The Fed Chairman Just Told Markets to Stop Watching Him

Kevin Warsh opened his first Jackson Hole address as Federal Reserve chairman with a joke about hiking trails, offered an outline of his remarks, and then told the audience what to call it: not forward guidance.

The joke was the thesis. Over the following forty minutes, delivered on 28 August to mark his hundredth day in office, Warsh set out a case for why the practice of telling markets what the Fed intends to do has outlived its usefulness, why he does not intend to supply a reaction function in its place, and why inflation running at 3.7 percent means, in his words, that there is work to do.

Markets took the hawkish half. Some large banks brought forward their expected timing for rate increases. The speech was widely read as the clearest signal yet that the next move is up.

That reading is correct and incomplete. The rate signal is the news. The framework change is the story, and it will outlast the current cycle.

What he actually dismantled

Forward guidance became a standing feature of Federal Open Market Committee statements in December 2008. Warsh was on the Board at the time and helped introduce it. His argument now is not that it was wrong then but that it was a crisis instrument that stayed in service after the crisis ended.

His objection has three parts, and they are worth separating because they are not equally strong.

The first is about clarity. Guidance in normal conditions, he argued, risks creating ambiguity in the name of clarity, and overcommitting to future decisions can lead markets, businesses and households astray. He cited 2021 specifically, when guidance may have slowed the policy response to rising inflation, and pointed to work by Christina and David Romer making that case.

The second is about institutional freedom. When policymakers make what he called quasi-commitments on rates through the cycle, they constrain their own ability to make the right call when the decision arrives. This is the self-interested argument and he did not pretend otherwise.

The third is the most interesting and the least discussed. Warsh described a hall-of-mirrors problem: the Fed reads market prices to understand conditions, markets read Fed guidance to set prices, and the two end up reflecting each other rather than the underlying economy. If both parties are looking at each other, both are more likely to miss a genuine turn in events.

He then made an argument about who pays for that. Market participants, he noted, are unlikely to bear the largest cost of the hall-of-mirrors problem. If the Fed misjudges inflation or the economy, the damage falls on people without financial assets, not on the traders who were closest to the signal.

Whatever one makes of the policy, that is a serious argument and it deserves to be engaged rather than dismissed as communication preference.

What he refused to replace it with

The obvious institutional compromise was available and he declined it.

If a chairman will not offer guidance, the standard alternative is an explicit reaction function: a stated rule linking incoming data to policy response, so markets can do the arithmetic themselves. Warsh rejected this too, on the grounds that economic understanding is not precise enough to support a mechanical rule, that the factors most relevant to policy change over time, and that reaction functions work better in theory than in practice.

He committed instead to a set of principles. Trends matter more than individual data points. Supply-side conditions can only be inferred, never observed, which makes the demand-supply balance an imprecise judgement. The 2 percent PCE target is fixed. Price stability is not self-executing and inflation is not necessarily mean-reverting, which is a direct rejection of the view that elevated inflation resolves itself. Short-term rates are the predominant tool and unconventional policy should be reserved for genuine crises.

He also revived something the Fed has largely set aside: money matters. Warsh said explicitly that money growth, from the central bank and from the banking system, deserves attention as a signal, a position long out of fashion in the institution he now runs.

"I stand here today committed to a discipline, not to a decision," he said in closing. That sentence is the operating principle, and it is a genuinely different offer from the one the Fed has made for eighteen years.

The economic read underneath it

The assessment section is where the hawkish interpretation comes from, and the numbers support it.

Warsh described himself as impressed by the economy's overall performance. Business capital expenditure is rising rapidly, with the four-quarter change in equipment and intangibles investment around 9 percent, its fastest since 2021. He attributed more than half of this year's capex growth to the AI buildout. S&P 500 profits have grown more than 20 percent over the past year, with margins elevated against history. Real consumer spending has increased more than 2 percent over four quarters, and private domestic final purchases, which he considers a cleaner signal than GDP, has risen at nearly 3 percent this calendar year.

On employment he was unambiguous. Unemployment at 4.1 percent, roughly unchanged for two years. Four-week average claims near their lowest in decades. Low turnover, which he attributed partly to the large-scale rematching of workers and employers after the pandemic, and low monthly job gains as a natural consequence of slow labour supply growth rather than weakness. He named recent graduates as an area of concern and otherwise judged the labour market consistent with full employment.

Then the problem. Twelve-month PCE inflation at 3.7 percent, with the six-month change at 4.1 percent, meaning the recent trend is worse than the annual figure. Core measures elevated on both PCE and CPI.

His method for reading underlying inflation is the most useful disclosure in the speech, because it tells you what he is watching. Warsh disaggregates all 199 components of the PCE basket and counts how many are rising above 3 percent. Over twelve months, 54 percent were. Over six months, 49 percent. Both are far below the post-pandemic peak near 77 percent and both remain well above the 32 percent that prevailed in the two decades before the pandemic.

He acknowledged that summer readings came in better than expected and said plainly that they do not tell him underlying trends have meaningfully improved.

Then the line that did most of the work with markets: he said he would be hard pressed to describe broad financial conditions as restrictive. Credit spreads near the low end of historical ranges, strong issuance, and the July Senior Loan Officer Survey showing commercial and industrial lending standards on the easier end of their range. A central bank that considers policy non-restrictive while inflation runs at 3.7 percent has told you which direction it is leaning without issuing guidance.

He also accepted the institution's responsibility for what he called 65 months of sustained elevated inflation, placing it squarely with the central bank. Coming from a sitting chairman about his own institution, that is an unusually direct admission and it functions as a commitment device.

What this changes for anyone planning against rates

Three practical consequences, and none of them depends on whether the Fed hikes in September.

Optionality has repriced upward. Treasury and hedging strategies built over the past decade were, in effect, positions on a communicated path. That path is no longer communicated as a matter of doctrine, not as a temporary posture. The right response is not to guess harder. It is to structure financing and hedges so that being wrong about direction costs less, which means shorter tenors, more explicit triggers, and less conviction embedded in any single assumption.

The inflation composition data is now a leadership tool, not an economist's footnote. Warsh told the market exactly which measure he acts on and how he reads it. Anyone with pricing decisions to make can run the same analysis: what share of your input basket is rising above 3 percent, over six months and twelve. That is the question the chairman is asking, and businesses that ask it about their own cost base will be reading the same signal he is.

Financial conditions are the variable to watch, not the statement. Warsh has said he considers current conditions non-restrictive. Credit spreads, issuance volumes and lending standards are therefore the leading indicators of Fed discomfort. If spreads stay tight and issuance stays strong while inflation stays above target, the case for tightening builds regardless of what any statement says, because the chairman has told you that is the mechanism he is watching.

The argument against him

It should be stated, because it is not weak.

A central bank that says less does not eliminate uncertainty about its intentions. It relocates the cost of that uncertainty onto everyone who has to plan around rates: corporate treasurers, small businesses pricing multi-year contracts, households considering mortgages. Critics have said the approach has left markets confused about the Fed's intentions, and Warsh defended it directly at Jackson Hole rather than treating the criticism as a misunderstanding.

There is also a coordination argument. Forward guidance was never only about information. It was a tool for making policy more effective by shaping expectations, and expectations do real work in the transmission mechanism. Removing the tool may make each individual decision freer while making the cumulative effect of policy weaker.

And a chairman who declines both guidance and a reaction function is asking to be trusted on judgement. Warsh's response is that credibility rests on results rather than explanation, and he closed on a line from Chuck Yeager to that effect. That is a coherent position. It is also one that cannot be evaluated until the results arrive, which is a considerable amount of institutional credit to draw on in advance.

The read

The Federal Reserve has spent eighteen years telling markets what it expected to do. It has now stopped, on purpose, and explained why at length.

The immediate consequence is a hawkish tilt: inflation at 3.7 percent, conditions the chairman does not consider restrictive, and a stated standard that requires confidence inflation is moving to target clearly and at sufficient speed before the pressure comes off.

The longer consequence is that a category of planning input has been withdrawn from the economy. Businesses that treated Fed guidance as a free forecast now have to produce their own, and the ones that had already stopped relying on it are, for the first time in a long while, at an advantage.

Warsh committed to a discipline rather than a decision. Everyone downstream of him now has to do the same.