Don’t shoot the economists
- The Death of Economics by Paul Ormerod
Faber, 230 pp, £14.99, March 1994, ISBN 0 571 17125 7
What are we to make of a book entitled The Death of Economics, the second half of which is called ‘Towards the Future of Economics’? Paul Ormerod’s book has already provoked excited reviews from people thrilled to have their prejudices against economists and economic forecasters confirmed by one of the profession, an economic forecaster himself for ten years, who proclaims in his Preface that conventional economics is ‘deeply flawed’, and that its prescriptions for the problems of inflation and unemployment are ‘at best misleading and at worst dangerously wrong’.
One bad reason why economists are unpopular and economics is widely regarded with distaste and distrust might be that economics is about cost – not mere financial cost but the underlying cost of decisions in terms of other possibilities foregone. ‘There’s no such thing as a free lunch’ is quintessentially an economist’s observation – and it isn’t endearing. Keynes’s remark that ‘in the long run we are all dead’ speaks more to our hedonistic hearts but, significantly, he has been taken to task for it by much of the economic profession ever since. There is a sense in which economists are unlikely to be popular even when – or especially when – they are practising their subject well. In the same way a good finance director should often irritate a good company board; and everyone should have a bad word for the Treasury most of the time, if it is doing its job. Another bad reason for disliking economics is that economic reasoning, because it is concerned with long-term general consequences rather than immediate effects, often leads to conclusions at variance with common sense. In fact there is a much greater degree of agreement among the profession on appropriate ways of analysing economic problems than is generally recognised.
None of this, however, even begins to explain the present attitude to the profession. Economists have brought on themselves an intensity of distaste well beyond the call of duty. There has been, over the past thirty or forty years, a quite monstrous trahison des clercs; and over the past twenty, some dramatic failures of economic prescriptions and policies. Perhaps the most fundamental way in which society has been let down – well identified by Ormerod in the first half of his book – has been through the prostration of academic economists before the false god of mathematics. In March 1901, in a letter to the statistician A.L. Bowley, the great Alfred Marshall wrote: ‘In my view every economic fact, whether or not it is of such nature as to be expressed in numbers, stands in relation as cause and effect to many other facts: and since it never happens that all of them can be expressed in numbers, the application of exact mathematical methods to those which can is nearly always a waste of time, while in the large majority of cases it is positively misleading.’
These wise words derive particular force from the fact that Marshall, like Keynes who felt similarly, had his original training in mathematics and only came to economics – through a concern for society and its problems – after obtaining a degree in the Cambridge mathematics tripos. After nearly a hundred years Marshall’s dictum remains a necessary but now largely unheeded prophylactic against the idea that the scientific, the difficult, the grownup part of economics is to construct a mathematical theory which, on certain assumptions, will explain a number (ideally all) of the known economic variables.
Exactly the opposite is true. The hard part of economics is bringing intelligence and judgment unremittingly to bear on the mass of intractable data – much of it inherently unquantifiable – which can neither be encapsulated in a model nor assumed away. One has to espouse the scientific spirit while accepting that the subject can never be a true science. Not only is there no possibility of controlled or repeatable experiments, there is the further difficulty that no recommendation can ever be value-free. Every economic policy or decision will affect different people in different ways; and even if one could be absolutely certain what all the effects would be, to believe one could rank the desirability of outcomes in an objective way is still a deeply fallacious piece of utilitarianism. If it is hard for even thoughtful economists to live with this moral relativism, it is impossible for most politicians, who are apt to believe that There Is No Alternative.
These are fairly longstanding difficulties. In the past twenty-odd years things have got a lot worse. The quintupling of oil prices at the start of the Seventies, together with the break-up of the fixed exchange rate system, delivered a shock to the world economy unequalled in peace-time this century, except for the 1929 Crash. Policy-makers in most countries faced the twin possibilities of cost inflation and demand deflation, both potentially very severe. Roughly speaking, they opted for the first alternative in the Seventies and, when levels of inflation had reached seriously unacceptable levels, for the second in the Eighties.
The evident deterioration in economic performance in the Seventies caused serious disillusion with all previous approaches to economic policy. The deep yearning for simple solutions, always and everywhere just below the surface, overwhelmed politicians, opinion-formers and bien-pensants generally and led to the grotesque popularity of naive monetarism (and pretty naive fiscal policy) in the late Seventies and early Eighties. A lot of economists who should have known better allowed policy-makers to believe that the inflationary dragon could be slain without more than a very temporary blip in interest rates or unemployment: an outcome that would be achieved simply by controlling a single, comfortably abstract variable – the figure representing the nation’s money supply.
The result in the UK was a very sharp recession followed by, in sequence: the abandonment of monetarism; an unsustainable boom fuelled by borrowing; the adoption of another simple cure-all policy – viz, fixing the exchange rate with Europe for ever; another serious recession; and the forced abandonment of the second nostrum. For good measure, official Treasury forecasters failed to predict the current recession, even while we were plunging into it; and confidently asserted that if we came out of the ERM interest rates would have to rise.
Ormerod’s main target is the Thatcherite devotion to a competitive market-driven view of the economy, as encapsulated in her famous dictum ‘There is no such thing as society.’ He mounts a powerful attack on the neo-classical equilibrium model developed by Walras in the mid-19th century, and much used since as the theoretical basis for the view that completely free competition in all markets will produce an overall optimum. Ormerod rightly does not rest his case on the difficulty in practice of achieving perfect market structure, perfect information etc. More formidably, he cites theoretical work showing the difficulty in principle, even in an ideal world, of specifying appropriate sets of prices; and the theoretical indeterminacy of the outcome if any part of the system should not meet the necessary criteria.
A good demolition job then, but a curiously blinkered one. Ever since it came to be realised early in this century that increasing returns were not only possible but likely in many forms of production, a very great body of work – in this country particularly associated with Cambridge – was produced attacking the notion of an ideal competitive equilibrium. But Ormerod hardly acknowledges that. He is also strangely determined to show that the whole approach of those who followed Walras and developed the marginal equilibrium theory was a priori and deductive; they started with a theory and if the facts didn’t fit, so much the worse for the facts. In this he is concerned to contrast the neo-classicals with the early economists – Smith, Ricardo and Marx – who, says Ormerod, were concerned with how economies grew, who confronted the evidence before them and who were more concerned with the moral and social framework of the economy than with how resources could be deployed most efficiently. Ormerod is extremely keen to acquit Adam Smith of any responsibility for Margaret Thatcher. This is all fairly odd. Marshall (barely mentioned by Ormerod), who did more than anyone else to develop the marginal approach, was deeply concerned to collect and analyse the facts about industry and the economy with a view to improving economic welfare. And The Wealth of Nations is full of (very good) arguments for competition.
What Ormerod wants to do, it is clear, is lump all economists since his great trio of Smith, Ricardo and Marx into an all-embracing ‘equilibrium analysis’ camp. So keen is he to do this that he barely distinguishes monetarists from Keynesians. But notwithstanding the fact that many models can and do incorporate both ‘Keynesian’ and ‘monetarist’ relationships, no understanding of the events of the past twenty years is remotely possible if the two views are simply elided. A good example of his approach is provided by his remark on the first page of Chapter One that ‘in Western Europe the economics profession eulogised the Exchange Rate Mechanism and monetary union.’ This is breathtakingly false. Some UK economists saw value in ERM membership for the UK – though most of them would have specified fairly carefully the conditions they considered necessary for its success. But many including in particular those free-market adherents whom Ormerod so despises (for example, Alan Walters), were from start to finish opposed to the ERM and predicted – even over-predicted – the trouble that would arise from it.
Ormerod’s dissatisfaction with what he regards as present-day economics centres on the general failure to understand, predict and alleviate unemployment. He is right to be concerned about this and about the irritating way in which many economists continue to believe in a relationship between inflation and unemployment, expressed as the Non-Accelerating Inflation Rate of Unemployment – even though this entity, like one of the particles of modern physics, is never actually observed and seems continually and unpredictably to change its value. Ormerod’s observation of the changing relationship between unemployment and inflation over time leads him to his major thesis, his start for a new economics. It is that relationships in a complex economy are likely to be non-linear: a perception which would put economics alongside meteorology and epidemiology. One feature of non-linear models is that the relationships between variables depend on the starting point: thus, the effects of a change in interest rates will differ according to whether it occurs in recession or boom. This very plausible idea leads Ormerod to lay great emphasis on shocks to the economy after which things do not return – even allowing for a time-lag – to their original positions or relationships. The oil price shock was one such. The financial deregulation of the Eighties another.
But if this is an interesting thought, what is one to do with it? Here, suddenly, Ormerod goes over to the enemy. He says that ‘a striking characteristic of non-linear analysis... is that the essential dynamics of the phenomena being investigated can often be captured in a small number of well-chosen equations.’ He then develops what he calls an illustrative model of the UK economy, with which he sets out to predict unemployment. It consists of just three equations (one of which looks suspiciously like an identity). To solve them for the post-war UK, one needs to know, among other things, the capital/output ratio of the economy in 1948; the rate of depreciation of the capital stock; and the proportion of profits invested. These are extraordinarily slippery concepts, highly abstract and virtually incapable of non-controversial measurement.
So we are back with simple-minded abstract models of the economy. We are also, it turns out, back with simple policy prescriptions. It is no use, Ormerod says, for policy-makers to rely on conventional macro-economic tools to reduce unemployment. (Nothing in his book remotely supports this: the most his arguments suggest is that conventional policy will not be able to bring unemployment back to pre-Seventies levels – but we are at present way above these.) The social framework must be changed. And he has some suggestions about what to do – none, it should be said, original, and none supported by serious argument. Work-sharing ‘is a policy whose time has come’. We should imitate the higher levels of ‘social cohesion’ in Germany and Japan and the greater protection their firms enjoy from hostile takeovers. Generally, as one would expect, he urges revolt against market disciplines, while adding, rather surprisingly: ‘of course this does not mean that anti-competitive policies are always sensible.’
Sadly, despite much excellent castigation of error, some fruitful insights and a welcome despatching of much Thatcherite nonsense, this book turns out to be at bottom another – quixotic – addition to the long line of simple prescriptions for the already seriously overdosed British economy. But we don’t need simple prescriptions. We need complex prescriptions. It is no use shooting the economists. When you disagree with an economist you are arguing as one. The point is to better than he is.