By Dr. Harold A. Black
blackh@knoxfocus.com
haroldblackphd.com
At the end of March, the stock market panicked when the rate on the two-year Treasury briefly rose above the rate on the ten-year Treasury bond. This phenomenon is called an inversion of the Treasury yield curve and is thought to predict a recession. One market watcher said that the inverted yield curve meant that a recession was two-thirds likely in the next year and 98 percent likely within two years.
What is the Treasury yield curve? It shows how interest rates on Treasury’s vary with the time it takes the instruments to mature. The curve plots yield against time to maturity and typically slopes upward. That makes intuitive sense. One would expect that the longer it takes an instrument to mature, the greater the risk of holding it and the higher should be the yield. In Finance, it is generally assumed that the yield curve embodies expectations about the future. Consider that an investor has the choice of buying short term securities or longer term securities. Shorter term instruments are less risky so why would the investor buy a longer term instrument? One explanation is that the yield on the longer term instrument must be greater than if the investor bought a series of shorter instruments over the same time to maturity as the longer instrument. Thus, there will be a positive spread between long rates and shorter rates.
Suppose there is inflation and the inflation is not “transitory” and is expected to last longer. If the Fed moves to raise interest rates it does so by selling Treasury bills which are short term. The investors holding longer term instruments are no longer being adequately compensated because their bonds pay a fixed return and the differential between the shorter term rates and the longer term rates narrow. Of course, newly issued long bonds will pay higher rates but if shorter rates increase more, then the yield spread narrows and, like last month, could go negative.
Since 1955, an unambiguously downward sloping yield curve has preceded every recession. However, not every yield curve is unambiguously downward sloping. The curve could briefly invert as it did last month rather than invert for a longer period. The curve could invert in one time segment and not throughout its length. Most research shows that when the curve does invert, recessions occur with a lag – an average of 18 months. The lag has been as short as three months and as long as three years. What is interesting is that today both consumers and the market anticipate that inflation will linger which makes a recession more likely. Much depends on the actions of the Fed and whether it will counter the profligate spending of this Administration.
Also, the Fed’s keeping long term rates low since 2008 means that the yield curve is flatter than it would be otherwise so a run up in short rates may not augur a recession. However now the Fed is pushing up short term rates while selling off its portfolio of accumulated longer term instruments. There is a higher likelihood of more uncertainty and turmoil in the financial markets over the next year or so.
Lastly, the market looks at the yield spread on the two year Treasury and the 10 year Treasury. But is that the relevant spread? Researchers at the Federal Reserve of San Francisco contend that the market is looking at the wrong rate spread. They find that the difference between the three month Treasury bill and the 18 month forward rate is more predictive than the difference between the two year and ten year Treasury. That measure is not inverted and has been found to be more predictive than the yields looked at by the market. Fed chairman Powell agrees and said in a speech that the shorter end of the yield curve is more predictive than the longer end. Currently the shorter spread is steep while the longer end is flat. Thus, the shorter end spread is not predicting a recession in the near term. Powell says that although he looks at the traditional spread, he pays more attention to the shorter end differential. So, is the market wrong? That is an empirical question and only time will tell.
