“There are several reasons why the economy is not amenable to simple modelling. Or rather why, when it is subjected to simple modelling – by economists – the results should be treated with a judicious degree of caution. To be specific, the results should be interpreted as applying to the particular model, not necessarily to the economy.
First, there are a huge number of economic – and non-economic – factors continually interacting. Many of these interactions are two-way – where one factor will change, causing other things to change, and then these changes themselves affect the initial factor – causing a new cycle of interaction, and so on.
Second, the nature of the causal mechanisms themselves alters over time.
Third, one may sometimes find a causal mechanism from one variable to another that appears absolutely stable, so one can predict that if a certain event happens it will always be followed by the same consequence. But if something new is introduced – for example a policy intervention suggested by the economist who has discovered this stable relation – then that may well cause the previously stable correlation to break down. This may sound rather esoteric, but it became crucially important when, for example, policy makers accepted the claim that there was stable relationship between money and prices, so that if the money supply was controlled, inflation could be eliminated painlessly. In the event, as governments clamped down on the money supply, people just used the existing money stock more intensively. This hadn’t happened before, but it did now. This phenomenon – that intervening on the basis of past behaviour can actually change that behaviour in the future, thus undermining the intervention – came to be known as ‘Goodhart’s Law’. (After Charles Goodhart, an economics professor at the London School of Economics and a member of the Bank of England’s Monetary Policy Committee, who had pointed out that this was indeed likely to be the result of the Conservative Government’s monetarist policies in the 1980s.)
Fourth, many of these ‘laws’, causal mechanisms, call them what you will, depend on what decisions actually come to be taken by various people in the economy (and, indeed, in other economies). So an economist might predict that since a fall in interest rates will reduce the cost of borrowing to make an investment, investment will therefore increase if the interest rate is reduced. And so, often, it does. But if, on the other hand, investors decide not to invest after all, say because of their uncertainty about the future, then that’s that. And if the decision is a general one, then our economist will be proved wrong. End of story. This is why Keynes described ‘animal spirits’ as playing a part in investors’ decisions (meaning their business and financial instincts). Indeed, the terms ‘bear’ and ‘bull’ market are used to describe the stock exchange when it’s on the way down or up, respectively.
And fifth, many of the things that economists are analysing – what the effects of changing interest rates will be, or which will be the richest ten countries in five, ten or twenty years time – are simply unknowable. The answer will depend on what happens to a whole range of other factors, about which we can’t be sure.
These difficulties are not, in our view, what put people off the subject. But they do lie behind one of the off-putting factors, which is, rather, that too many economists are either unaware of, or else forget, the above. They talk as if the results of simple models can be translated directly into policy for the real world. Students often object to this – rightly – saying that ‘economics is completely unrealistic’. The problem, though, is not economics – it is the misuse of economics.”
Michael Kitson and Jonathan Michie (2000), The Political Economy of Competitiveness, Routledge, p.3-4.