Common Property Resources, Storage Facilities and Ownership Structures: A Cournot Model of the Oil Market

During the last decade economists have tried hard to convince Meadows and his followers that market forces do not tend to exhaust our natural resources too rapidly as compared with the requirements of an intertemporal Pareto optimum. It is well known now that, under well-defined property rights and correct price expectations, a competitive extraction industry brings about a Pareto-optimal depletion path, and it has been shown that oligopolistic and monopolistic market structures may even produce a bias towards conservation. 1 It is still an unresolved issue, however, which extraction path will result if property rights are imperfectly defined. Following the pioneering work of Gordon (1954) there has been a large body of literature demonstrating that, in the extreme case of free entry and myopic profit maximization, there are strong reasons for excessive extraction.2 But this literature hardly yields reliable conclusions for the perhaps more realistic case where firms derive their behaviour from intertemporal optimization and where there is a common ownership, but limited access to the resource. The basic intuition behind the common-pool problem is that the individual agent chooses an extraction policy other than the one he would choose under well-defined property rights since he is afraid that part of the resource he leaves underground will be extracted and sold by others.3 Despite the simplicity of this intuition, there does not seem to be any publication that analyses it within an intertemporal equilibrium model with rational and far-sighted agents. This is the motivation for the present approach. The approach studies the common-pool problem in the context of an oligopolistic world market for oil. It posits a given number of identical firms extracting oil from holdings between some of which there is mutual seepage. The firms sell the extracted oil in the same market or they store it in privately owned, seepage-proof tanks. Each firm derives its behaviour from intertemporal optimization under perfect foresight of the time paths of the model variables, but Cournotesque conjectures about the reactions of its competitors. Except for the assumption of storage facilities, this specification is closely related to models previously studied by Khalatbari (1977) and Kemp and Long (1980, essay 10).4 The two models are very similar but they yield strikingly different conclusions. While Kemp and Long find that the extraction path chosen by the firms is Pareto-optimal, i.e. that the price of the resource rises at a rate given by the market rate of interest, Khalatbari contends that there is over-extraction. Kemp and Long convincingly demonstrate that Khalatbari's model suffers from an inconsistency in the way the individual firm expects its rivals to respond to its own actions, and they argue that this inconsistency is the reason for the divergence in results. Thus the casual reader gets the