Fridson: Split risk profiles of high-yield investors

This commentary is written by Martin Fridson, a high-yield market veteran who is chief investment officer of Lehmann Livian Fridson Advisors LLC as well as a contributing analyst to S&P Global Market Intelligence.

What is the present market stance of high-yield investors? Judging by their acceptance of spreads on the lowest-rated issues that are narrow by historical standards, they are fairly unconcerned about risk. By certain measures that we discuss below, however, they are considerably more cautious. This week’s piece explores the currently split personality of the high-yield buyer.

"+355" means different things at different times
One basic demonstration of the above-described contradiction is detailed in the table below. The shaded line shows the March 12, 2021, option-adjusted spread (OAS) of +355 basis points on the ICE BofA U.S. High Yield Index and its breakdown between the spreads on the BB+B and CCC & Lower sectors. Similar breakdowns are provided for all month-end dates prior to 2021 on which the all-rating index’s OAS was +355 bps plus/minus 2 bps (see note 1).

Within this peer group, the current spread is composed of the second narrowest CCC & Lower OAS and the widest BB+B spread. It is no great stretch to say that today's high-yield risk preferences are different from those of September 2017. At that time, investors demanded 576 incremental bps for owning the lowest-rated sector, versus just 358 bps now. Such disparities are important to keep in mind when comparing high-yield spreads across time.

Here is another take on the spread data that reinforces the message of a prevailing dual attitude toward risk. At +310 bps, the narrowness of the present BB+B OAS would place it at the 26^th percentile of the historical observations (monthly, December 1996 to December 2020). By comparison, the CCC & Lower OAS of +668 bps would put it at the 10^th percentile. In effect, investors are buying BBs and Bs on the premise that one-quarter of the time in the past, credit risk has been lower than it is now, then turning around and buying CCC-C bonds on the premise only one-tenth of the time in the past has credit risk been lower than today.

Spread curve is misaligned with spread
Another indication that investors are of two minds involves the high-yield spread curve. This concept is discussed in "Spread curve returns to positive slope: valuation analysis." That piece provides links to the basic research that formed the basis of Martin Fridson, Yaxian Li, and Kai Zhao, "Solving the spread curve puzzle," The Journal of Fixed Income (Summer 2018), pp. 38-47.

In brief, we define the high-yield spread curve as the OAS of the 10 to 15 year segment of the high-yield index minus the OAS of the 3 to 5 year segment (note 2).

Most of the time, the spread curve is negative, i.e., the short-term spread is higher than the long-term spread. We explain that by the fact that in default, recoveries are the same for all holders in a given seniority class, without respect to which maturities they own.

MORE FRIDSON: Pitfall of high price on recent high-yield bond performance

As long as investors perceive nonnegligible near-term default risk in the market, they demand extra spread on short-term issues. Otherwise, the dollar prices on those issues would rise to levels that would put them at a great disadvantage, relative to holders of longer-dated bonds, in terms of expected recovery as a percentage of price paid. Only if investors perceive no material near-term default risk will the spread curve turn positive. Under those conditions, investors focus exclusively on the point that an issuer’s cumulative default probability rises with the passage of time. That makes the issuer’s long-dated bonds more likely to default at some point in their lives than its short-dated bonds.

The chart below documents that the spread curve is negative in the majority of months, by far. Positive curves are associated with high levels of risk tolerance, as indicated by low spreads-versus-Treasuries on high-yield bonds. Indeed, the ICE BofA U.S. High Yield Index's OAS explains 69% of the variance in the spread curve during the period covered in the chart (correlation [R] = -82.96%, R² = 68.82%).

The regression formula that relates the high-yield spread to the spread curve on the same date is:

Spread curve = -0.42 x OAS + 79.62 bps

Plugging the March 12, 2021, spread of +355 bps into the formula produces an expected spread curve of -137 bps. The actual curve on that date was -73 bps, connoting substantially greater optimism than the risk premium (OAS) suggests. The 64-bps disparity between the expected and actual spread curve does not qualify as an extreme, defined as one standard error (= 75 bps), but it can be fairly described as a misalignment.

Interpretation
Throughout this piece we have spoken of high-yield investors as if they were a monolithic group. Is it truly likely, though, that the very same portfolio managers who are content to buy CCC & Lower bonds based on one assumption about the level of credit risk in the system are making a very different assumption when deciding what to pay for BBs and Bs? A more plausible explanation for the current ratings-based breakdown of the high-yield spread is that demand is segmented. In fact, that may be the case in every period.

It is beyond the scope of the present piece to determine which investors are chasing yield on CCC & Lower bonds, while accepting a less negative spread curve than credit conditions warrant, and which are exhibiting caution about what they will pay for BB and B bonds. Clearly, though, market participants such as mutual funds, insurance companies, hedge funds, and pension plan sponsors have diverse investment objectives and strategies. Liability matchers may be responding very differently than short-run total return maximizers are to current market forces and opportunities.

Drilling down into these differences could provide useful intelligence regarding the future relative returns of rating- and maturity-based sectors. Even when spreads on higher-rated and lower-rated bonds are properly aligned, CCC-Cs can be expected to fall more sharply when credit risk increases. That performance gap is likely to be magnified if the present relative pricing pattern remains in place when the next major downturn occurs.

Research assistance by Bach Ho and Ducheng Peng.

ICE BofA Index System data is used by permission. Copyright © 2021 ICE Data Services. The use of the above in no way implies that ICE Data Services or any of its affiliates endorses the views or interpretation or the use of such information or acts as any endorsement of Lehmann Livian Fridson Advisors LLC's use of such information. The information is provided "as is" and none of ICE Data Services or any of its affiliates warrants the accuracy or completeness of the information.

Notes
1. In relating the BB+B and CCC & Lower spreads to the All HY spreads shown in the table, keep in mind that the relative proportions of the two subcategories have varied over time.

2. Technically, we use the option-adjusted spreads of the ICE BofA 10-15 U.S. Cash Pay High Yield Index and the ICE BofA 3-5 Year U.S. Cash Pay High Yield Index. ICE Indices do not provide comparable maturity breakdowns on the ICE BofA U.S. High Yield Index. Dependence on the subdivisions of the U.S. Cash Pay High Yield Index makes no significant difference. That index currently accounts for 99.22% of the market value of the ICE BofA U.S. High Yield Index, which includes payment-in-kind and zero-coupon as well as ordinary cash-pay issues.