<<<	articles
It is time to jettison the forecasts Samuel Brittan Financial Times 21/12/07 More than 30 years ago, Denis Healey, a UK Labour chancellor of the exchequer, said he wanted to be to economic forecasters what the Boston Strangler was to door-to-door salesmen. Unfortunately, he did not succeed. Indeed, I have had an uneasy foreboding since the mainline economic forecasts at the beginning of 2007 evinced a remarkably favourable prospect. As recently as this summer, the editorial in the June Organisation for Economic Co-operation and Development Economic Outlook said: "The current economic situation is in many ways better than what we have experienced in years". I single out this forecast because it was one of the clearest and best-reasoned ones. The authors were well aware of the risks, and even had a "box" on subprime mortgage markets. Nevertheless, they decided to stick with their projections. Similar optimism was exhibited by the main official national forecasts. Such forecasts do not have the advantage of being always reliably wrong. They remind me of that mythical stockbroking firm that predicted the equity market better than its competitors. Then its run of good luck came to an end and friends of the firm asked what had happened. The answer was: "We had a partner whom we counted on to be always wrong and now he has retired." I have been emboldened to make these remarks partly by the scepticism shown in his memoirs by Alan Greenspan, the former US Federal Reserve chairman, who admitted only that forecasts performed a useful consistency check. He was always more interested in what was really going on. So much so, in fact, that journalists used to wait in mock amusement for what he said about indicators such as "freight-car loadings", which he felt offered important clues to the real economy. But I do not need to look at avowed heretics. The December OECD Economic Outlook contains an evaluation of the organisation's recent projections. The conclusion, unsurprisingly, is that they have not done too badly. Then comes the admission that "in general they have displayed a limited ability to anticipate business cycle turning points". But this is precisely when they are most needed. What is really wrong with the conventional forecasting process is unintentionally explained in the UK 2007 pre-Budget report. "The essential building blocks" of the forecast are estimates of the trend level and rate of growth of output and of the "output gap". This gap estimate, together with an assessment of the various components of activity, leads to an "informed judgment on the short-term path of the economy back to trend". Again, this does not matter if nothing much is happening, but will be very misleading if the normal rate of unemployment or surplus capacity, compatible with stable inflation, is changing, as it was so dramatically in the 1970s and less dramatically in the past decade and a half in the opposite direction. I am sorry to become a little technical. But the essence of the so-called monetarist counter-revolution did not depend on the predictive value of particular measures of money supply. Its main lesson, as I tried in vain to explain at the time, was the folly of trying to use monetary or fiscal policy to determine real variables such as output and employment. Lord Burns, long-time UK Treasury chief economist, tried to substitute the expression "nominal framework" for the new type policies. He meant that government authorities could try to influence the flow of expenditure in the economy, but could not determine how far any increase in the national income would be divided between beneficial increases in output and wasteful rises in inflation; and that was even before globalisation had become a dominant force. The Burns formulation did not catch on. Mainline forecasters have gone back to trying to predict real variables – indeed, they never really stopped – and have regarded any emphasis on monetary flows as a rightwing political aberration. When I allow such thoughts to escape, I am normally asked: "What is your alternative?" One of the UK Treasury's economically literate, but non-technical, officials, the late Sir Frank Figgures, despite his name wanted not forecasts but intelligence of what was going on in key parts of the economy. In addition, contingency plans covering the worst and best that could happen have a role to play in a world where the much-desired crystal ball does not exist. What do I then say about private sector forecasters? Who am I, as a market-inclined political economist, to censor the tastes of their customers? Friedrich Hayek once said that he knew few people who had made money from acting on economic forecasts, but a good many who had made it from selling them. It is difficult for those in any profession to stand out against the spirit of the times. And, in the course of predicting the unpredictable, the better forecasters often come across interesting information about the structure of the economy. But I suspect that many of their customers are mainly interested in finding out how official policy is likely to change; and to do that it may be helpful to stick to forecasting methods not a million miles removed from those employed by the so-called authorities.
<<<	articles
Published on www.samuelbrittan.co.uk Contact - samuel dot brittan at ft dot com