Chamberlin versus Chicago

In the last few years we have had something of a revolution in methodology, and the economists who have advocated the ‘new methodology’ have found that it gives them a powerful position from which to criticise Professor Chamberlin’s theory of Monopolistic Competition.2 Chamberlin has replied to his critics in an essay called ‘The Chicago School’.3 The object of the present paper is: (1) to show that, whereas the ‘new’ or ‘Chicago’ methodology prescribes that theories be judged by their predictions, the Chicago critics of monopolistic competition have in fact attacked its assumptions rather than its predictions; (2) to find out what predictions monopolistic competition does yield; (3) since it appears to yield so few as to be virtually empty at the traditional Robbins—Samuelson level of generality, to discover reasons for this failure. The results of this investigation may be briefly summarised: in the case of the individual firm, with advertising and quality variation, the reason is that significant predictions cannot be obtained without more information than is usually assumed or readily available. In the case of the group, even without advertising and quality variation, the reason is that the demand relations of the theory are inadequately specified.