When the Bank of Japan raises rates, it does not make money scarce — it cannot, with a balance sheet swollen by years of quantitative easing. It simply pays banks more to hold the reserves it already created, and arbitrage does the rest. That single fact explains almost everything about how a modern BoJ hike works, why it can be implemented in a single morning without selling a single bond, and why the transmission to your screen — TIBOR, lending rates, JGB yields, bank stocks — follows a predictable sequence.
The question is timely: Japanese short-term rates are a live policy variable again, and as of the 2026-09-18 session the yen was trading around 157 to the dollar, modestly weaker on the day. But this guide is not about any one meeting — for the current target level, the vote, the statement, and any dissents, go straight to the BoJ’s own published decision. It is about the machinery that makes any such decision real — machinery that works very differently from the textbook, and from the 1990s.
The target: the uncollateralized overnight call rate
The BoJ’s operating target is the mujintanpo call rate — the uncollateralized overnight call rate, the interest rate at which banks lend yen to each other overnight, without collateral, in the interbank call market (call shijo). Trades are largely intermediated by tanshi companies, specialist money-market brokers. After each Monetary Policy Meeting, the BoJ issues a directive instructing its Financial Markets Department to encourage the call rate to remain at around the target level, and it publishes the realized rate daily on its website.
Two things are worth internalizing. First, the target is an overnight rate — the shortest possible maturity — and everything longer is market-determined expectation built on top of it. Second, the word encourage in the directive is now close to a formality: under the current framework the BoJ does not need to fight the market to hit the target. Here is why.
The lever: interest on reserves, not reserve scarcity
The instrument that actually moves the call rate is the complementary deposit facility (hokan toza yokin seido) — the system under which the BoJ pays interest on the excess reserves that banks hold in their current accounts at the central bank. When the board decides to hike, the operational step is administrative: the rate applied to those balances is raised in line with the new target. No bond sales, no reserve draining, no open-market struggle.
The logic is pure arbitrage, often called a floor system:
- No bank will lend overnight in the call market at a rate meaningfully below what it can earn risk-free by leaving the money parked at the BoJ. The interest-on-reserves rate therefore acts as a floor under the call rate.
- No solvent bank will borrow at a rate meaningfully above the BoJ’s standing lending facility (the complementary lending facility), which acts as a ceiling.
- With reserves abundant — the legacy of years of quantitative easing — the call rate settles at or just under the floor, and moving the floor moves the market.
The tiering history matters for reading the fine print. Between 2016 and 2024, under the negative-rate framework, reserves were split into three tiers (a basic balance, a macro add-on balance, and a policy-rate balance) earning different rates, precisely so that a negative marginal rate could steer the call market without taxing the entire reserve stock. Since the 2024 exit from negative rates, the framework has been simplified to essentially a single rate on excess balances. One residue remains: some call-market participants, such as investment trusts, do not have access to the deposit facility. They will lend slightly below the floor rather than earn nothing, which is why the realized call rate typically prints a touch under the policy rate. This is the well-known leaky floor, and it is normal — not a sign the BoJ is missing its target.
Why this is not the 1990s
In the pre-QE era, the mechanism for steering rates was fundamentally different, because reserves were scarce. The BoJ steered the call rate by fine-tuning the daily supply of reserves through money-market operations — adding or draining funds via the tanshi brokers — so the call rate was a genuine market-clearing price of a scarce commodity, anchored loosely by the official discount rate. Hiking meant engineering scarcity, and the call rate could overshoot or undershoot while operations caught up.
Today the BoJ still holds a large share of the outstanding JGB market and banks sit on an enormous stock of excess reserves. Scarcity-based control is impossible at that scale, which is exactly why the floor system exists. Two practical consequences follow:
- Rate policy and balance-sheet policy are separate dials. A hike does not require, and does not imply, faster JGB run-off. Quantitative tightening proceeds on its own slower track. Expecting bond sales on hike day is the single most common mechanical misunderstanding.
- Hikes raise the central bank’s own interest bill. Raising interest on reserves means the BoJ pays more on its enormous liabilities while earning fixed low coupons on its legacy bond portfolio — an accounting and political issue, not an operational constraint, but one that shapes the debate around every hike. For commercial banks, by contrast, the net effect is not automatic: it depends on how quickly asset yields reprice relative to deposit costs, and on the shape of the yield curve.
The transmission chain, stage by stage
| Stage | Instrument | What happens and what to watch |
|---|---|---|
| Same day | Overnight call rate | Reprices to just under the new interest-on-reserves rate; check the BoJ’s daily publication. |
| Days to weeks | TIBOR | The Tokyo Interbank Offered Rate — the term interbank benchmark — resets to reflect the new overnight path; it is the reference for much corporate floating-rate lending. |
| Weeks to months | Short-term prime rate, deposits | Banks reprice the short-term prime lending rate; floating-rate mortgages, the majority of Japanese housing loans, typically reset on semi-annual schedules, so households feel hikes with a lag. |
| Continuous | JGB curve | The front end moves with the expected policy path; the long end adds term premium and is distorted by the BoJ’s own holdings. |
| Continuous | Equities and FX | Bank margins benefit; long-duration, leveraged sectors (real estate, utilities) face pressure; the yen channel feeds exporters. |
On the equity leg, sector behavior tends to follow duration logic when rates are in focus. As a dated illustration from the 2026-09-18 session (TOPIX-17 ETF proxies): Power & Gas fell 2.60% and Real Estate 1.42%, lagging the market — classic rate-sensitive laggards — while bank shares were slightly lower rather than euphoric, a reminder that deposit repricing and curve shape matter as much as the level of the policy rate.
Calculation method: reading the 2-year JGB against the policy rate
The cleanest way to see what the market expects the BoJ to do next is the 2-year JGB yield, and you can do the arithmetic yourself. We deliberately leave the policy-rate input for you to fill in from the primary source, so the method stays valid at any target level. Using Ministry of Finance official yields as of 2026-09-17:
- Take the policy rate. Use the current target for the uncollateralized overnight call rate, taken from the BoJ’s latest published directive — the directive itself, not a secondhand figure.
- Take the 2-year JGB yield. MOF’s daily file shows 1.868%.
- Subtract. The gap between the 2-year yield and the overnight target is the market’s message.
- Interpret. A 2-year yield approximates the average overnight rate expected over the next two years, plus a term premium. If the 2-year yield sits well above the current target, the market is pricing the overnight rate to average materially above today’s level over the horizon — in other words, further tightening is expected within two years. The arithmetic alone cannot tell you how many hikes, when they land, or where the terminal rate sits, because it cannot separate the expected path from the term premium; that split requires an explicit model.
- Know the limit of the method. Do not run this calculation on the 10-year (2.993% on the same date). At that maturity, term premium and the distortion from the BoJ’s own bond holdings dominate; the 2y–10y slope of about 112bp is not a pure read on policy expectations.
Common mistakes when tracking a hike
- Conflating hikes with QT. The policy rate and the balance sheet are separate instruments on separate timetables.
- Expecting instant loan repricing. Legacy fixed-rate loans do not move; floating mortgages reset on a schedule. Transmission to the real economy is measured in quarters.
- Assuming bank stocks always rally. If deposit rates reprice quickly or the curve flattens because the market reads a hike as near-terminal, net interest margins can disappoint.
- Treating the small gap between the call rate and the policy rate as a failure. The leaky floor is a structural feature of participant access, not a control problem.
FAQ
Does the BoJ sell bonds to raise rates?
No. Under the abundant-reserves floor system, a hike is implemented by raising the interest rate paid on banks’ excess reserves at the complementary deposit facility. Bond run-off (quantitative tightening) is a separate, slower policy track.
Why does the call rate print slightly below the policy rate?
Some call-market lenders, such as investment trusts, cannot access the BoJ’s deposit facility. They accept a rate a few basis points below the floor rather than earn nothing, pulling the traded rate slightly under the target. This is expected behavior.
How quickly do households feel a hike?
The overnight rate moves the same day and TIBOR within weeks, but the majority of Japanese mortgages are floating-rate loans tied to the short-term prime rate and typically reset semi-annually — so the household cash-flow impact arrives with a lag of months.
What is the single best free data source for tracking implementation?
The BoJ publishes the realized uncollateralized overnight call rate and each policy directive on its website, and Japan’s Ministry of Finance publishes official daily JGB yields across the curve — together they let you verify both the anchor and the market’s expected path.
Sources
- Bank of Japan — Monetary Policy Meeting decisions, statements, and directives (primary source for the current target, the vote, and any dissents): www.boj.or.jp/en/mopo/mpmdeci/index.htm
- Bank of Japan — call money market data and the complementary deposit facility: www.boj.or.jp/en/statistics/market/short/mutan/index.htm
- Ministry of Finance Japan — official daily JGB yield table (source of the 2026-09-17 2y/10y figures): www.mof.go.jp/english/policy/jgbs/reference/interest_rate/index.htm
- JPX — equity market and investor statistics (sector and turnover data): www.jpx.co.jp/english/markets/statistics-equities/
- Nikkei Indexes — official index data: indexes.nikkei.co.jp/en
Disclaimer: This is an information and analysis publication, not investment advice. See our Methodology for data sources, standards, and our corrections policy.