What Warsh Killed, and What the Market Built in Its Place

For a month, this letter has run one argument: that Kevin Warsh's first act as Chair would not be a rate decision but a demolition. That the thing he would kill first was forward guidance — the decade-old machinery by which the Fed tells the market, in advance and in detail, what it intends to do. We called it "What He Kills First." Wednesday, in a forty-two-minute press conference, he killed it.

He did it almost gently. The funds rate held, unanimously — no drama there. But Warsh declined to submit his own economic projections to the quarterly release. He scaled the policy statement back to something closer to a telegram than a memo. And across the forty-two minutes, he said remarkably little about where rates go next, in either direction. A reporter looking for a steer left without one. This was not a man hedging. This was a man removing, on purpose, the instrument his four predecessors spent fifteen years building.

The thesis was right. That is the easy part of this note, and the least interesting. What matters now is the part the thesis only gestured at: once you remove the Fed's guidance, something has to set the market's reaction function. On Wednesday we learned what that something is. It is the market itself. And the market's first act, handed the keys, was to reprice the front end of the curve with a violence the funds decision alone could never have produced.

Warsh said little. His committee said plenty. He may have withheld his own dot, but the Summary of Economic Projections still went out — and it came back meaningfully more hawkish than March.

The median projection for the funds rate moved from 3.4% in March to 3.8% in June. That is not a rounding artifact; that is the center of gravity of the Committee shifting toward higher for longer, and possibly higher from here. Underneath the median, the distribution is starker than the headline: nine participants now see at least one hike before year-end, eight see no change at all, and exactly one sees a cut. Read that again. The entire dovish wing of this Committee has been reduced to a single dot.

This is the central irony of Warsh's debut, and it is worth sitting with. He stripped out his own guidance precisely so the Fed would stop projecting false precision about a path it cannot know. But the vehicle he left running — the SEP, the dots, the collective forecast — is itself a guidance instrument, and on Wednesday it spoke louder than any sentence he could have delivered from the podium. He muted the conductor and left the orchestra playing. The market did not hear a careful, non-committal Chair. It heard nine hawks.

The repricing was immediate and it was surgical. The policy-sensitive two-year yield leaped to roughly 4.20% — its highest in more than two years. Fed funds futures, which a week ago carried a comfortable lean toward eventual easing, flipped: a September move is now firmly on the table and a hike is fully priced by October. The probability of a cut at the next meeting, on July 29, has been marked down to essentially zero.

That is the front end doing exactly what the front end is built to do — pricing the near-term path of policy off the freshest signal it has, which on Wednesday was nine hawkish dots and a Chair who refused to talk anyone out of them.

But here is where a desk separates from a headline. The long end did the opposite.

Overlay Wednesday morning's curve on Thursday's and the shape of the move is unmistakable. The front end is up eleven to fifteen basis points across the one- to three-year sector. The belly is up modestly. The ten-year is essentially pinned, sitting in the 4.45–4.50% zone it has occupied for weeks — a middling level, neither here nor there. And the thirty-year fell, easing back to the high-4.80s/low-4.90s, its lowest in over a month. The curve did not shift up. It pivoted, roughly around the ten-year, and flattened hard. 2s10s came in from about 37 basis points to the mid-20s; 5s30s compressed by a similar dozen. This is a textbook bear-flattener, and it is the single most important thing that happened on Wednesday.

Why does it matter? Because a parallel selloff would have told you the market simply got more hawkish across the board. A bear-flattener tells you something far more specific and far more interesting about what the market now believes.

Isolate the change by tenor and the logic resolves. The yield response declines monotonically as you move out the curve — plus fifteen at the one-year, fading to zero at the ten-year, and turning negative at the thirty. Two different markets pricing two different things off the same press conference.

The front end priced the obvious: more restraint, sooner. Nine dots, a hot economy, sticky inflation — the near-term path is higher, so the short rates that track it went higher.

The long end priced something subtler, and this is the read I want subscribers to carry out of this note. The thirty-year is the market's purest statement about long-run inflation. It rallied. It rallied while the front end was selling off on hike fears. The only coherent way to hold both of those facts at once is this: a Fed that credibly commits to "delivering price stability" — and that removes the dovish guidance safety net it used to extend — is a Fed the long end can trust not to let inflation run. So the inflation risk premium embedded in the long bond came out, even as the front end braced for tighter policy.

Put plainly: the market read Warsh's silence-plus-price-stability-vow as more pain now, less inflation later. The bear-flattener is not a contradiction. It is the most articulate thing the Treasury market has said all year. Restraint at the front, credibility at the back. Warsh said almost nothing, and the curve translated the nothing into a coherent regime.

That is an extraordinary outcome for a Chair who claims to dislike sending signals. He sent the loudest one available — I will not flinch on inflation — and he sent it by refusing to say anything else.

And yet. There is a second Warsh in the transcript, and the market is not pricing him at all.

For all the price-stability iron, Warsh could not fully suppress his own lean — and his lean is dovish. He went out of his way to stress that the inflation problem is supply-driven: the energy shock out of the Iran conflict, not an overheating domestic economy. He stated flatly that he does not see a "cruel choice" between inflation and the labor market — code for we can hold growth without crushing it. He was openly optimistic that AI will lift productivity, the same argument he has used in the past to make the case for cuts. He noted that current rates are restraining real sectors, housing in particular. And he announced a task force to re-examine inflation measurement from "first principles" — which matters more than it sounds, because Warsh has long preferred trimmed-mean inflation to core PCE, and a Fed that changes its inflation yardstick can change its conclusions without changing its rhetoric.

So we have a genuine split screen. The committee — the dots, the SEP, the nine hawks — is tightening-biased. The Chair — supply-side framing, no cruel choice, AI optimism, housing restraint, a measurement review — reads dovish. The market, handed both, sided with the committee and sold the front end. It priced the dots, not the man.

That gap is the trade for the rest of the year. If Warsh's worldview is right — that this is an energy-driven, supply-side inflation that fades as the war premium bleeds out — then the front-end repricing is an overshoot, and the nine hawkish dots are forecasting a hike cycle that the data will not justify. If the committee is right — that inflation is broad, sticky, and five years above target — then Warsh's dovish lean is a liability and his price-stability vow will be tested in public. Both cannot be true. The bond market has placed its chips on the committee. I am not sure it has read the Chair closely enough.

Here is what makes this regime genuinely two-sided rather than merely hawkish. The macro tape underneath the FOMC is pulling hard in both directions at once.

On the disinflationary side: oil has collapsed. WTI is back to roughly $77 and Brent to the high $70s, down sharply on the week as the US–Iran interim peace deal gets signed and the war premium drains out of the barrel. Energy was the proximate cause of the inflation scare Warsh kept pointing at; energy is now deflating in real time. The single biggest argument for the hawkish dots is softening underneath them as they print.

On the inflationary side: the domestic economy refuses to cooperate with the doves. Headline CPI is still running 4.2% year-on-year, core 2.9%, and retail sales came in genuinely hot — a 6.9% annual pace that does not describe an economy in need of relief. Growth is tracking 2.7%. This is not a backdrop that begs for cuts. It is a backdrop that, on the data alone, justifies a Committee leaning toward holds and hikes.

Sticky domestic demand, collapsing energy, a Chair who blames supply, a committee that fears breadth. Into that — a maximally ambiguous, high-variance macro picture — Warsh chose to remove the Fed's guidance. The timing is either brave or reckless, and we will not know which for two or three data cycles. What we do know is that with the Fed's voice turned down, every CPI print, every retail sales number, every barrel of crude now lands on the curve with more force than it used to. Volatility is no longer something the Fed will pre-empt with a steer. It is something the market will discover the hard way, release by release.

The critics lined up fast on Wednesday, and they are not wrong to. The most useful frame came from those who have watched this movie before: it's 1995 again — a Fed that makes the market parse its reaction function from scraps rather than handing it over. Three real risks follow.

The first is loss of narrative control. By saying little, Warsh leaves more in the hands of the market — and the market can run somewhere he does not want it to go, forcing him to either break his own rule and talk, or live with a repricing he never intended. He has traded flexibility for grip, and grip is not a thing you miss until you need it.

The second is the vacuum. Warsh may prefer silence; his colleagues do not. Twelve regional Fed presidents and the Board governors speak constantly, and the most vocal cohort right now is the hawkish one. A quiet Chair does not produce a quiet Fed. It produces a Fed narrated by whoever is loudest — and at the moment, that is the nine dots, not the man at the top.

The third is shelf life. The price-stability pledge works only as long as the data lets it. If inflation turns higher from here, words stop being enough; at some point the market needs to see action behind them, or it stops believing the vow. A Chair who has staked his credibility on delivering price stability while refusing to say how has given himself a very small margin for the data to misbehave.

Duration — 5. Eased. The long-end rally is the tell: the market is buying the price-stability vow, and inflation risk premium is leaking out of the back of the curve. Lower than last week.

Curve — 8. The action. A double-digit bear-flattening pivot in a single session, 2s10s through the mid-20s. This is where the regime is being written, and it is the most fragile point on the board.

Supply — low / not the story. The week's coupon supply is behind us and the 5-year TIPS reopening clears today into a market entirely focused elsewhere. Not a source of pressure this week.

Credit — 3. Still asleep. IG broad OAS around 75bp, the 1–3Y bucket inside 50, high yield well-behaved. Credit has not so much as blinked at the front-end repricing — which is either admirable composure or the same complacency we keep flagging. The low number remains the divergence.

Composite — 6.5. The binary event resolved, which takes one kind of uncertainty off the table. But it has been replaced by a structurally higher-variance regime: a quiet Fed, a loud macro tape, and a market that now sets its own reaction function release by release. The number holds; the character of the risk has changed.

Warsh did exactly what this letter said he would: he killed forward guidance first. But the lesson of Wednesday is not that the thesis was right. It is what rushed in to fill the space. Strip out the Fed's voice and the market does not go quiet — it gets louder, more reactive, and more willing to price a regime off nine dots and a four-word vow. The front end now prices hikes. The long end prices credibility. And the gap between Warsh-the-committee and Warsh-the-man sits unpriced, waiting for the data to settle the argument.

No committee, however unified, can force a market to see the world its way for long. From here, the numbers decide — not the rhetoric, because there barely is any. That was the point. We will find out this summer whether it was the brave call or the reckless one. Either way, the era of being told is over. The era of having to read it yourself has begun.

Published by Rich Petruzzo, CFA. Informational and educational only — not investment advice or a solicitation, and not individualized. Views are the author's own as of the publication date and not those of CFA Institute. Data believed reliable but not guaranteed; past performance is no guide to future results. CFA® and Chartered Financial Analyst® are registered trademarks owned by CFA Institute.