FRBSF Economic Letter
2018-07 | March 5, 2018
The term spread—the difference between long-term and short-term interest rates—is a strikingly accurate predictor of future economic activity. Every U.S. recession in the past 60 years was preceded by a negative term spread, that is, an inverted yield curve. Furthermore, a negative term spread was always followed by an economic slowdown and, except for one time, by a recession. While the current environment is somewhat special—with low interest rates and risk premiums—the power of the term spread to predict economic slowdowns appears intact.
One of the most pervasive relationships in macroeconomics is that between the term spread—the difference between long-term and short-term interest rates—and future economic activity. A negative term spread, that is, an inverted yield curve, reliably predicts low future output growth and indicates a high probability of recession (Rudebusch and Williams 2009). This relationship holds not only in the United States but also for a number of other advanced economies (Estrella and Mishkin 1997). The term spread is one of the most reliable predictors of future economic activity among a wide range of economic and financial indicators and, as such, is closely watched by professional forecasters and policymakers alike.
Over most of the current recovery, particularly in 2017, the yield curve has flattened. As of the end of February, the difference between the ten-year and one-year Treasury yields stands at only 0.8%. The Federal Reserve, which affects short-term interest rates, is continuing its path of monetary policy normalization. In its Summary of Economic Projections from December 2017, the median projection of the federal funds rate rises from its current 1.4% to 3.1% in 2020, even slightly overshooting its long-run projected value of 2.8%. Many observers and forecasters therefore expect the term spread to shrink even further, including the possibility that it could turn negative.
The question then naturally arises whether this development may signal a rising probability that a recession could begin. Some commentators have argued that the relationship between the slope of the yield curve and the business cycle may have changed due to the unique current circumstances, including the unusually low risk premiums holding down interest rates. This Economic Letter revisits and updates some of the empirical evidence for the predictive power of the term spread and addresses the question of whether this time may indeed be different.
The yield curve and the business cycle
The predictive power of the term spread is immediately evident from Figure 1, which shows the term spread calculated as the difference between ten-year and one-year Treasury yields from January 1955 to February 2018, together with shaded areas for officially designated recessions. Every recession over this period was preceded by an inversion of the yield curve, that is, an episode with a negative term spread. A simple rule of thumb that predicts a recession within two years when the term spread is negative has correctly signaled all nine recessions since 1955 and had only one false positive, in the mid-1960s, when an inversion was followed by an economic slowdown but not an official recession. The delay between the term spread turning negative and the beginning of a recession has ranged between 6 and 24 months.
Figure 1
The term spread and recessions
Note: Gray bars indicate NBER recession dates.
While historical circumstances differed for these episodes, the patterns of past yield-curve inversions were remarkably similar: The decline in the term spread was generally driven by a pronounced increase in short-term interest rates. Long-term rates, on the other hand, typically moved much more gradually and either increased slightly over those periods or even declined....MUCH MOREHT: FT Alphaville's Further Reading post, March 7
