The Federal Reserve meets Sept. 15-16, and for the first time in years the question is not whether it cuts. It is whether it hikes.
Two weeks ago, prediction markets said no. On Aug. 15 a contract on Kalshi paying $1 if the Fed raises rates in September traded at 26 cents, a 1-in-4 shot. As of Sept. 2 that same contract is 58 cents. A rate hike is now the favorite.
This is the story of how that happened, what three different markets are saying about it and what it means if you carry a credit card balance.
The flip
Here is the Kalshi September contract, day by day, from our Fed Rate Tracker capture.
For 12 days the number sat between 25% and 34%. Then on Friday, Aug. 28, it went from 30 cents to 51 cents in one session. It has not looked back.
That is not a slow change of opinion. That is a market hearing something and repricing on the spot.
What it heard
The chain of events is short and each link is public.
July 29. The Fed holds rates at 3.50%-3.75%. The vote is 9-3, and all three dissenters want a hike, not a cut. A 9-3 split is unusual. It told anyone paying attention that a hike was already on the table.
Aug. 26. The July PCE report, the Fed's preferred inflation gauge, prints 3.7%. The forecast was 3.6%. The Fed's target is 2%. Core inflation, which strips out food and energy, is 3.3%. Nothing in that report gave the committee a reason to relax.
Aug. 28. Fed Chair Kevin Warsh speaks at Jackson Hole, Wyoming. "Inflation is running above our 2% target," he says. "So the Fed's predominant focus right now should be on prices." He does not say the word hike. He does not need to. CME FedWatch odds of a September increase go from about a third before the speech to better than half after it, and Kalshi's contract makes the same move the same afternoon.
That is the logic chain. A divided committee, a hot inflation print and a chair who chose the biggest stage of the year to talk about prices and nothing else. The market added it up.
Oil is the reason inflation will not come down on its own. The conflict with Iran has kept crude around $80 a barrel through a cycle of flare-ups and cease-fires, and every flare-up shows up at the pump a week later. That is the pressure Warsh was talking about.
Three scoreboards, one game
Kalshi is not the only place pricing this. Polymarket lists the same meeting, and the CME fed funds futures market, where banks and hedge funds hedge interest rate risk, has been pricing Fed decisions for decades.
Here is where all three stood on Sept. 2.
| Market | Hold | Hike 25 bps |
|---|---|---|
| CME fed funds futures | 36% | 64% |
| Kalshi | 39% | 58% |
| Polymarket | 43% | 56% |
Everyone agrees on the direction. They do not agree on how sure to be. Futures traders, the professionals, are about 8 percentage points more confident in a hike than the crowd on Polymarket. Kalshi sits in between.
That gap is the interesting part. All three markets are looking at the same PCE report and the same speech. They came to different numbers anyway. One of them is closer to right than the others, and on Sept. 16 we find out which.
We started capturing all three markets side by side on Aug. 30, and the September meeting is the first one we can grade on all three. We keep the scorecard on the September 2026 meeting page. Right now it reads zero meetings graded, because that is the truth. It will read one on the evening of Sept. 16.
It is not one and done
The part that should get your attention is not September. It is what the market says comes after.
| Meeting (Kalshi) | Hold | Hike 25 | Hike 50 |
|---|---|---|---|
| Sept. 16 | 39% | 58% | 2% |
| Oct. 28 | 33% | 54% | 8% |
| Dec. 9 | 16% | 38% | 39% |
Read the December row. The market gives the Fed a 16% chance of sitting still, and a 39% chance the rate ends up a full half point higher than it is today. Two hikes by year-end is the base case. Three is not off the table.
A note on that table, because it matters. Kalshi's contracts for October and December ask a different question than the September one does. They ask where the rate will be after that meeting compared with today, not what the Fed does at that specific meeting. Two quarter-point hikes in September and October land in the same "50 bps" bucket as one half-point hike in October. So the December row says "we are probably two hikes higher by December," which is the plain-English read anyway. Our tracker separates the two kinds of contracts so nobody compares apples to oranges. The full explanation is on the KXFED settlement page.
What a quarter point means for you
A quarter of a percentage point does not sound like much. Here is the honest math.
Credit cards. These move first and fastest. Card rates are pegged to the prime rate, and prime moves the day the Fed does. On a $10,000 balance, a quarter point is about $25 a year. Not a crisis. But the average card rate is already north of 20%, and this would be the first increase after a stretch of cuts.
Home equity lines and variable-rate loans. Same path as cards, usually within one or two billing cycles.
Car loans. A new $35,000 loan at a quarter point higher costs roughly $4 more a month over five years. If you already have a fixed-rate car loan, nothing changes.
Mortgages. This is the one people get wrong. A 30-year fixed mortgage does not follow the Fed directly. It follows the 10-year Treasury yield, and Treasury yields already moved. After the July hold, long-term yields jumped and the 30-year Treasury hit its highest level since 2007. Mortgage rates near 7% are the result. A September hike is mostly priced in there already.
Savings. The one piece of good news. High-yield savings accounts and money market funds tend to pass along rate increases within a few weeks. If you have cash sitting in a big bank account paying nothing, this is a decent time to move it.
What could change the number before Sept. 16
Two data releases sit between now and the meeting, and both can move the contract 10 points in a morning.
The August jobs report comes Friday, Sept. 4. A weak number gives the doves an argument. A strong one takes it away.
The August consumer price index lands the week before the meeting. The July reading was 3.4%. If August comes in hotter, the hike odds go to the 70s. If it surprises to the downside, expect the market to give some of the last two weeks back.
How to read all of this
If you have never looked at a prediction market before, here is the entire rulebook.
The price is the probability. A contract at 58 cents means the market thinks there is a 58% chance. If the Fed hikes, the contract pays $1. If it does not, it pays zero. Nobody is guessing. People are putting money on it, and the price is where the buyers and sellers meet.
That is why we treat these numbers as a forecast, not a poll. A poll asks what you think. A market asks what you will pay.
We also do not pretend the market is always right. On Aug. 15 it said 26% and it was about to be very wrong. What it is good at is updating fast. The moment Warsh finished speaking the number moved, hours before the analyst notes went out. That speed is the product, and it is why we built the tracker.
Where to watch it
The Fed Rate Tracker has the live board, the day-by-day history behind the first chart in this piece and the inflation and jobs indicators the committee is looking at. The Inflation Tracker prices the CPI report itself.
If you use Claude or ChatGPT, you can ask it directly. Our MCP server gives any AI assistant the same live numbers, free, no key required. Type "what are the odds the Fed hikes in September" and it will pull the board.
And if you run a site or a newsletter, the live board is yours to embed. One line, no signup, updates itself.
<iframe src="https://predictionmarketspicks.com/embed/fed-rate-tracker/article?theme=indigo" width="100%" height="400" style="border:0;max-width:728px" loading="lazy" title="Live Fed rate decision odds"></iframe>We will grade all three markets on the evening of Sept. 16 and publish who was closest. Then we start the clock on October.
All prices as of Sept. 2, 2026, from PredictionMarketsPicks' 30-minute capture of Kalshi, Polymarket and CME fed funds futures. Polymarket prices are from the international book. Prediction market contracts involve risk of loss. This is not financial advice.