Welcome

Newly appointed Bank of Japan (BoJ) governor, Ueda Kazuo has guided the bank through a year marked with moderate inflation, Yen depreciation, and the end of negative interest rates in Japan. Governor Ueda seems intent on shrinking the BoJ balance sheet. In the past twelve months the bank has stopped buying ETFs and JREITs, decreased the amount of JGBs on the balance sheet, and overseen two rate hikes as the rest of the Western influenced world is cutting rates. It is unclear if the BoJ will be able to continue shrinking the balance sheet while keeping markets at levels they are comfortable with in 2025, as a weak Yen and inflationary pressures are likely to persist.

Overview

• Key events and commentary
• Updates on data visualization
• Charts
• 2024 takeaways
• Scenarios for 2025

Key Events and Commentary

Over the past twelve months a number of important events have transpired related to the BoJ. Some were decisions made by the bank, like raising the official policy rate (equivalent to US Federal Funds rate), and others were external events that had a large impact on bank policy and thought, like the rapid Yen rally in August.

On March 19th the official policy rate was moved from -0.1% to 0.1%. The BoJ also ceased purchasing ETFs and JREITs on this date, and stated a plan to taper corporate bond purchases.¹

This policy rate change was more symbolic than anything else. It marked the end to negative interest rates in Japan, and only affected institutions with an account at the central bank. Similarly, ending ETF and JREIT buying signals a step back from the BoJ in supporting markets, but their relatively limited size (37t and 656b Yen respectively) means limited impact on day to day operations in these markets.

During the first few days of May, the Ministry of Finance (MoF) intervened in the USD/JPY currency market, spending 9.7t Yen to take the pair from 160 to 153.

Yentervention – When the MoF uses foreign exchange reserves to intervene in a Yen currency market. Usually characterized by a sharp move in the exchange rate that costs trillions of Yen, and is quickly retraced.

The May 2024 Yentervention was quickly forgotten, with USD/JPY levels returning to just shy of 160 by the end of May. The MoF’s hand was forced by the falling currency, which was becoming a PR crisis as savings were eroded, energy costs rose, and tourists flocked to the country while much of the public felt priced out of traveling abroad.

From July 11 – August 5 the Yen appreciated by 12.5% against the US Dollar in a gigantic move. At the beginning of this move the MoF spent 5.5t Yen to strengthen the currency, which may have set off the cascade of Yen buying.

This is one of the most successful Yenterventions ever, as it started a chain of buying from other market participants and led a to a sustained change in price levels. Side effects included a global risk off, as the Nasdaq, Nikkei, other global indices, and Bitcoin sold off while bonds sharply rallied.

On July 31st the official policy rate was moved once again, this time from 0.1% to 0.25%. On this date the BoJ also decided to begin tapering its scheduled JGB purchases, targeting a 400b Yen reduction in purchases each quarter until April 2026. ²³

In a most likely coordinated move, the BoJ followed the July Yentervention with a surprise rate hike. All else equal, raising interest rates will increase demand for bonds as they provide a higher yield. This effect, or at least the psychological belief in the relationship provided a further push to the Yen, which is needed to buy JGBs.

The 1-2 punch by the MoF and BoJ was the most important period of 2024 for JGB yields, as they dropped rapidly from above 1% (dangerously close to the flexible cap of 1.10%) all the way down to 0.75%, and remained below 1% for months following. The combo of rate hikes and Yentervention led to a calm period for the Japanese financial puppeteers, which I wrote about in ‘Room to Breathe‘.

As the year wraps up, things appear to be reaching reactionary levels. We have the Yen at 158 and the 10-year JGB yield comfortably above 1%.

Before we get to charts and now that we have an overview of this years key events, I want to briefly touch on my work analyzing and visualizing the BoJ’s JGB holdings data.

Updates on Data Visualization

There has been a lot of progress made over the past month in two areas: Charting (looks, options, labels) and creating a cohesive program to explore the data in a variety of ways. For more background read this and this.

Awesome update 1: The charts look a lot better! The y-axis is no longer in scientific notation, x-axis ticks are no longer overlapping on higher timeframes, the colors were intentionally chosen, along with other small updates!

Update 2: Completely rewiring the exploration programs. As written about in one of the blog posts linked above, I used to have a bunch of stand alone exploration programs. This meant lots of repeated code, different user input conventions, and limited charting output.

Over the past month these disparate programs have been combined into a single user interface accessible via the command line and featuring robust input checking, charting optionality, and much more. Here is a list of the new programs capabilities:

• Can show total holdings, purchases, or change in holdings (y-axis).
• Option to view by date or CUSIP number (x-axis).
• Input results can be charted, saved to a csv, or output in terminal table.
• Input is combined to a single line with robust defaults and help messages.
• Short cut to view top 10s over any time period. Largest changes, purchases, holdings, AND least purchases, changes (negative), holdings.
• Short cut to fill in a template with charts and text to summarize the changes since the previous report.

It feels great to make progress here, as I want to be proud of the charts I’m putting out there, and want to spend as little time as possible creating comprehensive charts and updates as possible. Of course, there is still work to do.

Charts

These charts are all generated using data from December 29th, 2023 – December 20th, 2024. Chart 1 shows the total JGB holdings of the BoJ on each report date. Note the BoJ is setup to end the year with less JGBs on the balance sheet than it started with.

Chart 2 shows the net change in all JGB holdings of the BoJ since the previous report. The large negative changes are indicative of a bond the BoJ had a large holding of rolling off the balance sheet. Rarely does the BoJ actively sell bonds, however their portfolio of JGBs is so large that to simply maintain the size they have to purchase more than roll off in a given period, which has not been happening in 2024.

Seeing the net change is very interesting, and it is nice to see the total purchases since the previous report in isolation (i.e. ignoring roll offs or the rare selling), which is what Chart 3 shows.

The line charts (Chart 1 – 3) are useful when plotting data on even intervals, like report dates. The next series of charts (4 – 7) will shows largest and smallest net changes and purchases visualized as bar charts and ordered by report date.

Chart 4 shows the ten largest positive net changes in all holdings of the BoJ since the previous report, by report date.

Chart 5 shows the ten largest purchases of all JGBs by the BoJ. This top ten is fairly different from Chart 4, and shows the value in having both metrics.

Chart 6 displays the smallest values for net change in holdings by report date. This is useful for showing the dates with negative net changes in JGB holdings, and other small values. You can see that the four large negative bars occur in a predictable manner. This is because 10-year JGBs are issued in three month intervals and the BoJ controls such a large amount of the market that whenever one rolls off it has a significant effect.

Chart 7 is the equivalent of chart 6, but for purchases. The ten smallest purchases of 2024 are shown, ordered by report date.

The charts could go on and on. These first seven simply provide an overview of how the BoJ has interacted with the JGB market in 2024. We have yet to dive into the nuance of things like:

• Does the BoJ buy more JGBs when 10-year yields are above 1%?
• Does the BoJ buy more JGBs when the Yen is weak?
• Has the BoJ focused on one part of the yield curve?

Does the BoJ Buy more JGBs when 10-Year Yields are above 1%?

These are the 2024 dates when the 10-year JGB yield started or ended the day above 1%:

• May 22 – June 5.
• June 10 & 11.
• June 25 – July 31.
• November 7 – Present (December 30).

To answer this question I will average the purchases since previous report for each report that was released during the dates above, and compare that average to the average of all other 2024 reports.

Eleven reports fall within the dates listed above and twenty-four that fall outside of them. When I average the purchases of all JGBs since the previous report for these two sets, this is the results:

• 1.720t Yen for reports occurring when the 10-year yield is above 1% (Chart 8).
• 1.886t Yen for reports occurring when the 10-yera yield is below 1% (Chart 9).

The BoJ did not buy more JGBs when the 10-year yield was above 1%. But maybe these results are too high level, as they look at all JGBs. Perhaps we should just look at the 10-year? Here are the results:

• 335b Yen for reports occurring when the 10-year yield is above 1%.
• 484b Yen for reports occurring when the 10-year yield is below 1%.

An even more notable divergence! Separating report dates by yield above/ below 1% may be too simplistic – what if there is a huge buying op that pushes yields below 1%, so the report date showing those purchases falls into the ‘below 1%’ group even though the majority of the days and purchases since the previous report were in the context of 1%+ 10-year yields? Regardless, the simple method I used is still informative.

Does the BoJ Buy more JGBs when the Yen is Weak?

First, what qualifies as ‘weak Yen’ needs to be defined. For the sake of this analysis we will choose 155+ exchange rate against the USD. This gives us a reasonable amount of reports to compare – if we went with 160 there wouldn’t be many reports.

These are the 2024 dates when USD/JPY was above 155 at open or close:

• April 24 – May 1.
• May 8 – July 24.
• November 13 – 15, 20, 21.
• December 19 – Present (December 30).

We’ll do the same thing as we did when looking at buying and JGB yields. The averages for the two groups of report dates and all JGBs can be seen below:

• 1.742t Yen for reports published while USD/JPY > 155.
• 1.876t Yen for reports published while USD/JPY < 155.

For the 10-year, instead of all bond types, we get these numbers:

• 407b Yen for reports published while USD/JPY > 155 (Chart 10).
• 451b Yen for reports published while USD/JPY < 155 (Chart 11).

The choice of 155 USD/JPY level was somewhat arbitrary. Future analysis could adjust this level to see if there is a point where the BoJ does start buying more bonds. This level most likely exists, but changes over time making the timeframe a critical consideration as well.

Has the BoJ Focused on one Part of the Yield Curve?

Something I’ve noticed as I look at various 2024 charts of BoJ JGB holdings for different durations is that lesser watched JGBs (i.e. not the 10-year) have had volume increases, while the 10-year has had a volume decline.

It’s worth reminding/ informing readers that the 10-year volume dwarfs any other duration’s volume. A shift away from the 10-year, to controlling other parts of the yield curve would be interesting, and possibly tell us something about the BoJ’s focuses and concerns.

There are a number of ways we can explore this shift. Since we are looking at 2024 in this piece, I will simply share charts of each bond type’s change in total holding value over the course of the year (Chart 12 – 18). I am omitting Green Bonds due to extremely low volume, and won’t show Inflation Indexed or Floating Rate JGBs as the BoJ is allowing them to roll off the balance sheet.

It is evident from the above charts that the BoJ has been focused on controlling the long end of the yield curve. Increased holdings of 20, 30, and 40-year JGBs is contrasted by reduced holdings of 10, 5, and 2-year JGBs.

Future analysis of the BoJ’s purchases along the duration scale would probably be best done comparing ratios for each year. What percentage of all purchases were 10-year JGBs in 2024, and how does that compare to the number for 2014?

2024 Takeaways

The BoJ was able to avoid raising the 10-year JGB yield cap and decrease the amount of 10-year JGBs on the balance sheet. This is a clear win. The serenity with yields above 1% speaks volumes of the new BoJ regime, one which is comfortable with higher yields and is seeking to loosen the grip it holds over the JGB market.

This move towards higher interest rates was confirmed when the BoJ brought the short term interest rate up to 0.1% and then up to 0.25% a few months later. This most likely indicates the bank believes we are in a higher inflation regime, though it is possible they are simply setting themselves up for a cut in the future.

When looking at the change in total holdings over the course of the year, it is clear that the BoJ is more concerned about long-term rates than short. The reason for this could be many things – fear of hurting consumer mortgage rates, wanting to control the inflation narrative, or very possibly something I’m not thinking of.

By almost any metric, the BoJ is ending the year in a worse place than it started. Benchmark yields (10-year, short term call rate) are up and the Yen is not in a great place as it creeps back towards 160. While having a light hand in the market over the course of 2024 is admirable, it seems like the exception not the rule.

Scenarios for 2025

In light of the above, it seems probable that the 10-year JGB is either a) allowed to break through 1.10% and move up towards 1.5%, or b) the BoJ is forced to do a quite large ‘open market’ operation in the first half of 2025.

If Trump forces a USD devaluation in order to make American export oriented jobs more competitive, we may see the Yen struggle to break through 160 and even fall back towards 100-120, where it was when Trump left office in 2020.

If global fiscal policy remains loose, we are likely to see inflation return to consumer staples, raw materials, and energy. This would spell trouble for Japan, as it would necessitate a hike in interest rates.

In a world where the BoJ is forced to deal with JGB selling past their comfort level, it is possible they end up with lots of higher yielding bonds on their balance sheet relative to the current makeup. This would spell trouble in future years as the interest expenses rise.

My base case is that the Bank of Japan will keep doing what its been doing: meddling in the bond market and keeping yields where they are comfortable with them. Signs of worry will be clear when purchases see a significant jump. Applying this train of thought to 2024, it seems the BoJ was not too concerned with developments during 2024, especially at the short end of the yield curve.

References

1. https://www.boj.or.jp/en/mopo/mpmdeci/mpr_2024/k240319b.pdf ↩︎
2. https://www.boj.or.jp/en/mopo/mpmdeci/mpr_2024/k240731b.pdf ↩︎
3. https://www.boj.or.jp/en/mopo/mpmdeci/mpr_2024/k240731a.pdf ↩︎

Comments

Join the discussion on '2024 Review – Bank of Japan'