The reliability of macro-economic forecasts

In Germany since early 1981 a consensus has emerged predicting that an economic recovery is due for the second half of the year. This forecast of an upturn was put-forth for 1981 and for 1982. Despite these consensus, the forecasts proved to be wrong. Nonetheless in autumn 1982 the consensus forecast for 1983 again explicitly included a recovery starting mid-year, and again the forecast seems to be proving to be wrong. This time, however, the failure is not forecasting a recovery, only to find out its not there, but rather miscalculating in the other direction as the economic recovery has already been' on the move since the beginning of 1983.