Bonus Episode: The Yield Curve Whisperer Weighs In

Episode Summary

Executive Summary: This episode features economist Campbell Harvey explaining how he developed the yield curve recession predictor, why the 10-year minus 3-month Treasury spread remains powerful, and why he believes it is signaling slower growth but not necessarily a hard recession this time. The discussion contrasts theory, historical performance, and current labor, housing, and policy conditions that may blunt recession risk.

Main Topics: Origins of the yield curve recession model (Priority: 5/5): Harvey recounts how a student internship task to forecast GDP led him to bond markets and ultimately to his PhD thesis on the yield curve as a recession predictor. Why the 10-year minus 3-month spread matters (Priority: 5/5): The episode explains Harvey’s preferred yield curve measure, why he chose liquid Treasury maturities, and why he prefers the original model despite later variants. Historical track record and methodology (Priority: 5/5): Harvey describes the model’s in-sample and out-of-sample success, the use of quarterly averaging, and the absence of false positives in his preferred specification. Intuition behind the signal (Priority: 4/5): The guests explore why inversions predict recessions: flight to quality, expectations of slower growth and inflation, and the effect on credit conditions and bank lending. Why the model may be less reliable outside the U.S. (Priority: 3/5): Harvey argues that market manipulation and policy distortions weaken the signal in other countries, while Canada shows some predictive power relative to the U.S. Why this cycle may be different (Priority: 5/5): Harvey argues the current inversion may be a false signal because labor markets remain unusually tight, the recession-risk transmission mechanism is muted, and firms are reacting early to the widely publicized warning. Fed policy, QE/QT, and technical distortions (Priority: 4/5): The conversation addresses how central bank balance-sheet operations and forward guidance may add noise to the curve and potentially affect its predictive power.

Key Arguments: Harvey’s yield curve model was born from a practical forecasting problem and relies on a simple economic intuition rather than a complex econometric system. The 10-year minus 3-month Treasury spread is his preferred measure because it uses liquid instruments, avoids inflation noise better than level rates, and historically produced no false signals in his work. The model’s historical performance is unusually strong: it predicted recessions in sample and out of sample, including the 1988 non-recession call after the 1987 crash. Yield-curve inversions are consistent with flight-to-quality behavior and expectations of weaker growth and lower inflation. A flatter curve can also tighten bank lending, reduce credit creation, and reinforce downturn dynamics through the credit cycle. The model is weaker at forecasting recession depth than recession occurrence or duration. Harvey believes current labor-market excess demand, healthier household balance sheets, and stronger bank/housing fundamentals make a hard landing less likely. He also thinks the curve’s public prominence may change behavior, encouraging firms to de-risk early and potentially softening the downturn. Fed policy is still a major wild card; Harvey argues the Fed may over-tighten and cause recession despite improving inflation trends. Technical factors such as QE/QT, global capital flows, and clearer forward guidance can add noise but are not, in his view, the main explanation for a potential false signal.

Data Points: Number of U.S. recessions since World War II referenced: 12 - Used to illustrate the historical recurrence of yield-curve inversions before recessions. Original yield curve forecast record: 8 out of 8 - Harvey’s preferred 10-year minus 3-month model predicted four recessions in sample and four out of sample, with no false positives. Initial sample result: 4 out of 4 - Harvey’s dissertation showed the yield curve predicted all four recessions in his historical sample. Out-of-sample result: 4 out of 4 - After publication, the model correctly anticipated four more recessions. Lead time: 6 to 18 months - Typical lag between inversion and recession onset. Duration signal: Highly correlated - Harvey says the length of the inversion is strongly linked to recession duration. 3-month averaging window: One quarter - Harvey uses a quarterly average inversion, not a single day or week, as the signal threshold. Current inversion timing discussed: October of last year / end of December code red - The guests discuss the then-current inversion and Harvey’s threshold for a recession warning. Job openings to unemployed: Very high / record level - Cited as evidence of unusual labor-market tightness reducing the odds of a sharp unemployment spike. Expected 1988 GDP growth forecast: 4.2% - Harvey’s contrarian call after the 1987 stock market crash, which proved correct. Fed balance sheet size: About $9 trillion - Used to underscore the scale of QE/QT and potential impact on Treasury yields. Pre-financial-crisis Fed balance sheet: $400–500 billion - Comparison point for how much larger central bank market presence has become. Shelter weight in inflation measures: 40% of PCE deflator; 33% of CPI - Used to explain why housing/rent changes affect inflation and policy with a lag. Forecast publication cost: 25 cents - Harvey contrasts his simple model with expensive econometric services. Canadian growth relation: Difference between Canadian and U.S. yield curves was powerful - Used to show the yield curve can predict relative growth differences in a closely linked economy.

Pivotal Quotes: "It is naive to think that this model will never produce a false signal." — Campbell Harvey: Explaining why the current inversion may not guarantee an imminent recession despite the model’s historical success. "There is a buffer. And that means we're not going to see a spike." — Campbell Harvey: Discussing unusually strong labor-market excess demand and why unemployment may rise without a severe recession. "The big wild card, in my opinion, is what the Fed is going to do." — Campbell Harvey: Warning that policy overshoot could still push the economy into a harder landing.

Implications: Listeners should treat the yield curve as a strong but not standalone signal: it still warns of slower growth, yet labor, housing, and balance-sheet conditions may reduce recession severity. Fed policy remains the key risk.