I gleaned from the Guardian’s liveblog of the Treasury Committee hearing this morning that Carney disclosed that the Bank of England’s Monetary Policy Committee had discussions on whether to publish interest rate forecasts.
A majority were not in favour, so they are sticking to the status quo. I think it’s hugely positive that such discussions were had, and that it has been disclosed that they took place. Whether or not to publish such a forecast seems to me to be a matter that has to be left to the MPC itself, as compelling it to do so against its will would open the door to unacceptable infringements on operational independence, and anyway be impractical. So putting the decision to some kind of informal vote seems like the best way to manage disagreement.
Such a vote puts all members in an interesting predicament. We might reasonably presume that MPC have to have one communications policy. The minority who prefer publishing forecasts would further reveal the forecasts of those who did not if they went ahead anyway and disclosed forecasts of their own vote. [For example, MPC member Jan Vlieghe’s comments about the need for perhaps 6 interest rate rises over 3 years in his own world view can be used to narrow down our distribution on the concealed views of the other members’]. Yet on the presumption that every interest rate vote only makes sense as an implementation of an entire interest rate plan, the majority who vote against publishing deprive the minority of a key piece of evidence that they can use to defend their own vote and be individually accountable for it.
The reason given for not publishing by the majority, we are told, is that there was a fear that such forecasts would be interpreted as ‘promises’. The idea is that the forecast would explain what would happen to interest rates if the economy develops as MPC anticipate; so that if things do not turn out as expected, rates also will not move in line with the forecast. And the concern is that the initial conditionality of the rates path on the forecast will not be appreciated, and the change of path interpreted as breaking of a promise. That might be bad because it might signal that the promise-breaking related to the pursuit of the inflation target itself.
These concerns are not unfounded. But to make a point about the process governing this discussion, stated simply like this they are not yet evidence-based. Such an important decision for the MPC surely ought to flow from some kind of thorough report and analysis, with each side explaining and substantiating its position fully. One would expect the MPC to take evidence from other central banks who have chosen to publish those forecasts, and from outside observers who consume and interpret the Bank of England output. If this discussion took place in the context of such an exercise, the Bank should publish the proceedings, supporting papers and evidence, and minutes of the relevant meeting. If not, we might reasonably wonder why not. Given that the MPC membership turns over slowly, we might also be given forward guidance as to when the decision would be reviewed.
On the substance of the ‘promise-breaking’ argument, there are counter-arguments, made many times in this blog and by others.
Because monetary policy, and the shocks it is seeking to offset, take time to have their full effects, monetary policy making must entail making forecasts, including those of the interest rate and asset purchase instruments. If the policy making process does not involve such a process, it must be deficient, and sooner or later will be revealed to be. If it does, then at some point, when enough people realise this, they will see the communications process as being about attempting to obscure or conceal the instrument forecasts. If MPC are worried about being seen as promise breakers, then it is worth thinking through the possibility that outside observers may twig that a good way to get away with breaking promises is to hide or at least obscure the promise being made.
To some extent the communications process now adopted, which involves frequent coded references to the concealed interest rate forecast, already runs the risk that the MPC majority fear – of making promises that are not kept. That is why, fairly or unfairly, Carney is frequently referred to as the ‘unreliable boyfriend‘. Think of the status quo communication policy as revealing the forecast, but injecting a little bit of noise amounting to a few policy changes here and there. On occasion, you are going to be accused of breaking the ‘promise’ even when you haven’t, precisely because of that noise. [Or rather, on occasion, events will turn out such that the hidden interest rate forecast has not changed much, but because you did not signal that forecast accurately enough, observers will incorrectly update their estimate of your forecast and wrongly conclude it has changed when it hasn’t.] Dave Ramsden is reported to have remarked that published forecasts would be ‘lost in translation’: arguably the same loss is being experienced currently.
Previous to today’s remarks, Ben Broadbent was reported to have commented on the matter to the effect that the Bank should not have to ‘spoon feed’ markets, and that the general public at large was not so concerned about the short run outlook for interest rates.
These remarks are not persuasive.
The Bank does not disclose enough about its interest rate reaction function [how it responds to events], nor its interpretation of the ambiguous monetary policy mandate, nor its interpretation of the state of the economy for that matter [we don’t have working versions of the forecast plus judgement to scrutinise], for outside observers to be able to forecast what the MPC itself will do if its own forecast comes to pass. So an interest rate forecast is not properly described as ‘spoon feeding’.
It is also not appropriate to appeal to service to the wider public as a justification for concealing its interest rate forecast. Much of the wider public has mortgages, and their forecasts of their own disposable income will depend on the outlook for interest rates. By contrast, I doubt that all that many people are interested in this quarter’s vintage of the answer to the question ‘how this time are we forecasting inflation to converge back to 2 per cent?’, a caricature of the Inflation Report that is sometimes not far from the truth.