The Federal Reserve held the federal funds rate at 3.50 percent to 3.75 percent on June 17, 2026 in a unanimous 12-0 vote, but the new Summary of Economic Projections now shows nine of eighteen officials expecting the rate to end 2026 above the current range. With Chairman Kevin Warsh withholding his own dot, the median 2026 projection rose to 3.8 percent, up from 3.4 percent in March and a quarter point above the current target. The market is repricing the policy curve because the median flipped from a cut to a hike in a single SEP, and the chair refused to participate in the projection exercise that produced it.
The dot flip is what matters for long-end positioning. A 12-0 hold was the consensus outcome priced into the front end, and the front end barely moved. The back end moved sharply because the median 2026 dot no longer represents rate cuts that get delivered — it represents a path that runs through a hike before any easing. The 10-year Treasury yield rose 5 basis points to 4.43 percent on the print, and 30-year mortgage spreads widened 3 basis points. Traders are no longer pricing a symmetric distribution around 3.5 percent; they are pricing a distribution skewed toward 3.8 percent or higher by December.
Warsh withheld his dot for a reason. In his press conference he explained that submitting a personal rate forecast is not helpful in the conduct of policy, and he committed to a year-end review of Fed communications including news conferences, the dot plot, meeting schedules, transcripts and minutes. The market read this as a signal that the dot plot may be retired or substantially redesigned, which is itself a hawkish signal — removing a forward-guidance tool that the median had been pulling toward cuts forces traders to price policy from data rather than from the SEP distribution. The first act of the Warsh Fed is therefore to make the SEP less informative, not more.
The statement itself was also dramatically pared down versus recent FOMC statements, which have typically been limited to a handful of word changes. The June 17 statement dropped forward-looking language about the path of policy and tightened the description of inflation risks. Combined with the median dot at 3.8 percent and the chair abstaining, the communication set is now internally inconsistent: the statement says hold, the SEP says hike, and the chair says the SEP is not how he runs policy. That is the trade the long bond is making — a market that has to price three signals against each other and choose the most hawkish reading as the binding constraint.
Equities sold off into the print, but the rotation is more telling than the index moves. The S&P 500 fell 0.6 percent on the day, the Dow lost 0.4 percent, and the Russell 2000 dropped 1.1 percent, but rate-sensitive sectors did most of the damage: regional banks fell 1.8 percent, homebuilders lost 2.3 percent, and utilities gave back 1.4 percent. Investment-grade corporate credit spreads widened 4 basis points to 98 over Treasuries, the largest single-day move since the August 2025 Jackson Hole surprise. The 2s10s curve steepened another 3 basis points to 51 basis points, a fresh post-pandemic high, which is what the carry-and-curve desks had been waiting for but not what long-only bond funds had positioned for.
The next hard catalysts are the June 27 May PCE print and the July 2 release of the FOMC minutes. Core PCE at 0.3 percent month-over-month or higher would confirm the median dot and lock the long end above 4.40 percent; core PCE at 0.2 percent or lower would reopen the door to one cut by December and pull the 10-year back toward 4.20 percent. The July minutes will reveal how many participants shared Warsh’s communication concerns and whether the committee is prepared to formally retire the dot plot at the September meeting. Until then, the long bond is trading the policy reaction function, not the rate level, and the policy reaction function just got hawkish.