The 10-year just cleared 5% with a hike already priced — the bond market isn't waiting on the Fed to break

Six business days ago the story was a 5% ceiling that long-end buyers defended. By Tuesday morning that ceiling had broken: the 10-year yield touched 5.041% intraday and held above 5.025% on the day, per CNBC, the highest closing-equivalent print since the summer of 2007. The 30-year traded to 5.40%, the 2-year to 4.69% — a 33bp 2s10s slope that is positive but still well below the 85bp long-run average.

What makes this episode different from October 2023's failed flirt with 5% is that the front end is moving in the same direction. CME FedWatch put the odds of a 25bp hike at the September 15–16 FOMC above 92%, per CNBC's live coverage. Duration risk and a hawkish Fed are being priced simultaneously — a regime that didn't exist when the 10-year last touched 5%, because back then the Fed was still cutting into the financial crisis.

The July 28–29 FOMC minutes, released August 19, documented the most fractured committee of this cycle: a 9–3 vote to hold the funds rate at 3.50–3.75%, with three members preferring an immediate 25bp hike. The same minutes noted that the market was already fully pricing a hike by September and another by end-Q1 2027 — pricing that, until CPI confirmed it, was treated as a tail bet.

August CPI, released September 11, validated the hawks. Headline inflation stayed at 3.4% year-over-year and core around 2.4–2.5%, per Trading Economics and the WSJ coverage. The Reuters September 4–9 economist poll, captured here, shows the share of economists seeing at least one hike this year more than doubled from August (only 30% saw a hike next week; the rest expected hold, but a clear majority now expected hikes by year-end).

When only duration is repricing (the 2023 episode), long-duration equities and growth stocks lose. When the front end is repricing with it, the damage spreads into anything whose business model depends on a falling or stable rate environment: rate-sensitive banks with floating-rate loan books, REITs whose dividend yield competes with the 10-year, utilities leveraged into capex, and homebuilders already absorbing 6.7%+ mortgages.

For the major banks, the curve shape matters more than the absolute level. JPMorgan Chase, Wells Fargo and Bank of America all reported Q2 2026 (period ending June 30) net interest income above year-ago: $25.5B at JPM (+6% YoY), $12.3B at WFC (+5% YoY), and $16.0B at BAC (+9% YoY), per each issuer's Q2 earnings release. A steeper curve extends that tailwind by letting them reinvest maturing securities and loans at higher rates; a flat-and-falling curve breaks it. The yield curve is now positive and steepening — the opposite of the inversion that preceded the 2023 regional-bank stress.

Banks fund themselves at the short end and lend at the long end; a positive, steepening curve with the Fed hiking is the textbook earnings tailwind. But the same move is a tax on every variable-rate borrower. Wells Fargo's Q2 supplement (filed July 14, 2026) shows commercial real estate balances down 1% quarter-over-quarter as the bank's own commentary cites 'lower interest rates' as the segment headwind — which means the September move to 5% reverses that drag.

On the liability side, the cost of the 30-year auction is now north of 5.4%. Long-dated corporate issuance that priced through the summer at 5.0–5.1% has to roll at 5.4%+ in 2027. Term-premium decomposition puts roughly half of the September 10-year move in the 'risk and uncertainty' bucket — the ACM 10-year term premium sat at 0.76% in late August per CEIC and ~1.02% on the StreetStats composite — meaning supply concerns and inflation risk, not rate-cut expectations, are doing most of the work.

At a 10-year yield above 5%, the cost-of-capital math for dividend-paying equities flips. Per Yahoo Finance, fewer than 4% of S&P 500 stocks now yield more than the 10-year — the Treasury has reclaimed the yield crown for the first time since 2009. That pulls capital away from REITs and utilities whose entire thesis is dividend income.

Homebuilders absorb the mortgage channel directly. D.R. Horton's TTM net debt/EBITDA sits at 1.24x with return on equity at 12.8%, per the company's TTM through Sep 15, 2026 — healthy, but leverage-priced. KB Home and Lennar operate at similar leverage profiles, and each additional 25bp in primary mortgage rates historically clips 3–5% off single-family housing demand.

Over days–quarters, the binary is the Fed. A 25bp hike on September 16 with hawkish-dot guidance validates the move; a hold with a strong dissent count is the surprise that could knock 30–40bp off the 10-year in a session. Either way, the term premium is unlikely to compress quickly — supply pressure and geopolitical risk are not FOMC-driven.

The regime that matters for 1–3 year positioning is not 'high rates' — it is 'high rates with a Fed that hikes into them because inflation refuses to cooperate.' That has structural implications: bank NIM stays elevated but loan demand softens; REITs must grow AFFO/distributable cash flow to defend their spread over the 10-year; utilities need capex visibility (transmission, data centers) to justify any multiple expansion; and long-duration assets stay range-bound until either the Fed pivots hawkish-to-dovish or the term premium normalizes.

The history suggests patience. The 2007 10-year print above 5% held for several quarters before the GFC broke it lower, and rates did not revisit 5% for sixteen years. A repeat — with the Fed now actively hiking to defend it — would imply either a 2027 policy reversal or a growth scare. Either scenario rewards balance-sheet quality and punishes leverage, which is why the long-duration iShares 20+ Year Treasury ETF has underperformed cash despite a 5%+ headline yield.