Neoclassical abuse

Terence Hutchinson, now emeritus professor at Birmingham University, has made distinguished contributions over many years both to the history of economic thought and to debates on economic methodology. These interests, outside the conventional mainstream of high mathematical theory, required intellectual courage to follow so consistently, for Hutchinson is a long-standing critic of much of what passes for respectability in academic economics.

The appearance of this volume, when increasing doubt is being cast on the paradigms of orthodox economics, is therefore timely. The book consists of 13 separate papers, almost all written in the past 15 years. Seven of the essays are on aspects of economic thought up to and including the late 19th century, and the remainder deal with various issues in methodology.

The papers on the history of thought are fascinating, not least for the insights provided in the detailed notes to each chapter. Yet, despite the clarity of the writing, the non-economist will find the essays hard to appreciate to the full. For example, there is an impassioned attack on the early 19th century economist David Ricardo, who remains one of the greats. Hutchinson refers to the "preposterous case" of Ricardo and to his "dangerously active (methodological) heritage". These are important arguments, but require prior knowledge to follow well. Other chapters include two on Jevons, a hero of the author, and some intriguing pages on economics before Adam Smith.

Perhaps the chapter of widest appeal to the general reader is the final one, written in 1993 and whose title forms the title of the book. This brings together a number of themes which run throughout the essays on methodology.

A key criticism by Hutchinson of contemporary orthodox economics is the way in which purely abstract, formal reasoning has been elevated in status within the profession. He rightly identifies the current source of this as the American graduate schools, whose dominance of conventional economics is now almost complete.

The degree to which the world can be understood purely by abstract reasoning rather than by empirical deduction is an important question which is common to many disciplines, not unique to economics. But the great theorists such as Archimedes and Newton have had acute powers of empirical observation. Further, in the natural sciences, theoretical models are ultimately subjected to empirical tests of the greatest rigour.

Hutchinson reminds us forcefully that not only do the same standards not apply in orthodox economics, but many such academics have found it "prudent to withdraw from the dangers of prediction, so as to protect their 'professionalism', or 'rigour' ".

He points out, equally clearly, that the founding fathers of conventional economics did not subscribe to these views in any way. Alfred Marshall, for example, who occupied a Cambridge chair for many years around the turn of the century, made a major contribution to the development of neoclassical economics, the currently dominant methodology.

Yet Hutchinson demonstrates Marshall's contempt for the "calamitous notion" that abstract economics was proper economics. Marshall, in common with Leon Walras, the other great originator of neoclassical theory, believed that the dominant aim of economics was to contribute to a solution of social problems. In Marshall's view "intellectual toys" diverted attention from the real world, in which poverty causes degradation and to the study of which our energies should be devoted.

Hutchinson sympathises with a second methodological view held not only by Marshall, but by all his Cambridge successors from Keynes onwards right up to the most recent few years, that the use of mathematics in economics causes more problems than it solves.

Here I must part company to some degree. The strictures that the use of maths can easily lead us away from constructive work are surely correct, as a glance at almost any academic economics journal confirms. But, properly applied, the use of mathematics can enable great progress to be made.

The most devastating attacks on the neoclassical framework have come from mathematical economists, whose conclusions the profession refuses to face. The model of competitive markets is central both to neoclassical theory and to the dominant trend in world economic policy. But its key results hold only in a timeless world. Radner showed as long ago as 1968 that in a model in which the future exists, the existence of competitive equilibrium requires quite literally infinite computing power for every individual. Newbery and Stiglitz showed in 1982 that in a world with time a competitive equilibrium is not in general a Pareto optimum, yet students are still taught that it is.

This difference of opinion does not in any way detract from the book, for its arguments precisely stimulate such thinking.

Paul Ormerod is visiting professor of economics, University of Manchester, and author of The Death of Economics.