Monetary easing and the Bank of Japan's balance sheet

October 13, 2017

The Bank of Japan has been struggling with sluggish underlying inflation (the CPI excluding food and energy has remained at around 0%) despite strong economic growth recently. Even though inflation is likely to reach around 1% toward the fall of 2017, mainly due to the base effect related to oil price movements, the distance from the 2% price stability target will remain large. According to the latest monthly ESP Forecast survey,1 the average inflation projected by 42 forecasters was 0.6% in fiscal year 2017, 0.8% in fiscal year 2018, and 1% in fiscal year 2019 (excluding the direct impact of the scheduled consumption tax hike). Moreover, the longer-term average inflation for fiscal year 2019-2023 is only projected to be 1.1%. This indicates that the 2% price stability target is unlikely to be achieved in the foreseeable future (see also Cecchetti and Schoenholtz 2016). Given that adverse impacts from the current monetary easing framework are also mounting, the Bank will eventually need to make more sustainable policy decisions to reach underlying inflation of 2%.

On this front, there are essentially two important issues discussed in this column: the sequence of monetary policy toward normalisation of monetary easing, and the impacts of the Bank of Japan’s normalisation on its balance sheet. Regarding the former, the process could be divided into two phases: the first phase where the amount of net purchases of Japanese Government Bonds (JGBs) will gradually reduce towards zero, and the second phase where a rise in short-term interest rates takes place and is followed by a cut in the size of JGBs. The cut in the amount of net purchases of exchange-traded funds (ETFs) and real estate investment trusts (REITs) will be discussed as well. This column will focus on these issues as well as the Bank of Japan’s balance sheet issues.

The first phase: The process to reduce the pace of monetary expansion

Before raising short-term interest rates, the Bank of Japan is expected to remove the 10-year yield target (set at around 0% since September 2016) and reduce net JGB purchases. Conceptually, if the Bank reduces the 10-year yield target, it could reduce the amount of JGB purchases. In my view, the following course of action could be one possibility, considering the consistency related to the conduct of monetary policy and market operations.

First, the Bank raises the 10-year yield target from around 0% currently or introduce a range (such as 0-0.5%). Second, it begins to reduce the annual pace of net JGB purchases explicitly and gradually to the amount of net issuance (about ¥50 trillion for fiscal year 2017). Third, it eliminates the 10 year-yield target range. Finally, it reduces the annual pace of net JGB purchases to zero and maintains the outstanding amount of JGB holdings by reinvesting repayments of principal on the JGBs held by the BOJ.

While the Federal Reserve engaged only in a steady reduction in asset purchases (‘tapering’) from January to October 2014, the Bank of Japan needs to work on both removing the 10-year yield target as well as on tapering. This complicates the Bank’s task, since the 10-year yield target was set at an excessively low level so that the removal of the target may generate an overshooting of the long-term yields and thereby cool down the economic recovery. If such a yield overshooting lasts long, the Bank may end up purchasing more JGBs rather than purchasing less.

The Bank of Japan’s tacit tapering of JGB purchases

With regards to the pace of reducing asset purchases, the cut in the amount of net JGB purchases toward ¥50 trillion is relatively easily done since this amount is equivalent to the amount of net JGB issuance by the Ministry of Finance (based on the initial budget for the fiscal year 2017 and including Fiscal Investment and Loan Program bonds). This is because the Bank of Japan does not need to purchase the JGBs from existing holders any longer, at least on the aggregated data basis. With the amount of Bank’s purchases of around ¥50 trillion, there is sufficient demand for JGBs by domestic commercial banks and institutional investors. Banks want to hold JGBs mainly due to the limited demand for credit by the private sector, while deposits are rapidly growing. Insurers and pension funds need JGBs due to the asset-liability match as well as the small size of private sector debt securities markets.

Indeed, the Bank is ‘tacitly’ tapering its JGB purchases from September 2016 by reducing the annual pace of increase toward ¥50-60 trillion currently – although its official statement on monetary policy stresses a continuation of about ¥80 trillion. In my view, the main reason why the Bank shifted its official guideline for market operations from the monetary base (quantity target) to the yield curve control in September 2016 was to reflect this challenging environment and shift market attentions from the volume of JGB purchases to the JGB yields. So long as the Bank continues to purchase ¥80 trillion, greater demand will lower 10-year or longer yields into the negative territory. This will aggravate the profitability and viability of banks, insurers, and pension funds, thereby raising financial stability risk. The near-term issue will be on the timing for the Bank to officially admit the de facto tapering by removing the words “about ¥80 trillion” from the Statement on Monetary Policy.

The more challenging issue is associated with a cut in the annual pace of net JGB purchases from around ¥50 trillion toward zero or completing the tapering process. This will be more challenging because the JGB market needs to find alternative large, stable investors that could replace the Bank of Japan’s investment. The supply (issuer) side or the amount outstanding of JGBs issued will continue to grow every year due to the persistent fiscal deficit. Meanwhile, there is uncertainty with regards to the demand (investor) side, partly because the Government Pension Investment Fund reform on basic portfolios has reduced demand for the JGBs. Moreover, once the Bank begins to reduce the amount of JGB purchases toward zero, many financial institutions (including insurers) might refrain from investing in the JGB market for a while and take wait-and-see attitude until the market prices stabilise – in fear of evaluation losses caused by a sharp rise in long-term yields.

The process to tighten monetary policy and shrink the balance sheet

Then, the process to start normalisation or raising short-term interest rates will begin — roughly in line with the sequence that the Federal Reserve has been pursuing (Federal Reserve 2014). This involves four steps.

  • First, the Bank of Japan will unify three types of interest rates on excess reserves – roughly the commercial banks’ current account balances with the Bank (–0.1%, 0%, 0.1%) – into 0.1% to take a first step toward normalisation. Also, the Bank is expected to reintroduce the target range on the uncollateralised overnight call rate of around 0%–0.1% (this policy rate target range was removed when the Bank adopted quantitative and qualitative monetary easing in April 2013, and changed the main operating target for money market operations from the uncollateralised overnight call rate to the monetary base). The target range on the uncollateralised call rate and the interest rate on excess reserves are expected to rise together following the Federal Reserve.
  • Second, during normalisation, the Bank will intend to use liquidity-absorbing operations including (a) the issuance of equal or less than 3-month Bank of Japan bills, (b) the sale of JGBs and Treasury Bills with repurchase agreements, and (c) outright sales of Treasury Bills. 
  • Third, after raising the target range on the uncollateralised overnight call rate for some time, it will begin to reduce the amount of reinvesting repayments of principal on the JGBs held by the Bank of Japan.
  • Fourth, the size of the Bank’s assets is likely to be reduced to the level that would be necessary to implement monetary policy efficiently and effectively—that is, the projected amount outstanding of notes in circulation (based on the projection on nominal GDP growth rates) as well as the appropriate size of excess reserves to be estimated by the Bank. 

Reducing the annual pace of ETF and REIT purchases

Regarding ETFs and REITs, the timing of reducing the annual pace of these purchases could be determined separately form the above-mentioned sequence. At the 16 June 2017 press conference, Governor Haruhiko Kuroda also explained that it is theoretically possible to reduce the ETF purchases before the 2% price stability target is achieved and before the discussions related to the exit policy begin (Kuroda 2017). However, he added that it is generally unthinkable that such a cut will take place before 2% inflation is achieved, since the ETF purchase is a part of monetary easing tools to achieve the 2% target.

The annual pace of the Bank of Japan’s purchases of the ETFs (and REITs) is about 6 trillion (and 90 billion) in March 2017. Even though this amount is well below the daily transaction volumes in the stock market and the amount of JGB purchases, the impact of the Bank’s actions is large, since the stock market participants have been conducting transactions by taking into account the Bank’s actions. The Bank of Japan is widely viewed as tending to purchase ETFs (and REITs) when stock prices fall. This has given a sense of security to investors by reducing downside risk as well as generating transactions that attempt to anticipate the Bank’s moves.

Stock market participants largely view that a further increase in the Bank’s ETF purchases will lead to a situation where market prices of individual companies do not necessarily reflect firms’ specific information and fundamentals. An important issue in the future, therefore, will be how these markets will respond to change in the Bank’s policy toward normalisation. Given that Japan’s economy currently maintains a strong momentum and that the current stock prices (over 20,000 yen on Nikkei 225 and around 1,600 points on TOPIX) remain reasonable based on PER, some long-term oriented investors suggest that the Bank should taper the annual pace of ETF purchases now. So long as stock prices are not overvalued and the economy is in the recovery phase, they claim that there is always demand for stocks, which could take over the Bank’s purchase position. Other market participants, however, view that the Bank’s tapering is likely to trigger a large sell-off of Japanese stocks.

Meanwhile, many market participants hold the view that the real estate markets are likely to remain active and favourable at least until before the 2020 Tokyo Olympic Games, due to an expected increase in construction activities (such as for sport facilities, hotels, and restaurants) in the Tokyo Metropolitan Area. Nevertheless, the declining trends of the TOSHO REIT index since early 2017 is worrisome. The Bank of Japan may find it difficult to taper REIT purchases since it may aggravate the market conditions.

So far, this column has focused on the sequence of monetary policy normalisation. Another important issue is the impact of normalising monetary easing on the Bank of Japan’s balance sheets.

Raising short-term interest rates and net income loss

When the Bank begins to raise the interest rate on excess reserves, it is likely to result in net income loss on the income statement. This is because interest payments by the Bank to commercial banks with regards to the amount of excess reserves (roughly the current account balances on the liability side of the Bank’s balance sheet, as shown in Figure 1) will begin to rise and eventually exceed interest income mainly from the JGBs held by the Bank (the asset side of the Bank’s balance sheet). A continuation of net income loss for several years will wipe out accumulated provisions (reported on the liability side of Bank’s balance sheet) and then legal reserves (reported on the total net assets of Bank’s balance sheet), resulting in negative equity. Legal reserves, stated capital, and provisions constitute the Bank’s capital base.

Figure 1 Bank of Japan’s balance sheet

Given this background, there have been growing calls from the media and public for the Bank of Japan to disclose simulation results with various exit scenarios, which it conducted internally. Governor Kuroda has repeatedly rejected the call, stressing that relevant information should be provided when the exit is approaching by referring to the case of the Federal Reserve, which had released its exit principles in September 2014, before the first hike of the federal funds target range in December 2015. In a press conference on 16 June 2017, Governor Kuroda also stressed that premature disclosure could risk disturbing the markets, by pointing out that in the case of the Federal Reserve the early disclosure of the exit principles in 2011 was revised in 2014 with the sequence from a decline in the size of assets to a short-term interest rate hike reversed (Kuroda 2017).

No remarks on net income loss under the new Bank of Japan Act

One growing concern in Japan is that there is no provision specified under the (new) Bank of Japan Act – which was revised in June 1997 and has been in effect since April 1998 – regarding the case of net income loss and associated possible actions by the government. This is in contrast with the previous (old) Bank of Japan Act, whose supplementary provisions specified that the government must supplement the Bank of Japan’s losses if legal reserves are not sufficient to cover such losses. The removal of the provisions under the new Act is a reflection of the views of major central banks that it is important to maintain the soundness of their balance sheet to maintain operational independence. In case a central bank’s net income loss continues and draws down its own capital, this may necessitate recapitalisation by the government and thus it may lose operational independence. Therefore, the Bank established its own accounting rules to maintain financial soundness.

The Bank of Japan’s stated capital is 100 million under the old and new Acts. The government contributes 55% of the stated capital and the private sector the remaining 45%. In principle, this stated capital is expected to remain intact or recapitalisation by the government is not envisaged – even in the case of net income losses. This view seems to remain unchanged both under the new and old versions of the Bank of Japan Act. This suggests that the government needs to find other ways to support the Bank of Japan financially. One possibility is that the government may end up buying back JGBs from the central bank at higher prices.

The case of net income loss and negative equity

The Bank of Japan’s current account balances amounted to ¥350 trillion in June 2017. To simplify the issues, let me consider two simple scenarios.

The first scenario is that the three types of interest rates on these reserves are unified at 0.1% now and this level is maintained for a long time since 2% has not yet been achieved. The Bank’s payments will then rise from ¥187 billion currently to ¥350 billion annually. Since the annual interest income from JGB holdings is at present a little over ¥1 trillion, net income will remain positive and thus the adverse impact on the central bank’s balance sheet won’t be materialised.

The second scenario is that the Bank of Japan achieved about 2% now and will start to raise the interest rate on excess reserves rapidly toward 1%. Then, the Bank’s interest payments will jump from ¥350 billion to ¥3.5 trillion. Given that interest income on JGB holdings remains unchanged, the income statement is expected to deteriorate. In order to avoid net income loss, the first thing that the Bank will do is to utilise the accumulated provisions of about ¥4.6 trillion in March 2017 (Shirai 2017). When these provisions are used up, the central bank will face net income loss. The Bank could then use legal reserves (about ¥3.1 trillion in March 2017). After those reserves are used up, the Bank’s balance sheet will face a negative equity. The Bank of Japan will probably need to get support from the government and discuss how to find financial resources to pay for the general and administrative expenses and costs (about ¥190 billion in 2017), and pay interest to commercial banks on excess reserves. During the period with net income loss, the payments of (a) dividends (5% of share face value, amounting to $5 million in fiscal year 2016) and (b) payment to the government (¥481 billion) are expected to be suspended.

The period during which the central bank would be faced with net income loss is likely to be long and it is likely to take many years before net incomes turn positive. If the Bank of Japan starts to purchase shorter-term JGBs now, this period will be shortened. This is because it takes less time for the Bank to replace old JGBs with new JGBs with a higher coupon rate through the reinvestment process and thus a period of low-income flows is likely to continue for less time.

To conclude, discussions on the Bank of Japan’s balance sheet problems are relevant to the case of raising the interest rate on excess reserves. This is likely to occur in the more distant future as 2% inflation is unlikely to be achieved any time soon. Therefore, current discussions should be focused on how to taper the JGBs, ETFs, and REITs without causing severe disturbances to the markets – if the central bank follows the appropriate sequence, as set out above.

One growing concern is that the Bank of Japan may never able to reduce the amount of net JGB purchases toward zero so that the negative interest rate will remain permanently – if the Bank follows the appropriate sequence. Given that the negative interest rate policy has been subject to strong criticism in Japan, this means that the central bank may find it inevitable to eliminate the negative interest rate while the JGB tapering process continues slowly. This may flatten the yield curve since short-term yields will rise and longer-term yields will remain under downward pressures. To avoid a further distortion on the yield curve, raising the 10-year yield target with a range as suggested by this article could be one important decision that the Bank of Japan has to make. 


Cecchetti, S and K Schoenholtz (2016), “The Bank of Japan at the policy frontier”,, 7 December.

Federal Reserve (2014), Policy Normalization Principles and Plans, 17 September.

Kuroda, H (2017), The summary of Governor Haruhiko Kuroda’s Press Conference on June 16. 

Shirai, S (2017), “Mission Incomplete: Reflating Japan’s Economy”, Asian Development Bank Institute.



Article Categories:
Economic Institutions Blogs

Leave a Comment

Your email address will not be published. Required fields are marked *