The 10-year pushed past its January-2025 ceiling to roughly 4.81%, the highest since November 2023, inside a bond selloff that is global rather than US-specific.
A synchronised move across the US, UK, German, French and Japanese long ends cannot be a US fiscal story, because those five sovereigns are not running one fiscal policy. What they share is a term premium — the compensation for holding duration — and a term premium repricing on the same session in five markets means the price of duration itself moved, not any one issuer's credit. Oil is the obvious candidate: an energy shock raises expected inflation everywhere simultaneously and is the one input all five curves hold in common.
Scenarios 2
This is the oil shock passing through, and it unwinds with the oil shock. Brent's four-session run is the whole explanation; if Hormuz transit normalises, the global term premium compresses back and the US 10-year's "highest since November 2023" print is a spike rather than a level.
Directly falsifiable on two series this map already carries daily: Brent's daily print and the 10-year's. If Brent gives back the move toward the high $80s and the 10-year does not follow within a week, the oil explanation is insufficient and something structural is being mispriced as a shock.
Precedent — The 2022 global bond selloff after the Ukraine invasion. Yields rose together across the G7 on an energy-driven inflation impulse, and the correlation broke apart within two quarters as each central bank reverted to its own domestic data.
The energy shock is the trigger and the AI buildout's issuance is the reason it is not absorbing. Duration is being supplied faster than it is being demanded — September corporate issuance forecast at $215bn on top of a record August — so an inflation impulse that would once have been absorbed now moves the curve.
The discriminating test is relative: if US long-end yields rise more than the UK, German and French moves on the same session, the excess is US-specific supply rather than the shared inflation impulse. This map has Bloomberg's own count of 55 days above 5% on the 30-year and named analysts (Briggs, Misra) arguing buybacks are being dwarfed by AI-buildout supply — the disagreement is already on the record and dated.
Precedent — The 2003-2007 GSE and structured-credit issuance wave. Steady duration supply from a single sector compressed the market's capacity to absorb shocks, and the effect showed up as an unusually sharp reaction to news that would previously have been routine.
Context
capital-context.yaml's rate_regime reading is a hawkish hold at 3.50-3.75% built on domestic data — the July payrolls miss, weak retail sales, the Cook removal fight. It has no cross-border term-premium leg at all, which is precisely the gap yesterday's JGB crossing exposed and today's five-market move widens.
Threads 1
This is generated reasoning, not a sourced fact — the mechanism and scenarios above are the model's read on what this item could mean for capital flows, tagged with its own confidence rather than stated as settled. Back to Global Capital.