Tuesday, September 29, 2009

Getting it wrong about the CS-CPI

- by New Deal democrat

One of the most interesting alternate measures of the economy was first posted a couple of years ago by Tim Iacono of The Mess that Greenspan Made. The CS-CPI is the consumer price index, with the Case Schiller house price index substituted for owner's equivalent rent. In that manner it captures the impact of price changes in the biggest asset most consumers will ever buy, on the inflation rate.

I have mentioned several times how research into the Roaring Twenties and Great Depression showed that it was when the year-over-year rate of inflation (really, deflation) bottomed and started back up, that the economy started to recover. I correctly suspected that the same scenario would play out this summer.

But I always suspected that the "real" bottom in the economy might be when the CS-CPI bottomed, meaning that the deflationary pressure on all things bought or sold by average Americans, including houses, was beginning to ease. I hadn't read any updates about the CS-CPI since February, but a post within the last week by Mish, claiming that the CS-CPI was at an all-time low, caused me to take another look.

Well, it turns out that Mish simply reprinted the very same graph, and the very same text, he had used back in February -- and it makes a big difference.

I have calculated the CS-CPI for January of this year (cited by Mish) and compared it with the CS-CPI now (calculating OER as 25% of total CPI, and once obtaining the value for the remaining 75%, re-doing the calculation using the Case-Schiller index, just updated this morning). It turns out the CS-CPI bottomed earlier this year. In January the CS-CPI was (-10.1%). Now it stands at (-5.3%).

In summary, the deflationary pressures on average American families are decreasing. And Mish's latest post is simply wrong.

3 comments:

Joe Zupan said...

NDD,

You go! I stopped reading Mish some time ago because all too often he seems to be 1) unremittingly negative about EVERYTHING!; and 2)a bit lazy about supporting his many, many, many assertions.

Once again, kudos and thanks.

SilverOz said...

I do not really like the CS-CPI, as I think OER actually tells us more about what is going on with current prices than CS does. My reasoning for this is that the vast majority of homeowners simply aren't seeing their costs change on a yearly basis and thus the effect of actual house prices on inflation is best measured through OER as opposed to CS.

Gary said...

There are big disconnects between rental prices and home prices. Technically they should both be included via some share of the market calculation but I don't see anyone advocating this. Of course, the CPI has been messed with since Reagan several times under both Democrats and Republicans to keep Social Security payments and other government outlays lower. See shadowstats.com.