The accuracy of OECD forecasts of the international economy: balance of payments

This paper examines the accuracy of forecasts of the international economy made by the OECD. Our large data set, comprising some 5500 pairs of forecasts and outcomes, includes one-, two-, and three-step ahead semi-annual forecasts of eight components of the balance of payments for the G7 economies over a 20-year period. There is considerable variation in the accuracy of these forecasts. Although they are generally superior to naive and time-series predictions, there are some marked exceptions particularly as the forecast horizon lengthens. Forecasting error is overwhelmingly non-systematic. However, our study reveals numerous instances of forecasts which could be improved by a simple linear correction, or by incorporating information contained in known, recent forecast errors. The OECD's forecasts of services and private transfers, and official transfers are cause for particular concern: the accuracy of these forecasts is low, often below that of corresponding time-series forecasts, and rationality tests indicate that they are most prone to inefficiency and inconsistency.