The Bond Market Cast the Deciding Vote

The Federal Reserve held interest rates steady this week. The bond market disagreed — and the bond market is the one I'd listen to.

On 29 July, the FOMC voted 9–3 to keep the federal funds target at 3.5–3.75%. Three regional presidents — Beth Hammack, Neel Kashkari and Lorie Logan — wanted a quarter-point hike instead. The statement was unusually short and unusually blunt: inflation “remains elevated relative to the Committee's 2 percent goal,” followed by a single flat sentence — “The Committee will deliver price stability.” Kevin Warsh, two months into the chair, has made clear he intends to give markets fewer signals about what comes next.

The market answered anyway. The 2-year yield slipped about 4 basis points. The 30-year rose more than 9, to roughly 5.19% — close to levels last seen in 2007. Equities gave back their gains, with the S&P 500 closing down around 1.5%. A curve that steepens like that, on a day the Fed talks tough, is not steepening on growth optimism. It is steepening on doubt.

A rising long end is a credibility charge, not a rate call. Short rates tell you what the Fed intends. Long rates tell you what everyone else believes about the Fed's intentions. When the front end says “they're done” and the back end says “then inflation isn't,” you are watching investors quietly reprice not the economy but the institution. Warsh framed the decision as a review rather than a pause. Thirty-year money read it as a blink.

In fifteen years of building credit-risk software in the US, I never once saw a small business priced off the fed funds rate. Nobody's working capital line, equipment loan or invoice facility keys off the policy rate directly. They key off the long end, plus a spread that widens the moment lenders feel uncertain. So while the headlines say “rates unchanged,” the actual cost of capital for the smallest, thinnest-margin borrowers in the system went up this week. It always does when the term premium moves. Those borrowers just find out last.

There is a serious case on the other side, and I'll give it its due. The IMF's July update cut 2026 global growth to 3.0% and raised headline inflation to 4.7%, attributing the revision chiefly to energy and food costs after the Middle East conflict. You cannot fix an oil price with a policy rate. Hiking into a supply shock destroys demand you actually needed and lands hardest on the same small businesses I'm worried about. Warsh's patience is defensible.

But five consecutive years above target stops being a supply-shock story and starts being an expectations story. And once expectations drift, you don't get them back cheaply.

The practical warning is for the models. Nearly every credit-scoring and PD framework in commercial use was calibrated on the 2010s — an era of falling rates, compressing spreads and reliable disinflation. Feed that model a world where the long end has a floor under it and the term premium is a live variable, and it will underprice risk with tremendous confidence. In India, where I work on MSME credit ratings, the exposure is even sharper: thin files, informal cash flows, and a global rate anchor no domestic lender controls. When US long yields sit near 2007 highs, capital gets expensive in Pune and Coimbatore too.

My read: the era of assuming the discount rate eventually falls is over, and the risk frameworks built on that assumption are the next thing to break.

If your cost of capital never returns to what you planned around — what in your business stops working first?