Talk loudly and carry a little stick: the ECB’s communication policy

The ECB, through its President, Jean-Claude Trichet, is back in the game of pre-announcing future interest rate changes. I would have thought that the experience of August 2007 would have cured them of this urge for a bit longer. On August 2, President Trichet flagged a rate rise for September by using the ‘strong vigilance’ code words: “… strong vigilance is therefore of the essence to ensure that risks to price stability over the medium term do not materialise.” Then events, dear boy, events intervened in the form of the August 9 eruption of the financial crisis, and the pre-announced rate hike was hastily shelved.

The wordsmiths at the ECB appear to have been busy in the mean time coming up with a new collection of code words. Strong Vigilance is no longer with us, and neither is his weakling half-brother, Mere Vigilance. Instead the Governing Council is reported in the Introductory statement of June 5th to be “… in a state of heightened alertness…”. Taken at face value this means no more than that the majority of the members were awake during the meeting, because the President had removed the decaf.

The communications policy of the ECB is dreadful. The Introductory statement, written well before the rate-setting meeting starts, at least has the virtue of being carefully if not skilfully crafted. It should just be e-mailed out to the media and put on the ECB’s website. The Q&A is an uninformative, noise-amplifying, de Gaulle-style press conference – an ego-trip for the ECB President, with his side-kick, the VP sitting mumly next to him. No light is shed on anything that matters. Quite often the waters are muddied by verbal infelicities and ambiguities.

The only communication that matters for economic efficiency is the communication of the ECB’s decision rule, mapping the past, the present and uncertain future contingencies into future rate actions. Pre-announcing (or hinting vigorously at) the next policy rate decision is pointless, because next month’s rate decision in and of itself has a negligible impact on anything that matters for macroeconomic stability. Because the pre-announcement can never be a firm commitment (vide August 2, 2007), it also risks undermining the ECB’s credibility.

The next rate move matters greatly to those who wager multiple billions euro worth of bets on it, that is, to the financial markets. Indeed, it may be a matter of commercial life and death to many of the players. I don’t think it is the job of the ECB to create a three-ring circus to facilitate the organisation of lotteries whose payoffs depend on its future actions. By only pre-announcing the next move, the central bank also encourages and feeds the myopia of the financial markets, where the short term is the next trade, the medium term is lunch and the long term is the end of the trading day, when positions have to be squared.

There is no need for the central bank to ‘prepare the markets’ for the next rate increase. It is not the responsibility of the central bank to ensure that futures markets, the OIS market or other derivatives markets correctly anticipate its next action. It should of course not do anything to deliberately wrong-foot the market either. ‘Teaching the market a lesson’ through a short squeeze or whatever is counter-productive macho nonsense. But if the financial markets price in a rate move or rate level that is unlikely to materialise in the view of the ECB, it’s the market’s problem. In any case, for every loser there will be an equal winner somewhere – it’s a redistributional event among the shareholders of financial institutions and maong the bonus earning traders working in these institutions. The mispricing of the next rate move of the ECB has negligible implications for real consumption and capital formation decisions – the only things that ultimately matter for economic efficiency.

Some central banks, like those of Norway, Sweden and New Zealand, offer forecasts not only of inflation and real GDP, but also of their own actions going several years into the future. These tend to be point forecasts, rather than distributions of forecasts, generated using either stochastic simulations with one or more decision rules for the central bank (some variant of the Taylor rule, say), or through the Bank-of-England style imposition of a variance and a skew on the central projection. For the ambitious, they could be generated using dynamic optimisation methods. I am not opposed to this kind of information. Neither do I find it wildly informative. There is no substitute for the track record of actual rate decisions and the communication of the reasons for these decisions and of the information on which it was based. Anything else is, if not quite cheap talk, at least rather inexpensive talk.

Pre-announcements, forecasts etc. are not extremely cheap/gratis talk because they can affect the credibility of the central bank. If that vanishes, because the markets, having been bitten once too often, have become shy, the central bank’s capacity to leverage expectations of future policy rates will be impaired.

The credibility problem is a serious one, because the central bank is potentially afflicted by a serious ‘time consistency’ or credible commitment problem. Assume that, in current circumstances, increases in current and future interest rates would be costly because they would drive actual output below the path of potential output and actual employment below full employment (or what passes for it in the Euro Area). The price is deemed worth paying because inflation needs to be brought down from its current 3.6 percent rate to something below but close to 2 percent.

The central bank announces a bunch of future rate increases, starting, say, a couple of months from now, but does not increase the current rate. The markets believe the central bank. Medium-term interest rates go up over the horizon the markets believe the central bank will be raising the policy rate. These higher medium-term rates dampen demand and reduce inflation to just below two percent. The central bank has inflation where it wants it. Two months pass. There are no further shocks or new developments. Time for the central bank to deliver on the policy rate increases whose credible pre-announcement brought inflation down to the target level. But why should the central bank bother with the actual increases in policy rates? The pre-announcement has done the job on its own. The actual increase itself are not just redundant, they may well prove harmful. It may be that the central bank will be able to convince itself to deliver the no-longer-necessary policy rate increases, and thus to create a further slowdown and perhaps too low a rate of inflation, just to keep intact its future credibility. But that would not be an easy decision to make.

The dominant current view of the inflation process – the New-Keynesian paradigm – has exactly the features that I emphasize in the previous paragraph. In that paradigm, current inflation depends on past inflation and on current and anticipated future output gaps or deviations of the actual unemployment rate from the natural rate. The central bank can get inflation down if it can credibly announce future policies that will create future recessions. Of course, if the central bank does this successfully, it could find itself in the position that inflation is at its target, there is no output gap – the actual and natural unemployment rate coincide – but the central bank now has to deliver the actual recessions whose earlier anticipation brought down the inflation rate to its target level. Can you imagine the howls of fury coming from the Palais de l’Élysée if the ECB created a recession simply to invest in or maintain its own credibility for fighting future inflation, when actual inflation is on target and no inflationary threat may be on the horizon?

So life is difficult for monetary policy makers when private economic actions today depend at least in part on these private agents’ past and current anticipations of the future that can be manipulated by the central bank. But you don’t address this problem by dropping strong but non-binding hints about the next movement in the official policy rate. Instead you do this by explaining at length and in detail your view of the monetary transmission mechanism, by having a clearly stated objective, and by the track record of your actions, given the information available at the time.

This is perfectly consistent with, indeed in my view requires, the absence of Q&A sessions, press conferences and other noise-amplifying media events. Talk about the framework of monetary policy, about models and transmission mechanism, about properties of different policy rules, but don’t anticipate in your statements actual future policy rate moves in any way. Say nothing at all, but carry a massive stick.

Maverecon: Willem Buiter

Willem Buiter's blog ran until December 2009. This blog is no longer active but it remains open as an archive.

Professor of European Political Economy, London School of Economics and Political Science; former chief economist of the EBRD, former external member of the MPC; adviser to international organisations, governments, central banks and private financial institutions.

Willem Buiter's website