The U.S. Treasury bond yield curve has been in the news again after recently turning negative. The financial media is trumpeting this event as an additional indicator that a recession will likely follow with a further concomitant stock market decline.
I suggest that this indicator is not especially useful in predicting either one.
The yield curve is a plot of current U.S. Treasury bond interest rates (Y-axis) based on their maturities from two-year to 30-year (X-axis). Ordinarily, shorter-maturity rates are lower than longer-maturity rates because of the time value of money: When you buy a 10-year Treasury bond, you are lending money to the government for 10 years, hence the demand for a higher return as compared to buying a two-year bond.
On a graph, this would appear as an upward-sloping curve. Occasionally the slope will invert (that is, short-term bond rates become higher than long-term) when the Federal Reserve, which only has the ability to control short-term rates, raises them faster than the broad market’s view of what the long-term rate should be.
For simplicity, the Federal Reserve calculates the yield curve numerically as the difference between the two-year and the 10-year rates. As of this writing, the 10-year has fallen approximately 0.25% below the two-year. The last time an inversion occurred was in August 2019, and it was followed by a recession about six months later. In fact, every yield curve inversion since the mid-1970s has been succeeded by a recession. That’s why many view this event as a recession predictor.
I don’t believe the yield curve is a sufficient recession indicator for several reasons. First, the cause of an inversion is generally a Fed-generated interest rate increase, typically for the purpose of controlling inflation. If they raise rates too aggressively, a recession could certainly be an outcome.
But could it be a coincidence sometimes?
The 2019 inversion occurred after a series of rate increases in 2018. But the subsequent recession in 2020 was arguably the result of COVID-induced lockdowns and other economic disruptions, not a consequence of Fed actions. A pandemic was not on anyone’s radar in 2019. Furthermore, the small sample size (only six inversions over the past 45 years) does not provide enough data to conclude a cause and effect with any statistical significance.
Can you use a yield curve inversion as a market downturn signal?
Even assuming (1) it did portend a recession and (2) a recession is always accompanied by a market downturn, the challenge – as always – is the timing. The lag time between an inversion and a subsequent market decline has varied widely, from a low of seven months in 2020 to a high of 22 months between the December 2005 inversion and the consequent market correction beginning October 2007.
If you had treated each occurrence as a trigger to sell out of the market, you would have missed out on double-digit returns five out of the six times before the subsequent market decline began. If you had sold out after the August 1978 yield curve inversion, you would have given up a gain of over 120% before the market finally dropped.
As a recession indicator, I find the yield curve to be questionable. And as a market timing signal, it doesn’t provide enough information to avoid a market downturn without forgoing the market gain that often precedes it. A more reliable approach is to follow an investment strategy that is resilient enough to weather downturns regardless of the cause. That way you can avoid reacting to external events in an unnecessary and possibly even incorrect manner.
Los Altos resident Artie Green is a Certified Financial Planner and founder of Cognizant Wealth Advisors. For more information, visit cognizantwealth.com.
