The hold resolved exactly as priced. One dissent, not two. The dot plot median held. And underneath that unchanged headline number, the committee quietly moved hawkish in a way the tape is still digesting.
Hold. 3.50% to 3.75% maintained. One dissent from Miran, who wanted a 25bps cut. Waller, who broke from Powell in January, came back to the majority. First cut still pointing to Q4, with June hold at 64% on Polymarket and April hold at 94% on Kalshi.
Going into Wednesday, Kalshi had two or more dissents at 62%. That was the number to watch, not the rate decision itself. The rate decision was noise. The dissent count was signal.
One dissent came through. Miran, as expected at 98%. Waller did not follow.
That matters. Two dissents tells the market the committee is fracturing under the oil shock and pushing for relief. One dissent tells it Miran is an outlier and the rest of the board is reading inflation data the same way Powell is. The full vote had Powell, Williams, Barr, Bowman, Cook, Hammack, Jefferson, Kashkari, Logan, Paulson, and Waller all holding. Miran alone broke.
Ahead of the meeting, expectations ran as high as two to three dovish dissents. The committee closed ranks instead. Waller, priced at 74% to dissent, read the same inflation data as the majority and voted accordingly. That is a more hawkish outcome than the market had positioned for going in.
The headline number is 3.4%. Unchanged from December. If you stop there, nothing happened Wednesday.
Do not stop there.
14 of 18 members now see one cut or zero cuts for 2026. In December, only 7 members were in that range. Only 5 members still see more than one cut this year. The projection range runs from 2.50% at the low to 3.75% at the high.
The median held because the doves and hawks roughly offset. But the committee is concentrating around zero and one cut in a way it was not in December. That distribution shift is the actual dot plot story, and it will take a few days for the full-year cut count markets to price it properly.
The Fed also revised its economic projections. GDP now at 2.4% for 2026 and PCE inflation at 2.7%, both up from December. Median path: one more cut in 2027, long-run rate near 3.1%.
The three-way split from March 15 is still a three-way split. The dot plot did not resolve the uncertainty. It confirmed it.
One cut still leads. Zero cuts is within single digits. The market went into Wednesday with no consensus and came out the same way, which tells you the dot plot did exactly what it was supposed to do: confirm uncertainty rather than resolve it.
June hold leads at 64% on Polymarket and April hold sits at 94% on Kalshi. Pause-Pause into Q3, then maybe one cut in Q4. That is the base case the market has been building toward since January.
Stocks sold off into the close as Powell's press conference drew attention to how persistent the inflation risk is. His line on the war: too soon to know. That is not the language of a chair about to pivot.
The dot plot was little changed versus December. No change to the broader trajectory. Powell's last meeting is April 29. Warsh takes over in May with a clean slate, no recent cuts to defend, and a committee that is already moving his direction on the distribution.
| Signal | March 15 pricing | Outcome |
|---|---|---|
| Rate decision | 99 to 100% hold | Hold confirmed |
| Miran dissent | 98% | Dissented |
| Waller dissent | 74% | Voted with majority |
| Two or more dissents | 62% | One dissent only |
| Dot plot median | One cut | Held at 3.4% |
| First cut timing | Q4 consensus | Q4 consensus holds |
The rate decision and Miran were fully priced. Waller was the miss. The median held but the distribution underneath it shifted further hawkish than most year-end cut count markets had embedded going in.
The Fed meets Wednesday. The rate decision is already done. Prediction markets price a 100% hold for March 18 on nine-figure volume across Kalshi and Polymarket. There is no trade in the headline rate. The real market is in the dot plot, the dissent count, and the year-end rate path, where the two exchanges disagree by single digits and offer no consensus going into the meeting.
As of March 15, 2026, both Kalshi and Polymarket price a 100% hold for the March 18 FOMC decision. The rate decision is settled. The live trade is the dot plot, the dissent count, and the year-end rate path. One cut leads on both exchanges at 27-31%, with zero cuts and two cuts within 4 points. Kalshi prices two dissents at 62%. Stephen Miran at 98% and Christopher Waller at 74%. Polymarket's Fed Chair market has Kevin Warsh at 95%. Powell's chairmanship ends May 15.
The March hold is noise. What matters Wednesday is the dot plot. That determines whether the market anchors to one cut, zero cuts, or two cuts for the rest of 2026. Right now the tape has no consensus.
Both exchanges show a near-identical three-way split. One cut leads narrowly. Zero cuts sits within 2 points. Two cuts is not far behind. That kind of distribution means no one is confident, and the dot plot is the resolution mechanism.
When two independent exchanges with different user bases and infrastructure produce the same three-way split, that is genuine uncertainty. Not noise. The dot plot on Wednesday is the single most important input for resolving it. If the median dot moves down, one or two cuts become the anchor. If it holds, zero cuts becomes the floor.
Kalshi's Fed decisions sequence market, which prices the outcome path for March through June, tells a cleaner story than the full-year count. The market prices two more holds after March before the committee even considers easing. Pause-Pause-Pause at 59% is not ambiguity. It is conviction.
Goldman Sachs moved their first cut forecast from June to September last week. The contracts had that move priced before the sell-side note went out. Polymarket's rate cut ladder puts December at 78%, making it the highest cumulative cut probability of any single meeting this year. The market is saying: do not expect easing before Q4. And even then, one cut is the most likely outcome.
Prediction markets price a 90% combined probability that at least one governor dissents on Wednesday. That is not normal. A two-dissent hold would be an unusually strong signal of internal committee division going into an already uncertain year-end rate path.
Kalshi's Stephen Miran dissent market at 98% is essentially resolved before the meeting starts. The real information is in the Waller number. At 74%, Polymarket is saying there is a three-in-four chance Waller breaks from Powell. A two-dissent hold carries more information than a unanimous one. It tells you the committee is fractured, and the dot plot will show it.
That succession dynamic shapes everything Powell says at the Wednesday press conference. If he signals dovish concern about growth, the market reads it as a lame-duck hedge. If he stays hawkish on inflation, the market reads continuity into the Warsh transition. Every word gets filtered through the fact that the next Chair is already priced.
The January picture was a clean 2-3 cuts, zero dissent concern, and a soft landing consensus. That picture is gone. Here is where the tape moved.
| Metric | January 2026 | March 15 | Direction |
|---|---|---|---|
| Rate cuts priced (full-year) | 2-3 cuts consensus | Three-way split: 0, 1, 2 | Hawkish shift |
| Dissent probability | Near zero | 90% at least one | ▲ Up sharply |
| Recession odds | 20-25% | 28-31% | ▲ Up |
| First cut timing | June (GS forecast) | September (GS revised) | Pushed back |
| Inflation outlook | Cooling toward 2% | Re-acceleration risk (87c surge) | ▲ Up |
Every row moved in the same direction: hawkish. That is not five independent shifts. It is one macro regime change, driven by the oil shock and the inflation re-acceleration it triggered, propagating through every Fed-adjacent market simultaneously. The dot plot on Wednesday either confirms that regime or pushes back against it.
Everything that matters on Wednesday happens outside the rate line. Three things to watch.