Thomas and I wrote a second Economic Letter about the information in the yield curve for predicting recessions. Here we focus on what different measure of the shape of the yield curve—that is, which yield spread—appears to have the most information. We conclude that the classic 10-year minus 3-month spread is the most useful one. We also discuss the role of the term premium, and how to interpret this evidence. (Correlation is not causation!
Thomas Mertens and I wrote an Economic Letter on predictings recessions with the yield curve. There has been a lot of public discussion about whether a flat yield curve contains a strong signal about a future economic slowdown. We found that the predictive power of the yield curve for future economic activity and recessions is very strong, and remains strong even in the current environment with a low overall level of interest rates, a low term premium, and other somewhat unique circumstances.
Why have long-term interest rates been falling throughout much of 2017, while the Federal Reserve has been normalizing monetary policy? At first sight, the combination of rising short rates and falling long rates seems puzzling, and even vaguely reminiscent of the famous Greenspan conundrum. But this time around, there are some good reasons that explain the flattening of the yield curve, which I discuss in my most recent Economic Letter.