The thesis, in one sentence

I expect EUR/USD higher and USD/JPY lower through 30 September 2026 — but the compression in the US–Germany 2-year spread (from ~+147bp toward ~+115bp) now comes from the foreign legs rising, not the Fed cutting: Kevin Warsh's Fed is pinned to a hold by hot inflation and bond vigilantes, while the ECB and BoJ hike into the same oil shock. The dollar bulls are too Fed-centric.

// Market snapshot — 22 May 2026
EUR/USD
1.1608
near six-week range mid
USD/JPY
~159
wide carry priced in
DXY
~99
near six-week high
Brent
$104.52
off $114.44 spike; +~50% pre-war
US 2y
4.13%
Fed priced to hold
US–DE 2y spread
+147bp
compressing; German leg rising
Sources: Trading Economics · Google Finance · 22 May 2026

Kevin Warsh was sworn in as the 17th chair of the Federal Reserve at the White House today, 22 May 2026 — confirmed by the Senate 54–45 on 13 May, with President Trump openly "seeking interest rate cuts." The dollar greeted him firm: DXY sits near 99 (21 May 2026), within a hair of a six-week high, because the market reads Warsh as a credibility upgrade and the Strait of Hormuz — running at roughly 5% of pre-conflict shipping since the late-February air war on Iran — as a one-way safe-haven bid. Both reads are half-right, and together complacent. The dollar is being propped by an oil shock it does not control and a Fed chair the market has not yet figured out; underneath, the relative policy path is turning against it.

01The regime and the framework

This is cross-border policy divergence — the dollar driven by the relative trajectory of the Fed, ECB and BoJ — with an energy-driven reflation as the cross-cutting overlay. One shock, a Hormuz oil blockade with Brent at $104.52/bbl (22 May 2026, off a $114.44 spike earlier this month but still roughly 50% above pre-war), hits three central banks at three different points in their cycles. The transmission runs: same shock → divergent policy responses → 2-year differential compression → FX → equities.

The ECB held its deposit rate at 2.00% on 30 April but openly debated a hike, with euro-area inflation at 3.0%; the €STR strip prices roughly +50bp of hikes into year-end, the first as soon as its June meeting, to 2.25%. The BoJ held at 0.75% — a 30-year high — but three of nine board members dissented for 1.00%, and it lifted its FY2026 core-inflation outlook to 2.8% from 1.9%. Both are leaning against the shock: earlier in normalisation and more exposed to imported energy inflation, they read it as something to fight. The Fed is the constrained one — and that is where the new chair changes the picture.

02The Warsh Fed: pinned now, spring-loaded later

Warsh inherits a divided committee and a bond market daring him to blink. Wall Street prices a 97.2% hold at his first meeting on 16–17 June; the live debate is not whether he cuts but whether a faction pushes to hike — April CPI at 3.8% is the hottest in three years, and FOMC minutes show members "open to a hike." His signature is unusual: a "rate cuts plus active QT" mix — shrinking the $6.7tn balance sheet through asset sales while eventually lowering the policy rate, a return to what he calls a traditional monetary standard. He wraps it in independence language — the Fed must "stay in its lane" — while the administration that staged his swearing-in at the White House, the first there since the 1980s, makes no secret of wanting cuts.

// Warsh — two faces the market is pricing as one
The face the market is pricing — pinned Cutting into 3.8% CPI and a live oil shock, at his first meeting, with hawks on the committee and Powell still seated as a governor for two more years, would torch the credibility Warsh needs and hand the bond vigilantes a reason to drive the 10-year — already at a one-year high of 4.595% on 15 May — toward 5%. So he holds. Dollar keeps its prop. Correct, for June.
The face it is under-pricing — spring-loaded Warsh's mandate is to ease — that is why this administration chose him — and his QT is the cover: a smaller balance sheet is the hawkish offset that lets him argue a lower policy rate is not stimulative. The moment energy inflation rolls over, his Fed eases faster than Powell's would have. By end-September, that asymmetry outweighs the June hold.

The dollar is being propped by an oil shock it does not control and a Fed chair the market has not yet figured out.

Brighthedge — Macro thesis · May 2026

The market's error is extrapolating the June hold across the horizon. The reaction function is asymmetric: pinned now, predisposed to cut into the second half, because the political and intellectual incentives point the same way. For a divergence trade that resolves by end-September, that matters more than one meeting.

There is a darker version, and I will not bury it. If Warsh eases under visible political pressure while inflation is still hot, the bond market will not read it as benign convergence — it will read it as debasement. The long end sells off, the term premium blows out, and the dollar falls not because the gap closed cleanly but because confidence cracked. That outcome would make my FX call right for the wrong reason: EUR/USD up, but alongside a 10s30s bear-steepening, an equity drawdown and a gold bid — the opposite cross-asset picture to my base case. The instrument that tells the two apart is the US 10-year; I flag the level in the falsifier below.

One more thing the consensus is glossing: the market's relief at Warsh — the dollar and short-dated Treasuries rallied when Trump tapped him, because he is not a pure dove — is a real, dollar-supportive counter to my thesis. My answer is that the relief is about avoiding a worse dove, not an endorsement of a hawkish path; the actual mandate, and the committee arithmetic once energy peaks, still bend toward easing.

03The numbers the equity market isn't pricing

US equities are priced for the benign path only. The S&P 500 closed near 7,446 (21 May 2026) at 21.1x forward earnings — above its 5-year (19.9x) and 10-year (18.9x) averages — discounting +22.1% calendar-2026 EPS growth. That is a thin cushion against a hold-or-hike Fed, a 10-year at 4.56% (22 May), an oil tax on margins, active QT draining liquidity, and a new chair the bond market has not yet tested. The macro split that pins Warsh is visible in the prints: core PCE 3.2% (March) and headline CPI 3.8% (April) on one side; payrolls a soft +115k and unemployment 4.3% on the other, with Q1 GDP a middling +2.0% annualised. The US–Germany 2-year spread sits near +147bp (US 2y 4.13%, 22 May; German Schatz 2.66%, 11 May) — and the German leg is the one with room to rise.

04Cross-asset positioning — one frame

Cross-asset view — conviction matrix · 22 May 2026
Leg View Already priced Cleanest expression Conviction
Rates Near-term move is the German/euro front-end rising on ECB hikes; US front-end rally is an H2 story tied to Warsh's easing asymmetry. June hold 97.2%, hike tail rising; first US cut ~Sep. ECB hiking leg and H2 pivot under-priced. Long German front-end vs US; US 2s5s steepener for H2 cut. Medium
FX — EUR EUR/USD higher, led by the Bund leg, not the Fed. 1.1608; ECB hikes largely priced — edge is Fed's capped upside plus de-escalation optionality. Long EUR/USD — the cleanest divergence proxy. Medium
FX — JPY USD/JPY lower on BoJ hikes and eventual repatriation, fighting Japan's energy-import terms-of-trade headwind. ~159 prices a wide, durable carry. Long JPY via USD/JPY — same trade, most in tension. Low
Equities Neutral-to-cautious S&P at 21.1x; weak dollar is an EPS tailwind but won't offset high real yields, QT and a Warsh credibility test. 21.1x fwd P/E, +22.1% CY26 EPS growth — benign path fully baked in. Relative tilt ex-US vs US, not an outright short. Low
Conviction matrix Brighthedge Research · 22 May 2026

Note the downgrade: a week ago I held the EUR/USD leg at high conviction. The deeper read of the Hormuz blockade's severity — a structural euro-negative the rate channel only partly offsets — and a Fed now priced to hold-or-hike pull it to Medium. The direction stands; the timing is less clean.

Where the legs agree: the dollar's marginal driver is rotating away from the Fed and toward the foreign central banks and the oil price. Where they fight: on Japan (the terms-of-trade headwind), on equities (weak-dollar tailwind vs QT and real-yield headwind), and within rates — Warsh's QT is dollar-supportive at the margin even as his rate path is dollar-negative. The expression that survives the most cross-currents is long EUR/USD, and even that is a Medium call now, not the High it was a week ago.

05Three paths

Base — pinned, then easing 50%
EUR/USD 1.19–1.22
DXY 95–96  ·  USD/JPY 152–156  ·  S&P range-bound

Warsh holds in June and builds credibility; the ECB hikes and the BoJ inches to 1.00%; Hormuz de-escalates gradually into summer, energy rolls over, and Warsh signals a September cut. The 2y differential compresses from both sides.

Acceleration — clean de-escalation 15%
EUR/USD 1.24+
DXY ~92  ·  USD/JPY sub-150  ·  ex-US relief rally

A US–Iran deal reopens the strait, Brent toward $80. The safe-haven prop vanishes, Warsh cuts in September with cover, the ECB/BoJ sit on hold after hiking.

Break — two flavours 35%
EUR/USD <1.13
falsifier trips if escalation scenario plays out

Escalation / hawkish (~25%): Hormuz worsens, Brent >$130, stagflation; the dollar's safe-haven and energy-exporter status dominate; Warsh is forced to hike in July; the ECB/BoJ can't deliver hikes as growth craters. EUR/USD <1.13 — falsifier trips; USD/JPY >162; DXY >103.

Debasement (~10%): Warsh eases under pressure with inflation hot; the long end revolts, 10s30s bear-steepens, term premium blows out. EUR/USD up but via risk-off — equities down, gold bid — the wrong-reason version of being right.

Asymmetry: about 75% of the distribution has EUR/USD higher, but ~10 of those points are the disorderly debasement version, and the 25% escalation tail is the largest single move. Positive skew on the FX call, fat left tail on the dollar — size for the tail, not against it.

06The falsifier

This thesis is wrong if…

EUR/USD trades below 1.1300 at any point before 30 September 2026.

  • Observable — a spot FX level, no interpretation.
  • Specific — a number (1.1300) and a date (30 Sep 2026).
  • Not already true — EUR/USD is 1.1608 as of 22 May 2026.
  • Fatal — a euro below 1.1300 means the energy / safe-haven channel has beaten the policy channel; that is the core claim wrong, not one leg.
Yellow flags — drift, not death
  • The ECB holds at its June 2026 meeting instead of hiking to 2.25%.
  • Two consecutive US payroll prints above +175k.
  • Brent sustained above $115/bbl for two weeks.
  • Warsh flag: the Fed hikes in July, or the US 10-year closes above 5.00%. If 5%+ coincides with EUR/USD rising, the thesis is right on FX for the wrong reason — the debasement tail — and the rates and equity legs are wrong.

Correction trigger: if EUR/USD breaks 1.1300, Brighthedge runs a correction as its own piece — the oil channel won.

AAppendix — Sourced figures

Every figure cited inline — sourced and dated per Brighthedge standard
Metric Value As of Source
Warsh confirmation Senate 54–45; 17th Fed chair; sworn in at White House conf. 13 May / sworn 22 May 2026 CNBC / CNN / NPR
Warsh first FOMC 16–17 June 2026; 97.2% hold priced; hike tail rising ~22 May 2026 Fortune / Motley Fool
Warsh policy mix "Rate cuts + active QT"; shrink $6.7tn balance sheet; Fed must "stay in its lane" Apr–May 2026 Fortune / Invesco / CNBC
Powell status Stays on as governor (~2 yrs remaining term) — continuity, constrains Warsh May 2026 CNBC / NPR
Initial dollar reaction Dollar + short-dated USTs rallied on "not a pure dove" relief when Trump tapped Warsh Jan 2026 Bloomberg
Fed funds target 3.50–3.75% (held; easing bias; Miran dissent for cut) 29 Apr 2026 Federal Reserve FOMC
US 2y 4.13% 22 May 2026 Trading Economics
US 10y ~4.56% (spiked +14bp to 4.595% on 15 May; 1-yr high) 22 May 2026 CNBC / Trading Economics
US 2s10s ~+43bp (positive, narrow) 22 May 2026 Derived (10y − 2y)
ECB deposit rate 2.00% (held; hike debated) 30 Apr 2026 ECB
ECB year-end pricing ~+50bp hikes, first ~June to 2.25% ~May 2026 €STR forward / CNBC, Euronews
German 2y Schatz 2.66% 11 May 2026 Trading Economics
German 10y Bund <3.10% 21 May 2026 Trading Economics
US–Germany 2y spread ~+147bp (compressing; German leg rising) 22 / 11 May 2026 Derived
BoJ policy rate 0.75% (30-yr high; 3 of 9 members dissented for 1.00%) Apr 2026 BoJ / CNBC
BoJ FY26 core CPI / growth Core CPI 2.8% (↑ from 1.9%); growth cut to 0.5% Apr 2026 BoJ outlook
Japan 10y JGB ~2.77–2.79% (30y at record) Late May 2026 Trading Economics / CNBC
DXY ~99 (near six-week high) 21 May 2026 CNBC / streetstats
EUR/USD 1.1608 22 May 2026 Trading Economics
USD/JPY ~159 22 May 2026 Google Finance
Brent crude $104.52/bbl (spiked $114.44; ~50% above pre-war, +61% y/y) 22 May 2026 Trading Economics / Al Jazeera
Hormuz shipping ~5% of pre-conflict volume; blocked since 28 Feb air war on Iran May 2026 House of Commons Library / Wikipedia
S&P 500 ~7,446 (record 7 May) 21 May 2026 Trading Economics
S&P 500 fwd P/E 21.1x (5y avg 19.9x; 10y avg 18.9x) ~8 May 2026 FactSet
S&P 500 CY26 EPS growth +22.1% ~1 May 2026 FactSet
US core PCE 3.2% y/y (+0.3% m/m) Mar 2026 BEA
US headline CPI 3.8% y/y (hottest in 3 years, energy-led) Apr 2026 BLS
US payrolls / unemployment +115k / 4.3% Apr 2026 BLS
US Q1 GDP +2.0% annualised (advance) Q1 2026 (rel. 30 Apr) BEA
Counter — terms of trade Hormuz blockade structural; EZ/Japan large energy importers → EUR/JPY-negative May 2026 Macro identity / HoC Library
Counter — hikes may not arrive ECB cut growth outlook; BoJ cut FY26 growth to 0.5% — energy could stay their hand Apr 2026 ECB / BoJ
Counter — Warsh = dollar+ Fed priced to hold-or-hike; some members floating a hike; bond vigilantes constrain cuts May 2026 Fortune / Motley Fool
Source log — tan rows: Warsh data · red-tint rows: counter-thesis Brighthedge convention: every inline figure sourced here