If you’re a normal human being trying to read recent economic news, especially about inflation, you may be feeling confused right now. You may have seen reports that consumer prices are up 3.1 percent over the past year, which sounds bad; “core” inflation, which excludes volatile food and energy prices, was 4 percent, which sounds worse.
Yet many commentators are saying that the Federal Reserve’s fight against inflation is more or less done, and financial markets expect the Fed to reverse course soon, cutting interest rates instead of increasing them. How can this make sense?
In a technical sense, it’s all about the lags. In a deeper sense, there are a range of inflation measures, and which one you should choose depends on what question you’re trying to answer.
Here are two measures of inflation in recent years: the one-year change in the core Consumer Price Index and the six-month change in the core personal consumption expenditure deflator — a measure that’s conceptually similar but different in detail — expressed at an annual rate. The latter measure was only 2.5 percent for the six months ending in October, and most analysts expect that when the November number comes in on Friday, it will be close to 2 percent, the Fed’s target:
Why do these measures look so different? Part of the answer is that if inflation is falling rapidly, which seems to be the case, looking at the change in prices over a whole year literally puts you behind the curve: A lot of what you’re capturing is stuff that happened a while ago, rather than what is happening now.
In that case, however, why not look at month-to-month changes? Noise: There’s too much randomness in monthly data to make it a reliable indicator. Taking a six-month average is a compromise that cancels out a lot of the noise but gets you reasonably close to current events.
But that’s not the whole story. These are two measures of consumer prices. And right now the deflator — which the Fed normally prefers in any case — is a much better indicator than the Consumer Price Index of how the fight against inflation is going. Why? Again, the answer is lags, specifically involving housing.
The cost of housing makes up around a third of the Consumer Price Index and about 40 percent of core C.P.I. The Bureau of Labor Statistics measures housing costs using rents — the rents people actually pay if they are, in fact, renters, and an estimate of the rents they would be paying if they own their houses. Normally this procedure raises few problems.
But most renters are on leases, so the average rent people are currently paying lags behind market rents — what people pay for newly rented dwellings. This isn’t usually a big deal. But there was a huge surge in market rents in 2021-22, probably reflecting the rise in remote work: People working from home wanted more home to work from. This surge has now subsided, but it’s still filtering into the standard rent numbers. Here’s growth in “new tenant rental rates,” a newly developed series, and official rents:
What this tells us is that a lot of measured consumer price inflation reflects stuff that happened many months ago, not what’s happening now. And for technical reasons, the deflator puts a lower weight on housing, so it’s less affected by this lag.
Which of these measures is right? As I said, it depends on what question you’re trying to answer. The Fed is trying to decide whether it should raise or lower interest rates, so it’s looking for indicators of whether the economy is currently running too hot, too cold or just right. For that purpose something like the six-month change in the deflator is better than the annual change in the Consumer Price Index, which is strongly affected by factors that are now in the rearview mirror.
And this measure suggests that the economy is no longer running hot and may be getting colder. So it’s time to think about rate cuts.
There are, however, other questions we may want to answer — such as, what is happening to the purchasing power of U.S. workers? And these call for different measures (although I can’t think of any current question for which the one-year change in core C.P.I. is relevant).
If we’re asking about the past year, the natural way to answer this question is to compare what has happened to average earnings with overall consumer prices. And yes, real earnings are up.
But if you try to push things farther back — say, to the start of the Biden administration — you run into some weird data issues. Specifically, the pandemic temporarily caused many workers to lose their jobs — and the laid-off workers tended, on average, to have low wages. So the average wage of workers who still had jobs jumped, not because anyone received a real raise, but because the worst-paid had left the room. Average wages then fell as life returned to normal, not because people were facing pay cuts, but because low-wage workers were back in the average.
All of this makes it hard to pin down what has happened to workers’ earnings since 2020 or 2021, which is why many analyses focus on a longer time horizon, comparing earnings now with earnings before the pandemic. Here, for example, is a recent chart from the Treasury Department:
Wait, you say, you were just telling me to focus on six-month changes; now you want me to look at changes over four years? But there isn’t any real inconsistency here. We’re looking at different numbers because we want to answer different questions.
And the overall picture is actually pretty good. Inflation does seem to be coming under control without the high unemployment many economists thought would be necessary. Workers seem to have come through a turbulent period of the pandemic and inflation with higher purchasing power than they had before.
There is also a broader lesson. People often want to judge the economy by a single statistic, like the annual inflation rate. But numbers don’t speak for themselves. They can be used to help tell a story; they aren’t the story on their own.
The administration’s top economist on a good year.
Donald Trump denounces rising stock prices. Really.
European inflation is also falling.
Republicans and Democrats see very different economies.