Excluding the Endowment Effect?

I. Introduction

                The Coase Theorem has no doubt left an indelible mark on Law and Economics. The theorem proposes that, absent transaction costs, property will be allocated efficiently regardless of initial entitlement.[1] Widespread acceptance of this assertion has given rise to much legal analysis aimed at reducing transaction costs in order to lubricate bargaining and achieve efficient allocation of property rights. On the whole this is not a bad thing – transaction costs are a major obstacle to efficient bargaining and should be reduced. However, recent studies which indicate the existence of the endowment effect have lead many scholars to re-examine their initial assumptions regarding the importance initial entitlement.[2]

II. Background

                A. The Coase Theorem

                The main premise of the Coase Theorem is that, absent transaction costs, property rights will be efficiently allocated regardless of initial entitlement.[3] This conclusion leads to two further assertions named invariance and efficiency. Invariance states that, discounting transaction costs, the same efficient ultimate distribution will prevail. Efficiency states that, discounting transaction costs, the ultimate distribution will be efficient. The theorem is used to support the notion that clearly defined property rights will promote efficient distribution of those rights. One of the primary transaction costs is information – the ability and the cost of obtaining it. The importance of information is highlighted by the famous prisoners’ dilemma.[4] In the prisoners’ dilemma each actor would be better off cooperating instead of refusing to cooperate. However, absent such information and confidence, the dominant strategy is to not cooperate. Accordingly, much attention has been given to the reduction of transaction costs.

                B. The Endowment Effect

                The endowment effect indicates that “[p]eople are reluctant to part with their property, and the amount that they are willing to accept (WTA) to sell it generally far exceeds the amount that others are willing to pay (WTP) for it.”[5] The endowment effect is essentially the gap between a person’s buy and sell price regarding a specific good. This finding suggests that initial entitlement would play an important role in the efficient allocation of resources because specific valuations will be altered by ownership. In particular, two studies have led to conclusions which could prove influential for a law and economics analysis of the regulation of business transactions.

                C. The Endowment Effect and Differing Legal Regimes

                The first study conducted by Jeffrey J. Rachlinski and Forest Jourden compared the endowment effects in property and liability regimes.[6] A property regime is one of equity where one’s rights are typically protected by injunctive relief or excludability under the law – a court order mandating certain activity or a cessation of activity by the defendant. A liabilities regime is one of law where a plaintiff’s rights are protected by damages – the plaintiff receives monetary compensation for her loss. Rachlinski’s study was comprised of two similar tests which were further divided into three distinct fact patterns differing only by the remedy to be bought or sold.[7] For each fact pattern there were two groups – buyers and sellers. Buyers were given the opportunity to purchase a right associated with a piece of property.[8] Sellers were given the opportunity to sell a right associated with a piece of property.[9] Each individual had the opportunity to purchase/sell one of the following rights: a court order prohibiting interference with, a right to a large amount of damages which was coupled with a low probability of occurrence, or a right to a small damages amount coupled with a high probability of occurrence.[10] Each of the individuals only saw the right enumerated in their fact pattern. The Buyers and Sellers in both tests were informed that their purchase, or refusal to sell (obtaining or keeping the right) thereby attaching somewhat diminished utility from not having the money offered/spent.[11] Finally the dependent variable in the test was the willingness to sell/buy. Rachlinski and Jourden’s test indicated that rights protected by property rules tended to show a stronger endowment effect than those protected by liabilities.[12] The test which was measured by a “willingness to sell/buy” indicated that a property regime produced more people who were unwilling to sell than a liabilities regime produced.[13] Furthermore the difference between the willingness to buy under a property regime vs. a liabilities regime was almost non-existent.[14]

                D. The Endowment Effect and Incentives

                The second study by Christopher Sprigman and Christopher Buccafusco measured differing valuations of a non-rivalrous good between “authors”, “owners,” and “bidders.”[15] This test attempted to see if a creator (“author”) of a good would experience a larger endowment effect then an “owner.”[16] The “author” in this test wrote a poem which would be submitted to a competition to win a prize.[17] The “author” was told that there was an additional group (the “bidders”) who would attempt to purchase the chance at winning.[18] Each of the “authors” was told that the highest bidder would be assigned to their poem, and that they should indicate the minimum amount at which they would be willing to sell.[19] The “owner” was randomly assigned a poem and given the same options as the “author” (keep the chance at winning, or sell the chance). The “bidder” was given no initial entitlement but rather assigned a poem and given the opportunity to purchase that poem’s chance at winning.[20] Thus there was one group who was given initial entitlement by virtue of their effort, and another by virtue of simply being there, and the third group had no entitlement.[21] The three groups were subjected to three different tests involving the aforementioned options.[22] In the first test they were told that the winner would be selected subjectively (that the “best” poem would win). [23]The groups were only allowed to view the poem assigned to them, and asked their willing to sell/buy price.[24] In the second test the selection of the winning poem was the same as the first.[25] However, in this test they were allowed to see all of the poems, but still only given the opportunity to buy/sell the poem assigned to them.[26]  In the third test, the subjects were informed that the winner of the contest would be random (that it would be drawn from a hat). In this third test they were allowed to read the poems, but were reminded that the chance of winning was not contingent on the quality of the poems.[27] The results of the experiment indicated that “authors” and “owners” did not show any significant difference in the endowment effect.[28] The experiment also showed a rather significant endowment effect goods which would be covered by IP law, even when the profitability of the rights were clearly objective (luck of the draw) as opposed to subjective (quality of the poem).[29] The findings of the two tests, combined with the larger implication of the endowment effect (that initial entitlement does matter, creates variables, which thought seemingly at odds with, are not wholly incompatible with Coasin Bargaing, but would have to be accounted for.

III. Discussion

                According to classical economics, it is assumed that, when one purchases property both the property and the cash are being transferred to those who value it most – a major tenet of Rational Choice Theory.[30] For example, if person A sells a bicycle to person B for $10 then person A would typically be held to value the ten dollars more than the bicycle and person B would be held to values the bicycle more than $10. In such a transaction $10 would indicate A’s minimum sell price and B’s maximum buy price. It can be inferred that the minimum sell price indicates the point at which any less money received would have less value than the bicycle, and that the maximum buy price indicates the point at which any more money paid would outweigh the value of the bicycle.

                Alternatively, if A’s minimum sell price is $12 but B’s maximum is $10 then the transaction will not occur because neither party would accept the other’s price – having the bicycle is more valuable to A then $10 and having $12 is more valuable to B then having the bicycle. Therefore in this second scenario the rights would already be efficiently allocated. If the initial bicycle entitlement were granted to B instead of A, but they still retained the same aforementioned valuations, then according to the Coase theorem the right to the bicycle should transfer to A. This conclusion is only made possible if one makes several assumptions. First, one would have to assume that their valuations impute a preference – that because A is unwilling to accept anything less the $12 in exchange for his bicycle he would be willing to pay anything less than $12 to obtain it, and that because B is unwilling to pay more than $10 for a bicycle he would be willing to accept any amount higher to sell it. This first assumption derives A’s maximum buy price from his minimum sell price should he lack initial entitlement, and derives B’s minimum sell price from his maximum buy price should he enjoy initial entitlement. This assumes that the previously stated (and presumably “measured” values) can reliably indicate a predicted value should the situation change. This first assumption is founded on a larger assumption – valuations and/or preferences do not change according to initial entitlement. It is posited here that these assumptions are only possible if one discounts the endowment effect and that if either of these assumptions is incorrect they would threaten to undermine the conclusion of the Coase theorem. I would like to note that there are other assumptions that must be made in order for the Coase theorem to work (e.g. absence of transaction costs etc.); however these assumptions will not be addressed as they are beyond the scope of the paper.

                The bicycle example assumed that the valuation of the bicycle would remain the same whether or not the person was buying or selling. However, this may not always be the case. Numerous experiments and studies point to the existence of an endowment effect.[31] As mentioned above the endowment effect can best be described as the phenomenon that people are often willing to sell an item they own for a much higher price than they would be willing to pay for it.[32] The most famous experiment regarding the existence of the endowment effect was conducted by Robert Thaler and his colleagues.[33] In this test, Thaler gave mugs out to half of his students, and according to classical economic theory both the owners and potential buyers of the mugs should have valued them at the same price and about half of the mugs should have been exchanged.[34] However, the owners consistently over valued the mugs and only about a quarter of them were exchanged.[35] Jeffrey J. Rachlinski and Forest Jourden in their 1998 law review article provide an excellent example which analyzes the effects of initial entitlement on ultimate distribution and valuation.[36] In the example there is a homeowner who would be willing to purchase a pollution easement for $2,000 or less, but would only be willing to sell it for $6,000 or more.[37] In the same example, there is a factory owner who would be willing to pay $2,500 or less for this same right, and would be willing to part with it for $5,500 or more.[38] In this scenario, regardless of who receives initial entitlement, no transaction will occur, which in turn might lead some to conclude that the rights are in possession of those who value them most.[39] However, this conclusion is without merit because arbitrarily switching the initial entitlement would still yield no transactions – either person, if she/he is the owner, will refuse to sell because both willing to sell prices are higher than the other’s willing to buy price. Thus, in this scenario the rights are not necessarily in the hands of those who value them most, but rather merely who received them first.

                The stagnation illustrated in Rachlinski’s problem does not necessitate regulating the distribution of initial entitlement to avoid the endowment effect.[40] Such an undertaking would face very serious moral criticisms in addition to complications associated with cost and proof. Such a regulation would have to take into account that initial entitlement colors valuation of a specific right. In order to mitigate this effect the regulation would have to ensure that the initial entitlement did not interfere with the allocation of this right to the person who valued it most. In the aforementioned scenario, and lacking such regulations, the only way to do this would be to assign it to the party who valued it most. Determining which party valued it the most would be quite difficult as using or deriving the buy/sell prices of either party would be influenced by whether or not they owned it in the first place. In light of the difficulties it is apparent that regulation should not focus on forcing efficient distribution of initial entitlements in an effort to avoid the endowment effect, but rather encourage legislation that attempts to minimize the endowment effect itself. Thus, in a sense, Coasian bargaining may still take place but only if the initial entitlement does color the valuations such that it impedes Coasian bargaining.

IV. Analysis

                If one is to focus not on the efficient distribution of initial entitlement, but rather the original goal of allowing property rights to be efficiently distributed then one must first turn examine the nature of property. Property is often broken down in terms of rivalry and excludability. Rivalrous goods are whose consumption affects the consumption of that good by another person. A milk dud I eat cannot be eaten (at least in the same way) by another person. Excludability is the ability to bar someone from consuming the good in question.  Public Goods are goods which cannot be excluded, and whose consumption does not affect the consumption by another. Knowledge and ideas are two things which could conceivably fall into this category. When a person has a new and original idea it could technically be called a club good. In order to get access to this club you have to be the originator. However, if the originator of the idea lets people know and it becomes common knowledge it is no longer a club good but a public good. The rights granted by Intellectual property law essentially turn what would be a public good into a club good, or rather temporarily maintains its prior status. In the United States, the purpose of Intellectual Property Law is to “to promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.”[41] The presumption is that originators of ideas will lack incentives to share those ideas because they will often not be allowed to monetize the benefits of those ideas and thus may be less likely to spend the time to create new thing, or in the event that they do they will be more likely to internalize the benefit as opposed to sharing it in order to gain a marginal competitive advantage. Thus temporary excludability is often granted to Intellectual Property in order to promote its creation and dissemination.

                The experiment from Rachlinski’s paper suggests that the endowment effect is increased when the right is protected by a property regime and that the endowment effect is mitigated (or even eliminated) when that same right is protected under a liabilities regime.[42] The experiment showed that as the right being exchanged between parties moved from a property regime to a damages regime those willing to sell the right increased, whereas those willing to buy remained relatively constant.[43] Rachlinski suggests that the data retrieved from his experiments the notion that “the inability of other people to appropriate my things lies at the core of the notions of ‘ownership’ and ‘property’”[44] Thus by granting one the right to exclude it increases his unwillingness to sell it but leaves his willingness to purchase unaffected. This difference can be seen as a skewed attachment stemming from initial entitlement, and is indicative of the endowment effect because the willingness to buy is not reciprocally diminished. When combined with the findings from Sprigman’s paper – that the endowment effect does exist in intellectual property, and that it exists whether or not the profitability of that good is determined by chance or quality – it may hold that the endowment effect created in goods made excludable by IP law may be reduced if the coverage extends to protections involving liability without decreasing the willingness to purchase.[45]

V. Conclusion

                Many studies have indicated the presence of an endowment effect. This presence of this effect given the right conditions could lead to inefficient bargaining, and/or an inefficient allocation of property rights due to initial entitlement. If the Endowment Effect is determined to be a common barrier to efficient bargaining such that it warrants legislative regulation, then that regulation should forgo any possibility of forcing initial entitlement from one party to another. Alternatively, the regulation should first work on identifying the possible sources of the effect, and work to create laws which would allow the market to take care of the allocation of property rights. If the right to exclude does not incentivize the buyer to buy, and disincentivizes the owner to sell then it may prove to be an impediment to bargaining. Perhaps the court’s reluctance to apply equitable remedies is not merely following protocol rooted in the old rivalry between law and equity. Modern Courts have done away with the plaintiff’s matter-of-course right to an injunction.[46] Perhaps this is evidence that excludability is not necessarily required to maintain incentives and profitability of a Club Good which would otherwise be a Public Good.  

[1] See Robert Cooter and Thomas Ulen, Law & Economics, 85-1, 225, 268  (Pearson Addison Wesley 2008)

[2] See Daniel Kahneman, Jack L. Knetsch & Richard H. Thaler, Experimental Tests of the Endowment Effect and the Coase Theorem, 98 J. Pol. Econ. 1325 (1990); See Richard Thaler, Toward a Positive Theory of Consumer Choice, 1 J. Econ. Behav. & Org. 39 (1980); ”); Amos Tversky & Daniel Kahneman, The Framing of Decisions and the Psychology of Choice, 211 Sci. 453 (1981); Russell B. Korobkin, The Endowment Effect and Legal Analysis, 97 Nw. L. Rev. 1227, 1229 (2003); Jason F. Shogren; Seung Y. Shin; Dermot J. Hayes; James B. Kliebenstein 'Resolving Differences in Willingness to Pay and Willingness to Accept' The American Economic Review, Vol. 84, No. 1. (Mar., 1994), pp. 255–270; W. Michael Hanemann 'Willingness to Pay and Willingness to Accept: How Much Can They Differ?' The American Economic Review, Vol. 81, No. 3. (Jun., 1991), pp. 635–647; Jeffrey J. Rachlinski & Forest Jourden, Remedies and the Psychology of Ownership, 51 Vand. L. Rev. 1541 (1998); Christopher Sprigman & Christopher Buccafusco, Valuing Intellectual Property:  An Experiment (2010).

[3] Milton Friedman, Essays in Positive Economics, 15, 22, 31 (1953).

[4] Cooter, supra note 1, at 38-41.

[5] Sprigman, supra note 2, at 3.

[6] Rachlinski, supra note 2, at all.

[7] Id.

[8] Id.

[9] Id

[10] Id.

[11] Id.

[12] Id.

[13] Id.

[14] Id.

[15] Sprigman, supra note 2, at all

[16] Id.

[17] Id.

[18] Id.

[19] Id.

[20] Id.

[21] Id.

[22] Id.

[23] Id.

[24] Id.

[25] Id.

[26] Id.

[27] Id.

[28] Id.

[29]Id., at 30

[30] Friedman, supra note 3.

[31] Supra, note 2.

[32] See note 1 supra

[33] Thaler, Supra note 2

[34] See Thaler, supra note 2; see Sprigman supra note 2

[35] See Thaler supra note 2.

[36] Rachlinski, supra note 2, at 1555.

[37] See Id. at all.

[38] Id.

[39] Id.

[40] Id.

[41] U.S. CONST. art. 1, § 8

[42] Id.

[43] Id.

[44] Id.

[45] See Sprigman, supra note 2.

[46] See eBay v. MercExchange, LLC, 126 S. Ct. 1837 (2006)