Professor Campbell Harvey’s Yield Curve Research: How Inversions Predict Recessions
By Tim Hayes, Financial Advisor for the Public and Not-for-Profit Sector
Summary: Discover Professor Campbell Harvey’s research on how the inverted yield curve predicts future recessions, specifically focusing on the three-month treasury bill and the ten-year treasury bond. Explore the history of his research and why this indicator works.
Introduction
Campbell Harvey, a professor of finance at Duke University, is credited with finding that a recession follows the inversion of the yield curve. It has correctly predicted the last eight recessions.
An inverted yield curve means that shorter-term government bonds pay higher interest than longer-term government bonds. This is not how the market is supposed to work, so when it happens, it has been an excellent predictor of a future recession. Professor Harvey’s research specifically points to an inversion where the three-month treasury bill is higher than the ten-year treasury bond for a minimum of three months.
The spread between the three-month T-bill and the ten-year treasury bond inverted in October 2022. Recessions typically begin six to 18 months after the inversion begins and usually last for the length of the inversion.
Today, the three-month rate is 5.388%, while the ten-year pays 4.056%, an inversion of 1.24%. Imagine visiting your local bank and seeing that a three-month CD pays a higher interest rate than a ten-year CD.
The History of His Research
Professor Harvey began his quest when he was an intern at Falconbridge, a natural resources company based in Toronto, Ontario. In 1982, he was tasked with finding an indicator for forecasting GDP. With a limited budget and only nine weeks to complete the task, he knew he could not compete against the giant companies doing this type of forecasting, so he looked at the markets to try to find the answer.
The stock market was too volatile, but he found signals in the bond market: mainly that the yield curve inverted before recessions. Before he could present his finding, however, his whole group got laid off. In 1986, Campbell began his Ph.D. program at the University of Chicago using the yield curve as his dissertation’s focus.
Why Does It Work?
Professor Harvey believes that an inversion signals that investors are predicting an economic slowdown by buying the ten-year treasury bond, pushing that rate lower. At the same time, the Federal Reserve hopes to slow down the economy or lower inflation by hiking short-term rates, resulting in shorter-term rates going higher than longer-term ones.
Inversion stresses the banking system because banks are forced to pay higher interest rates on deposits, thus earning less on loans that originated when rates were lower or assets such as bonds bought by the banks when rates were lower. Under stress, banks become less likely to make new loans or buy new assets. Both activities curtail money creation, which is the blood supply for the economy—and reduction in money creation can have dire consequences.
One thing Professor Harvey needs to correct is his understanding of banking. He still believes in Jimmy Stewart’s “It’s a Wonderful Life” banking, where banks take depositors’ money and lend it to borrowers. That is not how modern banking works. (See Money Creation in the Modern Economy from the Bank of England (PDF)
Banks today make loans that create deposits without using depositors’ money to make those loans. They do need more reserves to back up the new deposits, and the cost of those is based on the rate of short-term bonds. So, the cost of those funds, together with the higher rate paid on deposits, could cause stress for banks.
The Professor Questions His Model
In a LinkedIn post on January 4, 2023, Professor Harvey said that he thought this time might be different because the job market was so strong that anyone getting laid off would soon find a new job. Especially since most of the layoffs were in the technology industry, these highly skilled people would have no problem finding another job.
He supplied one warning: His prediction that this time could be different was true if and only if the Federal Reserve stopped its rate hiking, which was further inverting the yield curve. As we know, that did not happen, and the professor now thinks the chances are excellent that his indicator goes nine for nine.
The Impacts of a Recession
Recessions mean people get laid off. People could lose their homes and put off having families. Politicians might get voted out of office. The stock market usually falls significantly.
Bonds, however, could pay dividends if the yield curve turns out correct and a recession is forthcoming, as the return on bonds resulting from interest rates falling due to the downturn lessens the blow of falling stock prices
Suppose the yield curve turns out to be wrong, and we avoid a recession. In that case, your percentage in stocks could continue to rise, and if inflation does not start growing again, bonds should do okay, just not as well as if there were a recession.
Conclusion
Professor Campbell Harvey developed the idea that a recession will likely follow when the rate of a three-month treasury bill exceeds that of a ten-year treasury bond. Professor Harvey’s research has been correct eight out of eight times. If it is right this time, the stock market could fall and people could lose their jobs and homes and have difficulty starting families.
I am a fiduciary financial advisor with over 30 years of experience helping public employees and educators with retirement planning, including 403(b) plans. I offer transparent, fee-based advice tailored to each client’s goals and needs.
“My goal is to ensure your retirement plan is built on your best interests, not a product sale.”
These are the opinions of Financial Advisor Tim Hayes and not necessarily those of Cambridge Investment Research. They are for informational purposes only and should not be construed or acted upon as individualized investment advice. Content provided via links to third-party sites should not be considered an endorsement of content that we cannot verify completeness or accuracy of.