Home > About the Bank > Speeches and Statements > Speeches 1996–2010 > Speeches 2000 > Remarks by Kazuo Ueda, Member of the Policy Board of the Bank of Japan, at the Geneva Conference on 'Asset Price Inflation: What to Do about It?' organized by the International Center for Monetary and Banking Studies and the CEPR on May 5, 2000

Remarks by Kazuo Ueda, Member of the Policy Board

Remarks by Kazuo Ueda, Member of the Policy Board of the Bank of Japan, at the Geneva Conference on "Asset Price Inflation: What to Do about It?" organized by the International Center for Monetary and Banking Studies and the CEPR on May 5, 2000

May 18, 2000
Bank of Japan

I assume that my role in this conference would be to discuss the Japanese experience on the relationship between asset prices and monetary policy during the last two decades or so. I am a bit hesitant in this role because the story is well-known.1 Hopefully there is something new in my remarks.

  1. See, for example, Ueda (1997) and Yamaguchi (1999).

The question I am supposed to address is: Was there anything wrong with Japanese monetary policy during this period, especially in relation to asset price movements? And, if so, why? It would be also nice if I could draw from the experience some general lessons for monetary policy making.

Let me briefly summarize some of the key features of the period. The sharp appreciation of the yen in the mid- to late 1980s through the economy into a mild recession. The Bank of Japan (BOJ) responded by cutting the discount rate five times between 1986 and 1987. The moves generated a quick recovery of the economy and a sharp rise in stock and land prices. The BOJ waited for about two years before started tightening in May 1989. It raised the discount rate five times until August 1990. Stock prices started to decline in early 1990 and were already down by about 50% by September 1990. Land prices started their descent in 1991 and are still declining now in 2000. The boom-bust cycle in asset prices generated the severe bad loan problem with serious negative effects on the economy. The 1990s became a lost decade for the Japanese financial industry and the economy.

Many critics argue that the expansionary policy in 1986-87 was overdone and the tightening in 1989-90 came too late. Let us for the moment follow this line of logic and consider why things went that way.

In a sense, the BOJ was responding too much to certain asset prices until around 1987, and not much to some others after then. In four of the five discount rate cuts in 1986-87, the BOJ officially acknowledged that concern over the deflationary impact of a higher yen on aggregate demand was an important factor behind the discount rate changes.

The concern by itself does not seem to have been a problem. The extent of the concern, however, was problematic. The same pattern was observed in the 1970s, especially in the early 1970s. The expansionary monetary policy at that time in response to a stronger yen led to a sharp rise in goods and service prices. Put simply, in the 1970s and 1980s, the BOJ seems to have been worrying too much about the deflationary effect of a higher yen on the economy and/or there were strong outside political forces on the BOJ to counteract such effects. Technically, it is not too difficult to find a significant coefficient on the exchange rate in the estimation of a policy reaction function of the BOJ even after controlling for something like expected inflation.2 This pattern played itself again in the mid- to late 1980s and probably made the monetary stimulus excessive.

  1. 2 See Ueda (1997), Table 9.4.

The BOJ, after the five consecutive cuts in the discount rate, sought ways to change the policy stance in the direction of putting more weight on avoiding future inflation. It in fact let short-term market rates go up in the summer of 1987, as the economy started to recover.

Then, however, came the Black Monday in October 1987. After that, there were no clear attempts to tighten for a while. Again, the BOJ was paying a lot of attention to movements in asset prices, this time a decline in foreign stock prices.

There was foreign pressure on Japan to maintain low interest rates. The pressure was evident in the communiqué issued after the Reagan-Nakasone and Reagan-Takeshita meetings in 1987 and 1988.

In 1988 other economies such as Germany started tightening. In Japan, the Nikkei 225 went above the pre-Black Monday level, real GDP was expanding at around 6%. The BOJ again sought ways to raise rates. But there were no friends. The reason was simple; inflation was not in sight. CPI inflation was only 0.7% in 1988.

In fact, there was optimism about Japan, its current account surplus, the Japanese style of management, and the increased presence of Japanese financial institutions in the world. People were pointing to increases in productivity. The managing director of the IMF arrived in Japan in February 1989 and said something like "there was absolutely no fear of inflation. Japan should keep its current monetary policy stance and play the role of the monetary anchor for the world."

Seeing some signs of inflation, the BOJ finally raised rates five times between May 1989 and August 1990. Did the moves come too late? It is difficult to answer the question. At least, one can say that the policy stance became somewhat backward-looking with no clear sign of inflation and no clear-cut recipe about how to respond to asset price inflation.

One could say, however, that there were no major mistakes on the inflation front. CPI inflation reached a peak in 1991 at only 3.3%. Though the judgment depends on one's view about the optimal inflation rate, not too many central banks would have done better on this front.

Those who are not satisfied with such an evaluation point to the serious bad-loan problem of the banking sector resulting from the boom-bust cycle in the stock and land markets, and its effect on the rest of the economy. There is an obvious point that the difficulty of the 1990s was mainly a result of the mishandling of the bad loan problem. Could a better monetary policy, however, have prevented the disaster of the 1990s from happening?

I do not think that inflation forecast targeting would have done it, if the forecast horizon was 1-2 years as is normally the case, for the reason I mentioned above.3 It would have been different with a much longer horizon such as ten years. BOJ officials in the late 1980s then would have thought about problems posed by the zero interest rate policy-a policy to be adopted ten years later to combat the deflationary forces stemming from the boom-bust cycle. To be so forward-looking is not easy, to say the least.

  1. 3Some adopters of inflation targeting, for example, the Bank of England, do not specify the time horizon for hitting the target. Formally, they attempt to hit the target at all times. It would be practically difficult, however, for central banks to have a long horizon of five to ten years in their inflation forecasts.

Another way out would have been to remember the historical lesson that a large boom-bust cycle in asset prices often leads to serious problems in the financial system with negative implications for the rest of the economy. In a sense this suggestion is similar to the first one above, but is more specific in its focus on the health of the financial system. It also leads to the view that the dividing line between monetary policy and prudential policy is not as clear-cut as some would argue. The problem with such a suggestion is, of course, it may result in an attempt to stop a healthy rise in asset prices.

A third way out might have been the Bernanke-Gertler solution.4 They argue that the BOJ should have increased the call rate in 1988 up to 8% or so against the actual of about 4%. But a quick reading of their analysis suggests that this conclusion is obtained because the Taylor rule, which they use to calculate the optimal call rate, suggests a rise in the rate in response to the sharp pickup in real economic activity then rather than to inflation. Raising rates just because real GDP is growing strongly but with no inflation is hard especially when favorable supply shocks are thought to be hitting the economy.

  1. 4 See Bernanke & Gertler (1999).

Thus, all these suggestions seem to be easier said than done.

What are the implications of the above discussion for the argument to set up a predetermined rule regarding the relationship between asset prices and policy instruments as seem to be claimed in the conference paper? In a sense, Japan failed by having something like this on the exchange rate as I argued. I hasten to add that more independence from outside forces would have helped the BOJ from acting in that way in the 1980s. In any case, it would be more important to analyze adequately what current asset price movements would mean for future inflation and GDP movements than to make policy instruments respond directly to asset prices.

In this sense I am close to the view of Bernanke and Gertler. As I pointed out above, however, one would require a very sophisticated version of inflation-forecast targeting, which either looks at a horizon of ten years or so, or at the delicate boundary between monetary and prudential policies. Or else, central banks need to be able to distinguish between supply shocks and demand shocks. In the end, optimal decision making under uncertainty would recommend some "response" to a sharp asset price inflation because of, to repeat, concern for future inflation stability. But the degree of the response would in no way be a pre-determined constant.

References:

  • Bernanke, B. & M. Gertler (1999) "Monetary Policy and Asset Price Volatility," Economic Review (Fourth Quarter 1999), Federal Reserve Bank of Kansas City.
  • Ueda, K. (1997) "Japanese Monetary Policy, Rules or Discretion? A Reconsideration," in Towards More Effective Monetary Policy, I. Kuroda (ed) , MacMillan.
  • Yamaguchi, Y. (1999) "Asset Price and Monetary Policy: Japan's Experience,"Remarks presented at the conference on New Challenges for Monetary Policy, Federal Reserve Bank of Kansas City.