The case for core inflation exceeding 2%
admin | October 4, 2019
This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors.
The drop in core inflation early this year spooked dovish Fed officials and contributed significantly to the case for the modest Fed easing seen this summer. More recently, it appears that core inflation has begun to make its way back to 2%, but many agnostic Fed officials remain skeptical, given that inflation has systematically undershot the 2% target for most of the decade. Parsing the core inflation data in several different ways suggests that core inflation should indeed return to 2%, probably far sooner than many policymakers expect.
A dip or a downtrend?
After briefly inching above 2% in early 2012, the year-over-year advance in the core PCE deflator, the Fed’s favored metric of underlying inflation, spent the next six years below the Fed’s target of 2%, averaging about 1.6%. Finally, in March 2018, the gauge returned to 2% and held within a tenth of that pace for the balance of last year. However, three straight soft monthly readings to start 2019 pushed the year-over-year advance back down all the way to 1.5%.
Fed Chairman Powell wavered on how to interpret the early-2019 slowdown in price gains. At his May 1 post-FOMC press conference, he indicated that “some transitory factors may be at work,” so that the FOMC expected a return to 2% “over time.” At the next FOMC meeting in June, however, Powell suggested that it would take longer than previously thought to get back to target. By the time of Powell’s Jackson Hole appearance in late August, he specifically argued that “inflation seems to be moving up closer to 2 percent.” In fact, he made that point twice in the speech.
Powell had it right the first and third time. The dip in core inflation in early 2019 has been followed by a period of considerably firmer price hikes. While the year-over-year advance should remain somewhat depressed until early 2020—when the extremely low readings from January through March drop out of the 12-month window—the monthly increases since April have been quite firm.
Outliers drove the dip
The FOMC is divided over the optimal indicator of underlying inflation. The most prominent and popularly recognized gauge is the core PCE deflator. However, a number of Fed officials, perhaps half of the Committee or more, have expressed a preference for the Dallas Fed trimmed mean PCE deflator. Whereas the core PCE deflator gets to an underlying gauge by arbitrarily labeling the entire food and energy categories as volatile and excluding them, the Dallas Fed trimmed mean calculation is much more systematic. The 178 line items in the PCE deflator are ranked by price change over one, six and 12 months and the top quartile and bottom quartile are excluded, leaving the middle 50% of line items. The trimmed mean inflation rate is calculated using only that middle half of line items.
This approach systematically eliminates the sort of special factors that have from time to time distorted the core PCE deflator, including examples like the steep plunge in cell phone services in 2017 and the drop in apparel prices in March and April of this year. When the core PCE deflator and the Dallas Fed trimmed mean gauge diverge, it means that a handful of outliers are driving the core measure either up or down. This has occurred several times in recent years, including in early 2019 (Exhibit 1).
Exhibit 1: Measures of core inflation
The narrative provided by the Dallas Fed trimmed mean is radically different than that of the core PCE deflator. On a 12-month basis, the Dallas Fed gauge has been between 1.9% and 2.1% for 19 straight months since February 2018. This suggests that the dip in the core PCE deflator in early 2019 was much ado about nothing.
The other compelling point is that in both of the previous recent cases when the core PCE deflator has fallen well below the Dallas Fed trimmed mean measure, it is the core PCE deflator that has corrected back up to the Dallas Fed gauge rather than vice versa. Indeed, even using the relatively sluggish year-over-year calculations, the core PCE deflator is already well on the way to rebounding back into line with the more stable alternative index.
Higher-frequency alternatives show the rebound
Another way of projecting the future course of the 12-month change in the core PCE deflator is to look at how the higher frequency alternatives are evolving. Unusually high or low monthly readings will continue to impact the year-over-year advance until they drop out of the 12-month window. Looking at higher frequency alternatives provides a preview of where the more broadly followed 12-month measure is headed. Recent monthly readings of the core PCE deflator have been unusually firm (Exhibit 2). In fact, the 3-month annualized increase has been well above 2% since June, and the August level of 2.5% matches the highest reading seen since 2012. Of course, this follows the extremely soft figures early in the year, which pushed the 3-month annualized pace all the way down to 0.5%.
Exhibit 2: Year-over-year and 3-month annualized core PCE deflator
Another way of thinking about the year-over-year measure is to recognize that it consists of the sum of the last 12 monthly readings. For each successive month, there is one new observation, replacing the year-ago month that drops out of the 12-month window. It is possible to have an early read on movements in the year-over-year measure, especially when there are unusually high or low monthly readings dropping out of the 12-month window. For example, the core PCE deflator posted a meager 0.02% rise in August 2018. Thus, even with a pretty pedestrian 0.14% advance for August 2019, the year-over-year advance accelerated by more than full tenth, from 1.65% to 1.77%. The last four monthly readings of 2018 were all between 0.14% and 0.19%, so the swings in the year-over-year advance should be relatively modest through the end of this year. I expect a gradual pickup, but the 12-month advance is likely to inch upward rather than jump, as it did in August.
However, unless the unusual softness in early 2019 is repeated at the beginning of next year, the 12-month advance should zoom higher after the turn of the year. The most popular gauge of underlying inflation has already accelerated from troublingly below 2% to within a few tenths of the target and is likely to push above 2%, possibly well above, by early next year before perhaps flattening out or even receding somewhat next spring and summer, when the hefty monthly readings recorded recently begin to drop out of the 12-month window.