Category Archives: endowment effect

Klass & Zeiler: Endowment theory is wrong.

Klass, G., & Zeiler, K. (2013). Against Endowment Theory: Experimental Economics and Legal Scholarship. SSRN Electronic Journal.

This is a remarkable article. The authors make a clear-cut claim and justify this claim by means of a strong and convincing argument, grounded in experimental research; they put their claim in historical context; they spell out some implications of their claim for future research; and all along they cover an impressive amount of literature from different disciplines. They know their shit back and forth.

The simple claim is that endowment theory, at least as it has been commonly interpreted in behavioral economics and, derivatively, in legal studies, should be rejected.

The core of their argument is straightforward. It is generally accepted that the endowment effect expresses itself in the laboratory by means of exchange asymmetries: for example, people who are randomly assigned the mug don’t want to trade it for the pen, and vice-versa. In valuation studies, it expresses itself in a gap between the Willing-To-Accept and Willing-To-Pay prices: the amount of money someone is willing to pay for a concert ticket is less than what the same person is willing to accept to give up a ticket that she already has. Plott & Zeiler (2005), however, in a series of studies, found that those gaps and asymmetries can be made to disappear by manipulating contextual variables. For instance, typically, when the endowed good is handed to the participant, she is told, “I’m giving you the mug.  It is a gift.  You own it.  It is yours.”  But this utterance might signal that the mug is very valuable. When one substitutes this phrasing with a simpler formulation such as “the mug is yours, you own it,” and one also tweaks other factors (decisions are anonymous, the pen and the mug are at the same physical distance from the subject, etc.), the endowment effect tends to vanish. The endowment effect, to sum up, is an artifact of the researchers’ faulty methods that goes away when one controls all relevant variables.

The article provides historical context for the emergence of endowment theory. Behavioral economists used the endowment effect as a paradigmatic spearhead against neoclassical economics. The clash (and the types of arguments used) reminds me of intercultural psychology critique of homo economicus (Henrich et al., 2007).

In the conclusions, the authors emphasize that simple experiments on human decision making do not translate automatically into normative or political recommendations. Rather, they contribute to corroborate or falsify a theory, and it is theories that can eventually have normative implications.

Klass and Zeiler’s description of the endowment effect changed my understanding of it. They emphasize that, for endowment theory to be true, the increase in value has to result only from pure ownership, that is, from the mere fact that one has an entitlement. Let’s say that there are two mugs, and I own mug 1 but not mug 2, and mug 1 and mug 2 are identical, and both are placed at the same distance from me, and there are no facts in the object history that create any bond between me and any of the mugs. Klass and Zailer seem to claim that, according to endowment theory, I should still value mug 1 more than mug 2, because of the sheer fact that I am mug 1’s owner.

That is: Klass and Zeiler emphasize that pure, “abstract” ownership should be responsible for the endowment effect, and not the history of how the object came to be mine. Thus, they seem to exclude value resulting from attachment to the object (“emotional” or “sentimental” value). By free association, I am reminded of the Kantian view of duty as resulting from bare respect for the moral law. That is: I behave morally not because I feel some subjective satisfaction or pride in being moral, or because some utilitarian calculation that moral behavior will produce the greatest good. I think and act in accordance with my duty because that’s what I ought to do, period. In endowment theory, my thinking and acting is affected by my right, period.

I previously thought (perhaps erroneously) that the endowment effect relied in a richer conception of ownership that included psychological aspects, such as the individual’s attachment to the specific object and the pride one feels for one’s belongings. In my opinion, this view is represented by Gelman & Davidson’s (2016) study on children’s preference for unique owned objects; McEwan, Pesowski, & Friedman (2016) research showing that children view owned objects as non-fungible; Gelman, Manczak, & Noles (2012) paper that claims to have identified an endowment effect in children 2 and 3 years of age: “The present findings suggest that positive evaluation of and preference for one’s own possessions is a basic cognitive disposition, even before children have experience with conventional economic transactions.” There are many other researchers in the field of ownership (Rochat, Ross, etc.) that seem to have this richer understanding of what owning an object means. Perhaps developmental psychologists have misread endowment theory; perhaps they’re talking about a different phenomenon.

This richer view of the subject-object relationship has many antecedents in psychology. One may think of attachment theory, Winnicott’s transitional objects, etc. These different theories are not akin to the cold inclusion of the ownership title in the algorithm that calculates value, but to the warm fact that I value an object because it became mine in a certain context, as result from a certain history. Objects are extensions of the self (here I could quote some twenty developmental psychologists); ownership of property constitutes an important aspect of the construction of self-identity and of boundaries between myself and others.

Hume says something similar in his book on Emotions, in his Enquiry Concerning Human Understanding. One is proud only of the things one owns. In Humean terms, the idea that this mug is splendid and the idea that it belongs to me are associated to each other. The fact that the things that relate to my self (my body, my watch, my horse, my children, my estate, my country) are beautiful or valuable increases my pride; conversely, if my self-esteem is high I will tend to consider that my stuff is the best, just because it’s mine.


Gelman, S. A., & Davidson, N. S. (2016). Young children’s preference for unique owned objects. Cognition.

Gelman, S. a., Manczak, E. M., & Noles, N. S. (2012). The nonobvious basis of ownership: Preschool children trace the history and value of owned objects. Child Development, 83(5), 1732–1747.

Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Mcelreath, R., … Sciences, B. (2007). “Economic Man” in Cross-cultural Perspective : Behavioral Experiments in 15 Small-scale Societies. Social Dynamics, 204(06), 795–815.

Klass, G., & Zeiler, K. (2013). Against Endowment Theory: Experimental Economics and Legal Scholarship. SSRN Electronic Journal.

McEwan, S., Pesowski, M. L., & Friedman, O. (2016). Identical but not interchangeable: Preschoolers view owned objects as non-fungible. Cognition, 146, 16–21.

Plott, C. R., & Zeiler, K. (2005, June 1). The Willingness to Pay-Willingness to Accept Gap, the “Endowment Effect,” Subject Misconceptions, and Experimental Procedures for Eliciting Valuations. Retrieved from



Susan Gelman on children’s preference for unique owned objects

Gelman, S. A., & Davidson, N. S. (2016). Young children’s preference for unique owned objects. Cognition.

This is an incredible and profound study on children’s attachment to objects, with implications both for cognitive and emotional development. It’s also a study that amazes me for the amount of work it required from researchers (who had to find brand-new replicas of more than 100 children’s attachment objects).

The experimental design is quite simple. Researchers asked 36 three-year-olds to choose between two toys for either themselves or the researcher: an old (visibly used) toy vs. a new (more attractive) toy matched in type and appearance (e.g., old vs brand-new blanket). Focal pairs contrasted an old toy that belonged to the child with a matched new object; control pairs contrasted toys the child had never seen before.

The conclusion of the study is that, by 3 years of age, young children place special value on unique owned objects. Children prefer their original objects to newer, better versions, but only in the case of the focal pairs (with their objects of attachment) and not with the control pairs (objects the child had never seen before). These findings are consistent with the view that possessions are extensions of the self.

In addition, these preferences hold for “sleep” objects (blanket, pillow) and toys representing an animated character (dolls, action figures) but not for inanimate objects (a car, a toy hammer, etc.) Uniqueness is valued for sleep objects and animate toys, but not for inanimate toys. Moreover, in 30 out of 31 cases, attachment objects had a proper name. Ownership, attachment and anthropomorphism (eyes, animated features, soft or furry texture) all combine to enhance children’s preferences for their own objects.

In addition, children seemed to understand that their special objects had value for them only in so far as they share a history with the object. That is, they did not attribute the researcher the same preference for the old (attachement) object. In this sense, they seem to understand the subjective nature of value.

The results are remarkable, among other things, because of the understanding of the abstract ownership relationships, the distinction between appearance and reality and the perspective-taking abilities involved in children’s responses. The authors also emphasize how attuned any child can be to minor features of an object that indicate that it is her unique object and not a substitute.

These findings also offer a different (experimental, cognitive) perspective on the phenomenon of “transitional objects,” first described by Donald Winnicott in the 1950s.

A remarkable study indeed.

Children see property as nonfungible


McEwan, S., Pesowski, M. L., & Friedman, O. (2016). Identical but not interchangeable: Preschoolers view owned objects as non-fungible. Cognition, 146, 16–21.

This is another great article by Ori Friedman’s team. They did three experiments to find out whether children see owned objects as fungible (i.e., as replaceable or interchangeable). In Experiment 1, children considered an agent who takes one of two identical objects and leaves the other for a peer. When considering a scenario where a boy took one of two identical objects home, and left the other for a girl, preschoolers viewed his behavior as more acceptable when he took his own item, rather than the girl’s.

In Experiment 2, children considered scenarios where one agent took property from another. When considering a scenario where a boy deprived a girl of her balloon, preschoolers judged it acceptable for the girl to take back her own balloon; but they judged it unacceptable for her to take the boy’s balloon, even though it was the only balloon available to compensate her.

Finally, in Experiment 3A and 3B, children considered scenarios where a teacher could give a child either of two objects to play with—an object that the child had recently played with, or another object that looked the same. When considering a scenario where a teacher could give a boy one of two identical-looking balls to play with, preschoolers were more likely to say she should return the ball that the boy had previously played with when it belonged to him, compared with when it was her own.

These findings indicate that children see property as non-fungible.

Previous studies showed that children at these ages show concern for owners’ rights to an object. McEwan et al.’s findings extend knowledge by showing that these concerns persist even when an identical replacement is available to the other. The fact that children at these ages already show intuitions of non-fungibility indicates that such intuitions are an early development, and perhaps foundational in people’s reasoning about ownership. People view ownership as granting people rights to particular objects (i.e., rather than to objects of a certain type).

These are all very relevant and important findings that add detail to current knowledge about the development of ownership.

One conceptual doubt. The fact that children say that it’s not ok to take the perpetrator’s object may not mean that children literally see identical objects as interchangeable. In my opinion, this last statement presupposes a “physicalist” view of the world, understood as a collection of free floating objects with certain physical characteristics that make them different or identical, and placed in certain positions within a 3D space. An alternative view is that children, when they respond to the interviewer, are judging the actions and intentions of the characters, in the context of a social situation that includes objects. And, as Gelman says, objects have histories. So children may think something like “it’s not ok to take someone else’s property even if they took yours first”. It is also more likely that they think in terms of particular objects, not in terms of classes or categories of objects. So the concepts of “identical” or “interchangeable” may not play a role in children’s reasoning. Also, the difference between responses to the balloon situation and the cookie situation might be due to the fact that children take into account the actions of the perpetrators, and perhaps her intentions. It’s not the same popping a balloon accidentally than eating a cookie purposefully.

Great article.

Children value ideas over labor

Text #15

Li, V., Shaw, A., & Olson, K. R. (2013). Ideas versus labor : What do children value in artistic creation ? COGNITION, 127(1), 38–45. doi:10.1016/j.cognition.2012.11.001

The procedure is simple: have an adult direct a child making a work of art (so that the adult is contributing the guiding ideas while the child is contributing “brute” labor). Then, reverse the roles: have the child supply the guiding idea while the adult follows directions and executes the work of art. Finally, have the child chose which final product she prefers to take home with her: the one that incorporated her effort or the one that reflects her idea?

In a second experiment, the researchers used a similar situation but now they tricked the subjects so that children believed that the drawing contained their ideas when it actually contained the adult’s idea (and vice versa, they believed that the drawing which they had actually created while being directed by an adult was the one that incorporated their ideas).

In a third experiment, they used a third person narrative to lay out a comparison between someone who contributes labor and someone who contributes ideas to the creation of an object. Who should keep the resulting product?

These studies demonstrated that by 6 years old, children value ideas over physical labor. Six year olds systematically chose pictures that contained their own ideas over pictures that contained their labor, even when they were merely tricked into believing that they had come up with the idea for a picture that they had not. Further, 6 year olds demonstrated a general appreciation of ideas – they not only valued their own ideas (Studies 1 and 2), but also privileged idea creators over laborers in a property dispute (Study 3). In contrast, 4 year olds appear to have preferred pictures that contained their specific idiosyncratic preferences. Four year olds preferred pictures containing their ideas, but also their idiosyncratic preferences in Study 1 and pictures they believed contained their labor but also their idiosyncratic preferences in Study 2. Further supporting this possibility, in Study 3 where idiosyncratic preferences could not play a role in selection, 4 year olds showed no bias for either a third-party idea creator or laborer. Six year olds, by way of contrast, sided with the idea creators in third-party case, even when they personally had no connection to the idea.

The age effect in these studies may exist because 6, but not 4 year olds, understand that ideas are valuable and can thus be owned.

In conclusion, the tendency to value ideas is present in childhood and may emerge between 4 and 6 years old. 6 year olds value ideas over labor even when making third-party judgments, favoring those who only contributed ideas as more deserving of a picture over those who only contributed labor.

Ownership, object history and endowment effect in 2 and 3 year olds – Gelman

Text #13

Gelman, S. a., Manczak, E. M., & Noles, N. S. (2012). The nonobvious basis of ownership: Preschool children trace the history and value of owned objects. Child Development, 83(5), 1732–1747. doi:10.1111/j.1467-8624.2012.01806.x

The paper presents two studies with 2 year-olds, 3 year-olds, and adults. The first study addresses the use of object history to determine ownership, while the second studies the endowment effect. The experimental design is very simple: assign objects (from certain object sets) to individuals, and then ask subjects a) which object belongs to whom, b) which object they like best.

1) Ownership is abstract… or nonobvious.

One a conceptual level, it is interesting that this paper emphasizes the fact that ownership is abstract. I have previously used that word, abstract, to refer to ownership in a recent paper of my own. But, actually, that’s not the word that Gelman et al. prefer to use when speaking of ownership; they describe it as nonobvious rather than abstract. Here’s what they say:

  • Ownership is of interest because it is a cognitive construction, not materially present in the owned object. As Snare (1972, p. 200) aptly stated, ‘‘[A] stolen apple doesn’t look any different from any other apple.’’
  • A mature concept of ownership includes an understanding that proximity, perceptual or functional features, and desirability, although potent factors, cannot by themselves determine who owns what. In other words, ownership is an invisible quality that can be traced by consideration of object history rather than by inspection of the properties of the object.

I don’t know why the talk about “nonobvious” rather than “abstract”. Abstract seems to be the common-sense word to describe invisible, relational features.

2) Object history.

In order to assess who owns what, human agents reconstruct the history of each individual object (“first I found it there, then I gave it to her, then she placed it in that red box… etc.). “Given the centrality of object history, it becomes particularly important to track where an object moves over time.”

“Experiment 1 demonstrates that 3-year-olds, like adults, construe ownership as a nonobvious property that does not reduce to outward perceptual or functional features”. “This doesn’t seem to be the case for 2-year-olds, who in certain experimental situations and with certain sets of objects do not seem to use object history in their ownership judgments.” “The present studies demonstrate that children as young as 3 years of age spontaneously attend to object history to determine ownership.”

3) Endowment effect

The authors claim that they are the first to demonstrate an endowment effect in children 2 and 3 years of age. “The present findings suggest that positive evaluation of and preference for one’s own possessions is a basic cognitive disposition, even before children have experience with conventional economic transactions.”

This is interesting. By the way, why do they say that the disposition to value one’s own stuff is “cognitive”? They call it a “cognitive disposition”. But you can make the case that such a basic way of being of humans and perhaps other species is not appropriately described as “cognitive”. We are dealing with something that is as emotional as it is cognitive. We value and defend what we are and what we have. This phenomenon (the endowment effect) might be linked even with territorial behaviors. I think there are deep existential, cultural and evolutionary motivations in this “will to possess” and to value what one has. The word “cognitive” would suggest a conceptual frame, an innate category, some “cold” rule or feature of our cognitive system. It’s not proved that that’s the best description of what’s going on here.

“Ownership confers special value on objects, across the life span. This finding extends beyond prior work in demonstrating that preference is for the particular object assigned (not just for that type of object).”

4) Endowment effect precedes object history tracking

“The most striking developmental change concerned the comparison between the two experiments. Three-year-olds and adults distinguish ownership from likability, reporting that they owned objects even when they did not like them (e.g., the participant-plain items). In contrast, 2-year-olds show no difference between the ownership task (Experiment 1) and the endowment task (Experiment 2). In other words, 2-year-olds conflate ownership with desirability, thus failing to grasp that a toy they do not like actually belongs to them.

5) Mutual exclusivity

“Participants showed a mutual exclusivity bias concerning ownership, rarely assigning an object to Zippy [a fictional character they introduced in some situations] that had already been assigned to another owner. Mutual exclusivity is a principle that young children adhere to in their word extensions (Markman, 1989; Markman, Wasow, & Hansen, 2003), and it is notable that this same principle applies outside the realm of labeling.”

One might speculate about mutual exclusivity as a common feature of ownership and language. Perhaps this is an innate rule or constraint of our cognitive system, that is applied to different domains as language and ownership? Or perhaps mutual exclusivity as a principle co-evolved with human society and the rules of ownership? There are theories that derive human’s symbolic functioning from social life, and in particular from the institution of ownership. Mutual exclusivity is essential for the institution of ownership: the very meaning of something being mine is that I, not you, control it; if something is mine it’s not yours. In order to claim something as your own, you need to put a mark on the object, or make a gesture, or use some other symbolic means, that might be at the very origin of the semiotic function in general and language in particular.