Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Reference-Dependent Valuation, Ownership, and Endowment Effects

Abstract. This fundamental research monograph examines reference-dependent valuation and the endowment effect as a family of experimentally observable relations between ownership, expectations, transaction roles, and the prices or exchanges individuals accept. It reconstructs the conceptual and historical research tradition, evaluates classical exchange markets and willingness-to-accept/willingness-to-pay studies, contrasts loss-sensitive preferences with expectation-based, psychological-ownership, attentional, memory-query, and reference-price explanations, and scrutinizes procedural validity and evidence from market experience. Original economic and psychological studies, methodological disputes, empirical investigations published through 2025, a 2026 theoretical interpretation, comparative evidence, neural correlates, and an integrative scientific review provide the evidentiary basis. The synthesis establishes that entitlement and reference contexts can change valuation, while the magnitude and mechanisms depend on experimental design, good type, reference-point formation, and institutional experience. It identifies discriminating causal experiments and durable outcome measures as major priorities for research in human and technologically mediated transactions.

Keywords: reference-dependent valuation; endowment effect; willingness to accept; willingness to pay; ownership; psychological ownership; exchange reluctance; loss aversion; reference points; expectations; query theory; market experience; behavioral economics; consumer psychology; transaction utility; Artificial Era

The Scientific Problem of Valuation and Endowment

Reference-dependent valuation describes changes in the subjective assessment of an outcome when the comparison standard changes. A good can be evaluated as something to acquire, retain, exchange, or relinquish; these roles alter the psychologically relevant counterfactual. The research problem is to explain how an otherwise comparable commodity can receive different valuations across these positions. An endowment effect is an experimentally observable pattern in which initial possession or ownership changes subsequent valuations or exchange decisions. It is established through comparisons between appropriately assigned owners and nonowners, not through the assertion that every owned possession must command a premium. The experimental demonstration and the causal interpretation of that demonstration are distinct scientific questions. Thaler (1980) introduced the concept within a descriptive analysis of consumer choice; Kahneman et al. (1990) later investigated randomized endowments in exchange markets.

This dossier studies reference states, ownership assignments, willingness to accept, willingness to pay, trading reluctance, and mechanisms connecting possession to transaction valuation. It concentrates on transactions and valuation outcomes, leaving the larger formal treatment of risky choice to prospect-theoretic research and the larger treatment of consumer price perception to its adjacent domain. The empirical corpus spans laboratory exchanges, market experience, psychological process experiments, neural correlates, and field observations. The central inference is conditional: estimates depend on elicitation design, decision role, characteristics of the good, expectations, and the participant's history. The evidence is organized as a critical integrative narrative of original studies and research reviews; no new experimental dataset, preregistered systematic-search protocol, or pooled meta-analytic estimate is claimed. Ericson & Fuster (2014) synthesizes the diversity of empirical results and model refinements.

The interpretation of a stated valuation depends on how the question creates an initial cognitive benchmark. In an influential set of experiments, Ariely et al. (2003) showed that arbitrary anchors can shift absolute valuations while relative demand patterns remain internally orderly. Their evidence concerns price anchoring and preference construction rather than an experimentally isolated ownership effect. The connection is methodological: apparent stability of a demand schedule does not necessarily show that the initial level of expressed value was determined independently of contextual cues. In an endowment experiment, experimenter-provided prices, remembered purchase costs, and initial indications of an item's value may alter what participants take as the appropriate comparison. A study that seeks to measure possession-induced valuation should randomize or measure these other anchors and examine whether ownership effects remain after controlling them.

Conceptual Vocabulary and Observable Indicators

Willingness to pay (WTP) is the maximum payment a person would accept for acquiring an entitlement under a specified decision arrangement. Willingness to accept (WTA) is the minimum compensation demanded to surrender an entitlement. In a clean experimental valuation task, both must be tied to the same object and clearly defined exchange rights; a large WTA–WTP difference is one observable indicator of valuation asymmetry. In an exchange paradigm, the outcome may instead be an exchange rate: how often participants voluntarily exchange a randomly assigned object for a comparable alternative. A third outcome is persistence with a default or status quo. These operationalizations share theoretical connections but require separate interpretations because prices, exchange decisions, and defaults can respond differently to incentives. Kahneman et al. (1991) discusses the connections among the effects; Samuelson & Zeckhauser (1988) establishes the broader status-quo literature.

A reference point is a state, expectation, entitlement, comparison price, or distribution against which a transaction is judged. Psychological ownership is a person's experienced sense that a particular object is mine; legal ownership is a rule-defined entitlement that may or may not coincide with that experience. Loss aversion names an asymmetry in the modeled effect of downward versus upward deviations from a reference state. Transaction utility concerns the perceived attractiveness of the terms of an exchange relative to a comparison price. Exchange reluctance is a choice outcome; psychological ownership, loss aversion, and reference prices are proposed explanatory variables. Treating the outcome as proof of any one mechanism would make the theory circular. The relevant task is to identify independent manipulations and measurements that discriminate explanations. Shu & Peck (2011) studies ownership and affect as distinct process variables; Weaver & Frederick (2012) tests a reference-price alternative.

Historical Formation of the Research Program

The modern research line developed from attempts to give consumer theory a better descriptive account of observed choices. Kahneman & Tversky (1979) established a reference-dependent account of risky prospects, in which changes relative to a comparison state rather than final asset positions are central to subjective value. Thaler (1980) then discussed endowment and opportunity-cost anomalies in ordinary consumer choice. The methodological shift was significant: judgments about owning and losing ordinary goods were brought into formal behavioral research. The chronology matters because prospect theory initially addressed risky decisions, whereas transaction experiments involving mugs, tickets, pens, and other goods examine exchanges that can be conducted without an explicit gamble.

Exchange studies made the proposal experimentally accessible. Knetsch (1989) reported asymmetric exchange behavior in real-goods conditions and linked the results to nonreversible tradeoffs. Kahneman et al. (1990) placed randomly endowed participants in markets and compared trade in consumption goods with trading of induced-value tokens. Tversky & Kahneman (1991) then supplied a formal theory of riskless, reference-dependent preferences. The historical sequence subsequently expanded in several directions: List (2003) studied experienced market participants; Plott & Zeiler (2005) tested procedural misunderstandings; Johnson et al. (2007) investigated ordered memory search; Peck & Shu (2009) investigated touch; Ericson & Fuster (2011) manipulated expectations; and Weaver & Frederick (2012) investigated reference prices. The development of the field consists of competing causal tests rather than a single immutable coefficient.

The Reference-Independent Economic Benchmark

A reference-independent consumer model ordinarily represents preferences over feasible bundles of goods and money. If income effects are negligible, transactions are frictionless, information is shared, and a person's ranking is unchanged by initial entitlement, the reservation price for acquiring a marginal good and the compensation required to relinquish the same good tend to be closely related. This statement is conditional. Even conventional theory permits divergence when income effects are substantial, goods lack close substitutes, wealth is constrained, or the acquisition and loss compare different utility states. Hanemann (1991) analyzes conditions under which WTA and WTP can diverge substantially without requiring a behavioral endowment mechanism.

The experimental purpose of random assignment is to narrow alternative explanations. If otherwise comparable participants are allocated mugs or alternative goods by chance, and if they understand the exchange and face real incentives, systematic undertrading relative to an explicit benchmark is evidence that assignment can matter. A causal effect of assignment on trade is narrower than a causal claim about loss aversion. Transaction costs, time costs, information asymmetries, strategic concerns, uncertainty, or incomplete understanding must still be considered. The benchmark should therefore describe predicted trading given the actual experimental allocation, price rule, and incentive mechanism. Kahneman et al. (1990) used induced-value markets as a comparison because known token redemption values provide a sharper benchmark than privately valued consumption goods.

Prospect Theory and Value Around Reference States

The psychological foundation is that an outcome's subjective significance can depend on its deviation from a current benchmark. In Kahneman & Tversky (1979), prospect theory analyzes risky gains and losses and their nonlinear valuation. Its later application to transactions extends the insight to exchanges that change the bundle of goods a person currently holds. Tversky & Kahneman (1991) represents a reference point as a position around which tradeoffs can become asymmetric. Under a loss-sensitive model, giving up an owned mug may be represented as a loss of the mug and acquisition as a gain. An identical monetary amount need not compensate equally for those two directions if changes are valued asymmetrically.

A common schematic expression is V(x|r) = u(x) + g(x-r), where u denotes ordinary consumption value and g denotes the reference-dependent component. Its exact interpretation varies among formal models and is not an empirically established universal equation. In a simple two-sided account, g has a larger marginal response to losses below the reference point than to comparable gains above it. For two-dimensional exchanges, changes in both money and goods matter; a naive one-dimensional ratio can conceal the structure of choice. The researcher must state how the baseline is assigned, whether it is fixed during elicitation, and whether people integrate or separately evaluate attributes. Kőszegi & Rabin (2006) shows why expectations and multiple dimensions change the interpretation of gain-loss utility.

Willingness to Accept and Willingness to Pay

WTA and WTP are role-dependent reservation prices rather than direct observations of utility. Their comparison can be expressed as a difference, WTA minus WTP, or a ratio, WTA divided by WTP. Both summaries require care: the ratio becomes unstable when WTP approaches zero, while an absolute difference depends on currency units and the value of the object. The distributions of individual valuations, the proportion of zero responses, and treatment-specific uncertainty intervals are often more informative than a lone mean ratio. A properly specified design should make seller and buyer rights clear, ensure that values are meaningfully comparable, and describe whether participants face hypothetical or incentive-compatible decisions. Becker et al. (1964) describes the influential single-response elicitation procedure commonly called BDM.

The WTA–WTP gap is sometimes used as a proxy for loss aversion. That inference is valid only under additional assumptions about the mapping from latent value to elicited prices, the stability of reference points, wealth and substitution effects, and the absence of role-specific misunderstandings. A price statement can reflect perceived fairness, market reference prices, anticipated regret, or reluctance to make a bad deal. Plott & Zeiler (2005) showed that instructional and procedural modifications could remove the gap for mugs in their setting. Isoni et al. (2011) found persistent lottery disparities under a related protocol, while Plott & Zeiler (2011) disputed their explanation and emphasized unstable beliefs about lotteries. The disagreement provides direct evidence that an elicited gap cannot identify one psychological mechanism by itself.

Random Endowment and Exchange Experiments

Randomized endowment studies begin by assigning ownership or possession of one of two goods, sometimes while holding tangible properties constant across groups. Participants then receive an option to exchange. Investigators compare the proportion retaining their initial item with a model of exchange expected from independent preferences. Knetsch (1989) demonstrated that direction of initial entitlement can alter exchange behavior. Kahneman et al. (1990) found less trading of randomly allocated consumption goods than expected under their benchmark, while induced-value token markets exhibited the predicted trading volume. These observations gave the endowment literature a strong experimental foothold because the study changed the allocation rather than merely asking people to recall valued possessions.

The interpretation requires examination of timing, task framing, familiarity with the goods, and the perceived legitimacy of the initial assignment. A participant who receives a good and is immediately asked to exchange it may infer that the experimenter has signaled a recommended option, or may regard the second decision as a choice under unresolved uncertainty. A genuine preference for the first item can also produce nontrade in individual cases. Assignment-level causation is established by sufficiently powered and properly randomized comparison across conditions; individual nontrade is not on its own diagnostic. Studies should report the distribution of initial preferences where feasible, the actual transaction conditions, and how alternative goods were equated. A stronger mechanistic test separately manipulates the ownership signal, the expected final state, and the valuation rule.

The Coase-Theorem Connection and Trading Efficiency

The Coase theorem motivates many discussions of the endowment effect because, in an idealized setting with specified property rights and zero transaction costs, the ultimate allocation of resources should be independent of who initially owns them when bargaining is efficient. The experiments of Kahneman et al. (1990) brought this intuition to simple laboratory exchanges. Their undertrading finding suggested that initial entitlements can shape allocations even with markets designed to permit trade. This supplies a welfare-relevant question: whether the initial distribution of rights alters final holdings because preferences are entitlement-dependent or because actual transactions involve additional costs and uncertainties.

The experimental benchmark is frequently simplified in popular accounts. An observed trade count below a predicted number does not directly establish how much welfare has been lost, because subjects' welfare depends on preferences, expectations, and information that the researcher may not fully observe. When participants find ownership itself valuable, keeping an endowed item may reflect a genuine contribution to experienced utility. Economic evaluations must specify whether welfare is defined through experienced outcomes, choice under a selected reference regime, or some stable ex ante preference. Questions of compensation for environmental or legal entitlements raise additional issues about substitution and rights. Hanemann (1991) supplies a conventional theoretical reason for large price differences, and Ericson & Fuster (2014) examines the difficulty of transferring laboratory conclusions into field markets.

The Object Is Part of the Causal Design

Endowment effects differ depending on what is allocated. A mug, a lottery, a consumption experience, a legal entitlement, a highly substitutable commodity, and a personal possession have different objective and subjective affordances. Participants may know the market price of a familiar item, be uncertain about the distribution of lottery outcomes, or attach memories and identity to an object. Consequently, a pooled statement that ownership raises all valuations conflates several exchange environments. The historical literature itself makes these distinctions: Kahneman et al. (1990) compared consumption goods with induced-value tokens; Plott & Zeiler (2005) examined mugs and lotteries; and Isoni et al. (2011) showed that procedural similarity does not guarantee identical findings across object classes.

The study of substitute availability also matters. If an owner cannot easily replace a good, compensation demanded to surrender it may naturally exceed an unendowed consumer's willingness to acquire it. In an experimental design, researchers should measure market knowledge, replacement opportunities, perceived quality, hedonic appeal, uncertainty, and sentimental association where these constructs are theoretically relevant. Randomization balances average participant characteristics but does not equalize a person's response to different object properties. An informative factorial experiment can vary entitlement and replacement ease separately, rather than attributing every increase in required compensation to loss aversion. These distinctions also define the boundary between valuation of ordinary goods and research on unique or identity-bearing possessions.

The BDM Mechanism and Incentive Compatibility

The Becker–DeGroot–Marschak mechanism is frequently used to elicit reservation prices. In an ideal buying version, a participant states the maximum amount they would pay, a price is drawn independently, and the participant buys if the stated maximum meets the drawn price. In the corresponding selling version, a stated minimum can determine whether compensation is sufficient for sale. Under suitable assumptions, reporting a true threshold is an advantageous strategy. Becker et al. (1964) introduced a single-response sequential method that became influential in valuation research. The logical property of a mechanism, however, does not guarantee that a novice participant understands its consequences.

Experimental instructions therefore become part of the validity problem. Research teams may use practice rounds, quizzes, feedback, concrete examples, and repeated opportunities to observe how the drawn price interacts with the stated threshold. Yet practice can itself alter preferences through familiarity, adaptation, or learning about market prices. Plott & Zeiler (2005) combined extensive procedural controls and found no mug WTA–WTP disparity; Isoni et al. (2011) and Plott & Zeiler (2011) debated the significance of lottery results. The scientific conclusion is that incentive compatibility must be demonstrated at both the formal and behavioral levels. Researchers should report the mechanism, the instructions, the comprehension checks, and the timing of ownership assignment so independent teams can evaluate measurement equivalence.

Procedural Artifacts and the Replication Debate

A major methodological turning point came when Plott & Zeiler (2005) argued that confusion about price elicitation could produce apparent valuation disparities. Their study used a protocol intended to address multiple previously identified procedural concerns and reported an absence of the classic mug gap. The finding challenged the practice of treating a WTA–WTP difference as automatic evidence of a structural feature of preferences. Subsequently Isoni et al. (2011) examined results for lotteries, including additional data and new experiments, and argued that disparities can persist under comparable training. Plott & Zeiler (2011) replied that uncertainty beliefs and interpretation of the lottery outcomes mattered. Both sides agree, in effect, that observable elicitation responses require careful behavioral interpretation.

This debate illustrates why replication must be described precisely. A direct replication attempts to reproduce the relevant procedural combination; a conceptual replication varies operationalizations to test the proposed mechanism. An experimental result can be robust within one design and sensitive across object classes, population pools, or elicitation settings. The appropriate evidence table identifies the manipulated property, assigned roles, exclusion rules, incentive rules, primary outcome, and inferential uncertainty. Apparent contradictions may then reveal moderators or competing explanations. Ericson & Fuster (2014) evaluates both the robustness dispute and theoretical developments; it provides a structured starting point for assessing evidence but does not replace direct reading of the underlying experiments.

Market Experience and Learned Exchange Behavior

The effect of market experience is especially informative because it identifies a plausible mechanism of adaptation. List (2003) compared behavior in functioning marketplaces and found that individuals with more trading experience tended to approach neoclassical exchange predictions. Because experienced traders may differ before entering markets, observational associations alone cannot isolate learning. List (2004) continued the investigation in a marketplace sample and reported that intense market experience was associated with more classical exchange behavior, particularly in willingness to part with endowments. These findings constrained any theory predicting an equally strong effect for all individuals regardless of repeated market participation.

A stronger causal test changes experience rather than merely selecting experienced participants. List (2011) used a framed field experiment that experimentally induced market experience and reported attenuation of the anomaly. The interpretation remains task-specific: learning may improve understanding of pricing rules, reduce the psychological significance of an initial entitlement, increase confidence that a replacement can be obtained, or alter expected transaction outcomes. A full mechanism test must discriminate among these channels, preferably by independently measuring expectations and transaction knowledge. Experienced market behavior is relevant to external validity because ordinary consumers accumulate very different kinds of experience. Selling collectibles repeatedly, managing investments, and selling a cherished personal object are psychologically different activities.

Expectation-Based Reference Points

A reference point may be an expected outcome rather than the asset physically held at the moment of choice. Kőszegi & Rabin (2006) formalized gain-loss utility around recent rational expectations, making anticipated outcomes an endogenous part of preference formation. This perspective changes the causal question: a person can perceive a loss when an expected acquisition fails to occur, even without prior possession, and can experience little perceived loss from relinquishing an item they always anticipated selling. Reference states may be distributions rather than single fixed values, requiring attention to anticipated probabilities of acquiring, retaining, or paying for an object.

Ericson & Fuster (2011) experimentally tested expectations in exchange and valuation tasks and reported that expectations influence reference points. Designs of this kind can cross actual ownership with an independent manipulation of what a participant expects to own at the end. That distinction is decisive when evaluating claims that mere possession is sufficient. The expectation approach also implies that reference points can move during a transaction: reading a product description, being told a price, or receiving information about likely outcomes may revise the expected state. Research should distinguish ex ante beliefs, beliefs elicited during the task, and retrospective justifications. Ideally, beliefs are measured before assignment or manipulated without confounding the attractiveness of the object.

Multiple Reference Points and Contextual Comparisons

A single benchmark is often a simplification. An owner can simultaneously compare an object with its acquisition cost, its replacement price, the price a friend paid, a sentimental valuation, a hoped-for resale price, and the expected amount a buyer will offer. The active reference may change with task framing. Weaver & Frederick (2012) argues that an apparent endowment effect can emerge when a proposed transaction seems unfavorable compared with a salient reference price. Ericson & Fuster (2014) notes that the rich psychological literature may require models incorporating multiple reference points. These models turn a broad label into specific testable hypotheses about the informational context of an exchange.

The measurement challenge is to determine which standard is active, rather than merely recording every possible standard after a decision. Experiments can provide independent price anchors, manipulate whether participants view a market listing, or vary whether a purchase or sale is expected. A comparison of changed reservation prices should then hold the object and material payoffs constant. It is possible for multiple processes to operate together: an owned good may trigger a perceived loss, while an offered price simultaneously seems insulting or unfair. The resulting valuation is not an arithmetic sum unless a particular model specifies the aggregation rule. Careful experimental design requires prediction of interaction effects, not just separate main effects.

Psychological Ownership and the Meaning of Mine

Legal rights and psychological ownership occupy different conceptual levels. A person can feel that a borrowed product belongs to them after using or customizing it, while a legal owner can view an object as inventory intended for sale. The sense of mine is therefore an empirical construct that can vary independently of documented title. Shu & Peck (2011) examined psychological ownership and affective reaction as distinct process variables in a program of experiments and proposed their involvement in endowment effects. That work gives a mechanism-oriented account of why the same legal allocation can have different psychological meaning across participants and goods.

Research on psychological ownership requires more than a single post-choice statement that the participant feels attached. If the ownership measure is assessed only after a participant has named a high selling price, the two responses may share demand characteristics or post-hoc rationalization. Experiments can manipulate control, contact, investment of effort, customization, or perceived entitlement independently of price and then measure both ownership experience and actual choices. A causal mediation claim requires particular care, including assumptions about unmeasured mediator–outcome confounding. The broader theoretical issue is whether the causal pathway runs through attachment, loss anticipation, attention, or several processes in succession. The phrase endowment effect names the behavioral outcome and does not itself settle that pathway.

Morewedge et al. (2009) directly challenged the inference that anticipated loss is always the necessary mediator of the endowment effect. Their experiments manipulated ownership and the prospect of relinquishment and argued that the ownership representation itself played a causal role in their conditions. Such evidence should be interpreted at the level of manipulated experimental distinctions, not as universal disproof of loss-sensitive preferences: a participant may possess an object without contemplating a sale, or anticipate losing an unowned but expected item. Ownership can reorganize beliefs about use, responsibility, and the legitimacy of retaining an object, while loss anticipation can become salient later when a transaction is proposed. Replication across real ownership, temporary holding, and expectation manipulations is especially valuable because wording about possession can simultaneously affect several psychological states.

Touch, Embodiment, and Product Interaction

Physical interaction can make an object salient and personally relevant before any purchase. Peck & Shu (2009) found that touching an object, or mentally imagining touch under selected conditions, could increase perceived ownership among nonowners. Their research also concluded that valuation depends jointly on perceived ownership and the valence of the tactile experience. A pleasant texture, an unpleasant sensation, and a neutral product encounter can therefore have different implications for the resulting appraisal. The study illustrates that embodiment is a candidate contributor to transaction valuation and that the sign of an effect may depend on the quality of contact.

Touch experiments are particularly useful because they can vary contact while keeping legal rights unchanged. The design still has alternative explanations: contact supplies diagnostic information about product quality, increases familiarity, focuses attention, and may change expectations of future use. A robust analysis should compare touch with equivalent visual and informational exposure, measure quality beliefs, and separate ownership ratings from purchase bids. The implications extend to stores, product demonstrations, and digital interfaces simulating manipulation, although extrapolation to online commerce requires direct testing. An artificial hand animation or a virtual unboxing event may change engagement without creating the same tactile and proprietary experience as actual possession.

A later investigation provides a more focused test of tactile access independently of proprietary rights. Raeva-Beri & Jansson-Boyd (2025) compared the valuations of buyers who were able or unable to touch a product and varied whether the tactile experience was pleasant. The published experiment reported increased valuations for a pleasant-to-touch good, particularly among participants high in instrumental need for touch. This result speaks to the valuation influence of haptic information in a nonowner role, and thereby constrains explanations attributing a touch-related price change exclusively to legal ownership. It also identifies the positive valence of tactile contact as a substantive boundary condition. The experiment does not establish equivalent effects for every product or for artificial interfaces simulating touch. Follow-up studies could randomize haptic input and ownership simultaneously, calibrate the perceived product quality, and observe completed transactions or returns after the initial appraisal.

Memory Search, Queries, and Constructed Value

The query-theory approach treats valuation as an active construction from information retrieved at the moment of judgment. An owner asked to sell may first search for reasons to keep a good, while a nonowner asked to buy may search for reasons to refrain from purchasing it. The order of memory search can affect which considerations are available and weighted. Johnson et al. (2007) investigated this mechanism and developed a query-theory account of value construction. The approach suggests a specific causal prediction: experimentally changing the order of reasons considered should influence subsequent valuation even when the objective good and entitlement are held constant.

Research on attention and memory offers an alternative to the idea that a stable latent utility is merely revealed by the request for a price. Eliciting WTA and WTP is itself a cognitive task with prompts, search strategies, and implicit comparisons. If a participant first lists benefits of retaining an object, their selling price may change because those benefits are now cognitively accessible. A controlled design can randomize the prompt sequence, collect response times or recall content, and compare valuations following matched exposure. Yet the resulting evidence must distinguish a process that constructs value from a process that simply expresses preexisting concerns. This distinction is best addressed by interventions and prospective predictions rather than by correlations between recalled reasons and prices.

Focusing on What Is Forgone

Buyer and seller roles systematically change what is foregone in a potential exchange. Carmon & Ariely (2000) proposed that buyers attend to the money they would give up, while sellers attend to the object or experience they would surrender. Such a focus can change stated prices without requiring that the owned item has acquired a stable intrinsic premium. The proposal connects valuation asymmetry to attention allocation and the semantics of transaction roles. The same person may use a different mental comparison when asked how much to pay for an object and when asked how little they would accept to release it.

A discriminating experiment changes the focus of evaluation while holding economic consequences fixed. It can prompt buyers to consider the benefits of possession and sellers to consider the uses of compensation, or measure the sequence of attributes that enter judgment. A change in the WTA–WTP gap following the manipulation supports a cognitive-processing explanation, but the analysis should control for changes in beliefs about quality, future use, and alternative goods. Attention may amplify loss-relevant information, making it compatible with some reference-dependent models rather than a logically exclusive alternative. The scientific question is how attention changes the operative inputs to valuation and when those changes survive real exchange incentives.

Reference Prices and Aversion to Unfavorable Deals

A participant can reject a trade because the quoted terms compare unfavorably with a learned or salient price, even if the physical object is not experienced as a painful loss. Weaver & Frederick (2012) tested a reference-price account and reported that interventions reducing disparities between valuations and comparison prices also reduced or eliminated endowment effects in their experiments. This redirects attention from ownership alone to the perceived quality of a bargain. A seller may regard a proposed price as too low relative to the market, and a buyer may regard the same number as too high relative to a familiar benchmark, producing a gap without stable disagreement about usefulness.

Reference-price explanations have clear empirical obligations. The relevant anchor must be measured or independently manipulated; otherwise almost any price can be retrospectively called unfavorable. Researchers can vary market-price information, historical purchase prices, and suggested prices separately while testing willingness to trade. The distinction between a transaction's hedonic or practical value and its perceived fairness is essential because perceived unfairness can persist even when a participant would willingly accept the economic consequences in another frame. A reference-price mechanism may coexist with attachment or expectation-based loss sensitivity, particularly for goods with both strong sentimental and clear market benchmarks.

Social Meaning, Fairness, and Normative Entitlements

An entitlement has social meaning as well as exchange value. Ownership can be understood as a legitimate claim, and an offer to purchase an object can be interpreted as respectful, opportunistic, or unfair. Kahneman et al. (1986) studied public judgments about market fairness and established that entitlement-related standards influence how price changes are evaluated. These findings concern norms of acceptable transactions rather than the experimental mug effect alone. They nonetheless matter because sellers may interpret an exchange offer relative to a fair-price expectation, while buyers evaluate the same offer relative to a different presumed entitlement.

To isolate normative mechanisms, an experiment should manipulate the stated reason for a price, the legitimacy of initial allocation, or the social relationship between participants. A randomly assigned object may acquire less proprietary meaning when the assignment is explicitly temporary, whereas a reward earned through effort can carry a claim of desert. Such predictions remain context-sensitive hypotheses unless directly tested. Social fairness can be a confound when a valuation experiment uses a human buyer who might be judged exploitative. Anonymous and computer-mediated arrangements reduce some interpersonal influences while introducing questions about trust and mechanism comprehension. Social meaning should be incorporated as a measured and experimentally variable part of the transaction environment.

A second social-cognitive mechanism concerns predicting valuations from another transaction role. Van Boven et al. (2000) investigated egocentric empathy gaps between owners and prospective buyers: people evaluating an object from one position may mispredict the valuation of an individual in the other position. This is different from a price gap obtained by directly eliciting owners' and nonowners' preferences. Miscalibration about an opponent's reservation value can affect negotiation outcomes even if both parties clearly understand their own preferences. Mechanism tests should therefore collect sellers' beliefs about buyers' WTP and buyers' beliefs about sellers' WTA independently of actual prices. Belief errors may be linked to attention to one's own forfeited resources and to assumptions about what others know or care about. The work connects endowment research with social perspective taking, strategic bargaining, and errors in interpersonal valuation.

Heterogeneity Across Individuals and Tasks

An average endowment effect may conceal participants whose prices show large asymmetries, little asymmetry, or the opposite pattern. Statistical analysis should report distributions and variation within and across people, not only mean ratios. Gächter et al. (2022) investigated riskless and risky loss-aversion measures in a nonstudent sample of car-manufacturer customers. The researchers reported substantial individual heterogeneity and a positive association between their riskless and risky measures, while noting that average riskless loss aversion exceeded risky loss aversion under their procedures. Such results support some cross-task consistency but do not imply that every individual has a single permanent numerical loss-aversion parameter.

Differences can also be driven by experimentally relevant characteristics: familiarity, market experience, replacement possibilities, attention, ownership experience, wealth, or prior expectations. These should be defined as moderators before analyzing a crowded set of exploratory interactions. Reliability of individual parameter estimates is a separate problem from the reliability of an average treatment effect. Estimating many free parameters from a small set of choices can produce unstable apparent traits. Hierarchical statistical models and repeated-session designs can estimate heterogeneity more effectively, provided model assumptions and out-of-sample predictions are examined. Ericson & Fuster (2014) treats variability across experimental contexts as a central issue in understanding the literature.

A large working-paper research program sharpened the problem of associating endowment gaps with measurable risky-choice loss aversion. Using incentivized surveys of approximately 4,000 U.S. adults, Chapman et al. (2023) examined WTA and WTP for lotteries and reported weak correlations between the two elicited valuations, along with little evidence that the gap tracked independent measures of loss aversion for risky prospects. The working paper is an original empirical contribution with a public institutional record; its status must remain explicit, rather than being silently treated as a peer-reviewed journal publication. Its findings are especially relevant to construct validity because a person may express different prices under buyer and seller roles without ranking as generally loss averse in conventional risky-choice instruments. This result creates a demanding empirical challenge for models that identify one stable latent loss-aversion coefficient across riskless transactions and lotteries. Measurement reliability, repeated decisions, and multivariate model recovery become central to any proposed reconciliation.

Neural Evidence and Explanatory Limits

Neuroeconomic studies examine whether valuation roles engage distinguishable patterns of brain activity. Knutson et al. (2008) used functional magnetic resonance imaging while 24 participants considered products and prices under buying, choosing, and selling conditions. The researchers reported neural activity patterns related to product preference and transaction role, including right-insular activity during selling that predicted individual differences in susceptibility to the effect. The study gives evidence of neural correlates temporally associated with the task. It does not establish that activity in one region uniquely causes an ownership premium, or that the same pattern must occur for every object and decision context.

The interpretive challenge is reverse inference: a region associated with anticipation of losses can also contribute to other salient, aversive, or bodily significant experiences. Experimental neuroimaging is strongest when combined with independently specified behavioral contrasts, preregistered predictions, and convergent methods. A mechanistic account must explain how information about entitlement, price, expectations, and product desirability is transformed into choices; a neural contrast alone does not specify that transformation. The integration of valuation neuroscience with psychological process experiments remains useful because it can test whether apparently similar WTA–WTP gaps are generated through different attentional and affective pathways.

Comparative and Evolutionary Perspectives

Endowment-like exchange asymmetries have been investigated beyond adult human markets. Brosnan et al. (2007) reported endowment effects in chimpanzees in experiments involving exchange of goods. Comparative evidence is relevant because an effect that emerges without human currency markets or explicit prospect-theory instruction may involve more general mechanisms of possession, foraging, or change avoidance. The conclusion must be restricted to the actual species, protocols, goods, and exchange conditions investigated. The presence of an exchange asymmetry in another species does not by itself demonstrate homologous psychological processes or the same operational reference point.

Comparative design is demanding. An animal can fail to exchange because of food preference, neophobia, handling costs, learned task rules, distrust, or an inability to understand the choice. Researchers need independent preference tests, counterbalancing of assignments, reinforcement controls, and comparison across objects. Such studies also caution against equating verbally reported psychological ownership with the full causal structure of possession. For humans, ownership judgments can be linguistically and culturally mediated; for animals, experimental inference rests more directly on observable choice patterns. A strong cross-species theory predicts under which ecological conditions retention should change and specifies how those predictions differ from simple avoidance of an unfamiliar exchange.

Field Evidence and External Validity

The translation from a laboratory exchange task to ordinary economic behavior requires analysis of real institutions. Consumers often face return policies, search costs, guarantees, changing market prices, unequal information, and social commitments. List (2003) and List (2004) are central because their research included functioning marketplaces and examined the role of repeated experience. List (2011) adds experimental evidence on acquired market experience. Together they demonstrate that the magnitude of an observed anomaly is linked to the structure and learning history of the environment. This prevents direct extrapolation from first-time mug trades to every real-life market.

Field observational evidence introduces confounding. Sellers may demand higher prices than buyers because they select better items for sale, have different outside options, or know private information about condition and quality. A difference between posted asking prices and offers is especially ambiguous because negotiations incorporate bargaining strategy and opportunity cost. Stronger tests use exogenous changes in entitlement, randomized transaction framing, repeated observations within the same participants, or natural experiments with credible assumptions. Ericson & Fuster (2014) reviews the limits of external validity. The practical scientific objective is to model market behavior under observable institutions while testing which psychological mechanisms survive after the economic conditions of the trade have been accurately specified.

Recent field evidence broadens the tested goods beyond small, familiar commodities. Tomicki & Kuśmierczyk (2024) observed students' willingness to exchange two forms of examination bonus functioning as risk-reducing instruments, one a point bonus and the other an additional-time benefit, with a delay between endowment and potential exchange. The researchers reported exchange reluctance for both kinds of instrument, despite the fact that the objects of exchange were intangible and had uncertain value to participants. The field setting increases ecological relevance for that particular decision, yet unfamiliar examination rules, perceived legitimacy, and risk-related preferences complicate a direct comparison with mug markets. The primary outcome is retention of a randomly assigned entitlement under a distinctive institution; it is not automatically an estimate of the same preference parameter as a consumer product WTA–WTP ratio. Stronger generalization requires multiple replications in nonacademic institutions and designs in which comprehension, perceived fairness, and the risks of each entitlement are carefully assessed.

Possessions, Personal Continuity, and Duration

The interval between receiving an object and deciding whether to relinquish it can change the psychological task. Immediate endowment paradigms emphasize a newly assigned entitlement. Long-held possessions can accumulate associations with routines, personal memories, interpersonal exchange, and identity. In that setting, a high selling price may reflect genuine use and attachment value as well as a reference-dependent response. Shu & Peck (2011) provides a basis for separately measuring psychological ownership and affect; Peck & Shu (2009) shows how even limited interaction may affect ownership experience. The duration of possession should therefore be recorded, not left as an implicit property of all owner conditions.

Longitudinal research can compare actual changes in valuation before and after acquisition, controlling for selection and learning. Owners may have originally chosen their possessions because those goods suited them particularly well; comparing them with a random group of nonowners can create selection bias. Random assignment followed by repeated valuations gives a clearer trajectory, though repeated measurement may alter expectations. Studies could compare objects that are used, displayed, customized, or stored without interaction. The construct of psychological ownership should be analyzed together with, but distinguished from, remembered usefulness and identity relevance. The research question concerns the development of a valuation pathway across time, rather than the assumption that all long-held goods necessarily become more valued.

Recent experiments test whether the duration of possession can be isolated from the type of possession. Colucci et al. (2024) examined hypothetical ownership over one day, one week, and one month for a tangible mug, an exchange good in the form of a gift card, and an intangible subscription. The online questionnaire results reported endowment-effect patterns across object types while finding less consistent relationships with imagined possession duration. The findings expand the object taxonomy and identify duration as an important variable, but the imagined-ownership design should not be interpreted as proof that actual possession for one month has the same causal consequence as thinking about it. Its principal implication is methodological: researchers need real-time longitudinal assignments, repeated measurements of ownership experience, and separate controls for familiarity, use, and expiration of contractual rights. The data and code have an identified university-repository pathway, making further independent analyses feasible.

Digital Commerce and Virtual Endowment

Digital markets create new forms of simulated possession. Products may appear in a saved cart, virtual wardrobe, customization screen, interactive preview, or digital inventory before legal purchase. These interfaces can affect expectations, familiarity, accessibility, and a user's imagined future relationship with an item. Peck & Shu (2009) demonstrates that sensory interaction and imagined touch can alter perceived ownership under studied conditions, while Ericson & Fuster (2011) demonstrates that expectations can serve as reference points. Both findings motivate testable research on digital acquisition environments, although neither directly verifies a general virtual-endowment effect across contemporary platforms.

An online experiment should randomize interface features while holding price, product information, and purchase opportunity constant. Candidate measures include actual purchasing, willingness to relinquish an item from a virtual inventory, return decisions, comprehension, and perceived entitlement. The design must distinguish selection into customization from the causal effect of customization itself. It should also assess whether a digital item's apparent ownership is durable or merely temporary salience. In algorithmically ranked markets, personalized recommendations can change beliefs about scarcity or expected acquisition, generating reference shifts distinct from attachment. This field requires direct causal and longitudinal testing before extrapolating established mug and touch studies to large commercial systems.

Experimental Design for Mechanism Discrimination

A credible mechanism-discrimination design should cross two or more causal manipulations with separate outcomes. One factor might randomize real ownership, a second the expected final allocation, and a third a salient reference price. Participants would then make incentive-compatible choices under identical economic rules. If actual ownership retains an effect after expectations and price anchors are independently manipulated, models that equate ownership entirely with those factors would need revision. If expectations dominate, an expectation-based account gains support. Ericson & Fuster (2011) and Weaver & Frederick (2012) provide motivating examples of this discriminating strategy.

The design must avoid contaminating one factor with another. Telling participants that an item is theirs may necessarily change expected retention; an experimental condition claiming to manipulate ownership alone requires a manipulation check for expectations. Likewise, inserting a reference price can signal quality or fairness and therefore change baseline utility. Causal identification is strengthened by direct pre-treatment measures of object appeal, clearly defined contrasts, credible randomization, prespecified primary outcomes, and adequate statistical power. The analysis should account for multiple outcomes and interactions rather than selecting the comparison producing the strongest result. Evidence supporting one mechanism does not exclude simultaneous contributions from others, so theoretical claims should state the tested conditions explicitly.

Measurement Quality, Sampling, and Statistical Reporting

Valuation data can be skewed, zero-inflated, and sensitive to outliers. Prices should be reported in original units alongside robust summaries, distributional graphics, treatment contrasts, confidence intervals, and the number of participants whose responses enter each analysis. If researchers compute WTA/WTP ratios, they should explain treatment of zero values and distinguish a ratio of group means from a mean of individual ratios. The two estimands are generally different. The outcome of a randomized trade experiment is often a binary exchange decision, which calls for analysis suited to the actual randomization and clustering structure.

The study protocol should state sampling frame, participant compensation, object assignment, manipulation order, practice instructions, exclusion criteria, and any repeated decision dependencies. Student convenience samples can test mechanisms while limiting population generalization. Nonstudent samples such as the customer population in Gächter et al. (2022) broaden the evidence base, but remain selected on observable and unobservable dimensions. For field studies, investigators should address differential attrition, selection into market experience, item quality, and negotiation dynamics. Blinding, where feasible, and transparent analysis code reduce researcher degrees of freedom. Reproducibility means that another team can regenerate the reported analyses from a documented procedure, rather than merely observing a superficially similar behavioral pattern.

Model Identification and Predictive Validation

Formal loss-aversion models, query-based accounts, psychological-ownership explanations, and reference-price theories are often able to reproduce a high seller price under some parameters. Fitting the same observed gap therefore provides weak discrimination. A theoretically informative study asks what each model predicts under interventions that were not used to tune its parameters. For example, an ownership-only explanation, an expectation-based explanation, and a bad-deal explanation diverge when actual entitlement, expected allocation, and transaction price are varied orthogonally. Kőszegi & Rabin (2006), Johnson et al. (2007), and Weaver & Frederick (2012) articulate mechanisms with distinguishable predictions.

Research programs should report parameter recovery tests, identifiability checks, predictive performance on held-out decisions, and sensitivity to the assumed functional form. A model with several latent reference points can fit many outcomes but may be impossible to identify from a single price response. Hierarchical models can partially pool individual parameters across trials while making uncertainty visible. Model comparison should reward both explanatory reach and prospective prediction. Theoretical pluralism is particularly productive when a mechanistic account specifies a domain of operation and predicts interaction with rival mechanisms. Research advances when alternative accounts lose or gain support under independently defined tests, not when all observed prices can be explained after the fact.

Scientific Reassessment and New Research, 2023–2026

Recent studies add a differentiated evidence base rather than a new universal estimate of the endowment effect. The large U.S. incentivized survey working paper Chapman et al. (2023) centers on lotteries and the weak association of WTA and WTP with one another and with risky-loss-aversion measures. By contrast, the hypothetical questionnaire of Colucci et al. (2024) examines three product categories and imagined possession intervals, while the field study of Tomicki & Kuśmierczyk (2024) concerns the exchange of institutionally specific examination benefits. The tactile experiment of Raeva-Beri & Jansson-Boyd (2025) isolates contact in a buyer role. These projects differ in causal treatment, material incentives, temporal horizon, participant selection, and outcome. A scientific synthesis that averages their claims verbally as if they were repeated measurements of one invariant effect would obscure precisely the dimensions these researchers are trying to identify.

Contemporary theoretical work also questions what counts as irrational in an endowment setting. Altman (2026) interprets possession-based changes in value through acquired product attributes such as familiarity, identity, and control, proposing that observed reluctance to trade can reflect context-sensitive utility rather than an error in judgment. This article is a theoretical reinterpretation rather than an experimental demonstration that all ownership asymmetries are welfare-improving. Its relevance is that welfare analysis must define the value function and account for genuine consequences of experience with an object. If possessing a good supplies additional services or identity-related benefits, a seller's valuation can change because their experienced opportunity set has changed. An empirical test must distinguish such real attribute changes from preference-construction effects generated by role framing and comparison prices.

The modern evidence invites three layers of discrimination. The first layer concerns the random assignment of entitlements and the causal effect of initial ownership on subsequent choice. The second concerns intervening mechanisms: expectation shifts, haptic information, identity or control, retrieval order, and perceived price fairness. The third concerns the long-run consequences for exchange, satisfaction, and the ability to predict future behavior. Findings can be strong at one layer and uncertain at another. For example, a field demonstration of exchange reluctance supports sensitivity to the entitlement under its institutional rules but does not separately identify the affective meaning of losing that entitlement. Likewise, a randomized manipulation of touch can demonstrate a price change in a buyer role without showing that the observed change persists after use.

The recent research record also makes source classification important. Peer-reviewed journal articles, institutional working papers, conceptual reinterpretations, and openly accessible materials with data repositories contribute different kinds of evidence. Chapman et al. (2023) is cited as a working paper, Colucci et al. (2024), Tomicki & Kuśmierczyk (2024), and Raeva-Beri & Jansson-Boyd (2025) as published empirical journal articles, and Altman (2026) as a published theoretical analysis. Where the available review is restricted to an abstract, the present synthesis makes only assertions supported by that abstract; it does not claim to have evaluated unpublished methods tables or unreleased raw data. This explicit provenance is part of scientific quality control, because sweeping claims about evidence quality can themselves become a source of inferential error.

Critical Appraisal of the Experimental Evidence

The strongest source of causal evidence is randomized variation of an entitlement or mechanism under incentives and explicit counterfactual comparisons. Classic randomized exchange studies establish that initial allocations can change trade in selected settings. Studies of trained elicitation establish that procedural comprehension can change observed valuation gaps. Market-experience research shows that behavioral patterns can attenuate with learning. Psychological ownership, touch, query order, expectations, and reference prices provide mechanism-focused manipulations. Kahneman et al. (1990), Plott & Zeiler (2005), List (2011), Shu & Peck (2011), and Weaver & Frederick (2012) form a particularly instructive sequence because they test different causal pieces of the broad phenomenon.

The evidence remains asymmetric across research questions. Many studies identify average treatment effects in specialized tasks, while fewer demonstrate that a single internal parameter travels reliably across goods, roles, populations, and institutions. Some studies have small samples by contemporary standards; the fMRI experiment in Knutson et al. (2008) is informative about a task-specific neural correlate but cannot validate a universal circuit. Reviews such as Ericson & Fuster (2014) aggregate conclusions qualitatively and highlight empirical disagreement. This dossier therefore supports the existence of entitlement- and reference-sensitive valuation under specified conditions, the importance of measurement design, and the relevance of multiple explanatory mechanisms. It does not infer a single immutable loss-aversion constant or a settled ranking of mechanisms for every transaction.

The research through 2026 further emphasizes the difference between replicating a familiar statistical pattern and identifying a portable psychological explanation. The empirical contributions of Colucci et al. (2024), Tomicki & Kuśmierczyk (2024), and Raeva-Beri & Jansson-Boyd (2025) establish effects in distinct hypothetical, field, and touch-manipulation settings, each with its own dependent variable. The working paper Chapman et al. (2023) identifies a construct-validity difficulty when buyer and seller valuations are only weakly related to each other and to elicited risky loss aversion. The theoretical analysis of Altman (2026) adds a welfare interpretation that cannot be settled by price gaps alone. Future evidence reviews should encode these sources by design and target construct, rather than collapsing them under a binary label of evidence for or against the endowment effect.

Generalization, Culture, and Institutional Setting

Cross-cultural generalization requires care because the meaning of legitimate possession, exchange, price fairness, and obligation varies with institutions and social practice. A laboratory endowment protocol administered in multiple places may not be measurement-equivalent if ownership assignments are understood differently or the goods have unequal familiarity. The relevant evidence standard is therefore comparative design with equivalent material incentives, local-language comprehension, and prespecified tests for interaction by institution or population. The original exchange studies are informative about the populations tested; their causal contrasts should not be automatically treated as population prevalence estimates for the world.

Generalization also concerns income, age, market participation, access to substitutes, and experience of financial constraint. Wealth can influence the ordinary utility consequences of payment, while a strong attachment to a particular object can make replacement difficult even with ample financial means. Cultural norms might shape how much an initial transfer is perceived as a gift, a temporary allocation, or a rightful possession. These mechanisms are testable, but cultural labels alone do not provide explanations. The rigorous approach is to specify the mediating institutions and the concrete transaction rules, then compare outcomes after accounting for economically relevant differences. Ericson & Fuster (2014) provides a framework for recognizing the boundaries of the field's current evidence.

Welfare, Consumer Protection, and Decision Architecture

An endowment effect may influence negotiations, return decisions, compensation for rights, product trials, and consumer choice architecture. Whether it benefits or harms the decision maker depends on the outcome and the welfare criterion. An individual who keeps a valued possession may be responding to real attachment and continuity; an individual who refuses a favorable exchange because of a transient procedural misunderstanding may experience avoidable costs. Observed deviation from a textbook exchange benchmark cannot alone settle the welfare question. Hanemann (1991) shows why conventional preferences can generate substantial WTA–WTP differences, while Plott & Zeiler (2005) shows why the elicitation procedure can matter.

For consumer protection, transparent terms and comprehensible choice mechanisms are high-priority research variables. Defaults and simulated pre-ownership can change behavior, making disclosure of actual entitlements important. Samuelson & Zeckhauser (1988) documents the broader influence of default or status quo positions; Weaver & Frederick (2012) demonstrates sensitivity to comparison prices. Interventions should be evaluated through real decisions, downstream satisfaction, complaints, returns, and heterogeneity of effects. A sales technique that increases initial purchases may simultaneously raise regret or returns. Scientific evaluation should therefore assess durable outcomes and autonomy, not simply conversion rates. These applications are hypotheses to investigate under specific institutional conditions rather than blanket conclusions about manipulative or beneficial design.

Human–Artificial Transactions in the Artificial Era

Artificial systems increasingly organize how people encounter, compare, and transact for goods and digital entitlements. A recommendation assistant can select which reference prices are shown, summarize likely resale value, frame a virtual product as already selected, or manage a portfolio of possessions on a user's behalf. These functions modify informational and expectation environments that are directly relevant to reference-dependent valuation. Ericson & Fuster (2011) demonstrates the importance of expectations; Weaver & Frederick (2012) demonstrates price-comparison sensitivity; and Johnson et al. (2007) investigates the order in which value-relevant reasons are retrieved. Their findings motivate hypotheses about automated choice interfaces but do not establish that current artificial agents exhibit human psychological ownership.

An operational research program would compare human valuation under human-curated, algorithm-curated, and neutral presentation conditions. It would measure actual trading, belief calibration, perceived ownership, and changes in reference state over time. If an artificial agent maintains a stable inventory or acquires a simulated entitlement, a researcher can observe whether its policy changes after a possession flag is assigned; that is a behavioral property of the model and environment. Attribution of feeling, loss experience, or psychological attachment requires an independent evidentiary framework. The Artificial Era expands the set of systems mediating reference points and transactions, while preserving the need to distinguish observable policy dependence from asserted subjective states.

Unresolved Problems and Research Priorities

The first open problem is identifying how multiple candidate reference points compete. A single decision can evoke current possession, expected possession, market comparison, and a fair transaction price. A next-generation experiment should record beliefs and knowledge before manipulation, vary the relevant benchmarks separately, and test a model's prospective predictions. The second problem is temporal stability: whether role-dependent valuations survive repeated real transactions, changes in replacement availability, and delayed reassessment. The third concerns the measurement of individual differences without relying on unstable ratios or a small number of lottery choices. Studies should report out-of-sample reliability and model uncertainty. Kőszegi & Rabin (2006), Ericson & Fuster (2011), and Gächter et al. (2022) identify key theoretical and empirical foundations.

Further questions concern how social rights are understood, why different goods produce different patterns, and how ownership experience interacts with attention and perceived transaction fairness. Neural and comparative results should be linked to randomized behavioral manipulations rather than used as standalone validation of a preferred psychological theory. Digital markets add questions about personalization, simulated possession, automated bargaining, and the persistence of expectations under algorithmic environments. The essential methodological priority is to study the generation of a valuation together with the conditions of its observation. A scientifically mature account must predict when a gap appears, when it disappears, which processes carry it, and how the resulting choice relates to the participant's own welfare.

Conclusion

Reference-dependent valuation is a demonstrable and theoretically productive property of many human transaction contexts. Initial entitlements, expectations, comparative prices, and cognitive processing influence how people evaluate acquiring, keeping, exchanging, and surrendering goods. The classical evidence of asymmetric exchange in randomly assigned consumption goods established the scientific importance of initial allocations. Subsequent research showed that the resulting asymmetry varies with experimental procedures, market experience, object classes, and reference states. Kahneman et al. (1990), Plott & Zeiler (2005), List (2011), and Ericson & Fuster (2014) jointly anchor this conclusion without requiring agreement on a universal mechanism.

The endowment effect is best treated as a family of measurable transaction-level outcomes requiring mechanistic explanation. Loss aversion, expectation-based preferences, psychological ownership, memory search, the focus on forgone alternatives, and reference-price evaluation are complementary or competing causal hypotheses depending on the intervention and formal assumptions. Evidence is strongest when the design changes a proposed cause independently of the price or exchange outcome. The resulting synthesis positions valuation research at the intersection of cognitive and social psychology, consumer behavior, behavioral economics, and institutional design. In the Artificial Era, the same methodological discipline is required to investigate the systems that organize reference points for human decisions and the observable policies of artificial agents operating in transaction environments.

Recent research extends the agenda to imagined possession of diverse goods, uncertainty-bearing field entitlements, tactile information separated from ownership, population-scale lottery elicitation, and explicit alternative welfare accounts. Colucci et al. (2024), Tomicki & Kuśmierczyk (2024), Raeva-Beri & Jansson-Boyd (2025), Chapman et al. (2023), and Altman (2026) illustrate the research directions without yielding a common numerical effect estimate. Across this corpus, the most productive explanatory move is to specify the entitlement, the reference state, the mechanism, the valuation instrument, and the duration over which consequences are measured. Such specification allows distinct traditions to contribute complementary tests and yields a cumulative, reproducible framework for understanding human valuation.

References

  • Altman, M. (2026). Why the endowment effect reflects unbiased smart decision-making and can increase the willingness to trade. Review of Behavioral Economics, 1–15. https://doi.org/10.1108/RBE-02-2025-1007
  • Ariely, D., Loewenstein, G., & Prelec, D. (2003). “Coherent arbitrariness”: Stable demand curves without stable preferences. The Quarterly Journal of Economics, 118(1), 73–106. https://doi.org/10.1162/00335530360535153
  • Becker, G. M., DeGroot, M. H., & Marschak, J. (1964). Measuring utility by a single-response sequential method. Behavioral Science, 9(3), 226–232. https://doi.org/10.1002/bs.3830090304
  • Brosnan, S. F., Jones, O. D., Lambeth, S. P., Mareno, M. C., Richardson, A. S., & Schapiro, S. J. (2007). Endowment effects in chimpanzees. Current Biology, 17(19), 1704–1707. https://doi.org/10.1016/j.cub.2007.08.059
  • Carmon, Z., & 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
  • Chapman, J., Dean, M., Ortoleva, P., Snowberg, E., & Camerer, C. (2023). Willingness to accept, willingness to pay, and loss aversion (NBER Working Paper No. 30836). National Bureau of Economic Research. https://doi.org/10.3386/w30836
  • Colucci, D., Franco, C., & Valori, V. (2024). The endowment effect with different possession times and types of items. Journal of Behavioral and Experimental Economics, 110, Article 102216. https://doi.org/10.1016/j.socec.2024.102216
  • Ericson, K. M. M., & Fuster, A. (2011). Expectations as endowments: Evidence on reference-dependent preferences from exchange and valuation experiments. The Quarterly Journal of Economics, 126(4), 1879–1907. https://doi.org/10.1093/qje/qjr034
  • Ericson, K. M. M., & Fuster, A. (2014). The endowment effect. Annual Review of Economics, 6, 555–579. https://doi.org/10.1146/annurev-economics-080213-041320
  • Gächter, S., Johnson, E. J., & Herrmann, A. (2022). Individual-level loss aversion in riskless and risky choices. Theory and Decision, 92, 599–624. https://doi.org/10.1007/s11238-021-09839-8
  • Hanemann, W. M. (1991). Willingness to pay and willingness to accept: How much can they differ? American Economic Review, 81(3), 635–647. https://www.jstor.org/stable/2006525
  • Isoni, A., Loomes, G., & 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., & 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., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–292. https://doi.org/10.2307/1914185
  • Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1986). Fairness as a constraint on profit seeking: Entitlements in the market. American Economic Review, 76(4), 728–741. https://ideas.repec.org/a/aea/aecrev/v76y1986i4p728-41.html
  • Kahneman, D., Knetsch, J. L., & 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., & 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://ideas.repec.org/a/aea/aecrev/v79y1989i5p1277-84.html
  • Knutson, B., Wimmer, G. E., Rick, S., Hollon, N. G., Prelec, D., & Loewenstein, G. (2008). Neural antecedents of the endowment effect. Neuron, 58(5), 814–822. https://doi.org/10.1016/j.neuron.2008.05.018
  • Kőszegi, B., & Rabin, M. (2006). A model of reference-dependent preferences. The Quarterly Journal of Economics, 121(4), 1133–1165. https://doi.org/10.1093/qje/121.4.1133
  • List, J. A. (2003). Does market experience eliminate market anomalies? The Quarterly Journal of Economics, 118(1), 41–71. https://doi.org/10.1162/00335530360535144
  • List, J. A. (2004). Neoclassical theory versus prospect theory: Evidence from the marketplace. Econometrica, 72(2), 615–625. https://doi.org/10.1111/j.1468-0262.2004.00502.x
  • 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
  • Morewedge, C. K., Shu, L. L., Gilbert, D. T., & 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
  • Peck, J., & Shu, S. B. (2009). The effect of mere touch on perceived ownership. Journal of Consumer Research, 36(3), 434–447. https://doi.org/10.1086/598614
  • Plott, C. R., & 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
  • Plott, C. R., & Zeiler, K. (2011). The willingness to pay–willingness to accept gap, the “endowment effect,” subject misconceptions, and experimental procedures for eliciting valuations: Reply. American Economic Review, 101(2), 1012–1028. https://doi.org/10.1257/aer.101.2.1012
  • Raeva-Beri, D., & Jansson-Boyd, C. V. (2025). What makes buyers willing to pay more? On the role of touch in the endowment effect. Journal of Behavioral Economics for Policy, 9(2), 29–38. https://sabeconomics.org/ojs/index.php/jbep/article/view/237
  • Samuelson, W., & Zeckhauser, R. (1988). Status quo bias in decision making. Journal of Risk and Uncertainty, 1, 7–59. https://doi.org/10.1007/BF00055564
  • Shu, S. B., & Peck, J. (2011). Psychological ownership and affective reaction: Emotional attachment process variables and the endowment effect. Journal of Consumer Psychology, 21(4), 439–452. https://doi.org/10.1016/j.jcps.2011.01.002
  • Thaler, R. (1980). Toward a positive theory of consumer choice. Journal of Economic Behavior & Organization, 1(1), 39–60. https://doi.org/10.1016/0167-2681(80)90051-7
  • Tomicki, F., & Kuśmierczyk, P. (2024). The endowment effect in a field study with risk-reducing instruments. Risks, 12(12), Article 194. https://doi.org/10.3390/risks12120194
  • Tversky, A., & Kahneman, D. (1991). Loss aversion in riskless choice: A reference-dependent model. The Quarterly Journal of Economics, 106(4), 1039–1061. https://doi.org/10.2307/2937956
  • Van Boven, L., Dunning, D., & Loewenstein, G. (2000). Egocentric empathy gaps between owners and buyers: Misperceptions of the endowment effect. Journal of Personality and Social Psychology, 79(1), 66–76. https://doi.org/10.1037/0022-3514.79.1.66
  • Weaver, R., & Frederick, S. (2012). A reference price theory of the endowment effect. Journal of Marketing Research, 49(5), 696–707. https://doi.org/10.1509/jmr.09.0103