Core PCE ticked up. The dollar didn’t.
April Core PCE printed 3.3% YoY against 3.2% prior. DXY closed lower, the 10-year held its rally. A move in the wrong direction on a hawkish print is the cleanest read on how positioning was sitting going in.
DXY 99.11 · EUR/USD 1.1644 · US10Y 4.47% · Brent $94.17 · XAU $4504
April Core PCE printed 3.3 percent year-on-year this morning, against 3.2 percent in March. On a clean hawkish print, the dollar would normally bid and the long end of the curve would sell off. Neither happened. DXY closed at 99.11, below the 99 handle we broke yesterday. The 10-year held its rally at 4.47 percent. A market that fails to move on a hawkish print in the direction the print should push it is telling you something about positioning, not about inflation.
The print, and how to read the wedge
Core PCE at 3.3 percent is firmly above the Fed's 2 percent target, but the composition matters more than the headline. The contribution breakdown showed services ex-housing slowing modestly while medical services and one-off categories did the lifting. Median PCE (Cleveland) and trimmed-mean PCE (Dallas) were both softer than the core read. If Warsh's preference for the trimmed-mean shows up in the Committee's framing, this print reads dovish, not hawkish.
That's the framing read. The price read is even cleaner: positioning going into the print mattered more than the print itself. The options-implied move into release was wider than realised; the risk-reversal skew had been favouring USD calls; the dealer community was net long dollars into the data. A hawkish-looking headline that fails to deliver above the expected tail triggers a rotation, not a continuation.
What rates and FX did, in order
- 10:00 ET, release. Core 3.3 vs 3.2 prior. DXY spiked roughly 25 basis points on the headline, then unwound within 20 minutes as the trimmed-mean component and the median were parsed.
- 11:30 ET. 10-year yield was back to its pre-print level. The term premium compression continued; the expected-rate component was unchanged.
- EU close. DXY at 98.99, broadly flat from the open and below the prior session's close. EUR/USD held above 1.165. The long-euro positioning we flagged a week ago got its validation, although still at the same risk-asymmetric level.
- US close. DXY at 99.11, the second consecutive sub-99 close. The first time since mid-April. The correlation regime we flagged is fully back: gold up to $4504 on the day, Brent down to $94.17.
Reading it through the smile
Through the dollar-smile lens, today reinforces the middle-of-the-curve regime. A hawkish PCE that fails to bid the dollar is the signature of a currency sitting in the trough where the rate-differential pull is weak and the safe-haven pull is dormant. The right-tail rate-differential channel needs a wider US-vs-G10 real-yield gap than today's print delivered; the left-tail safe-haven channel needs an actual risk-off event, which gold's bid hints at but does not yet confirm.
What the chain looks like now
Six pieces in the dollar thread, each one anchored to a real catalyst. The path through them, in order: the original call (22 May), the three-sessions-on update (25 May), the Warsh + Michigan piece (26 May), the Conference Board reaction (27 May), the break of 99 (28 May), and today's PCE confirmation.
What to watch next week
- JOLTS (Tuesday 2 June). The cleanest leading read on labour-market tightness. A quits rate below 2.0% would be the activity-channel cue.
- ISM Services. Headline matters less than the new-orders sub-index; we covered the diagnostic in the original dollar piece.
- ADP and nonfarm payrolls. NFP on Friday is the operational catalyst for the week.
- Warsh's first scheduled remarks. The first time he speaks at length in the new role.
- The DXY-Brent rolling 30-day correlation. If it crosses back below zero, the regime anomaly we covered in the correlation piece is fully resolved.