While much has been said about the inversion of the 3M-10Y yield curve (even as the 2s-10s remain "normal") with one bank after another presenting its pitch why the historical track record is no longer applicable and why "this time may be different" for this fail-safe recession indicator which has correctly predicted 6 of the past 6 recessions, less has been said about the actual mechanics of what curve inversion actually means.
As BMO chief economist Tom Porcelli writes in a note from this week, the reasons this metric “works” historically is because the long-end of the curve is supposed to reflect some longer-run speed limit for nominal GDP growth. Traditionally when the economy overheats, as the Fed tightens short-run policy too much to avert an overheating economy, long-run growth prospects will slip below current growth trends and a reflection of this in the rates-world would be an inversion.
However, the problem with the current inversion and the historical record is that according to BMO, "the yield curve at present is not a referendum on the path of economic growth in the United States, but rather a function of goings on globally." As evidence, Porcelli says to look no further than the strong correlation between US and German 10-year yields (or any other DM benchmark paper) in recent months.
The other issue, of course, is the prevailing argument that the Fed has overtightened which as Porcelli correctly observes is "almost laughable", especially when considering that just in October, Powell said that the Fed Funds rate is a "long way" from neutral. Even using the Fed's (rapidly dropping) estimate of the neutral rate, which as we noted recently is largely a function of the excessive debt in the system...
... policy right now is nowhere near past cyclical endpoints. And, when looking at the chart below, the BMO economist tells readers to note how the inversions that did lead to downturns lined up with a Fed Funds well north of neutral (unless of course netural is far lower than BMO's generous estimate).
There is another "critical reason" why BMO believes the signal from the yield curve inversion this cycle is irrelevant.
As touched on above, historically the inverted curve works as a good recession signal "because the 10yr part of the curve theoretically reflects nominal growth prospects going forward." However, this cycle, yields have decoupled from growth in a very material way. In fact, as shown in the next chart, nominal GDP growth in the US right now is running at about 5% and 10s are sitting at just 2.5%. This is why Porcelli repeats that in his view, yields have become more a function of global growth dynamics and have become anchored to low/negative sovereign yields abroad.
The immediate implication is that the United States is able to finance relatively good rates of growth at artificially suppressed interest rates, a beneficial development for an economy that runs on debt. But while the US may not be facing an imminent recession, at least according to the yield curve - which as a reminder has to begin steepening after the inversion for the real alarm bells to go off in the stock market...
... what the yield curve is telegraphing may be just as bad if not worse: according to BMO, contrary to conventional wisdom, this type of dynamic where GDP is running well above the 10Y yield has historically been very positive for asset inflation, "it was a big reason why the housing bubble was allowed to form – recall the conundrum," and inflation in general.
So, to summarize BMO's position, the bank is not on recession watch because of the yield curve inversion. Quite the opposite: "We are, more than any other point this cycle, on bubble watch."