Journal Issue
Share
Article

Responding to asset price booms

Author(s):
International Monetary Fund. External Relations Dept.
Published Date:
March 2003
Share
  • ShareShare
Show Summary Details

Interview with Bordo and Jeanne: Is “benign neglect” the right response to asset price booms?

IMF Survey: What are the main features of recent asset price booms and busts?

Bordo: We looked at asset price boom-busts in 15 OECD countries since 1970 and found a considerably lower incidence of stock market boom-busts than property price booms and busts. Out of 24 boom episodes in stock prices, only 4 were followed by busts: Finland (1989), Italy (1982), Japan (1990), and Spain (1990). Out of 20 booms in property prices, 11 were followed by busts. Thus, the probability of a stock market boom leading to a bust was 17 percent, but the probability of a property price boom leading to a bust was 55 percent. According to our measure, the United States, as a whole, never experienced a boom-bust in property prices. Such boom-busts tend to be more prevalent in small countries where the real estate market is more concentrated in the capital or major cities. We also found that banking crises tended to be associated with asset price boom-busts.

We also looked at the behavior of consumer price inflation, the output gap, and domestic credit during the boom-bust episodes. All three variables declined with, or following, the bust. This, we suggest, provides evidence that asset price boom-busts have significant and deleterious effects on the macroeconomy.

IMF Survey: Should central banks be concerned about stock and property price inflation as well as inflation in the price of goods?

Jeanne: This question increasingly preoccupies central bankers—rightly so in our view, after the Japanese experience of the 1990s and the recent slide in stock markets worldwide. Our answer is that, yes, in principle, there might be circumstances in which central banks should respond to asset prices and, more specifically, should restrict monetary policy preemptively in an asset price boom.

The right way to think about this question, we believe, is in terms of insurance. Restricting monetary policy in an asset market boom can be viewed as insurance against the risk of real and financial disruption induced by a later bust. This insurance obviously does not come free: restricting monetary policy implies a sacrifice of immediate macroeconomic objectives. But letting the boom go unchecked entails the risk of even larger costs down the road. It is the task of the monetary authorities to assess the relative costs and benefits of a preemptive monetary restriction in an asset price boom.

Some argue that monetary authorities should adopt a wait-and-see attitude and deal with the real and nominal consequences of a crash if and when one occurs, in the same way the authorities respond to other demand or supply shocks. But there is an important difference between external shocks and financial crises. Financial crises, unlike earthquakes, are in part inherent to monetary policy. Their severity is determined by the imbalances that build up in the boom phase, which, in turn, depend on the more or less accommodating stance of monetary policy. As a matter of logic, it is not optimal for the monetary authorities to ignore the risks inherent in their own actions.

IMF Survey: But it is difficult to assess whether an asset market boom will end up in a bust. Should this deter monetary authorities from restricting monetary policy preemptively?

Jeanne: It is a difficult task, but it should not deter the monetary authorities from trying. Obviously, this assessment must be based on an estimation of probabilities—we cannot demand that the authorities exhibit a significantly higher degree of prescience than the market. But taking again the insurance perspective, it seems clear that uncertainty about the sustainability of the boom is not in itself a reason for inaction—no more than a homeowner needs to be certain that his house will burn to take out fire insurance.

Of course, the monetary authorities may choose not to restrict monetary policy preemptively in an asset price boom. This would be the case, for example, if the monetary restriction required to contain the asset price boom were too severe. It is not a good idea to generate a recession now to reduce the risk of a recession later! This would also be the case if the problem can be better dealt with by other policies, such as banking and financial regulation. For example, raising the interest rate in response to a real estate boom may not be a good idea if the problem can be dealt with through more targeted measures in the mortgage market. So, inaction may sometimes be the best course of action, but this is not a justification for inattention. One should not take for granted that, as a rule, monetary authorities should ignore developments in asset prices and focus exclusively on inflation in the prices of goods.

IMF Survey: There is a growing debate about the links between monetary policy, asset prices, and financial stability. Where do you stand?

Bordo: The consensus view—presented in several papers by Ben Bernanke and Mark Gertler, and nicely summarized in a recent speech by Bernanke, in his new capacity as a Governor of the U.S. Federal Reserve—is that benign neglect is the preferred course. This view holds that asset price booms do not need the special attention of the monetary authorities. If the authorities follow the conventional inflation targeting approach based on the Taylor rule (where the interest rate is set as a function of deviations of forecasted inflation from its target and the output gap), asset price run-ups—to the extent they are captured by the inflation forecast and the output gap—will be offset by setting interest rates according to the rule. Moreover, if asset price busts do occur, they can easily be handled by reactive, lender-of-last-resort policy. A variant of this view by Stephen Cecchetti, Hans Genberg, and others says that, in the event of an asset price bubble, the central bank might want to explicitly incorporate asset prices into its policy reaction function.

Photo credits: Christian Charisius for Reuters, Page 81; Chip East for Reuters, page 81; Pawel Kopczynski for Reuters, pages 85; Denio Zara, Padraic Hughes, Pedro Márquez, and Michael Spilotro for the IMF, pages 86-87, 94-95; Cynthia Mar for the World Bank, page 89; STR for Reuters, page 89; Georges Gobet for AFP, page 91; and Claro Cortes IV for Reuters, pages 95-96.

Andrew Crockett, General Manager of the Bank for International Settlements, and Claudio Borio and Philip Lowe, economists there, oppose this view. They see asset price booms as a symptom of a fundamental macroeconomic imbalance common to episodes of stable prices, such as prevail today. In such an environment, credit booms can arise that will inevitably be followed by busts. This view holds that such booms, and especially asset price run-ups, should be dealt with through tight monetary and regulatory policy before they lead to a bust.

We place ourselves between these two positions. We see asset price boom-busts as low-probability events that can produce serious real consequences when the bust occurs. This, we argue, makes the case for preemptive policy in very special circumstances when it can be ascertained with a reasonable probability that the benefits of preemption (in terms of preventing a future costly bust) outweigh the costs of killing the beneficial effects of the boom (to the extent that it is a real productivity boom). When to do this, as Olivier mentioned, is a difficult empirical question. We feel that this issue should be addressed and resources should be devoted to doing so. We also believe that financial stability is an important policy issue that should not be treated as independent of overall macroeconomic stability.

IMF Survey: What is the optimal monetary stance in an asset market boom?

Jeanne: This question is much more difficult to answer in practice than in theory. In theory, the monetary authorities should weigh the cost of the monetary restriction to their immediate macroeconomic objectives against the benefits of reducing the risk of a crisis later. But what does this mean in practice?

In our paper, we explore the implications of this general principle in a very stylized two-period model. The model looks at an economy in which investors believe in the “New Economy” (a high rate of productivity growth) but still attribute some probability to an “Old Economy” scenario with lower productivity growth. If it turns out that the New Economy is an illusion, there is a market crash. The question is how the monetary authorities should respond to this situation in the boom phase, when it is still uncertain whether a crash will occur.

Although the model is quite simple, we find that the optimal monetary policy depends on economic conditions in a complex, nonlinear way. The preemptive monetary restriction seems to be optimal for an intermediate level of market optimism—when the boom is already visible but before it has too much wind in its sails.

This simple example is special in some ways, but it suggests a more general lesson. The optimal policy cannot be described in terms of a basic rule. The circumstances in which a preemptive monetary restriction is warranted cannot be summarized in the standard macroeconomic indicators, such as inflation or the output gap. They involve imbalances in the balance sheets of the private sector, as well as market expectations.

Complexity is not a good reason to ignore the issue, however. Well, let me take that back. Maybe it’s a good reason most of the time. But there might be times when the risks are too large to ignore. In other words, we are not saying that monetary authorities should routinely target the price of assets the same way that some central banks target consumer price inflation. But exceptional developments in asset markets may require occasional deviations from the rules that should prevail in normal times. That’s true not only in a crash but also in a boom.

IMF Survey: Are your model’s conditions for a proactive monetary policy met in the real world?

Jeanne: That’s an important question that deserves further research; in other words, you won’t find the answer in our paper! The evidence suggests that, historically, there have been many booms and busts in asset prices and that they are associated with a great deal of economic disruption. So this problem should certainly be of concern to policymakers. Calling for more research on this question is as far as we go in our policy recommendations.

Why didn’t we go further in developing the practical implications of our analysis? To a large extent, the reason is technical. We would need to simulate and compare alternative monetary rules in realistic models of the economy. And we think it is important for these models to involve the kind of nonlinearity and low-probability events that we emphasize in our study. Now, this raises serious technical difficulties: nonlinear dynamic models are technically challenging. To be realistic, the models cannot be too simple, and even simplistic models can give rise to very complex dynamics once they are nonlinear. However, nonlinearity seems difficult to abstract from when it comes to financial crises, so some ways must be found to solve or circumvent these technical difficulties. We hope our paper will encourage more research on this.

IMF Survey: And other implications for policymaking?

Jeanne: We believe that our analysis has important implications for monetary policy. It suggests flexibility is needed to account for extreme events and especially severe financial instability. We see these as nonlinear events that cannot simply be dealt with by the conventional inflation targeting framework followed by most advanced countries. We see financial stability as a potential Achilles’ heel for the current central bank paradigm, and we caution against complacency by central bankers who may believe that inflation targeting is all they need to do.

Copies of IMF Working Paper No. 02/225, “Monetary Policy and Asset Prices: Does “Benign Neglect” Make Sense?” by Michael D. Bordo and Olivier Jeanne, are available for $15.00 each from IMF Publication Services. See page 93 for ordering information. The full text is also available on the IMF’s website (http://www.imf.com).

Other Resources Citing This Publication