Fed Meets Sept. 15-16 After August CPI Held at 3.4%; Futures-Implied Odds of a Rate Hike Near 90%

The 0.1 That Moved the Market, and the 0.1 That Didn't

Two numbers came out of the same government report on Friday, Sept. 11, and they point in opposite directions. Consumer prices rose 3.4% over the past year through August, the Bureau of Labor Statistics said, exactly matching July's pace^[1]. Strip out food and energy, and the annual "core" rate actually cooled, dropping to 2.4% from 2.5%^[1]. That looks like inflation easing, not worsening.

But the same report also showed core prices rose 0.3% in August alone, one tenth of a point above what economists expected^[2][3]. That single-month overshoot is what moved markets. Within hours, futures pricing showed the odds of a Federal Reserve rate hike at its Sept. 15-16 meeting jumping to nearly 90%, up from about 70% the day before^[8].

Both numbers are real. Both come from the same BLS release^[1][2][3]. Which one you lead with determines whether the story reads as "inflation is easing" or "the Fed has no choice but to hike." That split is now playing out as a public fight between the president who appointed the Fed's chair and the chair himself.

A Number Only Traders Vote On

The "near 90%" figure making headlines isn't a poll of economists. It comes from CME Group's FedWatch tool, which derives odds from the prices of fed funds futures, contracts that pay off based on where the Fed's rate actually lands^[8]. Traders are putting real money behind a guess, so the number swings every time sentiment shifts.

And it has swung a lot. It sat near 33% before Fed Chair Kevin Warsh spoke at the Jackson Hole conference in late August, jumped to about 66% right after his remarks, eased back to around 60% in early September, then spiked to near 90% once the August inflation numbers landed^[8][10][11]. That's not a forecast anyone is accountable for. It's a running bet that keeps changing.

Warsh, whom President Trump named to chair the Fed in a nomination announced Jan. 30, 2026, gave traders the clearest signal yet at Jackson Hole^[9][12]. He said 54% of the 199 components in the Fed's preferred inflation gauge, the PCE price index, had risen more than 3% over the prior year^[9]. That gauge is different from the CPI report driving this week's headlines. It's a broader measure of consumer spending that the Fed weights differently and watches more closely^[9]. Warsh's point was that if only a few categories were rising fast, you could blame a one-time shock. When more than half of everything is running hot, he argued, that points to something the Fed's tools are actually built to fix.

The Fed Chair Trump Picked, Now Facing Trump's Pressure

Here's the part that reads like a contradiction until you sit with it: Trump chose Warsh to run the Fed, and Warsh may now raise rates against Trump's explicit wishes^[4][12]. In the days before the CPI release, Trump, Vice President JD Vance and Treasury Secretary Scott Bessent all publicly urged the Fed not to raise rates, according to CNBC^[4]. Trump has posted on Truth Social demanding cuts, called FOMC members "clowns," and threatened trade restrictions tied to rate policy^[4]. He's also said, more mildly, that Warsh will "do what he has to do^[5]." Warsh, for his part, has said the president has had no effect on his decisions^[4].

The administration's argument isn't just political noise. Much of 2026's inflation, they say, comes from tariffs and from an energy shock tied to the U.S.-Iran conflict, one-time price jumps that a rate hike can't undo^[9][10]. Raising rates, on this view, doesn't fix imported inflation. It just makes borrowing more expensive on top of it.

Vance has made the sharpest version of that case about housing. He's said Trump cares about interest rates because he wants Americans to afford homes, and Bessent went further, saying the U.S. housing market is already in a recession because of Fed policy^[6][7]. Freddie Mac's average 30-year fixed mortgage rate hit 6.76% for the week ending Sept. 10, 2026, the highest in more than a year^[13]. On a $100,000 loan, that works out to roughly $649 a month in principal and interest alone, before taxes or insurance^[6][7][13].

There's a wrinkle here that undercuts both sides a little. Mortgage rates track long-term Treasury yields more closely than they track the Fed's own short-term rate^[6][10]. So a Fed hold doesn't automatically make mortgages cheaper, and a hike doesn't automatically make them worse. The 10-year Treasury yield has been climbing toward 5% on its own, partly because investors are nervous the Fed might not act on inflation at all^[10].

Why the Fed Chair's Word Is the Whole Job

Underneath the specific numbers sits a structural problem that would exist no matter who chaired the Fed. A central bank mostly works by being believed. If investors think the Fed will tolerate high inflation, they demand higher yields to hold long-term bonds, and borrowing gets more expensive anyway, hike or no hike^[9][10]. That's why Warsh leans so hard on the breadth statistic: it's evidence he can point to that isn't just one bad month.

That same logic cuts against Trump's political timeline. The FOMC meets roughly two months before the November 2026 midterms, and polling has shown voters angry about both prices and rates^[4]. A rate cut now would ease pressure before Election Day, but it would also test whether markets still believe the Fed answers to inflation data rather than to the White House. If investors decide the Fed folded under political pressure, the reaction could show up not in short-term rates but in a further jump in those already-rising long-term yields^[10].

Trump himself has acknowledged the tension in his own appointment. He's said he wouldn't have chosen Warsh if he'd wanted a chair who'd automatically hike rates^[4]. That's an admission that the job doesn't bend to the person who filled it, at least not entirely.

What the Rest of the World Is Watching Instead

Outside the U.S., the political fight barely registers. What matters is the mechanical effect of U.S. rates on the dollar. When Warsh's Jackson Hole speech pushed hike odds up in late August, the dollar strengthened, gold gave back part of a roughly 14% gain it had made that month, and Asian stocks fell^[10]. When hike bets eased in early September, the pattern reversed: emerging-market stocks and currencies rallied as the dollar softened^[11].

The reason is straightforward. Higher U.S. rates make holding dollars more attractive, which pulls investment money out of emerging markets and toward the U.S. It also makes dollar-denominated debt more expensive for countries that owe it^[11]. None of that turns on whether Trump or Warsh wins this argument. It turns on where the rate actually lands.

Bank forecasts already diverge on that point. Deutsche Bank expects two quarter-point hikes this year, in September and December. Bank of America expects three, adding October, which would push the Fed's target range to 4.25%-4.50% by year's end, up from today's 3.50%-3.75%^[2].

The Same Report, Read Two Ways

Coverage of all this splits less on the facts than on which fact gets top billing. CNBC's framing asked what happens "if Warsh wavers," treating a decision not to hike as a possible failure of nerve rather than a defensible reading of mixed data^[4][5]. CBS News called a September hike "all but guaranteed," hardening a market probability into near-certainty in its own headline^[8]. Fox Business centered Bessent's housing-recession claim without weighing it against the inflation numbers the Fed is responding to^[6]. Fortune described the bond market as "daring the Fed to do something," language that makes a hike sound inevitable rather than chosen by a committee^[10].

None of these outlets got the underlying facts wrong. The CPI numbers, the FedWatch odds, and Warsh's Jackson Hole statistic all check out against the primary sources^{[1][2][3][8][9]}. What differs is which of the two truths, cooling core inflation or the monthly overshoot, gets the headline, and whether the Fed's coming decision reads as a response to data or a test of one man's resolve.

The FOMC meets Sept. 15-16. Whatever it decides, the 3.4% inflation rate and the 6.76% mortgage rate won't move overnight either way^[1][13]. A quarter-point change is small next to the gap between 3.4% and the Fed's 2% target, and its effects take months to show up. The question both sides are actually arguing over, whether this inflation is the kind interest rates can touch at all, won't be settled by one vote.