June 5 - June 18, 2020
- Despite an increase in volatility driven by renewed concerns regarding a resurgence of the coronavirus, leveraged credit markets produced modest gains.
- Flows and new issuance remain strong for high yield bonds, though not as much for leveraged loans.
- We briefly analyze the industry standard calculation of the recovery rate on defaulted debt, highlighting some of the limitations of such measure.
On June 5th, investors were greeted with a much better than anticipated May U.S. employment report, with an increase in non-farm payrolls of 2.5 million relative to consensus for a decline of 7.5 million. The report also reflected a decline in the unemployment rate from 14.7% in April to 13.3% in May, also moving in the opposite direction as forecasts, which called for an increase in the unemployment rate to 19%. Such a rate would have represented the highest level since the 1930s. In addition, retail sales for May increased by a record 17.7% month-over-month, handily beating expectations for an increase of 8%. Such reports added to an already optimistic outlook amongst investors for a strong economic recovery beginning in the third-quarter, driven by a largely successful reopening of the U.S. economy.
However, over the past two weeks, volatility returned to markets at times, with June 11th representing the worst daily performance for many risk markets, including leveraged credit markets, since the March sell-off. Such performance was largely driven by fears of a potential “second wave” of coronavirus cases occurring in the U.S. (and globally) as infections and hospitalizations began to accelerate in many southern and western states in particular. In addition, just the day before on June 10th, U.S. Federal Reserve (“the Fed”) Chairman Powell threw some cold water on the two-plus-month rally when he noted that the impact of COVID-19 on the economy could be long lasting. However, by June 15th, the Fed was back in the markets good graces, after it announced that it would begin buying corporate bonds as soon as the following day, consistent with the Secondary Market Corporate Credit Facility that it previously announced on March 23rd, which up until such point had only purchased ETFs. As part of this announcement, the Fed clarified its anticipated steps moving forward, pleasing the market as it essentially broadened the bonds eligible for purchase. Most importantly, by following through on the program as the market expected, the Fed retained its credibility amongst market participants, which may prove critical in the event that market conditions deteriorate again. Additionally, over the past two weeks, multiple countries, including the U.S., either enacted or contemplated further stimulus measures in an effort to bolster the economic recovery that may now be underway.
Notwithstanding these recent events, over the past two weeks, the high yield bond and leveraged loan markets generated gains of 0.42% and 1.06%, respectively, as set forth in Chart 1 below. During the period, CCC-rated high yield bonds outperformed B-rated and BB-rated bonds, as such segments produced gains of 1.60%, 0.04% and 0.35%, respectively. This marks the continuation of a trend that began in May of CCC-rated high yield bonds outperforming their higher-rated peers, after such bonds had significantly underperformed at the beginning of the crisis; however, on a year-to-date basis, CCC-rated high yield bonds are still lagging both B-rated and BB-rated bonds by a sizable margin. Similarly, within the leveraged loan market, CCC-rated loans were the top performers during the period, producing gains of 2.31% compared to B-rated and BB-rated loans, which returned 0.95% and 0.68%, respectively.
In addition, sector performance was mixed over the past two weeks, with just over half of the sectors in the high yield market generating positive returns, with the Energy sector the top performer while the Media sector lagged. Among leveraged loans, Retail was the top performing sector while the Consumer Durables sector underperformed. Furthermore, performance by issue size shows that smaller deals in the high yield bond and leveraged loan market continue to lag their larger peers year-to-date. However, during the period, high yield bonds smaller than $750 million in size outperformed those high yield bonds that were larger than $750 million in size, extending a trend that began in mid-April and represents a sharp contrast to performance observed during the earlier stages of the pandemic (see Tables 2 and 3 in the Appendix for more information). Meanwhile, for leveraged loans, performance by issue size over the past two weeks shows that loans smaller than $500 million in size modestly outperformed their larger peers. Lastly, year-to-date through June 18th, the ICE BofA US High Yield Index and Credit Suisse Leveraged Loan Index have produced losses of -3.20% and -3.80%, respectively.
Chart 1 – Leveraged Credit Daily Performance (%): June 5th – June 18th
Sources: ICE, Credit Suisse
Turning our attention to flow data, as Chart 2 below shows, high yield bond mutual funds tallied a meaningful $6.4 billion of inflows during the two weeks ending June 17th, though such inflows were heavily skewed towards the first week. This recent activity marks the 12th consecutive weekly inflow for high yield bond mutual funds, a period that includes some of the largest weekly inflows on record. Year-to-date inflows for high yield mutual funds total $31.9 billion as of June 17th. Given the low yields offered in other fixed income markets along with the Fed’s implicit backstop of “fallen angels” and high yield ETFs, positive flow activity over the past couple of months makes sense. Conversely, after experiencing a modest inflow the previous week, leveraged loan mutual funds saw a weekly outflow for the week ending June 17th. Such funds have seen net outflows in 77 of the last 82 weeks and year-to-date outflows total $20.9 billion.
Chart 2 – Weekly Fund Flows ($ million)
Sources: Lipper; J.P. Morgan
Meanwhile, despite the increase in volatility, the primary market for high yield bonds remains very active with over $40 billion in new issuance month-to-date, after April and May combined for over $80 billion in primary market activity. The Fed’s announcement that it will begin purchasing corporate bonds as soon as this week should support continued strong primary market activity in the high yield market. Additionally, after being relatively dormant during April and May, month-to-date new issuance in the leveraged loan market already exceeds the total issuance in May. However, the pace of leveraged loan new issuance still lags what would be considered normalized levels. Overall, the demand environment for floating rate investment products remains mired by the low interest rate environment, outflows from retail funds, and most importantly the relative lack of CLO formation. Nonetheless, after bottoming out in March, new CLO activity has shown improvement.
After the number of defaults declined in May following April’s record high (April was also the fifth largest monthly volume of defaults on record), the number of defaults in June has already matched May’s total (and is higher on a volume basis). High yield bonds account for the vast majority of default volume month-to-date, with leveraged loans only representing approximately 20% of the total. Including June’s month-to-date numbers, 2020 already ranks as the second highest annual default total on record, trailing only 2009. Unsurprisingly, given the significant decline in oil prices this year, the Energy sector has the largest number of defaults thus far in 2020, followed by the Retail and Consumer Products sectors, which each have been hit hard as a result of the ongoing pandemic.
As default rates in the high yield bond and leverage loan markets have increased in recent months, and with further increases on the horizon in the coming months/quarters, we thought that it would be timely to take a deeper dive into an important component of defaults, namely, the recovery rate. In this discussion, we are using the standard industry definition of recovery rate, which is based on the price that the defaulted issue is trading 30 days following the default, stated as a percentage of face value. For example, if a bond with a face value of $100 is trading at $50 when it defaults, and 30 days later such bond is trading at $30, the recovery rate will be 30% ($30/$100). Recovery rates, like default rates, are typically calculated over a trailing twelve-month (“TTM”) period. According to JP Morgan, in May 2020, the TTM recovery rates for both high yield bonds and first lien leveraged loans reached their lowest levels on record at 15.8% and 45.4%, respectively.
The point of this commentary is to provide insights into some of the limitations of the standard calculation of recovery rates in providing useful information, as well as to discuss factors that can cause recovery rates to increase or decrease. For starters, recovery rates are often compared to a longer-term average, which typically comprise 15-to-25 years of data given the relatively limited history of reliable data on the high yield bond and leveraged loan market (with even less for individual sectors). Furthermore, on a year-to-year basis, such data can be very volatile, as Table 1 below displays, making averages less reflective of the actual data. As a result, such averages are not very informative as to what recovery rate one should expect to occur in any given year. Table 1 below displays the average as well as minimum and maximum values for the recovery rates by sector, as well as the entire high yield market, based on data from the last 12 calendar years, and in our view provides a good example of the limitations of simply referencing average recovery rates when trying to assess the consequences of any particular defaulted security.
Sources: J.P. Morgan; Moody's Investors Service; Markit; S&P LCD
Recovery rates are based on calendar years 2008 -2019 and are issuer-weighted, based on prices 30 days after the default date
That being said, there is a clear pattern in the high yield market that recovery rates decline when default rates increase, as displayed in Chart 3 below. DDJ attributes the decline in recovery rates as defaults increase in large part to the market and economic environment that typically occurs during a spike in defaults. More specifically, based on the chart below, the last three significant increases in default rates (and corresponding decreases in recovery rates) occurred during recessions (specifically, 2000/01, 2008/09, and the current period). In a recessionary environment, the equity value of many defaulted companies is also depressed, and their near-term operational prospects are oftentimes not great, which reduces a company’s total enterprise valuation. As a result, estimates of the ultimate recovery on the defaulted debt also typically decline, making such debt less attractive to investors. In addition, the sell-off in the high yield market that generally occurs due to a recession is more severe and can last longer than a market correction or period of risk-off sentiment. In DDJ’s view, in an environment of heavy selling, many bonds, in particular bonds that are at risk of default, are sold more aggressively than would be the case in a more normalized market environment when there is a better balance between buyers and sellers. Such a dynamic can result in bonds defaulting at particularly low prices, requiring greater price appreciation over the following 30 days to otherwise boost the recovery rate as reflected by JP Morgan (and other similar market participants that track such statistic) as of such date. Moreover, when the number of defaults increases, there is a supply/demand imbalance with an elevated supply of defaulted debt relative to the demand of the specialist investors who target investments in such debt, further hampering any potential price recovery particularly in the short-term (30-day) timeframe. DDJ believes all of these factors contribute to lower measured recovery rates when defaults increase, though such lower recovery rates have little, if any, predictive power of the actual recovery an investor will ultimately recoup with respect to any particular defaulted security.
Chart 3 – U.S. High Yield Annual Default & Recovery Rates
Sources: J.P. Morgan; Moody's Investors Service; Markit; S&P LCD; LTM = twelve months through 5/31/2020
Recovery rates are issuer-weighted and based on prices 30 days after the default date
Most importantly, from DDJ’s perspective, recovery rates calculated according to industry standards oftentimes do not accurately depict the true recovery for holders of such defaulted debt. In our view, changes in the trading price of defaulted debt for a limited 30-day period post-default provide little information on how holders of such debt actually fared over the course of the default and, in many cases, ensuing restructuring. Investors are often, though certainly not always, aware that an issuer is going to default on its debt obligations before such default formally occurs. Such investors often make a decision to either sell prior to the default occurring (as some investment managers are prohibited by client guidelines from holding defaulted debt so they are motivated to sell before such default occurs) or to hold the debt through the default and associated restructuring, oftentimes playing an active role in the process. DDJ believes a much more meaningful metric for investors is to measure the performance of the defaulted debt after giving effect to the returns generated upon the ultimate disposition of the investment, which may occur some period of time following a successful workout or restructuring of the underlying defaulted security.
In summary, overall, the leveraged credit markets appear optimistic that the reopening of the U.S. economy will proceed on a relatively successful trajectory, with such optimism buoyed by improving economic data. That said, the markets are becoming more sensitive to key metrics that could signal a resurgence of the coronavirus, such as an increase in daily new cases, positive testing rates, and hospitalizations. As a result, periods of heightened volatility, similar to that experienced recently, could once again begin to occur on a more regular basis. However, the Fed has signaled that it remains committed to do whatever is necessary to contain the economic fallout should conditions worsen, and its recent actions have given the market no reason to doubt such commitment. In addition, any progress on medical treatments or a vaccine for the coronavirus has the potential to improve not only the sentiment of market participants, but the broader public at large. At DDJ, we remain optimistic, but also realistic, and continue to believe that a V-shaped recovery will be difficult to achieve, particularly given current levels of unemployment, and foresee a slightly bumpier path to recovery than perhaps what is being assumed by market participants more broadly. In the current market environment, we believe that the recent outperformance of lower-rated debt, which lagged the initial phase of the market recovery, should continue and that the year-to-date relative underperformance of such segment as compared with the higher-quality tier should compress over time.
Appendix I – Additional Charts as of June 18, 2020
Chart 4 – ICE BofA US High Yield Index Evolution of YTD Performance (%) by Quality
Chart 5 – ICE BofA US High Yield Index YTD Performance by Sector (%)
Chart 6 – Credit Suisse Leveraged Loan Index YTD Performance (%) by Quality
Source: Credit Suisse
Chart 7 – Credit Suisse Leveraged Loan Index YTD Performance (%) by Sector
Source: Credit Suisse
Chart 8 – ICE BofA US High Yield Index Option-Adjusted Spread in Basis Points (bps) – December 31, 1997 through June 18, 2020.
Chart 9 – ICE BofA US High Yield Index Cumulative Performance (%) by Quality: January 1, 2019 to June 18, 2020
Table 2: ICE US High Yield Index Total Returns and Issue Size Return Dispersion (%)
Table 3: CS Leveraged Loan Index Total Returns and Issue Size Return Dispersion (%)
Source: Credit Suisse
The information and views expressed herein are provided for informational purposes only, and do not constitute investment advice, are not a guarantee of future performance, and are not intended as an offer or solicitation with respect to the purchase or sale of any security. The inclusion of particular investment(s) herein is not intended to represent, and should not be interpreted to imply, a past or current specific recommendation to purchase or sell an investment. Any projections, outlooks or estimates contained herein are forward-looking statements based upon specific assumptions and should not be construed as indicative of any actual events that have occurred or may occur. This material has been prepared using sources of information generally believed to be reliable; however, its accuracy is not guaranteed. Investing involves risk, including loss of principal. Investors should consider the investment objective, risks, charges and expenses carefully before investing with DDJ.
Past performance is no guarantee of future returns.