Saturday, February 20, 2010

CPI + Velocity = Trouble

Beginning of the year economic blues in the US? I think so. Just looking over Spencer's CPI post; here is an excerpt (the first paragraph):
The CPI report was encouraging. The total CPI rose 0.2% and the year over year increase is only 2.6%. Although real average hourly earnings fell, real weekly earnings were unchanged.

The core CPI actually fell for the first time since 1982, bring the year over year change in the core CPI to 1.6%. The 6 month SAAR for the core CPI is 0.8%. Despite all the worries about inflation the normal pattern is for the best cyclical reading on the core CPI to occur in the first year or two after a recession. If the economy follows the normal pattern, the core CPI should continue to moderate for another year or two.
My first thought is that I don't think that this is encouraging at all; and I'm not alone. Core prices fell; these prices are typically very, very sticky. For example, shelter prices are biased upwards in their calculations, but have been declining or unchanged for every month since August 2009. I know that the output gap is not directly observed, except by proxy in the capacity utilization numbers or the unemployment rate; but it must be huge to do this to housing costs.

Look at it differently: the velocity of money improved in October and November of 2009...

... but then took a step back in December of 2009. If this trend continues, non-energy prices are sure to back down much further. There's just no support for price action at this time - the Fed can't pull back... it probably should be putting more in.

Rebecca Wilder

Note on data: Macroeconomic Advisers now publishes a blog where they make available their calculated monthly GDP series (nominal and real) to the public (thank you).


  1. Stop wishing, hoping, waiting. We are definitely, as of this January, in a new bear market. Money flows presage a double top.

  2. rebecca, karl denninger says he found an error in the CPI computation...

    another sinister plot developing?

  3. Errors & errant data are common. People (not the publishers), who monitor their time series, usually catch them.

  4. Hi rjs,

    I am not so sure that this weighting holds after the numbers are seasonally adjusted. The unadjusted numbers seem to be okay.


  5. Or even fudge them (as in Greece!)!!