As a result of large declines in goods and energy prices, both the all items Consumer Price Index and the ex. food and energy CPI fell for the month of November. The headline measure was down 2.6% at an annual rate (a.r.), while traditional core declined 0.6% (a.r.). Meanwhile, the less volatile Median CPI computed by the Federal Reserve Bank of Cleveland increased by 1.6% (a.r.) for the month. Over the last 12 months, inflation was +1.8%, +1.1%, and +2.0% as measured by the all-items, ex. food and energy, and median CPI’s respectively. Inflation is now roughly equal to the size of most estimates of measurement bias, so this really is price stability.

The details of this morning’s report contain some surprises for most people, especially considering that most forecasters where expected the CPI to show an increase this month. First, the fact declines in goods price inflation continued to increase. Given the depreciation of the dollar (which I feel acutely paying for my dinners with euros here in Frankfurt) it is surprising that goods prices continue to fall. Core goods (commodities excluding food and energy commodities) fell at an annual rate of 4.2% for the month, which is bigger than the average drop over the last 3, 6, or 12 months. Meanwhile, core service prices rose an extremely modest 0.5% (a..r.) in November – far below the 2.6% average increase over the past year. This second fact is somewhat less surprising given the continued slack in the labor market and that wages are the largest cost in producing most services.

The deceleration in owner-equivalent rent (OER) is not what was expected either. (Once again, OER is the median good.) While there is some evidence that the housing and rental market has cooled down a bit, the fact that OER is a twelve-month moving average attenuates this significantly. And then there is the technical issue that arises because of the way energy cost is treated in the computation of rent. In the August edition of the inflation update I noted that since landlords pay utilities in many cases, the BLS assumes that the "rent" component equals what renters pay minus the cost of energy. This means that when energy prices fall, as they did this month, "rent" rises – and so should the OER that is computed from a survey of rents. Instead, OER inflation fell.

Overall, this morning’s report really is quite soft. It allows us to extend further our estimate of the "considerable period" before the FOMC will see the need to raise its federal funds rate target. It is now extremely difficult to see how conditions for a rate increase can emerge before early 2005. But then things are likely to move quite fast, with the funds rate going back to its equilibrium level of something like 5% by early 2006. (A good estimate of that long-run level of the funds rate is the productivity growth rate of 2½ to 3% plus the FOMC’s implicit inflation target, which I believe be between 2 and 2½%.) This makes me wonder about the current level of longer-term interest rates. Today, the constant maturity 10-year U.S. Treasury bond rate is 4¼%. But, during "normal" times, this rate is 1 to 2 percentage points above the funds rate, so by early 2006 it will have to be between 6 and 7%. Maybe there’s something wrong with my arithmetic, but even taking today’s inflation report into account, long-term bonds look like they are over priced. I guess I’ll keep my 30-year fixed rate mortgage, at least for now.

Consumer Price Inflation, Various Measures
(Through November 2003, all data s.a. at an annual rate)

For previous updates, as well as my occasional essays on current policy issues,

Please visit my home page:
Steve Cecchetti's Homepage

(Note: If you have trouble viewing the tables, you may prefer looking at them in html at
The Inflation Update: November 2003
or you can download the pdf file
The Inflation Update: November 2003 PDF Format
These also include the table used to construct the Median CPI.)