Stock and bond market investors reached totally different conclusions today when they read the March CPI report released by the BLS.
Prices were up +0.6% in March, 3.1% over a year ago, a 4.3% annual rate of increase for the first quarter. The big number spooked the stock market; the DJIA (-115), S%P 500 (-15.9) and Nasdaq (-18.6 were all trashed. But the bond market yawned. The ten year treasury yield closed at 4.18% for the day, just where it opened, as did the 30 year Treasury at 4.55%. What gives?
In this case the bond market got it right. There is nothing in the CPI report to suggest that demon inflation is back. The big culprits were energy (+4.0% in March, +12.4% for the year), and Transportation (i.e., energy, +1.9% in March). Taken as a whole, it still looks like CPI inflation will remain 2-3% for some time.
Communication prices continued to fall (-0.2%) across the board, in spite of all the whining going on in Congress this week about mergers and too much concentration in the telecom sector. The communications and IT sectors are incredibly competitive today.
The reason people are continually spooked by the monthly price reports is they don’t have a clear understanding about how and why inflation expectations impact interest rates. Too much textbook; not enough thinking.
I have been interested in this subject for 35 years. I wrote my Ph.D. dissertation on the subject. It was years later, after I had written a book on the subject as well, that I figured out I didn’t have a clue how it worked.
The textbooks today all tell you the nominal interest rate is equal to the real interest rate plus the expected rate of inflation. They refer to this as the Fisher equation, after Irving Fisher, who supposedly invented it. Most economists believe Fisher meant that changes in inflation expectations drive nominal rates up and down, leaving real rates–hence the real economy–unchanged.
The truth is very different. In 1972, while writing my dissertation, I read every single work Irving Fisher had published, then I corresponded with his son and read his unpublished manuscripts and notes as well. (OK, call me anal.) What Irving Fisher actually believed was that changes in inflation expectations had an even bigger impact on real rates than on nominal rates, and that they had important effects on the real economy by creating credit cycles. In this sense Fisher’s work was essentially Austrian, closely tied to Wicksell’s natural rate and, somewhat later, Keynes own rate of interest.
I finally worked out the linkage between inflation expectations and interest rates on a train while travelling to give a lecture to the GE Pension Fund, run by my friend Dale Frey. Dale was so smart and such a perfectionist in his own work that I wanted to get it just right.
The connection between inflation expectations and interest rates works through the prices of storable tangible goods. If a person wants to transfer purchasing power from this period to next period, he must do so by holding something that lasts longer than one period. We call such things assets. The investor has a lot of choices. He can hold a stack of $20 bills. He can hold a bond (IOU) promising a fixed number of dollars next period. He can hold shares of stock. Or he can hold existing real goods, which he can sell or use next period.
Asset markets will force asset prices to the levels that make these alternative strategies for transferring wealth have roughly comparable results. In particular, abstracting from credit risk and tax considerations, the total return on holding the bond (the nominal interest rate) should be about equal to the total return on holding quantities of real goods, which is equal to the sum of their service value (you can live in a house or drive a car), less depreciation (goods wear out), plus appreciation, less storage and maintenance costs (rent on a garage). The appreciation term is where inflation expectations come in. If people expect the price of their holdings of houses, cars, and other tangible assets to increase by 10% next year, then the nominal interest rate must be that much higher for bondss to remain competitive with real goods as an asset.
Note that we are referring to the appreciation of the stock of existing goods, because that is what people can actually hold. The CPI does not measure this, nor does the PPI, the GDP deflator, or any other price index measuring the price of new products.
59.76% of the CPI is comprised of service prices–haircuts and guitar lessons. Services are inherently unstorable; their depreciation rate is too high. For that reason, real interest rates calculated from CPI data misrepresent inflation pressures on interest rates. For the same reason, the yield on TIPS, inflation-indexed Treasuries, tells us very little about inflation expectations or pressures on nominal rates, because the Treasury uses the CPI, not storable tangible goods prices, in their adjustments of principal.
What really matters for interest rates is sustained increases or decreases (think Japan) in the prices of land, commodities, and long-lived structures. Not one-time increases like we saw last year, sustained increases. This requires the support and cooperation of a central bank. That is not going to happen in the US today.
When the stock market gets spooked by a CPI report, like today, think of it as a going back to school sale and load up on the stocks of your favorite companies and sectors. That’s what I did today.