Three of Five Voting Fed Presidents Wanted a Rate Hike

Why It Matters

The Federal Reserve’s own statement records the July 29 decision as a 9 to 3 vote to leave the federal funds target range at 3-1/2 to 3-3/4 percent. It names the three who voted against, and all three wanted rates higher rather than lower. Beth M. Hammack of Cleveland, Neel Kashkari of Minneapolis and Lorie K. Logan of Dallas “preferred to raise the target range for the federal funds rate by 1/4 percentage point at this meeting”, per the statement.

Quartz reported the same afternoon that all three are regional bank presidents. That detail is the story, because of how the committee is built. The FOMC seats 12 voters. Seven are members of the Board of Governors, one is the president of the New York Fed, and 4 come from the remaining 11 Reserve Bank presidents on a one-year rotation. Five presidents therefore hold a vote in any given year. Three of those 5 dissented.

The two weeks since have gone the other way. Charles Schwab reported the CME FedWatch odds of a September increase at 78.8% on the morning of the vote. They were 60.1% by the end of the press conference. By Sunday morning 24/7 Wall St. put the same number at 25%. The Market Context reported on August 7 that payrolls fell 23,000 in July while the jobless rate fell too, which is the other set of figures the nine were reading.

Three dissents were a majority of the five presidents who hold a vote

Counted against the whole committee, three dissents out of twelve is 25%. Counted against the group the dissenters actually belong to, it is 60%. The Federal Reserve’s published 2026 membership list gives seven governors, Chairman Kevin Warsh among them, and five presidents: John C. Williams of New York, Hammack, Kashkari, Logan and Anna Paulson of Philadelphia. The presidents split three to two in favour of raising rates. The governors went seven to nothing the other way.

Narrow it once more and the concentration is sharper still. New York votes permanently, so only four seats rotate, and this year they went to Cleveland, Minneapolis, Dallas and Philadelphia. Three of those four rotating presidents voted to raise rates. Forbes described the dissenters on August 11 as “three governors”. The distinction is precisely what the vote turns on, because governors are appointed in Washington and confirmed by the Senate while presidents are selected by their own regional boards.

The arithmetic that follows is simple and it is arithmetic rather than prediction. A majority of a twelve-member committee is seven. Three votes for a quarter-point increase are already on the record. Four more would have to come from the nine who voted to hold, and seven of those nine are the governors who moved as one.

The most hawkish vote in nearly ten years lowered the odds of a hike

CNBC reported that July 29 was the first time since September 2016 that three policymakers dissented with a unified view of which way rates should go. A hawkish surprise of that size would ordinarily push the priced odds of a hike up. They fell 18.7 points that afternoon, from 78.8% to 60.1% on Schwab’s reading of CME FedWatch.

The reason is that the two numbers answer different questions. The dissents told the market that three people want higher rates. The vote told it that nine do not, and Warsh’s press conference added nothing to lean against. He called the discussion a “good family fight” while stressing the committee’s commitment to price stability, per Schwab’s account. He also defended giving markets less guidance, saying participants are “learning to play the ball and not the referee”.

Then the data did the rest of the work. The Bureau of Labor Statistics put headline inflation at 3.4% over the twelve months to July, down from 3.5% in June, with core at 2.5%. The energy index fell 1.5% in July after falling 5.7% in June, which is two consecutive monthly declines in the component that opened the gap. The Market Context reported on August 12 that energy is 7.4% of the CPI basket and 29% of the inflation rate.

Six published probabilities for the same question differ by 53.8 points

A reader checking the odds this weekend would have found 78.8%, 65%, 63%, 34%, 32.3% and 25%, all describing a September rate increase. The spread between the highest and the lowest is 53.8 points. None of the figures is wrong, and three mechanical differences explain the gap.

The first is the date. The odds moved 53.8 points in eighteen days, so a probability without a timestamp describes an unknown moment. The second is the horizon. Schwab’s 63% figure, published August 14, is the chance of at least one increase by the end of the year across three remaining meetings, not the chance of one in September. The third is the instrument. J.P. Morgan Wealth Management cited fed funds futures for its 65% on August 5, while 24/7 Wall St. cited betting markets for Sunday’s 25%.

The same care applies to the underlying data. The 24/7 Wall St. piece published Sunday morning cites gasoline up 26.7% and energy up 15.7% over twelve months. Those are the June figures. The July release, out on August 12, put gasoline at 24.6% and energy at 14.7%, so both are 2.1 and 1.0 points lower than the numbers in circulation. The Market Context reported on August 10 that oil settled up 5.3% while the S&P 500 closed down 0.06%, which is the same pattern of a price and its narrative separating.

All three dissenters lose their vote at the first meeting of 2027

The Fed publishes the rotation years ahead. Its table gives the 2027 voting banks as New York, Chicago, Richmond, Atlanta and San Francisco. Cleveland, Minneapolis and Dallas are all absent, so every one of the three dissenters stops voting when membership changes at the first regularly scheduled meeting of next year.

Three meetings remain before that happens, on September 15 and 16, October 27 and 28, and December 8 and 9, per the Fed’s own calendar. Two of the three carry a Summary of Economic Projections, which means September and December each publish a fresh set of dots showing where every participant sees rates going. Schwab’s head of macro research and strategy, Kevin Gordon, put the consequence of Warsh’s reduced guidance plainly, saying watchers “will have to be more manual in tallying the hawks versus doves”.

The committee’s own statement gives them something to tally against. It attributes elevated inflation “in part reflecting supply shocks that have driven price increases in certain sectors, including energy”, which names a cause the funds rate does not reach directly. A further supply shock arrives this week, and The Market Context reported on Sunday that a 50% tariff on Canadian goods takes effect Wednesday covering no cars.

What to watch

1. The minutes of the July 28-29 meeting, due Wednesday August 19. Every 2026 set has been released exactly 21 days after the second day of the meeting. Read the passage describing the arguments of those who preferred a higher target range, and count how many participants are described as sympathetic rather than the three who formally dissented.

2. The Kansas City Fed’s Jackson Hole symposium, August 27 to 29, where Schwab expects a speech from Chairman Warsh. Read whether he characterises the energy component of inflation as transitory or as broadening, because the statement’s phrase “supply shocks” is the hinge of the internal argument.

3. The July personal consumption expenditures price index, scheduled for August 26. Read the core reading against the Cleveland Fed nowcast of roughly 3.3% that 24/7 Wall St. cited, since that model and the published series have been telling different stories.

4. The August producer price index on September 10 and the August consumer price index on September 11, the last two inflation prints before the meeting. In the CPI release, read the energy index line in Table A for a third consecutive monthly decline.

5. The FOMC statement and Summary of Economic Projections on September 16. Read the vote tally first and the dot distribution second, because the projections show every participant’s view while the tally shows only the five presidents and seven governors who hold a vote.