U.S. Inflation Is Still Hotter than Pre-Pandemic. And that's Okay.
Underlying measures of prices are still rising about one percentage point or so faster than they were in 2017-2019. But there is nothing wrong with this new normal, even if may not be fully priced in.
Federal Reserve officials’ preferred measure of inflation—the Personal Consumption Expenditure (PCE) price index—rose just 2.2% over the past 12 months. Over the past six months, PCE inflation has come in at just 1.9% annualized, and by just 1.5% annualized over the past three months. Little wonder that Fed officials have shifted their focus from worrying about inflation to preempting any potential further weakness in the job market.
While that may be the correct decision, many of the underlying inflation measures that are supposed to strip out the noise from volatile and idiosyncratic components imply that prices are still rising about 1 percentage point faster than they were in the years immediately preceding the pandemic. That is broadly consistent with the latest data on nominal wages and incomes, which suggests that it is more representative of the underlying trends than a few good months of headline inflation prints.
The current situation is not necessarily a problem—many policymakers and economists believed that prices were rising too slowly before 2020—but it is worth bearing in mind when thinking about where interest rates might end up once the Fed finishes its “recalibration”. If 2.5%-3% is the new 1.5%-2%, more or less, that could have implications for interest rate pricing above and beyond Jerome Powell’s belief “that the neutral rate is probably significantly higher than it was [before the pandemic].”
How Durable is the Disinflation?
About 56% of the total PCE price index consists of what I have been calling the “supercore” components. This excludes:
Housing, which is known to have serious lags because of how often rents are sampled (16% of the total)
Groceries and energy, which tend to be much more volatile than other prices even if they follow similar long-term trends (12% of the total)
Motor vehicles and parts, which were unusually disrupted by the pandemic and have since been undergoing a production boom (3% of the total)
All “non-market-based” goods and services that are imputed in the absence of observable transactions, including margins on used vehicles, gambling, net consumption of non-profits, some insurance products, and many financial services (13% of the total)
Historically, this narrower measure has been slightly more volatile than overall PCE on a month-to-month basis, but significantly less volatile on all other time horizons. It tends to run a bit slower than overall PCE, because both housing inflation and non-market non-housing services inflation tend to run faster than other price increases. During the worst of the 2021H2-2022H1 price spikes, supercore PCE inflation accelerated about 2 percentage points (annualized) less than the broader index relative to their prepandemic baselines.
Given this, the recent convergence and simultaneous disinflation of the two measures is therefore somewhat surprising. If supercore PCE inflation remains stable around 2% a year and if prior relationships reassert themselves, then total yearly PCE inflation might be expected to settle around 3%, instead of landing back at 2%.
A closer look at the components involved and the specific timing of price increases provides some potential insight into what is happening.