This chapter is in two main parts. The first contains a survey of economic developments in 1982. It begins with a brief account of economic policy; succeeding sections deal with output and demand, inflation, and the balance of payments. It concludes with a discussion of the experience of individual industries over the past year, and since the beginning of recession in 1979. The second part of the chapter contains a short-term forecast to end-1984.

Expected twelve months ago to be a year of slow but continuing recovery, 1982 proved in the event to be one of stagnation, at a level of output, for the OECD countries, slightly lower than that of 1981. Hopes for rather better things had been based in part on stock movements and in part on some minor policy relaxations. In fact, however, the self-perpetuating influences of slow growth in the preceding years proved to be stronger than the self-reversing ones.

Economists may wince when politicians compare public finance to the management of the housekeeping money or businessmen draw analogies from the experience of their firms; but it is then up to the economist to provide some alternative and better framework for public discussion of the Government's borrowing requirement and the size of the national debt.

Replacement ratios measure the relationship between income out of work and income in work. A ratio of 50 per cent for example would indicate that an employed person would have only half the income if he were unemployed. Interest in these ratios may be a reflection of concern either about living standards or about the incentive to work. For some purposes it is the current level and dispersion of these ratios that matter, for others it is necessary to build up a consistent set of ratios for a run of years in the past. The construction of time series is of particular importance if one is trying to assess the effect of the benefit system on wages or on employment.

Economists who regularly use econometric models are well aware of the problems which arise if those models are not updated frequently. Updating is more than mere mechanical revision in the face of new data. This article will reveal that quite substantial changes have taken place in the National Institute's model as a result of a comprehensive re-estimation exercise, undertaken over the last year, which encompassed most of the behavioural equations in the model. Although there will not be space to describe all of these changes in full, the next section provides a brief account of the main developments. More information on some of the equations is given in an appendix, and the final section of the main text describes the results of a range of simulation exercises using the whole model.