The Fed meets October 27–28, and it just did something it hadn't done since 2023

On September 17, 2026, the Federal Reserve raised its target interest rate for the first time since July 2023 — undoing the last of six consecutive cuts. The next decision arrives on October 28. It comes with no updated forecasts, and there is no November meeting behind it, so whatever the committee decides stands for roughly six weeks.

Start here: what the Fed actually decides

The Fed does not set your mortgage rate, your car loan, or the yield on a 10-year Treasury. It sets one thing: a target range for the overnight rate banks charge each other to borrow cash for a single night. Right now that range is 3.75% to 4.00%, and the rate actually trading inside it — the “effective” rate — is 3.88%.

Everything else follows indirectly. That overnight rate is the price of the safest, shortest loan in the economy, so it anchors what banks pay you for deposits and what they charge for short-term credit. Longer-term rates (mortgages, corporate bonds, the 10-year Treasury) are set by investors guessing where the overnight rate will average over years. That is why the Fed can cut and mortgage rates can rise on the same day: the Fed moved today’s price, and the market changed its mind about the next decade.

Two other things to know before reading any Fed coverage:

• There are eight scheduled meetings a year, each two days long, with the decision published at 2pm Eastern on the second day. The October meeting runs October 27–28.
• Only four of the eight come with forecasts. Four times a year the Fed publishes a Summary of Economic Projections — the famous “dot plot” showing where each official expects rates to go. October is not one of those. December 8–9 is. That means October will be a bare decision plus a statement, giving markets less to chew on and more to argue about.

The September surprise, in one chart

The Fed spent two years cutting. From 5.50% in July 2023, it delivered six reductions totalling 1.75 percentage points, bottoming at 3.75% in December 2025. Then it held for nine months — and in September 2026 it went the other way.

When the 2-year yield sits above the Fed’s own rate, the bond market is not agreeing with the Fed. It is front-running it.

What the committee is staring at: three inflation numbers that disagree

If you read one inflation headline this month you probably saw 3.35%. That is year-over-year CPI for August. It is also only one of three gauges the Fed watches, and right now they point in different directions — which is precisely why October is a genuinely close call.

Why do they disagree? Three reasons worth knowing:

• Different baskets. CPI measures what urban households buy out of pocket. PCE covers a broader set of spending and includes things paid on your behalf — most importantly healthcare bought through insurance. When medical and financial-services prices move, PCE feels it more.
• Different weights. Housing is about a third of CPI and far less of PCE, so a cooling rent market pulls core CPI down faster than core PCE.
• “Core” means excluding food and energy — not because they do not matter, but because they swing so hard that including them tells you about last month’s oil price rather than about underlying inflation. The 4.17% headline peak in May 2026 came mostly from the volatile half.

One more angle that cuts the other way. Look at the recent pace rather than the annual comparison: over the three months to August, CPI ran at an annualised 0.18% and core CPI at 1.97% — at or below target. Core PCE’s three-month pace was 3.05%. The year-over-year figures are still carrying last spring’s spike; the three-month figures say the most recent months were much calmer for consumer prices and still hot for the Fed’s own measure.

And the job market got tighter, not looser

The Fed has two jobs, not one: stable prices and maximum employment. Through late 2025 the second gave it cover to keep cutting — unemployment reached 4.5% in November 2025. It has fallen every few months since.

The bond market has already voted

You do not have to guess what traders expect — you can read it off the 2-year Treasury. A 2-year yield is roughly what investors think the overnight rate will average over the next two years, plus a small premium for tying money up. So compare it to the Fed’s current rate and the gap is the market’s forecast.

What does 0.87 points of gap mean in plain terms? If you ignore the term premium, a 2-year yield that much above today’s rate is consistent with the market expecting several more quarter-point increases over the next couple of years — and pricing a decent chance of the first arriving soon. You should not ignore the term premium entirely: some of that 0.87 is compensation for risk rather than a forecast, which is why this is a direction, not a decimal.

The longer end agrees. The 10-year is at 5.18% and the 10-year minus 2-year spread is a positive 0.31 — no inverted curve, no recession klaxon. Yet 10-year breakeven inflation, what bond investors expect inflation to average over a decade, is just 2.34%. Read together that is a market saying: rates stay high for a while, and the Fed wins in the end.

Three ways October 28 can go

We are not in the business of predicting central banks, and anyone quoting you a precise probability is quoting a market, not a fact. But the cases are clear enough to lay out, along with what each would actually do.

Our read, stated as a judgement rather than a fact: the weight of the evidence leans toward a hike or a hawkish hold, because the two measures the Fed has said it cares most about — core PCE and the labour market — are both pushing the same way, and the absence of a November meeting raises the cost of waiting. The strongest argument against is that core CPI’s recent pace is already at target, which would make another increase a policy mistake visible only in hindsight.

Why it matters for your money

If you have a mortgage to get. The Fed does not set 30-year mortgage rates — the bond market does, and it has already moved. The 10-year Treasury went from 3.97% in February to 5.18% now, which is why borrowing got more expensive long before September’s hike. A hike on October 28 mostly matters through what it says about the next year of rates.

If you have savings. This is where the overnight rate passes through almost directly. Money-market funds and high-yield savings track it within weeks. At a 3.88% effective rate, cash is paying real money for the first time in a generation — and if you have been meaning to move an idle balance out of a 0.01% checking account, the gap has rarely been wider.

If you carry a credit-card balance or have a variable-rate loan. Card APRs are usually the prime rate plus a margin, and prime moves in lockstep with the Fed. A quarter-point hike shows up on your next statement without anybody sending you a letter.

If you own stocks or bonds. Higher-for-longer rates reprice everything that behaves like a bond — utilities, property, dividend payers, and bonds themselves — while cash-rich businesses that need no borrowing are relatively insulated. That split is not theoretical; it is exactly what the past month delivered, documented in the bond-market note. And one of the reasons the Fed’s inflation read is so awkward right now is tariffs, which push prices up once and then drop out of the annual comparison — the subject of the tariff note.

Data appendix

Every figure above with the date it belongs to. The series codes are public and free — each one returns exactly these numbers.

Pulled from FRED (Federal Reserve Bank of St. Louis) and the Federal Reserve’s published meeting calendar on September 26, 2026. Three-month annualised rates are our own arithmetic on the same index levels. For where options-market volatility stands into the meeting, the Radar re-scores a liquid universe on live data, and the market’s weather report explains what it is reading.