The key consideration for the December rate decision is how the FOMC will assess the underlying inflation rate at that time and the balance of risks and uncertainty associated with that inflation estimate. The core measure of PCE inflation is usually taken to be an estimate of the underlying rate. That allows the Committee to look through volatile oil and food price shocks, but there may be other price-level shocks to components of core price levels that would be appropriate to look through. There was an apparent price-level shock, centered on March. A price-level shock would temporarily lower core inflation, though there would be a lingering effect on the 12-month core inflation rate until the shock moves out of the 12-month window. In this case, the Dallas trimmed-mean PCE inflation rate is likely a more reliable measure of the underlying rate than core inflation, as it simply drops outliers—which in this case includes wireless communications in March. We present inflation scenarios with alternative monthly inflation rates after March to see what the Committee could see at the December meeting for core inflation, both over a 12-month window and the annualized rate from March through October, as we believe annualized core inflation over this shorter horizon will get special attention at the December meeting.
The Slowdown in Core PCE Inflation
The core measure of the PCE inflation rate is usually viewed as a measure of the underlying inflation rate, and the inflation measure that should guide monetary policy decisions. In Figure 1, we show the one-month and 12-month inflation rates for core PCE inflation since January 2016. Note that the 12-month rate from July 2016 to February 2017 was 1.8% or 1.9% and we viewed that as a measure of the underlying inflation rate at that time. Indeed, in that case, the FOMC had virtually achieved its inflation objective! But not so fast. The 12-month rate fell after that, reaching 1.4% in July 2017. Is the 12-month rate the measure we should now identify as the underlying inflation rate today? If so, dramatically different implications for monetary policy.
Price-Level Shocks and Underlying Inflation
A shock on the price level—as opposed to price inflation—can have a sharp, but transitory, effect on inflation. Indeed, that’s why the FOMC focuses on core, as opposed to headline, inflation. But there are components other than oil and food that are occasionally subject to price-level shocks. Today, it’s clear that there was a sharp decline in the price level for wireless services in March. In Figure 2 we show monthly readings for the price level for wireless communications from January 2017 through today. The March drop stands out, literally like a sore thumb! We also show in Figure 2, the arithmetic contribution to 12-month core PCE inflation from the drop in the wireless communications component. However, at its greatest, this price-level shock only explains two-tenths of the ½-percentage-point slowdown in the 12-month core inflation rate, and just one-tenth today. There will be a persistent, though modest, negative contribution to the 12-month rate from the March 2017 price level shock through March 2018, when it will fall out of the 12-month calculation for core inflation. This will help.
The Dallas Trimmed-Mean Measure of PCE Inflation
The Dallas trimmed-mean measure of PCE inflation is yet another measure of core inflation. Rather than simply looking through shocks to oil and prices, it looks through price-level shocks in the data at every point, identified by a move in those components’ inflation rates that are outliers. That measure becomes especially valuable when there have been price-level shocks to components other than oil and food. In Figure 3, we show this measure, along with the core PCE, both in terms of 12-month inflation rates. Note the trimmed
mean measure has also declined, in this case, also from a 1.9% peak (from October 2016 to February 2017) and to 1.6% in July 2017. At this juncture, the trimmed-mean rate provides a more reliable measure of the underlying trend and we take its signal as a decline in the underlying rate of inflation from 1.8% or 1.9%, to 1.6% or 1.7%. We expect the Committee will be monitoring the trimmed-mean measure closely as it heads toward the December meeting.
Inflation Scenarios and the December Rate Decision
We conclude with some scenarios about how underlying inflation may look to participants at the December meeting. We have argued that the Committee is likely to give special weight to the annualized rate of core inflation over the period from March 2017 to October 2017, because inflation over the period since the price level shock in March would reflect any rebound to the underlying rate after that shock. So we show three scenarios in Table 1, beginning with alternative paths for monthly inflation readings and, for each scenario, both the corresponding 12-month inflation rate and the compound annual rate from March to each successive month between July and October.
Sources: BEA, Authors’ Calculations.