The Doves Lost Their Bet — Inflation Data Drove Hawkish Conviction and Punished Complacency: A Rate Hike Is on Deck

Growth held up better, −0.49% against −1.10% for value, but the real split was inside growth: the seven largest companies rose 1.42% while the broad growth basket fell. Value still leads by nine points for 2026. In a week when rates repriced sharply higher, the market sold what it could, small companies, cyclicals, health care, and kept what it trusts most. That is the narrowing described below, now at its most extreme of the year.

A word on what the yield curve is, because this chart repays it. The dark line is the gap between what the government pays to borrow for ten years and for two. In a normal cyclical expansion the 10-year yields more than the 2-year, producing an upward-sloping curve and a positive spread. When the 2-year yields more than the 10-year the curve is inverted, a historical recession warning. The gold line is the 2-year yield itself, which tracks what the market expects the Federal Reserve to do. The red shading on the chart marks how every inversion since 1990 was resolved: the 2-year fell, the curve steepened, and a recession followed, because the Fed was cutting into weakness. That was the bond market’s reliable warning through the entire forty-year bull market in bonds. The orange shading marks the other way an inversion can end: the 2-year rises and the curve flattens because the Fed is tightening. That happened in 1977 to 1981, when an energy shock forced a central bank that had paused to start again. A very similar dynamic has emerged in 2026. The lower panel shows this episode in detail. The curve was inverted from mid-2022 to September 2024 and un-inverted, as usual, because the 2-year fell when the Fed began cutting. What was unusual is how little it fell: from 4.96% to a low of 3.38% this February, about 160 basis points over 22 months, against declines of 430 to 570 basis points in every prior modern cycle. It never collapsed. Since February it has risen 125 basis points, and the curve has flattened from the front. Why this matters for a portfolio: the modern playbook says a re-steepening curve is the moment to prepare for recession and buy long bonds. This curve is not giving that signal, and the last time it behaved this way, long bonds were the worst place to be for several years. The 1970s comparison is not a forecast, and that episode did eventually end in a Fed-induced downturn. But it argues for the front of the yield curve over the back, for inflation protection over duration, and against treating this steepening as the recession warning it would have been in any of the last four cycles.

Our scenario assumptions are unchanged this week; they are reviewed monthly, and we expect to move the framework to 2027 earnings estimates in the coming weeks. On a central estimate of roughly 21.5 times about $340 of expected earnings, fair value sits near 7,310. At 7,656.98 the index trades about 4.7% above that center, down from 5.6% a week ago as the index slipped. The comparison on the other side widened sharply against equities: the earnings the index produces relative to its price is about 4.44%, while a risk-free 10-year Treasury now pays 4.96%. Investors are accepting roughly 52 basis points less to own stocks than government bonds, against 37 a week ago, because yields rose 18 basis points while the index fell less than 1%. That is the widest negative gap of the year.

The divergence we flagged last week resolved the way we noted it typically does: in favor of the market-based gauge. The Fear and Greed Index deepened to 33, well into fear territory and its lowest reading since the spring, and the retail survey that had swung bullish a week earlier swung back to net bearish. The VIX rose to 15.84, still low by any historical standard but its highest in six weeks. To boot, the cost of deep crash protection, as measured by the SKEW index, rose for a fifth straight week to 154.49. Cheap volatility with expensive tail protection has been the pattern for over a month; this week the front of that trade finally began to reprice.

Sources: U.S. Department of the Treasury; ICE BofA Indices; CME Group.

All eyes are on one event this week: the Federal Reserve’s policy rate decision Wednesday at 2:00 p.m., including updated economic forecasts, with investors keenly focused on Fed Chair Warsh’s press conference that will immediately follow. A quarter-point increase is 86% priced, so the decision itself should not surprise; the statement, the projections and the Chair’s framing of what follows are what markets will trade. Retail sales arrive the same morning, and Friday brings the quarterly expiration of stock-index options and futures, which tends to amplify whatever direction the week has taken.

Last week this letter described a market that had bet on a cool inflation report. On Friday the report came in warm, with core consumer prices rising 0.3% against a 0.2% forecast; the bet was lost and markets reacted accordingly. The odds of a rate increase at Wednesday’s Federal Reserve meeting rose from roughly 56% to 86%. The 2-year Treasury yield climbed 26 basis points on the week, the 30-year topped 5.35%, its highest since 2007, and the futures market now prices a peak policy rate near 4.56%, almost a full point above where it sits today. Crude oil closed at $99.99, up 20% in two weeks, with the premium for near-term barrels at a record $19. The inflation data drove hawkish conviction and punished complacency, exactly in that order.

Equities fell less than 1%, and that is the least informative number of the week. Small-cap stocks fell 2.4%, health care names 3.6%, and eight of eleven sectors declined. The share of S&P 500 members above their 50-day average fell to 39%, from 70% four weeks ago, the sharpest deterioration in participation of the year. The index was held within a point of its 50-day average by the seven largest growth companies with the heaviest weightings in the index, which rose 1.4%, and by almost nothing else. Fear deepened, the weakest corporate borrowers repriced for a second week, and the cost of deep crash protection rose for a fifth. Three of our sector calls changed, two more went on watch, and technology’s upgrade is explicitly tentative because the volume behind it is thin.

This week’s chart puts the move in the yield curve into fifty years of context, and the point is worth stating plainly for readers who do not follow the curve. Short-term rates rose faster than long-term rates this week, so the curve flattened. In every recession of the modern era, 1990, 2001, 2008 and 2020, the curve did the opposite as the downturn approached: short rates collapsed because the Federal Reserve was cutting into weakness. That signal is absent. The nearer historical rhyme is 1977 to 1981, when an energy shock forced a central bank that had paused to tighten again, and the curve flattened from the front. That episode did eventually end in recession, but it was one the Fed caused rather than one the bond market foresaw, and it took years rather than months. On valuation, the S&P 500 index still trades about 4.7% above our base-case fair value, and the gap between what stocks yield on earnings and what Treasury securities pay in interest is now the widest of the year in bonds’ favor. Recent developments have made this clearly not a time for complacency. These developments are also not yet enough to warrant material changes to portfolio exposure. Wednesday’s FOMC decision, and likely more importantly, what Fed Chair Warsh says at 2:30 p.m. ET that afternoon, could change that.

Portfolio Implications