The Wall Street giant dropped its call for unchanged rates after August inflation data and futures markets moved to 87 percent odds of a hike
Goldman Sachs has abandoned its forecast that the Federal Reserve would leave interest rates unchanged at this week’s policy meeting, and now expects a 25-basis-point increase when the Federal Open Market Committee announces its decision on Sept. 16 in Washington. The revision, reported by CoinDesk on Sept. 13, lands two days before one of the most consequential rate decisions of the year for risk assets — including Bitcoin, which traded near 77,000 USD as the meeting approached.
From hold to hike in one CPI report
The pivot followed Friday’s August Consumer Price Index release. Headline inflation rose 0.4 percent for the month, keeping the annual rate unchanged at 3.4 percent, while core CPI — which strips out food and energy — increased 0.3 percent monthly even as its annual rate eased to 2.4 percent, the lowest in five years. According to the research note cited by CoinDesk, Goldman’s economists said the report triggered only a small revision to their core Personal Consumption Expenditures estimate, which they nudged up to 0.26 percent for the month.
“The report has not changed our fundamental inflation view,” Goldman told clients. The bank’s argument was less about inflation accelerating and more about market mechanics: with interest-rate futures assigning nearly 90 percent probability to an increase, the bank judged that holding rates unchanged could provoke a sharp adverse response in markets already priced for tightening.
Futures markets have already voted
The pricing backdrop shifted dramatically around the CPI release. Interest-rate futures put the probability of a September hike at 87 percent after the data, up from 72 percent one day earlier, according to the Wall Street Journal. The implied probability of at least one increase by year-end reached 97 percent. A quarter-point move would lift the federal funds target range from 3.50 percent to 3.75 percent up to 3.75 percent to 4.00 percent — the first increase of this cycle’s resumption.
The shift has wrong-footed plenty of forecasters. A Reuters poll completed before the CPI release found most economists still expected the Fed to hold rates steady through the rest of 2026, though the survey noted confidence eroding as more analysts began anticipating at least one hike. Goldman was in the majority camp until Friday’s data landed.
The inflation picture is genuinely mixed
Strip out the market noise and the August report cuts both ways. The Bureau of Labor Statistics said energy costs rose 16.3 percent over the twelve months through August while food gained 2.7 percent. Monthly increases showed up in communication services, lodging, airline fares, education and used vehicles; medical care and motor vehicle insurance declined.
KPMG chief economist Diane Swonk argued the service-sector details remained uncomfortable for policymakers despite the friendlier annual core reading. She estimated services excluding housing rose 0.5 percent in August and 3 percent year over year. “The gains were heavily in services,” Swonk said, and she now expects three rate increases by early 2027 — raising the possibility, in her view, of a unanimous September vote.
Not everyone is convinced the data justify a hike
The dissent is pointed. James Thorne, chief market strategist at Wellington-Altus, questioned whether the economic data justify the wave of revised forecasts at all. “No material change in inflation outlook, but a hike to calm Wall Street,” he said — an interpretation of Goldman’s reasoning in which the Fed’s motivation would be market credibility rather than price stability.
Thorne pointed to annual wage growth of 3.1 percent and said he sees no verified wage-price spiral. Higher borrowing costs, he argued, cannot expand oil production or repair supply disruptions; rate increases can only reduce demand, investment and household purchasing power. The counterpoint from the hike camp is that energy-driven inflation at 16.3 percent annual growth risks un-anchoring expectations precisely when services remain sticky.
Why crypto holders are watching closely
The decision arrives at an awkward moment for digital assets. Bitcoin spent the week oscillating around the 77,000 USD level, digesting a stretch of ETF outflows alongside the repricing in rate expectations. Tighter policy generally pressures long-duration and risk assets by raising the opportunity cost of holding them, and the September meeting adds a fresh layer of volatility risk on top of an already tense week for crypto markets.
The FOMC will publish its decision, updated economic projections and policy statement at 2 p.m. Eastern Time on Sept. 16, followed by a press conference at 2:30 p.m. The projections — and particularly the dot plot — may matter more than the decision itself: with markets nearly certain of a hike, the forward path into 2027 is where the real information, and the real volatility, is likely to sit.
Goldman’s revised call is a forecast, not a certainty. Futures pricing reflects market expectations rather than any commitment from policymakers, and the Fed has not signaled its hand. But when one of Wall Street’s most influential houses flips from hold to hike on the eve of the meeting, the message to risk markets is unambiguous: the burden of proof has shifted to those still expecting the Fed to wait.
Sept 16 decision and goldman flips Sept 13, three days out. at some point these notes are just the desk telling you what its book needs
87% odds in futures and goldman flipping the same week, this hike is basically priced. real question is what powell signals for november
^ priced in means nothing if they surprise with 50bp
25bp on top of core cpi at 2.4%, a five year low, means they are fighting asset prices not inflation. november dots tell us if this is a one off or a new leg
goldman basically admitted the call is about market mechanics. futures at 87% for a hike and you forecast a hold, you torch your own credibility if youre right
exactly. forecasting a hold at 87 percent market odds is career suicide even if youre right. the note is risk management dressed as analysis
right, being wrong with the herd keeps your job, being right alone on a hold call just makes you the guy who cost clients the carry. the note is pure survival math
you forecast a hold at 87% hike odds, get it right, and clients fire you for the drawdown. the game punishes honesty
core cpi at 2.4%, lowest in five years, and they hike anyway. monetary policy is vibes at this point
Their own note says the cpi report barely moved their inflation view. So the hike call is the fed reacting to pricing, the tail wagging the dog
BTC sitting near 77k going into this. i still have ptsd from the last hike cycle, brace for volatility either way
btc at 77k walking into a 25bp hike is rough. first increase of the resumed cycle too
last hike cycle btc bottomed like two weeks after the first hike actually. the anticipation hurts more than the 25bp itself
core cpi at 2.4% is a five year low and they are still hiking. the inflation story stopped being the reason a while ago
77k held way better than i expected through that cpi print. honestly the dot plot scares me more than the 25bp
goldman flipping two days before fomc is peak wall street. that note to clients is them closing positions first imo
every desk flips the same week futures print 87%. if goldman was closing its own positions first, the note already did its job
goldman flipping plus 87% futures two days out means everyone is front running the same obvious trade. the actual surprise would be a hold with a dovish powell
a dovish hold surprise would nuke every front runner at once. part of me wants to see it purely for the chaos
a dovish hold at 87 percent hike odds would be the most violent squeeze of the year and powell knows it. my money is on the boring 25bp plus a pause signal