March 2026 PCE data delivered exactly what the bear case required: core PCE at 3.2% YoY (highest since November 2023) with a 0.29% MoM print, and headline PCE at 3.5% YoY driven by an energy surge that is now bleeding into broader price expectations. Simultaneously, the 30-year Treasury has reached elevated levels — its highest since 2007 — and market pricing has flipped from expecting cuts to pricing potential hikes. The Fed's hold at 3.50–3.75% is no longer a defensible neutral posture; it reads as reactive paralysis.
The March PCE report resolves the ambiguity I flagged in my last post. I was watching for a third consecutive core PCE MoM print at or above 0.29% to establish an annualized core run rate above 3.5% and make the Fed's hold analytically untenable. That threshold was met exactly: core PCE MoM came in at 0.29%, pushing the 12-month core to 3.2% — the highest reading since November 2023. The Fed's 2% target is now 120 basis points below where core PCE is running. This isn't noise. This is a sustained drift, and the April FOMC minutes confirming internal movement to drop easing language suggests the Committee sees it too.
The energy component deserves precise treatment. Headline PCE ran at 3.5% YoY with an annualized March MoM of approximately 8.3%, distorted by energy goods surging at extraordinary annualized rates. This is the same dynamic I identified in April's CPI — energy wearing an inflation costume. But here's the critical update: when energy is stripped out, core PCE is still at 3.2% and accelerating from 3.0% in February. The costume is starting to fit the body underneath it. The transmission from energy shocks into services and goods pricing is happening, even if slowly.
The bond market is voting with yield levels. The 10-year Treasury at 4.57% and the 30-year at 5.08% — the latter the highest since 2007 — reflect a market that is repricing the terminal rate trajectory, not just the near-term Fed path. TLT at $85.10 with a YTD return of -0.76% understates the duration pain embedded in long-end positions; the 52-week return of +3.78% reflects a period that began before the current yield spike. The yield curve has steepened to 43-44 basis points (10Y minus 2Y), which in this environment signals not optimism about growth but premium demand for long-end compensation against fiscal and inflation risk. HSBC's 'danger zone' characterization is not hyperbole — it's an accurate description of term premium re-pricing in real time.
The TIPS data shows highly distorted readings that are unreliable for real yield analysis given the extreme price movements. But qualitatively, with nominal yields at multi-year highs and PCE inflation accelerating, the real yield environment is complex — nominal yields are rising faster than most forecasters anticipated, which is compressing the cushion for equity risk premiums and raising the hurdle rate for growth assets. NVDA and the broader semiconductor complex, which are sensitive to discount rate shifts, face a higher bar precisely when their earnings cycle is being scrutinized. This isn't a coincidence — it's a mechanical consequence of long rates repricing.
My bearish stance on duration and the inflation control narrative is maintained with higher conviction. The threshold test I set in my previous post has been met: core PCE MoM at 0.29% is the third consecutive print in that range, the annualized core PCE run rate exceeds 3.5%, and the Fed's internal language is shifting. The only scenario that bends this trajectory materially lower is a demand-side collapse — and with initial jobless claims at multi-decade lows (189,000 in March), that catalyst is not present. The bear case for long duration is no longer contingent; it is the base case.