The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

« What, Exactly, Do Central Banks Control? | Main | Household Inflation Expectations: No Improvement »

October 14, 2005

The CPI Report: Very Interesting

The headline CPI report for September can be described as nothing short of eye-popping.  From the Federal Reserve Bank of Cleveland:

Earlier today, the BLS reported that the seasonally adjusted CPI for all urban consumers rose 1.2% (15.7% annualized rate) in September.

Adding fuel (no pun intended) to the core-or-not-to-core debate, the inflation signal coming out of the index excluding food and energy was completely benign:

The CPI less food and energy rose 0.1% (1.2% annualized rate) on a seasonally adjusted basis.

That same message also came through the Cleveland Fed's core inflation measure, the median CPI:

According to the Federal Reserve Bank of Cleveland, the median Consumer Price Index rose 0.1% (1.7% annualized rate) in September.

The story in a table:

Percent Change From Previous Month
Apr. May June July Aug. Sept.
CPI 0.5 -0.1 0.0 0.5 0.5 1.2
CPI less food & energy 0.0 0.1 0.1 0.1 0.1 0.1
MEDIAN CPI 0.2 0.3 0.2 0.2 0.2 0.1
Percent Change, Last 12 Months
Apr. May June July Aug. Sept.
CPI 3.5 2.8 2.5 3.2 3.6 4.7
CPI less food & energy 2.2 2.2 2.0 2.1 2.1 2.0
MEDIAN CPI* 2.3 2.4 2.3 2.3 2.3 2.3

So, the trend in the headline CPI is moving up, the trend in the CPI ex food and energy is moving down, and the median is hanging in right where it has been most of the year.

What to make at this?  I hate to say this, but it's hard to tell.  A sense of why it is so hard can be appreciated by looking at the variation in the individual prices measured in the CPI:


That variability is being driven by price changes spreading in both directions.  In other words, more slow-changing prices as well as more fast-changing prices:


My favorite inflation expert, Mike Bryan, sums it up this way:

... more than 40 percent of the retail prices increases were 3 percent or more, while at the same time almost 20 percent of the index showed either no change or outright declines. In other words, we have a very "tail heavy" distribution of price adjustments--prices are showing either troubling increases, or softness, but very few are in the middle, moderate range. This means that these price averages (including the median) could swing, perhaps dramatically, in either direction over the next month or two. Consider that the 16 percent trimmed-mean CPI was up 3.6 percent last month and 2.5 percent over the past 12 months--big jumps from August.

The "trimmed-mean CPI" just means the inflation rate that is calculated after eliminating those prices that fell the most and those prices that rose the most.  It is a calculation that is similar to one done by the Dallas Fed with the PCE price index.  The Dallas gang found the same large jump in that measure of core inflation between July and August, so this is not an isolated observation.

To make matters more confusing, Mike notes this, from the BLS details:

Of those components outside of energy that showed large price increases in September, many were in areas where seasonal factors at this time of year are large, which means that they may be subject to some sampling problems. Notable among these were large price hikes for education (tuition and books) and new cars (the rebates on some models came off.)

Put it together and whaddya got?  Not so clear.  The market seems to have interpreted the report in a positive light...

The September consumer price report offered some relief to a market concerned by building inflationary pressures in the economy.

... but I'm not sure I'm quite ready to join that parade.

Honorable Mention: The pictures above come to you via the good graces of the-always-hustling-on-CPI-day Linsey Molloy.

UPDATE: If you pay any attention to anything other than the overall inflation numer, Barry Ritholtz wants to smack you

The Capital Spectator has a sense of deja vu. TCS suggests that between inflation and the short-term pain restrictive monetary policy to fight it, he is most concerned about the former. James Hamilton sees things differentlyWilliam Polley is just plain concerned.

Daniel Gross focuses on the drop in real weekly earnings.  So does pgl at Angry Bear. 

Everyone's Illusion surveys the bond market reaction to all the economic news of the day.  The Skeptical Spectator also puts the CPI in the context of the day's full set of economic news, and discusses the role of trade patterns in muting some of those individual price changes.

The Prudent Investor thinks we should be more the money growth numbers.  (PI shares my blog reading list, too.)

October 14, 2005 in Inflation | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference The CPI Report: Very Interesting :


Please excuse this dumb question:

One factor greatly excluded is purchased housing, as owner-equivelent-rent has stayed flat while prices have skyrocketed.

What would it take to measure the REAL CPI for housing? Some mix of renting vs new purchase vs sustained service (some ARM, some not, effects of appreciation on property tax)?

What would this look like?

Posted by: Nicholas Weaver | October 14, 2005 at 05:16 PM

Well, sure, let's esclude food and energy, cause nobody really NEEDS those thing, do they?


Posted by: donna | October 15, 2005 at 09:13 PM

Nicholas -- In the current environment I guess that it would take something like rents rising faster than energy costs to get the owener-occupied rental equivalents to show really substantial increase. My understanding is that that these rents are imputed exclusive of things like implied utility costs. So if those costs are rising faster than rents, that damps the rate of increase in the the owner-cccupied housing service imputations. There are probably those out there with a better handle on this than I, though. (You can read about the BLS approach to calculating owner-occupied rental equivalents here: http://www.bls.gov/opub/mlr/1996/12/art5full.pdf)

donna -- I took my shot at defending the core here: http://macroblog.typepad.com/macroblog/2005/09/fiddling_with_c.html

Would it help if I told you I buy gas too, and therefore feel -- I mean really feel -- your pain?

Posted by: Dave Altig | October 16, 2005 at 06:48 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search

Recent Posts



Powered by TypePad