July 2013 newsletter - The rediscovery of Classical economics
01 Jul 2013
In recent issues the Newsletter has published a number of articles on what passes for ‘good’ economics and what students should be taught, in the light of the recent crisis.* In this article, David Simpson makes a case for replacing ‘equilibrium economics’ with a more classically focused approach.1 We leave it to readers to note the similarities in some of David’s recommendations to the suggestions made by the Bank of England — Government Economic Service steering group, summarised by Diane Coyle in our April issue.
In this essay, I shall put forward an alternative to equilibrium economics, the paradigm that has dominated the mainstream of economic thought for the best part of a century.2 The progressive refinement of that theory during the twentieth century delivered determinacy of solutions together with simplicity of structural form, but the costs of this achievement have been substantial. In order to achieve the desired simplicity of formal structure for what is essentially a mechanistic metaphor, most of the essential ingredients of a market economy have been removed. All elements of human behaviour, most institutional arrangements, change of any kind that comes from within the system, together with the possibility of increasing returns or path dependency have disappeared. It is difficult to relate equilibrium theory to the empirical processes of an actual market economy, and so it provides us with a poor understanding of how the contemporary economy actually works.
Equilibrium theory and policymaking
This has had implications for policymaking. The hubristic belief that the business cycle had been conquered by recent developments in theory and practice must surely in part be attributable to the displacement from the academic curriculum half a century ago of the study of business cycles by macroeconomics, itself a version of static equilibrium theory. Likewise, if the study of the processes of economic growth had not been abandoned at around the same time in favour of modelling ‘equilibrium growth’, (an oxymoron if ever there was one), perhaps policymakers might not be quite as helpless as they seem to have been recently in the face of political demands for the restoration of economic growth.
While equilibrium theory focuses our attention on issues surrounding the efficient allocation of a given set of resources amongst competing uses at a single moment of time, it does not address those features of a market economy that have principally impressed themselves on human history. These include the ability to sustain growth in aggregate productivity over long periods of time, periodic fluctuations in total output and employment, the existence and behaviour of markets themselves and the incessant change in the range and quality of both producer and consumer goods and services.
It is difficult to exaggerate the inappropriateness of using the concept of ‘equilibrium’ to try to analyse a market economy. A market economy is never at a state of rest. It is essentially restless, as Marshall understood.3 So what is the alternative method of analysis? The alternative is what may be called classical economics.
Defining ‘classical economics’
By ‘classical economics’ I do not mean that largely discredited body of doctrines, including the labour theory of value, the wages fund doctrine and ‘laissez-faire’, that faded away towards the end of the nineteenth century. I refer to an intellectual tradition that began with Adam Smith, was continued by Marx, Menger and Marshall, Schumpeter and Hayek and in the present day is represented by theorists of complexity.
The hallmarks of this classical tradition are principally three. The first is the belief that the growth of the economy, rather than relative prices, should be the principal object of analysis. Coupled with that belief is an understanding of the market economy as a collection of processes of continuing change rather than as a structure, and that the nature of this change is self-organising and evolutionary. Finally there is a conviction that economic activity is rooted in human nature and the interaction of individual human beings.
The differences between classical theory and equilibrium theory can be summarised in the following terms. Classical theory focuses on change and growth within open, dynamic nonlinear systems that are normally far from equilibrium. Equilibrium theory, on the other hand, analyses the theory of value within closed, static linear systems that are always in equilibrium. As to the essential nature of economic activity, classical economics makes no distinction between micro- and macroeconomics. Patterns of activity at the macro level emerge from interactions at the micro level. Evolutionary processes provide the economy with novelty, and are responsible for its growth in complexity. In equilibrium theory micro-and macroeconomics remain separate disciplines, and there is no endogenous mechanism for the creation of novelty or growth.
The behaviour of human beings in classical theory is analysed individually. People typically have incomplete information that is subject to errors and biases, and they use inductive rules of thumb to make decisions and to adapt over time. Their interactions also change over time as they learn from experience. In equilibrium theory, individual behaviour is assumed to be homogeneous and can be modelled collectively. It is assumed that humans are able to make decisions using difficult deductive calculations, that they have complete information about the present and the future, that they make no mistakes and have no biases, and therefore have no need for adaptation or learning.4
Carl Menger provides the principal link between the earlier classical school and their twentieth century successors. Menger’s name is frequently associated with that of Jevons and Walras, each having independently formulated the principle of the determination of price by marginal utility, but his analysis needs to be carefully distinguished from theirs. Whereas Jevons and Walras drew their methodological inspiration from classical mechanics, Menger had little interest in the concept of equilibrium. In addition to his emphasis on the subjective basis of economic activity, he made two other noteworthy contributions to economic theory. He noticed the emergence of social institutions as a consequence of what we should now call self-organising evolutionary processes. He also proposed an amended version of Smith’s theory of the progressive division of labour, one that foreshadowed Allyn Young’s later contribution.5 It was left to Hayek to build the intellectual bridge from Menger’s work to the theory of complexity.
Classical economics and complexity theorising
Following the publication in 1952 of his study of the mind, Sensory Order, Hayek began to cite works in cybernetics and systems theory. In 1964 he published his paper ‘The Theory of Complex Phenomena’, although that did not contain a fully worked out theory. His linking of the concept of what he called a ‘spontaneous order’ with the idea of a complex system provided him with a vehicle to re-organise his earlier arguments about methodology, He devised the concept of a hierarchy of complex phenomena to provide a common framework for the analysis of different classes of phenomena. At some time in the 1970s he is thought to have communicated with Ilya Prigogine, the Nobel Prize-winning chemist who was then becoming aware of the ubiquity of self-organising systems in natural phenomena, and pioneered the development of nonlinear dynamic systems to analyse them.6
By the 1980s complex adaptive systems had come to be regarded as forming a universal class in the natural world, with many common behaviours observed across traditional disciplines. In 1984 a number of natural scientists set up an Institute in Santa Fe, New Mexico to act as a centre for interdisciplinary research into the properties of such systems. One or two interested economists like Kenneth Arrow and Brian Arthur joined them. Unaware of Hayek’s earlier work, but following the lead of his scientific colleagues, Arthur began to model economies as complex adaptive systems. In an interview he gave in 1996, Arthur is quoted as saying:
Right after we published our first findings we started getting letters from all over the country saying, ‘You know, all you guys have done is rediscover Austrian economics’. I admit I wasn’t familiar with Hayek and von Mises at the time. But now that I've read them, I can see that this is essentially true.7
The congruence of Austrian economics and complexity theorising is indeed remarkable:
• Austrians see market institutions as ‘spontaneous orders’ that have emerged from the self-organising processes of the economy. What Austrians called ‘spontaneous orders’ correspond to the aggregate patterns that emerge from complex adaptive systems.
• Both see economic systems as dynamic processes involving direct interactions between individuals. The fundamental Austrian principle of subjectivism insists upon an economic analysis that looks at things from the perspective of the individual human being. The corresponding methodological principle in complexity theory is agent-based reasoning.
• Austrians and complexity theorists recognise that agents may be heterogeneous in their objectives and in their behaviour. Both tend to model agents as rule-followers. Although Austrians believe that human beings act purposefully, they may follow rules of thumb to achieve their objectives. In both systems of thought, agents adapt their behaviour as a result of their interaction with one another.
• The Austrian principle of ‘verstehende’ or ‘understanding’ claims that one cannot describe human action without reference to human meanings. In other words, human action cannot be fully explained in terms of physical laws alone. The same point can be expressed in the language of algorithmic information theory.8
Algorithmic information theory also helps us to understand how limited is our capacity to predict the behaviour of complex systems, a favourite theme of Hayek’s. When we are dealing with complex phenomena, says Hayek, we can generate pattern predictions telling us that in certain general conditions, a pattern of a certain kind will appear. On the other hand, specific predictions are normally out of reach.9
Then there is the mutual recognition of ‘bounded rationality’, or the limits to human knowledge and powers of cognition and the computing power of the human brain. Recognition of these limits is a long-standing principle in the classical tradition. It formed the basis of the arguments deployed by Mises and Hayek against the possibility of effective central planning in the socialist calculation debate of the 1930s.
So it is the theorists of complexity, not the neoclassical school, who are the rightful contemporary inheritors of the classical tradition in economic thought. Some people might suppose from the similarity of their names that the ‘neoclassical school’ is close to the ‘classical’. In fact, they are more nearly exact opposites.10
The implications for the teaching of economics
First of all, courses in economic history and in the history of economic thought should be a required part of the curriculum for every student of economics. The study of economic history, like the study of complex systems, reveals the importance of context in understanding economic behaviour. The study of the development of economic thought helps us to appreciate the weaknesses as well as the strengths of a theory. Macroeconomics should be downgraded, and give way to the study of business cycles.
Secondly, more space should be found for the analysis of dynamic processes at the expense of static theory. The study of the processes of economic growth should be restored to centre stage. This probably means a greater emphasis on nonlinear algebra.
At the same time, the limitations inherent in applying mathematics to economics need to be acknowledged. The importance of the human factor and of human institutions in economic activity means that more attention needs to be given to non-quantitative methods of analysis.
* These include 'Teaching Economics after the Crisis'; 'Teaching Evidence-based Economics'; 'What's the Use of Economics'; 'Economics Education after the Crisis'.
1. The ideas in the article are elaborated in more detail in David Simpson (2013), The Rediscovery of Classical Economics, Cheltenham, UK and Northampton, MA, USA: Edward Elgar. David Simpson was Professor of Economics at the University of Strathclyde from 1975 to 1989, and thereafter Economic Adviser to Standard Life.
2. Within the term ‘equilibrium economics’ I include Neoclassical and neo-Keynesian theories, and any others that use the static equilibrium framework of analysis.
3. Metcalfe, J S (2006), ‘Evolutionary economics’ in T Rafaelli, G. Becattini and M. Dardi (eds), The Elgar Companion to Alfred Marshall, Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing, p.651.
4. Beinhocker, E D (2006), The Origin of Wealth, London: Random House, p. 97.
7. Cited in Vaughn, K I (1999), ‘Hayek’s theory of the market order as an instance of the theory of complex adaptive systems’, Journal des Economistes et des Etudes Humaines, 9 (2/3), pp.241-56.
8. Koppl, R (2009), ‘Complexity and Austrian economics’ in J B Rosser Jr (ed), Handbook of Research on Complexity, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, p. 405.
9. Hayek, F A (1967), ‘The theory of complex phenomena’, in Studies in Politics, Philosophy and Economics, London: Routledge and Kegan Paul, pp. 22-42.
10. ‘Hardly an author can be found, not even Keynes himself, who is so much the exact antipode of Milton Friedman in every part of the economist’s theoretical vision as Carl Menger.’ Streissler, E and W. Weber (1973), ‘The Menger tradition’, in J R Hicks and W Weber (eds), Carl Menger and the Austrian School of Economics, Oxford: Clarendon Press.