Home > Announcements > Speeches and Statements > Speeches 2001 > A paper based on a speech given by Kazuo Ueda, Member of the Policy Board of the Bank of Japan, at the semi-annual meeting of the Japan Society of Monetary Economics on September 29, 2001 (Japan's Liquidity Trap and Monetary Policy)

Japan's Liquidity Trap and Monetary Policy

Based on a speech given by Kazuo Ueda, Member of the Policy Board, at the semi-annual meeting of the Japan Society of Monetary Economics held at Fukushima University in Fukushima City on September 29, 2001. It is a translated and revised version of the Japanese text released in October.

December 27, 2001
Bank of Japan


I. Introduction

I left my career in academia to become a member of the Policy Board, the highest decision-making body of the Bank of Japan (BOJ), in 1998. Since then, I have received various policy recommendations from academics, which I greatly appreciated, but I must say that some of them were based on misapprehensions due to insufficient knowledge of the actual economic situation and monetary policy. On the other hand, many in the academic world are clearly discontented with the BOJ's allegedly timid approach to monetary easing. Filling the gap between the views of the BOJ and the academic world is not easy, but I hope my speech today will contribute to better understanding between the two.

Simply put, economists seem to have been advocating propositions that hold true under normal circumstances, that is, when interest rates are well above zero. The BOJ has been groping for policy options that are valid near a liquidity trap. For example, some outsider asserted that if the BOJ cannot push down money market interest rates any further due to the zero bound, it should next simply increase the monetary base. It is almost a definition of a liquidity trap, however, that an increase in the monetary base by itself produces only minor effects on the economy. Life would have been very easy for the BOJ had simple increases in the monetary base led to a significant rise in the general price level. Instead, the BOJ has been thinking very hard and carried out a variety of policy measures, attempting to overcome the difficulty generated by the liquidity trap. In this speech I will first present theoretically what central banks can do near a zero interest rate, and then look back at the BOJ's past monetary policy. I regret, however, that I cannot comment in detail on the future course of monetary policy for obvious reasons.

II. Monetary Policy Near Zero Interest Rates

A. Constraints on the BOJ's Monetary Policy During 1995-1999

There have been broadly two major constraints on monetary policy since the mid 1990s. One of them has been the limited downward flexibility of the overnight call rate because the rate had already fallen below 1 percent in the second half of 1995 (see Chart 1). The other constraint has been the limited power of whatever monetary easing measures implemented to stimulate the economy because of the non-performing loan problem of the Japanese financial system.

In the face of near zero to negative rates of inflation during the period, many economists, including academics, have expressed the view that prices would rise if the BOJ increased the monetary base sufficiently. However, this view does not seem to fully take into consideration the two constraints mentioned above. Consider the equation "Hv=PY" where H, P, Y, and v represent the monetary base, the general price level, real GDP, and the velocity of the monetary base, respectively. The view that prices and nominal GDP (PY) will respond to increases in H is correct if v is stable. However, as shown in Chart 2, v has declined steadily since the early 1990s. Thus, nominal GDP has either increased only moderately or decreased despite significant increases in the monetary base. At times it was even difficult to increase the monetary base simply because people did not want it any more.

Normally, an increase in the monetary base induces declines in interest rates, which in turn stimulate the demand for goods and services. The increase in the monetary base supports lending by financial institutions to firms and households purchasing goods and services. However, this mechanism has been impaired since the mid 1990s because of very low interest rates and the non-performing loan problem.1 In the following I will discuss the BOJ's monetary policy mostly in relation to the first of the two constraints, i.e., the zero bound on nominal interest rates.

  1. This point is easier to understand using a theoretical model based on the assumption that bonds and bank loans are imperfect substitutes for each other, rather than a simple IS-LM model. Bernanke and Blinder (1988) use such a model to show that a downward shift in the credit supply function brings about deflationary effects on the economy.

B. The Liquidity Trap and Monetary Policy Options

The "liquidity trap" has been defined in many different ways, generating some confusion. Here I would like to simply define a liquidity trap as a situation where there is no room for nominal interest rates to decline. According to this definition, Japan's money market has been nearly in a liquidity trap for a few years. As for long-term interest rates, however, it is difficult to judge whether they can decline any further beyond recent levels.2

Three options for further monetary easing can be considered when money market interest rates are near zero. First, if the short-term rate is very low but not zero percent, the rate can be lowered further to as close to zero as possible by increasing the monetary base. Second, the BOJ can influence expectations of economic entities by promising to continue monetary easing into the future. Some have called this the commitment or policy duration effect. Third, the BOJ can carry out unconventional operations by purchasing assets other than short-term Japanese government securities.

Let us discuss the three options more carefully. First, it is possible to reduce money market interest rates to as close to zero as possible by increasing the monetary base until the economy falls literally into a liquidity trap. After entering the trap, it becomes almost meaningless to merely increase the monetary base. People hold money instead of short-term financial assets as they derive utility from the liquidity of money even at the cost of forgone interest. In equilibrium, the marginal utility of holding more money and the opportunity cost of holding money instead of interest-bearing assets, that is, the nominal interest rate, must be the same. Thus, when money market interest rates are virtually zero, the marginal utility of holding the monetary base is almost zero, which means, it is simply meaningless to supply more money, or people are satiated with the liquidity service money provides.3 In fact, undersubscription, the situation where the total amount of bids in money market operations falls short of the total offer, has been frequently observed in the BOJ's funds-supplying operations.

The second policy option, the duration effect, is the commitment to maintain near zero interest rates or an ample supply of liquidity not just in the immediate future, but far into it. Such a commitment will contain expectations of a future rise in short-term interest rates. Current long-term interest rates, which are approximately the average of expected future short rates, will go down accordingly and affect prices of other financial assets and in turn the demand for goods and services.4 Krugman has also argued that the commitment will increase inflation expectations of economic entities and cause a fall in real interest rates, thus directly stimulating the demand for goods and services.5

The third policy option is for a central bank to purchase non-traditional assets such as government bonds, foreign currencies, corporate bonds, stocks, or real estate which are more imperfectly substitutable for base money than are short-term government securities. As stated above, central bank operations that amount to the exchange of perfect substitutes produce little effect on the economy. Such non-traditional operations are effective because they directly alter the prices of the assets in question. Possible benefits and costs of this monetary policy option, however, are extremely uncertain.

More specifically, economists have long discussed whether long-term government bonds are perfect substitutes for short-term government securities. If they are, they do not qualify as a candidate for non-traditional operations. In fact, the second option above, the commitment in terms of policy duration, produces largely the same effect as direct purchases of long-term government bonds. However, if long-term government bonds are not perfect substitutes for the monetary base or short-term government bills, in other words, if there are risk premiums on long-term government bonds, a central bank may be able to influence long-term interest rates by purchasing long-term government bonds.

Let me make some comments on operations involving foreign exchange. First, in Japan the responsibility for operations in the foreign exchange market to stabilize exchange rates is assigned by law to the Ministry of Finance, and not to the BOJ. Second, there is a view that the effect of interventions on the foreign exchange rate is reduced if they are sterilized. It is, however, almost meaningless to discuss the distinction between sterilized and unsterilized interventions under a zero interest rate. This is because the difference between the two only amounts to whether to conduct the BOJ's operation in short-term government securities and allow an increase in the monetary base: base money and short-term securities are almost perfect substitutes for each other as discussed earlier. Finally, for the same reason, the Ministry of Finance is in a position to carry out a non-traditional monetary policy operation under a zero interest rate. That is, it can buy foreign exchange by issuing treasury bills. The operation is almost the same as the BOJ buying of foreign exchange by expansion of the monetary base.

  1. 2Keynes' notion of a liquidity trap is a situation where long-term interest rates become so low that people start to expect that they can only rise and thus prefer to hold money rather than bonds to avoid capital losses. In such circumstances, long-term interest rates cannot go lower. However, whether such a situation can be a stable equilibrium is an interesting topic for future study. Taking the case of Japan, the ten year Japanese government bond (JGB) rate fell to below 0.8 percent in September 1998 but quickly rebounded to above 2 percent in December 1998. The situation in that autumn may or may not have been a liquidity trap.
  2. 3As described in a later section, the BOJ has recently driven the overnight rate to what is probably a minimum positive level, but not completely to zero. For this or for other theoretical reasons, people may still derive some benefit from larger money balances. In fact, the current BOJ policy stance, though it does not fully rely on such effects, encompasses such a possibility. Simulations on the size of such effects carried out by, for example, McCallum (2000) suggest that the effects, if any, are very small.
  3. 4More precisely, one might argue that, as long as rational market participants consider it appropriate for a central bank to maintain the short-term interest rate at zero percent, they consider that a central bank will most likely do so. Hence, the central bank's commitment to continue its easy monetary policy into the future is effective only if it promises to maintain the interest rate at an extremely low level for some time even after it becomes appropriate to raise it. Even if the central bank does not decide to go this far, however, its strong commitment to continue the monetary easing stance into the future will reduce uncertainty in future monetary policy and produce easing effects. For details, see Ueda (2000).
  4. 5See, for example, Krugman (1998).

III. Monetary Policy from the Late 1990s to the Present

A. The Liquidity Crisis of 1997-98 and Monetary Policy during the Period

Based on the theoretical analysis presented earlier, I will next discuss monetary policy implemented in the late 1990s to the present, starting with the period between 1997 and 1999. During 1997-1998 the failure of Japanese financial institutions such as Sanyo Securities in 1997, followed by Long-Term Credit Bank of Japan and Nippon Credit Bank in 1998, generated a sense of crisis in the Japanese financial system. The international financial crisis triggered by the Russian government's default on its bonds added to the stress. Risk premiums and liquidity demand increased sharply across financial markets. Thus, in the international financial markets, the credibility of Japanese banks deteriorated and the Japan premium, the additional interest rate Japanese banks have to pay when raising U.S. dollar funds, expanded to unprecedented levels. In the money market in Japan, the term premium expanded reflecting an increase in financial institutions' liquidity demand. Moreover, some banks had to pay much higher term premiums than others. The economy went into a serious liquidity crunch. Bank lending and spending on goods and services dropped precipitously.

The BOJ responded to the crisis by fully implementing the three monetary policy options mentioned earlier. It sharply increased the supply of liquidity to induce a decline in money market interest rates, and the overnight rate declined from around 0.5 percent in September 1998 to close to zero percent in March 1999. In April 1999, the BOJ announced its commitment to maintain the zero interest rate until deflationary concerns had been dispelled--the use of the duration effect. The combination of a zero interest rate and the duration commitment has sometimes been dubbed "the zero interest rate policy." To support these measures, the BOJ carried out some non-conventional operations; it expanded the scope of CP repo operations, and added corporate bonds and asset-backed securities to the list of collateral eligible in the BOJ's market operations. It also refinanced private banks' loans in proportion to the increase in their volume.

Even after April 1999, calls for more easing measures, usually under the heading of "quantitative easing," remained strong. However, such calls seem to have been based on a misunderstanding of the working of monetary policy being pursued at that time. The reasons are as follows. First, as mentioned earlier, quantitative easing by way of increasing the monetary base is theoretically of little value for the economy in a liquidity trap, i.e., when money market interest rates are virtually zero percent. Second, even Paul Krugman, a strong advocate of "quantitative easing" in regard to Japan's monetary policy, admits that quantitative easing through increasing the monetary base is by itself ineffective for the economy in a liquidity trap and proposes other alternatives. His idea is for the BOJ to commit to future easing, for example, to permanent rather than temporary increases in the monetary base, which will have an impact on current inflation expectations. Furthermore, he points out that such a policy can alternatively be carried out by promising to maintain the interest rate at a low level even after the inflation rate starts to rise.6 The zero interest rate policy the BOJ implemented from April 1999 was an example of exactly such a policy, and can also be considered as a weak form of inflation targeting. In other words, the BOJ was indeed conducting a "Krugman-type" quantitative easing. It was unfortunate that this point was not fully understood by many economists. To be fair, however, the BOJ's commitment was a little ambiguous and less ambitious than Krugman's proposal. He proposed adopting a high inflation target of 4-5 percent but the BOJ's commitment was a much more moderate one of "until deflationary concern has been dispelled."

  1. 6See Krugman (1998).

B. Economic Recovery Leading to the Raising of the Overnight Call Rate Target in 2000

The liquidity crisis in 1997-98 gradually subsided due to measures such as the maintenance of the zero interest rate policy,7 injection of public funds into financial institutions, and enhancement of the credit guarantee system for small firms. As a result, business fixed investment and private consumption recovered. Over time, in the technology-related sectors, a virtuous circle supported by the worldwide IT boom started to develop involving exports, production, profits, and business fixed investment, bringing the economy gradually into a moderate recovery phase. In the summer of 2000, some began to view the economy as capable of achieving higher than potential growth. Thus, the majority of the BOJ's Policy Board members thought the output gap was beginning to narrow and supported an increase in the uncollateralized overnight call rate target. The rate was raised to 0.25 percent in August 2000.

However, in order to terminate the zero interest rate policy, it was first necessary to determine the optimal level of the policy interest rate. Even if the output gap was shrinking, and thus the optimal policy interest rate was on the rise, it could not be determined whether the level was positive or negative. Under the zero interest rate policy, it was possible that the optimal interest rate was actually negative even though the policy interest rate was close to zero percent, since the latter could not go lower. I estimated the optimal policy interest rate using the Taylor Rule, and the result was that the optimal rate was rising but the level was still negative. This was the main reason why I voted against terminating the zero interest rate policy at the Monetary Policy Meeting in August 2000. However, it should be noted that the result of the calculation could differ depending on the assumption used, and thus it is possible that the optimal policy interest rate calculated using a different assumption would have been positive. It is extremely difficult to accurately calculate the optimal policy interest rate, and this seems to be an area requiring further research by economists.

  1. 7This experience is an example of the strong power of the central bank to contain a liquidity crunch. I did state earlier that the non-performing loan problem weakened the BOJ's power to stimulate the economy. With a weak banking system, increased supply of the monetary base by the BOJ does not smoothly lead to increases in loans. But the BOJ was still capable of accommodating a surge in liquidity demand arising from a credit crunch and easing the crunch.

C. The Slowdown of the Economy from the Fourth Quarter of 2000 and the Introduction of a New Policy Framework in March 2001

Toward the end of year 2000, it became clear that the U.S. economy was starting to slow down as a result of the end of the IT boom, with obvious implications for the Japanese economy, especially its manufacturing industry. In addition, the non-manufacturing industry remained weak, and fiscal policy continued to tighten moderately, and thus the economic outlook became cautious. Against this background, the BOJ lowered the uncollateralized overnight call rate target and the official discount rate in February 2001. In March, the BOJ introduced drastic monetary easing measures: the main operating target for money market operations was changed from the uncollateralized overnight call rate to the amount outstanding of financial institutions' current accounts (henceforth, reserves) at the BOJ. The target amount was set at 5 trillion yen against around 4 trillion yen earlier in March. The BOJ also announced that it might increase the amount of its outright purchases of JGBs when necessary to provide liquidity smoothly. And in terms of policy duration, the BOJ made a commitment to maintain the new framework until the consumer price index (CPI) registered stably a year-on-year increase of zero percent or more.

This new framework of monetary policy includes the three monetary policy options mentioned in section II above, and it is quite similar to the previous zero interest rate policy. Money market interest rates have declined to close to zero percent as a result of the large injection of liquidity. The commitment in terms of policy duration is a common feature of the two policies.

However, there are some differences between the new monetary policy framework and the previous zero interest rate policy. First, under the new framework, money market interest rates can rise above zero percent, for example, due to a surge in liquidity demand toward the end of an accounting period or on days like September 12. In this sense, the zero interest rate policy with its explicit pre-commitment to a zero interest rate had a stronger easing effect. Second, some board members hoped that a substantial increase in reserves would generate additional easing effects on top of the effects expected as a result of lowering interest rates.8 Third, the BOJ's commitment in terms of the duration of the new framework is more explicit than that of the zero interest rate policy, and for this reason, it has a much stronger effect. Finally, under the new framework, an increase in the purchase of JGBs may lower long-term interest rates through a decline in the risk premium.

  1. 8See footnote 3. As argued in section II, however, the theoretical case for such an effect seems very weak.

D. Experience with the New Framework

Using the new framework, the BOJ has made a few policy moves since March. By August 2001, the economic outlook had deteriorated further. In response, the BOJ raised the reserve target to 6 trillion yen from 5 trillion yen and increased the purchases of JGBs by 50 percent. In early September, uncertainty mounted in the financial markets due to the terrorist attacks in the United States, and as a result liquidity demand increased sharply. To meet this demand, the BOJ temporarily increased reserves to around 8 trillion yen after September 12. Given that the demand for liquidity remained high into the next week, the BOJ formally decided to increase the target to "over 6 trillion yen" at the Monetary Policy Meeting on September 18.9 The ambiguity in the target reflected the perception that the instability in the demand for liquidity made it difficult for the BOJ to choose a specific target. Since then, the amount of reserves has fluctuated between 7.5 to 14 trillion yen.

We may provisionally summarize the effects of the policy moves since March as follows:(1) The increases in liquidity have driven short-term rates down to nearly zero and together with the duration effect, they have produced similar effects on asset prices as did the previous zero interest rate policy. That is, interest rates in the money markets and short to medium term bond markets have declined to unprecedentedly low levels. Between March and early June, the yen weakened against other currencies and stock prices rebounded from a low in early March.

(2) Rather unexpectedly, the daily average overnight call market rate has come down to as low as 0.001 percent as opposed to 0.02 percent in the previous zero interest rate period10 (Chart 4). This has certainly been a result of the massive liquidity injection.

(3) It is perhaps fair to say that we have not yet found any clear-cut evidence of favorable effects on the economy of a larger supply of liquidity, aside from any effects lower interest rates have generated. The reserve level in early December is roughly double the level in July. The monetary base is expanding at around 15 percent year on year. Yet the only clear-cut effect of this increase in reserves has been a small decline in money market rates as pointed out above.

(4) Despite the semi-inflation targeting nature of the current framework, inflation expectations have been declining continuously as shown in Chart 3.11

(5) Long term interest rates have not declined in response to our easing. Rather, they rose in August as the BOJ decided to buy more JGBs.

(6) The unprecedentedly low levels of money market rates have in a sense impaired the smooth functioning of the market mechanism and have lowered the usefulness of the amount outstanding of reserves as an indicator of the easing stance of monetary policy.On points (1) and (3), I must add that given the usual lag in the effects of monetary policy on the economy, we will continue to watch whether any additional effects of the current policy stance will be generated in the future. We will discuss points (4) and (5) more carefully below.

Point (6) would require some more explanation. As a general rule, suppliers of funds need to pay commissions to brokers, pay fixed costs for using the payment system, use various other resources for decision making, worry about borrower credit risks and thus are reluctant to make investments below a certain level of the interest rate.12 The threshold rate may differ among suppliers. Once the market rate falls below the highest threshold rate, some suppliers no longer provide funds to the market and are content with holding funds at their accounts at the BOJ which pay no interest. Those requiring liquidity then find it difficult to satisfy their need fully in the market and turn to the BOJ for more funds. The BOJ is obliged to respond by supplying more reserves.13 The larger the amount of funds stuck in the accounts of those with high threshold rates, the looser will be the relationship between the total amount of liquidity and the extent of easiness of monetary policy.

More specifically, after September 11, the demand for short-term dollar funds surged among Japanese financial institutions and institutional investors. Initially, the increase in demand was the result of concern over the U.S. payment system following the terrorist attacks. Over time, investors' attention shifted to buying medium to long-term U.S. bonds by borrowing in the dollar money market. Many Japanese investors thus entered into yen-dollar swap contracts with foreign banks to borrow dollars in the short term. On the other side of the transaction, many foreign banks acquired large amounts of funds in yen. Those foreign banks with high minimum required returns have stopped investing the yen acquired in the Japanese money market. These banks have still been making money because the average yen funding cost in the swap transactions has been negative for quite some time. As a result, a significant portion of funds supplied to the market by the BOJ has not been channeled into those requiring liquidity. During the reserve accounting period of October 16-November 15, the average amount of reserve supply was around 9 trillion yen, 50 percent above the August level; yet there were days when money market rates went up, if only slightly.

At this point, it is hard to know how long such malfunctioning of the money market will continue. The lingering non-performing loan problem seems to have added to the demand for liquidity as well. Given such a background, the BOJ has had difficulty in choosing a specific reserve target. Consequently, the policy directive since September has been to maintain reserve supply at above 6 trillion yen. Put differently, the current operating procedure places more emphasis on interest rate stability than before September.14

  1. 9The BOJ also decided to reduce the official discount rate to 0.1 percent.
  2. 10As discussed below, the 0.001 percent level is probably a positive minimum for the overnight rate. See also footnote 12.
  3. 11The small rise in the inflation expectation for 2001 in March was recorded before the policy change was announced on March 19.
  4. 12After allowance only for broker commissions and user fees for the payment system, an investor must lend at least 150 million yen to earn a positive net interest income at an overnight rate of 0.001 percent. And, he or she will make less than a dollar per day.
  5. 13Of course, a straightforward application of the current framework would have been to leave the reserve supply unchanged and let interest rates rise. The expected increases in money market rates, not just the overnight rate, however, seem to have been disruptively large. An alternative response would have been to raise the reserve target sufficiently. But then, the target would have become very unstable.
  6. 14At the Monetary Policy Board Meeting on December 19, the BOJ, in response to further deterioration in the economic outlook and signs of stress in the financial system, decided formally a new target range for reserves of between 10 to 15 trillion yen. It also decided to increase the purchase of JGBs by 200 billion yen per month. Additionally, it announced a few measures directed at easing strains in money and capital markets, including more intensive use of repo operations in the CP market. The first two measures were quite similar to those adopted in August. The wide target range for reserves, however, reflects the problems the BOJ has been facing with the use of the target as discussed in this section.

E. Increased Purchases of JGBs, Inflation Expectations and the Long-Term Interest Rate

As pointed out above, the increase in the purchase of JGBs did not lower long-term interest rates in August. This may not be surprising because the BOJ's commitment in terms of policy duration, which has essentially a similar effect to purchase of JGBs by the BOJ, had already been in place since March. Long-term interest rates, however, rose after the policy decision in August. One possible interpretation of what happened is that the market, observing, in addition to the monetary easing in August, the intensified pressure on the BOJ to adopt inflation targeting, became nervous about inflationary risks in the future. Chart 5 shows an interesting pattern of movements in 1-year implied forward rates calculated from the yield curve on JGBs. Implied forward rates rose between the 13th and 15th of August, the period around the decision on monetary easing, and rose further on the 21st when the debate on inflation targeting intensified. Thereafter, implied forward rates declined to the 24th, except for those starting nine years ahead, in response to negative statements about inflation targeting by Prime Minister Koizumi and others. The above observation, though not sufficient as proof, is highly suggestive of the relationship between changes in the amount of the BOJ's purchases of JGBs and/or the possibility of the BOJ's adoption of inflation targeting on the one hand, and the response of long-term rates on the other.

Meanwhile, as mentioned earlier, consensus forecasts of inflation continued to decline.15 The apparent contradiction between survey measures of inflation expectations and long-term interest rates may be explained by the differences in forecast horizons or differences in expectations among different economic agents. Whatever the reason, the observed combination of increases in long-term interest rates and the continued decline in survey measures of inflation expectations may well suggest a deflationary phenomenon.16 Of course, such a combination may not necessarily be the outcome of a policy move like the one in August, but it should be borne in mind as a possible risk.17

  1. 15A similar case can be found in the experience of the United Kingdom. When the Bank of England increased its operational independence in May 1997, the expected future inflation rates (implied forward rates calculated from Index-Linked Treasury Bonds) especially those of four to five years ahead dropped immediately. However, the survey of the general public around that time showed that the expected inflation rate rose in line with the actual inflation rate (Chart 6). In a similar vein, Englander and Stone (1989) discuss heterogeneity of inflation expectations for the U.S. and point out, for example, "during the first three quarters of 1987, financial market inflation expectations and interest rates rose sharply in anticipation of inflationary pressures which did not emerge, while household inflation expectations moved less pessimistically." They conclude that "to the extent that incorrect forecasts of accelerating or decelerating inflation affect interest rates and compensation growth, these forecasts may contribute unforeseen contractionary or expansionary impulses to the economy."
  2. 16For example, a rise in long-term interest rates causes financial institutions holding a large amount of JGBs to incur a capital loss.
  3. 17As of the writing of this article, the bond market has shown no significant response to the December policy move described in footnote 14.

F. Discussions of Inflation Targeting

As is clear from what I have said so far, the current framework for monetary policy has some elements in common with inflation targeting. That is to say, the BOJ bases its decision to terminate the framework on the rate of increase in the CPI. On the other hand, the framework cannot be considered a complete version of inflation targeting in the sense that the BOJ does not indicate the target timing of the rise in the CPI above zero percent. Nor has it made clear that it will implement additional measures of monetary policy in response to deviations of expected inflation from a target level.18

I consider inflation targeting a possible policy option once the inflation rate recovers to a normal level. There is a contrary view that if the BOJ indicates a specific timing for achieving an inflation rate of zero percent, even in the current deflationary situation, it will succeed in raising the expected inflation rate and thus produce monetary easing effects. With severe limitations on instruments to ease monetary policy, however, I fear that an announcement of the target date by which to achieve, say, zero percent inflation, would have either no effects on the market or be counterproductive. The market may lose confidence in a central bank announcing a hard-to-hit target. Moreover, such an announcement could be deflationary as mentioned earlier if the expected inflation rate of firms and households does not increase while at the same time long-term interest rates surge as a result of anticipation of inflationary risks in the more distant future.19

  1. 18It may be seen that the medium- to long-term inflation rate target implicit in the current framework is not zero percent but slightly positive. The reason is as follows. Assume that the BOJ will continue to maintain the nominal interest rate at virtually zero percent even when the economy starts recovering and the actual inflation rate increases to around zero percent. The real interest rate will also decline to close to zero percent, lower than at present, providing more stimulus to the economy than before at that point. As a result, the CPI inflation rate will very likely continue to rise to above zero percent.
  2. 19The BOJ could, however, hit any inflation target if the Bank of Japan Law were amended so that it could purchase any assets, for example durable consumer goods. The BOJ would then be using inflation targeting to avoid a runaway inflation

IV. Conclusion--Possible Coordination Between Monetary and Other Policies

Finally, I would like to discuss some possible other government policy options that can be combined with the current monetary policy stance.

In general, prices are not determined by monetary policy alone. Prices are determined by the supply of and demand for goods and services. Taking for example the demand side, it is influenced by not only monetary policy but also by fiscal policy and other determinants of the components of aggregate demand.20 As I have argued, the power of monetary policy to affect the level of prices is currently very limited.

The current monetary policy framework promises to maintain money market interest rates near zero until the inflation rate rises above zero percent. The timing of termination of the framework will depend much on other government economic policies including the stance of the fiscal authority. At one extreme, the government can implement drastically expansionary fiscal measures, such as the so-called non-Ricardian fiscal policy, if it considers the cost of the current deflation to be very serious. Deflation may stop as a result. But it is uncertain whether the economy will be on a balanced, sustainable recovery path. Alternatively, the government can speed up its reform efforts, including attempts at the resolution of the non-performing loan problem. With this option, it might take longer to achieve an inflation rate of zero or above; but the economy could be in better shape after a while. It is the government's decision whether to adopt either of the two options or to formulate a better policy mix.

  1. 20As an example of the role of fiscal policy in achieving price stability, a recovery in fiscal discipline has been traditionally considered important in stamping out hyperinflation in developing countries. According to a recent theory, the Fiscal Theory of the Price Level, non-Ricardian fiscal policies may be used to avoid a deflationary spiral in a liquidity trap. For example, the government carries out a large tax cut, but promises not to match this with future tax increases. Households' disposable income goes up on a permanent basis, as also do consumption and the price level. The rise in the price level reduces the real value of existing government debt and makes ends meet for the government. See Woodford (2000). The applicability of this theory to the current Japanese situation, however, remains to be seen.


Bernanke, Ben and Alan Blinder. "Credit, Money, and Aggregate Demand," American Economic Review Papers and Proceedings, Vol. 78, No. 2, 1988, pp. 435-39.

Englander, Steven A. and Gary, Stone. "Inflation Expectations Surveys as Predictors of Inflation and Behavior in Financial and Labor Markets," Quarterly Review, Federal Reserve Bank of New York, Vol. 14, No.3, Autumn 1989, pp.20-32.

Krugman, Paul. "It's Baaack: Japan's Slump and the Return of the Liquidity Trap," Brookings Papers on Economic Activity, No. 2, 1998, pp. 137-205.

McCallum, Bennett T. "Theoretical Analysis Regarding a Zero Lower Bound on Nominal Interest Rates," Journal of Money, Credit and Banking, Vol. 32, No.4, Part 2, November 2000, pp.870-904.

Ueda, Kazuo. "The Transmission Mechanism of Monetary Policy near Zero Interest Rates: The Japanese Experience 1998-2000," speech given at a conference held at the Swedish Embassy in Tokyo on September 22, 2000.

Woodford, Michael. "Fiscal Requirements for Price Stability," NBER Working Paper Series, No. 8072, National Bureau of Economic Research, 2000.

  • Chart 1 Short-Term and Long Term Interest Rates. The details are shown in the main text.
  • Chart 2 Ratio of Nominal GDP to the Monetary Base (Four-Quarter Moving Average). The details are shown in the main text.
  • Chart 3 CPI Forecast from the Consensus Forecast. The details are shown in the main text.
  • Chart 4 Short-Term Interest Rates in 2001. The details are shown in the main text.
  • Chart 5 1-Year Implied  Forward Rates in August 2001. The details are shown in the main text.
  • Chart 6 (1) Inflation Target and Changes in Expected Inflation in the United Kingdom. The details are shown in the main text.
    Chart 6 (2) Implied forward inflation rates. The details are shown in the main text.