question.create-answer
David Round
November 2016.
199
How effective is economic forecasting?
Answer
Comment
0
Follow
1
1 answer
Share
THE AUTHOR OF THE QUESTION HAS APPROVED THIS ANSWER

Well, in many cases it doesn’t seem to be very effective! But economic forecasters aren’t fortune-tellers or soothsayers. Their predictions are conditional on certain things happening or not happening, as well as on a model or theory.

One of the great forecasting errors was the 2008 recession. There hasn’t been a recession as long as this ever – the recession of the 1930s in the UK was certainly shorter than that of 2008. One line you can take on this is that few could have predicted the refusal of the US monetary authorities to bail out Lehman Brothers, and that really pushed the economic system over the edge.

The other line, which is the one I’d take, is that we didn’t quite realise how irresponsible and incompetent the banking systems had become – and, not having realised that, and the consequent sensitivity of the banking system, we couldn’t predict the full effects of the Lehman shock.

In the case of forecasting the effects of Brexit, the pessimists will say, quite reasonably, that we don’t actually know what the terms and conditions of Brexit are going to be – they may be a good, or a bad shock – in the next few years. The real error around Brexit was in saying that the world would immediately collapse if Britain voted to leave the EU. I’d say it was a middle-class conspiracy of ‘right thinking’ people behind that one!

"It’s fairly clear that much of the economics profession had the wrong rules that they thought were applicable to the Brexit vote, and therefore their predictions were wrong."

It’s fairly clear that much of the economics profession had the wrong rules that they thought were applicable to the Brexit vote, and therefore their predictions were wrong. Economists claimed that the sky would immediately fall on our heads after the vote, and it didn’t. The Bank of England being a case in point – they saw it as being necessary to reduce interest rates which were already unprecedentedly low, but they needn’t have done so. They only did it because they made incorrect predictions about the terrible things that would happen and the loss of confidence in the British economy that would occur if people voted to leave – and they were clearly wrong.

An example of a way in which forecasting can work helpfully is when we predict that national minimum wage legislation could increase unemployment, if the national minimum wage is pushed high enough. But the question really is how high must the national minimum wage rise before it has an adverse effect on employment in Britain? At the moment, it seems we’re still testing to see how far we can push it. The forecast is that if you push it far enough, you’ll get higher unemployment. But how high is too high? Opinions differ on that.

Another issue where forecasting could reasonably be applied successfully is on the question of what would have happened if Britain joined the Euro. It would have tied the British economy to the economies of other countries which each have different needs and rules and policies – so that would clearly have had an adverse effect. Linking Greece and Germany, Portugal and the Netherlands, for example, in the same monetary union is ridiculous; they’re throwing away the advantages of having their own monetary authorities and their own exchange rates. And if you look at Britain and Sweden both staying out of the Euro, it’s been rather a successful policy for them.

"We do know some things, and we can therefore make some plausible predictions. The consequences of Greece staying in the Euro, for instance."

But we do know some things, and we can therefore make some plausible predictions. The consequences of Greece staying in the Euro, for instance. It was a great squeeze on their economy and caused a lot of hardship for their people. That was entirely predictable, and indeed it was predicted. What I personally didn’t predict was that Greece would put up with it for so long or how much support they would get from the Eurozone to keep them in it. I thought they’d have left long the zone ago.

So, to summarise, I emphasise that economic forecasting assumes that certain events happen and that certain rules or theories are valid. If those events don’t happen, then predictions based on them won’t be correct. And if forecasters hold the wrong rules or theories, then also their forecasts will be ineffective – as they have been so often recently.