Ok, so this is one of those things that’s intuitive but simultaneously noteworthy (usually when something’s intuitive, it’s not worth mentioning by virtue of the fact that it’s self- evident).
First, a (very) brief recap.
IG has been outperforming HY for months. This is something we’ve documented extensively. You want to keep track of this if you’re in junk, because it suggests that even if the bottom hasn’t fallen out yet (i.e. even if dip-buying in risk assets causes spreads to snap back tighter after episodic bouts of widening tied to things like North Korea), the market is getting more concerned relative to IG.
The reason for what you see in that chart is simple: the disconnect between fundamentals and spreads in HY is perceived as more glaring than any such disconnect in IG. More colloquially: they’re both probably in a bubble, but the HY bubble is more egregious and/or less sustainable. Here’s Goldman:
The two key ingredients underpinning a Long IG vs. HY spreads relative value view are: valuation and fundamentals. On the valuation side, the relative excess premium provided by HY bond spreads remains quite depressed. The excess premium net of our own estimates of expected losses offered by B-rated bonds vs. their BBB-rated counterparts remains close to its post-crisis lows. On the fundamental side, the higher exposure of the HY market to secularly challenged sectors and a decelerating second derivative for US growth should continue to favor IG over HY spreads.
Ok, now to the point. Note the last bolded bit in those excerpted passages. What do you imagine that might mean for dispersion within HY versus IG?
Here’s Goldman again:
Exhibit 1 [shows] the spread differential between the 90th and 10th percentiles, or the inter-decile range, normalized by the median spread level across all the bond constituents of the iBoxx IG and HY indices. The message from Exhibit 1 is straightforward: when accounting for current tighter spread levels, bond-level dispersion in the HY market appears to be gradually approaching the highs seen in early 2016. By contrast, IG dispersion remains at the low end of the range of the past few years.
Is that a problem?
Well, at the risk of stating the obvious, that suggests there are a lot of land mines in HY and to the extent dispersion in junk continues to make new highs, the implication is that some of those are exploding.
Make of that what you will.