Paul Krugman
Wonking Out: Demand, Discombobulation and Prices
"So, Democrats did much better in the midterms than most pundits expected, and markets celebrated, with stocks surging.
OK, not really. What sent the stock market soaring on Thursday was a good inflation report, which led investors to believe that the Federal Reserve won’t raise interest rates as much as expected and that the U.S. economy may manage a soft landing, that is, get inflation under control without a severe recession.
If you ask me, markets overreacted to one good month. Core inflation, which excludes food and energy prices, came in low for the month, but the monthly rate of core inflation is quite noisy. On the other hand, the traditional measure, inflation over the previous year, lags too far behind a rapidly changing economy. Looking at inflation over the past three months has seemed to be a reasonable compromise. And if you do that, the good news from the latest inflation report doesn’t look like that big a deal:
On the other hand, while markets may have made too much of one month’s data, they were arguably too pessimistic before, not looking under the hood of disappointing inflation numbers to see the underlying reasons for optimism.
The inflation debate has often been framed as a confrontation between economists, myself included, who argued that high inflation was “transitory” and those who predicted that it would be high and sustained. Obviously, I and other optimists have been wrong so far, so in saying anything optimistic now I have the problem of having been the boy who cried “no wolf.”
But there is a different way to frame the debate, which I think is helpful right now — it’s about demand versus discombobulation (sorry for the technical jargon).
The simple story about inflation is that it’s about too much money chasing too few goods — spending growing faster than the economy’s ability to produce, so the extra demand shows up in higher prices. The Federal Reserve believes that there is a lot of truth to that story, so it has been raising interest rates in an attempt to choke off private spending.
What has made inflation analysis complicated these past two years is the discombobulation factor as the economy recovers from the effects of Covid-19. The pandemic was extremely disruptive, in ways that went beyond lockdowns. First, consumer spending shifted suddenly from in-person services to goods — overstressing supply chains and sending freight rates soaring — then back again. Then the surge in remote work fed a huge increase in demand for living space, as people both spent more time at home and sought more room in which to work.
Why is discombobulation inflationary? Why doesn’t it just mean higher prices for the stuff people demand more of and lower prices for the things they want less? The basic answer is that many although not all markets respond asymmetrically to shortages and surpluses.
Prices tend to rise quickly when demand exceeds supply; they often fall reluctantly when supply exceeds demand. Labor prices, that is, wage rates, rarely fall even in the face of high unemployment, because employers fear that wage cuts would damage morale and productivity. (Truman Bewley had an excellent book on this some years back.)
So in a time when there are huge swings not just in the overall level of demand but in its composition — what people are spending their money on as well as their total spending — we would expect to see high inflation, because prices for the things that people aren’t buying won’t fall as fast as prices for the things they are buying rise.
As a result, at least some of our inflation has reflected discombobulation rather than a fundamental excess of demand. And discombobulation will eventually diminish — in fact, it’s diminishing by the week. The question is how much inflation reflects too much demand, not just pandemic-related strangeness.
In 2021, I, along with many other economists — including those in the Biden administration — tried to adjust for disruptions by looking at measures of prices that, in addition to excluding volatile food and energy prices, excluded pandemic-sensitive sectors like used cars and hotel rooms. But the list of pandemic-affected sectors just kept growing, and the risk of motivated reasoning — choosing what to include and what to exclude so as to confirm one’s preferred narrative — became too great.
For the most part, I’ve given up on trying to find the “right” price index. Instead, these days I mostly look at wages as an indicator of underlying inflation.
Average wage data were also discombobulated in 2020 and to some extent in 2021, because Americans who lost their jobs were typically lower-wage workers, which caused average wages of those still employed to spike during the worst of the pandemic, then fall as people came back to work. But that has been less of an issue more recently. And high wage growth over the winter of 2021-22 convinced me that overall demand was, in fact, too high.
But wage growth has been coming down. In fact, over the past three months it has been less than one point higher than it was before the pandemic, when it averaged around 3 percent:
That’s encouraging. But how do we reconcile those numbers with core inflation, which is still running far above prepandemic levels?
The most likely answer is that core inflation now largely reflects a rapid rise in the official estimate of shelter costs:
These estimates, in turn, are largely determined by the rental rate on apartments — because rents are a major item in the Consumer Price Index and because another, even bigger item, the “equivalent rent” measure the Bureau of Labor Statistics uses as the cost of owner-occupied housing, basically piggybacks off actual rents.
But here’s the thing: Most renters have leases, lasting a year or more, so the average rents people are paying lag far behind the rents paid by new tenants, which reflect the current state of the market. The B.L.S. has studied this issue, but doesn’t produce data on a regular basis. There is, however, a lot of evidence suggesting that new-tenant rents, which surged last year as part of the general discombobulation, are now growing much more slowly. They may even be falling.
Since shelter is about 40 percent of core inflation, it’s therefore not hard to make the case that “true” core inflation is now well below 4 percent, making it more or less consistent with the wage data.
The bottom line is that while you don’t want to make too much of one month, inflation does indeed seem to be falling. We’re not all the way back to acceptable levels, but the progress is real."
No comments:
Post a Comment