Behind the Numbers: PCE Inflation Update, December 2010
This update, prepared by Dallas Fed Senior Economist Jim Dolmas, provides an in-depth analysis of the latest personal consumption expenditures (PCE) inflation data. Updates will be posted monthly, following the release of the official PCE data by the Bureau of Economic Analysis. NOTE: Terms in bold are defined in the Inflation Update Glossary.
Boosted by a sharp increase in the price of gasoline, headline PCE posted a one-month annualized inflation rate of 3.8 percent in December, its highest one-month rate since June 2009’s 6.8 percent. The 12-month headline rate ticked up from 1.1 percent to 1.2 percent, still quite low by historical standards. For a 12-month calendar period—December to December—only 1961, 1998, 2001 and 2008 showed lower headline inflation rates than 2010’s 1.2 percent.
In contrast to December’s robust one-month headline rate, core PCE posted a negligible gain of just 0.4 percent at an annualized rate, down from 1.1 percent in November. Over the last half of 2010, only November recorded a non-negligible rate of increase (one that, expressed as a monthly rate, would round to 0.1 percent or better). The 12-month core rate ticked down from 0.8 percent in November to 0.7 percent in December.
Core PCE suggests that, apart from November, the second half of 2010 was characterized by a nearly zero rate of underlying inflation. The impression one gets from the trimmed mean, though, is a bit different. December’s trimmed mean inflation rate came in at an annualized 1.2 percent, not far from November’s 1.4 percent. Both readings are up from October’s very low 0.2 percent. In fact, for the trimmed mean, over the last half of 2010, it’s the negligible October number that looks aberrant. The 12-month trimmed mean rate held steady at 0.8 percent for a third straight month.
To be sure, both the core and trimmed mean suggest that underlying inflation is quite low and is lower now than it was in the middle of 2010. They disagree on whether that trajectory is still pointed downward. That disagreement will only be resolved as more data accumulate over the next few months.
Gasoline Accounted for Most of December’s Headline Rate
From November to December, the price index for gasoline and other motor fuels (an average of all grades of gasoline and diesel fuel) increased 7.7 percent, or at an annualized rate of roughly 144 percent. Compared with December 2009, the motor fuels index is up nearly 14 percent, though it remains about 20 percent below its July 2008 peak.
Of December’s headline PCE rate of an annualized 3.8 percent, roughly 3 percentage points came solely from gasoline and motor fuel. Based on weekly data from the Department of Energy, we can expect another positive boost—though about half as big—in January. Given four weeks of DOE data, the average retail price of gasoline is tracking at about 3.3 percent above its level for December. That number is not seasonally adjusted. The normal seasonal pattern in gasoline prices would call for a slight decline—about 0.2 percent—in January. A 3.3 percent increase when the normal seasonal change is a 0.2 percent decline means a seasonally adjusted increase of 3.5 percent—just under half the size of December’s increase.
Price Gains for More-Processed Food Items Smaller in December
The index of food and beverages purchased for off-premises consumption—food at home, for short—increased at a 1.6 percent annualized rate in December, up from a 1.3 percent rate in November, and a bit above the component’s 12-month rate of increase, also 1.3 percent. The December gain combines a 3.5 percent annualized increase, on average, for items at the less-processed end of the food spectrum—led by a nearly 50 percent annualized increase in the price of fresh fruit—and a 1 percent annualized increase for more-processed items. This is consistent with the pattern we’ve seen over the past year, during which our index of less-processed items rose 4 percent and our index of more-processed items rose a scant 0.2 percent.
As regular readers of the “Inflation Update” know, we tend to pay more attention to the behavior of food prices at the more-processed end of the spectrum, as a signal about underlying inflation trends. In that regard, while December’s 1 percent annualized increase is a step down from November’s 1.8 percent, it’s still consistent with the general pickup in rates of increase we observed over the second half of 2010, off a string of declines over the first half of the year.
A Few Items Had Outsized Impacts on December’s Core Rate
As noted above, core PCE increased a scant 0.4 percent, annualized, in December. Underlying this increase was a fairly steep drop in prices for core goods (–1.5 percent, annualized) and a very tame increase in the prices of core services (1.0 percent, annualized). On a 12-month basis, the core index is up just 0.7 percent (down from 0.8 percent in November), with core goods and services registering 12-month changes of –0.9 percent and 1.3 percent, respectively.
The price index for financial service charges, fees and commissions made the largest negative contribution to the core in December. That index fell at a roughly 13 percent annualized rate and alone subtracted about 0.4 annualized percentage points from the December core rate. The indexes for sporting goods, computers and communication services also made significant negative contributions, each subtracting about 0.2 annualized percentage points.
At the other end of the spectrum, the price index for the final consumption expenditures of nonprofit institutions serving households—surely the PCE component with the most awkward name—made the largest positive contribution (about 0.3 annualized percentage points), while the very volatile indexes for jewelry and air transportation each added about 0.2 annualized percentage points.
Rent Growth Continues to Firm
In last month’s “Inflation Update,” we were encouraged by continued firming in the rates of increase for rent and owners’ equivalent rent (OER), two items that have very large weights in the core PCE index. Rent growth in December remained robust, coming in at an annualized 2.7 percent rate, after posting an annualized 2.6 percent rate in November. Those are the fastest one-month rates of growth we’ve seen in nearly two years. January 2009, with an annualized 3.2 percent, was the last time rent growth exceeded November and December’s rates.
OER, which tends to follow rent, increased at a 1.1 percent annualized rate in December, its third consecutive month at or above an annualized 1 percent rate.
Trimming Out Big-Impact Items Shows More Stable Underlying Inflation
As discussed above, December’s negligible core reading suggests a return, after a one-month respite in November, to a pattern of near-zero rates of inflation—in fact, at the monthly rates often reported, rounded to one decimal place, the core rate has been zero in five of the last six months of 2010. November’s 0.1 percent monthly rate (or 1.1 percent annualized) looks like an aberration.
Imagine for a moment, though, what the picture would look like if the signs of the contributions of some of the big-impact items listed above were reversed—if the index for financial service charges made a +0.4 percentage point contribution, or the index for the expenditures of nonprofits made a –0.3 percentage point contribution. Flip a few of the pluses to minuses (or vice versa), and we might have a very different impression of the December numbers—we might see instead a continuation of the firming that seemed to occur in November, or we might see an even starker signal of continued disinflation.
When a few items can have such outsized influences, it may be a good idea to abstract from all extreme price movements. This is what the trimmed mean does, excising all the big-impact items listed above, and more. The picture that emerges from the trimmed mean is a bit different from the one that comes out of the core—December’s 1.2 percent trimmed mean rate looks a lot like November’s 1.4 percent rate. If there’s an aberrant month, it’s October, with its very low 0.2 percent rate.
Number of Falling-Price Components Little Changed from November
Consistent with the stability of the trimmed mean over November and December, the number of falling-price components—out of the 178 components we keep track of to construct the trimmed mean—was little changed in December. Sixty-three of the 178 components, or about 35 percent of them, registered declines in December, compared with 65 (or about 36 percent) in November.
Both of those readings are down significantly from October’s 85 decliners—about 48 percent of the basket—which was a record high in our data, going back to 1977.
January 31, 2011