The Bank of England Could Use Some New Fans

Using fan charts to communicate a range of probabilities would reduce uncertainty.

Some charts might help.

Central bankers devote a huge amount of time and resources to communicating with markets, companies and the public. So why do the Bank of England's communications still end up confusing markets and distorting decisions in the real economy?

The BOE's latest inflation report marked a new low on the communication front. The bank explains how the outlook for interest rates has changed by comparing two invisible lines: a forecast for rates that the bank hasn't published, and the forecast the bank didn't publish three months ago. And the communication on how far rates could rise in this hiking cycle is so vague now as to add little value. Investors are left to figure it out for themselves, which leaves financial conditions -- bond yields, equity prices, the value of sterling -- at the mercy of the market's often flawed understanding of what the BOE will do.

It doesn't have to be this confusing. Of course, central banks cannot give cast-iron promises about how they will set rates in the future. But they can describe how they expect rates to evolve given their macroeconomic forecasts.

Better still, central banks can explain how policy will respond to unexpected events through publishing fan charts that show potential outcomes, reducing unnecessary uncertainty about the future path of interest rates. That takes care of the traditional worry that the market may misunderstand forecasts as promises. Central banks already use fan charts for growth and inflation forecasts; they should use them for interest rates, too.

A fan chart illustrates the probability of various outcomes for the future. For example, in this stylized version of a fan chart reflecting various outcomes for the main bank rate in the future, if economic circumstances identical to today's were to prevail on 100 occasions, then in any particular quarter the rate would lie within the darkest central band and within each pair of the lighter bands on 30 occasions (and can fall outside the purple area on the remaining 10 occasions):

Source: Barwell/Yates

The BOE currently puts itself in the worst of all worlds. There is no forecast for the bank rate, its main interest rate, let alone any indication of the range of possible outcomes for a variety of future data; so there are none of the benefits of full transparency. Instead, there are statements that could all too easily be misconstrued as promises ("two hikes in three years") with all the potential costs that might entail.

There is an additional problem with this communications strategy. The BOE might one day need to convince the markets that it plans to change the way it sets interest rates. The pledge to keep rates "low for too long" might be one of the few effective ways to stimulate the economy when interest rates next hit the lower bound. But the success of that strategy hinges on the credibility of the commitment. It is imperative that the BOE does not teach markets that its words carry little weight in the meantime.

This communication problem is only going to get more complicated. The BOE will soon have to start explaining how fast it plans to sell the bonds it purchased through its program of quantitative easing (QE), as well as how fast rates will rise and how these two policy instruments will be used in tandem to tighten policy. It will also have to provide clarity on where it sees both interest rates and the balance sheet leveling out. For this, the BOE should provide a fan chart for its balance sheet (a stylized version below) as well as one for interest rates. 

Source: Barwell/Yates

The bank now has a number of options for unwinding QE, all of which would require clear communication. It could follow the Federal Reserve, which has essentially pre-committed to a certain pace of balance sheet run-off, and then varies interest rates as circumstances dictate. Alternatively the BOE could focus first on unwinding the balance sheet -- reducing it through asset sales or, more slowly, by not reinvesting maturing assets. It would then reduce the effects, good and bad, that QE continues to engineer. The bank rate would not rise until the balance sheet was normalized.

A third option would be to use asset sales as a pressure valve for when the bank needs to tighten fast. For example, the bank could reassure households that whenever circumstances warrant an increase in the policy rate by more than 50 basis points in a calendar year, then the excess tightening will be delivered instead by selling bonds. Forward guidance that the hiking cycle will be gradual is thus made credible.

Once the bank has decided on a particular strategy for using the two instruments to tighten policy -- raising rates or selling assets -- it should publish forecasts for both policy instruments as well as growth and inflation, each placed inside fan charts and accompanied by a comprehensive policy statement. Publishing a more complete description of the state of contingent planning won't get rid of all uncertainty. But it would remove the current state of ambiguity and confusion.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

    To contact the authors of this story:
    Richard Barwell at rbarwell2@bloomberg.net
    Anthony Yates at anthony_yates@yahoo.com

    To contact the editor responsible for this story:
    Therese Raphael at traphael4@bloomberg.net

    Before it's here, it's on the Bloomberg Terminal.