An Evaluation of the Forecasting Ability of the Norwegian National Budgeting System
SINCE the end of World XVar II the Norwegian government has published forecasts of next year's values of several macroeconomic variables. These projections, mixed sometimes with plans and targets, are published in January or February of the year to which they refer in an official document called the National Budget. It is the purpose of this paper to evaluate the administrative system that sets up the National Budgets by comparing its forecasting ability with that of other simpler forecasting methods. In order to evaluate a forecasting system some criteria of success must be developed. Such criteria can be derived from the goals which the system has set for it, from comparisons with the results obtained from other countries using different methods or from comparisons with results obtained by different methods applied in the same country. The first method was used by the author and more comprehensively by P. J. Bjerve to study the Norwegian budgeting system.1 The second has been used by the United Nations Economic Commission for Europe and more systematically by H. Theil.' The third method is utilized in this paper. Each method has its advantages and drawbacks. An evaluation based upon absolute rather than comparative standards can locate sources of error and suggest improvements, but it cannot determine whether the whole effort was worth the cost, given what might have been done by other methods. A comparison of the results obtained by different countries is inconclusive unless it can be shown that the difficulty of forecasting and the seriousness of a given error does not vary greatly among the countries comnared. A comnarison of the results of applying different forecasting methods to data obtained from the same country represents an improvement, so long as no more information is utilized than was available to the policy makers at the time they made their forecasts. The conclusions reached by this study suggest that fairly simple, naive methods of forecasting would have done just as well, if not better in some cases, than the National Budgeting system. This conclusion must be modified to recognize that forecasting was only one of the tasks which the budgeting system was meant to accomplish. But it is of significance because this system is in common use as a forecasting tool by governmental organs and businesses throughout the world and because, in Norway, circumstances were particularly favorable for its use.
- DOI
- 10.2307/1924138
- Volume
- 45 (1)
- Pages
- 23
- Export
- BibTeX
- Sources
- openalex crossref