Clients have sent over a piece from a buy-side economist arguing that inflation is reaccelerating and that the Fed may have to reverse its rate cutting program in response. With financial conditions seemingly quite accommodative and with the real-side data flow strong, I would not be inclined to fight the idea that medium-term rates risk is tilted to the upside, particularly given that the tariff issue has not gone away.
However, the inflation news most recently has been more benign than my colleague characterizes, in part because he places far too much emphasis on CPI-based measures of underlying inflation, which are basically irrelevant. The Fed will ease on Wednesday, probably hint that the timing and speed of future rate cuts is more data dependent, and yet probably also continue to imply that further rate cuts are in store. The recent inflation news greenlights this approach. The uncertainties here relate to the strength of the economy, financial conditions, and to the tariff issue.
Fed must confront arrow in the data
Source: Bloomberg, Federal Reserve Bank of New York
NY Fed data are actual to November. University of Michigan data are actual to the December preliminary.
In this note, though, I want to get into an admittedly secondary aspect of the hawkish argument, not because it is particularly important, but more because I have a differentiated take on it. Note in the chart above two prominent measures of inflation expectations in the real economy, as opposed to financial markets, assessing things at both the short-term and medium- to longer-term horizon. My colleague is concerned by the fact that the 5- to 10-year inflation expectation in the Michigan Survey looks to be about 10 basis points higher than it has tended to be during the most recent period of steep inflation. He may have unnerved himself by putting a swoosh arrow under the time series to indicate that he already knows it will be higher in the future. Just don’t do that.
The value in telling the data what to do through chart annotation aside, I would just point out that the Fed cannot actually do anything about the 5- to 10-year inflation expectation, particularly if that is pushed higher by concerns that there may be a politically-driven regime change in the conduct of monetary policy. Trump may or may not neuter the Fed, but raising interest rates now to deliver lower inflation in two years will have zero effect on where inflation might be 5 to 10 years from now.
Closely related, you may recall from a paper that circulated a couple years ago senior-Fed staffer Jeremy Rudd arguing that even within the so-called “canonical” model, of which Rudd is skeptical in part because of the prominence it assigns to inflation expectations, it is the short-run inflation expectation that would presumably be relevant to policy. I share his view on this, taking it as bloody obvious, at least within that standard model. Here is the relevant passage from page 6:
Source: Federal Reserve as linked above, page 6
So, then, why do Fed officials so frequently refer to the long-run inflation expectation? I confess that this is a bit of a mystery to me and that I must speculate in a way that involves reading motives, which I realize can be irksome. It is possible that longer-term inflation expectations, as picked up possibly in, say, the Michigan survey, might be interpreted as an indication of how Americans assess the moral rectitude of even incumbent monetary policy makers. The one-year inflation expectation might be dismissed as reflecting a bygone, possibly delivered by a shock. But the longer-term inflation expectation is clearly a comment on policy, perhaps interpreted as today’s policy makers. If we don’t trust them, then that number will go higher, which might hurt their feelings. That may sound like a stretch, but to me it is the least bad explanation.
Either way, though, there is little that a central banker can do about the risk of regime change. Perhaps the only thing they can really do is deliver the right policy to make the economy turn out well enough. And that does not likely involve reacting to the University of Michigan’s 5- to 10-year inflation expectation, although I doubt the Fed would ever say this. It would sound defeatist, as if they were casting blame, even if fairly.
Let me conclude with a caveat that actually reinforces the main point here. The argument above takes it as given that the Michigan Survey being 10 bps above recent normal is actually picking up something real. Market based measures of forward inflation expectations strongly reject that assumption. But I thought it might be fun to argue in the subjunctive. Even if it were true…
