Economists are known to be poor forecasters of recessions. The slope of the yield curve, on the other hand, has a better record. A common assertion among economic and financial commentators is that an inversion of the yield curve almost always precedes an economic downturn. In fact, without being precise about which part of the yield curve, the yield curve inverted before every recession since 1970. Thus, when on March 22nd the yield on 10-year Treasury bonds fell below that on the 3-month variety for the first time since 2007, a recession in the horizon became nothing but a forgone conclusion to many.
Although history is not always prologue, we take a deeper look at the interaction of yield curve slope, investor expectations, and monetary policy. The shape of the yield curve embeds investor expectations about the future level of short-term interest rates. When investors expect a future economic downturn, they also anticipate that the Federal Reserve (Fed) will ease short-term interest rates in response. Hence, a higher risk of recession drives forward-looking investors to bid up the price (and reduce the yield) of longer maturity bonds, which can invert the yield curve.
To isolate and magnify the most salient aspects of the relationship between the slope of the yield curve and future economic performance, we organize this Strategy Insight as Frequently Asked Questions.
What do we mean by an inverted yield curve?
An inverted yield curve denotes a situation in which long-maturity rates are below short-maturity rates, or in other words, the term spread is negative. The academic literature often measures the term spread as the difference between the yield on the 10-year Treasury note, which reflects long-term views of bond investors, and the yield on 3-month Treasury bills (10Y-3M), a proxy for the federal funds rate targeted by the Federal Open Market Committee (FOMC). Market practitioners and commentators often use the difference between 10- and 2-year yields (10Y-2Y). Both measures exhibit a similar behavior with respect to the business cycle, but some argue that the 10Y-3M spread is the most useful indicator of future recession1. This term spread became negative on March 22, 2019.
Why can the yield curve invert before recessions?
Investors’ expectations and the Fed’s changes in policy stance given views on the economic cycle interact in the process. Recall that the interest rate on a long-term security in part reflects the expected path of future short-term rates. This path is affected by both expectations about the economic cycle and monetary policy. When investors expect an economic downturn, they expect the FOMC to cut the policy rate in the future. Expectations of lower future short-term rates reduce longer-term rates today as investors increase demand for longer-term securities. Also, concerns about a downturn tend to emerge deep into the Fed’s tightening cycle; (i.e., following an increase in short rates). Hence, higher short-term rates, coupled with expectations of rate cuts in the future, can result in an inverted yield curve. To the extent that investors’ forecasts of a downturn are correct, such changes in the slope of the yield curve will correspond to a higher probability of a future recession.
Where does the Fed fit in all this?
The notion of a Fed’s reaction function in its current form strengthened under Alan Greenspan. Since 1987, the U.S. experienced three recessions, each of them preceded by a sustained yield curve flattening that led to inversion. The yield curve inverted for a brief period in September 1998, just before the Fed’s meeting of September 29th when they eased rates. They eased rates again between meetings on October 15th, and one more time at the November 17th meeting. These policy actions reduced market volatility and lifted subsequent economic activity, leading the Fed to resume its tightening cycle in 1999.
The short inversion of 1998 is seldom counted as a false positive; (i.e., inversion not followed by a recession). In hindsight, the distinctive feature about this episode is the fact that inversion did not occur at the peak of the tightening cycle, as the Fed eased rates amidst the possibility of a downturn and high volatility in financial markets (remember Long-Term Capital Management?). In retrospect, the Fed’s reaction may have postponed the recession.
How long after inversion did recessions begin?
Since 1970, it took 303 days on average from the first time the 10Y-3M spread inverted until a recession started. Furthermore, using data on the 10Y-3M spread, the Chicago Fed National Activity Index (CFNAI), and the National Bureau of Economic Research (NBER) recession indicator, we estimated the probability that a recession will begin in 12 months. The exhibit to the right shows a time series of these probabilities and the actual recessions. At today’s values for the 10Y-3M spread and the CFNAI, the likelihood of a recession in 12 months is 36%, about 10% higher than our estimate a month ago.
What about monetary policy and the business cycle?
Consider the realized Federal Funds Rate (FFR) and a policy rule based on unemployment and inflation to determine a suggested FFR. Note in the exhibit to the right that the two differ by the most in the second half of the 1990s and after the Financial Crisis.
In the former period, the suggested FFR stayed an average of 1.5% below the actual FFR, suggesting that perhaps the policy rate was too tight. As mentioned before, the Fed paused and partially reversed the tightening cycle in in late 1998.
In the latter period, the severity of the Great Recession caused the suggested FFR to be negative. One might reconcile the difference with respect to the actual FFR, which is subject to zero lower bound, arguing that multiple rounds of Quantitative Easing amount to a negative effective policy rate. In the post-crisis period, low unemployment caused the suggested FFR to be above the actual rate, a reflection of a cautious Fed which was focused on a gradual approach to policy normalization and concerned about inflation running below target.
What are the fundamentals saying about the economy?
Neither the Conference Board Leading Indicators nor the CFNAI are at recession levels or predicting one, although they have moved lower over the course of this year. But both indicators have been more bearish on two occasions in the post-Crisis era.
How do we see the above developments in a portfolio context?
We do not make explicit macro predictions. Our view is that there is reason for caution about the strength of the economy, but no clear sign that we are headed into a recession. It has become clear that the Fed is also taking a more cautious tone and may move to ease if signs of a deeper slowdown emerge. Our portfolio positioning in credit remains conservative, as a result of less appealing valuations, not based on cycle predictions. Just as in 4Q18, we are well-positioned to take advantage of a cyclical market move should it come to pass.
Andrew P. Hofer
Head of Taxable Fixed Income
Neil Hohmann, PhD
Head of Structured Products
Jorge Aseff, PhD
Head of Quantitative Research
Issuers with credit ratings of AA or better are considered to be of high credit quality, with little risk of issuer failure. Issuers with credit ratings of BBB or better are considered to be of good credit quality, with adequate capacity to meet financial commitments. Issuers with credit ratings below BBB are considered speculative in nature and are vulnerable to the possibility of issuer failure or business interruption.
Opinions, forecasts, and discussions about investment strategies represent the author's views as of the date of this commentary and are subject to change without notice. References to specific securities, asset classes, and financial markets are for illustrative purposes only and are not intended to be, and should not be interpreted as recommendations.
Investing in the bond market is subject to certain risks including market, interest-rate, issuer, credit, and inflation risk; investments may be worth more or less than the original cost when redeemed. Income from municipal bonds may be subject to state and local taxes and at times the alternative minimum tax.
NOT FDIC INSURED NO BANK GUARANTEE MAY LOSE VALUE
IM-06219-2019-04-08 Exp. Date 07/31/2019
1 Bauer and Mertens, 2018, Economic Forecasts with the Yield Curve. Federal Reserve Bank of San Francisco (FRBSF) Economic Letter.
2 NBER is National Bureau of Economic Research