The 10-year Treasury yield surged to 5.12% on September 23, rising 16 basis points on a session and reaching its highest level since July 2007. The S&P fell 0.65%, the Nasdaq fell 1.1%. The driver was an S&P Global flash PMI showing composite business activity expanding at its fastest rate since July 2021. Markets repriced the October FOMC meeting from approximately 55% hike probability to 70%. The Federal Reserve’s target rate currently sits at 3.75–4.00%, set on September 16. The next FOMC meeting is scheduled for October 28–29. The bond market is running one meeting ahead of the central bank.
1. The Data Point
The S&P Global flash PMI for September 2026 showed composite US business activity expanding at its fastest rate since July 2021. The services component drove the move; manufacturing showed renewed strength. [Established — CNN Business, “10-year Treasury yield hits 5.1% for first time in 19 years,” 23 September 2026; Charles Schwab, “Blistering Yield Rally Overshadows Trump-Xi Talks,” 25 September 2026.] The 10-year yield surged 16 basis points to close at 5.12%, the highest since July 2007 — outpacing the 5.04% reached before the September 16 hike.
The Sounding No. 47 analysis — “The Vigilante Rate” — described the September 16 sequence: the 10-year hit 5.04% before the FOMC acted, then the Fed followed. That piece identified a structural shift in which the bond market was setting the rate the Fed then ratified. September 23 extends that pattern into the next cycle. The bond market has moved above the September 16 pre-hike peak before the FOMC has met again. The implication is unambiguous: markets believe the September 16 hike was insufficient, and they are applying the same forward pressure that forced the September decision.
2. What 5.12% Actually Prices
October hike probability moved from approximately 55% to 70% in a single session. [Established — CNN Business, 23 September 2026, citing CME FedWatch repricing.] The Federal Reserve’s last public guidance, in the September 16 dot plot, showed 16 of 18 officials projecting at least one additional hike in the November–December window. October was not the consensus meeting for that hike.
The PMI print has changed that calculus by demonstrating that the September 16 hike has not slowed economic activity. A central bank that hiked on September 16 and faces an economy expanding at its fastest pace since July 2021 one week later is a central bank that must decide whether its tightening has had any effect at all — or whether the supply-side inflation drivers are simply immune to the demand channel through which rate increases operate. [Assessed with moderate confidence — analytical inference from PMI timing relative to policy cycle; standard monetary transmission lag literature supports this framing.]
The structural answer, as the Purser has tracked from Sounding No. 29 onwards, is that a substantial share of current US inflation is supply-side in origin: energy from Hormuz, tariff pass-through from Canada and US-China trade policy, and now a PMI-confirmed demand component that was never weak enough to justify the pause the prior analysis assumed. The Fed faces a situation in which its rate increases are insufficient to suppress supply-side inflation, while the demand side of the economy has remained strong enough to produce PMI readings that themselves become inflation inputs. [Assessed with moderate-high confidence — consistent with the August CPI 3.7% print above consensus, confirmed in Sounding No. 39 analysis.]
3. The Summit Non-Effect
Xi Jinping arrived at Joint Base Andrews on September 23. On the same day, the 10-year yield surged 16 basis points and equities fell. The Busan truce extension — announced as the summit output on September 24 and 25 — produced no bond market relief. [Established — Charles Schwab market update, “Blistering Yield Rally Overshadows Trump-Xi Talks,” 25 September 2026.]
The Sounding No. 51 Purser analysis — “Détente Is Priced. Breakthrough Is Not” — identified the relevant asymmetry: the rally on September 23 had already priced cooperation. What it had not priced was resolution. A two-month truce extension is cooperation, not resolution. The unresolved variables — Taiwan, rare earths, Iran, AI chip controls — are not in the price. The bond market looked at the same summit output the equity market received, and concluded that the inflation trajectory had not changed.
The Sounding No. 51 prediction was that “the S&P 500 will give back at least half of the September 23 rally within five trading sessions of the communiqué.” The S&P fell 0.65% on September 23 and is down incrementally further through the summit’s conclusion. Whether this constitutes the predicted retreat is a borderline case; the full five-session window runs to approximately September 30. [Assessed — the Sounding No. 51 Ledger prediction on equity retreat is tracking to be confirmed but the resolution window has not closed; the five-session clock from the communiqué of approximately September 25 runs to September 30–October 1.]
4. The Fed’s Communication Problem
The Federal Reserve does not meet until October 28–29. Between today and that meeting, three data releases will materially affect the rate decision: the September jobs report (scheduled for October 3), September CPI (scheduled for approximately October 13), and the first Q3 GDP advance estimate (scheduled for approximately October 30 — after the meeting). [Established — BLS and BEA release calendar; Kiplinger, “What to Look Out for in Economic Data This Week,” September 2026.]
The Fed’s communication challenge is that 70% hike probability in the market is itself a tightening instrument — the anticipation of a rate increase raises borrowing costs before the Fed acts, which is part of how forward guidance works. But if the October data show any weakness — a softer jobs number, a cooling CPI — and the Fed does not hike at the October meeting, it faces the same credibility problem the Sounding No. 37 analysis identified: a central bank perceived as having been pressured out of a decision that market prices had already made.
Kevin Warsh, as the Fed chair who was “bullied” by the bond market into September 16, now faces the same dynamic one meeting earlier than the dot plot projected. The Sounding No. 47 Ledger prediction — that the 10-year would test 5.20% within five trading sessions of a December hike — may resolve earlier than December if October proceeds as September did. [Assessed — conditional on a further yield move above 5.20% occurring before a second hike; this is now a live scenario rather than a December-only scenario.]
Prediction: The 10-year Treasury yield will remain above 4.90% through the October 28–29 FOMC meeting, absent a materially weak September jobs or CPI print; if October CPI (released approximately October 13) prints at or above September’s 3.7%, the FOMC will hike 25 basis points at its October 28–29 meeting, raising the federal funds rate to 4.00–4.25%, ahead of the dot-plot projection of a November–December move.
Confidence: Moderate (yield floor above 4.90%) / moderate (October hike conditional on CPI at or above 3.7%). The primary failure mode is a September jobs report showing significant labour market deterioration on October 3, which would provide the Fed political and analytical cover for a hold at October 28–29 regardless of the inflation data. The Sounding No. 44 Ledger prediction on the November hike conditional remains open and would need to be revised if the October hike occurs ahead of schedule.
Resolution: October 13 (CPI data) and October 29 (FOMC decision). Check: BLS for CPI; CME FedWatch for probability evolution; Federal Reserve press release for rate decision.
Bottom line: A 5.12% 10-year yield on a PMI print, one week after the September 16 hike, is the bond market saying the hike was not enough. October hike probability at 70% is the market saying the next move will come before the dot plot scheduled it. The summit produced a two-month truce extension and changed nothing about the inflation inputs. The Fed faces a choice between following the bond market into an accelerated tightening path, or attempting to hold against a market that has already priced the move — which is exactly the position it found itself in on September 14, and which resolved in one direction.