Wrong variables

IMF seers should admit ignorance - or get a grip

A study of the last five semiannual forecasts from the International Monetary Fund makes for depressing reading - about the global economy and also about the models economists rely on.

The IMF, meeting in Washington this week, has become less optimistic. In October 2011, it expected global GDP growth of four per cent in 2012. The actual outcome was 3.2 per cent. For 2013, the first cut prediction in April 2012 was 4.1 per cent. This week's forecast was 2.9 per cent. There have been two attempts to forecast growth for 2014: a figure of four per cent in April is now down to 3.6 per cent.

The IMF has also lost confidence in the long term. Over the last two years, its forecast for world GDP growth five years from now has fallen from 4.9 per cent to 4.1 per cent. Almost all the decline comes from more modest expectations for developing economies, where the growth outlook has slipped from 6.7 per cent to 5.5 per cent.

To put it simply, the changes in the IMF's expected future corresponded with changes in the present. That suggests the fund could, essentially, abandon its model with hundreds of inputs. An extrapolation of current readings, adjusted by educated guesses about the economic cycle, would provide almost the same results.

That's not to say the IMF is much worse than other forecasters. All of them suffer from excessively narrow analysis. They are good on investment rates and inventories, and they make a stab at fiscal multipliers - although how changes in government deficits feed through into growth seems to vary more than many economists would like. But all these models leave a lot out.

They mostly assume that financial markets work smoothly, so they miss credit gluts and squeezes. They underestimate demographic factors, so they overestimate growth in countries with declining numbers of young people. And the models largely ignore institutional factors. Weak governments and a lack of faith in the rule of law do a lot to explain softening growth in poorer countries. Forecasters may always be the butt of jokes because they will often be wrong. But meteorologists have improved their accuracy, largely by collecting and examining more data. Economists at the IMF and everywhere else could try harder, too.