Japan’s 10-year government bond yield crossed 3% on September 1 — the first time since September 1996 — while the BOJ prepares to hike at its September 17–18 meeting. The world’s funding source is repricing, and every asset built on cheap Japanese money gets marked down with it.
Last week’s memo ended on a watch item: if the BOJ hiked into three-decade-high long yields and JGB outflows followed, “the debasement story stops being about one currency and becomes a global re-rating of fiat.” That is exactly what this week delivered — starting with the line itself.
| Date | Event | Level / Size |
|---|---|---|
| Sep 1 | Japan’s 10-year government bond yield briefly surpassed 3% — the first time since September 1996, a thirty-year high (Reuters)[1] | >3.0% |
| Sep 2 | Tokyo stocks fell sharply as renewed US strikes on Iran sent oil surging and bond stress hit the market — the Nikkei closed down nearly 1,900 points (News on Japan) | -1,889 pts |
| Sep 2 | BOJ Governor Ueda said rate hikes remain “on the table at every meeting” and that he hoped to debate economic and price risks with his board this month; markets priced an 80–90% probability of a hike to 1.25% (Reuters, WSJ, Euronews)[6] | 80–90% odds |
| Sep 3–4 | Brent crude settled at US$92.68/barrel on Friday after renewed US-Iran strikes — up ~7.6% for the week; US diesel hit a record average of $5.85/gallon (BT, AAA)[5] | $92.68/bbl |
| Sep 3–4 | The yen rallied to its strongest in a month — The dollar fell 1.5% to ¥156.17 on Thursday as markets raised bets on BOJ hikes and carry-trade unwinding began; Japanese officials warned against weak yen and stood ready to intervene (BT, ST)[4][7] | ¥156/USD |
| Sep 4 | The US side of the same tape: the 10-year Treasury near 4.79% and 30-year mortgage rates up to ~6.7% — global long-end stress is synchronised, not Japan-specific (BT)[2] | ~4.8% / 6.7% |
The headline number — a yield that had spent thirty years below the line now trading above it — matters less than what it implies: the funding cost of the world’s largest source of cheap money is no longer cheap.
The yen carry trade is the most important plumbing in global markets, and it runs on one input: Japanese borrowing costs. For two decades that input was near zero. The mechanics are simple — a borrower raises yen at ~0%, converts to dollars or other currencies, and buys higher-yielding assets abroad. The spread between what they pay in Tokyo and what they earn elsewhere is the profit, and it funds a huge slice of global risk-asset demand: US equities, emerging-market bonds, private credit, real estate.
Three structural facts make this week’s move more than a headline:
1. The BOJ no longer owns half the market it used to. At the end of last year, the Bank of Japan’s share of outstanding JGBs dipped below 50% for the first time in years[10]. Its balance-sheet runoff has been transferring government debt from a central bank that never sells into a market of price-sensitive buyers. When the marginal buyer is a fund with an alternative universe, yields respond to global conditions instead of domestic policy alone.
2. Foreigners are now the dominant voice in JGB trading — far beyond their holdings. Overseas investors hold roughly ¥94 trillion of JGBs (about 8% of outstanding), up from ¥67 trillion when the BOJ began cutting purchases in mid-2024 and just ¥41 trillion before quantitative easing. But their footprint in price discovery is much larger: as of end-2025 they accounted for just under half of all cash JGB transactions and over 70% of JGB futures volume (BOJ).[8][9] The same investors who bought at 0.5% are now the ones deciding whether 3% is enough — and they can exit faster than anyone else.
Source: Bank of Japan — “Japanese Government Bonds Held by the Bank of Japan” (statistics).
3. The carry trade has two kill switches, and both just turned on. First, funding cost: as JGB yields rise toward 3%, borrowing yen to invest abroad gets more expensive every quarter. Second, exchange rate: this week the yen strengthened to its highest in a month — meaning the overseas profits of existing carry positions shrink in yen terms even before any selling begins. Funding up, returns down: that is a forced-rebalancing setup for the largest cross-border position in global finance.
Governor Ueda has spent this week doing arithmetic in public. The September 17–18 meeting is now priced at an 80–90% probability of a hike to 1.25%. His own words — hikes “on the table at every meeting,” with economic and price risks on this month’s agenda — leave little room for a pass.[3][6]
The constraint is not inflation in the textbook sense; it is the fiscal side. Japan carries the highest gross government debt among major advanced economies, and with 10-year yields above 3% for the first time since 1996, the cost of rolling that debt has moved from a footnote to a line item. Every 25 basis points of policy tightening raises the servicing bill on a balance sheet measured in trillions — which is why the BOJ’s pace matters more than its level.
The second constraint arrived this week via oil. Brent settling at US$92.68 after renewed US-Iran strikes — up ~7.6% for the week, with diesel at record levels[5]. It feeds directly into Japan’s inflation math: an energy importer with no meaningful domestic supply converts every dollar of crude into imported inflation, which argues for tightening even as it squeezes household budgets and corporate margins. The BOJ is being pushed from both sides — fiscal cost pushing toward caution, oil-driven inflation pushing toward action.
The market’s read so far: hike in September, keep the guidance hawkish, and let the yield curve do the rest of the work. That combination — a policy rate at 1.25% with long-end yields above 3% — is the steepest Japanese curve in decades. It is precisely the environment that makes foreign JGBs attractive to global buyers while making yen carry expensive for global borrowers.
The transmission channel is not “Japanese stocks fall.” It is “the funding cost of everything that borrowed cheaply rises” — and the portfolio implications follow from there.
1. An Asia tilt (EIMI/CNYA) has a direct exposure to this trade. Faster Japanese tightening has historically been a headwind for regional equities: as carry positions unwind, funding flows reverse out of Asian risk assets first, and yen strength squeezes the earnings of Japanese exporters inside broad EM funds. The briefs this week flagged exactly that — if the September hike lands with hawkish guidance, expect the unwind to extend rather than stop at one session.
2. Singapore sits on both sides of the plumbing. As a reserve-currency hub and home to MAS’s diversified reserve management, local markets absorb global flows — September just posted SGX’s highest monthly turnover in five years as that flow arrived. The same channel now carries carry-unwind outflows. For a Singapore-based portfolio, the practical read is not “exit Asia” but “don’t be long duration and long EM simultaneously without knowing which leg breaks first.”
3. Energy is the inflation input that complicates every rate path. Brent near $93 with diesel at record levels means CPI stickiness in the US, Europe, and Japan alike — which kills rate-cut hopes everywhere and keeps long-end yields elevated even if equities stabilise. For withdrawal-rate math: a portfolio priced for falling rates is now pricing for a world where they don’t.
4. The cheap-finance era’s beneficiaries are the names to watch, not the ones to chase. AI infrastructure debt, private credit, and leveraged real estate were all built on the assumption that Japanese funding would stay cheap indefinitely. That assumption is now a variable with a September 17–18 resolution date. The question for any equity-heavy stack is not whether these sectors fall — it is whether your book’s duration matches the new cost of money.
The bear case for “Japan breaks the world” has real substance, and it starts with Japan’s own track record of absorbing shocks that looked systemic.
1. Japan has done this before without a global crisis. The 1990s saw yields fall to zero after a lost decade; the 2006-07 tightening cycle pushed the BOJ from 0.5% to 0.75% and the yen strengthened sharply — and global markets absorbed both. A 3% 10-year yield is high by Japan’s modern standards, but it is not unprecedented in Japanese history, and the last two episodes did not produce a carry-trade collapse of the kind now being priced.
2. The BOJ still holds the QT dial. Even after its share of JGBs dipped below 50%, the Bank can slow or pause balance-sheet runoff to manage supply — a tool it has used repeatedly since 2024. A central bank that controls both the policy rate and the pace of its own selling is not a passive participant in its own bond market, and “yield spike” scenarios assume it does nothing.
3. Intervention risk cuts both ways — and officials are ready to use it. Japanese authorities warned this week against weak yen and stood ready to intervene[7]; the same machinery can be pointed at FX moves in either direction if they become disruptive. A government that will defend the currency is also a government that can dampen the carry-unwind’s most violent expression — at the cost of credibility, but with an immediate effect on volatility.
4. The US side may absorb the shock before Japan does. With the 10-year Treasury near 4.79% and mortgage rates above 6.7%, American borrowers are already feeling the long-end squeeze. If US fiscal or political pressure forces a yield cap on the other side of the Pacific — as last month’s Treasury buyback program attempted — the global tightening story becomes asymmetric. Japan-specific positioning unwinds faster than a synchronised scenario would suggest.
Four events in the next two weeks will tell you whether this is a regime shift or a positioning spike:
1. The BOJ’s September 17–18 decision — and the guidance that accompanies it. A hike to 1.25% is already priced at 80–90%; the information content is in the statement. Hawkish guidance (further hikes signaled, QT continuation) extends the carry unwind into Q4. A softer-than-expected statement would deflate the trade quickly — and test whether JGB yields hold above 3% on fiscal grounds alone.
2. Whether foreign JGB outflows follow the yield spike. The watch item from last week’s memo, now live: if overseas holders — roughly 8% of outstanding debt but nearly half of all cash trading volume — start rotating out. The rotation is visible in MOF weekly flow data[9] and in the yen’s reaction to each print, the debasement story becomes a global fiat re-rating rather than a Japan-specific curve event.
3. Oil through the Strait of Hormuz. Brent near $93 after renewed US-Iran strikes is the inflation input that keeps every central bank on the hawkish side of its dilemma. A sustained break above $100 makes synchronised global tightening a base case; a quick de-escalation removes the second pillar under the BOJ’s corner and gives the yen-carry trade room to rebuild.
4. Whether US long-end yields stabilise or extend. The 10-year near 4.79% with mortgages above 6.7% means the US is already pricing a higher-for-longer regime. If US yields keep rising while Japan hikes, global duration gets squeezed from both ends and risk assets feel it first. If US yields cap — via fiscal action or demand destruction in credit — the squeeze becomes Japan-specific and manageable.
The thirty-year line was not drawn because 3% is a magic number. It was drawn because everything below it was built on the assumption that Japanese money would stay cheap forever. That assumption now has a price, a meeting date, and a probability attached to it — and markets are starting to mark down every asset that borrowed against it.
This analysis is based on publicly available data as of 2026-09-05. For related coverage, see When the Treasury Becomes a Market Maker and The Debasement Trade.