This short speech was to open a session on the subject at Te Papa on Thursday July 3, 2007. (revised)
Keywords: History of Ideas, Methodology & Philosophy;
Is economics a science? In a sound bite, the answer is ‘yes’ and ‘no’. If you rely on the media for your knowledge of economics (and everything else), then I have given all you want. You are welcome to go. Those who remain will have to deal the complexities of the real world, in which sound bites are not enough.
The word ‘science’ has a number of meanings. Frequently is used to mean a body of knowledge, as in the phrase ‘Marxism is the only true science’. How you tell whether such bodies of knowledge are true, is something of a puzzle – to me anyway. Since economics is a body of knowledge, it is in this sense a ‘science’.But there are many others.
However this session takes place in the Museum of New Zealand and is sponsored by the Royal Society of New Zealand, so in this context ‘science’ refers to a narrower, more specific notion. It is a body of knowledge which develops on the basis of the principles of the scientific method. What do we mean by scientific method.
I did my science degree at the University of Canterbury, where the thinking of Karl Popper loomed large, even though he had left twenty years earlier. It is his analysis which seems to me to be the best account of those principles of scientific method.
At the heart of Popper’s account is that a science is a set of propositions about the objective world, accumulated through the ages, which could be wrong, and are subject to empirical investigation. While thus far they have not been replaced, they are subject to an ongoing critical review with the  possibility that there might be a better set of theories. The Popperian scientist has to therefore humbly accept that her or his current set of theories may be proven wrong, and indeed the scientist has a duty to challenge them in order to improve them. Yet that humility is tempered by the enormous pride the scientist has in the achievements of science, the progress it has made over the centuries, and its contribution to understanding and to the welfare of humanity.
Such is that achievement of science in this narrow sense, many bodies of knowledge which are not organised on scientific principles like to call themselves ‘science’ to attract the prestige that the sciences have gained from their successes. Hence the ambiguity of the term in everyday use.
Where does that place the body of knowledge which we call economics? What I want to argue is that within economics, there is a component which is truly a science with a development subject to scientific method. But this is not true for all of economics, nor do all economists practice the critical process of the scientific method, even to those parts of their discipline which can be properly labelled ‘science’ in the narrow sense.
Perhaps there is no other body of knowledge – psychology possibly excepting – where this tensions between its scientific foundations and the rest of the body of knowledge is so profound.  Moreover, economics is a science dealing with very complex systems. The only challenger there might be ecology.
A second major problem is that economics cannot be an experiment based science. There are ethical problems, it would be incredibly expensive, and impractical – how could you run separate experiments on the same economy, even if you were allowed and could afford it? There are some economic experiments but they cover only limited parts of the subject. However, astronomy, biology, and geology are accepted sciences even though they are not primarily experimental. They have evolved non-experimental methods for investigating the real world. So has economics. (The statistician in me knows that econometricians have thought more about some aspects of non-experimental inference than anyone else in statistics.)
So in principle we can have a scientific economics, that is a subject which investigates critically economies in the objective world economy according to scientific principles. But there is also a part of the economic body of knowledge which is not scientific. Sometimes it overwhelms scientific economics.
Perhaps the problem occurs because people want to apply economics to social issues. We dont separate these activities out as we do for the applied science of engineering from the pure scientists of physics. Applying economics to questions of public policy is often thought more important than investigating the empirical phenomenon. A few years back I was talking to an ex-Treasury economist about the economic issues underpinning the broadcasting reforms. My colleague impatiently cut to the quick. ‘That all aside, Brian, you do agree with the reforms dont you?’.
Or consider a recent paper by an American economist who argued that since most people disagreed with his policy conclusions, they must be stupid and irrational. As it happens his policy conclusions involve assumptions which may or may not apply. It is perfectly rational to make other assumptions: others do.
Here one is moving from ‘positive’ economics – the way the world is – to ‘normative’ economics of what it ought to be. As the great eighteenth century philosopher (and also an important contributor to economics) David Hume argued you cant get from the empirical world to the moral world without inserting some moral assumption. Getting an ‘ought’ from an ‘is’ is like getting blood from a stone. Yet people try.
I am struck how in an earlier age, some economists would have been comfortable as preachers, relying on the ‘Good Book’ – a body of knowledge interpreted via the non-scientific principles of faith – to promise us damnation if we do not follow the preacher’s path to holiness.
Such an approach distorts judgements about the real world. Recently I was at a seminar in which a young economist stated a proposition, and the presenter, a very able research economist – a scientist – innocently asked whether he had any evidence for it. The reply, tediously long, was about the young economist’s (not very interesting) opinions on what he thought were the policy implications of the proposition. When cut off by asking whether he had any empirical evidence, he shamefacedly said he did not. He had been arguing that an empirical proposition was true because it conformed to his political and policy preferences. The test of truth did not involve scientific principles, but the ideological one of that it conformed to a preset view of the world.
One consequence of this approach is the quality of an economist’s scientific work becomes judged by whether it is consistent with the policies of those in power. At this point the scientific content of the body of knowledge of economics collapses. Some of its propositions may be empirically true, but they are no longer evaluated by scientific means.
Sadly, perhaps as a result, those economists you are most likely to hear in the media are these ideologues (or perhaps those who are paid by their employers to represent a particular view). Journalists do not have time, nor generally the expertise, to allow a economist to explain the uncertainties of their science.
Keynes famously remarked, that ‘it is better to be vaguely right, than precisely wrong’ The media prefer the opposite. The public is presented with economics as a certain body of knowledge dominated by ideology rather than science.
This presentation has not been a sound bite. Its content is larger than an hour of television news stripped of its adverts, the weather, sport, jocularity and trivia. Little enough of any residual is devoted to economics anyway. Yet I have hardly introduced the subject.
Much of what I have said tonight has been very abstract, without sufficient practical illustrations. After we have had coffee, I will answer questions which will discriminate between economics as ideology and economics as science. I hope to surprise you by demonstrating that there is a real scientific element in economics – yet too often we ignore it in favour of the sound bite of ideology.
I am grateful to Robert Nola, for some very useful comments on an earlier version of this paper.
Go to top