“Kee” Points with Jim Kee, Ph.D.

  • The Slope of the Yield Curve
  • Bullish and Bearish Flattenings
  • Difficult to “Connect the Dots”


The Slope of the Yield Curve

“Does the flattening yield curve signal a recession?” That was the title of an article in this weekend’s Barron’s, which has become required reading for most investment professionals. That could well be as much a statement of the profession as it is an endorsement of that financial newspaper (I’ll elaborate shortly). The difference between long-term and short-term interest rates is known as the “slope of the yield curve”, or “the term spread.” Much of the literature about the slope of the yield curve being a reliable predictor of future economic activity originated in the 1980s, but reference to it goes back to economist Rueben Kessel’s work in the mid-1960s (Federal Reserve Bank of New York). Kessel argued that the yield curve tended to flatten or invert at or near cyclical peaks in economic activity. The interest rates referred to in most yield curve discussions are 10-year Treasury yields at the long-end, and either 3-month Treasury rates or 2-year treasury rates at the short end. Reasons for the relationship between the slope of the yield curve and the economy often vary, but the original assertion was that longer-term rates are averages of expected future short-term rates, which would be lower if future growth (and the demand for credit) was lower. In other words, when growth is expected to slow, future short-term rates would be expected to fall. Subsequent research has shown that the predictive power of the slope of the yield curve varies over different periods; that is, it is (like many relationships) period-specific.


Bullish and Bearish Flattenings

And that is because interest rates, short and long, can be determined by different things in different periods. One of the first pieces I read on the slope of the yield curve, which continues to influence my thinking to this day, was written in the 1980s by economists Arthur Laffer and Victor Canto. A piece titled “The Yield Curve: The Long and the Short of it” that would subsequently be copied endlessly, the authors discussed how you could have a “bullish flattening” or a “bearish flattening,” depending upon the circumstances. For example, it is often useful to view the short-end of the yield curve as being primarily influenced by growth, with the long-end mostly reflecting inflation expectations. When growth is strong and inflation is falling, as happened during the 1980s, you could indeed have a “bullish flattening” of the yield curve (Federal Reserve Bank of San Francisco). This basic notion, that the meaning of a change in some variable depends upon what’s driving that change, is fundamental to economic analysis but often overlooked. Perhaps the most intuitive example is the price of oil. Whether or not an increase in the price of oil is a good sign (bullish) or a bad sign (bearish) depends upon what’s driving it. If economic growth is leading to an increase in the demand for oil and bidding up its price, then the higher price means things are going well. But if a supply restriction - perhaps because of reduced OPEC production like we had in the 1970s – is responsible for the higher price, then that would be a bearish signal. Likewise, a fall in the price of oil would be bullish if it was due to an increase in supply, and bearish if due to a fall in demand. This example is relevant to any price change, including currencies, interest rates, wages, etc. The implications of any price change depends upon whether that price change is being driven primarily by demand or by supply.


Difficult to “Connect the Dots”

Now, back to Barron’s. Each weekend issue of that paper also features an interview, and this past weekend’s interview was with economist Arthur Laffer. That’s the same Laffer who wrote the piece on the yield curve I mentioned above back in 1986. There’s no mention of Laffer in the yield curve article, and no mention of yield curves in the Laffer article! That is, the yield curve article’s author had no idea that the issue’s interview candidate was an early expert on the subject. And the interview author had no idea that a piece was being written on yield curves, or he didn’t know that his subject had published on the subject three decades ago. That shows just how hard it is to really “connect the dots” in an age of unprecedented information flow. Also, I’ve mentioned previously that the long end of the yield curve today is probably being influenced by international forces, as rates are lower overseas. The Barron’s article pointed this out as well, citing London-based “G+Economics” head Lena Komileva: “A surplus of investment funds looking for returns in low-yield global markets results in a cap on longer-term yields and a flat yield curve.”