Long ago I studied international economics with the great Charles Kindleberger, famous not just for his insights but also for his bon mots. One line I took to heart — he was talking about the balance of payments, but it applies equally well to many subjects in economics — was that people always want a single number, but what they really need is a story.
So it is with the issue of the moment, inflation. There are many measures, enough so that — to cite another Kindleberger quip — you can always find a way to justify either optimism or pessimism, according to your temperament. But what’s the story?
The answer, I’d argue, is that after a few confusing months early this year, the big narrative from last year — “immaculate disinflation,” inflation gradually ramping down to an acceptable rate even though we haven’t had the recession some economists insisted was necessary — is back on track. The big question now is whether, having discovered that we didn’t need a recession, we’ll get one anyway.
What’s the basis for my assertion? As I said, there are many measures of inflation. These measures tend to tell the same story when things are either very bad or very good. When inflation was very high in the 1970s, all measures agreed that it was very high. During the long period of relative stability from the mid-1980s to the 2008 financial crisis, and again for most of the 2010s, all measures said that inflation was fairly low.
In the turbulent post-Covid era, things have been more muddled. Inflation measures can diverge because of the way they treat hard-to-measure things like the price of financial services and, most famously, the weight they place on the cost of housing, which seems to reflect market conditions with a long lag, and more.
That said, there’s a pretty good case for focusing on the measures preferred by the Federal Reserve, which has many years of experience in trying to make inflation-related policy decisions. Let me give you a couple of variants.
First is a measure that Jerome Powell, the Federal Reserve chair, has cited favorably: core market-based personal consumption expenditures. To parse that a bit, the “core” means that it excludes volatile food and energy prices; “market-based” means it excludes items that don’t actually trade on markets, whose prices are purely imputed. Here, for the past year, is the inflation rate for this price index, measured at two frequencies, month by month at an annual rate and over the previous year:
As you can see, monthly measured inflation rates (the blue trend line) are wildly unstable, and it’s always hard to know whether a big swing represents a real change or just statistical noise. A spike in prices at the start of 2024 created a lot of uncertainty: Was inflation making a comeback, or was this just companies resetting their prices at the beginning of the year, a phenomenon that “seasonally adjusted” data is supposed to correct for but may not do so adequately? Well, the falloff in inflation since then has tilted the interpretation toward the idea that this was a statistical blip. Indeed, the annual rate of inflation (the red trend line) has continued its steady decline.
As it turns out, one thing “market-based” inflation still includes is the imputed cost of owner-occupied housing. As many of us have pointed out, official measures of housing costs are very much a lagging indicator, reflecting a surge in rents that ended more than a year ago. So it may make sense to exclude housing costs from your inflation measure — not because they don’t matter to families, but because a measure excluding shelter may be a better predictor of future inflation. If we do that, the above chart would look like this:
By this measure, inflation — both monthly and annual — is already more or less at the Fed’s target rate of 2 percent.
Of course, there are other measures. In the past I’ve cited the New York Fed’s estimate of Multivariate Core Trend Inflation, which has been revised up and now shows a blip earlier this year, but has reversed that rise and is currently running at 2.8 percent. A proprietary measure from Goldman Sachs is more optimistic and shows us very close to the Fed’s target:
Given the knottiness of official inflation numbers, I also find it useful to look at soft evidence — what businesses are saying. The Fed regularly surveys businesses around the country, publishing the results in the Beige Book. The most recent edition says “Prices increased at a modest pace over the reporting period.” How does this compare with what the report said on the eve of the Covid-19 pandemic, when everyone considered inflation well under control? The language from the January 2020 edition was almost identical: “Prices continued to rise at a modest pace during the reporting period.”
Overall, it looks as if underlying inflation is probably between 2 percent and 3 percent and the hot numbers earlier this year were a false alarm. We may or may not have brought inflation all the way back to the traditional (but arbitrary) target, but inflation really doesn’t look as if it should be a major preoccupation at this point.
I am, however, beginning to get a bit worried about an economic slowdown.
There’s nothing out there that screams “imminent recession,” but there are straws in the wind. Consumer spending, adjusted for inflation, fell slightly in April. A widely followed report on manufacturing hinted at developing weakness. Again, we’re not talking alarm bells yet, but the balance of risks has clearly shifted.
So it’s time to stop obsessing about inflation, which increasingly looks like yesterday’s problem, and start worrying about the possibility of a recession as the economy’s strength finally begins to erode under the strain of high interest rates. So yes, I think the Fed should start cutting rates, and soon.
Quick Hits
When were the good old days?
Murder is plummeting.
Economic news coverage may be turning up again.
Social mobility has been declining.