Irreconcilable Differences in Tests of Hotelling's Theory of Exhaustible Resources

Empirical tests of the theory of exhaustible resources have provided mixed results regarding empirical validity of the theory. This paper considers differences in the primary testing methodologies and alternative measures of the in situ resource price. We demonstrate that in situ price measures common in the literature are theoretically equivalent for the vertically integrated firm. We also show that empirical shadow price test methodologies imply the same restricted econometric model, while the unrestricted models diverge. We find that the Halvorsen and Smith (1991) test methodology is relatively efficient, and, using data on natural gas wells, reject the unrestricted transition equation of traditional methodology. We conclude that discrepant empirical test results may be an artifact of the econometric testing methodologies rather than a reflection on the validity of the theory.