Belfast Telegraph

... so, what's the view from Dublin?

By Brendan Keenan

There was a bit of fuss the other day over claims that the IMF (International Monetary Fund) failed to predict a single one of the recessions around the world since 1999. But actually, it doesn't say on the tin that it will.

That is kind of convenient, since the Fund's member governments would not care to be told that their economy was due to start contracting in, say, 18 months' time. Not unless it was a certainty. Which of course it isn't.

A better reason for this lack of dire predictions is that recessions cannot be forecast. Nor indeed can recoveries.

Our recent experiences should make us well aware of this. Mathematically, the Irish growth rates of the early 2000s could not go on for ever. But when would they stall - and how? Hard landing or soft landing?

The occasional politician and property developer could be found to say there would be no landing at all. That could be dismissed although, as any nervous flyer can testify, no landing ever feels soft. But one could apply only probabilities as to whether what was coming would be a crash, a recession or merely a downturn. As for when - forget it.

It has been the same at the other end. There were people who said there would be no recovery at all, and hardly any who forecast it would be early and vigorous, in the way it turned out to be. Behavioural economics says we are more prone to pessimism than optimism.

It's all about behaviour. Perhaps one day, those behavioural economists will be able to predict when people will take fright, or decide they have run out of money, or conclude that the sun is shining again. But I doubt it.

The problem is simple enough. What is known as forecasting is actually an, admittedly sophisticated, extrapolation of past trends and current conditions into the future. In in a recent edition, The Economist newspaper outlined the sophistication, but also did some measures of forecasting accuracy with, naturally, the IMF as guinea pig.

The conclusion would seem to be that medium term forecasting - never mind long-term - is pretty useless. (The Economist could not resist quoting JK Galbraith, "The only function of economic forecasting is to make astrology look respectable.")

Belfast Telegraph