In the last 10 months we have seen value in the fixed income market change substantially, whether we are judging value by comparison to historical levels, adjusting for changing default expectations, or simply on a relative basis between sectors as different parts of the market have been directly or indirectly stimulated by central bank buying.
This last point has continued to be a focus for us at TwentyFour, given the wide range of fixed income assets we invest in, and has at times led us to find some of the less widely held parts of the market more attractive.
We frequently use the iTraxx Crossover index as both a guide to European high yield movements (since it measures a portfolio of the 75 most liquid HY names in euros) and as a benchmark for mezzanine European CLOs. This latter comparison is for two reasons. First, CLOs are securitisations backed by leveraged loans, whose credit profile is similar to high yield (though loans are typically secured and with a lower recent default and loss rate than HY bonds, which are typically unsecured). The Crossover is also the index used by most CLO traders to hedge their market risk. So from both a fundamental default point of view, as well as a correlation point of view, it is a reasonable comparison to make. Thus when we look at the long term trends in both the Crossover and CLOs, we can measure the basis between them to see whether one looks particularly good value versus the other at any point in time and compare it to the long term average.
At the moment CLO spreads are tightening, as they are right across credit. However, we haven’t had any CLO primary deals come to market so far in 2021, so we are operating in a bit of an informational vacuum, at which point these cross-asset class comparisons can be useful.
At 378bp, the current spread pick-up of BB rated CLOs over the Crossover (the average rating of which is BB-) is around 80bp wider than its historical average of 297bp, while the pick-up on B rated CLOs is 604bp, around 100bp wider than its historical average. In other words, CLOs appear to be even better value relative to European HY than they normally are, and by quite an attractive margin. A buyer of B rated CLO notes now at a yield of around 8.5% would have a very nice coupon for income, but as a five-year bond if the basis to high yield was to tighten in the near term (and absent a general widening in spreads) the buyer could expect some attractive capital gain as well.
It is probably worth reiterating that even in the depths of Q2, as we reran our CLO pool performance models under increasingly negative assumptions, our worst case result was for CLO equity to take all the pain on defaults and for rated CLO bonds to receive all interest and principal due on them. Subsequently the performance of the loan pools has materially surpassed expectations – both ours and the market’s – and we continue to have high conviction in the credit performance of these positions.
We are expecting a general tightening of credit spreads in 2021 as investors can no longer get much yield from rates exposure, pushing credit tighter. This should imply a compression in spreads across the board, both between asset classes and across ratings bands. Put more simply, we think the spread basis of CLOs to the Crossover should move inside the long run average this year, implying there could be plenty of value to extract from the former.