When you get an energy shock, it's going to show up no matter how you calculate a price index. Among our unconventional measures, the ones that filter out volatile or extreme moves, naturally, look less bad. But there's not a one of them that isn't registering some degree of shock. Fine -- that's just a question of calculation. What counts is what it means. A shock is not the same thing as inflation. And the central bank standard playbook for fighting inflation with higher rates isn't appropriate.
US Unconventional CPI
US Unconventional CPI
Friday, April 10, 2026
The unconventional metrics use terms like "flexible" and "sticky" as more sophisticated lines of demarcation between "headline" and "core." But with fuels so dominant in March's statistics, it all works out the same. "Flexible" measures of inflation soared. "Sticky" measures slightly improved.
US Unconventional CPI
Thursday, March 12, 2026
The alternative measures all agree -- we're in a trading range for inflation very near the Fed's target. Remember, it's not 2.5% for CPI. The 2% target is for PCE. But here's what makes it so hard to figure now -- PCE is running above CPI, which almost never happens. Yes, the index with the lower target is running above the one with the higher target! There are basically two differences between the indices -- CPI overweights housing and underweights health care, and it just so happens housing inflation is (finally) abating, but health care inflation is not. But the fact that the analysis comes down to such picayune factors tell us that inflation should not now be a front-and-center issue for the Fed
US Unconventional CPI
Friday, February 13, 2026
The problem with looking at so many unconventional indicators of inflation is you get many divergent analyses, and you're not sure which to go with. But thinking like a juror in a civil trial and going with the preponderance of the evidence, we have to say we see some acceleration in the month of January as an isolated event, but smoothing with year-on-year returns a near-unanimous verdict that inflation is continuing to cool.
US Unconventional CPI
Tuesday, January 13, 2026
The unconventional CPI measures shed no new light this month. They confirm inflation was higher in December than the rest of the quarter -- but that year-on-year measures are relaxing. At this point the Fed's preferred measure, Personal Consumption Expenditures inflation, has still not updated since September, so the most important one in the mix is missing.
US Unconventional CPI
Thursday, December 18, 2025
Wow. All the unconventional measures confirm yesterday's good CPI news. Sharp deceleration across the board, no matter how you slice it or dice it.
US Unconventional CPI
Friday, October 24, 2025
The unconventional data this month gives a conventional result -- inflation is sticky, but nothing like an "unusual and extraordinary threat," to borrow the language of the International Emergency Economic Powers Act. And they all confirm that energy -- gasoline, that is -- was the culprit in September.
US Unconventional CPI
Thursday, September 11, 2025
There's no way not to see an acceleration of inflation in the unconventional data series.
US Unconventional CPI
Tuesday, August 12, 2025
The alternative CPI measures are a little harsher than plain-vanilla CPI. There is a sharp division between "flexible" and "sticky" sub-aggregates. The "flexible" are above target and the "sticky" are below. So how are we to understand this? Aren't the "flexible" ones the goods and services that are most sensitive to changing conditions? Shouldn't they have more predictive value?
US Unconventional CPI
Tuesday, July 15, 2025
Interesting paradox in the June unconventionals. The so-called "sticky" components heated up in June, while the so-called "flexible" components didn't.