Short- and long-term forecasting by the Netherlands Bureau for Economic Policy Analysis (CPB): Science, witchcraft, or practical tool for policy?

CPB forecasts for the next year and for the next period of government should be seen as well-motivated estimates based on all recent information, plausible assumptions and expected trends. These assumptions and trends are partly based on econometric estimation methods and insights from economic theory. However, the economy is a chaotic system in which small events can have major effects at short notice. Even when major uncertainties and risks are well understood, it remains impossible to indicate where and when a little spark will kindle a great fire or when the bubble in (e.g.) the housing or stock market will burst. The more distant the look into the future, the more uncertain are the forecasts. For such long-term analyses, the CPB employs scenarios, extended sensitivity analyses and identification of major political choices. Policy making is like sailing in fog. The regular set of CPB forecasts helps to look forward and to monitor whether a change of course is necessary. Despite fundamental uncertainty about the future, the CPB forecasts provide a good base for political discussions and decision making, like a coalition agreement, budget and wage rate negotiations and defining a long-term policy strategy. These forecasts inform Dutch society, reduce transaction costs in economic and political decision making, and foster consensus on economic and fiscal policy.