If you crave social prestige, now is not a good time to be an economic forecaster. And understandably so.

The average error of year-ahead GDP forecasts is typically around 1 percentage point. And when economies are hit by a large, novel shock – the classic was the 1973/74 quadrupling of oil prices – the error can be four times as big. Moreover, turning points are notoriously ill-forecast. US GDP fell in 26 of the past 200 quarters, yet never once was this foreseen by the 3-quarter-ahead Survey of Professional Forecasts.