Experimenting with the Coase theorem

ABSTRACT

Kahneman, Knetsch, and Thaler's [(1990). Experimental tests of the endowment effect and the Coase theorem. Journal of Political Economy, 98, 1325–1348] experiment on the Coase theorem disrupted a string of experimental successes in the 1980s. The source of their refutation is the endowment effect which generates a reluctance to trade. We use Steven Medema's recent benchmark interpretation of the Coase theorem to subject their experiment to methodological scrutiny, generating four distinct explanations of their findings. We find that their explanation is the only one at odds with the theorem. While Kahneman, Knetsch, & Thaler argued that they undermined the Coase theorem, their result is constrained by the exclusion of the rationality assumption and the adoption of the invariance-efficiency criterion. There is no immaculate Coase theorem and therefore no single experimental test that can falsify it. Instead, different experiments test specific deviations from a benchmark theorem.

KEYWORDS:

• Coase theorem
• transaction costs
• efficiency
• experiments
• decision costs
• rationality

JEL Codes:

• B21
• C91
• D23
• D61
• D82
• D91

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Ramzi Mabsout is currently assistant professor of economics at the American University of Beirut. His research interests range from ethics and economics, to the methodology and history of economic thoughts, to applied research in behavioral economics. He is particularly interested in: (i) the history and methodology of utility theory in economics; (ii) the crossroads where economics and psychology meet including the epistemic implications of the experimental turn in economics; (iii) the continuously evolving relationship between economics and ethics as both disciplines absorb the transformative effects of experimentation.

Hossein Radmard is assistant professor in economics at the American University of Beirut. He received his Ph.D. in economics from West Virginia University. His research fields are development economics, political economy, and institutional economics. He is particularly interested in studying informal institutions such as religion.

Notes

1 Samuels (1974 [1992, p. 85]) argued the Coase theorem is a partial equilibrium model that neglects general equilibrium implications, a tautology that misconceives rights, a doctrine rather than a scientific theorem in which ‘ideology [is] at one of its most esoteric levels’. More recent critiques by Farell (1987) and Canterbery and Marvasti (1992) consider the theorem is circular while Usher (1998) claims it is either tautological, incoherent, or wrong and Ventura, Cafiero, and Montibeller (2016) that it is tautological or false. Halpin (2007) posits it is logically inconsistent. Medema (1996) responds to Canterbery & Marvasti whereas Allen (2015) responds to Usher and Halpin.

2 See Coursey, Hoffman, and Spitzer (1987), Harrison, Hoffman, Rutstrom, and Spitzer (1987), Hoffman and Spitzer (1982, 1985, 1986), Harrison and McKee (1985), Prudencio (1982), and Schwab (1988).

3 Claims made, respectively, by Farber (1997), Samuels (1974 [1992]), Hoffman and Spitzer (1993, p. 62), Rachlinski and Jourdain (1998, p. 1546), Zamir (2015, p. 22), and Medema (1997).

4 See Bertrand (2010, pp. 980–1), Coleman (2002: chapter 3), Medema (1999, p. 213), Samuels (1974 [1992]) for explanations as to why Coase implicitly uses Pareto efficiency.

5 A similar definition is adopted by Allen (2015, p. 379) ‘if transaction costs are zero, then the allocation of resources is independent of the distribution of property rights’ and Korobkin (2014, p. 1231) ‘if transaction costs are zero, the assignment of a legal entitlement by the state will not affect the ultimate ownership of that entitlement’.

6 Specifically, ‘a random allocation design was used to test for the presence of an endowment effect. Half of the subjects were endowed with a good and became potential sellers in each market; the other half of the subjects were potential buyers. Conventional economic analysis yields the simple prediction that one-half of the goods should be traded in voluntary exchanges. If value is unaffected by ownership, then the distribution of values in the two subgroups should be the same except sampling variation. The supply and demand curves should therefore be mirror images of each other, intersecting at their common median. The null hypothesis is therefore that half of the goods provided should change hands’ (Kahneman et al., 1990, p. 1328).

7 KKT repeat the experiment with mugs and envelops with an unknown amount of money and find a similar reluctance to trade.

8 It is not 100% as in the bargaining session over induced-tokens because the utility surplus for market goods is related to individual preferences and some subjects may prefer money to chocolate or may dislike chocolate.

9 While the high WTAs were also reported in experiments where the good was allocated randomly by KKT, in these other experiments the exchange mechanism was not bargaining as it ought to be to test the unstructured bargaining version of the Coase theorem.

10 In section 2 we listed versions of the Coase theorem by Calabresi and Medema (but see also Hoffman & Spitzer, 1982) that account for rationality as a separate assumption from zero transaction costs. The inclusion of rationality as requirement for zero transaction cost can be found in Buchanan and Tullock (1962), Coase (1960, 1988), Cooter (1982), Calabresi (1991), Barzel (1985), Hovenkamp (1990, p. 787), Illing (1992), Pratten (1997), and Schlag (1989) among others. Coase (1981, p. 187), specifically, considers studying a world without transaction costs is equivalent to ‘augurs divining the future by the minute inspection of the entrails of a goose’. He later repeats that it is not worthwhile to spend much time studying a world without transaction costs (Coase, 1988, p. 15). One may speculate that Coase considers transaction costs ubiquitous because they include decision costs as he had originally argued in 1960: ‘In order to carry out a market transaction, it is necessary to discover who it is one wishes to deal with, to inform people that one wishes to deal and on what terms, to conduct negotiations leading up to a bargain, to draw up the contract, to undertake the inspection needs to make sure that the terms of the contract are being observed, and so on’ (Coase, 1960, p. 15). Coase (1988, p. 6) is even more specific subsequently and argues that transaction costs include ‘search and information costs, bargaining and decision costs, policing and enforcement’. Dahlman (1979), whose definition of transaction costs Coase approved, argued all transaction costs ‘have in common that they represent resource losses due to lack of information. Both search and information costs owe their existence to imperfect information about the existence and location of trading opportunities or about the quality or other characteristics of items available for trade … Therefore, it is really necessary to talk only about one type of transaction cost: resources losses incurred due to imperfect information’ (Dahlman, 1979, pp. 147–8). More recent definitions re-emphasize that transaction costs should more specifically account for problems of rationality and knowledge (Calabresi, 1991, p. 1211). Barzel (1985, p. 5) contends transaction costs are determined by product information which are homogenized in Walrasian models to eliminate exchange costs. Schlag (1989, p. 1690) identifies information about the products, associated markets, and consumer preferences as the informational prerequisites for market based transaction cost analysis.

11 It appears to be a problem that plagues behavioral economics. For discussion see Wilkinson and Klaes (2012, pp. 204–205).

12 Smith's postulate of non-satiation is that ‘given a costless choice between two alternatives, identical except that the first yields more of the reward medium (usually a currency) than the second, the first will always be chosen (preferred) over the second, by an autonomous individual i.e. utility is monotone increasing function of the monetary rewards’ (Smith, 1976, p. 275).

13 Horowitz and McConnell (2002) find that ordinary goods have lower gaps (with the highest gaps reported for health/safety and public non-market goods followed by ordinary private goods, lotteries and cash redeemable token and money). Similarly, Sayman and Onculer (2005) run a meta-regression with the gap as dependent variable and a range of explanatory variables among which the availability of substitutes. They find that the availability of substitutes is statistically significant and negative, the more substitutes the goods has the smaller the WTA/WTP gap.

Additional information

Funding

This work has been supported by the University Research Board (URB) of the American University of Beirut.