A quarterly magazine of the IMF

September 2008, Volume 45, Number 3

Picture This

Economic Forecasts: Hard to Rely On?

Forecasts of economic growth can help policymakers and the private sector plan ahead and make decisions. But how accurate are these forecasts? The answer is that forecasters often fall short in predicting a country's growth path, particularly around turning points in economic activity. And now that emerging markets are playing a larger role in driving global growth, forecasting growth has become even more difficult.

Tracking errors
The contribution of the Group of Seven (G-7) major industrial economies to global growth has declined—from about 50 percent in 1990 to 20 percent last year. Meanwhile, the seven largest emerging economies (the EM-7) have seen their contribution to global growth rise from 25 percent in 1990 to approaching 50 percent. This development poses a challenge for forecasters, who have much more experience with forecasting G-7 economies.

Growth in the G-7 economies has been only half as volatile as in the EM-7 economies, and data for the G-7 are more reliable and timely. So how have forecasters fared in tracking growth in the EM-7 economies? Evidence from a group of private sector forecasters shows that errors made in forecasting growth in these economies over the past decade have been larger than those for the G-7 economies. This year the task of predicting global growth has been even more perilous as many economies may be at turning points.

Recessions vs. reality
For the G-7 economies, on average, forecasters had no inkling of recessions a year ahead, although they suspected a slowdown by October. Only in June of the year in which the recession occurred was a decline in economic activity forecast, but it underestimated the eventual decline. As the year was ending, around October, forecasts of recessions caught up with reality.

Forecasts of recessions in the EM-7 economies show the same pattern as for the G-7 economies. Forecasts made in February of the year of the recession sense trouble, but it is only in June that there is a serious marking-down of forecasts, followed by a catch-up with reality by October. The task of forecasters is made more difficult because data on growth are prone to revisions. In addition, forecasts are conditional on policies and key prices (for example, oil prices), which can be difficult to predict. Such challenges might explain why, in the United States, the National Bureau of Economic Research often decides on the official start date of a recession well after the recession has ended.

Anticipating recoveries
Although forecasters—including those from the official sector—are often off target in predicting recessions, their record of predicting recoveries is somewhat better. For the G-7 countries, on average, private sector forecasts of growth for the following year tend to be marked down a bit once a recession is fully recognized. The gloom carries over into the start of the year but dissipates by the summer; by October, the forecasts of recovery match what actually transpires.

For the EM-7 economies, the pattern is similar, except that forecasters have been more pessimistic than warranted about the extent of the recovery—the forecasts made in October of the year of the recovery are still quite a bit below the actual outcomes.

Using forecasts correctly
This does not mean that forecasts of economic growth have no value. Without them, policymakers would be operating "without radar." But it does suggest that users of forecasts might be better served by paying greater attention to the description of the outlook and the associated risks than to the central forecast itself. Reassuringly, it is becoming more common to show how much uncertainty there is about whether the central forecast will come true. It is particularly useful to be explicit about the downside risks to a growth forecast because it can provide a wake-up call for policies and actions needed to keep those risks from materializing.