In January 1959 registered unemployment in Britain reached 620,000, 2.8 per cent of the labour force, and the highest figure since the War. The first issue of the Review appeared in the same month, and its conclusion was that if no further Government measures were introduced, a small rise in output was to be expected. In the event, even allowing for subsequent policy changes, this forecast proved to be a particularly bad one, so that for many years to come a superficial inspection of the sequence of forecasting errors was bound to suggest improvement! 1959 was to turn out to be the year in which: ‘You never had it so good’; in October, in a third consecutive General Election victory, the Conservative government increased its majority from 60 to 102. In 1959 also the yield on ordinary shares fell momentarily to 4 per cent and was subsequently to remain below the yield on Consols. Some concern was being expressed about ‘Stop-Go’ which was soon to lead to experiments in national planning, and some people were worried about the persistence of inflation, but all in all the wind seemed to be set fair for an increasingly affluent society. Confidence in the economy was matched by the assurance of economists that they knew what they were about. Economic research would put quantitative flesh on the bones of the Keynesian model which was not seriously challenged. That early forecasting error would be quickly rectified once the new team got into its stride.

This short essay begins by contrasting—in a rather broad brush manner—the two major types of explanation of the core of inflation in the United Kingdom. At one time these might have been labelled cost-push and demand-pull. It is now recognised, however, that the distinction cannot be made quite so sharply. It is perhaps better to speak of degrees of exogeneity of inflation—independence, that is, from other purely economic factors such as the pressure of demand and unemployment. The case of those who see important exogenous push elements in inflation may be labelled the eclectic—or, perhaps, the newspaper reader's—view. The other comprises the rather academic set of views associated with monetarist economists.

For good economic policy, two things are needed. First, there has to be clarity about objectives. The purpose of policy is to improve outcomes. The policymaker, therefore, has to know what an improved outcome would look like. There is, I think, not much dispute about this requirement: it is common ground that it would have been an improved outcome in Britain if unemployment and inflation had been lower and economic growth had been faster. There may be differences about the importance to be given to the various objectives: but that is a fairly minor matter.

In recent years economic forecasting has come under increasing attack for two main reasons. The first is that since 1973 the economic outlook for the United Kingdom has been worse than most people in the economy would have liked. As the purveyors of bad tidings, economic forecasters have been thought, albeit largely unconsciously, to bear some responsibility for the outcome they expect. It is implied that if it is possible to point out the dire consequences of any current policy it should be equally possible to point to a better alternative policy, and hence enable the policymaker to avoid the undesirable outcomes. This very plainly has not happened; judged by the criteria of economic growth, price inflation and unemployment the United Kingdom has done worse since 1973 than in any period of comparable length since the second world war. Indeed comparisons have been made between the last few years, 1979–81, and those of the slump in 1929–33. On that occasion, however, there was no corresponding fear of rising prices. The second source of attack has been on grounds of accuracy. Despite the great advances in technique and the considerably enhanced complexity of macroeconometric models, forecast errors have not fallen markedly during the last decade and have in some instances increased.

The NIESR has been for 25 years at the centre of the British macroeconomic policy debate. This seems to be a good occasion to look back critically over that debate and the NIESR's role in it. That role it has always taken very seriously, even though it has often involved painstaking exposition to non-economists of well known economic issues. In this it has continued a tradition actively followed by Keynes, of policy commitment and persuasion as well as the pursuit of research and its dispassionate dissemination. This tradition is frowned on across the Atlantic; there ‘scholars’ are not encouraged to dirty their hands with policy, at least until retirement. So be it; here our society expects greater involvement by its scholars, and he that pays the piper must call the tune.

This paper is concerned with two limited but complex questions: to what extent was the recession of 1979–81 foreseen in the forecasts made while I was Editor in 1977–9, and does the experience of that episode provide any general lessons for forecasters?

The domestic forecasts, which were finalised in the second week of May, make no allowance for the economic repercussions of the Falklands crisis, which have been small so far. The direct impact on total UK trade of the embargo on exports to Argentina will be negligible (see page 36). Although other effects, such as those working through a higher level of defence expenditure or the state of confidence in financial markets, are potentially of greater importance, these are impossible to predict at the time of writing.

The recovery in economic activity which began last spring was interrupted during the winter months. Industrial production fell in November and December and was unchanged in January. Although output during this period was affected by strikes and unusually severe winter weather, there was only a modest rebound—a rise of ½ per cent—in February, when output was 1¾ per cent lower than in October and only 1½ per cent higher than its low point of May last year.

In the United States in particular, the long-awaited upturn in economic activity has been further postponed even though inflation has been running below the rates that had generally been expected. Industry in the OECD countries may well have produced less in the first quarter of 1982 than in any quarter of 1981, and we no longer expect their total output to increase by more than about ½ per cent for the year.

Steel is the basic material of industrialisation; and also of war. For Britain in the eighteenth century, iron and steel was the cornerstone of the industrial revolution; for Germany, a century later, the steel industry was the foundation of the militarism of Bismarck. Both countries supplied steel rails for America's westward expansion in the third quarter of the nineteenth century, before the emergence of America's own steel industry. Until the 1880s the British iron and steel industry was dominant. By the turn of the century both America and Germany had overtaken Britain as a steel producer. Today Britain has the smallest of the three industries. In 1979, a relatively good year, 21.5 million tonnes of crude steel were made in Britain, compared with 46 million in Germany and 123.3 in America.