Escaping the Reality Test: How Macroeconomic Forecasters Deal With 'Errors'

Repeated experiences seem to have brought clear-cut evidence that predicting future economic reality is impossible. The mere existence of an activity such as forecasting therefore turns out to be puzzling: how do forecasters manage to fend off criticisms and persuade themselves and others of their own credibility and that of their activities?The contradiction between what had been anticipated and what actually happened is a common challenge for all belief-based practices, and serves as the basis for comparing present-day macroeconomic forecasters to the magicians that Henri Hubert and Marcel Mauss analysed in the early twentieth century. Rather than presenting the condescending claim that sophisticated macroeconometric models barely conceal primitive forms of reasoning, the aim of this chapter is to suggest that, though different, magic and forecasting shareimportant traits. Both are activities whose results become apparent somewhere down the line, and in each case their performance relies on precise forms of representation, purposively designed materials, defined sets of procedures, and trained professionals