Kevin Warsh gave his first semiannual monetary policy testimony to Congress this week, less than two months into his term as Chair of the Federal Reserve. He used it to make an argument rather than to deliver a status report — and that choice is itself the most informative thing about it.
What was said
Warsh described inflation as “a tax on the American people and businesses,” told legislators that the Fed intends to be rid of it, and said the inflation surge of the last five years “will be a thing of the past” if the central bank gets policy right.
He called for a “regime change” in the Fed’s approach, and criticised the flexible average inflation targeting framework adopted in 2020, describing it as a mistake.
Notably, none of this was accompanied by a commitment about rates. That combination — strong language about the objective, no commitment about the instrument — is the standard structure of central bank communication, and it is what makes these appearances easy to over-read.
What the testimony is for
The semiannual appearance exists because the Federal Reserve is an independent agency exercising substantial power under a mandate granted by Congress. Independence without accountability is difficult to justify in a democracy, and the testimony is the accountability mechanism.
Its practical function differs from its formal one in a way worth understanding.
- It is the one setting a Chair cannot control. A press conference follows a decision and takes questions from journalists who need continued access. Congressional questioning is adversarial by design, comes from people with no such constraint, and covers whatever the questioner wishes.
- It is directed at two audiences simultaneously. The immediate audience is a committee. The real audience is the market, which parses every phrase for a change in the reaction function.
- It is a defence of the framework, not of a decision. Individual meetings are explained in statements and press conferences. Testimony is where the underlying approach is argued for.
Why the framework criticism is the substantive part
Of everything said, the criticism of flexible average inflation targeting carries the most weight, because frameworks outlast decisions.
The 2020 framework changed how the Fed pursued its target in a specific way: rather than aiming at 2 percent from wherever inflation happened to be, it would allow inflation to run moderately above target for a period following shortfalls, so that the average over time came out at 2. The reasoning was that persistently undershooting a target drags expectations below it, and that asymmetric responses to overshoots had contributed to that undershoot.
The criticism is that the framework was designed for a world of persistently low inflation and was adopted immediately before conditions reversed — and that a commitment to tolerate overshoots is a poor position from which to respond to an actual overshoot.
Whether that is fair is genuinely contested. The relevant point for a reader is that a framework change is more consequential than any single rate decision, because it alters how the Committee will respond to every future situation. Markets price the path of rates, and the path depends on the reaction function.
The limits on what a Chair can deliver
Rhetoric about intent should be discounted against the structural constraints on the office.
The Chair holds one vote of twelve. Framework changes require the Committee, and the Committee was assembled by someone else — governors serve fourteen-year terms specifically to outlast any individual chairmanship. A framework review is a multi-year process involving staff work, public consultation and internal negotiation.
There is also the constraint nobody can appoint away. Monetary policy reaches the economy with lags of one to two years. A chair acting decisively today is making a decision whose consequences arrive well into the future, on forecasts that are unreliable at that horizon. That discipline applies regardless of conviction, and it is the reason chairs have frequently governed differently from how their appointments were interpreted — a point we set out when Warsh took office in May.
How to read one of these
- Compare the prepared statement against the previous one. It is drafted with the knowledge that it will be compared word by word. Changes are deliberate.
- Watch what gets declined. Questions a Chair will not answer mark the boundaries of the current position more precisely than the answers given.
- Note conditions stated in the negative. “We would need to see substantially more evidence before…” is a specific bar, and more informative than any statement of intent.
- Ignore the political framing. Much of the questioning is directed at constituents rather than at eliciting information, and the answers to those questions carry no policy content.
- Weight the vote over the words. What the Committee actually does at its next meeting, and how the vote splits, is the test of whether rhetoric describes a majority position or one person’s preference.
That final test arrives at the end of this month, when the Committee next meets.
Sources: reported remarks from the Chair’s semiannual monetary policy testimony, July 2026.
