Kevin Warsh has removed the one cue markets leaned on hardest. The Federal Reserve chair has discontinued forward guidance, the long-standing practice of signaling in advance where the central bank expects interest rates to move, on the argument that the tool is not well suited for the current policy conjuncture [1]. Every economic release now lands directly on traders rather than being softened by weeks of prepared Fed messaging.
Forward guidance was never a promise. It was a way of steering expectations, letting borrowers, employers, and investors price in a likely path before any vote was cast. Warsh's view is that pre-committing to a direction ties the Fed's hands when the data is noisy, and he would rather keep the options open. The effect on everyone downstream is that the planning aid is gone while the uncertainty remains.
The cost of that silence showed up immediately in how the market handicaps the next move. There is no Fed meeting in August; the next decision comes September 15 and 16. In the span of a single week, the probability traders assigned to a September rate move ran from 36% before a Warsh appearance, to 56% after he spoke, to 68% the following Monday, then back to roughly 58% by Thursday [1]. That is a swing of more than thirty points driven not by fresh economic data but by parsing a single official's remarks.
The professional forecasters are not converging either; they are pulling in opposite directions. Bank of America expects three quarter-point rate hikes through the end of the year. Citigroup expects three rate cuts by January [1]. Those are not two shadings of the same call. They are mirror-image bets on whether the Fed's next job is to cool the economy or cushion it, and both are being made without the guidance that used to narrow the range.
The disagreement extends to the data itself. Ahead of the July jobs report, still pending at publication, the consensus and the outliers sat far apart: Dow Jones put the consensus near 83,000 nonfarm payrolls, FactSet at 100,000, and high-end estimates reached 120,000, while Vanguard, citing pension data, projected just 18,000 [1]. For scale, the last confirmed print, June 2026, came in at 57,000 payrolls with unemployment at 4.2%. A forecast range that runs from 18,000 to 120,000 is not a market reading the economy; it is a market guessing at it.
Someone benefits from all of this. Traders who profit from data-driven volatility do better when each release detonates rather than fizzles, and a Fed that keeps its options open keeps its own room to maneuver. The people carrying the cost are households and borrowers, who now face rate decisions with no official read on the direction and no way to plan against them. Removing the Fed's signal does not remove the risk. It shifts that risk onto everyone left to guess, from the mortgage shopper to the small employer deciding whether to hire before September.
