In the Wake of the Crisis: Leading Economists Reassess Economic Policy

3. Lessons for Monetary Policy

International Monetary Fund
Published Date:
March 2012
  • ShareShare
Show Summary Details
Otmar Issing

During my time at the European Central Bank, I had a number of stimulating discussions with Olivier Blanchard, and when we disagreed—which happened from time to time—I always thought twice before continuing with my dissenting view. But he has encouraged the contributors to this volume to come up with controversial statements about the global economic crisis that began in 2008, so I have done what I would have done anyway: I examine in this chapter what I see as a flawed postcrisis consensus.1

Consider a statement by Lars Svensson, who is one of the gurus of the concept of inflation targeting, that shows how a strategy can be immunized against any critique: “In the end, my main conclusion so far from the crisis is that flexible inflation targeting, applied the right way and using all the information about financial factors that is relevant for the forecast of inflation and resource utilization at any horizon, remains the best-practice monetary policy before, during, and after the financial crisis.”2 According to this statement, if flexible inflation targeting has not worked as expected, either it was not applied properly or some information was missing. But the strategy was fine. In this way, you can continue with such concepts indefinitely, making mistake after mistake.

Inflation targeting was instrumental in bringing down inflation worldwide since the mid-nineties, especially in emerging countries. Some elements of inflation targeting are and should remain in consensus among all the central banks in the world—namely, the commitment to maintaining a quantitative definition of low inflation or a definition of price stability; to adopting a forward-looking policy; to presenting their views, strategy, and actions in a transparent way; and to communicating those to the markets and to the public in general. These important elements are part of inflation targeting but are also important elements in all cases in which there is no explicit inflation-targeting strategy.

One open question concerns the role of the inflation forecast. Historically, inflation targeting started with a simple concept, and its beauty was that you could look into your inflation forecast and know what to do. This was the starting point. Over time, though, the concept turned out to become more and more complicated. It soon was called inflation targeting with judgment and later flexible inflation targeting. Defining flexible inflation targeting as Lars Svensson has done is a nirvana approach.

One problem is that the inflation forecast is supposed to be a comprehensive summary of all elements that are relevant for inflation in the future. I doubt that this argument is valid.

Another issue concerns the time horizon of inflation targeting. At a 2003 conference at the Bank for International Settlements (BIS), inflation targeters discussed the time horizon for the inflation forecast. Initially, it was one and a half years, and then it was extended to two years. Lars Svensson asked, “Why not six or seven years?” I wonder what an inflation forecast with such a time horizon could really turn into. All concepts of inflation targeting are based on inflation forecasts in which money and credit do not play an active role. They are a passive part of the forecast but are irrelevant once it comes to monetary policy decisions.

Another issue related to time horizons is to what extent monetary policy should react to or try to deal with asset-price developments. I call this the Jackson Hole issue because it was presented at several conferences of the Federal Reserve Bank of Kansas City in Jackson Hole (e.g., 2002, 2005). Its three key points are that central banks should not target asset prices, should not prick a bubble, and should follow a mop-up strategy after a bubble has burst. I think these are all self-evident, and no central bank would act differently.

My question is always (and not only from hindsight) whether this is all. Having monetary policy dealing with asset prices only on the way down is an asymmetric approach. It implies that as long as asset prices go up, the central bank avoids responsibility. If asset prices collapse after a bubble bursts, then the central banks have to come to the rescue. I think that this asymmetric approach implies the risk of a series of ever-increasing bubbles because the causes that have led to the bubble are never corrected.

We should not repeat the mistakes of 1929 and the following years. When we are in such a situation, we have to provide the necessary liquidity. But does a totally asymmetric approach that implies a kind of guarantee from the central bank to rescue failed financial institutions not create moral hazard?

Whenever I criticize the asymmetry of this approach, I am asked how bubbles can be identified and whether central banks can know better than markets. But the issue is not identifying the appropriateness of the stock price of a company but rather assessing macroeconomic developments. A number of studies from the Bank for International Settlements and also from the European Central Bank have shown that nearly all bubbles in history have been accompanied or preceded by strong growth in money or credit.

Looking at money and credit can help the central bank to identify a risky macroeconomic situation, which leads to one of the questions that Olivier Blanchard has raised: to what extent should monetary policy contribute to containing financial instability? A major contribution would be to avoid the emergence of unsustainable developments in money and credit. But leaning against the wind might have two conceptual consequences—trying to identify asset-price misalignments and financial imbalances and leaning against unstable developments in money and credit. Both lead to the same concept.

One argument against such an approach has been that the central bank has only one instrument—the interest rate—and that this instrument is too blunt a tool to deal with asset-price developments. Interest rates would have to be raised substantially, which would create immense costs for the real economy.

But that is not a consequence of the concept of leaning against the wind. If started early enough, even small changes in the interest rate might have a substantial effect. In situations of emerging strong credit developments and unsustainable developments in asset prices, a number of financial actors typically are confronted with misalignments in maturity, so there is a maturity mismatch situation in which leveraging is going up. And changing interest rates is also a communication device that the central bank can use to signal that it cares, and it could support its communication by its publications and speeches.

With this kind of early action, it is not simple to know when to trigger. But there also is a risk in not acting soon enough. Acting early might also work against herd behavior.

As Olivier Blanchard prophetically said at the conference for my departure from the European Central Bank in 2006, “I worry that we have been lulled—or we have lulled ourselves—into a sense of complacency which is not warranted. There are still many issues we do not understand, and these may come back to bite us with a vengeance in the future.” He has also noted the elegance or beauty of models, and I think this beauty has contributed to the complacency. The initial concept of inflation targeting, for example, looked elegant. In 1999, when the European Central Bank started with monetary policy, it was criticized by academics across the board who did not like its concept, which was not elegant or seen as state-of-the-art. They asked for this approach to be framed in one model, but that was not possible. There is a Nobel Prize waiting for anyone who can combine money and credit quantities with the usual concept of inflation targeting.

I think inflation targeters are aware of this issue. They are trying to include a financial sector in their model by talking about frictions. This research concept one day may lead to better approaches in the future, but for now, it does not help central banks. If the inflation-targeting approach is augmented with market frictions, what is the result? What advice can be offered to monetary policy decision makers?

I think a better answer is an approach that tries to bring in analysis of money and credit. The European Central Bank3 called it monetary analysis, but from the beginning—not just from hindsight—the bank used the term for more than comparing M3 with its reference value. All aspects of monetary developments and credit were examined. This was extended over time. A book recently published by the bank demonstrates the extent of current monetary research.

We need to take other lessons from the global economic crisis. For me, dealing with the zero bound for interest rates by raising the inflation target is not a convincing approach for several reasons. Blanchard and his colleagues have dealt with the costs of inflation and have argued in favor of indexing the tax system, but this is difficult to implement and is not as efficient as stable money.4

But my main concern is the unavoidable loss of credibility of the central bank. Why should people believe that an upper limit of, for example, 4 percent inflation would not be increased again next time? How can inflation expectations be anchored with such an approach? The widening of the range would increase volatility. It would foster short-term activism. Finally, more leeway for reducing interest rates might be required than is needed with a smaller range. But this aspect of the problem has to be discussed further.

This debate also leads to the issue of dual versus single mandate. No central bank in the world would ignore developments in the real sector, such as unemployment. But what the central banks can deliver in the end is price stability and nothing else. Responsibility for financial stability finally is in the hands of regulatory and supervisory authorities.

Price stability and financial stability must not be seen as a tradeoff. Financial stability must be dealt with in the context of monetary policy that is geared to maintain price stability.

For a deeper analysis see Otmar Issing, “Lessons for Monetary Policy: What Should the Consensus Be?,” IMF Working Paper 11/97 (International Monetary Fund, Washington, DC, 2011).

Lars E.O. Svensson, “Flexible Inflation Targeting: Lessons from the Financial Crisis” (The Netherlands Bank, Amsterdam, September 21, 2009).

European Central Bank, “Enhancing Monetary Analysis” (Frankfurt, 2010).

Olivier Blanchard et al., “Rethinking Macroeconomic Policy,” Journal of Money, Credit, and Banking 42, no. 26 (2010).

    Other Resources Citing This Publication