More on the 5-Year Minus 2-Year Spread's Inversion

Tuesday, we shared our perspective on the inversion of the 5-year minus 2-year US Treasury yield curve—an article that seems to have struck a chord with readers, generating a good deal of reaction and follow-up questions. Most seemingly agree: The selection of two points in the yield curve’s middle doesn’t amount to inversion of a meaningful spread. However, several readers pose the question: Isn’t this an early warning sign of inversion to come in more-telling spreads, like the 10-year minus 3-month or 10-year minus effective fed-funds rate? In our view, it may be. But that isn’t helpful for investors. You don’t need an early warning sign of an inverted 10-year minus 3-month curve. It is an—imperfect—early warning sign.

Without retreading too much ground we covered Tuesday, yield curve inversion—short-term rates topping long-term rates—does matter. A lot. Banks borrow short term to fund long-term loans, so yield curve spreads heavily influence loan profitability. The reason the 5-year minus 2-year spread is suboptimal, in our view, is the maturities don’t approximate banks’ funding costs. Nor is five years exactly what we would call a long-term loan. More meaningful curves like the 10-year minus effective fed-funds rate or the 10-year minus 3-month aren’t inverted presently.

But is the 5-year minus 2-year spread hinting inversion of the 10-year minus 3-month looms? Perhaps. Historically, inversion of the 5-year minus 2-year spread has preceded most inversions of the 10-year minus effective fed-funds rate or the 10-year minus 3-month. But the thing is, this isn’t useful information for an investor sizing up the equity market.

Even the more meaningful spreads frequently invert in latter-stage bull markets—often well before bull market peaks. We covered this in an August post, sharing the table below to illustrate this point: Yield-curve inversion is not a timing tool. Full stop.

Exhibit 1: It Is Often a Long Way to the Top After Initial Yield Curve Inversion

Source: Federal Reserve Bank of St. Louis and Global Financial Data, Inc., as of 7/21/2018. *Inversion occurred after the bull market’s peak.

The same holds for 10-year minus 3-month inversions. Hence, inversion of even meaningful yield curves isn’t an automatic sell signal. Much less yield curves that (imperfectly) hint at inversion of other yield curves! Treating the 5-year minus 2-year spread—or the 5-minus-3, 10-minus-5 or 10-minus-2, for that matter—as a sell signal runs the risk of being way too early.