert, D. T., and Wilson, T. D. (2009). Bad riddance or good rubbish? Ownership and not loss aversion causes the endowment effect. Journal of Experimental Social Psychology, 45(4), 947–951. https://doi.org/10.1016/j.jesp.2009.05.014.
Plott, C. R., and Zeiler, K. (2005). The willingness to pay–willingness to accept gap, the 'endowment effect,' subject misconceptions, and experimental procedures for eliciting valuations. American Economic Review, 95(3), 530–545. https://doi.org/10.1257/0002828054201387.
Ross, G. R., Kim, E., and Meloy, M. G. (2026). Owners' willingness to accept in the sharing economy. Journal of Marketing, 90(2), 177–192. https://doi.org/10.1177/00222429251332889.
Tunçel, T., and Hammitt, J. K. (2014). A new meta-analysis on the WTP/WTA disparity. Journal of Environmental Economics and Management, 68(1), 175–187. https://doi.org/10.1016/j.jeem.2014.06.001.
Wang, Y., Yan, X., and Li, H. (2025). A systematic review of the willingness-to-accept and willingness-to-pay disparities in empirical studies in the healthcare field. Archives of Public Health, 83, Article 212. https://doi.org/10.1186/s13690-025-01699-w.