Foreign Exchange Intervention Funds left Unsterilized in the Market: New measures for overcoming deflation

WATANABE Tsutomu
Faculty Fellow, RIETI

Has the BOJ been pushed into a position where it can no longer control prices? This was a question I received from one of my research colleagues abroad. In emerging markets, central banks sometimes fail to control hyperinflation despite taking various countermeasures, and prices show uncontrolled fluctuation like a "kite with a broken string." The questioner wanted to know whether a similar situation is being observed in Japan regarding deflation.

What is currently happening in Japan is a mild deflation, with prices declining by 1%-2% per year. This is something different from the kind of deflationary spiral which was an issue of concern in the past. Therefore, it is not appropriate to treat the issue at the same level of the hyperinflation in emerging economies. On the other hand, the total amount of decline, accumulated from the 1990s, is significant even if the decrease in each year remains small. It is also true that the Japanese government and the Bank of Japan ( BOJ ) have not been successful in addressing the long-standing deflation.

♦ ♦ ♦

My colleague is not the only person who thinks that prices are left uncontrolled in Japan. The yen keeps rising in the foreign exchange market, as if the investors were taking a shot at Japan struggling to control prices. In response to such market behavior, the government and the BOJ implemented major yen selling and dollar buying intervention on September 15th, the first time since spring 2004. At the same time, the government and BOJ executed "non-sterilization," which means leaving the yen supplied to the market by the intervention unsterilized, instead of absorbing them.

Now, what sort of implications does this non-sterilized intervention have?

Under traditional monetary policy, central banks control the money supply and adjust the interest rate. However, in Japan, the rate has remained at nearly zero since 1999.

One of the functions of money is the facilitation of settlement arising from the buy-sell transactions of goods. Interest is the consideration paid for this liquidity service provided by money. However, since 1999, due to the large-scale supply of money by the BOJ, the marginal utility of this liquidity service dropped to zero as money supply reached a saturation point. This situation is reflected in the zero-interest rate, which represents the price of the liquidity service.

As monetary policy under a zero-interest rate is an unexplored frontier for central banks, a trial and error process is unavoidable. Researchers are also lost in this untouched territory, and consensus on how we can control prices in a zero-interest economy has yet to be established.

However, even in such a situation, there exists a hypothesis called the "strong yen theory" that has gained the attention of researchers from the early stage as a hypothesis capable of explaining Japan's deflation. For example, Ronald McKinnon, Professor at Stanford University, and Kenichi Ohno, Professor at the National Graduate Institute for Policy Studies, point out that the deep-rooted expectation for yen appreciation has induced the deflation. In fact, the Japanese economy has been repeating a pattern of soaring yen and subsequent drop of prices.

Regarding the relationship between yen appreciation and deflation, many researchers argue that in order to prevent deflation, yen appreciation must first be avoided. For this purpose, they support a major foreign exchange intervention. Nevertheless, from the traditional common sense on monetary policy, the concept of using foreign exchange intervention as a preventative tool against deflation is considered as going astray. This is because the traditional theory understands monetary policy and foreign exchange intervention as separate and independent policies without any interrelation.

In developed countries, monetary policy is implemented through setting a target as the policy rate (i.e. the overnight unsecured call rate in Japan) and adjusting money supply on a daily basis to achieve the target. However, if yen supply in the market increases due to yen selling interventions, a gap will arise between the policy rate and the target rate. To avoid such a situation, central banks carry out open market operations to absorb the intervention funds; this is called "sterilized intervention."

A number of experimental studies prove that the central banks of developed countries implement sterilization almost exclusively. As long as sterilized intervention is adopted 100% of the time, foreign exchange intervention will have no impact on yen supply. In that sense, those two factors are mutually independent. This is common knowledge that appears in international macro economic textbooks. However, this is common knowledge only when the interest rate is positive. If the money market is at the saturation point and the interest rate is zero, a different mechanism applies.

♦ ♦ ♦

How do intervention and monetary policy interact in the zero-interest market? The answer to this question was demonstrated through the major yen selling intervention by the government and the BOJ between 2003 and 2004. This large-scale intervention, named the "Great Intervention" by John Taylor, Professor at Stanford University, poured a total of 35 trillion yen into the market.

The figure shows an estimate of the impact that the 1 trillion yen selling intervention had on the yen supply at that time. At settlement, which takes place two days after intervention, 600 billion yen out of 1 trillion yen supplied to market through intervention had been absorbed by the BOJ sterilization operation, and 400 billion yen remained in the market. Afterwards, the amount of money in the market dropped to 250 billion yen four days after intervention, 170 billion six days after intervention, and kept decreasing gradually until it nearly disappeared 20 days after intervention. In other words, the average period that intervention fund remained in market was 20 days at that point. If we do the same analysis for the time before the major intervention, the intervention fund had been almost 100% absorbed by the settlement date, two days after intervention.

Figure: Period that 1 Trillion Yen Intervention Fund emained in the Market Figure: Period that 1 Trillion Yen Intervention Fund Remained in the Market

(Source)"Foreign exchange intervention in the quantitative easing period", Watanabe Tsutomu, Yabu Tomoyoshi, Financial Review , Vol. 99, February 2010

The factor that enabled the retention of intervention funds was the saturation of money. The BOJ had been implementing a quantitative easing policy since 2001, prior to the major intervention. When the major intervention began, the target balance of Demand Deposit of BOJ was set at 15-20 trillion yen, and even exceeding the target was permitted.

Due to such a large amount of money supply, the money market was already saturated and the overnight call rate was zero. Under such conditions, the overnight call rate will no longer decline, even if the yen fund supplied through yen selling intervention stays in the market. Thus, the goal of monetary policy can be achieved even though the yen selling intervention fund remains unsterilized. This is the big difference from a market with positive interest rates. The implication of this mechanism is that the "common sense in the positive interest rate market" that "an intervention should be immediately sterilized" no longer applies.

Paul Krugman, Professor at Princeton University, compared the zero interest rate market to Alice's world of "Through the Looking Glass," where sense and absurdity intermingle. Non-sterilized intervention is definitely one example of his metaphor.

What is the effect of non-sterilization? I carried out a research project with Yabu Tomoyoshi, Professor at Keio University, measuring the impact of intervention on the foreign exchange market, using the data of the major intervention. The results have shown that the effect of intervention depend on whether the preceding interventions in the three months before the intervention were sterilized or not. The impact of yen selling intervention (per 1 trillion yen) on the foreign exchange market (i.e. percentage decline of the yen) was 1.9% without any sterilization, 1.5% if 20% of the fund was sterilized, and 1.1% if 40% was sterilized. The exchange rate dropped from around 82 yen to 85 yen against the dollar with the approximately 2 trillion yen intervention implemented on September 15th, which was approximately the same level of effect as expected of a case without sterilization.

♦ ♦ ♦

Why does sterilization make a difference to the impact of intervention? One may immediately come up with the possibility of yen depreciation caused by the remaining intervention funds. However, given that the money market is at saturation point, this hypothesis does not work. Another possibility is the effect of "expectation." For the purpose of simplification, we will assume that no intervention will be sterilized and the funds will remain in the market forever. This means that the funds stay in the market even in the distant future, where the economy has come out of recession and normalized. Since interest rates should have turned positive when the economy normalized, additional yen supply through intervention would place a downward pressure on future interest rates. This drop of interest rate would trigger future yen depreciation, but if this were factored into investors' expectations, it would result in the current weakening of the yen.

Of course, there were some sterilization operations even at the time of the major intervention, and the funds were not abandoned forever. Nevertheless, we may infer that, on one level or another, such effect through the expectation of market players on foreign exchange fluctuation was present, and this added to the impact of intervention.

It goes without saying that intervention should not be implemented without considering the situation in other countries. Comparison between intervention and other money supply measures is essential. However, in the zero interest market, we have a new option of non-sterilized intervention, which may bring in an effect different from those in the positive interest rate market. This should be considered as one of the promising measures for recuperating control of prices.

>> Original text in Japanese

* Translated by RIETI.

September 27, 2010 Nihon Keizai Shimbun

February 7, 2011