The deceptive softness in America's producer-price report

The producer-price report published on September 10th contained a line that read as relief. The core gauge of wholesale inflation, which strips out volatile food and energy, rose 0.2% in August, less than the 0.3% that analysts had forecast. For anyone hoping the Federal Reserve was near an end to tightening, that looks like progress. It is the least informative number in the release.

The headline index of prices charged by American producers rose 0.4% on the month, in line with forecasts, and 5.4% over the past year — a tenth of a percentage point above the expected 5.2%, and up from 4.8% in July, which was itself revised higher. The gap between a soft core and a hot headline is not a statistical quirk to shrug off. It is a map of where inflation now comes from: not from broad domestic demand, which a cooling core would signal, but from shocks hitting the least controllable parts of the price system.

Energy did the lifting, up 4.2% on the month as diesel prices jumped 24.1%, with oil trading above $100 a barrel after the renewed hostilities between America and Iran. Portfolio-management fees, a barometer of the frothier corners of finance, fell sharply on the month yet remained almost a fifth higher than a year earlier. The pattern is telling. This is inflation forced on the economy from outside — by conflict in the oil market, by tariffs that firms pass through rather than absorb, and by the ravenous demand of data-centre construction for electricity and commodities — rather than an overheating that spending restraint could tame.

That distinction matters more than the monthly wobble, because it frames the decision the Federal Reserve faces at its meeting on September 15-16. Policy is already restrictive in name: the benchmark rate sits in a range of 3.50-3.75%, and investors, pricing with the CME's FedWatch tool, put a roughly 62% probability on a quarter-point increase next week. The case for holding rests on the soft core, the argument that the Fed should look through an energy spike. The case for hiking rests on the rest of the report and on the central bank's own credibility. Inflation has now run above the 2% target for more than five years; the Fed's chairman, Kevin Warsh, used a hawkish speech at Jackson Hole to limit his own room to waver. For an institution that spent 2025 vowing to restore its authority over prices, another meeting with prices accelerating is politically and institutionally costly. The soft core gives it an excuse to pause; the headline gives it a reason not to.

For the ordinary investor the temptation is to read the single benign number and conclude that the worst is over. The trouble is that markets did not read it that way at all: equities turned lower and Treasury yields moved sharply higher on the report, the reaction of investors who saw confirmation that rates will stay high, not relief. That is the mechanism that should govern portfolio thinking. When the Fed is raising into an inflation regime that will not die, the discount rate applied to future earnings stays elevated. Long-duration assets — the growth stocks whose value rests on profits far in the future, and long bonds — are the most exposed, even when a monthly statistic flatters. Companies passing tariffs and energy costs through to customers protect margins in the short run while handing consumers the bill, which is precisely how a trading shock becomes a persistent one.

There is honest uncertainty here. The government changed how it measures some services prices from August, and economists at Morgan Stanley note the shift muddies how cleanly PPI translates into the Fed's preferred PCE gauge; the Bureau of Economic Analysis will publish revised PCE figures on September 30th, just after the vote, giving the October meeting the clearer inflation picture. What would change the reading is not next month's core print but the sources of the shock: oil back down, tariffs unwound, energy stabilising for several months. Until then, the disinflationary story of the past year has been overtaken by something older and more stubborn. A price index rising above target for half a decade, fed by war and by trade barriers, does not stop being a problem because one stripped-down gauge misses a forecast by a tenth of a point.