Evolutionary economics and the competition between scientific paradigms

Inspired by a recent article on the modern development of evolutionary and institutional economics (Hodgson 2007), what follows is an attempt to classify some of the main areas within what has been called evolutionary economics, and to say something about how economists, philosophers, and social scientists have influenced one another's thinking.

The study of economics has two objectives; first, to develop theory to attempt to explain and predict human economic behaviour (economic theory), secondly to provide economic actors or agents with tools enabling them to conduct business and public operations more efficiently (applied fields). Of these, the second is the less problematic. The discipline of economics is continually providing economic agents with practical working tools to enhance organizational performance and efficiency. Much of this is done under the heading of management, and in close collaboration with practicing businesspeople. It is the former objective which is a cause for concern. The larger methodological question is what basis we can found the discipline of economics on, to give its models predictive power. Are there any such models?

The choice of physics as a model for the development of economic theory, a methodological direction which has been particularly dominant since the Second World War, has increasingly been criticized by economists, and not only by evolutionary economists, but by members of a variety of schools. Many of these critics see biology as an alternative methodological direction that merits investigation. Modelling economics on biology is not a novel idea; it is an attempt to revisit a number of questions which were left behind at the turn of the twentieth century. Thus the fundamental question is whether the concept of evolutionary economics was abandoned prematurely, or for good reasons.

The French philosopher and mathematician René Descartes inspired two lines of scientific thought. One was abstract, mathematical, and mechanistic; it led to significant advances in knowledge thanks to men like Leibniz and Newton96. The other approach explored the development of our living world with everything in it, from insects to animals. This second approach was taken forward by men like Buffon (1749), Lamarck (1809), Cuvier (1812), Wallace (1876), Darwin (1872), and Wegener (1915). In these terms we can say that evolutionary economists are trying to show where the former line of thought falls short when applied to the understanding of economic behaviour, and where the second line may be of help.

Adam Smith (1776) is often used as a reference by the neoclassical or marginalist school of economic thought. We shall argue that Smith, Thomas Malthus, and Alfred Marshall (1890) were in fact all inclined towards the evolutionary approach. If that is so, it means that the neoclassicals are not so much "classical" as "neo". The "marginalist school", which is a better term for the neoclassicals, might also be called the "mechanical approach", as compared with the evolutionary approach. The marginalist school, or marginalism, studies marginal concepts in economics: problems related to marginal cost, marginal productivity, marginal utility, the law of diminishing rates of substitution, and the law of diminishing marginal utility. Marginal calculations were a natural direction to follow once the physics paradigm had been selected.

The evolutionary model is implicit in Marshall's Principles of Economics, even though he did not incorporate the idea into his more formal theories. That was part of the problem for evolutionary economists at the turn of the century: they had not succeeded in producing applicable theories and models, but mostly left their analyses on the descriptive level. So when it came to building a scientific platform on which the positivist study of economics could stand it was the French economist Léon Walras who was chosen. Walras and his successors had mathematicized the Newtonian system.97 They could offer the discipline of economics a rigorous methodology which promised to deliver elegant answers, all in the spirit of the natural sciences. The underlying assumption was that if this method had worked wonders for the natural sciences then it should do the same for the social sciences. In other words, their answers promised to be more precise than what economists had delivered before; and that promise was delivered. The fact that the new models and their predictions often failed to correspond to actual economic behaviour was mostly due to their assumptions. They were nevertheless far better than nothing (a point which continues to be a main argument for the marginalists), and hence the evolutionary perspective was gradually lost from the discipline of economics (Boulding 1981: 17). However, it soon became clear that the problem was no longer one of precision, but of relevance. In other words, the answers were detailed and elegant and might have been correct, but they did not correspond to the economic realities.

Later, with Paul Samuelson - whose models essentially involved stable parameters and a dynamics based on stable differences or differential equations - economics became even more Newtonian, less Darwinian (Boulding 1981: 84). If it were not that current economic theories have still not demonstrated themselves to be the relevant predictive tools that economists had hoped for, our scientific journey would probably have ended at that point. But it continues.

The best philosophical foundation for economic research seemed to many to be a renewal of utilitarianism. The rehabilitation of economic theory was due to the Austrian Carl Menger - known to students today for his theory of supply and demand. Menger's essential aim was to discover the laws determining prices and to initiate discussions of supply and demand, human needs and marginal utility (Schumpeter 1992: 84). The biggest flaw in his assumptions is that Man is not entirely hedonistic, his nature is not wholly fixed and predetermined:

He is not simply a bundle of desires that are to be saturated by being placed in the path of the forces of the environment, but rather a coherent structure of propensities and habits which seeks realisation and expression in an unfolding activity (Veblen 1898: 11).

Both Karl Marx and Menger were much influenced by Ricardo. Menger gave rise to what has today become mainstream economics, but that was not his original role. Menger was at one time the outsider, at a time when Marx and the German historical school led by Gustav von Schmoller represented the consensus within the discipline of economics.98Critique of the "mechanistic approach" is by no means new either. In his 1875 book The Character and Logical Method of Political Economy, the Irish classical economist John Elliott Cairnes disputed Jevons's idea that economic truths are discoverable through mathematical reasoning (op. cit.: vi). What can do is illustrate and simplify conclusions that have been reached by other methods, or in his words:

I have no desire to deny that it may be possible to employ geometrical diagrams or mathematical formulae for the purpose of exhibiting economic doctrines reached by other paths. (op. cit.: vii)

The reason why mathematics can have only limited application to economics is twofold. First, "its close affinity to the moral sciences brings it constantly into collision with moral feelings" (op. cit.: 3). The second is even more fundamental: maths is ultimately by nature just another language, even if of course much more precise than ordinary languages. But precision by itself does not help. In the same way as we do not solve a problem by translating it into a foreign language, maths by itself cannot solve economic problems. It can only express what is already there in a simpler and clearer form. Progress using maths in the social sciences only comes through our ability to see and handle ideas more easily. The advantage is the same that came from the development of symbolic logic. Both mathematics and symbolic logic are very helpful in summing up what we have already discovered, but we have to draw the inferences for ourselves.

Why has physics not provided a successful cornerstone for the social sciences? When we compare the results of the social sciences to those of the natural sciences, we find that social phenomena are more difficult to study, less tangible, less physically observable. Social systems are just too complex if we hope to pin down individual behaviour; they contain too many variables, with too many possible and often irrational outcomes, to be explained via physics and mathematics alone. More important, our mathematical approaches are not capable of treating the element of change - what is often referred to in the scientific literature as the dynamic aspect. Newtonian and Cartesian numerical mathematics, which has dominated the study of economics for a century now, is unsuitable for the more structural and topological relationships found in evolutionary systems, except insofar as the topological relationships can be mapped and converted into numerical relations (Boulding 1981: 86).

Economic theory as developed in the twentieth century builds on a number of mechanistic assumptions. These assumptions were first criticized by Herbert Spencer in his 2 volumes book "the principles of sociology" (In Peel, 1972: 6), who held that they must be wrong because "it assumes the character of mankind to be constant". Or put differently, the problem is that "existing humanity" does not exist, but is constantly changing. Change is the law of all things, true equally for a single object as for the entire universe; all things are mutable: shells into chalk, sand into stone. "Strange would it be, if, in the midst of this universal mutation, man alone was constant, unchangeable" (op. cit.: 7). Everything is in a state of continual change or fluctuation, even the things we think of as most stable. Dynasties and private fortunes seldom last more than a few centuries; even a stone monument has a limited life. We seem to have a cognitive difficulty with change, probably because we constantly need to find order in our everyday lives. We have a strong need to live and find our balance in the present, hence we prefer to think in terms of constants rather than of fluctuation. This seems to be the way we are born. In much the same way, we do not feel the earth speeding round the sun, and that is good: if we did, we would not be able to concentrate on anything else. In other words, we seem inclined to think in the linear terms of a static, mechanistic world perspective. Likewise, we think we can have knowledge of the future, but we cannot. Instead we are continually surprised; and to top it all we are not surprised that we are constantly surprised. Within rational choice theory we might define these observations as a set of rationality errors. They mark a biological limit to our understanding of the real world, i.e. of Kant's Ding an sich.

From the above one might take it that we are confronted with an either/or choice between marginalist and evolutionary approaches. To the extent that these premises are not contradictory, the method used should be whichever method has the strongest predictive power in each particular case of economic behaviour. It is not a question of either Newton and physics or Darwin and biology.101So far as we can tell to date, evolutionary economics is not necessarily, and not necessarily always, a replacement for neoclassical economics. For instance, it seems that the evolutionary approach is more suited for studying economic behaviour over the long term, when the element of change becomes most significant. There are many problems, e.g. of production that are simple enough for marginalist calculations to be of value, but they seldom include problems of international business.

To complicate the question further, in many cases marginalists and evolutionary economists will both espouse the same methods or theories. So for instance game theory is seen as a marginalist contribution by some, because it can be highly quantitative, but as an evolutionary approach by others, because it is dynamic and does not seek to maximize a given set of variables. Game theory can also be studied from either a mathematical or a non-mathematical perspective, as in the writings of von Neumann and Morgenstern (1944) on one side and Axelrod (1984) on the other.

In the remainder of this chapter we shall examine the development of the evolutionary approach to economics in more detail.

< Prev   CONTENTS   Next >