NOTE TO MACROBLOG READERS: Dave has allowed me another guest shot at his blog today. The comments are mine, not his, and certainly not those of the Federal Reserve Bank of Cleveland or the Federal Reserve Board of Governors. ~ Michael F. Bryan
On Tuesday, monetary policymakers warned that “possible increases in resource utilization as well as elevated energy prices have the potential to add to inflation pressures.” Clearly, inflation is on the FOMC’s radar screen. And I would guess that all across the Federal Reserve System, inflation sirens began blasting within minutes of the release of the November CPI report. It’s true that retail prices dropped sharply last month as energy prices retreated. (Gasoline and home heating fuel posted huge declines, for example.) But good news on the cost of living was a bit hard to see elsewhere in the data and the CPI excluding food and energy posted a troubling 3 percent annualized increase for the second straight month.
In my office at the Federal Reserve Bank of Cleveland, I have thumb-tacked to my wall a yellowed headline from the front page of a local newspaper. It reads; “Prices up; Inflation still down”. I’ve kept it over the years because it’s a seemingly incongruous statement. But if you’ve read any of my earlier posts on inflation (like here), you’ll know it’s an assertion I often make. That is, it’s common for the CPI to go in one direction and underlying inflation to go in the other. And that’s just what we got in this morning’s CPI report, which showed a sharp drop in overall retail prices last month, while at the same time hinting at a rise in the inflation trend.
In the retail price reports of September and October, I noted to my boss that the underlying distribution of price changes was unusual. Nothing, it seems, was average. This little chart makes my point.
In the fall, retailers were either showing price restraint (increases of less than 2 percent) or lack of it (price increases of more than 3 percent.) Only 1 in 20 of the weighted CPI components actually showed a price increase in the vicinity of 2 to 3 percent, where virtually every economist thinks the underlying inflation trend is. I warned my boss that under these conditions, judging the inflation trend is especially difficult. With so much “weight” on either tail of the price-change distribution, my favorite measures of inflation (the trimmed mean estimators that you can read about here) have a tendency to flop back and forth, whipsawed by the two gigantic tails on either side of the price-change distribution and making them uncharacteristically unstable inflation indicators. I told her that I wasn’t at all confident that we were seeing much of a rise in the inflation trend, but warned that this sanguine view could change quickly as the data began to settle down and take on a more characteristic distribution.
And what we see in the November data is that characteristic distribution. Note that unlike this fall, the November retail price report shows a predominance of price increases in the 2 percent to 3 percent range. This is the kind of distribution that I believe makes the trimmed mean measures of inflation more reliable gauges of the momentum in the retail price data and deserving of more consideration when judging whether a shift in the inflation trend is developing.
So, what do these measures of core inflation tell me about this morning’s CPI report? The median CPI and the 16% trimmed mean CPI were both higher than they were in the fall (the median CPI rose 2.6 percent and the trimmed mean 2.9 percent while their 12-month growth rates also ticked up, to 2.4 percent and 2.6 percent, respectively.) These measures tell me to be a little less comfortable with my long-standing view that inflation has been holding steady around a 2 percent trend and that inflation (as measured by the CPI) may be creeping upward to 2 ½ percent. No, I don’t think the inflation trend is quite as high as the 3 percent level suggested by the CPI excluding food and energy. Further, I want to make clear that this inflation warning is based mostly on one month’s worth of data—not nearly enough on which to make a very definitive statement. Still, while I am not yet prepared to argue that inflation is on the rise, this morning’s November CPI report certainly caused the warning siren in my office to go off.