An Oxymoron For 2018

The words “numerous” and “idiosyncratic” do not work well together beyond a certain point.

That’s something we pointed out a couple of weeks ago during what turned out to be a short-lived selloff in high yield.

What was particularly interesting about that episode was the lengths folks were going to in an effort to explain away spread widening by reference to weakness in certain pockets of the market. Communications was a sore spot last month and to the extent everyone was nervous about the mini-rout in junk spilling over and derailing the equity rally, it was useful to be reminded that the trouble was concentrated in certain sectors.

That said, there are multiple sectors facing secular challenges in HY. The term “idiosyncratic” ceases to make sense for an asset class when either i) too many sectors within that asset class become challenged, ii) certain sectors face challenges acute enough to imperil the entire asset class, or iii) both.

So there’s something paradoxical about saying that there are “numerous” pockets of “idiosyncratic” risk. And indeed, there’s something strange about saying that idiosyncratic risk can become a directional driver of risk appetite (something Goldman has suggested). How can something which is “idiosyncratic” become “contagious”? That’s a question EM investors deal with every, single day. Country-specific risk is ever-present in EM and depending on the circumstances, there’s always the potential for instability in one locale to end up denting sentiment across the entire complex.

On Thursday, BofAML was out with their European credit conviction year-ahead outlook and one of the sections in the piece is called “contagious idiosyncrasy: an oxymoron for 2018?”

We’re not going to spend an inordinate amount of time on this because it’s Saturday and as such, no one cares, but we did want to excerpt a few passages as this is an important point in a world where the central bank backstop for credit markets (most clearly manifested in CSPP) will presumably be withdrawn over time. Here’s BofAML: