New Experiments in Industrial Organisation

Abstract

Experimental economics has devoted much of its relatively short life to the study of issues of industrial organisation, and has achieved notable successes, of considerable value to the development of both the theoretical and empirical aspects of the subject. The general message emerging from the market-level experiments is that the existence of, and convergence to, the (market) equilibrium of neoclassical theory is, in general, well-supported by the experimental evidence; on the other hand, the general message emerging from the individual level experiments is that the foundations of the neoclassical market theories are seriously flawed. This paper argues for new types of experiments in industrial organisation, ones that are theory-suggesting rather than theory-testing, and ones that are more ill-defined than those carried out to date. There is a need to shed some light on how people tackle extremely complicated ill-defined problems as opposed to experiments confined to testing well-defined theories of simple problems with complicated solutions.

I am particularly grateful to the ESRC, the Leverhulme Trust and the Fondazione No Vanzi for financing the experiments discussed in this paper. I would also like to pay tribute to Patrizia Sbriglia, who has carried out much of the work on The Combination Race, and to Cristina Pitassi, who has worked with me on the Oligopoly Experiment discussed at the end of the paper. Finally, my especial thanks to Norman Spivey, for writing the software for two of the three experiments discussed here, and to Jo Hall for typing this paper faster than I could write it.