My main takeaway from today’s FOMC even is that the Chairman of the Fed is an adult who knows how to delay gratification.
Powell is now a bit more confident still that the peak funds rate was put in place six months ago and his base case is that the 12-month inflation rate will fall to what is implied by the six-month rate. So, the Fed’s forecast is definitely benign / dovish, leaving aside whether it is correct. However, he does not want to declare victory (implicitly) by cutting the funds rate until he is very confident that inflation will stay down, especially with demand growth looking solid. This means that March is definitely not the base case. They would need a surprise to go in March, which Powell said pretty much explicitly. But they do expect to ease, perhaps even by as much as is currently priced. Powell made some references to the median rates dot from last meeting, but it was perfunctory. We cannot infer from that that their expectation is less than is priced. I will have some more later, but that would be the gist of it.
I want to return here to a point I made earlier about how the real wage is not really 5% below trend or below normal at all, in the base case. I mentioned that the disagreement between Jason Furman and me on this issue could be accounted as disagreements about productivity, the appropriate price deflator for use in this context, and the difference between Average Hourly Compensation and core ECI growth since just before the Covid shock.
With the advantage of a bit more time, I am able to quantify that the main issue here is definitely the last item. Since the Covid shock, the ratio of AHC to core ECI has risen by 4 percentage points. This explains 80% of my disagreement with Furman about the labor share.
It is convenient that the source of the disagreement is so concentrated in the one issue, because it means you, dear reader, can just take a view on the one thing. My contention is that AHC is too volatile on a quarter to quarter basis to be used as a way of determining underlying nominal wage growth or nominal broader compensation growth. I am well within consensus on that. But when thinking about whether the labor share is depressed or normal, we do have to take account of the fact that in level terms, it is total compensation, not just its nonvolatile component, that is relevant to the labor share and thus to the notion that any measure of real wages might be “depressed” in an economically meaningful sense.
Separately, and secondarily, notice that AHC relative to ECI has a trend of about 50 bps a year. This reflects that some component of productivity growth is due to mix shift, workers moving into relatively high productivity sectors over time. This does not mean that we should watch AHC rather than ECI to get a sense of underlying wage growth. Quite the opposite as mentioned. But when thinking about chopping trend productivity off the ECI growth rate, you first probably need to add 50 bps to that wage growth rate. You will not necessarily agree with me on this, which is fine. Maybe you think even the level of the ratio is distorted from the correct signal by stock options realizations. But at least now you know what the source of the debate here is. Conveniently, it is mostly just the one thing.
Data are effectively actual to Q4, although the growth rate of AHC is consensus. We get the data tomorrow.
