On September 16, the Federal Open Market Committee raised the federal funds rate by a quarter point, to a range of 3.75% to 4%. The vote was 12 to 0.
Seven weeks earlier, the same committee had held rates steady, 9 to 3, with three members dissenting in favor of exactly the increase that September delivered. The minutes of that July meeting record that several other participants, who don’t vote this year, favored a hike as well. By September the dissenters had won the room.
Why the shift? Not because the data got easier. Core inflation, as measured by the Fed’s preferred index, the price index for personal consumption expenditures (PCE), was 3.3% in July, marking the fifth year above target. The August consumer price index came in at 3.4%, though most of that was gasoline, up 27% from a year earlier; core CPI ran at 2.4%. On the other side of the ledger, payrolls grew by 162,000 in August and unemployment held at 4.1%. But the hiring rate in the Bureau of Labor Statistics’ job openings survey sits at 3.2%, a level we have not seen outside of a recession, and 27% of the people out of work have been out of work for more than six months. Growth in the second quarter was 1.5%. This is a low-hire, low-fire labor market, and it is stable until it isn’t.
So, the committee had two ways to be wrong. Hold rates, and it accepts a sixth year of inflation above target with expectations at risk. Raise them, and it tightens into an economy growing at 1.5%, with inflation mostly in gasoline, and the labor market isn’t hiring. Either choice could be criticized, and both criticisms could be defended by pointing to the same sentence in the Federal Reserve Act: one camp reading maximum employment, the other stable prices, both in good faith.
A rule that justifies both answers is not doing the work of a rule.
The committee chose anyway. It chose price stability, and it chose unanimously. In his press conference afterward, Federal Reserve Chair Kevin Warsh said that because inflation has run above target for more than five years, “our predominant focus is on the price-stability side of our mandate.” That is a priority, stated plainly by the chair. It is the priority I would have chosen. And it appears nowhere in the document the committee has written to tell the public how it makes decisions.
A Silence That Used to Be Affordable
When Congress wrote the dual mandate into the Federal Reserve Act in 1977, it never said whether maximum employment or stable prices took precedence. That ambiguity was affordable for most of the years since. Why? Because most of the shocks the Fed faced during that time came from demand.
When a demand shock happens, output and prices fall together. And when demand runs hot, both rise. The objectives point in the same direction, so there’s no conflict to adjudicate and no one asks which takes precedence.
Supply shocks are a different story because they split the mandates. An oil embargo, a war that threatens a strait, on-again-off-again tariffs, a working-age population that stops growing due to immigration policy; each of these factors simultaneously pushes up prices and pushes down output. The mandates pull in opposite directions, and the charge from Congress doesn’t specify which should prevail.
A rule that justifies both answers is not doing the work of a rule.
Paul Volcker walked into that in 1979. What did the framework give him? Nothing.
So, Volcker supplied the answer himself. He decided price stability was paramount, and he raised rates until the economy broke and then held them there while unemployment reached 10.8% in late 1982, the highest rate since the Great Depression. Of course, this did not sit well with many people. Home builders mailed him two-by-fours they had no houses to build with, and a congressman from Texas introduced a resolution to impeach him. Despite all these complaints, Volcker held his ground.
Volcker’s strategy eventually succeeded at the cost of a serious recession.
What was the playbook he followed? The ambiguity in the mandates gave him nothing to lean on, so what carried the decision was one man’s judgment, held under pressure. Every chair since has inherited the benefit of what Volcker did, and none has felt the need to write down the priority that made his stand necessary in the first place.
Credibility built over decades is not the personal property of whoever holds the chair.
Why This Surfaces as a Communication Problem
In July, Warsh named five task forces to review how the Fed operates: communications, the balance sheet, data, productivity and jobs, and inflation frameworks. He has recruited serious people to serve on them, and he has said he expects them to report by the end of the year.
At Jackson Hole in August, he said forward guidance was adopted during the financial crisis and “has overstayed its welcome.” In normal times it “risks creating ambiguity in the name of clarity.” Then the line the whole review seems to rest on: “A quieter Fed, more purposeful in its communications, is better able to meet its objectives.”
He’s right about the diagnosis. Take the dot plot in the Summary of Economic Projections, or SEP. Forward guidance and the dots grew as elaborate as they did because the committee was trying to signal a path it couldn’t fully commit to. At any given meeting, either mandate could become binding, so the guidance came wrapped in caveats, and the market took the dots as a promise the committee could never make. Warsh has stopped submitting a dot of his own and dropped forward guidance from the statement, which ran nine sentences on September 16.
Here’s where I part company with him. A priority is not forward guidance. Forward guidance is a promise about the path: Rates will stay here until something happens. A priority is a rule about what governs when the two objectives conflict. You can strip every word of guidance from the statement and still tell the public which mandate wins in a supply shock. Warsh did exactly that on September 16. He just did it in a press conference, in a single sentence, rather than in the committee’s framework.
Shorten the statement and the Fed says less. Withhold the chair’s dot and it reveals less. Neither move tells anyone what the committee will do the next time inflation and employment pull apart, because that was never a problem of wording.
A quieter Fed is not a clearer one. You cannot communicate your way out of an objective you have not written down.
The Fed Already Owns the Document That Would Fix This
Where would the committee write it? Not in new legislation.
The Statement on Longer-Run Goals and Monetary Policy Strategy belongs to the committee. The FOMC adopted it in January 2012, rewrote it in August 2020, rewrote it again in August 2025, and reaffirmed it, unchanged, this past January. Every one of those actions was the committee acting on its own authority, with no act of Congress required.
A quieter Fed is not a clearer one. You cannot communicate your way out of an objective you have not written down.
The Inflation Frameworks task force, led by Greg Mankiw, Thomas Sargent, and William White, should use that document to state the priority the chair has already stated. When price stability and maximum employment come into conflict, I believe that price stability governs, and maximum employment is pursued conditional on it.
The European Central Bank and the Bank of England both work this way. The Bank of England’s statute does it in one sentence: Maintain price stability, and “subject to that,” support growth and employment.
To be clear, employment remains within the mandate under this approach, and it recognizes that price stability is what makes durable employment possible, not a rival to it. A committee that lets inflation expectations come unanchored hasn’t protected anyone’s job. It has postponed the cost, and the cost comes back measured in people out of work rather than in basis points.
A written priority wouldn’t have made the September decision easy. The committee would still have had to weigh expectations against a real cost in employment. But everyone, inside the room and out, would have known which question was being decided.
What I am proposing is a testable claim, so let it be tested. If the ordering is written down and understood, then after the next supply shock the spread between the highest and lowest professional forecasts of short-term interest rates should narrow, and long-horizon inflation expectations should move less when energy prices move. The Philadelphia Fed’s Survey of Professional Forecasters will show the first. Breakevens in the inflation-protected Treasury market will show the second.
How much they’d move, I don’t know. That they should move at all is the part that can be checked. If neither does, the change failed, and someone should say so plainly.
The Best Argument Against This
Now the counterargument to this proposal is that Congress wrote both objectives into the statute and meant it. A committee that reorders them in a document it drafts for itself is editing the instructions it was handed, and there’s a risk that an institution that has done that once has told everyone it might do it again.
This objection is serious, and I won’t pretend otherwise.
How do I defend my proposal? Consider that the committee already provides a tiebreaker. It supplies one at every meeting where the two objectives are in conflict, and it calls that judgment; it is just unwritten and unacknowledged. July and September were both such meetings. In July, the committee split. In September, it agreed. In neither case did it say what rule it was applying.
Writing the priority down doesn’t hand the Fed a power it lacks. It makes visible a power the Fed has been using in private. And a written priority is far easier for Congress to examine and overturn than a habit nobody has to admit to.
There is also a second objection that deserves an answer. In 2008, and again in 2020, the FOMC leaned hard toward the employment mandate, and it was right to do so. A strict hierarchical ordering of mandates sounds like it would have slowed the committee’s response to the shocks hitting the economy.
I don’t think this is what the committee would’ve done, and I was there for the 2020 decision. The ordering I propose is conditional: Employment is pursued if price stability is reasonably secure. In both episodes price stability was secure, and expectations, at least in the markets, were running too low rather than too high. The ordering I propose would have endorsed exactly what the committee did.
Why This Year
Energy supply has again become a geopolitical instrument. Tariff policies change faster than anyone can forecast them. And the working-age population is growing slowly, while restrictions on immigration have pushed the payroll growth needed to keep unemployment steady close to zero.
An institution that has agreed on a priority and will not write it down has not finished the job.
Supply shocks are arriving more often, not less, and the neutral rate that guides much of monetary policy has drifted higher. The committee’s own median estimate of the longer-run federal funds rate now sits at 3.2%, up from 2.5% as recently as 2023, and the 10-year Treasury touched 5% the week of the September meeting, from 4.2% at the start of the year. The country’s fiscal situation adds to that pressure and leaves little room to absorb a mistake. The net interest payments we are making as a nation passed $1 trillion in the first eleven months of this fiscal year, up about 9% from a year earlier.
So, the pressure runs in both directions at once. Outside the Fed the calls are for cuts, and inside the committee the votes have been for hikes. Both camps cite the same mandate, and both are entitled to.
Whatever else that tells you, this is not a problem that belongs to one party.
What the September Vote Was Really About
Unanimity under political pressure can be a warning sign, but only when the committee is agreeing with the pressure. This one voted against it, 12 to 0. That is an independent institution doing its job.
What should concern all of us is that the committee has now agreed on a priority it hasn’t written down. Twelve people voted for price stability first. The chair said so. The framework statement still lists the two objectives side by side with no word about which governs.
The committee will face the same choice at its next meeting, and the September projections suggest most participants expect to raise rates again this year. Each time, it will have to supply the tiebreak itself.
An institution that has agreed on a priority and will not write it down has not finished the job.
The task forces report at the end of the year. Rewrite the framework statement once more and put on paper what the chair has already said aloud.
Volcker had to provide that answer from his own conviction. So, it turns out, did Warsh. Their successors should not have to.
