Deutsche’s Kocic: ‘We’re Not Very Far From The Red Zone’ On Equity-Bond Correlations

Rising yields look to be taking a toll on equities, just a day after stocks rose in tandem with the Treasury selloff.

The narrative here is all at once clear and muddy. On the muddy side, there’s considerable debate about how far long end yields can rise given assumptions about i) structural disinflationary factors, ii) a “broken” Phillips curve and iii) the continual suppression of the term premium.

What’s clear, though, is that equities can only stomach sharply higher yields if rate rise is i) seen as an expression of and in step with economic outperformance and ii) doesn’t happen too quickly.

On Wednesday, it was clear that the bond selloff was initially catalyzed by blockbuster U.S. econ data, but a day later, it seems like market participants are a bit anxious about the rapidity of rate rise and also about the prospect of Friday’s jobs report exacerbating inflation fears at a time when the Fed is seen as increasingly data-dependent.

Ultimately, the question is this: At what point does the stock-bond return correlation flip positive (equity-rates correlation flip negative), leading to diversification desperation?

Early Thursday, we brought you some lengthy commentary on this from Goldman and we also highlighted some analysis from the latest missive by Nomura’s Charlie McElligott.

In the interest of fleshing things out further on a day when “Will stocks buckle under the weight of rising yields?” is the only question that seemingly matters, we thought we’d also feature a couple of quick excerpts from a client note by Deutsche Bank’s incomparable Aleksandar Kocic, who begins by reminding you that for quite a while, stocks and bonds have been a kind of “have your cake and eat it too” deal:

Equity-rates correlations tend to be generally positive. Either we have a recession where bonds do well and stocks underperform or a recovery with rally in stocks and sell off in bonds. Regardless of which mode the market is in, we know what to do — we exploit the correlations.

Well, in the post 2008 recession, this was not enough. Both stocks and bonds did well. And nobody complained, because everyone was happy. While it lasted. The problem is clearly the reversal of that mode that takes us from rally in both stocks and bonds to sell off in both assets. This is the most feared scenario. And, to make things interesting, we are not very far from this regime.

Again, that is the question. When does the correlation flip, dragging us all kicking and screaming into a scenario where 60/40 and risk parity are imperiled by a simultaneous selloff in equities and bonds? Here is Kocic’s answer:

In the chart below we illustrate correlations between rates and SPX returns as a function of the terminal level of the bond risk premium (BRP) at the end of each observation window. Between approximately 250 bp and 300 bp over r*, these correlations change sign such that a higher BRP accompanies lower equities. Noise around estimates of r* creates additional uncertainty around the precise equivalent level of yields at which this happens, but starting with the midpoint of the HLW and LW model r* estimates (currently around 70 bp), the implication is that the “red zone” for the change in correlation is a range from roughly 3.20% to 3.70% 10y Treasury yields.