Goldman Sachs Bets the Fed Can Skip an October Hike

Goldman Sachs pushed its forecast for a second Federal Reserve rate hike to December after August core PCE inflation printed at 3.0%. The bank’s economics team, led by Jan Hatzius, now sees little chance of a move at the October 27-28 meeting and a strong chance the committee decides no extra increase is needed at all.

That is a wager against the Fed’s own September forecasts, against Governor Michael Barr, and against a month in which households spent far more. The September 16 hike still stands. The path after it does not.

The 3.0% Core Print That Moved Goldman

The Bureau of Economic Analysis said the August PCE price index rose 3.4% from a year earlier, with the core index, which strips out food and energy, up 3.0%. On a monthly basis, headline prices rose 0.3% and core prices rose 0.2%.

Economists in a CME Group Econoday survey had looked for 3.7% and 3.3% on the annual figures, and for 0.4% and 0.3% on the monthly ones. The miss was real. So was the annual rewrite sitting underneath it. The release folded in the 2026 update of the national accounts, with revisions running back to January 2021.

AUGUST PCE AGAINST THE FORECAST

Hatzius’s team told clients the report, and the downward revisions from those measurement changes, imply core PCE will rise 3.0% on a fourth-quarter to fourth-quarter basis in 2026. That sits 0.4 percentage points below the 3.4% median in the September Summary of Economic Projections.

Spending did not cool with the price indexes. Current-dollar PCE jumped $190.8 billion, or 0.9%, the product of $114.1 billion more on goods and $76.7 billion more on services. Personal income rose only $66.6 billion, or 0.2%. Real spending was still up 0.6%. The saving rate was 4.1%.

A print can be below forecast and still leave inflation a full percentage point above the Fed’s 2% goal. The annual rewrite made the year-over-year rates look calmer than the July vintages did. It did not make 3.0% core into 2%.

Why October Is Now the Long Shot

Goldman had spent the second half of September in the early-hike camp, looking for a second quarter-point increase at the October 27-28 meeting. The bank flipped after two things arrived in order: New York Fed President John Williams in Buffalo on September 29, then the PCE report on September 30.

Williams, who votes at every FOMC meeting, said the September increase bought the committee time. He still allows for one more move late in the year if the economy tracks his forecast. He did not treat the next gathering as a live decision.

With the policy action we took at our September meeting, there is no need for urgency, and we have time to gather more information.

John C. Williams, President, Federal Reserve Bank of New York, University at Buffalo

He added that if the outlook holds, “one further upward adjustment of the federal funds target range may be appropriate late this year.” Goldman took the no need for urgency after September line as a veto on a back-to-back hike, then used the inflation report to move the second increase to December.

Rob Kaplan, Goldman Sachs vice chairman and a former Dallas Fed president, had already sketched that calendar on September 23. He said that if he were in the room he would skip October unless the data forced a move, then look again in December, and that markets may be pricing too many hikes. Hatzius’s note is the economics team catching that house view, then going a step further with the “strong chance” that even December comes off.

Barr Has Not Folded His Hawkish Case

The other vote Goldman has to beat is Barr’s. Hours before Williams spoke, the governor told the Detroit Economic Club that inflation progress had been knocked off course by energy prices and a surge in AI-related investment. He said he does not yet see a clear trend toward a timely return to 2%.

His base case, in remarks prepared for that event, is that further policy adjustments are likely to be needed to bring inflation down on time. He made the same point in Chicago on September 23, after arguing the Fed had been “out of position” heading into the September meeting and that risks to the inflation target had risen while risks to the labor market had receded.

Williams is not arguing that prices are fine. He put this year’s inflation at 3.5%, sees it only just above 2% next year, and does not get back to the long-run goal until 2028. He named three forces that lifted inflation by about a percentage point over the past year and a half.

WHAT WILLIAMS SAYS IS STILL IN THE PRICE LEVEL

• Tariffs: He said they are no longer adding to goods inflation, with the caveat that new duties could change that.
• The war and fuel: Fighting in the Middle East and tight refining capacity are lifting crude and the spread of gasoline and diesel over oil.
• AI goods: Demand for the hardware behind the buildout is beating supply, and those costs are starting to show up in other products.

He also said those shocks have not yet spilled into broader, lasting inflation, that housing services have slowed, and that the labor market is not adding pressure. Barr looks at the same mix and wants more restriction now. Goldman is underwriting Williams.

Sixteen Officials Still Pencil In Another Hike

On September 16 the FOMC raised the federal funds target range by 25 basis points to 3.75% to 4%, the first increase since 2023 and the first under Chair Kevin Warsh, who took office in late May. The vote was 12-0. The statement said inflation remains elevated and that the move would support a timelier return to 2%.

The projections released that afternoon are the sheet Goldman is fading. 16 of 18 officials who submitted a funds-rate forecast saw at least one more quarter-point increase in 2026. Warsh, as in June, did not put in a dot. The median appropriate rate was 4.1% at the end of 2026 and again at the end of 2027. Median 2026 inflation was 3.7% on headline PCE and 3.4% on core. Median GDP was 2.3%. The jobless rate was 4.1%. Officials did not get inflation back to 2% until 2029.

WHERE GOLDMAN AND THE SEPTEMBER DOTS DIVERGE

Kaplan put a second move at 4% to 4.25% near his read of nominal neutral. He also said the interest-sensitive parts of the economy, housing and autos among them, are already sluggish, while AI capital spending and defense are not going to slow on the funds rate. Oil near $100 a barrel, against the $60s a year earlier, and a 10-year Treasury yield above 5% are the other weights on that call.

A December increase would complete the median 2026 dot. Dropping it would leave the September hike as the only tightening of the year, with core still at 3.0% on the new books.

Prediction Markets Flip Toward a Hold

Traders had priced an October hike near 70% in the days before Williams spoke. After his speech those odds fell toward a coin flip. After the PCE report they fell again. On October 1, Polymarket put no change at the October 27-28 meeting at 65.5% and a 25 basis-point increase at 33.5%.

OCTOBER AND DECEMBER ON POLYMARKET, OCTOBER 1

• October hold: 65.5% for no change at the October 27-28 meeting.
• October hike: 33.5% for a 25 basis-point increase, down from about 70% before Williams.
• December hike: 73.5% for a 25 basis-point increase, which still matches Goldman’s dated base case.
• Bitcoin: The coin traded above $85,000 on September 30 as the October hike came off.

The market is not buying Goldman’s overlay that the second hike itself may vanish. It is buying the delay. That is why December still sits as the favorite even as October has become the long shot. Risk assets treated the repricing as a pause in tightening, not as a Fed that has finished the job.

The awkward sequence is hard to ignore. The committee raised rates on September 16. Two weeks later the preferred inflation gauge printed below forecast, after an accounts rewrite that pulled the July core annual rate down to 3.0% as well. Some of the “softness” is how the bureau now prices non-market services. Some of it is the month. Spending of 0.9% is the part of the report that still looks like demand.

Payrolls on October 2 Can Still Undo the Shift

Goldman’s October call is a forecast, not a lock. The committee still has a jobs report, a CPI report, and another PCE print before it sits on October 27-28. A strong labor market was the caveat in the same note that killed the October hike: if hiring stays firm, officials can argue they are in position to go again.

THE DATA STILL AHEAD OF THE NEXT TWO MEETINGS

1. September 16, 2026: FOMC raises the funds rate 25 basis points to 3.75% to 4% in a 12-0 vote, the first hike since 2023.
2. September 29, 2026: Williams says there is no urgency; Barr, in Detroit, says more tightening is likely.
3. September 30, 2026: BEA prints 3.0% core PCE; Goldman moves the second hike to December.
4. October 2, 2026: September employment report, the first labor test of Goldman’s new calendar.
5. October 27-28, 2026: FOMC meeting Goldman now expects to be a hold.
6. October 29, 2026: September personal income and outlays, the next PCE reading, due after that meeting.

If payrolls on October 2 come in hot, Barr’s case gets easier and Goldman’s October skip gets harder to defend into the meeting. If they cool, Williams’s “totality of the data” line gets another month of cover, and the December decision becomes a fight over whether 3.0% core, on rewritten books, is close enough to wait.

Hatzius has dated the next hike and, in the same sentence, offered clients a way for that hike not to happen. The September dots still sit at 4.1%. The 0.4-point gap between Goldman’s 3.0% core path and the committee’s 3.4% is the whole argument. The next prints decide which number the Fed believes.

Disclaimer: This article is news reporting and analysis of Federal Reserve policy, bank forecasts, inflation data, and market prices, and it is for information only. It is not investment advice, a recommendation to buy or sell any security, cryptocurrency, or futures contract, or a prediction of what the FOMC will do in October or December. Readers should consult a licensed financial adviser or investment professional before making decisions that depend on interest rates, inflation, or asset prices. Figures, probabilities, and policy forecasts reflect the cited sources as of the dates given in the piece and will change with new data and new FOMC decisions.