In light of the burgeoning Treasury selloff, you can probably expect the old stock-bond return correlation debate to start heating up again.

As a reminder, you don’t want that correlation to flip positive when bonds are selling off. That creates the dreaded “diversification desperation” and plays havoc with balanced portfolios and risk parity.

Last week, risk assets were buoyant in the face of rising yields, in part because the decomposition of rate rise was somewhat benign and perhaps even bullish to the extent you’re inclined to take the widening in breakevens as a sign that the reflation narrative is getting some traction again.

The problem, though, is that while reals were steady last week, they’ve generally been in the driver’s seat since the lows in August. The purple highlight in the following chart shows nominals breaking higher last week even as reals were steady, but the recent trend is clear.

(Bloomberg) 

Of course this is anything but cut and dry. Higher inflation expectations or, “worse”, evidence of inflation pressure from a series of hot AHE prints or maybe several months of above-consensus CPI numbers, will stoke concerns about “excessive” Fed tightening and that, in turn, could push up real yields. There’s more on that in “Credit Suisse’s Kasper Bartholdy Takes A Wrecking Ball To ‘Global Liquidity Shortage’ Narrative.”

For the time being, it’s worth remembering that if it’s the stock-bond return correlation you’re concerned about, it’s the speed of rate rise that matters, not so much the level of yields. This is conversation that took center stage back in late January/early February and Goldman revisited it briefly on Monday evening.

“Investors often express concern about the level of yields, however, we find over the shorter term the speed of the rate move is a crucial input for equity-bond correlations”, the bank writes, adding that “historically, we find that when the UST 10y yield has a 2-standard deviation move over a 3-month period the equity-bond correlation usually turns positive.”