The probability of a recession in the United States just went up, materially.
In a piece titled “5 Reasons 2019 Is The Most Important Year For Housing In 10 Years” I argued that the broader economy was on the shortlist of critical factors that would impact the U.S. housing market. While the probability of a recession was approximately 15% in late 2018, today it is closer to 30%.
It’s a worthwhile exercise to explore the question: “What is causing U.S. recession probabilities to increase?”
Imagine, for the sake of simplicity, that your bank says it will pay 2% on your savings account for the next three months. Then imagine that the same bank will only be willing and able to pay 1.5% on that same savings account for the three months beginning a year and a half from now.
The “bank” that is currently offering lower short-term rates in the future is the United States of America. This phenomenon, a negative near-term forward spread, has traditionally been an ominous sign for the U.S. economy.
In the fictional example above, the negative near-term forward spread, the scenario where the bank is willing to pay a lower interest rate in the future, is 0.5%. However, in the actual financial markets, those interest rate percentage numbers are closer to 2.39% on today, and 2.38% for 18 months from now—a mere negative near-term forward spread of 0.01%.
While this may not appear concerning, if this current interest rate phenomenon does not change soon, and aggressively, the U.S. economy will likely enter recession within the next 12 to 15 months, according to analysis by the Federal Reserve.
Nine months ago, Federal Reserve economists Eric Engstrom and Steven Sharpe authored a research report titled “(Don’t Fear) The Yield Curve,” where they made a compelling argument: that near-term forward spreads (the difference between three months rates expected in 18 months minus the three months rates today) have been a better predictor of recessions than the traditionally-watched, long-term yield spreads (10-year yields minus 2-year yields).
U.S. Near-Term Forward Spread & Long-Term Yield Spread 1997 – 2019 OMICELO, LLC
The chart above, which captures the daily short-term and long-term yield spreads from 1997 through March 2019 (analysis adapted from Bloomberg, The Federal Reserve, and Omicelo) shows that (a) negative near term spreads have been a precursor to recessions, and (b) near-term spreads are lower than they have been at any point since the Great Recession in 2009.
The economists’ argument was timely last summer because long-term yield spreads (10-year yields minus two-year yields) had dropped dramatically, but the near-term forward spreads (the difference between three months rates expected in 18 months minus the three months rates today) remained in a comfortable range that had been established since the last recession ended in 2009-10. Last summer’s recession probability, according to their model, was approximately 15%.
Each time near-term forward spreads have gone negative, according to Engstrom and Sharpe’s analysis, the probability of recession rose to between 30 percent and 50 percent. Even absent perfect insight into their models, it’s fair to speculate that the probability of recession has risen to at least 30 percent in the last three months.
Things could be different this time, but over the last approximately 50 years, there were seven times that near-term forward spreads have gone negative, and six of those times have resulted in recessions within about a year.
Let’s hope it is, indeed, different this time.