The Federal Reserve is losing the war on inflation, and a 25-basis-point hike in December will not win it. The Federal Open Market Committee should raise the federal funds target range by 50 basis points to 4.25–4.50% and say plainly that more may follow.

October’s consumer price index, published on November 10, showed a third straight core increase of 0.3% month on month. Price gains are spreading to more of the basket, and the Fed’s own preferred gauge is heading for 3.3% year on year.

While most market voices are still expecting a 25-basis-point hike in December, with another in January or March 2027, the gap between their diagnosis and their prescription is the problem. When the evidence says inflation is breaking upwards, a timid step is not prudence. It is the same mistake the Fed made in 2021, made again on a smaller scale.

The argument for a 50-basis-point hike at the December FOMC meeting rests on four facts. Inflation is not falling; it is speeding up. It is broadening, not narrowing. The new drivers (hardware costs related to artificial intelligence, freight and diesel pass-through, and a floor under shelter inflation) will not fade on their own.

And at today’s rate, the Fed’s real policy stance is close to neutral, not restrictive. A central bank that is not restrictive while core inflation in the personal consumption expenditures price index is heading for 3.3% year on year is not fighting inflation at all.

Tripple Jump
Core CPI rose 0.296% month on month in October, against a consensus forecast of 0.2%. One should call it a “hat trick of strong prints”, the third consecutive increase of 0.3%. Headline inflation exceeded inflation-swap pricing by 12 basis points, the largest miss since January 2024. The two-month average of upside surprises is the highest since February 2023.

The miss was not confined to one rogue category. Some 68% of 194 CPI components rose in the month, the widest spread since September 2025. Every block of core inflation ran hot at once:

• Goods. Core goods rose for a fourth straight month, even with used cars down 1.2%. Excluding new and used vehicles, core goods jumped 0.45%, the second-largest rise since the pandemic. Furniture rose 1.2%, appliances 1.3%, smartphones 1.4% and TVs 3.3%. Prescription drugs posted their first increase since December 2025.
• Supercore services. Prices rose 0.399%, after 0.422% in September and 0.510% in August. That is roughly a 5% annualised run rate in the part of the basket most tied to wages and demand. Recreation services rose 0.9%, motor insurance 0.7% and hospital services 0.8%.
• Shelter. Owners’ equivalent rent, which was meant to be the reliable disinflation engine, rose 0.32% after 0.35%, and this time the strength was spread across regions.

The doves will point to quirks: a New York ferry fare rise, parking fees, a utility adjustment in owners’ equivalent rent. Strip them out, and the picture barely changes. The resilience in many items suggests a risk of core inflation being stickier than generally thought. When two-thirds of the basket is rising, idiosyncrasy is no longer the explanation.

Fed’s Speeding Preferred Gauge
The Fed targets overall PCE inflation, not CPI inflation, and the translation makes the picture worse, not better. Used cars, which fell, carry less weight in PCE. Prescription drugs, which rose, carry more. Estimates point to a strong October core PCE reading of around 0.34% month-on-month.

That would lift year-on-year core PCE inflation to about 3.3%, the highest since 2023, from 3.12% in September. The momentum is worse than the level. The monthly path runs as follows: 0.262% in August, 0.317% in September and an estimated 0.343% in October, each reading higher than the last.

Three-month annualised core PCE inflation would reach 3.75–3.8%. Core PCE goods are tracking 0.41% month on month, the strongest since February. Supercore PCE, the measure least exposed to supply shocks, is estimated to have held near 0.4% month on month for three months running.

Put simply, underlying inflation is now further above 2% than it was in the spring and moving away from it. The FOMC’s own September median forecast put core PCE inflation at 3.4% in October–December 2026 compared with a year earlier.

Although the formal 2% target applies to overall PCE inflation, a central bank projecting underlying inflation at 1.7 times that level more than five years into an overshoot has not won anything.

Three new fires that will not go out by themselves

The comforting story is that this is a passing supply shock. The research says otherwise. Three separate forces are pushing prices up at the same time, and each has staying power.

1. AI is now an inflation story, not just a growth story. The surge in memory-chip costs earlier this year is still working through to shelves. Smartphones rose 1.4% in October, the largest monthly gain this cycle, as Samsung repriced its flagship line. Market estimates suggest higher memory costs will add 20–40 basis points to core PCE inflation in total, and only about half has passed through so far. Elevated input costs are likely to continue pushing consumer electronics prices higher. The other half is still in the pipeline.
2. Diesel is on a long haul to the consumer. Brent has eased to about $91 a barrel from above $110, but that headline relief is misleading. Distillate prices and refining margins remain elevated. The persistence of high oil and diesel prices matters more for inflation than any near-term dip. Bulky goods with high transport margins, such as furniture, appliances and toys, are exactly the items now rising.
3. Airline fares are up nearly 20% since the end of 2025 and have added about 0.2 percentage points to core CPI inflation this year. Postage and courier fees rose about 2% in October on US Postal Service and FedEx fuel surcharges. The diesel spillover has appeared earlier than most analysts had expected.
4. The shelter brake has failed. For two years, the disinflation case rested on rent. But now rents on continuing leases are rising faster than rents on new leases, and new-lease growth appears to have bottomed out. The estimated underlying run rate is 0.27% month on month for owners’ equivalent rent and 0.23% for rents, roughly 3% annualised. Zillow’s market-rent measure has turned firmer. There is little incremental disinflation to come from shelter.

None of these depends on a mistake in the data. Together, they mean the economy has lost its main disinflationary offset, housing, just as two fresh cost shocks arrive. Demand is not cooling enough to absorb them: both the manufacturing and services indices from the Institute for Supply Management are well above 50, and US data strengthened in October at the fastest pace since May 2021.

Behind The Curve
The FOMC paused in October with the federal funds target range at 3.75–4.00%. Governor Lisa Cook and Cleveland Fed President Beth Hammack dissented in favour of a hike. Recent remarks also point to a hawkish drift:

• Governor Christopher Waller set the test in advance on November 10: if October core CPI came in at around 0.3% or higher, “a 50 basis point hike in the policy rate should be considered at the December meeting.” It came in at 0.296%, which rounds to 0.3%. The test has been met.
• Dallas Fed President Lorie Logan has called for at least two more 25-basis-point increases and says “it may be possible to move more rapidly this winter than the FOMC has moved so far this fall.”
• San Francisco Fed President Mary Daly, long a dove, says the Fed is “more likely than not to need additional adjustments.”
• Vice Chair for Supervision Michelle Bowman now says “there is still work to do.”

So, support for further tightening is broadening. The argument is increasingly about speed, and that is where the Fed keeps getting it wrong. Its default is a steady 25 basis points a meeting or a hike every three months. At that pace, the Fed would reach the level the data already justify only in January or March 2027, depending on the pace of subsequent increases.

That is the pattern of a central bank chasing inflation rather than leading it. Each meeting, it does the minimum the last data release demanded. Each release then shows it did too little. Even cautious forecasters see it: the December debate has already shifted from “zero versus 25 basis points” to “25 versus 50 basis points”. The question for the FOMC is whether it shapes that shift or is dragged along by it.

Why Caution Fails
“The shutdown distorts the numbers.” A few analysts warn that October and November core PCE inflation will be mechanically too strong because of last year’s government shutdown. That affects the year-on-year comparison. It does not explain the month-on-month readings, which are clean, or the 68% breadth, or supercore inflation at 0.4% for three months.

“Three months is noise; six months looks stable.” Six-month services inflation of 3.2% annualised is not a reassuring number. A six-month average is stable partly because it still contains the soft June–July readings. Waiting for the longer average to confirm the shorter one means acting only after the acceleration is locked in.

“Pass-through will fade: oil is down, tech repricing is half done.” Perhaps. But diesel and jet fuel have not fallen with petrol, and half the AI cost shock is still to come. Meanwhile, there is likely to be no further help from shelter. Betting that every shock fades at once is a forecast, not a plan.

“Financial conditions have tightened and housing is softening.” Some cite the recent equity sell-off and weaker housing as reasons for a gradual path. Yet the ISM indices are above 50 and the overall data flow is the firmest since 2021.

With productivity growth above 3%, the neutral rate may be higher than the Fed assumes, so a 4% nominal rate may be too low even in an economy with 2% inflation, let alone one with core inflation at 3%. If neutral is higher, the Fed may not be tight at all.

“The FOMC prefers 25-basis-point steps.” That is a habit, not an argument. Gradualism makes sense only when it is unclear how many hikes are needed. Today, the debate is increasingly about how fast to tighten. Once the direction is clear, the central bank should move policy rates quickly towards equilibrium and slow the pace when it gets close. What is right to be done cannot be done too soon.

50 over 25
A simple test of the degree of restraint is the real federal funds rate, calculated here as the policy rate minus core PCE inflation. On that measure, the Fed appears to be barely leaning against inflation at all.

 Measured against current momentum, the real policy rate is roughly zero. Even after a 25-basis-point hike, it would be below 0.5%. That suggests little restraint, particularly if the neutral real rate has moved higher. A 50-basis-point hike is the minimum needed to get a real rate of more than 1% against the year-on-year number.

Three further reasons favour moving now rather than in two steps:

First, the cost of delay compounds. Every month the real rate sits near zero, inflation expectations have room to drift. Inflation risk premiums fell around the midterms, but that calm will not survive another two hot readings.

Second, it buys optionality. A 50-basis-point move in December gets the Fed to a range of 4.25–4.50% by year-end. The FOMC could then pause in January and March and watch the data from a position of strength, not chase it.

Third, it resets credibility. Waller set a public test, and the data met it. Dismissing it now would tell markets the Fed’s conditional guidance means little. Delivering on it tells them the reaction function is real.

The risk of overdoing it is small by comparison. If inflation does cool next year, the Fed can stop or reverse quickly from a higher base. If it does not, a Fed still at a target-range midpoint of 4.125% in March will be facing a far harder and costlier job.

Move Big, Move Now
The evidence points one way. Core inflation is running at 0.3% a month for a third straight month. Core PCE inflation is heading for 3.3% year on year and nearly 3.8% on a three-month annualised basis. Price rises cover two-thirds of the basket. The AI shock is only half absorbed, diesel costs continue to feed through, and shelter has stopped helping. Against that, the Fed holds a real policy rate of roughly zero on current momentum.

The FOMC has spent this autumn doing just enough to match the last data release, and each new release has shown it was not enough. That is what losing a war on inflation looks like: not a dramatic surge, but a steady drift that policy keeps failing to catch.

In December, the Fed should raise the federal funds target range by 50 basis points to 4.25–4.50%. It should signal that further moves depend on whether the monthly core readings return to a pace consistent with 2% annual inflation, about 0.17%.

And it should lift its interest rate projections for 2027, so markets do not mistake one bold step for the last. A half-point now is the cheaper option. The alternative is a longer, more painful tightening cycle in 2027.