The Coase Theorem is false on purely economic analytic grounds. Furthermore, the conventional interpretation of Coases message, and the policy prescriptions which follow from this do not observe the strict qualification which Coase placed on his result. There is, therefore, a divergence of meaning between the orthodox statement of the Coase Theorem and what Coase intends (as revealed in his 1989 retrospectives: Notes on the problem of Social Cost.). This may also be dissonant with what he meant in his original 1960 article.
The initial allocation of legal entitlements does not matter from an efficiency perspective so long as they can be exchanged in a perfectly competitive market.To illustrate, consider Coases famous example whereby locomotives emit sparks which set fire to farmers fields. Suppose the legal intervention is by an initial governmental legislative allocation of spark emission permits, which allows the railway to emit only a restricted amount of sparks. Assume these can be traded with the farmers. While the law apparently controls the extent of the damage, the theorem dissents, stating that it is the market which determines the final efficient allocation of permits between farmers and the railway, and hence the extent of the damage.
This is because the farmers and the railway will face the above aggregate marginal willingness to pay and marginal cost curves for the trade in permits. Starting at, say, an initial allocation Q1, the individuals most preferred level of emission reduction, firms will buy permits until the efficient outcome Q* results. Indeed Q* will result regardless of the initial allocation of legal entitlements.
The rationale for Stiglers (1966) definition of the Coase Theorem is now evident: Under perfect competition, private and social costs will be equal. Coase (1988) argues that zero transaction costs are a suitable proxy for perfect competition and notes that zero transaction costs were assumed in his original (1960) paper. An alternate definition by Cooter, then, is the best indicator of what Coase means: The initial allocation of legal entitlement does not matter from an efficiency perspective so long as the transaction costs of exchange are nil.
Regardless of the inadequacies of the zero transaction costs approximation, the initial allocation of entitlement always has distribution consequences for wealth. As a result, demand for other goods and services will be dependent on the initial allocation of property rights. The shapes of our aggregate marginal willingness to pay and marginal cost curves will then change, being dependent on the demand for other goods and services. Consequently, the efficient outcome Q* is dependent on the initial legal decision. This amounts to outright theoretical falsification of the theorem.
Varian (1987) has noted, however, that in the very special circumstances, where preferences are quasilinear, the strict version of the theorem may hold. Quasilinear preferences suggest the absence of income effects, implying demands for the emission permits are independent of the income distribution. This is seen in the form of a horizontal contract curve yielded in an Edgeworth Box analysis of the situation.
Varian doesnt acknowledge Arrows aforementioned objection, however, which would deny the theorems veracity even in this special case.
A final objection to the theorem is that the notable externalities of our age, such as ozone depletion and nuclear pollution, affect a large number of people, implying large transaction costs. This further consolidates the political irrelevance of the theorem.
It also indicates the removal of Coases present position from the, at times, rash and barely qualified espousal of its result, by his disciples. Consider Stiglers (1966) description of the original statement as a profound article and its result as a remarkable proposition to us older economists who have believed the opposite.
It also suggests a divergence of Coases present position from what he once meant. Circumscribing criteria seem comparatively de-emphasised in the original article, in which he states that the usual courses of action (on externalities) are inappropriate, in that they lead to results which are not necessarily, or even usually desirable.
Coase, R (1988) The Firm, the Market and the Law
Cooter (1987) The Coase Theorem in The New Palgrave
Stigler, G (1966) The Theory of Price
Varian, H 1987 Intermediate Microeconomics