Kashkari's long-end diagnosis: a supply problem the Fed does not plan to solve

Declining to rank the drivers ^[5] is the part that costs money. Deficit-driven issuance and AI-related capital spending ^[4] push the long end the same way but resolve on entirely different clocks. Borrowing needs persist until fiscal policy changes. A data centre build ends when the buildings are finished. A thirty-year buyer who believes the construction story dominates is running a carry position with a visible exit; one who believes the borrowing story dominates is underwriting a steeper curve indefinitely. Kashkari's framing lets both investors hold the same bond for opposite reasons, which is how a crowded trade forms without anyone noticing.

The instrument mismatch deserves spelling out. Treasury Secretary Scott Bessent's expanded buybacks retire older, less liquid issues to smooth market functioning ^[7]. That is plumbing. If Kashkari is right that trading is already orderly ^[1], the plumbing does not need smoothing, and the pressure on long-dated bonds that the operation is meant to ease sits in the level, which repurchasing off-the-run paper is poorly built to move.

Nor is there help arriving from the short end. Kashkari has positioned himself among the more hawkish FOMC members, arguing for modest tightening against inflation risk ^[8], with any move gated on employment and inflation readings rather than bond market volatility ^[9]. By his own account the swing factor is the data: upcoming inflation and employment reports will show whether the rest of the committee shares his composure ^[12].

Take the quoted levels at face value. Near 4.7% at ten years and above 5% at thirty ^[2] leaves at least 30 basis points of positive slope at the far end of the curve ^[13], at yields the market has not consistently held since the mid-2000s ^[3]. On the Fed's telling, that slope is being set by how much paper has to be placed and how much of the AI build wants the same pool of savings ^[4], with the committee's contribution limited to keeping inflation expectations anchored ^[6]. The marginal buyer of long paper is therefore being paid by supply, not by an expected policy turn, and should size the position on that basis rather than on a rescue nobody at the Fed has offered.