NextFin News - Bond traders are pricing in a Federal Reserve rate hike this Wednesday with a conviction that history says they can trust. Interest-rate swaps tied to Fed meeting dates show more than a 90% probability that Fed Chairman Kevin Warsh and his colleagues will lift the benchmark policy rate by a quarter point, from the current 3.50%-3.75% range — roughly 23 basis points of tightening already embedded in prices. The question is no longer whether the Fed will move. It is whether a market this certain has left itself any room for the surprise that actually moves prices.
The Road to 90% Was Anything But Straight
The near-certainty priced into Wednesday's meeting is the end point of a volatile six weeks, not a stable consensus. After the Federal Open Market Committee held rates at its July 28-29 meeting — a 9-3 vote in which Dallas Fed President Lorie Logan, Cleveland's Beth Hammack and Minneapolis Fed Chief Neel Kashkari dissented in favor of a quarter-point increase — traders were fairly confident a September hike was coming. Then the July employment report showed the U.S. economy shedding jobs, and the odds collapsed. On CME's FedWatch tool, the probability of a hold sat near 60% in early August; on the Kalshi prediction platform, a hold was seen at 65%. By late August, after Warsh's Jackson Hole keynote, the decision was being called a coin flip.
The reversal arrived with the inflation data. The August Consumer Price Index rose 0.4% month over month, matching forecasts, with the annual headline rate at 3.4% — flat from July and in line with the consensus of economists. The problem for doves sat underneath the headline: core CPI, which strips out volatile food and energy and is the measure the Fed scrutinizes most closely, rose 0.3% for the month against an expected 0.2%. On a year-over-year basis core inflation eased to 2.4% from 2.5%, but the monthly acceleration — a tenth of a percentage point above consensus — is what moved the policy needle.
Wholesale prices reinforced the picture. The Producer Price Index advanced 0.4% in August, in line with expectations, but ran 5.4% year over year, above the 5.3% estimate and up from a revised 4.8% in July. Core PPI came in at 0.2% month over month, slightly below the 0.3% economists expected — one small crack in an otherwise consistent story. Energy led the headline strength, with gasoline jumping 3.9% on the month and up 27.4% over the past year, accounting for more than a third of the headline increase.
The repricing was abrupt. Hike odds jumped from roughly 70% before the consumer-price report to about 85% immediately after, then pushed above 90% on CME FedWatch. Kalshi showed an 83% likelihood that the federal funds rate would sit above 3.75% after the September meeting. By Monday, interest-rate swaps — the instrument that tracks the Fed's own preferred forward curve — had priced in about 23 basis points of tightening. That is the market equivalent of a locked door: at 90%, the only remaining question is whether the Fed walks through it or breaks the frame.
Why History Sides With the Bond Market
The claim that this level of conviction "has proven right for decades" rests on a simple institutional reality: the Fed does not like to surprise a market that has already moved.
When swaps and fed funds futures price a rate decision at 90% or better, the central bank faces asymmetric pressure. Delivering the hike costs nothing in credibility; defying the market forces the committee to either explain why traders are wrong or accept a disorderly repricing across every duration of the yield curve. Over the past three decades the Fed has built a regime of predictable tightening — moving in 25-basis-point increments, telegraphing shifts through speeches, and using its dot plot and press conferences to narrow the gap between its intentions and market expectations. In that regime, a 90% print is less a forecast than a constraint.
The historical record supports the asymmetry. In the 1994-95 tightening cycle, the Fed raised rates faster than investors anticipated, lifting the funds rate to 6% by early 1995 — the opposite of a dovish surprise. In the 2004-06 cycle, the committee opened with "measured pace" language and delivered 17 consecutive 25-basis-point increases. The pattern is not that the Fed always does what markets want. It is that once pricing reaches the 90% threshold, the committee's bias tilts toward confirming rather than contradicting it.
There is also a mechanical reason the threshold matters. Fed funds futures do not measure opinion; they measure money. A 90% implied probability means traders are willing to put capital behind the view that the effective federal funds rate will average above the current range midpoint for the month. When enough contracts trade at those odds, the effective rate itself begins to drift toward the priced level ahead of the meeting — and the Fed, which targets the effective rate, finds itself steering into a current that already exists. The market, in effect, has already done part of the tightening for the committee.
The bond market is not predicting the Fed. It is building the conditions that make the Fed's decision cheaper.
The Case Against: A Labor Market Cracking Faster Than Inflation Admits
The strongest argument against a hike is not that inflation is tame. It is that the labor market may be weakening faster than the price data admit, and that hiking into that weakness risks a policy error that shows up in inflation only after it is too late to reverse cleanly.
The July employment report, which showed the U.S. economy losing jobs, is what sent hike odds down to a coin flip in the first place. August's nonfarm payrolls print of 162,000 was solid, but it sits on top of a pattern in which the Federal Reserve's own Survey of Professional Forecasters expects unemployment to drift from 4.4% this quarter to 4.5% by the first quarter of 2027. A central bank that tightens into a softening labor market inherits a problem that does not appear in the CPI until months later. That lag is the classic trap of restrictive policy: by the time the damage is visible in the inflation data, the committee has already overshot.
There is also a dissenting signal inside the Fed itself. Governor Christopher Waller said last week that if August figures showed inflation coming down and core prices rising 0.2%, he would support holding rates steady. New York Fed President John Williams set a similar bar: monthly inflation readings consistently at 0.2% on the Fed's preferred Personal Consumption Expenditures measure would indicate inflation is coming down. The August core CPI print of 0.3% crossed both officials' thresholds — but only by a tenth of a percentage point. That is thin ice for a decision that will shape policy for months.
Wall Street is pushing back too. Goldman Sachs warned that markets remained too hawkish even as inflation cooled, and a retail-investor poll found 43% of respondents expecting the Fed to hold rates steady. A poll of economists conducted September 4-9 showed about 70% expecting no change at the September meeting — a survey that predates the CPI report but captures the institutional baseline the committee was debating against. The gap between that 70% pre-CPI baseline and today's 90% is the entire story of what the inflation print changed.
Then there is the oil problem. Diesel prices hit a record $5.85 a gallon in early September, and renewed Middle East hostilities are crimping global oil distribution. The textbook Fed response to an oil supply shock is to look through it. But with inflation above the 2% target for five and a half years — on top of tariff shocks and pandemic aftershocks — the committee may not have the luxury of looking through another energy spike. A second-round effect, in which higher diesel costs feed into freight rates and then into goods prices, is exactly the channel that turns a relative-price shock into persistent inflation.
"The policy question is now not so much whether August CPI broadly confirms the summer improvement in the inflation data but rather whether that summer improvement provides sufficient reassurance for the Fed to look through a renewed oil supply shock with outsized impact on diesel and other refined products," said Krishna Guha, head of central bank strategy and economics at Evercore ISI.
The Second-Order Trade: A Priced Hike Is Not a Priced Outcome
Here is what the 90% print does not tell you. A rate hike that is fully priced in is not the end of the trade — it is the setup for the next one.
The first-order effect is mechanical and already in the price: a 25-basis-point hike lifts short-term rates, tightens financial conditions, and strengthens the dollar. The second-order effect is what happens to the yield curve and risk assets once the decision is delivered. If the Fed hikes and signals that more tightening is coming — if Warsh's press conference confirms the "couple of rate hikes by the end of the year" that BMO Capital Markets senior economist Jennifer Lee now expects — then the front end of the curve has further to go, and the 10-year Treasury, which sold at a 4.834% yield in the September 9 auction (up from 4.683% at the prior sale on August 12, the highest borrowing cost for that maturity since 2007), can push toward 5%.
But if the Fed delivers the 25-basis-point move and strikes a neutral tone — taking the hike as insurance against sticky inflation rather than the opening of a new campaign — the market's 23 basis points of priced tightening have nowhere to go but down. That is the asymmetry traders are buying into: the odds of the hike are 90%, but the odds that the hike is the beginning rather than the whole story are materially lower. In options terms, the market has bought the direction and left the magnitude unhedged.
The third-order implication cuts across asset classes. A confirmed hiking cycle that stops at one or two moves is bullish for the dollar and bearish for long-duration bonds, but it is not necessarily bearish for equities. The alternative — a Fed that falls behind on inflation and has to tighten aggressively later — is the scenario that actually breaks risk assets. The 1994-95 cycle ended in a soft landing; the policy error that preceded the 2001 recession came from waiting too long, not from moving early. Preventive tightening that is front-loaded and then paused is the least damaging path for growth assets, because it buys the committee optionality without forcing a hard landing.
The market has priced the decision. It has not priced the sentence that follows it.
What Comes Next, and What Would Break the Thesis
The base case is a 25-basis-point hike to a 3.75%-4.00% range, delivered with language that keeps a December move alive. The hawkish upside is a hike accompanied by a dot-plot shift or press-conference tone that confirms at least one more increase this year; that would push the 2-year Treasury, which jumped 0.118 percentage point to 4.348% after the Jackson Hole speech — the largest single-day gain following a Fed chair's Jackson Hole address since Alan Greenspan in 1996 — back toward its post-speech highs.
The downside case is narrower but real: a split decision with multiple dissents, or language that explicitly flags labor-market deterioration as the reason to pause. The specific falsifying signal is this: if core PCE prints at or below 0.2% month over month for two consecutive readings while payroll growth falls below 100,000, the structural-stickiness argument collapses and the market's 90% conviction will look like the top of a cycle rather than the start of one.
For the broader market, the beneficiaries of a delivered hike are the dollar and the front end of the Treasury curve; the exposed are long-duration bonds, growth equities with distant cash flows, and borrowers in rate-sensitive sectors. The European Central Bank's September 10 decision to raise its deposit facility rate to 2.50% from 2.25% and its main refinancing rate to 2.65% from 2.40% adds a global dimension — this is not a Fed-alone tightening, which limits how far the dollar can run but reinforces the direction of travel for global yields. When the ECB and the Fed tighten in the same month, the cross-current that usually cushions rate-sensitive assets weakens.
Inflation above target for more than five years has left the Fed with no clean options, only less-bad ones. A 90% implied probability is not a forecast the Fed might defy; it is a constraint the committee has already accepted. The real trade begins when the statement ends.
Explore more exclusive insights at nextfin.ai.
