Central banks are no strangers to criticism, but we wonder if the Bank of England was remotely prepared for the smackdown it was just given by FT economics editor Chris Giles:
The forecasts used by the Bank of England to set interest rates are biased and contain little useful information, a Financial Times audit has demonstrated.
Despite having hundreds of economists working in the Bank, and the most sophisticated suite of economic models in the UK, the monetary policy committee’s forecasts since 1997 have achieved no better outcome than if the committee had simply predicted the average level for inflation and growth over the 13-year period.
Like securities analysts, the BoE seems to have a perpetual bias towards optimism. Giles again:
This bias stems from the Bank’s default forecasting practice that assumes good times will last, but that when times are bad they can only get better.
Giles goes on to explain that the BoE has a few other specific problems to deal with, but the usefulness of macroeconomic models more broadly has been the source of considerable commentary in the past few weeks–mostly because it was fundamental to the debate on whether the US stimulus was effective.
And as for whether macroeconomic models can be used to predict the future, some of the recent criticism from bloggy economists has been targeted at the the forecasts coming from the private sector. Tyler Cowen wonders why there are so many of these forecasts given that they are similar to a public good, and Eric Falkenstein, well, just go read him and thank us later.
To be fair, we should make the obvious point that the BoE isn’t alone among the world’s central banks in getting forecasts wrong, though its apparent track record over 13 years is impressive in this regard. For another very recent example, here’s a line from today’s FOMC statement that makes strategic use of the passive voice:
Nonetheless, the Committee anticipates a gradual return to higher levels of resource utilization in a context of price stability, although the pace of economic recovery is likely to be more modest in the near term than had been anticipated.
Regardless, there seem to be only a couple of possible responses when presented with evidence that a forecasting model hasn’t worked. One is to try building a better model. But it might be unclear just what “better” would entail. In the case of the BoE, with its hundreds of economists and “sophisticated suite of economic models”, this would seem especially problematic.
The other option is to admit that even newer and more inclusive models are unlikely to be much better, and therefore policy decisions should be based on something other than predictions of the future. This would be the obvious choice if it were clear what those “other” things would be. A solution isn’t immediately obvious, at least not to us.
By the way, the Giles article does include this one (rather weak) defense of the BoE:
Independent economists say that while these problems are undesirable, they are understandable be-cause, as Simon Kirby of the National Institute of Economic and Social Research says, everyone missed the recession and “forecasting is very, very difficult”.
Okay, but maybe it would be better to focus a bit less on whether accurate forecasts are “very, very difficult”, and more on whether they are simply impossible.
Related links:
FT audit casts doubt on Bank’s forecasts – FT
Can we predict a financial crisis? Does it matter? – FT Alphaville