Time to rip it from him, warn economists
Last week’s Autumn Statement was a headache for the chancellor.
He was forced to revise his figures yet again. Growth down. Borrowing up.
But he can plead it’s not his fault. After all, he doesn’t come up with the forecasts – that is the role of the Office for Budget Responsibility.
Headed by Robert Chote, the OBR has exclusive control over the forecasts the chancellor can use. The OBR was set up in 2010 to supply independent forecasts to prevent political meddling.
The most crucial forecast is GDP growth. The fan diagram, pictured above, is nicknamed Chote’s Flamethrower of Uncertainty.
You can see why the flamethrower name is apt. The OBR says growth in 2016 could be as high as 4.5% or back in recession. That’s enough to torch any sense of confidence.
Worse, Chote seems to waive his flamethrower around.
In June 2010 the OBR forecast growth in 2012 would be 2.8 per cent. Now it expects a minus figure.
Last week the OBR was forced to increase public borrowing by £104bn over the next five years compared to its forecast in May.
The hapless Chote was grilled by Paxman on Newsnight, who asked “what the point of you is” when the forecasts were “wrong”.
There are two other problems with the OBR, other than its vagueness and shaky record in forecasting.
“I like the Swedish example which is a composite and makes no forecasts of its own”
David “Danny” Blanchflower
The first is that under the current system, it is a monopoly.
The use of a single forecaster ignores the principle of the “wisdom of crowds”. In his book of that name James Surowiecki tells the story of how the crowd at a county fair correctly estimated the weight of an ox. No individual could do so, but by averaging all the guesses the correct weight was derived. By contrast the current system relies on just one guess, that of the OBR.
Second, the chancellor is given little discretion to decide whether the OBR’s judgements are plausible. He may have gotten wind of political developments which would materially affect growth forecasts, such secret sovereign default negotiations, but, in theory, he must ignore these and use the OBR’s charts. Okay, in practice he can do, but god help the chancellor who admitted to ignoring OBR data.
David Blanchflower, a former member of the Monetary Policy Committee, tells me these shortcomings mean the OBR is failing to justify its existence:
“Hard to see they have generated any credibility given that their forecasts have been so hopeless- not truly independent,” he says via Twitter.
So what’s to be done?
The chancellor could switch to a composite system. By fusing the estimates of, say, 20 forecasters he would radically lessen the impact of an error by any single contributor.
Blanchflower urges the chancellor to adopt a new system:
“A composite is better – I like the Swedish example which is a composite and makes no forecasts of its own.”
He points to a paper by Swedish economist Lars Calmfors, on how the Swedish system works (link). The paper warns the UK system may not even be properly independent. Since the OBR must provide the macroeconomic forecast in the government’s budget bill: “This makes it impossible to avoid a continuous, behind-closed-doors interaction with the Treasury on numbers, which could easily turn into a negotiation process.”
Jonathan Portes, the respected director of the National Institute of Economic and Social Research tells me:
“Think I agree. OBR could then stick to assessing fiscal sustainability given an independent composite forecast.”
On the other hand, Chris Giles, economics editor of the FT, joined in to tell Portes, Blanchflower and myself:
“Oh come on – consensus forecasts no better than OBR. Would make no difference, but be less accountable”
He points out the OBR has the merits of being “transparent” and the OBR has “hugged” consensus. Which isn’t to say he’s a fan of the current system. Chote’s flamethrower comes in for sharp criticism:
“OBR does fan charts, but they are as useless as the BoE versions. Reality worse than previous errors”
Sadly, nothing seems to work very well.
As this chart shows, Gordon Brown’s tenure as chancellor was dogged by the most horrendous optimism. The dotted lines show his debt forecasts at each budget; the thick line shows what actually happened. Delusional forecasting like this partly explains how Britain racked up such a huge structural debt during the long boom years.
Perhaps Chote should keep a print-out of this chart. The next time he’s forced to rejig his forecasts he can remind his critics that yes, the OBR is flawed, but hey, it ain’t as bad as the old system.
And besides, at least we can see the OBR’s methodology. Monetary Policy Committee forecasts are still opaque. Blanchflower points out: “OBR has been more transparent than the MPC who still wont tell us anything about their model”.
Even Chris Giles concurs: “Agreed. There is a potential conflict between OBR and MPC forecasts – serious, but no info”.
Leave a Comment