The Fed's decision lands Wednesday at 18:00 GMT and the rate itself is a near-certain hold at 3.50% to 3.75%. The event is everything around it: Kevin Warsh's first meeting as Chair, a Summary of Economic Projections that may erase 2026's last cut, a dot plot he may decline to sign, and a press conference that doubles as the opening statement of a new communication regime.
For two years the only question at an FOMC meeting was how fast disinflation would let the Federal Reserve cut. That question is gone. After an energy shock pushed inflation back up the hill and May CPI landed hot at 4.2%, the Fed arrives at its June meeting with the cut debate replaced by a hold-or-hike debate. Today it will hold. The rate is not the story.
The story is everything stapled to the decision. This is Kevin Warsh's first meeting as Chair, the meeting carries a fresh Summary of Economic Projections and the dot plot that goes with it, and it ends with a press conference that is really the opening night of a new way of running the Fed. This is the full institutional read: what actually gets decided at 18:00 GMT, why the dot plot is the event and not the rate, the chair's own dot that may be missing, the words in the statement that move the tape, what to listen for in the press conference, and the cross-asset map for the dollar, gold, rates and risk.
The statement and the projections drop at 18:00 GMT (2:00 PM ET), with Chair Warsh's press conference at 18:30 GMT. Markets price the rate decision itself as a near-certain hold in the 3.50% to 3.75% range, so the volatility will come from the dots and the tone, not the headline. Into the print gold sits near $4,333, the dollar holds most of its 2026 gains with EUR/USD pinned around 1.1600, the US 2-year yield is at about 4.05% and the 10-year near 4.43%, a curve already leaning to a Fed that is not in a hurry to ease. The energy scare that drove the inflation reacceleration has cooled with the Gulf de-escalation, which pushes the remaining inflation risk onto the domestic core, exactly the part of the basket Warsh has said he wants to measure differently.
The mechanical decision is the easy part. The Fed has held the target range at 3.50% to 3.75% through the spring, and for today CME FedWatch has priced no change at roughly 99%. There is no serious case for a move in either direction at this meeting: the inflation data are too hot to cut into and the labour market is too firm to force the Fed's hand toward easing, while an actual hike with no prior signalling would be a communication shock no central bank delivers casually. So the funds rate is settled before the meeting begins.
Because the rate is a foregone conclusion, the meeting is decided on three other documents: the statement and how its language shifts, the Summary of Economic Projections with the dot plot, and the press conference. The market will trade all three in sequence, and the three do not always agree. That is the setup for a classic FOMC whipsaw, where the first move is faded and the real trend only forms once the statement, the dots and the chair's tone have been reconciled.
Disinflation was real until an energy shock broke it. Inflation had cooled toward the low 3s, then conflict in the Gulf hit the oil market and the energy index dragged the headline back up: CPI ran 3.3% in March, 3.8% in April, and 4.2% in May. The Fed's preferred family of gauges tells the same story from a calmer angle, with headline PCE around 3.8% and core PCE near 3.3% year on year, both still well north of the 2% target. Producer prices have been hotter still. Whatever ruler you use, inflation is not where a central bank starts cutting.
The other half of the reaction function is the labour market, and it has refused to crack. Payrolls have kept printing in positive territory (the May report came in around +172K with unemployment near 4.3%), comfortably above the level that would pressure the Fed to ease to protect jobs. A central bank cuts when growth is the risk; right now inflation is the risk and growth is fine. So the market has done the logical thing and repriced the path: cuts have been pushed into 2027, and at the hawkish extreme traders have at times priced a real chance of a hike before this cycle is done. That is the backdrop the dots have to ratify or reject today.
Four times a year the Fed publishes its Summary of Economic Projections: each official's forecast for growth, unemployment, inflation and, most watched of all, the appropriate path of the funds rate. Those rate forecasts are plotted as the dot plot, one dot per participant for the end of this year, next year, the year after and the long run. June is one of those meetings. The March projections still pencilled in at least one cut for 2026. The job of today's dots is to tell the market whether that cut survives.
The base case across the street is that it does not. The median 2026 dot is expected to shift up to show no cuts this year, formalising in the Fed's own hand the hawkish repricing the market has already done. That sounds technical, but it is the single most market-moving number in the release, because it converts a vague "higher for longer" into a dated, official path. The questions that decide the reaction are how far the median moves, how tightly the dots cluster, and whether any participant goes all the way to penciling in a hike.
Here is the twist that makes this dot plot unlike any other. Wall Street widely expects the new Chair, Kevin Warsh, not to submit his own dot at all. He has been openly sceptical of the dot plot for years, arguing that it masquerades as a forecast while functioning as a commitment, boxes the committee into forward guidance it later regrets, and gives markets false precision. Having taken office only in late May, he also has cover to sit this first projection round out. A dot plot missing the chair's dot is a strange object: the median still prints, but its signalling power is diluted because the most important participant has declined to anchor it.
The dot plot only works as guidance if the market believes it reflects the chair's intentions. If Warsh withholds his dot, a hawkish median becomes harder to take at face value, and the burden of guidance shifts almost entirely onto the press conference and the statement. In other words, a quirk of process turns the chair's spoken word into the dominant signal of the day, which is exactly the regime Warsh appears to want.
We read the 2026 median against three thresholds, because the gap between them is the whole trade:
Two second-order signals matter as much as the median. The first is dispersion: a tight cluster says the committee is united, a wide scatter says the path is contested and any single data point can swing it. The second is the vote and any dissents: a unanimous hawkish hold is far stronger than a hold with one or two members dissenting toward cuts, which would expose a dovish bloc waiting for an excuse to ease.
To trade this meeting you have to understand what its chair is trying to build. Warsh did not arrive to tweak the funds rate by a quarter point. He arrived with a view that the Fed's communication regime itself is the problem: too much forward guidance, too much false precision, a dot plot that commits the committee to paths it cannot honour, and an inflation framework he considers too forgiving. We covered the inflation-measurement half of this in the May CPI preview: his stated preference to judge inflation by trimmed-mean and median gauges rather than core PCE. Today is where the communication half goes live.
Expect the early signature to be subtraction rather than addition: less pre-commitment, a leaner statement over time, fewer promises about the path, and more emphasis on optionality and data dependence. Withholding his dot would be the first concrete move in that direction. The risk he is balancing is credibility. He takes the chair under visible political pressure for lower rates, in the middle of an oil-driven inflation scare, which means every dovish-sounding word will be scrutinised for whether the Fed's independence is intact. The cleanest way for Warsh to establish authority on day one is to sound willing to hike if inflation broadens, even while holding, because that is the stance a market cannot accuse of bending to politics.
The communication overhaul has a twin, and it matters just as much for how today's projections read. Warsh has been blunt that he does not fully trust the way the inflation number is built, dismissing the Fed's preferred core PCE gauge as only a rough approximation even after food and energy are stripped out. His stated preference is to lean on trimmed-mean and median measures instead. Because June is a projections meeting, the inflation line in the Summary of Economic Projections is the first place that preference can surface, so it is worth understanding exactly what these gauges are and how they are built.
In any given month the prices in the consumer basket do not all move together. A handful of categories swing violently (energy, airfares, used cars, hotel rooms) while most move modestly. If you plot the cross-section of monthly price changes, component by component, you do not get a tidy bell curve. You get a distribution that is fat-tailed and skewed, usually to the upside, with a long right tail of a few categories spiking. The headline inflation rate is simply the expenditure-weighted mean of that distribution, and a mean is exquisitely sensitive to the tails: one large outlier with a non-trivial weight can drag the whole number around even though it says nothing about the broad, persistent pace of inflation underneath.
That is the statistical problem trimmed estimators are built to solve. The insight, formalised for inflation by Michael Bryan and Stephen Cecchetti in the 1990s, is that when a distribution has fat tails the sample mean is a noisy, inefficient estimate of its centre. A trimmed mean, which discards the most extreme observations before averaging, or the median, the exact middle, is a far more stable estimate of the central tendency, which is the thing a central bank actually cares about. Ex-food-and-energy core is a crude special case of the same idea: it always throws out the same two categories by rule. A trimmed mean throws out whatever is extreme this month, wherever it happens to sit in the basket.
The calculation is a weighted percentile trim. Step by step, for a given month:
Written compactly, with Iα,β the set of components left after the trim:
Two special cases fall straight out of this. Set α = β = 0 and you trim nothing, so you are back to the ordinary headline mean. Push the trim all the way in from both sides until only the component at the 50th percentile of weight survives and you have the median:
Take a deliberately simplified basket of five components. The real indices have hundreds, but the mechanics are identical. The weights sum to 100% and the price changes are annualised one-month rates. Note the energy spike. (Illustrative figures, not the actual May data.)
| Component | Weight | Annualised 1-month change |
|---|---|---|
| Core goods | 20% | +1.0% |
| Food | 14% | +2.8% |
| Shelter | 33% | +3.6% |
| Core services ex-housing | 25% | +4.2% |
| Energy | 8% | +60.0% |
The headline number is the weighted mean of the whole column: 0.08·60 + 0.33·3.6 + 0.14·2.8 + 0.20·1.0 + 0.25·4.2 = 7.6% annualised. Energy alone contributes 4.8 of those 7.6 points. One category with an 8% weight has hijacked the entire print.
Now sort by price change and accumulate the weight: core goods (cumulative 20%), food (34%), shelter (67%), core services (92%), energy (100%). The median is the category where cumulative weight first crosses 50%, which is shelter at 3.6%. For a 16% trimmed mean you cut 8% of weight off each end: the bottom 8% comes out of core goods, the top 8% removes all of energy. Averaging what is left and re-normalising the weights gives ≈3.3%:
Same month, same data, three answers: 7.6% on the headline, 3.6% on the median, 3.3% trimmed. The energy outlier that drove the headline simply never reaches the trimmed gauges.
This is not theoretical; two Federal Reserve banks publish exactly these series every month. The Dallas Fed Trimmed Mean PCE applies the procedure to the PCE basket with an asymmetric trim, cutting roughly 24% of the weight from the lower tail and about 31% from the upper tail. The asymmetry is deliberate: the cross-section of PCE price changes is right-skewed, so a symmetric trim would leave a systematic upward bias, and the cut points were chosen to track the underlying trend with the least noise. The Cleveland Fed publishes the Median CPI (the 50th-percentile case above) and a 16% Trimmed-Mean CPI (8% off each tail). These are the gauges Warsh has pointed to as a cleaner read than core PCE.
The payoff is statistical. Because it ignores the violent tails, a trimmed or median gauge has much lower month-to-month volatility than the headline, it is harder to fool with a single energy spike or a one-off used-car jump, and historically it has been a better predictor of where inflation settles over the following year. That is exactly why it appeals to a chair who wants to look through an energy shock without being accused of cherry-picking which categories to ignore: the trim is a rule applied to the whole distribution, not a hand-picked exclusion.
But the gauge is not a dove in disguise, and this is the nuance the desk watches. Trimming only cools the read when the hot prints are genuinely isolated in the tail. If the strength is broad, shelter firm, core services sticky, goods re-accelerating, then there is no narrow tail to cut: the middle of the distribution is simply higher, and the trimmed mean and median come in hot too. History is the warning. There have been stretches when the trimmed median ran hotter than core PCE, which would have argued for a more hawkish stance, not a softer one. The trim is a filter for noise, not a discount on inflation.
Bring it back to the dots. The force that dragged inflation toward 4% is energy, the textbook tail a trimmed gauge discards, so if the strength is concentrated there the trimmed and median readings sit well below the headline and below core, and they hand a chair who wants a patient hold exactly the cover he needs, both in the press conference and in the inflation line of the SEP. If instead the pressure is broad, his own preferred gauge turns against him and the hawkish case strengthens. So the question that decides the path is not only how high the projected inflation number is, but how wide the underlying pressure is. We walked through the same lens applied to a single CPI print in the May CPI preview.
The statement is short and almost every word is deliberate. The algorithms diff it against the prior version in milliseconds; the desk reads it for the direction of five things:
| What to watch | Hawkish read | Dovish read |
|---|---|---|
| Inflation language | "Remains elevated", emphasis on broadening or upside risks | "Eased", framing the spike as energy-driven and temporary |
| Balance of risks | Risks tilted to inflation; readiness to tighten if needed | Risks "roughly balanced", attention back on employment |
| Policy bias | Any hint of a tightening bias or a higher bar to cut | Language preserving an eventual path lower |
| The vote | Unanimous hold | One or more dissents toward cuts |
| Balance sheet | No change to runoff; QT continues | Any signal that runoff slows or ends |
The most likely outcome is a statement that keeps inflation described as elevated and risks tilted to the upside, with no explicit hiking bias yet, leaving the heavy lifting to the dots and the press conference. A hawkish surprise would be the statement itself introducing a tightening bias; a dovish surprise would be a softening of the inflation language or a visible dissent.
At 18:30 GMT the new Chair takes questions for the first time, and if he has withheld his dot, this is where the actual guidance lives. The market will be listening less for new numbers than for posture. Does he sound more worried about inflation or about growth? Does he leave a hike genuinely on the table, or treat the next move as still more likely down once the energy spike passes? How does he handle the dot plot itself, does he defend it, downplay it, or signal he wants to change it? And how does he field the inevitable question about political pressure for cuts without either picking a fight or appearing to capitulate?
First press conferences set a reputation, and Warsh knows it. The base case is a deliberately measured, optionality-heavy performance: firm on the 2% goal, unwilling to declare victory, careful not to pre-commit. The hawkish tail is that he explicitly entertains a hike if inflation broadens. The dovish tail is that he leans on the energy-driven, temporary framing and sounds in no rush, which the market would read as the cut path quietly surviving behind a hawkish set of dots. Expect the statement reaction and the press-conference reaction to differ, and expect the second one to set the tone into the close.
A press conference is not a speech, it is cross-examination, and the market trades the answers in real time. These are the questions Chair Warsh is almost certain to get on Wednesday, and the tell in each direction. The skill is not predicting the question, it is knowing in advance which way each answer pushes the dollar, gold and the front end.
| The question | Hawkish tell | Dovish tell |
|---|---|---|
| Is the next move more likely a cut or a hike? | Keeps a hike genuinely on the table, calls it "data-dependent in both directions" | Frames the next move as still down once the energy spike passes |
| Is the inflation pickup just energy, or is it broadening? | Stresses breadth: shelter, core services, signs it is spreading | Calls it narrow and energy-driven, a relative price shock to look through |
| Why withhold your dot? Do you still back the dot plot? | Signals he wants to downgrade or reform the dots and forward guidance | Defends the projections as useful, business as usual |
| How do you respond to political pressure to cut? | Stresses independence, refuses to ease into above-target inflation | Any phrasing that sounds accommodative to the pressure |
| What would it take to hike? | Names conditions: broad core, unanchored expectations | Deflects, sets a high bar, emphasises lags |
| Is policy restrictive enough today? | Hints it may not be | Says policy is sufficiently restrictive and patient |
Watch the clustering of answers more than any single line. A chair who repeatedly leans hawkish across unrelated questions is sending a consistent signal; one who softens on two or three in a row is quietly opening the door, whatever the dots said twenty minutes earlier.
A new chair is an unknown quantity, and the first meeting is where the market calibrates the new communicator. When Jerome Powell took over in 2018, his plainer, less scripted style took several meetings for the market to read, and more than one early press conference produced outsized intraday swings purely on tone rather than substance. The lesson for Wednesday is specific: Warsh is both new and openly trying to change how the Fed talks, so the risk of a translation error between what he means and what the tape hears is unusually high. That argues for wider ranges than a routine meeting, a bigger gap between the statement reaction and the press-conference reaction, and more reversal risk through the European close. The first instinct of the market is often wrong at a regime-change meeting; the second move, once the style is decoded, is usually the one that holds.
Warsh inherits the chair amid open political pressure for lower rates, and that subtext sits under every word. The bind is real: a chair who sounds willing to cut into a 4% inflation print invites the charge that the Fed is being steered, while a chair who holds firm and keeps a hike on the table buys credibility but risks a public clash. For markets the read is direct. Any sign of capitulation to pressure would tend to weaken the dollar and steepen the long end as an inflation-risk premium is rebuilt, a bear steepener, even as front-end cut odds rise. A credible, visibly independent hawkish hold does the opposite: it supports the dollar and keeps the long end anchored because the market trusts the 2% anchor. Warsh's most likely path is to defend the Fed's independence explicitly, and that in itself is a mildly hawkish signal, because it says the inflation fight comes before the political calendar.
The whole regime rests on one inversion that traders have to internalise. With the bar to cut now set by inflation rather than growth, strong economic data are hawkish, not reassuring. A firm labour market does not bring cuts closer; it removes the Fed's reason to ease and keeps the hike option alive. That is why every upside inflation or activity surprise lands directly on hike odds and the dollar, while downside surprises produce relief rallies that tend to fade, because one soft print does not undo an oil shock or change a committee whose red line is inflation. Into a projections meeting, this asymmetry is amplified: the dots can ratify the hawkish path in a way a single data point cannot.
With the rate fixed, the scenarios are built around the dots and the chair's tone rather than the decision. We frame three.
| Scenario | What lands | Dollar (DXY / EUR/USD) | Gold | Rates & risk |
|---|---|---|---|---|
| Hawkish | Dots show no 2026 cuts and a hike on the horizon, or Warsh leaves a hike clearly open | Dollar firms, EUR/USD breaks below 1.1600; 2027 first-cut pricing pared back | Pressured toward 4,260 then 4,220 as real yields rise and the safe-haven bid keeps fading | 2Y yield up, curve flatter; equities and crypto sell, long-duration tech hit hardest |
| Base | Median moves to zero 2026 cuts, statement steady, Warsh measured and non-committal (dot possibly withheld) | Dollar holds firm, EUR/USD heavy but defends 1.1600 | Chops around 4,300 to 4,360; the trimmed-inflation framing caps the hawkish follow-through | Front end little changed; risk drifts, the 2027 cut path intact |
| Dovish | Dots keep a 2026 cut, or Warsh leans on the energy-is-temporary framing and sounds in no rush | Dollar dips, EUR/USD pops back above 1.1700 | Relief rally toward the 200-day near 4,433 | Yields ease, curve bull-steepens; equities and bitcoin bid |
The greenback goes into the meeting holding most of its 2026 gains, with EUR/USD pinned around 1.1600 after the inflation-and-energy story did the heavy lifting earlier in the quarter. In an inflation-led regime the dollar's reaction is skewed: a hawkish set of dots or a hike-open press conference extends the trend, while a dovish read produces a dip that has repeatedly faded. The level that matters is 1.1600 on EUR/USD; a clean break and hold below opens further dollar upside, while a defended 1.1600 on a dovish FOMC is the first warning for dollar bulls.
Gold is the cleanest two-way trade and it sits near $4,333 into the decision, below the 200-day moving average around 4,433 that now acts as resistance rather than support. Two forces meet here. The headwind is a hawkish Fed, higher real yields and a firm dollar, all of which weigh on a non-yielding asset. The tailwind is gold's role as an inflation and debasement hedge plus a residual geopolitical bid. A hawkish FOMC presses gold toward 4,260 and then 4,220; a dovish read lets it reclaim the 200-day and squeeze toward 4,433. The dots decide which side of that line gold closes the week.
The front end is the purest expression of the meeting. The 2-year yield near 4.05% is the cleanest hike-odds gauge; watch it more than any single word. The 10-year near 4.43% and the 30-year near 4.93% carry more growth and term-premium information. The 2s10s slope is already positive and modestly steep, which tells you the market expects the Fed to hold rather than ease imminently. A hawkish FOMC lifts the 2-year and flattens the curve as near-term cuts are priced out; a dovish FOMC bull-steepens it as the front end rallies on a surviving cut path.
Equities trade the textbook rate reflex, with long-duration growth and technology the most sensitive, so a hawkish FOMC is a headwind for the Nasdaq in particular and a tailwind for nothing. Crypto trades as a high-beta liquidity proxy: it leans risk-off into a hawkish surprise and rallies on a dovish relief, while keeping a partial debasement bid that can blunt the downside. Both sit on the back foot into the meeting after the hawkish repricing, which means the bar for a relief rally on any dovish wobble is low, but so is the floor if the dots confirm the hike camp.
One line in the statement that rarely makes the headline can matter for liquidity: the pace of quantitative tightening. The base case is no change, with runoff continuing in the background. But Warsh has long-standing views on the size and role of the Fed's balance sheet, and any hint that runoff slows, ends, or is up for review would be read as a quiet easing of financial conditions even with the funds rate on hold. It is a low-probability surprise at this meeting, but it is the kind of detail that separates a desk read from a headline read, and it is worth checking before reacting to the rate line alone.
An FOMC decision is a top-tier volatility event, and we treat it as a sequence of structured inputs rather than a single number: the statement diffed against the prior one, the 2026 and 2027 median dots against March, the dispersion of the dots, the vote, and then the press conference weighed against all of it. Expect a whipsaw. The first move trades the statement and the dot headline and frequently reverses once Warsh starts speaking, so the edge is usually in the second move, after the market has reconciled the page with the podium. Into the print we watch the 2-year yield and EUR/USD 1.1600 as the fastest tells, and we let the dollar and gold confirm the direction once the press conference is done.
A near-certain hold, a dot plot that is expected to delete 2026's last cut, a chair who may not sign it, and a first press conference that is really the start of a new communication regime: that is the meeting. The funds rate is settled; the path, the tone and the framework are not. The dollar holds the cards with EUR/USD on 1.1600, gold leans on its inflation-hedge floor while it fights the 200-day, the 2-year yield is the scoreboard, and the genuine game-changer is whether Warsh formalises a hawkish path in the dots or quietly keeps the cut alive in his words. Trade the dots, listen to the chair, and wait for the second move.
Figures are prior official releases, consensus and analyst estimates and market levels (sources include the Federal Reserve, the BLS and BEA, CME FedWatch, Trading Economics, and reporting on Chair Warsh's public remarks and the June meeting from CNBC, Yahoo Finance, Reuters, the Conference Board and Kiplinger) as of publication on 17 June 2026, and are subject to revision. Market levels are drawn from TTerminal's own data as of the latest close into the meeting. This article is TTerminal's own market analysis and is not investment advice.
The macro calendar, the FOMC cross-asset reaction, the dollar, gold, yields and the risk map, all live across forex, indices, commodities and crypto.
Start 7-day free trial ↗