There is always much pretend around the Winter Solstice, especially among the adults. On Wall Street we like to make a show of holding ourselves to account for our calls over the past year, for example. I got “only” 6 — or at a huge stretch maybe 7 — of 10 right!
But let’s follow the wisdom of Jack Palance and focus on the one thing. My biggest miss during the past year was a speculation that a sustained decline of underlying inflation in the goods and services markets would probably require an easing of labor market conditions delivered through below-trend aggregate demand growth. In the event, the GDP grew at an above-potential rate, as evidenced by the employment / population ratio rising; and core inflation fell steeply anyway. There may be some “lags” involved here, as I will allude to below. But they were not supposed to take more than a year to play out.
In the chart below, the red dashes indicate the readings of 12-month core inflation and the Employment Gap for November 2022, although the figures are current vintage. You would have to work hard to see any ambiguity here. On the most central macro debate of 2023, the most straightforward read of the evidence shows that the hawks were wrong.[1]
These are standard, not cherry picked
Source: Blanchflower and Levin (2015), Bloomberg, CBO, Federal Reserve Bank of St. Louis, Zillow, CoreLogic, FH calculations
Labor market and price data are actual to November. The red splotches indicate readings for 12 months ago.
But Wall Street and policy circles are full of ambitious people. And there still runs a fake debate about whether the immaculate disinflation, widely disparaged in real time, has actually happened. One approach is to use a Catholic definition of what labor market ease might involve. For example, some claim that the labor market eased, but that the evidence is in lower vacancies, lower quits, or even slower wage inflation.
For example, see here for private sector labor economist Guy Berger insisting that Paul Krugman is too quick to declare victory on his recombobulation thesis, which is from the immaculate disinflation school. According to Krugman, it is not that there was excessive demand (in “level” terms) that needed to be taken out. Rather, the economy was discombobulated. The evidence from 2023 supports that take and it is unseemly to quibble in retrospect.
In fairness, not all who quibble are throwing up rationalizations. For example, a year ago you could easily have found people arguing that matching frictions were a transitory source of inflation pressure and that merely bringing down demand growth to potential could deliver disinflation, simply by providing space and time for the matching frictions to go away. Economists at Goldman were in that camp. And if I remember correctly, I think Guy Berger may have been as well, much to his credit. It is more that in an important practical sense, there is little difference between the transitory matching frictions and recombobulation theses. They need not be presented as contrasts. As often, some of the confusion here can be cleared up by agreeing what the question even is.
And that brings me to how I am inclined to be a bit persnickety on this issue, because I can recall clearly what the main macro debate was a year ago. The hawks, among which I mostly count myself, were not saying, we need labor market ease through lower vacancies or a lower wage growth rate. No. We were insisting the unemployment rate needed to rise.[2]And on the evidence so far, it didn’t. Moreover, the fact that wage growth and quits slowed without the unemployment rising should probably be described as endogenous to the apparent validity of the immaculate disinflation view. For example, wage growth slowed because lower inflation reduced the need for nominal catch-up, and that was achieved without the higher unemployment rate that many thought would be required.
It would be pretty weaselly, then, to sneak out by saying well the labor market eased if you look at it properly, just as we insisted would be required! I think it is better to join, say, Olivier Blanchard (a very prominent hawk) in conceding ruefully that the standard take did not work in 2023.[3]
The issue is not the specific author but that the subject humiliates
Source: Olivier Blanchard, as linked above
So, where does this leave us looking forward? I would say that there is one dovish and one hawkish implication. On the dovish side, there is little question that the realization of steep disinflation in goods and services markets allows the Fed the luxury of risk managing away from economic weakness as well as from the inflation threat. This luxury has become affordable in stages over the past several months, as underlying inflation in the goods and services markets has marched lower. The equity market has responded to the development already, so my only forward-looking point here would be that I see little reason to believe that this reassessment must itself be second guessed. Confidence in the benign take will ebb and flow with the data, but the apparent reality here looks quite different from what we might have guessed a year ago.
The hawkish point is that the rates forward market may now be expressing too much confidence in the view that the immaculate disinflation has become the only plausible lens. Most of the disinflation in the goods and services markets can still be described as a result of the high-frequency implications of getting aggregate demand growth in line with supply, just as the optimists insist. Both the now received view and my 2 Stage Disinflation hypothesis agree on that point. However, it remains conceivable that the high-frequency developments have (understandably) distracted folks from a lower-frequency development that will encourage the Fed to deliver a rates ease, particularly with the most standard measures of the labor market suggesting it may still be tightening. That is, the central case remains that the labor market has overshot full employment. And it is easy to miss this point, assuming it is valid, if we watch vacancies, quits or the change of wage inflation, whose behavior can easily be seen as consistent with the high-frequency developments going one way, even as the lower-frequency developments move less spectacularly in the opposite direction.
Or to put this in plain English, the Fed’s risk management approach is now much more balanced, because of the fact of lower inflation. This is hugely important, as we have seen. But the idea that they will now shift to growth on seems still to be a stretch. So, the odds are that the rate cuts now priced will be delivered only if a failure to do so would likely imply meaningfully below-trend growth. And that seems like a decent hurdle.
[1] The two panels in the chart show what I have long taken to be the best summary measures of the economic variables under consideration in this context. There is no cherry picking of indicators here.
[2] This idea was pervasive and was behind Powell’s suggestion that there was no “easy” way to disinflate and that it would take bearing “pain.”
[3] Blanchard has been more direct on other occasions, again to his credit. What I love about the Tweet is that the responses to it presume he has some complaint about Greg Mankiw specifically. So good.
