Three Tests
in Four Days
Sep. 15 funding pressure from Treasury settlement and taxes · Sep. 16 FOMC · Sep. 17–18 Bank of Japan · three different events compressed into four days, with the two policy meetings pointing in the same direction
Two central-bank meetings weighing rate hikes, two days apart.Ahead of them comes a funding-market pressure day when settlement and taxes coincide · short-term funding remains quiet so far, while long-term yields have already moved higher.
What is packed into four days
This week brings three different events into a four-day window from Sep. 15 to 18 · Sep. 15 is the day of funding flows, while Sep. 16 and Sep. 17–18 are the days of policy.
The Fed decides the day after the settlement-and-tax squeeze, followed immediately by the BOJ · each decides whether to hike · all three are connected through bond markets, so the outcome of one changes the setup for the next.
Sep. 15 · Quiet so far
The Sep. 1 note identified four indicators to watch around Sep. 15 · through the final pre-settlement observation on Sep. 11, none shows a funding-stress signal.7
| Indicator flagged in NO.01 | What to watch | Latest published reading |
|---|---|---|
| SOFR − IORB | Whether the spread persists for several days | −3bp 09.11 · SOFR 3.62 · IORB 3.65 |
| Upper tail of SOFR distribution | Whether the 75th and 99th percentiles widen first | 3.67 / 3.69% Sep. 11 · 99th percentile −1bp d/d |
| SRF usage | Whether use of the standing backstop rises | Effectively zero Sep. 8–11 · max $21m |
| EFFR | Whether stress spills into fed funds | 3.63% Sep. 11 · unchanged since August |
If anything, funding conditions look easier than in August · SOFR fell from 3.66% on Sep. 3 → 3.65% on Sep. 4 → 3.64% on Sep. 8–9 → 3.62%, remaining below IORB · it had jumped to 3.68% on Aug. 31, the final business day of the month, on roughly $3.06tn of volume, but reversed within one day.7
Even in the final observation before settlement, SOFR remained below IORB and the upper tail actually narrowed · on Sep. 11 the 99th percentile fell 1bp to 3.69%, while the 75th percentile held at 3.67% · volume was $2.87tn · in other words, internal repo-market tension had not yet appeared in prices ahead of settlement.7
The $119bn settlement and estimated-tax payments occur on Sep. 15 itself · calm conditions through Sep. 11 are only a pre-signal, not confirmation of the BASE scenario from NO.01 (“nothing happens” or a one-day rise) · confirmation comes from Sep. 15 SOFR and EFFR published on Sep. 16, together with SRF usage.
Sep. 16 · Core inflation is at a 5½-year low. Why hike?
August CPI, released Sep. 11, showed headline inflation of +0.4% MoM and +3.4% YoY, in line with expectations · yet core inflation was +2.4% YoY, the lowest since March 2021 · only core MoM, at +0.3%, came in slightly above the +0.2% consensus.4
Energy is what split the picture · the BLS explicitly said gasoline rose 3.9% in August and accounted for more than one-third of the monthly increase in the all-items index · the energy index rose 2.1% MoM, while gasoline was up 27.4% YoY · against the backdrop of the U.S.–Iran conflict, Brent stood at $104.61 on Sep. 11 (+17.6% over one month · +56.2% over one year), WTI at $100.05, and only nine cargo vessels transited the Strait of Hormuz versus a June daily average of 30.4,8
The rise in hike expectations cannot be explained by inflation data alone · the current 3.50–3.75% target range has been unchanged for nine months since December 2025, and the policy turn began at the June FOMC · the median 2026 year-end dot rose from 3.4% in March to 3.8% in June · on the median, the path flipped from one cut to one hike, and 9 of 18 participants placed the year-end rate above the current midpoint (cut 1 · hold 8 · hike 9) · it was also Chair Warsh’s first meeting after taking office.15
The second repricing came at Jackson Hole on Aug. 28 · Chair Warsh upgraded his characterization of the economy to “strengthening,” said the labor market was consistent with full employment, and argued that tight credit spreads and private domestic demand made it difficult to call the current policy stance restrictive · the 2Y Treasury yield rose 11.1bp that day.9
The Sep. 11 CPI reaction was also concentrated at the front end · the U.S. Treasury 2Y yield rose 7bp to 4.63%, while the 10Y rose just 1bp to 4.96% and the 30Y fell 2bp to 5.35% · a curve shape in which the front end reacted much more strongly alongside the repricing of the policy path.14
Core inflation is at a 5½-year low, yet a supply-driven headline shock is adding pressure to hike · monetary policy cannot directly reverse the supply shock itself, so the key question is whether inflation expectations begin to move · the SEP inflation projections and rate dots released on Sep. 16 will show whether the Fed views the shock as temporary.
Sep. 17–18 · For the BOJ, the transmission path matters more than the odds
The Bank of Japan is expected to raise its policy rate at the Sep. 17–18 meeting from 1.00% to 1.25% · the case rests on Q2 GDP growth of +0.9% YoY, rising wages, inflation near 2%, and August producer prices up 7.6%.5,6
Japanese yields have already risen sharply · the 10Y JGB stands at 2.99%, up 138bp over one year, while the 30Y is at 4.05% · USD/JPY is 153.6, with the yen strengthening 3.7% over the past month · the U.S. Treasury Secretary said not to “bet against the yen.”6,10
It is important not to mix two separate transmission paths · if the U.S. and Japan each hike 25bp in the same week, the policy-rate differential does not mechanically narrow · a carry unwind is not opened simply by a smaller rate gap, but by a relatively more hawkish BOJ.
A rate hike itself is not a low-probability event · what is closer to the low-probability case is the liquidation chain that could follow · the surprise July 2024 hike helped trigger a large yen-carry unwind that shook global markets, but conditions are different this time: a hike is already substantially priced and the yen has strengthened over the past month · this is not a forecast, but the widest potential transmission path to flag when the U.S. settlement date, FOMC and BOJ decisions cluster together.5
The key is not simply whether the Fed hikes, but how long-term yields and the yen react when the U.S. and Japan move in the same direction in the same week.
What to watch
How it reaches Korea
How the Korean market has responded ahead of these three events depends on the window you choose · over the broader period since end-August, KOSPI actually rose and the won strengthened, while KTB yields increased across the curve.12,13
| Indicator | 08.31 | 09.11 | Change |
|---|---|---|---|
| KOSPI | 6,820.02 | 6,909.91 | +1.32% |
| USD/KRW | 1,368.6 | 1,345.9 | −KRW 22.7 |
| KTB 3Y | 3.84% | 4.01% | +17bp |
| KTB 10Y | 4.31% | 4.54% | +23bp |
| KTB 30Y | 4.53% | 4.71% | +18bp |
In other words, immediately before this week, equities and FX had held up while rates moved first · KOSPI briefly traded above 7,000 during the period, then fell 1.76% (−124.01pt) on Sep. 11 to close at 6,909.91, surrendering 7,000 · a session in which the oil spike, higher U.S. yields and CPI-event risk converged.13
Foreign investors sold a net KRW 17.2tn of KOSPI stocks from July through Sep. 10 · yet the index still rose because of other corporations · from Aug. 20 to Sep. 11, net KOSPI buying by other corporations totaled KRW 25.15tn; over the same period, Samsung Electronics (KRW 7.774tn) and SK hynix (KRW 18.7973tn) executed KRW 26.5714tn of share repurchases on the article’s aggregate figures, explaining most of that flow · over those three weeks, buybacks absorbed net selling of KRW 11.4tn by foreigners and KRW 15.77tn by individuals.13,17
The decline in USD/KRW also needs to be separated from pure won strength, because broader dollar weakness may have contributed.
By planned share count, progress stands at 55.9% for Samsung Electronics (29.8m shares / through Nov. 21) and 45.5% for SK hynix (10.95m shares / through Nov. 19) · at the current pace, purchases could finish before the scheduled November end dates.17
WGBI phase-in, by contrast, continues through the final November tranche · the equity-market cushion from buybacks may weaken first, followed in November by the end of mechanical incremental demand from new bond-index inclusion · tracking funds will still hold and rebalance after inclusion is complete, but it is worth monitoring whether the two markets’ flow supports weaken sequentially through the autumn.11,17
U.S. 30Y 5.35% · Japan 30Y 4.05% · Korea 30Y KTB 4.71% · a period of jointly elevated global ultra-long yields6,12,14
Bank of Korea base rate 3.00% (back-to-back hikes on Jul. 16 and Aug. 27) · there is no Monetary Policy Direction meeting in September; the next is Oct. 22 · even after the impact of the U.S. and Japanese decisions is known, roughly a month separates Korea from its next scheduled rate decision (an extraordinary meeting remains possible if conditions change abruptly)16
A BOJ hike would favor yen strength and dollar weakness, supportive for the won, but that effect could be offset if risk aversion rises at the same time
Brent in the $104 area · feeds into Korean inflation and growth with a lag through import prices and the trade balance8
Persistent foreign net selling in KOSPI · the Sep. 30 sixth WGBI tranche is the next bond-market checkpoint11
This week’s question is not “Will they hike?”
When two meetings already tilted toward hikes arrive two days apart, which moves first: the short-term funding market that is still quiet, or the long end that has already repriced?