The flat truth on the yield curve and recessions

Alarm bells rang for many investors when the U.S. Treasury yield curve recently inverted for the first time in roughly a decade. On March 22, the yield on the 10-year Treasury bond fell slightly below that of the 3-month bill. The classic 2-year-to-10-year curve remains positively sloped for the moment, but only barely.

Yield-curve inversions are rare occurrences in which short-term interest rates exceed longer-term rates. For several decades, these events have served as reliable predictors of a coming U.S. recession. The logic behind this link is that bond yields can be thought of as a proxy for growth expectations.

So if the market is looking for less economic growth down the road (10-year bond) relative to today (3-month bill), that is a forecast for a weakening economy — precisely the sort of environment that can culminate in recession.


Even though forecasts of “less growth” should not imply “no growth,” deteriorating expectations often build upon themselves, creating a vicious circle into recession. 

What has prompted the curve’s inversion?

From an economic standpoint, the flattening of the yield curve is hardly a new development. Fed rate hikes have lifted the short end of the yield curve over the span of several years and an aging cycle has been dimming longer-term growth prospects (reflected in a lower 10-year yield) for some time.  

At RBC Global Asset Management, we flagged 2019 as the likely year for the curve’s inversion some time ago.

The Fed meeting in March arguably provided the final push past the inversion finish line. On the surface, this claim seems illogical, as the Fed furnished a dovish rather than a hawkish decision. The 3-month yield fell rather than rose, on diminished rate hiking expectations.

But this was outweighed by an even larger drop at the long end of the curve, driven by nervousness about the Fed’s dimming growth forecast and mounting suspicion that the business cycle was drawing to a close. 


Why it’s premature to panic

While the inverted yield curve gives investors valid justification for caution, there are several reasons why an extreme response is probably unwise:

The curve is barely inverted 

The yield curve has merely inverted by a handful of basis points thus far. This fact doesn’t invalidate the signal altogether, but it means the signal is at the faint end of the spectrum and could well vanish with only a slight recalibration of the bond market.

The inversion hasn’t lasted long enough

Most econometric models of the yield curve require that the curve be inverted for a full quarter before formally triggering a recession signal. That has not yet happened, and there is a chance that it doesn’t happen at all given the limited extent of the inversion.

On average, a recession occurs about a year after the yield curve inverts. Granted, the historical experience has varied, from a short lead time of just half a year to a long lead time of nearly two years.

But the point, in all cases, is that an inverted yield curve doesn’t predict a recession tomorrow so much as it predicts one in about a year’s time. That leaves a bit of breathing room.

The bond market lacks a term premium

Normally, the yield curve is upward-sloping not just because of expectations for improving growth and rising policy rates but because longer-term bonds naturally command a term premium that sits atop this.

However, for reasons related to the legacy of quantitative easing and distortions arising from liability-driven pension funds, the term premium no longer exists today.

This is key because in the past, an inverted yield curve didn’t just mean that the market was pricing in a slightly worse economic environment in the future.


Instead, it meant the market was pricing in a much worse economic environment, as the term premium kept the longer end elevated until the outlook was truly dire.

Today, without a term premium, one could argue that the yield curve needs to invert more significantly than normal to furnish the same signal.

This is important, though let us equally acknowledge that there is an alternate specification of the yield curve that doesn’t rely on the term premium, and it has also inverted. Suffice it to say that the water is muddier than usual. 

The inverted yield curve is undeniably bad news, but it is not exactly a shock given its slow-motion arrival. RBC Global Asset Management has argued for some time that the probability of a U.S. recession is about 35 percent for 2019 and 40 percent for 2020.

This was once considered pessimistic, but is now interpreted as on-consensus or even optimistic relative to some market views and the output of formal recession models.

This is an imperfect investment environment, arguing for less risk-taking than at earlier points in the cycle. All the same, the yield curve could be lying, in which case risk assets such as equities could enjoy further life, particularly given their superior valuations to bonds.

Alternately the yield curve could be telling the truth, but that would still mean an average of another year of economic growth, with some of that presumably mapping onto market returns. The end is nearing, but it isn’t obviously nigh.

Eric Lascelles is the chief economist at RBC Global Asset Management.