Good morning. I don’t think I can remember a major data release that came in as closely in line with expectations as yesterday’s CPI report. Calling around to economists and strategists yesterday, all of them struggled to name something in the report that didn’t fit with their own or consensus targets. Not sure what that signifies, if anything.
As bad as we thought is plenty bad enough
Well, here we are:
The blue solid lines are core inflation, which excludes food and energy, compared with a year ago and with last month. The dotted green lines include all items. The annual rate of inflation is accelerating, for the third month in a row. Monthly increases are steady, with the interesting wrinkle that headline has fallen just below core, because food and energy prices both fell last month (a phenomenon most observers do not expect to last).
As expected, goods were the big driver. The egregious example remains used cars, which rose 37 per cent from the year before, handily beating the S&P 500.
But this data lags, and CPI rent will get higher before it gets lower. It currently reflects neither the huge spike in rents at mid-year or the cooling in rents we have seen very recently.
The reason for worry is that we have a good theory of why goods inflation might go away. The pandemic has pushed demand away from services towards goods; stimulus payments have meant total demand has not fallen, possibly even risen some; when the pandemic subsides, demand will shift back towards services, and goods prices will normalise.
But if strong demand and limited labour supply force wages higher, and those wage increases leak into services, that’s evidence we are heading for a big, sticky wage-price spiral.
Well, the wage-sensitive services I was warning about did accelerate in December, if not exactly sharply. Prices of everything from haircuts to house cleaning to legal services and funerals picked up from November, to greater or lesser degrees. These are volatile series, and we need more confirmation of this trend. But I do not like this one tiny little bit.
It is especially worrisome because real (post-inflation) wages are falling. Workers have good reason to respond to pervasive price increases by demanding more pay. Olivier Blanchard of the Peterson Institute put the problem this way in an email:
It does not have to happen this way, as Blanchard acknowledged. As Don Rissmiller of Strategas put it to me, there is a wage-price spiral happening already in some low-wage areas such as leisure and hospitality. But:
There is still a way out, especially for the high-wage professions. People are asking for flexible work. They don’t want more pay, they want to work at home Monday and Friday . . . it’s one more pressure-release valve. Certainly there is the risk, and the biggest risk we’ve had in a while, that we are heading into some sort of wage-price cycle. But there is a chance that the Fed can act now and stop the services inflation.
How can you stop inflation with less-negative real rates?
The Fed projects, and the bond markets expect, that at the peak of this rising cycle the policy rate will be about 2 per cent, or a touch more. That means the Fed and the market think the central bank will halt inflation while never pushing real rates above zero.
This is a strange idea, on its face. Think of it this way. If all goes as expected, in a year short-term rates will be all of 1 per cent higher than they are now, and still very negative in real terms. How is that going to constrain the economy in any meaningful way? It seems a stretch to think such a small change would constrain either consumer or corporate spending significantly.
The range of views on this puzzle is remarkably wide.
Paul Ashworth of Capital Economics argued that “even the Fed easing its foot off the accelerator a bit” can actually make quite a large impact on the real economy.
My colleague Martin Wolf disagreed. As a mechanical matter, he thinks the policy won’t cool the economy to any significant degree. He says it all comes down to credibility:
The actions themselves don’t really matter. What matters instead is the confidence that the Fed is serious about its goals. Then, if what it has done is not enough, the Fed will do more — much more. So, the signal is the policy and the signal on its own might be enough . . . but that only works to the extent that the intentions revealed are believable. The smaller the credibility of its intentions, the more the Fed will need to do to show it’s serious …
For 40 years the Fed has lived on [Paul] Volcker’s credibility. Maybe it will have to show it means it once again. That would be a nightmare. And that is also why letting inflation rip is dangerous. The more that needs to be done, the less credible the needed actions become. That is why a Volcker became necessary in the 1970s.
Rissmiller takes still a third view.
And inverted curve is hard to deal with, if you assume the future is more uncertain the present. It hurts confidence, and the financial sectors’ ability to make profits . . . you at least get a situation where, with an inverted curve people will become concerned and take less risk . . . and then if the tide goes out a bit, someone [a big creditor] has a problem, and then credit spreads widen, and it all becomes self-reinforcing.
These views likely have a fair amount of overlap, and are not mutually exclusive.