Fed holds at 3.50-3.75% in July 2026 — but three voted to hike

The Federal Open Market Committee held the target range for the federal funds rate at 3-1/2 to 3-3/4 percent on Wednesday, July 29 — but the vote was 9-3, and all three dissenters wanted a quarter-point increase. Beth Hammack, Neel Kashkari and Lorie Logan each preferred raising the range to 3.75%-4.00%. After June’s unanimous 12-0 hold, that is the loudest thing a statement without a dot plot could have said: the case for a hike that lived quietly in the June minutes now has three named votes attached to it.

The vote mattered more than usual precisely because this meeting was missing its usual headline generator. July is not a projections meeting — no Summary of Economic Projections, no dot plot; the next set arrives with the September 15-16 decision. In June, what moved markets was not the unanimous hold but the median 2026 dot jumping to 3.8% from 3.4% in March. This time there were no dots to read, so the signal migrated to the vote line, and the vote line delivered.

What the Committee decided, in one paragraph

The target range stays at 3.50%-3.75%, where it has sat all year. The interest rate on reserve balances stays at 3.65%, effective July 30, and the discount window’s primary credit rate stays at 3.75%. A cut was never seriously in play — the live choice was between holding and hiking, and three of twelve voters chose the hike. As of July 23, the effective federal funds rate was 3.63% and the bank prime loan rate — the benchmark most variable credit card APRs are built on — was 6.75%, the top of the target range plus three percentage points. Because the range did not move, neither does any of that machinery: nothing about Wednesday changes what your accounts pay or charge this month.

The division the minutes hinted at is now on the record

The June vote was 12-0, which read like consensus. The minutes of that meeting told a different story: participants judged that “upside risks to price stability remained elevated while downside risks to achieving maximum employment had moderated a bit” — in plain terms, more worried about inflation than about jobs. And there was this passage: “A few participants commented that, in light of these developments, there was a case for raising the target range for the federal funds rate, but those participants indicated that they supported maintaining the current target range at this meeting.”

In July, those participants stopped settling. Three dissents in favor of a hike is a materially stronger signal than the June minutes’ even split on where rates should end the year — “many participants” seeing the appropriate level within or slightly below the current range against “many other participants” seeing it above. A committee that was privately divided is now publicly divided, and that reframes September: the meeting with the projections is where this argument has to resolve into numbers. Our FOMC minutes reading framework covers the vocabulary; the dot plot guide explains why the September SEP now carries the weight.

The minutes also name what the Committee thinks is driving prices: “lingering effects of tariffs, supply chain disruptions related to the closure of the Strait of Hormuz, and strength in demand” tied to AI-related investment. That mix matters because supply-driven inflation responds poorly to rate hikes while demand-driven inflation responds well, and the Committee is still trying to work out how much of the current run is which — Wednesday’s statement kept the supply-shock language, now naming energy specifically, so the question remains open.

The four things we said to watch, and how each resolved

Before the meeting, we flagged four places the signal would live in the absence of a dot plot. Here is the line-by-line.

The wording: no retreat on inflation. June called inflation “elevated relative to the Committee’s 2 percent goal, in part reflecting supply shocks.” July keeps “elevated relative to the Committee’s 2 percent goal” and keeps the supply-shock language — now pointing at “supply shocks that have driven price increases in certain sectors, including energy.” The caveat we said to watch did not get dropped: the Committee still treats part of the current inflation as supply-driven, which is the part rate hikes fix least well. On activity, the language held steady at “expanding at a solid pace despite elevated uncertainty.”

The vote line: this is where the news was. We wrote that a single hawkish dissent would be a stronger signal than any adjective in the statement. There were three — Hammack, Kashkari and Logan, each preferring a quarter-point increase. The statement’s adjectives barely moved; the vote did.

The implementation note: quiet, as usual. The interest rate on reserve balances was maintained at 3.65%, effective July 30, and the primary credit rate at 3.75%. No technical adjustments, no change to how the balance sheet is being managed — Treasury principal continues to be rolled over at auction.

The press conference. Warsh spoke half an hour after the statement, with the transcript to follow on the Fed’s site. With three dissents on the record, the question that matters — how he characterizes September, the meeting where the disagreement has to resolve into projections — matters even more than it did going in.

The data the Fed did not have — and what it showed the next morning

Here is the detail almost no one mentioned. The Fed does not target the Consumer Price Index; it targets the PCE price index published by the Bureau of Economic Analysis. Going into Wednesday, the two measures were telling noticeably different stories.

June CPI, released July 14, came in soft: headline fell 0.4% on the month and cooled to 3.5% year over year from 4.2%, with core flat on the month at 2.6% annually, largely because energy prices dropped 4.9% in a single month. That is the number that generated “inflation is cooling” headlines. But the most recent PCE reading available to the Committee — May’s — had headline PCE at 4.1% year over year and core at 3.4%, a full point hotter than the headlines, and nowhere near a 2% target.

The June PCE report landed on July 30 at 8:30 AM — roughly eighteen hours after the decision. It cooled: headline PCE fell 0.1% on the month and eased to 3.7% year over year from 4.1%, while core rose 0.1% monthly and ticked down to 3.3% from 3.4%. Read against the vote, that cuts two ways. The three dissenters pushed for a hike without seeing a report that moved in the doves’ direction — and the nine who held got a data point in their favor the next morning. But one month of easing, with core still 1.3 points above target and the statement still blaming supply shocks in energy for part of the gap, is nothing like a resolved argument. The CPI-PCE gap narrowed from a full point to two tenths; the distance to 2% barely moved.

The labor picture may also have shifted underneath the June minutes, which said “payroll employment gains had strengthened this year.” The June employment report tells a cooler story: unemployment at 4.2%, down from 4.3% in May, but preliminary payroll levels implying a gain of roughly 57,000 jobs over the month, against implied gains near 129,000 in May and 148,000 in April. One soft month is not a trend, and a falling unemployment rate cuts the other way. Still, a Committee that hikes into a decelerating labor market is taking a different kind of risk than the minutes describe.

The arithmetic that should actually change your behavior

This is where the coverage usually goes wrong, so let us do the math instead of the adjectives.

A quarter point is worth less than you think. On a $25,000 emergency fund, a 0.25 percentage point move is $62.50 a year before tax. That is the entire direct consumer stake in Wednesday’s decision for a typical saver.

Your choice of bank is worth roughly twelve times more. According to FDIC data published July 20, the national average savings account pays 0.38%. The federal funds rate is 3.50%-3.75%. A saver earning the national average is collecting less than one-ninth of what the policy rate makes possible. Closing that gap on the same $25,000 is worth about $780 a year — more than twelve times what a quarter-point Fed move delivers. We track that distance month by month, with the cost broken out by balance, in the savings rate gap. The Fed decision is a headline; the account you keep the money in is the decision that compounds. If your cash sits with a broker, the brokerage cash sweep default is usually the worst-paying option on the menu and takes one form to fix.

The deposit market is already pricing something the Fed has not said. Look at the same FDIC table across maturities: 12-month CDs average 1.68%, 24-month 1.56%, and 60-month 1.36%. Banks are paying less to lock your money for five years than for one. An inverted deposit curve like that is the banking system quietly saying it does not expect today’s rates to last the decade — and Wednesday’s hold did nothing to change its mind. Practically, it argues against locking long right now and in favor of laddering the short end — the mechanics are in our CD ladder guide, and the HYSA vs CD break-even calculator will tell you where the trade actually pays for your numbers.

For borrowers, the Fed is not your main variable. The hold keeps prime at 6.75%. If September delivers the quarter-point hike the three dissenters wanted, prime goes to 7.00% and variable card APRs follow within a statement cycle or two — on a $5,000 revolving balance, about $12.50 a year. Meanwhile the APR itself is likely somewhere between 20% and 30%. The Fed is a rounding error against your rate and your balance; paying the balance down is the move, in either direction. For mortgages, Wednesday’s hold removes any argument for waiting on refinance relief — lender spreads, not the policy rate, are where the savings live, which is what our mortgage shopping guide is built around.

What to watch between now and September

The September 15-16 meeting is the next one with a Summary of Economic Projections, and after a 9-3 vote, that is the meeting where the disagreement has to resolve into numbers — three dissenters will have to put dots where their votes were. The first piece of evidence, June PCE, arrived the morning after the decision and favored the holders. Between now and September the Committee also gets the July employment report in early August, July CPI in mid-August, July PCE on August 26, and then the August employment report and August CPI before it convenes. That is a lot of data for a committee that has now shown us, on the record, how divided it is.

For savers, the practical posture is unchanged and slightly reassuring: no cut came this week — three votes leaned the other way entirely — so the high-rate window stays open, and the case for grabbing yield is about which account you use rather than about timing the Fed. If and when the turn does come, accounts reprice at very different speeds — sticky versus fast-mover HYSAs is worth reading before you need it, not after. And for context on how the cooler June inflation print filtered into savings yields, see our June CPI savings analysis.

Common questions about this meeting

Is there a dot plot at the July 2026 Fed meeting?

No. The Federal Reserve publishes its Summary of Economic Projections — the document containing the dot plot — at only four of the eight annual FOMC meetings. In 2026 those are March, June, September and December. July 28-29 is not a projections meeting, so the next set of dots arrives with the September 15-16 decision.

Who dissented at the July 2026 FOMC meeting?

Beth Hammack, Neel Kashkari and Lorie Logan. All three voted against the hold and preferred raising the target range by a quarter point, to 3.75%-4.00%. The decision passed 9-3. June’s decision, by comparison, was unanimous at 12-0.

Does the Fed target CPI or PCE inflation?

The PCE price index, published by the Bureau of Economic Analysis, not the more widely reported Consumer Price Index from the Bureau of Labor Statistics. The two can diverge meaningfully: June CPI cooled to 3.5% year over year while June PCE, released July 30, came in at 3.7%, with core PCE at 3.3%. The gap narrowed from the full point it had been in May, but the Committee’s preferred gauge still runs hotter than the one in the headlines — and core at 3.3% remains well above the 2% target.

What time was the Fed decision announced?

The statement was released at 2:00 PM Eastern on Wednesday, July 29 — the second day of the meeting — with Chair Warsh’s press conference at 2:30 PM Eastern. That is the standard schedule for every FOMC decision.

Did the Fed cut rates in July 2026?

No — and a cut was never the live question. The Committee held the target range at 3.50%-3.75%, with the only pressure running in the opposite direction: three of the twelve voters preferred a quarter-point increase. The last time the range moved remains the story of 2026 so far — it has been parked at 3.50%-3.75% all year.

How much does a quarter-point move actually change what I earn?

Less than most coverage implies. On a $25,000 balance, 0.25 percentage points is $62.50 a year before tax. By comparison, moving that same balance from an account paying the 0.38% national average to one paying something close to the policy rate is worth roughly $780 a year. The account you choose matters about twelve times more than the decision the committee makes.

Updated July 30, 2026, with the decision, the 9-3 vote, the statement comparison against June, and the June PCE figures released the morning after the meeting.