San Francisco Fed on the Predictive Power of Yield Curve Spreads

A FRBSF maco economic letter discusses the yield curve and recessions.

Although the U.S. economy is in a sustained expansion and many economic indicators show continued strength, concerns about a possible recession have emerged. One reason is the narrowing spread between long-term and short-term Treasury yields. A well-established regularity in postwar U.S. economic history is that an inverted yield curve—when long-term rates drop below short-term rates—is generally followed shortly afterward by an economic recession.

Financial commentators typically focus on the difference between the ten-year and two-year Treasury yield (10y–2y), because the former summarizes long-term perceptions and sentiment of bond market investors, while the latter is viewed as a reasonable indicator of the stance of monetary policy. But many other combinations of long-term and short-term yields are possible. The tradition in the academic literature has been to focus on the spread between the ten-year and three-month Treasury yields (10y–3m), going back to Estrella and Mishkin (1998), who documented the strong predictive power of this term spread for recessions and economic activity. Recently, Engstrom and Sharpe (2018) have argued that a spread of short-term Treasury rates—the difference between the six-quarters-ahead forward rate and the three-month yield (forward6q–3m)—might be preferable as a predictor because it focuses on expectations of the near-term path of monetary policy.

Term Spreads vs Recessions

The traditional 10y–3m spread is the most reliable predictor, and we do not find any evidence that would support discarding this long-standing benchmark as a measure of the shape of the yield curve. It is worth emphasizing again, however, that all of these term spreads are fairly accurate predictors and quite informative about future recession risk; the differences in forecasting accuracy are small.

On theoretical grounds, one may argue that, when gauging whether the bond market signals a future downturn, one should not compare the ten-year yield itself, but instead its expectations component, the average expected path of short-term interest rates over the next ten years, to the current short-term rate.

But what about quantitative easing (QE)? One might argue that long-term yields, particularly the term premium component, are significantly depressed due to QE programs by central banks around the world and the large balance sheet of the Federal Reserve. Much empirical research suggests that QE likely had quantitatively large negative effects on long-term rates and that some of these effects are still present and continue to push down the long end of the yield curve (Bonis, Ihrig, and Wei 2017). If long-term yields are still low because of QE, and if these effects contribute to the yield curve flattening but do not increase recession risk, then some part of that flattening may not be worrisome at all.

While this reasoning is plausible, these are, however, two big ifs. First, there is a lot of uncertainty around the effects of QE on interest rates, as evident from statistical measures of uncertainty, such as the wide confidence bands reported by Bonis, Ihrig, and Wei (2017) and others. In reality, the uncertainty is almost surely even larger than that, because these estimates assume a specific underlying model that is itself highly uncertain. Second, while a lower term premium contributes to easier financial conditions and therefore stimulates economic activity, this was also the case in the past and at times may have contributed to economic overheating, heightened financial stability risk, and ultimately higher recession risk.

In light of the evidence on its predictive power for recessions, the recent evolution of the yield curve suggests that recession risk might be rising. Still, the flattening yield curve provides no sign of an impending recession.

Cause and Effect

Furthermore, when interpreting the yield curve evidence, one should keep in mind the adage “correlation is not causation.” Specifically, the predictive relationship of the term spread does not tell us much about the fundamental causes of recessions or even the direction of causation. On the one hand, yield curve inversions could cause future recessions because short-term rates are elevated and tight monetary policy is slowing down the economy. On the other hand, investors’ expectations of a future economic downturn could cause strong demand for safe, long-term Treasury bonds, pushing down long-term rates and thus causing an inversion of the yield curve. Historically, the causation may well have gone both ways. Great caution is therefore warranted in interpreting the predictive evidence.

The above paragraph is silly.

Inverted yield curves do not cause recessions. Recessions are generally caused by the monetary excesses that preceded the recession. Recessions can also be caused by external shocks, wars, sanctions, etc.

The yield curve in an indication of something, not the cause of anything.

Conclusion

The yield curve has been a reliable predictor of recessions, and the best summary measure is the spread between the ten-year and three-month yields. Although this particular spread has narrowed recently like most other measures, it is still a comfortable distance from a yield curve inversion. In this Letter, we do not find an empirical basis for adjustments based on the term premium, especially in light of uncertainties about the possible effects of quantitative easing. Finally, when interpreting the yield curve evidence, it is important to remember that the predictive relationship in the data leaves open important questions about cause and effect.

Once again, the cause-effect is not open. Inversion does not cause anything.

That aside, I find the letter quite balanced. Yet, it is lacking. Recessions can occur without the yield curve inverting at all.

For proof, look at Japan.

One should not take it as fact that inversion is a necessary prelude to recession. The study failed to mention that point.

Mike "Mish" Shedlock

Comments