The New York Times
By Paul Krugman
| The news about inflation has been really good lately — almost too good to be true. In fact, it probably is too good to be entirely true. I say this partly on general principles, and partly because there are some specific sectors — notably medical care services — where the methodology used to estimate inflation is probably leading to some spurious apparent price declines. But the overall decline isn’t being driven by a handful of items; the distribution of price changes has shifted down across the board. Disinflation is real, even though we can quarrel about the specific numbers. |
| But where’s the good news coming from? Many commentators rushed to give credit to the Federal Reserve, which began raising interest rates early last year. But the more I look at that claim, the less plausible it seems. I’m not saying that the Fed was wrong to raise rates: The case for rate hikes seemed clear at the time, and at this point it’s starting to look as if the Fed may have done the right thing for the wrong reasons. More on that below. But the assertion that the Fed gets credit for disinflation seems at least mostly wrong, for two reasons. |
| First, we have a standard account of how tight monetary policy is supposed to reduce inflation — and it doesn’t fit the facts at all. Here’s how the process is described, for example, in one of the best-selling economics textbooks: |
 Krugman and Wells |
| Yes, that’s from Krugman and Wells, which is totally mainstream on these issues. And the problem should be obvious: We haven’t seen an extended period of high unemployment, or actually any increase at all in the unemployment rate. We can tell stories about how rate hikes might reduce inflation without any visible rise in unemployment, but they’re speculative and strained. |
| The second reason not to give the Fed credit is that we have lots of direct evidence for an alternative story about disinflation, which I think of as recombobulation. (No, I didn’t invent the term.) |
| Clearly, Covid-19 and its side effects discombobulated the economy — a point much emphasized by Team Transitory. It’s now clear, however, that we vastly underestimated both the scope and the duration of the discombobulation. There was much talk in 2021 about how a shift in demand from services to goods had overwhelmed supply chains, causing things like shipping costs and the prices of used cars to skyrocket. But these, it turned out, were only a small part of the story. |
| Housing was a much bigger factor. The rise of remote work, which began as a response to the pandemic but seems to have become permanent, led to a surge in the demand for housing. This in turn led to a corresponding surge in rental rates for new tenants. This surge peaked in the first half of 2022, and new tenant rental rates have flattened out drastically as the market normalizes. As I said, recombobulation: |
 CoreLogic |
| But the Bureau of Labor Statistics measures shelter inflation largely by looking at average rental rates, which lag behind market rates by a year or more, so this recombobulation is only now beginning to bring official inflation down. |
| Then there’s the labor market. The traditional Phillips curve, which relates wage increases to the unemployment rate, hasn’t worked at all in recent years. But other measures of labor market tightness, such as vacancy rates and, especially, the rate at which workers are quitting their jobs, have done tolerably well at predicting wage changes: |
 FRED |
| Now, normally there is a fairly close inverse relationship between unemployment and quits, a quit-rate version of the famous Beveridge Curve linking unemployment to vacancies. During the pandemic and its aftermath, however, quits were much higher than the normal relationship to the unemployment rate would have predicted, presumably reflecting the dislocation of labor markets in a time of wild and crazy changes: |
 Bureau of Labor Statistics |
| But here too we see recombobulation, with the relationship of quits to unemployment moving back toward the historical norm, which is probably the main reason wage growth appears to be slowing. |
| The point is that stories about recombobulation — the fading away of pandemic-era distortions — driving disinflation are clearly supported by the data. Claims that Fed tightening drove it are sketchier and much more speculative. |
| Another claim I’ve been seeing is that inflation would be much worse right now if the Fed hadn’t raised rates. This might well be true, but it’s also something of an evasion — it’s offering a counterfactual rather than answering the question of how we’ve achieved disinflation so far. |
| Which is not to say that the Fed was wrong to raise rates. In fact, given the inflation rate in early 2022, I don’t think the Fed had any alternative. The economy looked overheated at the time, and it was also reasonable to argue that tight money was essential to avoid a ’70s-type scenario in which inflation became entrenched in expectations — even if rate hikes meant risking a recession. |
| The funny thing, however, is that even though almost everyone was predicting a recession, it still hasn’t happened — but disinflation has. This suggests to me that the Fed may have done the right thing for the wrong reasons. |
| The economy’s resilience in the face of rate hikes suggests that overall demand has been stronger than anyone expected — possibly in part because Biden administration policies appear to have unleashed a huge wave of manufacturing investment. Without the Fed’s hikes, this might have led to an inflationary boom. So while the Fed was trying to cool the economy down, what it may actually have done was prevent the economy from getting even hotter — and thereby allowed the forces of recombobulation to reduce inflation. |
| I still believe that overall U.S. policy, both fiscal and monetary, has been much better than most people realize. But I now also think that when it comes to disinflation, which has so far been incredibly painless, we got lucky. |