Let’s forget the physics thing
The Financial Times features a discussion about economics between Martin Wolf and Gideon Rachman. It isn’t particularly lively, or a debate, as the FT bills it; more a friendly chat about what’s wrong with the dismal discipline.
Rachman says he thinks economics might not be a science and that its mathematical modelling should probably be replaced by the sort of empiricist trial-and-error that historians use. Wolf broadly agrees but says that we need a framework to talk about the economy, and he seems to imply that formal models are probably necessary.
Evolutionary biology, Wolf says, is a better approach than physicists for the sort of things that economists do. We can’t predict how the economy is going to evolve: “Let’s forget the physics thing, ” he says.
This is a radical view, perhaps more radical than might be suggested by the rest of Wolf’s work. Neoclassical economics, the dominant paradigm of modern mainstream economics, arose in the late 19th century directly from a desire to follow the scientific method, specifically the newly-emerging techniques of physicists. Metaphors like equilibrium, so central to microeconomics, are borrowed from physics. Jevons, Walras and Menger looked for a way of analysing society that produced universal, timeless truths, just like physics did.
The evolutionary approach of Thorstein Veblen explicitly confronted the neoclassical view, suggesting that more could be learnt from the methods of biology than from physics. Veblen lost. The neoclassicists won. Veblen is hardly taught any more, and the vast majority of mainstream contemporary academic articles spring from neoclassical principles rather than evolutionary economics.
So it is interesting that Wolf, who is decidely mainstream, claims to be jettisoning the physics metaphor in favour of something that sounds like Veblen.
Wolf argues that one reason for not abandoning models is that we can’t rely on experience alone to analyse the economy because experience constantly changes. We need some sort of analytical framework as a guide for policy, and for this reason, he thinks, we still need to assume that people are broadly rational, self-interested utility maximisers.
Wolf is right that we need some sort of framework for analysis, but we could easily throw out the fanatical formalism of modern economics whilst retaining an analytical framework. We could forego the assumptions of rational, self-interested utility maximisation and remain scientific.
Plenty of economists have put forward alternatives to these neoclassical obsessions, including Keynes, the Austrian school, Dierdre McCloskey and Tony Lawson. These economists all promote the use of an analytical framework, even if in different ways.
Of course economics should be a science, but science doesn’t mean reliance on highly mathematical analytical models that pretend to establish truths that apply forever. It should be quite possible to use the empiricist trial-and-error techniques of historians whilst maintaining some sort of overall framework for analysis. Neoclassical economics is only one small part of the broad tradition of economics, and it doesn’t have a monopoly on scientism.