A 5.9% Jump in Phone Plans Is Moving the Fed's Next Decision

The most important interest rate in the world may come down to how much you pay for your family's wireless plan. Core inflation in the U.S. rose 0.3% in August, a tenth of a point more than the 0.2% economists expected, and the entire surprise sat in a single category: wireless telephone services, which jumped 5.9% in one month — a record — all by itself adding about a tenth of a percentage point to the core reading.

That is weird only if you expect inflation data to be a clean thermometer. It isn't. The U.S. is in the middle of a genuine question about how to price a category of the economy that simply does not sit still, and the Bureau of Labor Statistics is the referee.

Wireless services count for only about 1.3% of the consumer-price basket, but they are among the hardest things in it to measure. Carriers don't sell one stable object. A plan is a bundle of talk, text, and data, sold in tiers with promotions, grandfathering, and free phones folded in, and providers keep restructuring the menu. The BLS has had to build a statistical machine just to keep comparing one month's plan to the next. Since July 2025 it stopped fielding its old price survey for this category entirely and instead buys web-scraped plan data from a market-research vendor and runs a hedonic regression to estimate what a given set of features is worth to consumers. It does not even quality-adjust the move from 4G to 5G — a new network is treated as a different price, not a better product.

So what actually happened in August? Economists at Bank of America, Barclays, and Pantheon Macroeconomics trace the jump to real carrier behavior: AT&T raised prices on certain plans starting in August, and T-Mobile retired legacy plans, which pushes grandfathering customers onto pricier current offerings. On its face, then, this is not a phantom — people really are being charged more. But whether a legacy-plan retirement counts as "inflation" or as telecom billing semantics is precisely the classification call the BLS gets to make, and the agency declined to say how it treated the moves (the BLS declined to specify how the carrier changes affected the data).

The reason any of this matters for you, right now, is that the print lands exactly on the Federal Reserve's decision boundary. The policy committee meets September 15–16 with the federal-funds rate at 3.50%–3.75%, after a sharply divided July meeting that split 9–3 in favor of holding, with the three dissenters voting to raise (Hammack, Kashkari, and Logan). Chairman Kevin Warsh has sounded hawkish, and markets were already pricing wild-card odds — near 56% for a quarter-point hike — which would be the first hike since July 2023 — before Friday's number arrived. Within that framing, the 0.3% core print is what "sealed the deal," even though almost everyone agrees the phone-plan spike was largely a one-time event. As Omair Sharif of Inflation Insights put it, strip out wireless and core inflation would have printed exactly where economists expected, at 0.20% — but the Fed "will not have the luxury of slicing and dicing the data."

If this feels familiar, it should. The same category once pulled inflation in the opposite direction. In early 2017, when Verizon reintroduced unlimited-data plans, the BLS decided the shift was a quality improvement that lowered the effective price of a plan, and wireless services fell about 7% in a single month — roughly five and a half points of that drop a statistical judgment call, big enough to move annualized CPI around a full percentage point between the low-1% and ~2% readings. The same plumbing, the same incentives of carriers to keep re-pricing the same thing, had made inflation look too cold. This time, with AT&T and T-Mobile doing the re-pricing, it makes inflation look too hot.

Here is the structural point worth keeping, because it is not going away. The one number that steers the entire yield curve — and, through discount rates, the value of virtually every stock you hold — can pivot on how a statistician classifies a T-Mobile legacy-plan retirement. The August beat is not fresh evidence that the economy's momentum suddenly accelerated; it is the market coordinating on a number as printed, and that number has a phone-shaped hole in it. Whatever the committee decides next week, the useful habit to keep is healthy skepticism of single-month inflation beats and misses that rest on this category — and an acknowledgment that when everyone is pricing off a single noisy measurement, the price itself inherits the noise.