MODEL DESCRIPTIONS
Descriptions of FridsonVision High Yield Strategy Models
Abbreviations Used Below:
bp = Basis Point
EM = Emerging Markets
ERM = Equalized Ratings Mix
HY = High Yield
IG = Investment Grade
OAS = Option-Adjusted Spread
Note on FridsonVision’s Equalized Ratings Mix (ERM) Methodology
A direct OAS comparison of the two regional indexes is not highly informative. For example, the EM is considerably higher-rated than the U.S. universe. Specifically, on May 31, 2024 the ICE BofA High Yield Emerging Markets Corporate Plus Index had a 21-percentage-point greater concentration of market value in BB and a 5-percentage point lesser concentration in CCC & Below than the ICE BofA US High Yield Index. EM consistently trades wider than the U.S., but the gap would be even greater if the U.S high-yield market were not lower-rated on average. A further complication arises in collecting data for a historical observation period, since the ratings mixes of both indexes change over time, reducing the comparability of observations on different dates. To address such issues, we normalize for differences in ratings mix when comparing spreads on regions and industries.
BB Distressed Bonds
Our historical research has found that that when 5% or more of the issuers represented in the ICE BofA US Distressed High Yield Index are rated BB, the market is overstating the default risk of those issuers’ bonds.
Evidence from the 2020 recession upholds the inference that at such times, BB bonds trading at distressed levels (+1,000 basis points or more above Treasuries) collectively represent an attractive risk-reward proposition. The ICE BofA US High Yield Index reached its cyclical peak OAS (measured monthly) on March 31, 2020 at +877 bps. On that date, 62 issuers were represented in the distress index with bonds rated BB and carrying option-adjusted spreads of +1,000 bps or more. More than two-and-a-half years later only one of those issuers—Ruby Pipeline LLC—had defaulted. That default, which was accompanied by a bankruptcy filing, did not occur until April 1, 2022. Accordingly, the March 31, 2020 BB distressed universe had a 0% percentage-of-issuers default rate over the succeeding 12 months versus an 8.52% rate for the U.S. speculative-grade-bond-only universe, as reported by Moody’s Investors Service.
CCC & Lower versus BB/B
The risk premium on the ICE BofA CCC & Lower US High Yield Index is largely a function of the ICE BofA BB/B US High Yield Index’s risk premium. That is, 78.8% of the variance in the CCC-C index’s OAS is explained by the BB/B index’s OAS. To calculate fair value for the CCC-C spread at a point in time we apply the following regression formula:
y = 2.34x +73.44
Where:
y = 2.34x +73.44
Where:
x = OAS of the ICE BofA BB/B US High Yield Index
y = OAS of the ICE BofA CCC & Lower US High Yield Index
We define an extreme as one standard deviation = 254 basis points. So, for example, if the formula produces a fair value estimate of +1,000 bps, then an actual OAS of more than +1,254 bps indicates that the CCC & Lower sector is relatively undervalued and an actual OAS of less than +746 bps indicates that the CCC & Lower sector is relatively overvalued. (Note that the CCC & Lower index does not include defaulted issues, which are rated D by Standard & Poor’s and Fitch Ratings.)
Distressed Debt’s Attractiveness
The distressed segment of the high yield universe consists of bonds with option-adjusted spreads of +1,000 basis points or more. It is tracked by the ICE BofA US Distressed High Yield Index. Our historical research has found that when the market-implied one-year default rate forecast exceeds (falls short of) Moody’s forecast by one percentage point or more, the distressed sector tends to return more than (less than) its historical average return over the succeeding year.
We recommend that investors who customarily include some distressed bonds in their high yield portfolios overweight (underweight) the distressed sector when the market-implied forecast exceeds (falls short of) Moody’s forecast by one percentage point or more. When the market-implied default rate forecast is within plus/minus one percentage point of Moody’s forecast, we recommend a Neutral weighting for the distressed sector. “Neutral” means the investor’s long-run target allocation to distressed bonds, which may be adjusted for short-term, tactical reasons.
To derive the Moody’s forecast for the one-year default rate on U.S. speculative grade bonds, we take the agency’s current percentage-of-issuers forecast for U.S. speculative grade debt (which includes loan-only issuers) and multiply it by the fraction consisting of the agency’s all-regions bonds forecast divided by its bonds & loans forecast.
We calculate the market-implied default rate using the following formula:
y = 0.133 times x-0.534
Where:
y = Distressed default rate
x = Percentage of issues in the ICE BofA US High Yield Index with option-adjusted spreads of +1,000 bps or greater (see note 1)
The market-implied default rate forecast for the next 12 months is:
Distress ratio times distressed default rate
Emerging Markets versus U.S
To determine relative value for these two regions on a rating-for-rating basis, we compare option-adjusted spreads on the ICE BofA Yield US Emerging Markets Corporate Plus Index and the ICE BofA US High Yield Index. We recommend overweighting (underweighting) EM when its OAS is unusually wide (narrow) versus its U.S. counterpart according to our Equally Ratings Mix methodology (see above).
Specifically, we calculate the median OAS within each rating category for each of the two regional indexes. We then weight the U.S. OAS medians the respective weights of the rating categories to produce a U.S. high yield OAS that we use exclusively for this comparison. Next, we weight the OAS medians for the EM rating categories by those categories’ percentages of the U.S. high yield index’s market value to produce an EM high yield OAS. We then subtract the U.S. high yield OAS derived in this manner from the EM high yield OAS derived in this manner to produce an OAS differential between the two regions.
The historical base period for this analysis produced the following quartiles for the distribution of the EM – US high yield OAS differential:
| Quartile | Basis Points |
|---|---|
| 1 | Equal to or greater than +354 |
| 2 | Equal to or greater than +182 |
| 3 | Equal to or greater than +123 |
| 4 | Less than +123 |
We recommend overweighting EM with the OAS differential is in Quartile 1. We recommend underweighting ERM when the OAS differential is in Quartile 4. When the OAS differential is in Quartile 2 or 3 we recommend a Neutral weighting on EM high yield. “Neutral” means the investor’s long-run target allocation to EM, which may be adjusted for short-term, tactical reasons.
Europe versus U.S.
To determine the relative value of these two regions, we break down the ICE BofA European and U.S. Non-Financial High Yield Constrained Indexes into their rating categories. Next, we calculate the median OAS for each regional rating group. We then calculate a market-weighted OAS for the U.S. index by multiplying its rating-category median spreads by the market weightings of its rating categories. By then multiplying the European rating-category median option-adjusted spreads by the U.S. rating-category percentages, we arrive at a valid means of comparing U.S. and European high yield debt on a rating-for-rating basis. It tells us what the OAS would be on the European index if its ratings mix were identical to the U.S. index’s.
We subtract the U.S. OAS calculated in this fashion from the European OAS calculated in this fashion and determine where that differential stands in the quintiles derived from our historical observation period:
| Quintile | Basis Points |
|---|---|
| 1 | Equal to or greater than +149.1 |
| 2 | Equal to or greater than +48.2 |
| 3 | Equal to or greater than +20.1 |
| 4 | Equal to or greater than -12.1 |
| 5 | Equal to or less than -12.0 |
We recommend overweighting Europe when the differential is in Quintile 1 and underweighting Europe when the differential is in Quintile 5. We recommend a Neutral weighting on European high yield debt when the differential is in Quintiles 2 to 4. “Neutral” means the investor’s long-run target allocation to Europe, which may be adjusted for short-term, tactical reasons.
Investment Grade versus High Yield
Our relative value analysis for these two asset categories is directed at investors who use both asset categories. It employs the option-adjusted spreads on the investment grade ICE BofA US Corporate Index and the ICE BofA US High Yield Index. We measure the impact of spread on relative return in terms of workout periods, asking:
How many quarters does it take for HY to overtake IG in total return if the quarter ends with the HY – IG OAS differential at x bps?
Empirical study found that when the high yield OAS exceeds the investment grade OAS by more than 700 basis points, there is a strong probability that high yield will beat investment grade in total return in the next quarter. We recommend overweighting high yield under those conditions. Further, we found that when the high yield OAS exceeds the investment grade OAS by less than 265 basis points, there is a substantial probability that high yield will underperform investment grade for the next 2.5 years or more. We recommend underweighting high yield under those conditions. When the high yield OAS exceeds the investment grade OAS by 265 to 700 basis points we recommend a Neutral weighting of high yield. “Neutral” means the investor’s long-run target allocation to high yield, which may be adjusted for short-term, tactical reasons.
Industry Relative Value
FridsonVision’s industry relative model covers the 20 largest industries (by market value) in the ICE BofA US High Yield Index. Directly comparing the yields or spreads on them is uninformative, as they are affected by differences in ratings mix among the industries, as well as changes in those ratings mixes over time. The real question to ask is, “Which industries are currently rich or cheap on a rating-for-rating basis?”
To answer that question, we use the Equalized Ratings Mix methodology described above. Each industry’s resulting rating-specific spreads are then assigned the same weightings to calculate an ERM-based OAS for each industry. In the graph of the monthly output of this analysis, we plot each industry on the vertical scale according to the percentage by which its ERM-based OAS exceeds or falls short of the peer group average.
On the horizontal scale we plot industries by their Net Ratings Prospects. Each issue within the industry subindex has ratings outlooks or watchlistings of Positive, Stable, or Negative from some or all of the following agencies—Moody’s, Standard & Poor’s, and Fitch Ratings. If, for example, two agencies have the issue at Stable and one has it at Negative, the issue is classified as Negative. One Positive, one Stable, and one Negative results in a Stable classification. Adding up the classifications for all issues in the subindex produces an industry Net Ratings Prospect, expressed as a positive, zero, or negative percentage displayed on the graph’s horizontal scale.
We draw a regression line through the 20 industries’ plot points. Industries located above (below) the line are cheap (rich) relative to their peers on a rating-for-rating basis. Also of interest are the few industries located in the graph’s northeast and southeast quadrants. Location in the northeast quadrant indicates that an industry is trading wide to its ratings despite the rating agencies’ signaling that its ratings are likely to improve on balance. Such industries offer good value, by our reckoning. Relatively poor value, conversely, is provided by industries in the southeast quadrant, indicating that they are rich to their ratings even though the rating agencies say their ratings are likely to decline, on balance.
Market-Implied Default Rate Forecast
See “Distressed Debt’s Attractiveness,” above, for our method of calculating the market’s implied one-year default rate forecast on U.S. speculative grade bonds. We have previously shown why the more common method of backing out the default rate forecast from the high yield index’s spread-versus-Treasuries is flawed. For one thing, almost all defaults within a 12-month period occur among bonds with spreads of +1,000 bps or more at the start of the period. An index spread of +600 bps, for example, could represent either a lot of bonds with spreads of +500 bps to +700 bps (0% default rate expected) or many with spreads in the neighborhood of +500 bps and a few at +1,000 bps or greater (positive default rate expected).
Undervalued Bond Model
Our Undervalued Model identifies non-distressed bonds within the ICE BofA US High Yield Index that are cheaply valued and therefore have a high probability of outperforming the ICE BofA US High Yield Index. We have back-tested the model’s recommendations in selected 12-month periods and found highly favorable, performance-enhancing results in both rising and falling markets. Note, however, that in the extreme bear and bull phases around the Global Financial Crisis, investors appear to have sold and then bought indiscriminately. A value-based selection model did not generate alpha in those periods. Investors should not utilize the Undervalued Bond Model if and when they foresee a crisis as severe as the 2008 downturn.
The details of our Undervalued Bond Model, which employs financial data on the issuers, is proprietary. Note, however, that past returns for periods detailed here, on bonds selected by our methodology, have received attestation by the institutionally recognized boutique performance measurement consulting and GIPS® standards specialist firm TSG (also known as The Spaulding Group). Click here for TSG’s attestation.
Each month we highlight a few currently cheap bonds. Subscribers can view the full listings by clicking here.
Fair Value of High Yield Spread
Our Fair Value Model for the high yield asset class is a multiple regression analysis for which the dependent variable is the option-adjusted spread (OAS) on the ICE BofA US High Yield Index that is in line with the prevailing values of six independent variables:
Credit Availability
The percentage of banks tightening credit for medium- and large-sized companies minus the percentage of banks easing credit for those borrowers, as reported in the Federal Reserve Board of Governors in its quarterly Senior Loan Officer Survey.
Capacity Utilization
Reported monthly by the Federal Reserve Board of Governors.
Industrial Production
Month-over-month change reported by the Federal Reserve Board of Governors.
Default Rate
Latest-12-months’ U.S. speculative-grade, bond-only, percentage-of-issuers rate, as reported by Moody’s Investors Service.
Five-Year Treasury Rate
Effective yield on ICE BofA Current 5-Year US Treasury Index.
Dummy Variable for Quantitative Easing
1 = In effect, 0 = Not in effect
For portfolios that include high yield bonds along with other asset classes, we recommend overweighting (underweighting) high yield when the actual OAS exceeds (falls short of) the model-estimated value by one standard error or more. To avoid distortion by outlier OAS values that arose during periods such as the Global Financial Crisis, we emphasize a version of our Fair Value Model that excludes data for recession months.