The Hold at 3.50-3.75% Sets the New Policy Anchor
Kevin Warsh’s first FOMC meeting delivered a unanimous 12-0 hold at 3.50% to 3.75% and a dot plot that, for the first time since 2012, places the median end-of-2026 rate above the current target range, at 3.8%. That is the new policy anchor, and it tells the market that the bar for dovish surprises has risen while the bar for hawkish surprises has fallen. The market reaction was textbook: two-year yields up 8 basis points to 4.05%, ten-year yields up 6 basis points to 4.32%, the dollar index through 104.20, and the 2s10s curve steepening to plus 27 basis points, the most since March 2023. Seventeen of eighteen officials who submitted dots judged the risks to inflation to be tilted to the upside, and the median 2026 PCE inflation projection jumped to 3.0% from 2.7% in March, the largest single-meeting upward revision since the September 2022 SEP.
Why 0.4% Core CPI on June 24 Is the Line That Decides the Path
Consensus has core CPI at 0.3% month over month and 2.9% year over year, with a 0.4% print triggering a wholesale repricing of the front end. A 0.4% core reading would push the three-month annualized rate to 3.1%, well above the median 2026 PCE projection of 3.0% that the SEP delivered on June 17, and would harden the case for a hike by the September meeting. The Atlanta Fed Wage Growth Tracker ran at 4.7% on the most recent reading, and the May Employment Cost Index, due release at the end of the month, is expected to print 0.8% quarter over quarter, both numbers consistent with services inflation that will not be contained without a restrictive policy stance. A print at or above 0.4% would also break the four-month sequence of disinflationary surprises that the committee was counting on, and the second-derivative effect of a hot core would matter more for the median dot at the September meeting than the headline level.
The Hawkish Statement Rewrite and the Reaction Function It Signals
The committee removed the phrase that had signaled a bias toward future cuts, dropped the language describing the labor market as having moderated, and replaced the customary reference to possible additional adjustments with a clause noting that the committee will consider the extent and timing of any move. In the parlance of the Federal Reserve, that is a hawkish rewrite, and the market is now trading the rewrite rather than the hold. The dot plot is now the policy reaction function, not the rate level, and the reaction function just got hawkish. A 0.4% CPI in a 0.3%-consensus environment is the asymmetric upside surprise the Warsh-era Fed would not absorb quietly, and the risk-reward on the long end of the curve has shifted decisively toward the steepening trades that defined the post-FOMC session.
How the Global Central Bank Map Reads the Print
Three other major central banks are watching the same release for different reasons. The European Central Bank cut its policy rate to 2.50% on June 5, the Bank of England paused at 3.75% in May, and the Bank of Japan held steady at 0.75% while signaling a possible October move. A hot US CPI print would widen the rate gap, lift the dollar, and pressure the yen back through 152 with the Ministry of Finance one bad print from intervention. A soft print would compress the gap, ease the dollar, and revive the cut narrative that the median dot just buried.
The dollar-funded carry trade that defined the first half of 2026 is now the trade most exposed to a hawkish surprise, and dealers are reporting that yen funding costs through one-month cross-currency basis have already widened by 12 basis points since the FOMC decision, the largest move in a single session since the August 2024 unwind. Either way, the trade is the long end of the US Treasury curve, and the position is to fade any move that runs more than 5 basis points in the first hour of trade. Warsh’s inaugural operational review, four internal task forces reporting by September, will set the policy reaction function for the second half of 2026, but the May CPI release on June 24 is the first real test of the response function he just inherited, and the first opportunity for the new chair to put his own voice on the data rather than the data of his predecessor.