Some economic yardsticks, like inflation and unemployment, are evergreen. Others, like “gross national product,” the predecessor to GDP, have expiration dates. Yield curve “inversions” belong to the latter group. The differences between yields of U.S. Treasuries of various maturities have been regarded as foolproof predictors of recessions for decades. This time, they have failed spectacularly, and closer study reveals several weaknesses in past performance as well. It’s time investors and economists look past the U.S bonds’ faulty forecasts.
The yield curve plots the yields on government bonds of different maturities. Typically, the curve slopes upward as maturities increase since investors require a higher yield as compensation for the risk of holding a bond for longer. In rare circumstances, though, yields on long-dated bonds fall below those of short-dated ones and the curve slopes downward or, in traders’ parlance, “inverts.”
That’s been the case in the United States for more than a year. The yield on two-year Treasuries climbed above that for 10-year government bonds in July 2022, causing a closely watched “two-ten inversion.” The reversal is the longest-lasting since 1980, and the spread between the two yields widened the most since 1981 in July. Investors interpret these signs as bad omens because they reflect fears that near-term economic risks exceed the dangers of holding longer-dated debt.
At first sight, they are right: Yield curve inversions have been a consistent predictor of future downturns. Every modern U.S. recession has been preceded by an inversion. The yield curve has only shown one false positive, in 1960, and only because the following slowdown wasn’t dramatic enough to constitute a recession, according to the San Francisco Federal Reserve.
Yet on closer inspection, the curve’s predictive powers aren’t as good as they seem. Inversions have preceded downturns by an average of 19 months, leaving plenty of time for unrelated events to spark a recession. That’s exactly what happened in 2019; the yield curve inverted that September, well before Covid-19 and the resulting economic shock hit the U.S. in March 2020. There’s also no evidence that yield curve inversions predict stocks will underperform Treasuries, according to a 2019 working paper by academics Eugene F. Fama and Kenneth R. French.
BAD TO WORSE
The yield curve’s limitations as a crystal ball have been laid bare by the latest economy cycle. Aggressive interest rate hikes by the Fed pushed short-term yields higher while fears of a recession down the line sent yields on longer-maturity debt lower. Yet the past 12 months turned received wisdom on its head, as inflation cooled to 3.7% and joblessness stayed historically low. Higher interest rates, it seems, are doing little harm to economic growth, and a growing number of economists are ditching their recession forecasts.
Inflation expectations may have blunted the yield curve’s predictive strength. While Americans’ expectations for one-year inflation generally moved with the official growth rate in consumer prices, their three-year expectations were far more muted, according to the New York Fed. By August 2022, households’ median three-year expected inflation rate was back to pre-crisis lows, suggesting most had concluded the price surge would be temporary. Such thinking would therefore push short-dated yields higher while leaving longer-maturity bonds relatively unaffected.
The latest reversal also has just as much to do with the past as it does with the future. Interest rates sat near zero for much of the 2010s. As such, the rate-hike campaign that the Fed began in March 2022 lifted short-term yields. Strong demand for the safety of longer-dated Treasuries, meanwhile, pulled late-maturity yields lower. Those two dynamics have made inversions a less reliable recession indicator over the last two business cycles, Morgan Stanley U.S. Chief Economist Ellen Zentner said last year.
BENDING THE CURVE
The Fed’s rate hikes aren’t the only monetary policy tool influencing the yield curve. The central bank bought nearly $2.4 trillion of Treasury bonds from 2008 to 2014 in a bid to lower long-term interest rates and aid the economy in the wake of the financial crisis. The onset of Covid-19 prompted the Fed to buy another $3 trillion of Treasuries. Much of those purchases were focused on longer-maturity debt, increasing prices for those notes and weighing down yields.
The central bank has since started to let its Treasury holdings run off its balance sheet. But its bond buying, also known as quantitative easing, had led the bond market to enter 2022 with a much flatter yield curve, and therefore a lower bar for which to reach a two-ten inversion.
The European Central Bank made a similar argument, with regards to the euro zone’s bond buying, in the minutes of its latest policy meeting, noting that the effect of its bond purchases “could reduce the predictive content of the slope of the yield curve for economic growth”.
U.S. fiscal policy has also skewed the curve in recent years. The Treasury issued nearly $21 trillion in notes and bonds in 2020, up 87% from the year prior, to help pay for massive pandemic relief. Issuance totaled another $20 trillion in 2021 and dipped to $17 trillion last year. Most of that debt was sold as notes with maturities ranging from two years to 10 years. As such, the influx of new supply and steady investor demand likely placed more downward pressure on longer-dated yields, especially relative to bonds with maturities of less than two years.
Quantitative easing, near-zero rates and massive fiscal stimulus helped the world’s biggest economies bounce back from Covid-19. Policymakers could deploy similar tools when the U.S. faces its next downturn. That may spark another deep, and likely harmless, yield curve inversion. Gaps between Treasury yields still give important insights into where bond traders are bracing for the most risk. But after substantial government intervention and the economy’s surprising resilience to higher rates, the yield curve’s crystal ball is all but shattered.
Source: Reuters (Editing by Francesco Guerrera, Sharon Lam, Aditya Sriwatsav and Oliver Taslic)