Krugman and Bond 36,000

Ten years, I criticized the Dow 36,000 thesis for its bad math. Now I very cautiously criticize Paul Krugman’s defense of the bond bulls.
Professor Krugman had a post earlier this week saying that the low yield on the 10-year T-bond is justified. He backed up this claim by taking the CBO’s 10-year projections for unemployment and core inflation and ran that through Greg Mankiw’s Fed rate rule.
The output was a projection for the Fed funds rate for the next 10 years. What’s interesting is that it shows that the Fed ought to keep rates near 0% for a few more years.
The problem I have with Krugman is what he did next. He used the 10-year forecast of short-term rates to arrive at his estimate of what the 10-year yield ought to be—2.6%.
The problem with this is that the yield on the 10-year bond is not solely the summation of 10 one-year bonds, or 40 three-month bills, or whatever time slice you like to use. There’s also an added “term premium.” This is what makes the yield curve, well…curve.
I looked at the historic spread of the 10-year bond over the effective Fed funds rate since 1954 (chart below). The spread has average 92 basis points. Bear in mind that this is a premium that’s over and above what the market has already factored for future rate increases. This is simply the premium you get for holding longer term debt.
Assuming a 10-year bond with a coupon of 2.6%, an increase in yield of 92 basis points translates to a drop in price of close to 8%. That may not sound like a lot but it’s a big move in the bond market and it’s especially large if you’re pricing in an historic bull market for bonds.
This ends today’s portion of my day when I criticize Nobel Prize winners in their field.

Posted by on August 27th, 2010 at 9:55 am

The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.