Money modellers miss the point

Economists must address the real world or risk losing their best students and researchersto other disciplines

What is wrong with contemporary academic economics? There seems little doubt that something is wrong, whatever yardstick we use. Its formalistic theoretical models are increasingly found to have little purpose, its empirically-based forecasting models do not forecast particularly well, and, perhaps most markedly of all, there is a significant disparity between the practices of economists and the methodologies they draw upon. The confusion generated by this situation results in unhappy compromises, the widespread discouragement of both criticism and approaches not favoured by the mainstream.

Some who recognise the failings of the formalistic modelling project at the heart of all mainstream economics have attempted a resolution. But there is little evidence of success. And many refuse to acknowledge that academic economics gives cause for concern. The result is that the project has continued largely unabated. The assessment by Wassily Leontief, a Nobel Memorial Prize winner in economic science, seems as relevant now as 15 years ago when it was made: "Professional economic journals are filled with mathematical formulas leading the reader from sets of more or less plausible but entirely arbitrary assumptions to precisely stated but irrelevant theoretical conclusions... Year after year economic theorists continue to produce scores of mathematical models and to explore in great detail their formal properties; and the econometricians fit algebraic functions of all possible shapes to essentially the same sets of data without being able to advance, in any perceptible way, a systematic understanding of the structure and the operations of a real economic system."

So what accounts for this situation? Modern economics is remarkable for the minimal concern shown either for tailoring methods to insights available regarding the nature of the social world, or for explicitly determining conditions under which chosen methods would be appropriate. Rather, methods are formulated according to their technical sophistication, or their conformity with prior conceptions of proper practice. Most economists seem merely to take for granted that if ever real-world problems are to be studied the chosen methods will necessarily "fit". In this sense reality is neglected in modern economics. And, in my view, it is this presumption that methods of economic analysis can be fashioned without explicit regard for the nature of social phenomena that accounts for the subject's numerous failings.

Certain contributors to one strand of mainstream economics - pure theorists - sometimes acknowledge that their methods do not (yet at least) shed insight on the world in which we live. Mathematical economists start their arguments with constructions along the lines: "assume a world of omniscient agents" or "let us assume three good economies lasting for two periods". The complexities of real people's motivations, lives and economic environments are rarely included. As Frank Hahn, a major theorist in this tradition, acknowledges: "When a mathematical economist assumes that there are three good economies lasting two periods, or that agents are infinitely lived I everyone can see that we are not dealing with any actual economy. The assumptions are there to enable certain results to emerge and not because they are to be taken descriptively."

Other theorists in this tradition admit that it is often difficult to see the point. Ariel Rubinstein, for example, says: "Economic theory lacks a consensus as to its purpose and interpretation. Again and again we find ourselves asking the question 'where does it lead'?'' But not all theorists are so forthcoming. More usually, critics of pure theory are dismissed as failing to understand the point. I myself confess to such a failing. If "theory" constructions are acknowledged as neither expressing nor approximating reality, and are not counterfactual scenarios that express real possibilities I do find it difficult to see their purpose.

A problem also arises from the contributions of many econometricians and others who work with economic data. Standard procedures in econometrics assume regularities of the "whenever event x then event y" form (or a variant that can be covered under a well-defined probability "law"). Implicitly, it is supposed that such regularities are everywhere to be found. This assumption is rarely questioned, despite limited success in uncovering enduring regularities of this sort. Indeed, outside astronomy, most such regularities are produced in conditions of experimental control, conditions which hardly seem available to the social realm.

Moreover, econometricians often interpret themselves as explaining data, or modelling "data generation processes", rather than analysing the events the data are designed to express or record. To conflate data and their objects is itself to discourage explicit investigation into the nature of reality, and so considered reflection upon proper investigatory procedures.

Of course some economists, albeit a seemingly small minority, are more alive to the situation that knowledge claims, including data, are never a straightforward mirror of reality. This minority is also prepared to criticise traditional economic procedures. But even among this group there has been a tendency to over-react, to suppose a dualistic separation between knowledge claims and their object, so that all we have is mere construction. It is easy enough to see that any such position, certainly as usually formulated, is untenable. For if everything is mere construction there is no basis for criticism, description or progress. Yet those who adopt this sort of position are regularly finding texts to deconstruct or practices to describe. Such stances presuppose a distinction between the contributions of the constructivists themselves and the equally real items they acknowledge in describing them. To fail to recognise this is to commit the same fallacy as is involved in reducing events to the data designed to express them.

Most contemporary mainstream economists veer towards one of the positions described above. In each scenario insights into the nature of reality are neglected in the process of determining the investigatory approach to take. In consequence, it is not surprising that economics is so widely perceived to be both failing and pointless.

The neglect of reality in modern economics is a situation that many economists have seemed content to live with. However, there are now mechanisms in place, perhaps for the first time, that may stimulate a change of orientation. This may be the occasion for academic economists to recover reality.

Until recently those who wished to engage in the academic study of aspects of economic activity had little option but to enter economics departments. This is no longer the case. Nowadays, possibilities for studying economic life as it occurs are increasingly being offered elsewhere, particularly in departments of management studies or business schools, but also in departments of human geography, history, sociology, land economy, accounting, and so on. Reports are appearing in the UK, Australia and especially North America of a significant decline in enrolments on economics courses (both at school and university) and a correspondingly increased take up in alternatives of the sort listed above. And there is evidence that the perceived irrelevance of formalistic economics looms large in the explanation. My own experience of teaching at an international summer school oriented towards a more realistic economics is that, of the more than 60 postgraduate students, lecturers and researchers attending, the overwhelming majority were, largely for this reason, in faculties other than economics.

In short, there has long been a need for economists to recover reality. This need has rested on the concern for relevance and understanding, and so on the implications of the discipline for society at large. But to this concern must now be added academic economists' self-interest. For if those of us in economics faculties do not soon move to recover reality, at least in our teaching, we may eventually find that we have lost most of our best students and researchers to other academic departments. In other words, we may find that we need to recover our status as a viable academic discipline as well.

Tony Lawson is a lecturer in economics at Cambridge University. This commentary draws upon the analysis in his book Economics and Reality, published this month by Routledge, Pounds 16.99.