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.
A quarter-century of empirical study of the high-yield risk premium (spread-versus-Treasuries) has identified its key determinants as the state of the economy, credit availability, and the Treasury yield, which is inversely correlated with the spread. Among many indicators of the state of the economy, including GDP, unemployment, and business confidence, capacity utilization stands out for its ability to explain the high-yield spread. The FridsonVision Fair Value Model, discussed below, also employs industrial production to characterize economic conditions.
Reported monthly by the Federal Reserve Board, the capacity utilization ratio covers the manufacturing, mining, and electric and gas utility sectors. The numerator is a seasonally adjusted index of output. The denominator is a capacity index designed to measure the maximum output that plants can maintain under a realistic work schedule, taking into account normal downtime and assuming the inputs required to operate the facilities are available.
Capacity utilization is of limited use in analyzing the social media networks, search engines, and other New Economy stalwarts of today’s stock indexes. In the high-yield universe, however, producers of raw materials and companies that convert them into physical goods remain prominent enough that capacity utilization shows a strong correlation with the asset class’ risk premium. Neither Facebook Inc. nor Google can be found in the ICE BofA US High Yield Index’s Technology sector. That subindex does contain old-fashioned makers of stuff, such as Xerox Holdings Corp., which manufactures printers, scanners, and accessories.
Bad things happen to such companies when the economy heads south, and demand declines. There are suddenly fewer units making contributions to cover fixed costs. Companies respond by cutting prices in an effort to boost volume. The pricing power that they enjoyed when they were running flat-out evaporates. Contribution per unit declines, further eroding profits. At the worst of a typical recession, one out of eight high-yield companies defaults on its debt within a 12-month period. The source of the trouble, the decline in units being produced to fill current demand, is captured in the capacity utilization ratio.
The current recession is hardly typical, either in its catalyst, a horrific pandemic, or in the policy response. By stating its willingness to buy corporate bonds, including even some speculative-grade ones—recent fallen angels—the Federal Reserve has provided unprecedented support to the debt market. The prevailing levels of capacity utilization and the option-adjusted spread (OAS) on the ICE BofA US High Yield Index demonstrate the extent to which the Federal Reserve’s extraordinary intervention has disrupted traditional links between economic conditions and valuations.
Capacity utilization, which came in at 71.4% in August, was not out of line for a recession month. The atypical thing was the high-yield index’s OAS on Aug. 31, +502 basis points. From the Dec. 31, 1996, inception date of OAS data on the ICE BofA US High Yield Index, through the end of 2019, there were five months in which capacity utilization nearly matched August 2020’s, i.e., that were in the range of 71.4% plus/minus 1%. August’s +516 bps OAS was far below the bottom range of OAS in those months, which was +561 bps to +1,812 bps.
Flipping it around underscores how out of whack the present Capacity Utilization Rate/OAS relationship is. Prior to 2020 there were nine months in which the OAS nearly matched August’s +502 bps, i.e., that were in a range of +502 bps plus/minus 10 bps. Mean capacity utilization in those months was 77.6%. If that does not strike you as a large disparity with August 2020’s 71.4%, consider this: The lowest capacity utilization recorded in those nine months was 74.9%. The difference between that figure and last month’s, 74.9% minus 71.4%, was 3.5 percentage points. That is equivalent to 95% of a standard deviation (3.7 percentage points), calculated for the full capacity utilization series since December 1996. In short, the Aug. 31, 2020 high-yield risk premium was, for all intents and purposes, extremely lower than the lower extreme previously observed when capacity utilization approximated August 2020’s level.
The high-yield spread is similarly out of line, by historical standards, with the credit availability measure employed in the FridsonVison Fair Value Model. As a result of these disruptions of traditional relationship, our Fair Value analysis currently shows the high-yield market to be absurdly overvalued, as indicated by the fever line in the chart below. The only way to make sense of current high-yield valuations is to acknowledge that the Fed’s unprecedented market intervention is maintaining prices at levels that would otherwise be unimaginable under current economic and financial market conditions.
For the record, our model’s fair value estimate for the ICE BofA US High Yield Index’s option-adjusted spread dipped from +1,300 bps in July, to +1,294 bps in August. That decline of six bps was more than offset, however, by a narrowing of the actual OAS by 14 bps, from +516 bps as of July 31, to +502 bps as of Aug. 31. The shortfall between the actual and estimated spreads consequently increased by eight bps. At 792 bps, the Aug. 31 disparity was a six-sigma event (a gap of 6.4 standard deviations).
Contributing to the slight decline in the estimated spread between July 31 and Aug. 31 was a rise in the five-year Treasury yield, which is inversely correlated with the spread, from 0.21% to 0.26%. In addition, capacity utilization edged up from 71.1% to 71.4%. On the other side, growth in industrial production dropped from 3.5% to 1.0%. Credit availability, which is reported only quarterly, was unchanged at a 71.3-percentage-point excess of banks that are tightening credit standards for medium-sized and large companies over those that are easing standards. The dummy variable for quantitative easing remained at one to indicate that this form of accommodative monetary policy remained in place. The default rate, a backward-looking variable with only a minor impact on the spread, increased from 5.9% to 6.2%.
These conclusions are drawn from the updated methodology presented in earlier analyses. In brief, we find that 80% of the historical variance in the ICE BofAML US High Yield Index’s option-adjusted spread (OAS) is explained by six variables:
- Credit availability, derived from the Federal Reserve’s quarterly survey of senior loan officers;
- Capacity utilization;
- Industrial production;
- Current speculative-grade default rate;
- Five-year Treasury yield; and
- A dummy variable for the period covered by quantitative easing (QE).
Each month we calculate a Fair Value spread based on the levels of these six variables. The extent of high-yield overvaluation or undervaluation is determined by the difference between the actual OAS and the Fair Value number (“estimated”). We define an extreme valuation as a divergence of one standard deviation (124.5 bps) or more from Fair Value. The monthly difference between the actual and estimated OAS is tracked in the chart above.
Covenant quality slightly better in August
On a scale of 1 (strongest) to 5 (weakest), FridsonVision’s adjusted version of Moody’s covenant quality score improved a touch, from 4.33 in July to 4.31 in August (see chart below). This move contrasted with a substantial worsening from 4.17 to 4.56 by Moody’s methodology.
The disparity in the latest results arose from a major shift in issuance toward Ba rated deals, which accounted for 71.2% of August’s issuance. That was up from 46.7% in July and double the historical average Ba share of 35.6%. As is typically the case, Ba issues had the weakest covenant quality among the three rating categories — 4.71 versus 4.22 for B and 3.75 for Caa. The bottom rating tiers accounted for just 1.9% of August’s issuance, down from 6.7% in July and far below the 18.1% historical average.
Because issuance shifted toward higher-rated, weaker-covenant deals, the Moody’s monthly series indicated a deterioration in covenant quality. FridsonVision’s methodology, on the other hand, showed that on a rating-for-rating basis, covenant quality held steady and actually improved a pinch. Note that the differences between the FridsonVision and Moody’ calculations tend to be minimized in the three-month-rolling series that the rating agency emphasizes in its reporting.
Research assistance by Lu Jiang and Zhiyuan Mei.
ICE BofA Index System data is used by permission. Copyright © 2020 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.