US CPI Hits 3.7% But Inflation Breadth Falls to 40%

US consumer prices rose 3.7% in the year to September, up from 3.4% in August and a touch above expectations of 3.6%, according to Wednesday's Consumer Price Index report. Prices climbed 0.6% on the month, in line with forecasts. Core inflation, which strips out volatile food and energy prices, rose to 2.6% from 2.4%, slightly above the 2.5% expected, with a monthly gain of around 0.3%.

Context — why inflation breadth matters more than the headline rate

On the surface, that is not the direction the Federal Reserve wants. Both headline and core inflation accelerated on an annual basis, and both printed above consensus. A 3.7% headline rate sits well above the Fed's long-run target, and the monthly 0.6% gain shows price pressure did not cool between August and September.

Bloomberg Economics read the report as less troubling than the headline numbers suggest. Anna Wong, its chief US economist, highlighted a measure of how widespread inflation is, rather than how fast it is running on average, and noted that it fell in September. That is a different question from the one the headline rate answers.

The gauge is built in a similar way to an inflation breadth measure based on PCE, the Fed's preferred inflation index, which Wong linked to Fed chair Kevin Warsh. That link matters because Warsh has signalled a preference for distribution-based inflation measures over simple averages.

The historical range gives the latest reading its context. During the 2022 inflation surge, roughly 80% of CPI components were rising faster than 3%, meaning price pressure was almost everywhere. Through much of the 2010s, when inflation was subdued, the share sat closer to 20%. September's reading sits between those two regimes, closer to the calm end than the surge.

Data — what the numbers show

The Consumer Price Index is built from hundreds of separate price categories, from rent and car insurance to eggs and airfares. The headline inflation rate is effectively a weighted average of all of them. A diffusion, or breadth, measure takes a different approach: instead of averaging, it counts.

Bloomberg Economics' version asks what share of CPI components are rising faster than 3% a year. It tracks that share over two windows. One uses price changes over the past 12 months. The other uses the past six months, annualised, meaning the six-month change is scaled up to a yearly rate. The six-month version reacts faster, so it often shows a turn first.

Measure	Latest	Earlier this year	2022 surge	2010s norm
Six-month breadth	~40%	~55%	~80%	~20%
12-month breadth	Eased by less	—	~80%	~20%

The six-month share has now dropped to around 40%, down from roughly 55% earlier this year. The 12-month measure has eased by less. That gap is the signal: the faster window is turning before the slower one, which is what a genuine narrowing in price pressure looks like early in the process.

The headline and core numbers, by contrast, both moved the other way. Headline rose to 3.7% from 3.4%, core to 2.6% from 2.4%. One month of divergence between the average and the breadth measure is not confirmation of anything on its own.

Analysis — what it means for markets and sectors

An average can be pulled up sharply by a handful of large items, such as a jump in energy or insurance costs. That is a very different situation from one where most prices are rising quickly. If fewer categories are running hot, inflation looks narrower and less entrenched. That is generally more comforting for the Fed, because narrow price shocks tend to fade, while broad-based inflation is harder to bring down.

The approach fits with how Warsh has talked about inflation. At his confirmation hearing he called PCE an imprecise tool and favoured trimmed-mean and median measures, which strip out the most extreme price moves to get at the underlying trend. A chair who prefers those tools is more likely to weight a breadth gauge than a single headline print.

For sector exposure, the read-through runs through rate expectations rather than any single company. If the Fed treats September's print as narrow, the pressure to tighten further eases, which supports duration-sensitive assets and rate-sensitive equity sectors. If it treats the print as broad, the opposite holds. The report does not name specific tickers, and no single issuer's guidance was cited.

The counter-argument is straightforward. Breadth tells you how widespread price rises are, not how large they are. A few big price increases in heavily weighted categories can still keep headline inflation high even as the share of hot components falls. One month's drop in a fast-moving six-month measure is also not a trend. If the share rebounds in the next few reports, the reassurance would fade quickly.

Positioning reflects that uncertainty. Desks that lean on breadth data are likely to treat September as a reason to fade the hawkish read of the headline. Desks that trade the print itself have the opposite incentive. Neither side has a confirmed trend yet.

Outlook — what to watch next

Watch whether the six-month breadth measure keeps falling in the next CPI report, and whether the 12-month line follows it lower. A second consecutive decline in the faster window would turn a single reading into a pattern, which is the threshold that matters for the Fed's reaction function.

The September PCE data will show whether the Fed's preferred index tells the same story. If the PCE-based breadth measure confirms the CPI-based one, the case for reading September as narrow strengthens considerably.

Listen to how Warsh and other Fed officials describe inflation. Language about price pressures becoming less broad would suggest this kind of measure is shaping their thinking. That language shift, not any single number, is the clearest signal available.

Frequently Asked Questions

What is inflation diffusion or breadth?

Diffusion measures count how many price categories are rising quickly, rather than averaging how fast they rise. Bloomberg Economics' version asks what share of CPI components are up more than 3% a year, tracked over both 12-month and six-month annualised windows. During the 2022 surge, roughly 80% of components cleared that bar. In the subdued 2010s, the share sat near 20%. September's six-month reading was around 40%, down from roughly 55% earlier this year.

Why did core inflation rise if price pressure is narrowing?

Core inflation strips out volatile food and energy prices and is a weighted average, so it can rise even when fewer individual categories are running hot. September's core rate rose to 2.6% from 2.4%, above the 2.5% expected. A small number of heavily weighted categories can lift an average while the broader count of hot components falls. Breadth and level measure different things and can move in opposite directions for a period.

What does this mean for retail investors?

Nothing here is investment advice, but the practical takeaway is that when the headline CPI number looks uncomfortable, it is worth asking how many prices are actually driving it. A headline driven by a handful of items behaves differently from one driven by most categories. Retail investors watching rate-sensitive holdings should track the six-month breadth measure alongside the headline rate, and treat one month's move as noise until the next report confirms it.

Bottom Line

September's CPI looks hot on the surface, but a falling breadth gauge gives the Fed a reason to look past it.