The ECB Called Its Own Hike a No-Brainer and the Curve Broke; We Re-Specify Our Own Hormuz Trigger and Find the Real Near-Zero in Gas, Not Crude

The ECB raised to 2.50% unanimously, told the market inflation stays well above target for an extended period, and declined to pre-commit to anything — so the market priced 88bp more tightening into September 2027. Add the hottest annual PPI of 2026 and a Houthi ground offensive that took a Red Sea port, and the US 2-year moved 17bp with the 10-year through 4.97%. But take the move apart before calling it a de-anchoring: real yields carried 10.5 of the 13.2bp, the 5y5y forward did not move, and gold fell 2.5%. This is orderly tightening, not a loss of faith. The US August CPI prints today.

(1) A unanimous "no-brainer", and the sentence that did the damage. The deposit rate went to 2.50%, the second hike since the Iran war began, and President Lagarde called the decision a unanimous "no-brainer" while stressing the euro-area economy's surprising resilience. The rate was ~99% priced and was never the event. The event was one sentence in the statement: "inflation is set to remain well above target for an extended period." Lagarde paired it with an explicit refusal to pre-commit — "can't anticipate what will be the next move" — and the market resolved that ambiguity hawkishly, now pricing a further 88bp of tightening with the peak in September 2027, with officials already floating October. The case was built by the energy channel: euro-area August HICP 3.3% with energy at +14.3%. A central bank tightening into a supply shock is defending the expectations channel at the cost of demand it never weakened — a deliberate, and defensible, choice.

Global Macro · ECB

(2) The long end broke to 4.97% — and the decomposition is the reassuring part. The UST 2-year jumped 17.0bp to 4.594% and the 10-year 13.2bp to 4.973%, with the 30y at 5.378% — a front-end move of that size on a day with no US data tells you the market repriced the Fed on the ECB's tone and the barrel. Now take it apart. Real yields carried 10.5bp of the 13.2; breakevens carried only 2.7bp to 2.413%; and the 5y5y forward barely moved at 2.321%. Gold fell 2.5% to $4,303 — which is what real yields do to a debasement hedge, and emphatically not what a fear trade looks like. We wrote yesterday that a breakeven-led move from here would be a materially worse regime than the one we describe. The test ran overnight and came back on the benign side. That does not mean the inflation problem is solved; it means the mechanism is orderly tightening rather than de-anchoring — and those two call for opposite duration decisions. The one genuine change in character: MOVE at 82.09 from 76.74. Rate volatility has woken up.

Global Macro · Rates

(3) The PPI ran hot — but the heat was diesel, not services. August producer prices rose 0.4% m/m for 5.4% y/y, the highest twelve-month reading of 2026, core 0.3%. The composition decides how much of this ever reaches core PCE, and the composition is lopsided. Final demand GOODS rose 1.1% with energy up 4.2%, and diesel alone, up 24.1%, accounted for more than a third of the entire goods increase. Final demand SERVICES rose just 0.1% — transportation and warehousing +2.3%, but trade services actually fell 0.2%. So the pipeline pressure is an oil pass-through, not a broadening. That is the single most useful thing to carry into today's CPI, because the components that translate most cleanly into core PCE are the services ones, and those stayed quiet. A hot headline with a soft services core is the most likely shape of today's print — and it is the shape that lets the Fed hold.

Global Macro · Inflation

(4) We re-specified our own Hormuz trigger, because the evidence behind it had a benign cause. Yesterday we armed a fifth Phase-1 trigger on a VLCC-exit stall — no very large crude carrier had cleared Hormuz outbound since 2 September. Today we found the mechanism we had not established. Gulf producers have for months been shuttle-shipping crude through the Strait and reloading it via ship-to-ship transfer — a documented, routine workaround. That is precisely how aggregate flow holds near 8.6 mb/d while laden VLCCs never appear in the outbound count, and it is corroborated by reporting that oil flows have rebounded in recent weeks. A trigger that fires on a workaround is a bad trigger, so we re-specify it to total crude clearance — direct plus ship-to-ship — which needs a flow print rather than a transit count to fire. We could have left it armed and looked prescient for another week. The observation was sound; the inference was not; we armed on the inference. Publishing that is the entire point of running triggers in the open.

Iran-Hormuz · SoH Monitor

(5) And the same evidence found the real near-zero. It is gas, and it has been for two months. No LNG carrier has transited the Strait of Hormuz since 11 July — sixty-one days. The monthly series reads 0 transits in March, 4 in April, 8 in May, 40 in June, then nothing. Qatar shipped 18 cargoes in the first six months of this war against 509 a year earlier — a 96% collapse and roughly $24bn of lost sales — with force majeure to European and Asian buyers extended into October and cancellations running into early November. Twenty-one LNG carriers sit idle in the MidEast Gulf (13 in ballast, 5 laden, 4 berthed), and Qatar has begun bringing empty ships home. The asymmetry has a physical cause, and it is the same fact that rescued crude: LNG cannot be ship-to-ship transferred. We spent two editions hunting a physical near-zero in the crude channel. It was in the gas channel the whole time. That relocates the European inflation risk from Brent to TTF — which is precisely where the ECB's actual problem lives, with European gas near four-year highs.

Iran-Hormuz · The gas channel

(6) A second chokepoint opened — and we are not booking it as Hormuz. On 10 September the Houthis seized the Yemeni Red Sea port of Mokha after a ground offensive, taking another point on the coastline overlooking Bab el-Mandeb, roughly 80km away, through which about 12% of world goods and a tenth of seaborne oil normally pass. Saudi Arabia answered with nearly forty strikes on Houthi targets the same day, and Pakistan warned Tehran that continued Houthi action risks triggering a joint defence arrangement involving Riyadh, Ankara and Islamabad — which would widen a US-Iran maritime war into a regional bloc confrontation. Brent went through $105 on the news and marks $108.08. Our discipline is to keep this out of the Hormuz ledger: it is regional premium, and that file is a claim about one Strait. But it changes what you should do with our openness number. A corridor at 43% is survivable largely because alternative routing exists. That assumption is now itself contested at its own chokepoint. And for the record: Kharg's export terminal was NOT hit — vicinity explosions only, no confirmed damage to loading infrastructure, storage or offshore moorings, and no export halt.

Iran-Hormuz · Second chokepoint

(7) Athens eased, but the banks went up against the tape. The ATHEX General Index closed 10 September at 2,697.58 (-0.15%) on turnover of about EUR394m, 36 advancers against 64 decliners. The banks index rose 0.45% to 3,200.83 — outperforming a European sector and a global tape that were both falling, which is a rate-hike beneficiary behaving like one the day its central bank hiked and signalled more. Alpha +1.72%, National +0.32%, against Eurobank -0.53% and Piraeus -0.33%; Eurobank separately had its outlook raised to positive by a ratings agency. The other side of the rotation reversed: Motor Oil -3.57% and Cenergy -2.85% gave back, while Athens International Airport +3.95% and GEK TERNA +2.80% led. The domestic cost side keeps hardening — unleaded near EUR2.10/litre, wholesale power toward EUR200/MWh — with new diesel and heating-oil measures still under consideration. Eight sessions to reclassification.

Greece · Market

(8) Greek owners are selling the steel as well as the paper. A thread worth pulling, because it now runs through two asset classes at once. This week two of the strongest balance sheets in Greek dry bulk issued equity at a discount — Safe Bulkers' EUR80.4m placement at EUR6.70, and Star Bulk's Athens book, which closes today at 16:00 with pricing due tonight against a EUR23.00-25.50 range. Now the same instinct is visible in the asset market: Greek press reports 316 vessels put to auction as major groups sell into elevated values, with five leading tanker sellers already transacting $1.71bn and reinvesting in newer tonnage, and a large owner exploring a $200m VLCC sale. Meanwhile Tsakos reported Q2 profit of $228m on record revenues and rising freight rates. Read the combination: operators are earning record numbers from the war premium and simultaneously monetising both their equity and their ships into it. That is not a bearish signal about rates — it is a statement about where these owners think mid-cycle asset values sit relative to today's prints. We re-mark our Star Bulk and Safe Bulkers cases on tonight's pricing.

Greece · Shipping

What We're Watching

• US August CPI, 08:30 ET today. Watch the SERVICES core, not the headline — the PPI's heat was diesel.
• Star Bulk's Athens book closes 16:00; final pricing tonight, Euronext Athens trading from 16 Sep.
• An LNG transit through Hormuz — none since 11 July; now the cleanest de-escalation signal in the file.
• Bab el-Mandeb — whether Mokha becomes a second closed chokepoint, or stays a premium event.
• Wed 16 Sep — FOMC, a 25bp HIKE modal at ~57%, with the SEP and the dots.
• 18 Sep rebalance / 21 Sep Developed Market reclassification for Euronext Athens.

The Bottom Line

Two legs we have carried for a week both paid: fade the dovish, do not fade the oil. The ECB hiked and called it a no-brainer, the PPI printed the hottest annual rate of 2026, the 2-year moved 17bp and Brent is $108. But the most useful thing in the tape is a negative result — the long-end move was real-led, the 5y5y forward did not move, and gold fell. We named a breakeven-led leg as the regime-changing tail yesterday and it did not fire, so this stays orderly tightening at a restrictive level rather than a de-anchoring, and duration near 5% gets more attractive on that reading, not less. On the corridor we corrected ourselves for the third edition running, and it was worth it: the VLCC-exit stall we armed a trigger on has a benign mechanical explanation in ship-to-ship transfer, and the same evidence located the genuine near-zero in gas — shut since 11 July, Qatar down 96%, because LNG cannot be transferred the way crude can. That moves the European inflation risk from Brent to TTF, which is where the ECB's real problem lives. Today the CPI decides Wednesday, and the line to read is services, not the headline.

Synthesised from: Delphic SoH Monitor (11 Sep 2026), Daily Market Commentary (11 Sep 2026), Regime Radar · Market levels: ~/Dropbox/delphic-market-data store parquets; US/Europe/Athens = 10 Sep cash close, rates/commodities/FX = 11 Sep marks · Greek single names: personal-portfolio/builds/prices.json (10 Sep ATHEX close) · ECB monetary policy decision and press conference, 10 September 2026; BLS Producer Price Index, August 2026 · Kpler (LNG transit halt, throughput, ship-to-ship practice); reporting on the Mokha seizure and Saudi response · Greek all-outlet scan (11 Sep): capital, mononews, powergame, euro2day, ot.gr, protothema, newmoney, iefimerida