Introductory remarks by Athanasios Orphanides, Governor of the Central Bank of Cyprus, at the Fifth ECB Conference on Statistics: "Central Bank Statistics: What Did the Financial Crisis Change?"
Frankfurt, 20 October 2010
I am honoured to be chairing the closing session of the fifth ECB conference on statistics which coincides with World Statistics Day. The topic of the session, "The role of statistics in central bank communication", is of particular interest because effective communication is a very important policy tool. For that reason, we must always seek ways to improve central bank communication and to understand how statistics can help enhance its clarity and effectiveness.
A key challenge is how to provide better information and how to help households and businesses better understand the economic and policy environment around us, and thus to take better decisions. This requires awareness of the limited capacity humans have for processing information, an inherent limitation we all face. When confronted with voluminous and unprocessed information, people are not always able to quickly analyse it effectively and to extract the most useful elements needed to improve decisions.
Policymakers need to be sensitive to communicating information where the useful content dominates the unavoidable noise in the statistics being communicated or where the uncertainty associated with the statistics provided can be explained reasonably well. Care is needed in deciding what is communicated and how this is done in order to ensure that communication reduces the uncertainty faced by the intended audience and facilitates better decisions. In other words, effective communication requires clarity. More communication is not always necessarily better, especially when it may inadvertently serve to mislead, rather than to inform on concepts that may be critical for good decisions.(1)
Statisticians are well aware of the imperfections associated with the creation of statistical series. When we describe the overall economy even under the best of circumstances, some irreducible uncertainty may remain on how well a statistical series can capture the underlying concept it attempts to measure. Statistics may only provide a proxy for a concept that may be critical for decisions. And the creation of aggregate data series invariably introduces noise that is due, for example, to breaks in concepts, imputations, aggregation limitations and so forth. Effective communication of the uncertainty surrounding statistics is thus an important but challenging task.
The crisis we have been going through has given us additional examples of challenges associated with central bank communication. Let me briefly mention two examples that also serve to remind us of the need to adapt the communication of statistics as circumstances change.
One example relates to the communication of the stance of monetary policy. As a result of the crisis, central banks around the world, including the European Central Bank, the Federal Reserve System and the Bank of England, have reduced their official policy rates to unprecedentedly low levels. They have also implemented a series of unconventional policy measures to engineer an additional easing of monetary policy beyond what would be reflected in those official rates. But how are we to communicate precisely what the monetary policy stance is in these circumstances? This presents a communications challenge that has not, in my view, been solved very clearly—some of the statistical series used to this end before the crisis are not necessarily as useful indicators of the monetary policy stance as they were then. For example, policy rates are no longer sufficient policy indicators. Under these circumstances, all other things being equal, an additional expansion of a central bank’s balance sheet would, for instance, constitute an easing of the bank’s monetary policy stance. Conversely, adjustments to unconventional measures without changing official rates could constitute a tightening of a central bank’s monetary policy stance. But it may be difficult in the given circumstances to communicate the monetary policy stance as accurately as was possible before the crisis, even if the central bank is as clear as can be not only on its policy rates, but also on the size and composition of its balance sheet and on any changes to other unconventional measures. This is an example of the importance of striving to adapt and find effective ways of communicating with the public in a changing environment.
The second challenge I would like to mention concerns the measurement of incipient macroeconomic imbalances. This is particularly important in the context of the efforts that are being made at the international level to improve macro-prudential supervision. As has already been highlighted at this conference, this is one of the key areas in which we need to develop new statistics and improve the statistics we already have. As examples, let me mention a number of issues where we need to improve our measurement so as to be able to provide better answers to the questions raised. When does credit expansion become unwarranted? How should we be measuring this concept of credit in order to understand it as a macro-prudential risk? When does leverage in the financial sector become excessive? What is the most appropriate concept of leverage we ought to measure to assess aggregate risks? These are all issues that have been identified as quite important in the aftermath of the crisis, but for which we need to develop better statistical tools to guide us going forward. This is essential to help us understand and avert crises in the future.
Statistics are critical for central bank communication and effective communication is a crucial element of central bank policy-making. But a number of questions remain with respect to identifying best practices. The two presentations by Andreas Ittner and Petra Geraats, and the corresponding discussions by Elga Bartsch and Frank Paul Weber, offer us different points of view and help guide us through some of the issues.
-------------------------------------------------------------
ENDNOTES:
(1) See Dale et al. (2008) for an example of pertinent trade-offs in central bank communications.
REFERENCES:
Dale, Spencer, Athanasios Orphanides and Par Osterholm (2008) "Imperfect central bank communication: information versus distraction", Central Bank of Cyprus Working Paper 2008-1.