A Coin-Flip Fed: When the Treasury Secretary Says Don't Hike, and the Market Prices One In

The weekend before a coin-flip Federal Reserve meeting, the man who pays the United States' bills stood up and talked policy around the one number he does not control. Treasury Secretary Scott Bessent said he would not speculate on what the Fed does on September 16 — then spent the rest of the answer explaining why it probably shouldn't do much. "We've seen a supply shock, and traditionally you don't raise [rates]... unless", he argued, adding that core inflation sits at "a very moderate level".

That is a Treasury secretary publicly coaching the central bank out of a rate hike, the weekend before a decision the futures market now prices as a toss-up. It deserves a closer read than the headline suggests.

A pause turns into a coin flip

The easing cycle ran through most of last year without drama. Under Jerome Powell, the Fed cut its benchmark rate three times from September 2025, reaching 3.50–3.75% by December — where it has sat since — and quietly ended its balance-sheet runoff around the same time. Then the baton passed. Kevin Warsh took over as chair, and roughly a hundred days in, at his Jackson Hole debut, he said the quiet part out loud: inflation was "running above our 2 percent target" — headline PCE at 3.7%, core at 3.3% — he called the objective "a firm, fixed target," then delivered the line that moved markets: "We must be confident that underlying inflation is moving to our objective, clearly and at sufficient speed. Otherwise, we have work to do."

In the hours after that speech, the market's September expectation flipped from holding steady with nearly seven-in-ten odds to a true coin flip — 48% to 56% on different platforms. The July meeting had already shown where the pressure was: the Federal Open Market Committee voted 9-3 to hold, with three officials formally dissenting in favor of a hike.

So who is right — the secretary or the chairman? The fair answer: Bessent is making the better economic argument, and Warsh is making the more important institutional one.

Why a supply shock is a bad reason to hike

Start with the economics. The inflation now is largely a supply shock. The U.S. and Israel have been in an escalating confrontation with Iran since the spring — at its peak it effectively interrupted shipping through the Strait of Hormuz — and oil climbed back above $90 a barrel in mid-August when a two-month ceasefire window lapsed. No rate hike can conjure a barrel of oil out of the ground. Higher rates fight supply-driven inflation the only way they can: by destroying demand, raising borrowing costs until someone's marginal purchase stops happening.

That is the trade a central bank should be reluctant to make when the rest of the economy is already cooling. The July jobs report showed payrolls fell by 23,000 and unemployment sat at 4.1%. Tighten into a supply shock and you run the 1970s play: prices stay high, growth disappears. Bessent's "traditionally you don't raise" is, on its own terms, a defensible piece of macro. Calling a supply-shock hike "unusual" is not spin.

The "unless" that keeps the chairman up at night

But the "unless" in Bessent's own sentence is the whole game. A central bank looks through a supply shock only while it trusts that inflation expectations stay anchored — that people won't start granting 3%-plus price increases on the assumption that more are coming.

That is exactly what Warsh says he is watching. Over the past twelve months, 54% of the components of the Fed's preferred price gauge rose by more than 3%. When that many prices move together, the problem stops being "energy" and starts being "psychology." July's core report still showed services grinding up, including a 1.2% monthly jump in financial services and insurance. And Warsh, playing the credibility card, put "65 months of sustained, elevated inflation" squarely on the central bank's own shoulders.

The real fight: Treasury vs. Fed

Here is the part that makes this more than a rates debate: it was never really about economics. Bessent's job is not fighting inflation; it is paying the government's bills — and he has decided the 10-year Treasury yield is too high. So the Treasury, which had barely bothered with buying back its own long-dated debt, doubled its buyback program in August and floated financing it from a roughly $950 billion cash pile, then signaled the buying could go further. In a vacuum that is ordinary debt management. In this context it is jawboning through the balance sheet — and long yields kept rising anyway, the 10-year pushing toward 4.8% at the end of August, levels that measure against the near-two-decade highs touched earlier in the selloff.

That is the loop with teeth. Every time the Treasury leans on rates, the pressure on a brand-new chairman — a Trump appointee — to prove he cannot be leaned on grows. Warsh must sound hawkish for institutional reasons even if he personally leans toward holding. And the more publicly he proves his independence, the higher the long end goes — which is the exact yield Bessent wants down. Two branches of government in a credibility contest, and the prize is the price of money. Fed watchers call the tax this fight puts on long bonds an "uncertainty premium".

Three markets, three answers

For an investor, this is the liquidity cycle's first genuine fork in more than a year. Through the cuts the direction was never in doubt: easier money, supportive conditions, lift everything. Now the market is seriously pricing the first rate hike since the last tightening cycle, and different markets are answering the same question differently.

Equities spent late August near record highs — the S&P 500 finished the week around 5,850, up nearly 4% — apparently betting on hold-and-soft-landing. The bond market, with the 10-year at 4.76% and climbing, is pricing hikes and fiscal risk. And bitcoin, the asset class most exposed to the liquidity cycle, has already run a full fear-to-hope round trip: it sits near $78,000, more than a third below its 52-week high of about $125,000, even after a 22% bounce in the past three weeks.

Three markets, three different answers. That divergence is the signal: nobody knows yet, and the models have not converged. When models disagree, the discipline is to wait and watch the lead indicators, not to take a side on a headline.

The watched items are mundane: oil (does the Strait keep moving, does the ceasefire hold), the next core inflation prints (does 3.3% start falling toward 3%), payrolls (does a soggy labor market make toughness unaffordable), and whether Bessent's buyback cash actually reaches the market or just gets re-borrowed as fresh bills. The decision lands September 16. Until the data picks a branch, a coin flip is a coin flip.

And keep one habit. When a Treasury secretary spends the weekend before a Fed meeting explaining why the Fed shouldn't raise rates, that is information about politics, not about the economy. It tells you the pressure is on. It tells you nothing about whether the data justifies action. Let the numbers do that job — they are the ones paying attention when the talking stops.