Recently, Mario Draghi was forced to realize a bit of “fallen angel risk”.
Of course “forced” is a bit of a misnomer. The ECB doesn’t “have” to do anything with regard to CSPP because after all, it’s their program.
But the Steinhoff debacle did underscore the fact that sooner or later, Draghi will have to make some tough decisions about how to handle soured (or souring) debt bought as part CSPP. There’s just no way around it because you know, shit happens and no matter how much due diligence you conduct, some things (like say, accounting irregularities) aren’t predictable.
The ECB undoubtedly realizes this which is presumably why they tweaked the language on the official CSPP webpage. Now it says this:
But it used to say only this:
The Eurosystem is not required to sell its holdings in the event of a downgrade below the credit quality rating requirement for eligibility.
The addition of the word “choose” probably should have been a warning sign to everyone and as BofAML recently wrote, this creates a situation where fallen angels could effectively become “falling knives” if the ECB becomes an active seller on a downgrade.
It’s not clear whether it’s a good idea for the ECB to pull support for imperiled credits when they’re already tapering APP. After all, one of the reasons why spreads are so tight has to do with the fact that Draghi has driven everyone down the quality ladder and that scramble ends up leaving everything priced to perfection with the sole exception of the shittiest names where the catalyst for a disaster is readily apparent.
That state of affairs suggests that as the ECB tapers, spreads on challenged sectors and credits will widen out commensurate with the extent to which they’re challenged. Or, put differently, the market will again start to serve as an semi-efficient mechanism for price discovery.
Exiting unprecedented accommodation is just as much an experiment as implementing it in the first place. That is, there’s no telling what’s going to happen and so, you’d be forgiven for questioning the wisdom of the ECB if they do indeed repeat the Steinhoff experience by becoming an active seller into a market that’s trying to figure out how to cope with a disappearing policymaker bid.
Well, Barclays is out underscoring all of this in a new note. Consider this:
We have argued previously that the ECB should refrain from selling bonds except as an active decision to tighten monetary policy, as we believe the ECB should be a policy setter not a policy taker. However, it was probably not difficult for the ECB to replace the Steinhoff bonds it sold, given that it only held €100mn according to the FT. This is a small fraction of current weekly CSPP purchase volumes and even smaller relative to overall asset purchase volumes.
On the other hand, credit deterioration is pro-cyclical; hence, if there is an economic downturn in 2019 or 2020, the ECB could find itself trying to sell bonds into a falling market as many credits suffer simultaneous credit deterioration.
Not to put too fine a point on it, but depending on the size of what they were selling, that could be trouble.
As far as that goes, there’s good news and bad news according to Barclays. First, the bank details what likely went into the ECB’s decision re: Steinhoff, but the bottom line is that the trigger for the sale appears to be tied to the likelihood of restructuring, which Barclays equates to a rating of Caa1 or lower. Here’s the bad news first:
Based on Moody’s historical transition matrix for corporate ratings and our estimate of the rating distribution across the CSPP’s bond holdings, we estimate that c.€250mn of bonds held by the CSPP will migrate to Caa1 or lower over the next three years – roughly equivalent to two more Steinhoffs.
The good news is that “relative to the overall corporate bond market, €250mn of CSPP selling is too small to pose a systemic risk,” Barclays says, before adding what I can only assume is a veiled (and ultimately successful) attempt at dry humor:
We agree with the market’s sanguine reaction: the selling of Steinhoff does not indicate an increased chance of the ECB selling credits that fall to Bs. This is fortunate, as we estimate it will hold c.€6bn of them by the end of 2020.