α
AlphaSteve
← Papers
Geopolitics

Geopolitics & macro: Is the market mispricing Warsh's reaction function on the energy shock?

2026-06-09 · long-form

Executive summary

The May jobs report flipped the rate debate. Markets now price roughly a 68–70% chance of a Federal Reserve hike by year-end T3. May CPI lands Wednesday and consensus has headline at 4.2% year-over-year, the hottest since April 2023, driven by the Iran energy shock T2. On the surface, a hike looks like the obvious next move.

The question this report answers: is the market reading the new chair's reaction function correctly, or is it extrapolating hot labor data into a tightening cycle that Kevin Warsh's own stated framework argues against? Warsh prefers trimmed-mean inflation, a measure that drops the extreme price moves — exactly the oil spike — from the read. By that measure he has said inflation sits much closer to 2% than the headline shows T3.

Our answer: the market is pricing the labor heat correctly and the chair's framework incorrectly. The higher-for-longer base case is intact — no cuts in 2026. But the modal outcome under Warsh is an extended hold, not a hike. The hike that markets price at ~70% is a supply-shock response from a chair who has told everyone he looks through supply shocks. Five-year breakeven inflation sits at 2.53%, expectations are anchored, and the same energy shock is now destroying demand: China's crude imports just fell to their lowest since October 2016 T1. Hiking into that is the policy error a trimmed-mean chair is built to avoid.

House view reconciliation

The current house view, after the 2026-06-05 PM update, reads: "hike scenario promoted from a non-trivial tail to a co-equal base case in market pricing; 'one cut at best' no longer describes consensus." The 2026-06-06 AM note added the probability-source caveat — Bloomberg's 85% and CME's 57%+ are different framings, and the precise figure should not be over-anchored — plus Goldman's counter-voice that hikes remain "unlikely."

This report confirms the higher-for-longer base case and the demand-destruction shape the 2026-06-06 PM note flagged. It does not confirm the market's hike-pricing as the kit's own forecast. The house view to date has described market pricing without separating it from the kit's view of the chair's actual reaction function. That gap is what this report closes.

The update: we register an explicit variant. Market hike-pricing at ~70% overshoots Warsh's stated framework. The kit's modal forecast under Warsh is an extended hold through 2026 — neither cut nor hike — with the hike a real but thinner tail than the tape implies. The discriminating event is the June 16–17 dot plot and Warsh's first press conference. Nothing here softens the "no cuts in 2026" call; it sharpens what sits on the other side of the hold.

The setup

Three things happened in the eight days since the last Tuesday long-form, and they pull in opposite directions.

The labor side ran hot. May payrolls came in at +172k against a Dow Jones consensus near 80k, unemployment held at 4.3%, and average hourly earnings rose 3.4% year-over-year — the strongest three-month hiring run in more than two years T1. It followed April JOLTS openings of 7.6 million, the highest since May 2024, and an ISM manufacturing print of 54.0, the highest since May 2022 T1. The producer side of the economy is accelerating.

The price side ran hot too, but for one reason. May CPI is expected at 4.2% headline year-over-year and 0.5% month-over-month, while core is expected at 2.9% year-over-year and 0.3% month-over-month T2. The 1.3-point gap between headline and core is the Iran energy shock. Strip energy and the inflation problem is far smaller than the headline number advertises.

The demand side, meanwhile, is cracking. China's seaborne crude imports fell to 6.36 million barrels per day in May from 8.10 million in April — the weakest since October 2016 — as the Hormuz disruption cut supplier access T3. University of Michigan consumer sentiment sits at a record low 44.8 T3. The energy shock that is lifting the CPI is simultaneously taking real activity down. That is the signature of a supply shock, not a demand-driven inflation.

Markets resolved this tension in one direction: ~68–70% odds of a hike by December, the 10-year Treasury at 4.56%, and the October meeting near a coin flip T3. The bet is that hot labor plus 4%-handle CPI forces the Fed's hand. That bet rests on an assumption about how Warsh reads inflation. The assumption is wrong.

The analysis

What Warsh actually said he would do

Warsh was sworn in on 2026-05-22 and chairs his first meeting June 16–17 T3. His stated method is not the headline CPI the market is reacting to. He prefers trimmed-mean inflation, which drops the largest price moves in both directions — this spring's oil spike included — to find whether prices are broadly rising. By the trimmed-mean read, Warsh has said inflation is much closer to the 2% goal than headline suggests T3.

This is not a fringe view inside the building. Governor Michelle Bowman has argued the Fed should not overreact to a temporary energy spike. The administration's economic adviser Kevin Hassett has backed Warsh's focus and played down the inflation risk from the oil shock T3. Dallas Fed President Lorie Logan pushes the other way and distrusts the trimmed-mean read T3. There is a real split. But the chair sets the reaction function, and this chair has told the market he subtracts the oil shock before he acts.

The market is pricing a hike off a number the chair has said he discounts. That is the core mispricing.

The supply-shock doctrine the market is ignoring

Standard central-bank doctrine is to look through a relative-price supply shock and respond only if it threatens to unanchor expectations or feed broad second-round inflation T2. An oil spike from a Hormuz blockade is the textbook case. It raises the price level, it is a tax on activity, and it fades when the supply constraint clears — unless workers and firms build it into wage and price-setting.

The test, then, is whether expectations are unanchoring. They are not. Five-year breakeven inflation sits at 2.53% T1, below the 2022 peak of 3.59% and consistent with anchored expectations even as the headline pushes toward 4.2%. The bond market is doing exactly what the doctrine predicts: it is letting the near-term print run hot while keeping medium-term expectations near target. A chair who looks through supply shocks, facing anchored expectations, does not hike. He holds and waits for the energy base effect to roll off.

The wage read supports the hold rather than the hike. Average hourly earnings at 3.4% year-over-year are firm but not accelerating into a wage-price spiral T1. Goldman's economists make this the off-ramp: subdued wages mean the hot hiring raises the odds of a longer pause, not a hike T3. The second-round channel that would justify tightening is not lit.

Demand destruction does the Fed's work

The strongest argument against a hike is that the shock is already self-correcting on the demand side. China's import collapse to a near-decade low is the clearest evidence T3. Brent is down roughly 20% from its 2026 peak even with strikes ongoing, and on June 8 WTI traded near $91 and Brent near $95 as the conflict cooled and Trump pushed both sides to de-escalate T3. OPEC+ added another 188k barrels per day to July quotas despite the supply risk T3. A supply premium that is destroying demand and meeting added supply is a premium with a half-life. Hiking to fight it would compound the demand hit the shock is already delivering — tightening into a slowdown the energy tax is creating.

This is the stagflation shape the 2026-06-06 PM note named: supply-driven prices up, real activity down. The correct policy response to that shape is not symmetric. A supply premium passes into headline CPI faster than the demand weakness shows up in growth, so the inflation print leads and the activity damage lags. A chair who hikes off the leading print risks being maximally restrictive exactly as the lagging activity damage arrives. Warsh's trimmed-mean framing is, in effect, a device to avoid that timing trap.

Why the ECB is not the read-across the market thinks

The market leans on the ECB as confirmation. Frankfurt is near-certain to hike 25 basis points to 2.25% on June 11, with at least one more priced by year-end, on eurozone inflation running above target T3. If the ECB hikes into the same energy shock, why wouldn't the Fed?

Because the two are not in the same position. The ECB held in April in a decision several members called close, and would have hiked then if it had been proposed — it is catching up to a tightening it deferred T1. Its policy rate at 2.00% going into June is far below the Fed's restrictive stance; a move to 2.25% is normalization from an accommodative base, not incremental tightening into a slowdown. The euro area also imports a larger share of its energy and runs a weaker currency channel, so the pass-through to broad inflation is more direct. The Fed is already restrictive, the dollar is firm, and the chair discounts the oil component. Same shock, different starting point, different correct response. Reading ECB hawkishness straight into Fed pricing is the error.

Variant perception

Consensus, as priced, is that the Fed hikes once by year-end at roughly 70% odds, with the June dot plot expected to validate the turn. The load-bearing assumption underneath that price is that Warsh treats headline CPI at a 4% handle as an action trigger.

Our variant: that assumption is the weakest link in the chain. Warsh has publicly committed to a trimmed-mean read that subtracts the oil shock, expectations are anchored at 2.53% on five-year breakevens, wages are firm but not spiraling, and the same shock is destroying demand fast enough to mean-revert the oil price. The modal outcome is an extended hold through 2026 — the kit's existing "no cuts" call, with the symmetric truth that there are probably no hikes either. The market has priced the hot labor data, which is real, into a price-side response the chair's framework rejects.

What supports the variant: the breakeven anchor, the China demand collapse, the wage read, and the chair's own words. What would falsify it: five-year breakevens breaking above roughly 2.9–3.0%, which would signal expectations unanchoring and force even a trimmed-mean chair to act AS-cal; a core CPI surprise that shows the inflation broadening beyond energy into services and shelter; or average hourly earnings re-accelerating through 4%. Wednesday's core print is the first clean test — core at or below the 2.9% consensus keeps the look-through case intact; a core surprise to 3.1%-plus is the first real crack in the variant.

The asymmetry favors the hold. If we are right, the market unwinds ~70% hike-pricing toward a hold and the front end rallies. If we are wrong, it is because core inflation broadened or expectations slipped — both of which the Wednesday and the following CPI prints will show before the December meeting, leaving time to adjust. The variant is falsifiable on a known schedule.

Implications for AlphaSteve

The top-down implication is that the front end of the curve is mispriced for a hike the chair's framework argues against, and the kit should lean toward the hold rather than chase the tape's hawkish turn. This does not change the "no cuts in 2026" base case — it defines the other boundary of it. The higher-for-longer regime is real; the incremental hike priced on top of it is the soft part. The discriminating event is eight days out: the June 16–17 dot plot and Warsh's first press conference, read against Wednesday's core CPI.

  • Portfolio: No position change triggered. The read argues against adding duration-sensitive risk that depends on cuts and against fading rate-sensitive names on hike fear that may not materialize.
  • Watchlist: Names whose theses depend on the front-end path — rate-sensitive financials, long-duration software — should be screened against the hold case, not the market's hike case.
  • Theses on the workbench: Any thesis carrying an implicit "Fed hikes in 2026" assumption should be re-checked against the extended-hold modal call.
  • Sectors: No sector-view shift this run. Energy demand destruction (China imports) is worth tracking as a leading tell on whether the oil premium mean-reverts.
  • House view updates: Update the Geopolitics & macro "US rate path" section to register the kit's variant — market hike-pricing at ~70% overshoots Warsh's trimmed-mean reaction function; modal kit forecast is extended hold, not hike.
  • Daily-scan adjustments: Add five-year breakeven (FRED T5YIE) crossing ~2.9% and core CPI surprise above consensus as the two falsifiers to watch on the scan.

Charts / data

Table 1 — Supply-shock vs demand-shock scorecard, as of 2026-06-09

Signal Reading Points to
Headline CPI (May consensus) 4.2% y/y T2 Hike
Core CPI (May consensus) 2.9% y/y T2 Hold / look-through
5y breakeven inflation 2.53% T1 Hold — expectations anchored
Avg hourly earnings 3.4% y/y T1 Hold — no wage spiral
China crude imports 6.36M bpd, low since Oct 2016 T3 Hold — demand destroying
Brent vs 2026 peak ~−20% T3 Hold — premium mean-reverting
Payrolls / JOLTS / ISM +172k / 7.6M / 54.0 T1 Hike — producer side hot

The labor and headline rows drive the market's hike-pricing. The core, breakeven, wage, and demand rows drive the kit's hold call. Warsh's trimmed-mean framework weights the second group.

Table 2 — What markets price vs what the chair's framework implies

Market pricing Warsh framework implies
Year-end 2026 hike odds ~68–70% T3 Lower; extended hold modal
Inflation gauge that matters Headline CPI (4.2%) Trimmed mean (near 2%) T3
10Y Treasury 4.56% T3 Rich if hold prevails

Sources

House view changes this run

  • Geopolitics & macro → US rate path: Register the kit's variant. Market hike-pricing at ~68–70% by year-end overshoots Warsh's stated trimmed-mean reaction function, which discounts the energy shock driving headline CPI. Kit's modal forecast under Warsh is an extended hold through 2026 — neither cut nor hike — with the hike a real but thinner tail than the tape prices. The "no cuts in 2026" base case is unchanged; this defines the upper boundary of the hold. Rationale: anchored 5y breakevens (2.53%), firm-not-spiraling wages (3.4%), China demand destruction (6.36M bpd), and the chair's own framework.
  • Daily-scan: Add two falsifiers — 5y breakeven (FRED T5YIE) above ~2.9%, and a core CPI surprise above consensus — as the triggers that would convert the variant from hold to hike.
  • No other house-view section changed this run.

Linked