May CPI printed 4.2% headline with core at 2.9% YoY — a meaningful divergence from core PCE at 3.4% that actually complicates the Fed's reaction function rather than simplifying it. The FOMC's June dot plot shift (median end-2026 rate now 3.8%, up from 3.4% in March, with 9 of 18 members penciling in at least one hike) confirms the policy bias has flipped hawkish. TLT at $85.52 is down 1.04% today and the path of least resistance remains lower into July 14.
The May CPI data landed at 4.2% headline YoY — the highest reading since April 2023 — driven almost entirely by energy, which surged 23.5% annually and 3.9% month-over-month. That's the easy part to dismiss. The harder part: core CPI came in at 2.9% YoY with a monthly gain of just 0.2%, actually undershooting the 0.3% consensus. Shelter at 3.4% annually is sticky but not accelerating. Core commodities declined 0.1%. On the surface, this looks like a benign core print.
But here's the tension that's been building since my last post: core PCE is at 3.4% YoY while core CPI prints 2.9%. That 50 basis point spread between the two core measures is historically wide and demands explanation. Part of it is methodological — PCE gives more weight to healthcare services and less to shelter than CPI does. Part of it may be that PCE is capturing services inflation that CPI is underweighting. Either way, the Fed's preferred gauge is 50bps hotter than what CPI is suggesting, and Warsh is staring at that discrepancy heading into the July meeting.
The June FOMC outcome confirmed what the PCE data was already telegraphing: the easing bias is dead. The dot plot median for end-2026 moved from 3.4% to 3.8%, the policy statement was stripped of any forward guidance leaning dovish, and Warsh conspicuously declined to submit his own dot — a signal that the chair himself doesn't want to be boxed in. The Fed held at 3.5%-3.75%, but the institutional posture has shifted from 'patient' to 'watching for the next hike trigger.' With the funds rate at 3.5%-3.75% and core PCE at 3.4%, real policy rates on the Fed's preferred measure are essentially zero — that is not a restrictive stance by any historical standard.
For duration, the setup remains unfavorable. TLT closed today at $85.52, down 1.04% on the session, and is up only 0.49% YTD despite brief rallies on softer-than-expected prints. The YTD gain is misleading — it reflects a volatile path, not a clean trend. SCHP at $26.24, up 1.16% YTD and 3.29% over the past 52 weeks, is marginally outperforming TLT on a risk-adjusted basis, which makes sense: TIPS are getting partial credit for realized inflation while nominal Treasuries face both duration risk and the possibility that real yields rise further if the Fed tightens. The TIPS outperformance isn't large enough to call it a conviction trade, but the directional logic holds.
The critical variable before September is the June CPI release on July 14. If core CPI converges toward the 3.2-3.5% range where core PCE is sitting, that removes the last structural argument for holding rates steady. If core CPI stays sub-3.0%, it gives the doves a lifeline — but J.P. Morgan's call for core goods inflation to run at 2% annualized in 2026 versus 0.9% in prior years suggests the tariff and energy pipeline isn't done feeding through. Energy near $100/barrel, geopolitical supply disruptions still active, and shelter inflation with no clear deceleration catalyst: the base case for core CPI moving higher into summer is stronger than the market is pricing.