Don't forecast recession

Duncan Weldon is forecasting a “double dip” recession. I’m not sure this is wise.
I don’t mean he’s wrong - he might not be. What I mean is that an economic forecast should not be a description of the future. It should instead be based upon an assessment of the costs of being wrong versus the benefits of being right. This, for example, is why well-established forecasters - such as those working the major investment banks - so often produce near-consensus forecasts whilst less well-known ones produce extreme ones. The former lose a lot from being wrong, relative to others, and so don‘t risk it. The latter, however, need to attract attention, and so the benefit of producing an eye-catching forecast outweigh the cost of being wrong.
By this criterion, Duncan’s forecast doesn’t stack up - or at least, it certainly wouldn’t if all of Osborne’s critics were to adopt it.
To see my point, consider two scenarios:
1. Osborne’s critics forecast a recession, but we get sluggish growth of, say, 1%.
2. They forecast growth of 1%, but we get a recession.
In both cases, the forecasts are equally wrong. But the pay-offs differ.
In (1) Osborne will say: “My critics forecast recession. Their warnings about the effect of my cuts were wrong. This shows you shouldn’t believe them.”
In (2), however, he can’t say this. But his critics can say: “We thought the cuts would have bad effects, but they turn out to have been worse than we thought.”
It’s better for Osborne’s critics to forecast mild trouble, therefore, than to predict doom.
You might object here that I’ve set up the problem wrongly, and a recession is, as Duncan says, the most likely outcome.
This isn’t really the point. There is a big risk to Duncan’s view. It is that bosses now are even more overconfident than usual - because surviving the recession has further boosted their inflated self-belief - and overconfident bosses are more likely to invest. If so, we might get sufficient growth in capital spending to offset weak public and personal consumption.
If this happens, it would in no way vindicate Osborne’s position: such investments will prove to be unprofitable in coming years. But it would appear to discredit those who forecast recession. The cost of this risk materializing, I suspect, outweighs the expected benefit from taking the risk.
This is not to say Duncan, as an individual, is wrong to take the risk. Maybe he reckons the uplift to his own reputation in the event of being right outweighs the risk of being wrong.
What I’m saying is that Osborne’s critics, in aggregate, should not forecast recession.

Comments

[Duncan]

I broadly agree with much of this and spent several days umming and erring before posting that article.

But the central point is that I don't think the OBR have got 4 things right.

Certainly were I a Labour frontbencher, I wouldn't be forecasting a recession.

[Leigh Caldwell]

Isn't it possible that the act of making the forecast could itself have an influence on current policies? If enough people agree that there is likely to be a recession, might the government's actions (subtly) change?

I don't propose that they'd publicly revoke the cuts or change course in a highly visible way, but it could change the emphasis of some future decisions. Indeed, I believe that some of the notionally growth-friendly aspects of this year's budget were indeed motivated in exactly this way.

[Cecily]

I agree with your view. I would not be forecasting recession either.

[CharlieMcMenamin]

You appear to have made a watertight case for established economic forecasters all agreeing that the Emperors New Clothes are lovely, but permitting themselves to disagree a trifle over whether the cuffs on his jacket have two buttons or three.

This may indeed be sensible professional behaviour - it may even, as Duncan frankly confesses, be sensible political behaviour for the Labour front bench.

But it's not actually 'forecasting' in the sense that most lay people understand it, is it?

[George Hallam]

"This may indeed be sensible professional..[or] political behaviour..
But it's not actually 'forecasting' in the sense that most lay people understand it, is it?"

I agree.

I'm not shocked by S&M's opportunism, just surprised that anyone could be so disarmingly frank.

I’ve enjoyed reading this blog. However, from now on, I won’t be able to take anything said in it seriously.

He who tells the truth is sure to found out.

[chris]

@Charlie, George - lay people don't understand forecasting because they don't realize that knowledge of the future is a contradiction in terms.
Any statement about the future is, by definition, untrue because it describes a state of affairs which does not exist. If you're going to tell untruths, you should at least do so in your own interests.
@ George - no-one is supposed to take this blog seriously, as I said ages ago:
http://stumblingandmumbling.typepad.com/stumbling_and_mumbling/2005/02/against_credibi.html

[CharlieMcMenamin]

@Chris:"..lay people don't understand forecasting because they don't realize that knowledge of the future is a contradiction in terms."

Upto a point Lord Copper.

I expect some weather forecasts to prove more accurate than others even if I don't expect them to always be 100% accurate. I can even cope with the idea of gross errors like Michael Fish and his Hurricane mistake once in a while.

But I don't expect weather forecasters to huddle together around an 'average' prediction unless they each, independently, look at the data and decide that's what the data seems to suggest *whatever* their professional colleagues might think.

&, yeah, I know you're doing Social Science and not Physical Science, so there are limits to this analogy. But I'm not sure the limits are that great here: it's about honesty in both cases. Honesty with the non professional audience.

P.S. I'm glad you're not serious-minded on this blog, nor self important. But to claim you're not serious underneath all the tongue in cheek stuff is pushing it a bit far I think.You may be the last Analytical Marxist left in the financial blogosphere, but some of us think we just about recognise the tattered flag you still fly on High Days and Holy Days.

[Paul Sagar]

"What I’m saying is that Osborne’s critics, in aggregate, should not forecast recession."

Even if that's the most likely outcome?

So, even intellectual honesty must be subsumed beneath the principle of projected political efficacy?

Ho-hum.

[Keith]

"Any statement about the future is, by definition, untrue because it describes a state of affairs which does not exist."

In The city of God Saint Augustine seems to think the future does exist as part of the devine plan already laid up by God. He seemed pretty sure about it!

What is being forecast is the evoluion of trends discernable in the present. So it maybe true or not; we merely cannot be certain. If it happens as anticipated it is true at the future time. When it will be in actual existence.

[Shuggy]

"Any statement about the future is, by definition, untrue because it describes a state of affairs which does not exist."

Some things that lie in the future are so probable, to describe a statement about them as being 'untrue' stretches the concept to the point of meaninglessness. It'll be 1 o'clock in the morning in a few minutes....

[Mark Fuller]

recession is a very bad forecast. If it was based in a valid and reliable information, it could be mentioned. But forecasting recession doesn't appeal to me as well.

[Glenn]

Its 0.5 per cent growth but the margin of error must be about + or - 0.5%!!!