Practical analysis for investment professionals
21 July 2023

The Predictive Power of the Yield Curve

“[O]ur mind is strongly biased toward causal explanations and does not deal well with ‘mere statistics.'” — Daniel Kahneman, Thinking, Fast and Slow

The predictive power of the yield curve is a widely accepted causal narrative. But the history of the yield curve shows that the causal correlation between long and short rates is actually quite weak. While long and short rates tend to move in the same direction, they do so at varying rates.

Subscribe Button

The debut of the Federal Reserve System in 1914 and the advent of modern central bank orthodoxy amid the Great Inflation of the late 1960s to early 1980s contributed to a divergence in how the market sets long and short rates. The yield curve’s predictive accuracy was decidedly mixed in the first half of the 20th century but was much more reliable in the second half — a shift that aligns with how the the US Federal Reserve has evolved over the years. 

During the 19th century and the first three decades of the 20th, yields for four- to six-month commercial paper were higher on average than those of prime long-term bonds. As the US Civil War gave way to peace and deflation, interest rate levels exhibited a downward trend. But towards the turn of the century, gold discoveries increased the money supply and sent rates higher. During this period, the market set interest rates based primarily on the supply and demand of loanable funds. The low interest rates of the post–Civil War era did not prevent eight different NBER recessions between 1868 and 1900.

But higher rates from 1900 to 1920 didn’t exert much of an influence over the economy either, with six different NBER recessions over the 20 years. A persistently inverted yield curve may have contributed to the high frequency of recession. After all, a negatively sloped interest rate term structure disincentivizes long-term investment.

Only after 1930 did positive yield curves become more regular. The 1929 stock market crash, the resulting shift toward greater economic planning by the state, and the integration of Keynesian economic policies later in the 1930s certainly shifted the slope of the yield curve. As short rates came onto economic policymakers’ radar, they introduced a new causal force that broke the link between short and long rates.

With the markets free to set long-term rates, the views of policymakers and the market on the state of the economy diverged. The Fed’s open market operations are, by their nature, countercyclical and lag the real economy. The market, on the other hand, is a forward-looking voting machine that represents the collective wisdom of the crowd. When the market thinks the Fed is too hawkish, long rates fall below short rates. When it perceives the Fed as too dovish, long rates rise well above their shorter counterparts.

Data Science Certificate Tile

Market prices are the best indication we have of future market outcomes. Why? Because of the potential rewards available. If the future is in anyway knowable, prices in a free market are the most effective crystal ball: Resources will be directed to take advantage of any mispricings. Financiers in earlier eras would not recognize a connection between long-term and short-term rates. They saw short-term lending as primarily concerned with the return of principal and long-term lending on return on principal. But the combination of Keynesian economic policies and the market’s discounting mechanism made the yield curve the predictive tool that it is today.

But it needs to be deployed with caution. It is not just the slope of the curve that matters but how it develops and how long the curve is inverted.

Cumulative Days of Yield Curve Inversion

Chart showing Cumulative Days of Yield Curve Inversion

Source: Federal Reserve Bank of St. Louis, NBER

The yield curve has inverted from positive to negative 76 different times since February 1977 according to the preceding chart — sometimes for months at a time, at other times for just a day — but there have only been six recessions. So, inversion alone is hardly an accurate oracle. Only when the market and the Fed veer apart for an extended time period, when the market expects significantly lower growth than the Fed, does the market’s recession expectations tend to play out. Given the efficiency of the market voting machine, this should hardly come as a surprise.

The yield curve is a popular recession indicator for good reason. But we need more proof of its efficacy, particularly when the signals suggest that Fed policy is too loose.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.

All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Image credit: ©Getty Images/ ardasavasciogullari

Professional Learning for CFA Institute Members

CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can record credits easily using their online PL tracker.

About the Author(s)
Joshua J. Myers, CFA

After a successful 20-plus year investing career, Joshua J. Myers, CFA, launched Cedars Hill Group to bring large market expertise to broader audiences. He primarily serves as an outsourced CIO/CFO for family offices, RIAs, and small-to-medium sized businesses. He started as an assistant trader at Susquehanna Investment Group during the Russian default and LTCM failure in 1998. Afterwards, he was head of fixed income at Penn Mutual Life Insurance during the global financial crisis (GFC). He traded distressed CMBS securities in the aftermath of the GFC at Cantor Fitzgerald and most recently was chair of the board for an oil production company during the COVID pandemic. He is a lifelong student of financial markets and writes about current events with a focus on the art of decision making and cognitive psychology.

Leave a Reply

Your email address will not be published. Required fields are marked *

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.