« Thinking About The CPI | Main | Poland Dodges A Bullet »
March 16, 2006
CPI -- Cool... CPI ex Food and Energy -- Cool... Median CPI -- Not So Cool
The market apparently liked today's report on consumer inflation in February, but if you feel any affection for the median CPI as a measure of core inflation, things are not as rosy as the reported numbers suggest. From the Cleveland Fed:
According to the Federal Reserve Bank of Cleveland, the median Consumer Price Index rose 0.3% (3.5% annualized rate) in February. The median CPI is a measure of core inflation calculated by the Federal Reserve Bank of Cleveland based on data released in the Bureau of Labor Statistics’ (BLS) monthly CPI report.
Earlier today, the BLS reported that the seasonally adjusted CPI for all urban consumers rose 0.1% (0.6% annualized rate) in January. The CPI less food and energy rose 0.1% (1.8% annualized rate) on a seasonally adjusted basis.
Over the last 12 months, the median CPI rose 2.5%, the CPI 3.6%, and the CPI less food and energy 2.1%.
The summary table:
A look at the distribution of individual price changes (weighted by expenditure shares) reveals why the median is giving a less benign signal than the average. Here's how things looked in January...
... and here is how they looked in February:
So, while there has been a noticeable shift away from very high rates of change toward very low rates of change, there has also been a large shift from the moderate 2-3 percent range to the less moderate 3-4 range. Personally, I'm not feeling a lot of love for this picture.
With thanks to Linsey Molloy, the data above, and more, if you are interested:
Download Histogram_0603.ppt
Download Histogram_0603.swf
March 16, 2006 in Inflation | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef00d834250e5a53ef
Listed below are links to blogs that reference CPI -- Cool... CPI ex Food and Energy -- Cool... Median CPI -- Not So Cool:
» macroblog: Median CPI is Not Cool from Economist's View
David Altig at macroblog has the inflation news, and he say it's not as rosy as many reports suggest: macroblog: CPI -- Cool... CPI ex Food and Energy -- Cool... Median CPI -- Not So Cool: The market apparently liked [Read More]
Tracked on Mar 17, 2006 2:52:36 AM
» Inflation concerns from Econbrowser
The Bureau of Labor Statistics reported that inflation as measured by the seasonally adjusted consumer price index for all urban consumers rose only 0.1% in February (a 1.2% annual rate),... [Read More]
Tracked on Mar 19, 2006 12:00:47 PM
Comments
Posted by:
bailey |
March 16, 2006 at 08:10 PM
I think you overprocess the data. Annualizing month to month data multiples tiny errors. Then it's hard to tease out seasonal and seasonally adjusted and other influences. I increasinglt like the idea of average consumers and reporting their costs. 24 year old single urban male renting and leasing a car. 35, family with one child and a starter home. Etc.
As an alternative, instead of the price of a 27" color TV, the price of the median TV. Or computer or automobile or loan. Then as tastes change the median will follow it.
Posted by:
Robert Cote |
March 17, 2006 at 12:02 PM
bailey -- All I saw were Governor Kohn's prepared remarks on whether or not a policymaker ought to respond to asset prices. Like most of his colleagues his answer (which he has given before) is probably not-- and I would have to say I concur. I think what you are referring to muct have been in Q&A.
Robert -- I completely agree that it is foolish to read a lot into one-month's data. And if you look at the full powerpoint or flash presentation that I posted above, you can see that the distribution has been shifting all over the place in the past year. But it did strike me that the reaction to the report was overwhelmingly positive -- I just wanted to say "not so fast."
Posted by:
Dave Altig |
March 17, 2006 at 05:14 PM
From the Cleavland Fed:
* Seasonal factors have been recalculated by the BLS to reflect developments in 2005. As a result, some of the figures in these tables differ from previous press releases.
I want changes like this to include what the numbers would be without the changes for at least a year. I don't recall where I saw it but I read that cumulative inflation is underreported by 71% relative to what it would be if we had stuck with the Carter era formula.
Posted by:
Robert Cote |
March 18, 2006 at 08:58 AM
Robert Cote, I saw & posted the change in inflation reporting since Carter because I find it astounding. The author is John Williams, Dartmouth MBA Economist turned Gov't. statistics tracker(maybe about same time Dave Altig was at Brown). http://www.gillespieresearch.com/cgi-bin/bgn/ will take you to Williams' home base.
Posted by:
bailey |
March 18, 2006 at 12:10 PM
Interesting, when combined with Kohn's statement today (as reported by Marketwatch.)
Obviously, the equity markets would LOVE to see the Fed stop raising rates YESTERDAY. I have no doubt they will then immediately start clammering about the need for the Fed to reverse recent increases. Kohn suggests that whenever the Fed stops it may not be inclined to further accomodate the markets. I loved it, until he moved into Greenspan "nonsense" speak leaving me wondering if even he has a clue what he's saying. (I do hope the Fed's taking note of how much of financial center profits are trading related (& how good they're getting at it.) Care to translate any of Kohn's comments, Dave?