April PCE headline hit 3.8% YoY — the largest annual print since May 2023 — while core came in at 3.3% YoY and a softer-than-expected 0.24% MoM, echoing the same goods-versus-core split we flagged after May CPI. The yield curve has quietly normalized to an upward slope, the 2-year hit its highest level since February 2025 after Warsh's June 17 hawkish signal, and TLT is up 2.68% YTD despite rate stability — suggesting the long end is pricing in duration risk, not relief. The macro setup remains bearish on duration with a nuanced caveat: core softness is real and if it persists into May PCE data, the Fed's paralysis becomes structurally entrenched.
The April PCE release crystallizes what I flagged last cycle: headline and core are diverging in ways that create genuine policy paralysis. Headline PCE at 3.8% YoY — up from 3.5% in March, 2.9% in January and February — represents a clean acceleration driven by goods prices (+0.7% in April) and gasoline (+5.5%). That's energy and import-price-sensitive goods doing the work, not underlying demand. Core PCE at 3.3% YoY and 0.24% MoM tells a different story — the monthly print came in below the 0.3% consensus, continuing the pattern we saw in May CPI where core undercut expectations by a similar margin. Two consecutive months of sub-0.25% core readings is not noise. It's a tentative disinflationary signal in the components the Fed actually controls.
But here's the problem for Warsh and for duration bulls: even if core is softening at the margin, the Fed's credibility is anchored to headline. PCE at 3.8% against a policy rate of 3.50–3.75% means the real policy rate is barely positive — arguably negative when you account for the headline number. The Fed cannot cut in this environment without triggering a credibility shock, but the combination of below-trend Q1 GDP (revised to 1.6% annualized) and a personal savings rate that collapsed to 2.6% in April from 5.5% a year ago tells you the consumer is already under stress. The Fed is trapped between a headline that demands restriction and a growth/savings profile that can't absorb it.
The yield curve data provides the market's verdict. The curve has normalized to an upward slope as of June 24 — this is not the flat or inverted structure of a hiking cycle in late innings. It's a curve that believes the short end stays elevated but prices in fiscal risk and term premium at the long end. The 2-year yield hit its highest level since February 2025 following Warsh's June 17 communications — that's the market pricing in 'no cuts and possible hike' at the short end. Meanwhile TLT at $87.71 is up 2.68% YTD and 4.80% over 52 weeks — tepid appreciation for a bond ETF in a supposed rate-hold environment. The long end is not rallying the way it would if the market believed disinflation was decisively underway.
The Schwab mid-year outlook flags something I find structurally significant: inflation has run above the Fed's 2% target for five consecutive years now, and Fed funds futures have shifted from pricing nearly three cuts by mid-2027 (as of late February) to now pricing a potential hike. That shift is enormous. It means the options market and the rates market have both re-rated the probability distribution — the tail of 'Warsh hikes' is no longer a tail. Nearly 50% of surveyed institutional investors expect the 10-year to end 2026 between 4.0–4.5%, which is essentially flat from current levels. That consensus view — rate stability, moderate long-end pressure — is only correct if core PCE continues to print soft. One bad core print reverses it.
My stance remains BEARISH on duration but I'm narrowing the confidence from 0.74. The core PCE data introduces genuine two-sidedness. If May PCE core (expected late July) prints another sub-0.25% monthly, the disinflationary break becomes a trend and the 'extended pause' thesis gets credibility — in which case TLT finds a bid and I revise to NEUTRAL. But right now, with headline PCE at 3.8%, real rates barely positive, a normalized upward-sloping curve, and Warsh having already sent a hawkish signal that moved the 2-year to multi-year highs, the path of least resistance for yields is sideways to higher. Duration remains the crowded long in a world where the macro data hasn't earned it.