Warsh Fed Hikes to 3.75%-4% and Puts Trimmed-Mean Inflation at the Center
The Federal Reserve lifted its benchmark interest rate to a 3.75%-4.00% range, its first increase under Chairman Kevin Warsh. New projections released alongside the decision put the policy rate in the 4.00%-4.25% range by the end of 2026 and in 2027. Reuters
The September increase had been set up in advance as the first hike of Warsh's chairmanship. Warsh dislikes giving guidance about the likely path of U.S. interest rates. Reuters That reluctance left the statement, the dot plot and his press conference to carry the weight of the forward signal.
At Jackson Hole on Aug. 28, Warsh recommitted to the Federal Reserve's 2% PCE inflation target and said elevated prices should be the central bank's main focus. CNBC In the same remarks he said the Fed wants to gauge whether underlying inflation is "rising, falling, or stuck in place." Federal Reserve Wall Street ended lower after Warsh reaffirmed the focus on fighting inflation. Reuters
The measurement question is now central to how that stance gets operationalized. Core PCE inflation is the Federal Reserve's long-time preferred inflation metric. Warsh has described core PCE inflation as a "rough swag." Core consumer prices exclude food and energy. WSJ
Warsh prefers trimmed mean and median inflation rates as inflation measures. Brookings Trimmed-mean estimators strip out the largest price moves in either tail of the distribution each month, while median PCE tracks the center observation. Both approaches are designed to filter transitory outliers without permanently excluding entire categories, in contrast to an ex-food-and-energy definition.
That preference predates the chairmanship. Warsh has said that inflation is a "choice." He created five task forces to examine Fed functions. CNBC
The broader context here is a potential shift in the reaction function, not just a 25 basis point move. If the Committee gives greater weight to trimmed-mean and median PCE, month-to-month policy judgments will depend less on whether food and energy are excluded and more on the breadth of price increases across the basket. A narrow but persistent overshoot in the middle of the distribution can keep underlying inflation sticky even when headline prints cool. Conversely, a few large outlier components can no longer drive the narrative on their own.
In my view, that makes communication the binding constraint. A chair who dislikes rate guidance but holds a hawkish inflation view forces markets to infer the trigger points from alternative inflation estimators rather than from a dot path. The September dots point higher, to 4.00%-4.25% by year-end 2026 and holding there in 2027. The dispersion around those medians, and which inflation series Warsh cites to justify them, will determine how restrictive that path actually is. For front-end pricing, the distinction matters. Core PCE, trimmed mean and median PCE can diverge for months. Watching which series the chair emphasizes meeting to meeting will matter as much as the rate decision itself.
