Fed Chair Kevin Warsh recently argued for alternative ways of looking at inflation, such as trimmed means or medians, which, in his words, capture the “underlying inflation rate” and “not … the one-time change in prices because of a change in geopolitics or a change in beef.”Before evaluating his claims, it is useful to explore the distribution of inflation and see what appears to be driving inflation. Headline inflation has been above the Fed’s 2% target for more than five years, and concerns are mounting that inflation could be persistently elevated unless the Fed tightens monetary policy.During the past twelve months, PCE prices have risen 3.8% (dark blue line), and over the past six months, that pace has accelerated to 4.8% at an annual rate (light blue line). Headline inflation is running well above the Fed’s 2% target, and the momentum is pushing inflation rapidly away from target. A critical question is whether the current pickup resembles the early stages of pandemic-era inflation or simply the unfortunate layering of multiple one-off price increases due to tariffs and higher energy prices.Looking Under the HeadlineAverage inflation—the headline number reported—is elevated, but the distribution of price changes suggests that the problem is relatively narrow. Behind aggregate PCE inflation are prices for 177 categories of goods and services. The categories vary in their share of total expenditure. For example, gasoline is 2% of total consumer expenditure, dining out is 5%, and owner-occupied housing is 11%. Putting that information together, we can then ask the following questions: What share of consumer spending saw prices fall? What share saw prices rise 5% or more? And then sort everything in between into buckets. The charts below do this using the annualized six-month change to capture the recent acceleration. The period from 1995 to 2005, when inflation averaged close to the 2% target, serves as a benchmark.PCE inflation is elevated for two main reasons: 1) there are too few categories with price declines (top panel), and 2) there are too many categories with 5% or more price increases (bottom panel). Other parts of the distribution have largely normalized since the pandemic surge in inflation. The shortfall at the bottom began to open in early 2025, while the excess at the top is a more recent phenomenon.Core goods and non-housing core services are critical to explaining the top and bottom of the distribution relative to the 2%-consistent benchmark. The current composition of price changes at 5% or more shows that two-thirds of the excess contribution comes from non-housing core services (blue bars) and one-third is from core goods (red bars). Energy is a rising contributor in the top inflation bucket, driven by the surge in gasoline prices since the start of the Middle East conflict, but it remains close to the benchmark contribution to energy. In the pandemic, the inflation surge began with a similar composition, but housing inflation was a key and lasting contributor to the excess. Housing is no longer contributing to the top bucket of inflation.The role of core goods in explaining the lack of declining prices is clear. During the pandemic, there was substantially less deflation in core goods than usual. Protracted supply chain disruptions led to sizeable increases in goods prices. The anomaly had largely resolved by late 2024, but then it reopened to a magnitude similar to that of the pandemic. The pattern at both the top and the bottom of the distribution has clear signs of tariffs. The large increase in tariff rates in 2025 has filtered through to goods prices. While the assumption is that these are one-time adjustments to price levels, the adjustment process has stretched over several months. The timing of price increases has varied across categories, too.Looking at the recent change in prices of specific core goods compared with six months ago reinforces the role of tariffs. Several items in women’s and girls’ clothing, footwear, and luggage have experienced significant price increases over the past six months and are highly exposed to tariffs. Price changes have slowed dramatically for other core goods that had risen earlier, such as furniture and major household appliances. These are consistent with the passthrough being largely complete and with a return to disinflation. Tariffs are not the only driver of core goods. Sharply higher prices for computer software and consumer electronics likely reflect strong AI-related demand and semiconductor shortages.The rising contribution of non-housing services to the top inflation bucket also highlights how higher energy prices bleed into core inflation. Over the past six months, price increases in several energy-intensive service categories such as airfares, transit, and taxis have accelerated sharply. Oil prices surged in early March, so these data through April only capture the initial adjustment to the energy price shock. Moreover, most of the outsized increases in this category, like hotels, health insurance, and domestic services, are likely unrelated to the conflict in the Middle East.In ClosingLooking at the full distribution of inflation, in line with Warsh’s comments, offers a less alarming perspective on the current direction of inflation than the headline numbers. Elevated inflation now is about unusual patterns in the tails, not the broad-based heat. Of course, the distribution cannot tell us whether supply shocks like tariffs or energy disruptions will lead to one-off price increases or spill over into a series of price increases, as they did during the pandemic. That will require ongoing monitoring of the data to determine.Bonus Links and a ChartExcellent explainers of the trimmed mean and why one should be careful with it now:“Skewness warrants caution as Trimmed Mean PCE inflation eases.” by Tyler Atkinson, Jim Dolmas, and Rebecca Zarutskie at the Dallas Fed.“So You Want to Talk About Trimmed Mean Inflation?” Vikas Patel and Skanda Amarnath at Employ AmericaDeep dives into the effect of tariffs on inflation. Multiple approaches to the question suggest that tariffs have pushed up consumer prices.“Detecting Tariff Effects on Consumer Prices in Real Time – Part II” by Robert Minton, Madeleine Ray, and Mariano Somale at the Federal Reserve Board.“Tracking the Short-Run Price Impact of U.S. Tariffs” by Alberto Cavallo, Paola Llamas, and Franco M. Vazquez at Harvard.The distribution chart from 1960 to the present includes earlier periods of inflation.Original Post