Han Dieperink: Yield curve cannot predict recession
Han Dieperink: Yield curve cannot predict recession
This column was originally written in Dutch. This is an English translation.
By Han Dieperink, written in a personal capacity
The yield curve in the United States has been inverted since July 2022 due to the rapid rise in short-term interest rates. Since June 2023, the eurozone has been in an inverted curve. Normally, the difference between the ten-year interest rate and the two-year interest rate is taken into account, but there are different variations on this. The long-term interest rate can vary from ten to thirty years and the short-term interest rate from three months to two years.
As soon as the long-term interest rate dives below the short-term interest rate, the curve is inverted. That should be a strong indicator of a recession, because the 'market' shows through long-term interest rates that central banks have done enough to get inflation under control. After all, a bond investor's biggest enemy is inflation. And until recently, everyone believed that the only way to curb inflation was to cause a recession. Yet the yield curve is a poor predictor.
The position of the yield curve
The yield curve is a line that plots the interest rates of bonds with different maturities and equal credit quality. In general, a yield curve is upward-sloping, with short-term bonds offering lower returns and long-term bonds offering higher returns. In other words, you should be compensated with higher returns if you tie up your money for a long time. Sometimes a yield curve can invert and slope downward. When this happens, short-term bonds have a higher yield than long-term bonds and investors are not rewarded for tying up their money for longer.
Yield curve as a predictor
Much has been written about the predictive value of the position of the yield curve (see for example the Journal of Finance, Annual Review of Financial Economics, or NBB Economic Review), perhaps also because economists themselves are very bad at predicting a recession. They are massively looking for guidance and the advantage of an inverted yield curve is of course that it is objectively measurable. Moreover, a high short-term interest rate (regardless of the position of the curve) also works as a self-fulfilling prophecy. Then there is a reversal of causality. An inverted yield curve is then not the predictor, but the cause of a recession.
This is in line with Knut Wicksell's ideas: if interest rates are high enough, a recession will automatically follow. The amount is relative, because it depends on the natural or equilibrium interest rate. Given the higher future growth potential of the global economy, the question is whether interest rates are already high enough for this. The faster growing parts of the global economy are becoming increasingly important and the AI revolution will further increase productivity and thus growth. Evidence that interest rates are not yet high enough is that there is currently an acceleration of growth in the US, which now also appears to be followed by higher growth in Europe and Asia.
Track record
All US recessions since 1970 have been preceded by an inverted yield curve. However, there were also many false signals and sometimes there was a very long time between the start of the recession and the inversion of the yield curve. There were false signals in 1966, 1978, 2006 and 2019. Now, after the inversion of the yield curve in 2019, a recession followed, but it was caused by the COVID-19 pandemic. That an inverted curve is associated with a recession can only be maintained by someone who dares to believe that the origins of that pandemic were caused in the financial world. If there had been a recession after every yield curve inversion, there would have been eleven recessions in the US since 1970. However, there were only seven. That is not a strong average and not statistically significant.
This time it's different
Since 2008, something special has been going on in the bond market. Due to the phenomenon of quantitative easing, the central bank not only intervenes on the short side of the yield curve (via the policy rate), but for a long time central banks were also the largest buyers on the long side of the curve. Central banks are still by far the most important players in that market. Moreover, it is not only the central banks that are putting pressure on the long side of the yield curve. For example, since the financial crisis, the capital requirements for commercial banks have been increased significantly, which also creates more demand on the long side.
Furthermore, institutional investors (pension funds and insurers) are increasingly focusing on liabilities this century, often regardless of the return. Under the heading of liability-driven investments, long-term bonds were purchased en masse at historic low interest rates, sometimes even with four times leverage. The curve has therefore become completely decoupled from the underlying economic development.
A fairly reliable recession indicator was the British stock market, mainly due to the many early cyclical companies in that market. Between 1997 and 2008 the curve was continuously inverted, without any significant slowdown in growth or recession. So there are many more explanations why the yield curve is inverted this time. This also reduces the chance that such an inverted yield curve will be a good indicator of a recession this time. As in recent years, there will be no recession this year.