Asset Prices and ECB Monetary Policy

The ECB and the Eurosystem should normally take asset-price movements and potential asset-price bubbles into account only to the extent these are deemed to have an impact on the in‡ation and output-gap forecasts that should guide monetary policy. Asset prices should not be separate target variables, additional to in‡ation and the output gap. The explicit or implicit objective of …nancial stability, including an e¢ cient payment system, is best handled with e¢ cient supervision, including a regular and transparent Financial Stability Report, with indicators of …nancial stability and early warnings for appropriate regulatory and supervisory action. Then, …nancial stability can be seen as a constraint on monetary policy that, under normal times, is not binding and does not a¤ect monetary policy. In crisis situations, however, …nancial stability may be a binding constraint on monetary policy and typically induce more expansionary monetary policy. To what extent, and how, should monetary policy in the Euro area take asset-price movements and potential asset-price bubbles into account? I believe the answer to this question follows from the general principles for good monetary policy, as explained, for instance, in Svensson [2] and [3]. Furthermore, whereas the principles for good monetary policy are simple, the practice of good monetary policy is di¢ cult. The same is the case for the question of how to take asset prices into account in monetary policy: the principles are simple, but the practice is di¢ cult. So, the principles of good monetary policy are simple: Perform ‡exible in‡ation targeting, which means aiming to stabilize in‡ation around an explicit low positive numerical in‡ation target with some weight also on stabilizing the real economy, which can be expressed more precisely as stabilizing the output gap, that is, stabilizing output around a measure of potential output. Brie…ng paper for the Committee on Economic and Monetary A¤airs (ECON) of the European Parliament for the quarterly dialogue with the President of the European Central Bank. I thank Kathleen Hurley for editorial and secretarial assistance. Expressed views and any errors are solely my own responsibility. Because of the lags between monetary-policy actions and the e¤ect on in‡ation and output, the best way to do this is to look forward and perform forecast targeting. This means setting the central bank’s instrument rate (more precisely, to choose an instrument-rate plan, a path for the current and future instrument rate) such that the corresponding in‡ation and output-gap forecasts “look good,”which in turn means that the in‡ation and output-gap forecasts approach the in‡ation target and zero, respectively, normally some 1–3 years ahead (but, more precisely, the whole future forecast paths should look good, not just the forecast at some …xed horizon). Although these principles are simple, as explained in Svensson [3], the practice of constructing forecasts, deciding on the appropriate instrument rate (plan), and communicating these to the general public and the market is quite complicated and di¢ cult. How do these principles apply to asset prices? A …rst issue is whether asset prices should also be considered targets of monetary policy. I believe they should not. The principles above refer to stabilizing in‡ation and the output gap as the objectives of monetary policy. In‡ation and the in‡ation target are typically speci…ed in terms of an index of …nal goods and services, a consumer price index or a variant thereof, such as a core measure. I see no good reason to include asset prices separately in the price index. Asset prices should not be separate target variables for monetary policy. But what about …nancial stability? Financial stability, including an e¢ cient payment system, is normally an explicit or implicit separate objective for the central bank. Clearly, asset-price movements and bubbles can threaten …nancial stability. I believe the best way to deal with …nancial stability is by e¢ cient supervision, including a regular and transparent Financial Stability Report, produced either by the central bank or by a separate …nancial-supervision authority, with various indicators of …nancial stability that serve as early-warning indicators for necessary regulatory or supervisory action. With such informed and forward-looking action, under normal circumstances, the risk of …nancial instability will be small, and it will not be a concern for monetary policy. In line with this, I believe that the objective of …nancial stability and an e¢ cient payment system can be seen as a constraint on monetary policy. Under normal circumstances, and with e¤ective supervision, this constraint is not binding and has no e¤ect on monetary policy. Only under abnormal circumstances, with very unfavorable shocks, gaps or mistakes in …nancial supervision, large …nancial fraud, etc., would …nancial stability be threatened and then be a constraint on monetary policy. Such a constraint would typically force the central bank to conduct more expansionary monetary policy, for instance, because sizeable parts of a weak