« New Economist: Productivity, the Fed, and the Use of Real-Time Data | Main | Edward Prescott: Privatization of Social Security Should Happen and Will Happen »

Saturday, October 22, 2005

Is Core Inflation a Systematically Biased Measure of Actual Inflation?

Does core inflation systematically underestimate actual inflation as many believe? Here are graphs of the CPI along with the CPI less food and energy, and the PPI for finished goods along with the PPI less food and energy through September of this year (the starting dates are different in the two graphs due to data availability). Core inflation is less than actual inflation recently due to rapidly increasing energy costs, but overall it's difficult to detect a systematic bias in either direction:

[Update: These are twelve month (year-over-year) inflation rates but the graphs don't change much with other measures, e.g. averaging monthly rates.]

[Update: I added graphs showing the differences between the all item and core rates of inflation for both the CPI and the PPI as well as a two-scale graph showing the accumulated differences for the two series. If there is systematic bias, the accumulated differences should trend upward or downward. Such trending is not evident. The deviations from zero are, however, highly persistent. The graphs are in the continuation frame.]

[Update: This Economic Scene from The New York Times discusses the use of core versus headline inflation, something that has been discussed extensively on blogs, e.g. macroblog has done a lot on this (Google ''core inflation'' at macroblog).

    Posted by on Saturday, October 22, 2005 at 12:52 AM in Economics, Methodology, Monetary Policy | Permalink  TrackBack (0)  Comments (7)


    TrackBack URL for this entry:

    Listed below are links to weblogs that reference Is Core Inflation a Systematically Biased Measure of Actual Inflation?:


    Feed You can follow this conversation by subscribing to the comment feed for this post.