The Importance of ECB Wording on Inflation – Euro Thoughts Sep 7-11 2009 – UniCredit Group

Last Thursday my boss, Marco Annunziata, did not hesitate to define President Trichet a true statesman for his wise and transparent conduct of the ECB September press conference. In the last few weeks, with his intervention in Jackson Hole, the Sep 4 presser, and the op-ed in the Financial Times the following day, Mr. Trichet has been able to reassure markets that the ECB is aware that the economic and financial cycles remain fragile, that the weapons of mass salvation employed over the last year need to remain in place (e.g., the decision not to apply a spread on the upcoming 12-month fixed rate-full allotment operation), thus steering market rates’ expectations in the desired direction. The publication of bearish staff forecasts has contributed to finish the job. The ECB stance remains quite accommodative and any decision on rates and on the unwinding of unconventional measures is postponed to next year.

Having criticized in a few occasions over the last two years the ECB communication, I feel the natural obligation to praise the more recent strategy, and to try to put it in a wider and more analytical context, also because since the publication of the research paper entitled “Read my lips – Assessing the recent ECB’s communication” in August 2008, we at UniCredit have been devoting great attention and analytical respect to the central bank’s communication policy. Most recently we have studied the issue more in depth and have developed a few tools that are now full members of our Eurozone economist’s toolbox.

When talking about communication, there are two important issues at stake: the accountability of the central bank, and the market impact of its communication. In turn, there are at least three reasons why communication may prove useful for central banks. First and foremost, communication may be a very direct and effective tool to influence expectations. Therefore, it plays a key role in improving the effectiveness of policy and, consequently, the economy’s overall performance. Second, communication may be used to reduce noise in financial markets. Greater disclosure and clarity over policy may lead to greater predictability of central bank actions, which, in turn, reduces the uncertainty in financial markets. Finally, communication is indispensable from the perspective of central bank accountability. As central banks have become more independent over time – and the ECB has proven more than once to be a “very independent” central bank – they have to pay closer attention to explaining to the public opinion what they do and what underlies their decisions.

Resorting to the econometric technique of the Principal Component Analysis (PCA), and following the methodology developed in Berger, De Haan and Sturm in 2006 (http://www.ifo.de/pls/guestci/download/CESifo%20Working%20Papers%202006/CESifo%20Working%20Papers%20January%202006/cesifo1_wp1652.pdf), Davide Stroppa and I have developed an index of ECB Communication based on the content of the Introductory Statement the ECB President reads every first Thursday of each month. More precisely, we have developed three sub-indices that try to capture in a single continuous variable the ECB assessment on real activity, inflation and money, respectively. The information contents of the three indices are consolidated in one single index of ECB communication. A simple regression shows that the communication index leads the refi rate by 3-6 months, with the coefficient being positive and highly significant, meaning the future monetary policy actions are an increasing function of today’s words. Then, we have tried to understand how communication can improve our understanding of the policy reaction function. We thus estimated four different equations, adding to our proprietary Taylor rule (with a survey-based measure of the output gap, and M3 growth to capture the two-pillar ECB strategy) the level of our communication index and the three sub-indices for activity, inflation, and money.

There are some interesting insights. First, all our specifications signal a significant disconnect between the fair value implied by the model and the actual value of the refi decided by the ECB in the second half of 2008. In particular, all the models tell us that the ECB should have cut rates beforehand since main indicators of slack were pointing south and were clearly indicating a strong need to ease monetary conditions. However, this is history. Second, we observe that the model employing the sub-index for inflation is the one delivering the best overall fit. Third, the analysis of the residuals reveals that including the sub-index for inflation (rather than the overall index) is the option that delivers the best performance in terms of forecasting error over the last two years. This is the most interesting result of the analysis: it suggests that the very tough anti-inflation rhetoric of summer 2008 signaled a deviation from the ECB’s usual policy reaction function. Fourth, our analysis also warns that rhetoric alone cannot act as a substitute for actual data: a Taylor Rule based solely on our communication sub-indices in substitution for actual data on growth, inflation and money, delivers worse results than the ones based on data. This would suggest that while words are important, as they allow a better understanding of future ECB’s actions, actual figures must not be dismissed too easily. To put it differently, ECB communication helps market participants understand better the bank’s policy reaction function, but it is then crucial for any ECB watcher to monitor the data closely and feed them correctly into the reaction function.

The final step of our analysis is the assessment of ECB’s communication to extract a meaningful policy signal. We thus estimated our four Taylor rules listed above by means of an ordered probit model which evaluates the probability that the next move will be a hike, a cut or simply no change in the refi rate, i.e. whether forecasts of the refi rate improve or worsen when ECB’s rhetoric is taken into account. The working of this model employs a conventional Taylor rule estimation and adds a correction mechanism to predict the likely change in the cash rate towards a long run equilibrium rate. We show that the benefit from including the overall index in the reaction function is relatively low. When the sub-index on inflation is added, the increase in the share of correct predictions is higher (from 83% to 88%), with a sizeable increase in the forecasts of interest rate hikes (from less than 30 to more than 60%).

The bottom line of my reasoning here is that as far as the refi rate outlook is concerned, it will be what the ECB says on inflation that will certify that the central bank has changed posture and is ready to hike rates. In my view, in June 2008, we witnessed a sort of sudden switch of the ECB’s assessment on the inflation outlook that resulted in Trichet surprisingly flagging a rate hike at the next meeting. A similar evolution in next months cannot be excluded a priori. From what we have shown above, when the balance of risks is roughly balanced on different fronts – as it is right now – it is rhetoric on inflation that indicates the eventual ECB direction. In a scenario of smooth unfolding of the recovery, with a gradual rise in inflation, ECB communication will not need to adopt discrete jumps toward a more hawkish posture. Always stay tuned but in that case, especially if the ECB staff doesn’t alter materially its inflation outlook in December, don’t worry on rates until mid-2010.