Last Wednesday the Federal Reserve raised its benchmark rate by a quarter point, to a range of 3.75% to 4%. It is the first increase since July 2023.

The vote was unanimous. In July the committee split 9–3, with three regional presidents dissenting in favor of exactly this move. This time nobody dissented in either direction, despite months of public pressure from the White House to cut.

That is the news. The part that belongs in your plan is underneath it.

Alongside the decision the Fed published its quarterly projections. Twelve of eighteen officials expect the rate to finish 2026 at around 4.1%, meaning one more increase. Four expect two more. Sixteen of eighteen expect at least one.

Read that and the obvious conclusion is that rates are going up again before Christmas. Probably right, and not the important number.

The important number is 2027. The median projection for the end of 2027 is also 4.1%.

Not lower. The same. The Fed’s own committee, in its own published projections, is not forecasting a single cut across the whole of next year. The 2028 median comes down to 3.9%, which is a quarter point over two years.

And the longer-run estimate — the level officials consider neutral once inflation is back at target — was raised from 3.1% to 3.2%. The destination moved further away, not closer.

There is a second line worth putting next to the first. Futures are pricing the funds rate at roughly 4.5% by September 2027, against the Fed’s own 4.1% median.

That gap is unusual and it runs the wrong way for planning purposes. Ordinarily the market prices a faster return to easing than the committee projects, because traders assume the Fed will blink when growth softens. Right now the market is more hawkish than the Fed.

It is also in a hurry. Futures are already putting the odds of another increase at the October meeting at better than even. The argument is no longer whether there is one more this year — the committee and the market agree on that — but whether it arrives five weeks from now or in December. Either way the base rate underneath your year-end close may not be the one in front of you today.

Whatever assumption sits in your 2027 model, there is no external forecast supporting a materially lower rate. Not the committee’s. Not the futures curve.

Most finance functions carrying a high-rate assumption have a second, unstated one behind it: that if rates stay high, something will eventually break, growth will soften, and the Fed will cut in response. That has been a reasonable thing to believe for two years.

The projections released on Wednesday removed the support for it.

The Fed revised its unemployment forecast down to 4.1% for each of 2026, 2027 and 2028, from 4.3%, 4.3% and 4.2% in June. It revised GDP growth up, to 2.3%, 2.4% and 2.2%. And it revised inflation up — headline PCE for this year to 3.7%, core to 3.4%.

So the committee is raising rates into an economy it believes is growing faster and employing more people than it thought three months ago. This is not a central bank tightening against a slowdown it expects to arrive. It is one that thinks the economy can take it.

There is no soft landing embedded in these numbers that rescues a plan built on cuts. There is a fairly strong economy with inflation that has not come down, and a committee that has decided to do something about the second without much fear for the first.

Three things change, and none of them is dramatic on its own.

Your incremental cost of debt. If you are refinancing anything in 2027, the base rate underneath it is now reasonably expected to be where it is today or higher. A plan built on a lower rate has an interest line that will not hold.

Your hurdle rate. Projects approved against a cost of capital assuming 2027 relief are being approved against a number that the Fed and the futures market both now dispute. That is worth revisiting before the capital plan is signed rather than after.

And your covenant headroom. Fixed charge coverage and interest coverage ratios built on a declining rate path have less room than the model shows. The time to find that out is during planning, not at the first test date.

Our take: the useful action this month is narrow and specific. Find the rate assumption in your 2027 plan, write it down, and put the Fed’s 4.1% median and the market’s 4.5% next to it. If your number is materially below both, you are not being conservative or aggressive — you are carrying a forecast that nobody who publishes one currently shares. That may still be the right call. It should be a decision rather than an inheritance from a model built in 2024.

Leadership is asking: are we getting value from AI? Which tools are worth the spend? Where are we exposed? Right now, most teams have no idea.

Harmonic Security Usage Explorer changes that.

You get a complete picture of how your organization uses AI, automatically categorized into custom tasks and use cases.

You’ll see the projects being worked on, who’s using what tools, where AI investments are driving value, and where employees are engaging in risky behavior.

CIOs can rationalize spending and cut wasted licenses. CISOs can pinpoint where risk exists and neutralize it. AI committees can show exactly how their efforts are paying off.

See it in action

In July we asked when the next Fed move would come. Of the 61 of you who voted, 31% said no move at all this year and 25% picked September.

The September group was right. The no-move group is now wrong, and that is the first time this tally has gone against readers since we started keeping it.

Worth being precise about what was and was not predictable. When that poll ran, futures put a September hike at around one-in-three. By late August it was 40%, then 57% after Jackson Hole, 69%, back to 60% on a strong jobs report, 86% after August inflation, and 94.5% the morning of the decision. Anyone holding a firm view in July was holding it without much support.

The readers who said September were closer to right than the market was at the time they voted. That is worth something, even in a week when the tally as a whole was beaten.

Four things every CFO needs to know this week.

· Seven percent. That is the share of finance leaders who describe themselves as very confident in interpreting AI outputs, in research from FERF and CrossCountry Consulting covering 197 executive-level finance leaders. The same study found 64% of organizations running active transformation initiatives, 80% of them citing automation and AI as a driver — and 46% with no formal AI governance structure at all. Deployment is running well ahead of the ability to check the work.

· Revenue growth has climbed from seventh to second. U.S. Bank surveyed 1,000 senior finance leaders at companies above $100 million in revenue. Cutting costs remains the top priority at 39%, up from 33% in mid-2024. But driving revenue growth is now second at 31%, having ranked seventh two years ago. Finance functions that spent three years defending the P&L are being asked to grow it, usually without a corresponding headcount.

· A January deadline most companies are treating as a 2028 problem. FASB’s expense disaggregation standard takes effect for annual periods beginning after December 15, 2026. For a calendar-year filer the disclosure lands in the FY2027 10-K, filed in early 2028 — but the data collection starts on the first of January, and you cannot retroactively disaggregate expenses your general ledger never tagged. That is roughly fourteen weeks away. We covered it in August; the deadline has not moved.

· You were split almost exactly down the middle. Last week we asked whether you would price an IPO before the midterms or wait. Of the 57 who voted, 25 would wait for December and 23 would price now — and only three would wait for 2027. So the real argument in the market is about six weeks, not a year. Anthropic and OpenAI, who have taken opposite sides of that question in public, are arguing over a narrower gap than the coverage suggests.

1. What rate is actually in the model? Not what you assume it is — what the treasury schedule was built with. In most finance functions the number lives in a tab somebody set up eighteen months ago and nobody has revisited. Find it.

2. What happens at 4.5%? Run the interest line at the market’s September 2027 pricing rather than the Fed’s. It is a ten-minute exercise and it tells you whether your plan is sensitive to a half point or indifferent to it.

3. Which covenants are closest? Fixed charge coverage, interest coverage, and any ratio with interest expense in the denominator. Model them at the higher path and find out which one tightens first, and in which quarter.

4. What are you refinancing, and when? Anything maturing in 2027 is now expected to reprice at or above today’s level. If the plan assumed otherwise, the gap is a real number and it belongs in the capital plan rather than in a footnote.

5. What would have to be true for rates to fall? Write it down in one sentence. If the answer is a recession, check whether your revenue plan assumes one — because a plan that carries lower rates and higher growth is carrying two assumptions that do not sit together.

SPONSOR THIS NEWSLETTER

Your name, in front of the CFOs and senior finance leaders who read The CFO Desk every Tuesday. The primary slot is available for Q4. Write to [email protected].

For three years the interesting question about interest rates was when they would come down. Every plan, every model, every refinancing schedule carried some version of an answer to it.

Last Wednesday the Fed published projections showing no cuts next year, a quarter point over two, and a neutral rate further away than it was in June. The market thinks that is too dovish.

The question has changed. The plans mostly have not.

One question before you go: what rate does your 2027 plan assume?

One click. And if you have a view on why, reply — we read everything.

The Editors hold no position in any company mentioned.

— The CFO Desk
[email protected] · thecfodesk.com
CAPITAL · FINANCE · LEADERSHIP