Why the Fed's forecast matters more than its promises

Over the weekend the Federal Reserve's new chairman, Kevin Warsh, used his first address to the central-bank world's annual gathering in Jackson Hole partly to tell investors to stop leaning on him. Inflation, he said, is still running at 3.7% and should be the Fed's "predominant focus"; if underlying price pressures do not move toward the 2% target "clearly and at sufficient speed," the bank will "have work to do." And in case anyone expected a roadmap, he renounced the tool investors have used for two decades to read him. Forward guidance -- telling markets where rates are likely to go -- has "overstayed its welcome," he said, and risks "creating ambiguity in the name of clarity."

One of the profession's most prominent Fed watchers thought that was roughly half right. Adam Posen, an American economist who served on the Bank of England's rate-setting committee and now runs the Peterson Institute for International Economics, says it is fine, even wise, for the Fed to refuse to commit to a path for interest rates. It is not fine, he argues, for the chairman to treat the Fed's published economic forecast as part of the same bargain. The forecast "gives you a framework for having a debate in the committee because without a forecast, everybody's just sort of blathering."

The exchange matters far beyond the seminar circuit, because a central bank that merely talks has already acted. Rates themselves are not that important to a portfolio until markets can anticipate them; nearly every price a retail investor watches -- bond yields, mortgage rates, the discount applied to future earnings in stocks -- is set in anticipation of the Fed's next move. How the Fed communicates is therefore as much a tool as the rate itself. Warsh's weekend represents the biggest change to that tool in a generation, and the question Posen raises is whether the new chairman is dismantling the wrong half of it.

To see the stakes, separate the two things Warsh is throwing away together. Forward guidance is a promise about the Fed's own future behaviour. Born of desperation after the 2008 crisis, when rates were stuck at zero and the usual lever had failed, it lets the central bank push borrowing costs down by committing that they will stay low for a long time. It worked at first, then quietly decayed into a fixture of market life: the "Fed put," under which investors priced every wobble as a reason for the central bank to ride to the rescue. Warsh's indictment of it is not merely stylistic. He blames his own institution -- "responsibility for 65 months of sustained, elevated inflation sits squarely with the central bank" -- and identifies a mechanism: the hall of mirrors. If markets price off the Fed and the Fed reads markets, both stop seeing the world. It is the strongest version of the case, and he deserves the concession.

A forecast is a different animal. The Fed's quarterly Summary of Economic Projections, with its "dot plot" of where each official thinks rates belong, is not a commitment about what the Fed will do. It is the collective judgment of eighteen people who sign their names, in public, every quarter, and live with being wrong. Publishing it does three jobs at once. It forces the committee to hold a view, which disciplines its debate. It makes disagreement visible. And it gives the public something it can test against events -- a forecast you can beat is a forecast that keeps the Fed honest. That is information, not a promise. The promised rate path created the seductive distortions Warsh is right to fear. The map never did.

Drop the promise and keep the map, in other words. What unsettles Posen is that Warsh appears to be dropping both. In June, the first projections of the new era arrived with a visible hole: nine of the eighteen officials who submitted estimates pencilled in at least one rate rise this year, while the chairman submitted none. The Fed still publishes its forecast; its leader has conspicuously declined to take part in it. His reasoning, delivered on Saturday, was that "providing forecasts to illustrate the Fed's reaction function works better in theory than in practice" and that "accuracy in economic forecasting is still just an aspiration." All true, and all beside the point; no one holds a weatherman to a guarantee.

The stranger part of the doctrine is that Warsh cannot stop forecasting, and does not try. In the same speech in which he disavowed the craft, he told the world the economy is growing briskly, that unemployment is a low 4.1%, and that the six-month rate of inflation is a troubling 4.1%. A quiet Fed, he promises, will be "more purposeful in its communications." The trouble is that a central bank cannot actually resign from being watched. It can only choose whether what it publishes is disciplined, collective and checkable -- or made up, chairman by chairman, on the spot. The choice between the two is what the fight is about.

Investors will notice the difference even before the committee does. The next meeting is on September 15th and 16th, and markets are already restive: at the July meeting, three officials -- Beth Hammack, Neel Kashkari and Lorie Logan -- dissented from the decision to hold rates at 3.5-3.75% and voted instead for an immediate hike. Traders before the Jackson Hole speech had priced next month's meeting at a little over a one-in-three chance of an increase; after the chairman's remarks they moved it closer to a coin flip, and the two-year Treasury yield jumped to a one-month high. For all Warsh's talk of markets "learning to play the ball and not the referee", the weekend was a vivid demonstration that they are still watching the referee's lips.

There is a genuinely uncomfortable irony beneath the rhetoric. Warsh defends his opacity by invoking the people it supposedly protects: the "hall-of-mirrors" blindness, he says, does "the most serious harm" to "those without financial assets", because a Fed locked in mutual reassurance with markets let inflation ripen on ordinary people's dollars. There is something in this. But uncertainty about the rate path is not free for the asset-less either. Mortgages, car loans, small-business borrowing and the volatility of savings all carry the price of guessing the central bank's next move. The cost of a less legible Fed falls most heavily on the households that can least afford to watch it closely. A vow of silence dressed as solidarity is still a vow of silence.

None of this requires agreement with Posen's own call, which is bolder than his attachment to forecasts. He thinks inflation will persist, that last autumn's 75 basis points of cuts left policy too loose, and that the Fed should raise rates twice before year-end, warning that waiting for December or January would be "probably too late" and "probably too little". That is a bet about a bet, and it can be wrong; the oil shock from the conflict with Iran, the AI investment boom and the trajectory of a 3.7% inflation rate are all live questions. The transferable lesson of Jackson Hole is institutional, not cyclical. Warsh is right to retire a promise that outlived its usefulness and to accept that a central bank earns credibility from results, not prose. But a claim to be judged by results, made in the same breath in which the institution stops publishing the one document by which its judgment can be checked, is not accountability. The healthy arrangement was on offer: drop the promise, keep the forecast. A chairman who cannot distinguish them will discover that a quieter Fed is watched more closely, not less -- and that the market's attention is a tax he cannot choose to stop paying.