The Marriner S. Eccles Federal Reserve building stands in Washington, D.C., U.S. (Photographer: Andrew Harrer/Bloomberg)

Fed Paper Questions ‘This Time Is Different’ Yield Curve Theory

(Bloomberg) -- The Federal Reserve Bank of San Francisco has bad news for those declaring an inverted yield curve is no longer a recession predictor.

Adjusting for the compensation investors demand to hold longer-dated bonds doesn’t invalidate the curve’s prognosis powers, according to a new research post published Monday.

Policy makers and market participants alike are watching as the gap between short- and longer-term rates narrows. Historically, recession has followed such inversions, though it’s unclear why that relationship exists.

Some, including former Fed Chair Janet Yellen, have suggested that depressed term premium -- the bonus investors require to take on the added risk of holding longer-dated bonds -- could mean the curve’s flattening doesn’t herald a downturn this time around. The logic is that if long-term rates are lower thanks to government bond-buying programs and other structural factors, the curve could invert with just mildly restrictive monetary policy. In the past, it took very restrictive Fed policy to push short rates above the long end of the curve. Against today’s changed backdrop, an inversion could be a less potent predictor of recession.

But San Francisco Fed research advisers Michael Bauer and Thomas Mertens push back on that theory. By subtracting out an estimate of the term premium, they obtain an “expectations only” spread between short- and longer-dated securities. By separating the two drivers, they find that inversion signals high recession risk whether it stems from a low term premium or low short-rate expectations holding down long-term rates.

“We do not find an empirical basis for adjustments based on the term premium,” they said.

Fed Paper Questions ‘This Time Is Different’ Yield Curve Theory

The gap between 2-year and 10-year yields on Monday narrowed to as little as 18.3 basis points, the smallest differential since 2007, having started the year more than 30 basis points higher. The term premium, a measure of the compensation that investors demand for holding longer-dated debt, remains mired more than 50 basis points below zero.

It’s “plausible” that long-term rates are lower because of the Fed’s crisis-era bond-buying program, they allow. But there are two big ifs. First, it isn’t clear how much of an effect bond buying has had on rates. Second, low term premiums may have contributed to overheating in the past, so even if quantitative easing is the cause, the result could still be heightened recession risk.

“There is no clear evidence in the data that ‘this time is different’ or that forecasters should ignore part of the current yield curve flattening because of the presumed macro-financial effects of QE,” the researchers wrote.

It isn’t all bad news. Today’s flattening yield curve “provides no sign of an impending recession,” they write. Bauer and Mertens are also careful to note that correlation is not causation: Because it’s not clear what ties inversions and recessions, “great caution is therefore warranted in interpreting the predictive evidence.”

©2018 Bloomberg L.P.