The yield curve

The chart posted below records weekly the difference between the yield on the 10 year US treasury note and the 2 year treasury note. When the 10 year note yields more than the 2 year note this difference is positive and the (interemdiate part of ) the yield curve is said to be positively sloped. A positively sloped curve prevails most of the time as you can see from the chart.

When this yield difference is a very big positive number the curve is said to be very steep. During the summer of 2003 this yield difference was +275, matching the difference recorded in 1992.

The steepness of the yield curve in 2003 is the key to understanding why longer term interest rates have not on average risen nearly so much as short term rates during the current period of Fed tightening.

The first thing to keep in mind is that the Fed directly controls only the Fed funds rate. This is the rate at which banks can borrow overnight to meet their reserve requirements. The funds rate in turn is tightly bound to the level of very short term interest rates (e.g. 3 months or less in maturity).

The rate on the 10 year note is not controlled by the Fed but instead is determined by the market's expectation about two things: the average level of the Fed funds rate over the next 10 years (with a heavier weight given to rates expected to prevail during the next two or three years) and the average level of inflation expected over the next 10 years.

During the summer of 2003 we had two unusual circumstances working together. First, the yield curve was showing record steepness (+275). Morover, this record steepness accompanied an unusually low level of short term interest rates, a level not seen since the aftermath of the Great Depression of the 1930's.

Consequently, there was plenty of room for the Fed to return short rates to normal levels without causing a big rise in the 10 year yield. (The normal level for short term rates is the expected rate of inflation plus the expected rate of economic growth - somewhere between 4% and 5% currently). Indeed, since the Fed action should only serve to decrease inflationary expectations, and since the Fed was only returning short rates to "normal" levels, one would not expect the yield on the 10 year note to change much at all.

This is exactly what happened. When the yield curve was at its steepest in August 2003 the yield on the 3 month treasury bill was about 1 %, the yield on the 2 year note was 1.8% while the yield on the 10 year note was 4.6%. Today, after 18 months of Fed tightening, the three month bill yields 2.9%, the 2 year note yields 3.6% and the 10 year note yields 4.3%. Note that the yields on the 3 month and 2 year paper went up about the same amount - about 200 basis points - while the 10 year note yield actually dropped 30 basis points during the same time.

So far all we have done is "predict the past". What about the future?. Well, you can check out my 2005 bond market forecast posted below. Since the curve has returned to more normal levels during the past 18 months (see chart below) I expect the 10 year notes to start moving more in sync with the shorter term maturities. Sometime this summer or fall I think that both short and long term rates will start moving up together in response to a renewed campaign of Fed tightening. Rates will probably reach their peaks late in 2006.