Global credit funds & CLO's
October 2020 | Issue 228
Published in London & New York.
Copyright Creditflux. All rights reserved. Check our Privacy Policy and our Terms of Use.
Opinion credit derivatives
Duncan Sankey
Gaining exposure to US IG credit through the index means giving up about 10% of potential value
Portfolio director and head of credit research Cheyne Capital
October 2020 | Issue 228
Shorn of rate risk (it will be back — but that’s a topic for another day), investment-grade credit spreads offer a compelling default risk premium under almost any scenario. Even at the pre-pandemic tights of February this year, spreads on the iTraxx Main (Europe) and CDX IG (North American) credit indices offered break-even compensation for cumulative five-year default rates of around 3.5-3.75%, a figure many times the average cumulative five-year default rate of 0.89% for any five-year period in the past 50 years.
Spread risk can be easily accessed through the credit indices, which are among the most liquid credit instruments and are easy to leverage. As long as the investor keeps to the on-the-run index, they should avoid major negative transitions in the credit quality of the constituents. In the worst case, they could always hedge out a dodgy credit by buying single-name protection.
So why pay up for an actively managed bespoke portfolio when selling protection on the indices is so easy?
Because, in fact, it isn’t. First, index spreads tend to undershoot the intrinsic value of the portfolio components. The spread basis (or skew) between the new CDX NA IG series 35 and the intrinsic value of its constituents is, at the time of writing, just shy of -5 basis points (it had been as much as -20bp on the previous series). This reflects the superior liquidity and frictional costs associated with trading the indices and a general preference to sell protection in index rather than single-name form.
From the start, investors that gain exposure to US IG credit through CDX NA IG are giving up about 10% of the value they could extract from the underlying credits.
Share this article:
Trim low spread names and buy rising stars
Also, the index has a preponderance of very low-beta, low-spread names. About 40% of the constituents of the on-the-run CDX NA IG index trade at or below 30bp, providing only 16% of the index’s intrinsic spread.
The value of taking exposure to these names is questionable and the application of leverage to them clearly inefficient. An active manager with access to a broader universe of investible CDS contracts would likely exclude these names, or at least downsize exposure to them and, in order to improve returns, upsize exposure to others within the index that research determined were a satisfactory credit risk. Conversely, an index investor is pretty much stuck with them.
An active manager might also seek to improve the overall spread by including a smattering of credits that research identifies as rising stars (credits transitioning from high yield to investment grade). These may well be included in a later iteration of the IG index, but will typically provide some spread uplift in the meantime.
quotation mark
A bespoke portfolio of IG credits can take advantage of the skew between indices and single names
Three names to hedge
At the other end of the spectrum, there are three names in the current on-the-run series of CDX IG that trade wide of 300bp. (The prior series contained five names that traded wide of 500bp.) It is admittedly unlikely that these would be candidates for default before the next roll, but they could come under further spread pressure before exiting. Good risk management might suggest hedging these names. However, to do so would trim another 5-6bp, or 10%, off the intrinsic spread of the portfolio.
Add these factors together and the proposition of accessing IG corporate spreads through the indices begins to look considerably less attractive. An actively managed fund, with selection informed by fundamental credit research, is better placed both to harvest default risk premium and defend returns.
Share this report: