US: JH – Our Comment Q&As

Our post Jackson Hole comments in Q&A format, following the discussions we have had with several of you over email and in meetings.

What was your expectation ahead of Jackson Hole?

Ahead of Jackson Hole, our expectations were fairly uncertain. We saw two roughly equally likely outcomes. One possibility was that the Chair would say essentially nothing new, resulting in a broadly neutral message. The other was a relatively hawkish outcome, particularly given the pressure at the back end of the curve. In the end, the message turned out to be a bit more hawkish than we had expected, although no forward guidance was provided, as expected.

Is the focus on distributions surprising?

The focus on distributions is not surprising, given that the staff has been emphasizing distributional measures for some time, which is also why we have been focusing on them ourselves. If anything, the surprising element is that Warsh appears to have genuinely changed his view on trimmed mean measures.

There are, however, some surprising details in the distributional statistics and the specific numbers he cited, particularly Warsh’s choice of “the two decades that preceded the pandemic” as the relevant benchmark. In our view, this is not an appropriate reference period, because core PCE inflation remained persistently below the Fed’s target over those two decades, averaging around 1.7%. A more appropriate benchmark would arguably be the 2000-2007 period (or 2016-2019 in CPI space), when core inflation was much closer to target.

Do you have any metrics for the distribution?

Certainly. This has always been one of our main areas of focus. On underlyinginflation.com, select an area, then “Distributions,” and finally “Empirical CDF Stats Table.” There you can find the full set of statistics describing the cross sectional distribution of month over month SAAR inflation rates. For example, the table reports the share of items whose prices are increasing at annualized rates above 2% or 4%.

The main difference relative to Warsh’s approach is the horizon. We focus primarily on the margin, using month over month SAAR inflation rates, whereas Warsh referred to six month and year over year changes.

Ok but are we at target then?

We reiterate what we have been saying for the past five years. The distribution does not appear consistent with a 2% inflation target. That said, some progress has been made in recent months, with several distributional statistics improving and now looking broadly similar to those observed during the 2016 to 2019 period. We nevertheless remain cautious in drawing strong conclusions, as the high volatility of the current environment has repeatedly shown that these dynamics can change very quickly.

Does the Fed need to hike then? Will they hike in the end?

The answer obviously depends on the broader macroeconomic equilibrium, the forecast, and the Taylor rule. We will shortly circulate a new FRB US baseline. Taking all these elements into account, the new baseline continues to imply no rate hike and instead suggests keeping the federal funds rate unchanged.

More importantly, measured inflation could be revised downward by around three tenths, which means that the most likely scenario at this stage is one in which the September SEP includes a downward revision to the core PCE inflation forecast for the current year. Conditional on such a downward revision, there have been only three instances in the history of the SEP in which the Fed nevertheless raised interest rates.

There is, however, an important distinction. Two of those three cases occurred in December SEPs, when the Q4 over Q4 forecast for the current year already incorporates a substantial amount of realized data. If we exclude December and consider only the March, June, and September SEPs, only one case remains, June 2017. There is no historical precedent for a rate hike at a September SEP meeting accompanied by a downward revision to the median core PCE inflation forecast for the current year.

One can, of course, argue that the current level of the federal funds rate is not restrictive enough. We will address this point separately and circulate new scenarios based on FRB US shortly.

Would you understand a hike in September?

Of course, anything is possible. However, if the FOMC were actually to raise rates in September with an SEP broadly similar to the one published in June, or even with a downward revision to the core PCE forecast, we would no longer be able to reconcile the decision with our understanding of the Fed’s reaction function. Ultimately, we believe this is the main source of the current confusion.

Contact

Want something more tailored?

We provide tailored consulting on ad-hoc projects.