|
|
Kunio OKINA, Director, Institute for Monetary and Economic Studies,
Bank of Japan Shigenori SHIRATSUKA, Senior Economist, Institute for Monetary and Economic Studies, Bank of Japan |
In his recent article2, Professor Koichi Hamada, Yale University, proposed Japan effect further monetary easing by not sterilizing foreign exchange intervention.
His proposal seems to be based on the following three arguments:
| (a) | Under a floating exchange rate regime, an increase in domestic demand will lead to the yen's appreciation which will tend to undermine the recovery through a decrease in external demand, thereby necessitating intervention in the foreign exchange market to prevent the yen from appreciating. |
| (b) | Since the effectiveness of intervention is largely negated by sterilization, the Bank of Japan should, depending on the economic situation, refrain from sterilizing intervention as long as it does not cause any disturbance in the short-term money market. |
| (c) | Unsterilized intervention is necessary since if the market believes that the Bank may not sterilize intervention, expectations may arise that the yen will not appreciate continuously. |
This paper addresses the issues regarding monetary policy in relation to some crucial points, which, in our view, are overlooked in Professor Hamada's arguments.
In response to
Professor Hamada's arguments
Argument 1: Intervention is necessary to prevent the yen's appreciation resulting from a recovery of domestic demand.
Further to (a) above, Professor Hamada pointed out that "the foreign exchange rate should be, to the extent that it affects Japan's economic activity, of great interest to the Bank."
We do not deny developments in the foreign exchange rate are important and the Bank has repeatedly asserted that they are factored into its assessment of the economic and financial situation for policy formulation purposes.
Generally speaking, foreign exchange fluctuations themselves
do not necessarily call for intervention. Indeed, if all countries
intervened to prevent their exchange rates from fluctuating,
there would not be any floating exchange rate regimes. Hence,
intervention can only be justified when the exchange rate overshoots
in light of economic fundamentals and there is good reason to
believe that such overshooting will exert a negative impact on
not only Japan's economy but also the global economy. It is,
however, not quite self-evident that any particular exchange
rate can be judged as overshooting at any particular point in
time. In this paper, for the sake of discussion, let us assume
that there was an excessive appreciation of the yen and intervention
was conducted.
Argument 2: Effectiveness of intervention largely diminishes if sterilized.
Professor Hamada argues that "the effectiveness of intervention will be largely negated by sterilization as if wiping out what the right hand did a moment ago with the left hand." This argument is somewhat inappropriate for several reasons.
First of all, within the framework of the Mundell-Flemming model which the earlier part of Professor Hamada's argument is likely based on, economic recovery under a floating exchange rate regime will induce upward pressure on domestic interest rates, thereby resulting in the appreciation of the yen. In such a situation, it would seem a logical consequence to prevent interest rates from rising through further monetary easing, thus stemming the yen's appreciation. However, the zero interest rate policy dictates that interest rates are maintained at zero even when there are upward pressures. Hence, it does not seem possible to advocate that sterilization would negate the effects of intervention.
In the latter part of his argument, Professor Hamada shifted his focus from changes in interest rates, which are the core mechanism of the Mundell-Flemming model, to changes in quantitative variables. His modified argument is "unsterilized intervention is effective because the relative ratio of money supply between the US and Japan changes."
It is not clear what is the specific model Professor Hamada had in mind when making his argument. However, his article seems to suggest either of the following two. In a general equilibrium model, when the monetary base increases the yen will depreciate in order to restore asset market equilibrium since interest rates do not change under the zero interest rate policy. Or, in a partial equilibrium model, when an increase in the monetary base expands money supply the yen will depreciate since the foreign exchange rate reflects purchasing power parity in the long run.
However, neither of these two arguments is correct under the zero interest rate policy. This is because when short-term interest rates are kept at zero as a result of the Bank's provision of ample liquidity, short-term government bills and the monetary base become almost complete substitutes.
Let me explain this in the context of a general equilibrium model. Sterilization means that the central bank absorbs an increase in the monetary base due to intervention using such operations as the sale of short-term government bills. When interest rates are zero, the monetary base and short-term government bills are almost complete substitutes for each other. Since, in a general equilibrium model for asset markets, equilibrium does not change by exchanging two assets which are almost complete substitutes, equilibrium under the zero interest rate policy will be the same whether or not intervention is sterilized. In other words, the equilibrium foreign exchange rate will not change even if the Bank increases the monetary base by not sterilizing intervention.
The same conclusion can be obtained in a partial equilibrium model. As long as the result in a general equilibrium model is not affected, an increase in the monetary base due to unsterilized intervention will not result in a corresponding expansion in money supply. However, there is one caveat. The above argument does not incorporate the effect stemming from expectations, which will be discussed later.
Professor Hamada rejects the Bank's assertion that "any
additional injection of liquidity would only result in further
accumulation of excess funds at financial institutions"
by saying that "such assertion is made by those who don't
understand the theory of policy mix." We feel that his arguments
miss the essence of the zero interest rate policy described above.
Argument 3: By not sterilizing intervention, expectations may arise that the yen will not appreciate continuously.
Another rationale to support the effectiveness of unsterilized intervention is its effect on market expectations. Professor Hamada states that "if the market believes that the Bank may conduct unsterilized intervention, expectations may arise that the yen will not appreciate continuously."
Given the massive flow of funds in and out of the money market, is it meaningful to make an issue of only those funds stemming from foreign exchange intervention? The figure below shows the factors prompting the outflow and inflow of funds since 1996. Since the actual amount of intervention is not disclosed even ex post, we have used the foreign exchange special account as a proxy, and its movement appears random in the figure. The amount of funds flowing into the market through the foreign exchange special account was only about 1 trillion yen, even in September 1999 when there were strong pressures for the yen to appreciate, and it is small compared with the monetary base of about 60 trillion yen and money supply of some 600 trillion yen (M2+CDs).
As implied in the above discussion of equilibrium models, the bulk of excess reserves currently accumulates in the accounts of tanshi companies (money market brokers), meaning that the transmission mechanism from monetary base to money supply is not functioning properly. Under such circumstances, we cannot envisage in the near future a monetarist view that the monetary base and money supply will increase proportionally. Furthermore, the 1 trillion yen in the foreign exchange special account is only 0.17 percent of money supply. Thus, even if money supply is forecast to increase several times the amount in this special account in the distant future, it is unlikely that intervention will have any significant impact on the foreign exchange rate.
As we have explained so far, Professor Hamada's proposal to prevent the yen's appreciation through unsterilized intervention is thin in terms of theoretical foundation. Professor Hamada rejects Professor Krugman's idea of "generating inflation in Japan"3 as "a policy based on illusion." In our view, the idea of "stemming expectations for the yen's appreciation by unsterilized intervention" also relies heavily on 'illusion' and such a policy is unlikely to have anything but a temporary and illusory effect.
Having said this, let us further consider the case where the Bank dares announce unsterilized intervention to affect expectations by taking advantage of 'illusion' in the marketplace. In examining this case, we immediately confront the operational problem that the actual size of intervention has never been released, even ex post in Japan. By contrast, in the US, for example, a report called "Treasury and Federal Reserve Foreign Exchange Operations" is compiled quarterly, which contains details of foreign exchange intervention such as execution dates, currencies used, amounts, and some background to the intervention. And it is published at the beginning of the second month after the end of each quarter.4 Since foreign exchange intervention in Japan is within the jurisdiction of the Ministry of Finance, the Bank, as an agent, cannot disclose such information at its discretion. As a result, even if the Bank announces unsterilized intervention, it cannot be held accountable for its announcement since the information related to intervention is not disclosed.
Therefore, even considering the impact on market expectations,
Professor Hamada's proposal seems to be accompanied by the following
significant shortcomings. First, he seems to overestimate the
effect of foreign exchange intervention, which is a mere fraction
of the flow of funds in the money market. Second, he proposes
that the Bank announce unsterilized intervention even if details,
such as the amount of intervention, which do not come under the
jurisdiction of the Bank, cannot be disclosed. Such announcement
is not compatible with the accountability of monetary policy.
An
alternative measure to counter the excessive appreciation of
the yen
Even if we reject Professor Hamada's proposal of unsterilized intervention, the question still remains as to what should be done if the yen appreciates excessively and the risk of falling into a deflationary spiral materializes. To this question, a theoretical answer would be quite simple if the excessive appreciation of the yen is clearly identified, namely, the government simply announces the unlimited purchase of dollars with the yen at a level which does not represent an excessive appreciation. The government would issue short-term financing bills to raise the yen funds necessary for dollar purchase intervention at negligible cost as long as the zero interest rate policy is in place.
The key to successful prevention of the yen's excessive appreciation is to obtain credibility from market participants with respect to exchange rate policy. If they truly believe that the Ministry of Finance will implement such drastic unlimited intervention, the appreciation of the yen will stop even without the execution of intervention. For this, it is not necessary to change monetary policy, only to change intervention policy. A combination of the zero interest rate policy and unlimited intervention could be a sufficient countermeasure against the excessive appreciation of the yen.
There would naturally be a couple of questions regarding such
a policy prescription. For example, is it possible to clearly
identify the divergence of the foreign exchange rate from the
level warranted by economic fundamentals? Is it a realistic policy
choice for Japan to de facto break from the floating exchange
rate regime by setting a ceiling on the foreign exchange rate?
Given the long history of the floating exchange rate regime being
adopted by major industrial countries, it is indeed a far-reaching
and dramatic challenge to introduce a policy of setting an exchange
rate ceiling to counter the yen's appreciation. However, should
we ever clearly define a situation where the yen has appreciated
excessively, this policy would have a much firmer theoretical
footing than the policy proposed by Professor Hamada.
Flexibility
in monetary policy and its limit
Now let us turn to the issue of monetary policy in Japan. Professor Hamada states that the Bank of Japan is "opposed to all pressures for any monetary easing." This statement is simply not true.
The zero interest rate policy since February 1999 was considered extremely unrealistic, putting aside the theoretical possibility, and therefore was totally unexpected when it was effected. Even under such circumstances, the Bank of Japan dared to make this extremely accommodative policy because the situation necessitated such action. And, ever since then, the Bank has been actively working on market expectations in response to the change in economic and financial conditions, and making efforts to provide ample liquidity.
Unfortunately, the extremely accommodative monetary policy coupled with expansive fiscal policy is yet to have the intended effects of bringing the Japanese economy firmly back onto a sustainable growth path. Why has it taken so long for the economy to recover? A simple answer to this is that there were significant structural problems that could not be solved solely by macroeconomic policy which primarily aims at stimulating the economy in the short run.5
The biggest structural problem we have been confronted with is the non-performing asset problem. In other words, erosion of the capital base has weakened the resiliency of various sectors in the economy. As lenders, financial institutions, faced with a shortage of own capital, have become prudent in taking on new risks in extending loans. As borrowers, firms have suffered capital erosion and have become cautious in making investments in new businesses as well as research and development. As a consequence, sluggish lending by financial institutions has continued for a long time.
The Bank of Japan succeeded in avoiding deflation and financial panic by drastic interest rate cuts and the use of the lender of last resort function. However, monetary policy cannot substitute for structural policy. Rather, one cannot deny the possibility that extreme monetary easing might have the side effect of delaying serious restructuring on the part of financial institutions.
In the meantime, the public has been rather slow in fully appreciating the gravity of the situation, and thus it took a fairly long time to pass the laws regarding the resolution of bank failures and to garner support for the injection of public funds. A dilemma for the central bank is that no matter how delayed implementation of structural policy and structural adjustment may be, should the economy actually face the risk of falling into a serious deflationary spiral, the bank has a responsibility to make efforts to avoid it.
In fact, if the economy were on the verge of tumbling into a deflationary spiral of the kind witnessed during the Great Depression, the central bank would do everything in its power to prevent it from happening while recognizing serious side effects. In such a case, serious consideration might be given to the drastic injection of massive liquidity, an amount which goes beyond the limit provided by ordinary monetary operations. Needless to say, such a policy, if implemented, would impose extremely high costs on the national economy, for example by undermining fiscal discipline.
Though clear signs of a self-sustained recovery in private
demand have not been observed yet, the economy had stopped deteriorating
and has recently started to improve, with exports and production
increasing. Therefore, we are definitely not in a situation which
would warrant the adoption of any extraordinary measures with
large side effects and which would only be taken if the economy
faced the risk of falling into a serious deflationary spiral.
Monetary policy
in the 21st century
Professor Hamada concluded his article by saying that "In the next century, the Bank of Japan should change its overly defensive attitude. I hope it will become a central bank with a variety of policy measures which it will use timely and appropriately, and also at times be able to work on public sentiment."
We totally disagree with his comment that the Bank of Japan is overly defensive, and it is a pity that Professor Hamada has come to harbor such an impression. Even though the Bank of Japan has been striving to provide ample liquidity in response to the change in economic and financial conditions, monetary policy is not a panacea. The government and the Bank of Japan should make the utmost efforts, in terms of their respective responsibilities and measures available to them, to counter the challenges facing the Japanese economy.
To better discharge its duty through the effective conduct of monetary policy, it is important for the Bank of Japan to refine monetary policy tools, provide appropriate prescriptions for various structural problems, which are treated as given in the conduct of monetary policy, and encourage their implementation.
In this regard, we should discuss frankly and constructively
the effectiveness and limit of various policy options, rather
than rely on policy effects that are merely the outcome of temporary
illusion. As part of such constructive discussion, we hope that
the press and academic circles will closely monitor whether the
Bank of Japan makes the utmost efforts in discharging its policy
responsibility and continue to voice whatever criticism and advice
they may have.
1 English translation of an article by the authors entitled "Hi-Futaika Kainyu no 'Sakkaku'," which originally appeared in Shukan Toyo Keizai, January 15, 2000.
2 Koichi Hamada, "Nichigin no Futaika Seisaku wa Machigatte Iru" (The Bank of Japan is Wrong to Take Sterilized Intervention), Shukan Toyo Keizai, November 13, 1999, pp. 72-76 (in Japanese).
3 Since Professor Krugman has written many works on this subject, we are not sure which one Professor Hamada is quoting from. However, for example, Professor Krugman emphasized the importance of credible commitment to future monetary expansion so as to generate inflationary expectations, and proposed that Japan should announce an inflation target over the long term, for example "4 percent inflation for 15 years," in "Further Notes on Japan's Liquidity Trap," June 1998, http://web.mit.edu/krugman/www/liquid.htm.
4 It should be noted that, since intervention would inevitably change the amount of foreign exchange reserves, it can be estimated to the extent that detailed information regarding factors contributing to the change in foreign exchange reserves is disclosed. In this regard, the IMF has created a disclosure template regarding foreign exchange reserves which is comprehensive, detailed, and to be compiled on a regular basis. Disclosure of foreign exchange reserves will proceed in line with this template from this spring among industrial countries and some emerging economies.
5 See Yutaka Yamaguchi, "Monetary Policy and Structural Policy: a Japanese Perspective," November 2, 1999, http://www.boj.or.jp/press/koen062.htm.