The Winning Trade In High Yield Corporate Bonds

 |  Includes: AWF, BLW, HYG, JNK, NHS, PHB, PHK
by: Ploutos

Over a long time horizon, BB-rated corporate bonds have outperformed their lower rated counterparts on an absolute return basis due to their lower default rates. An asset subsector providing higher absolute returns for lower variability of returns presents long-term investors the chance to generate alpha in the higher quality segment of the high yield universe. Below are annual total returns of the Barclays Capital U.S. High Yield index subdivided by ratings strata.

Examining loss rates by ratings category over a matched time horizon illustrates why BB-rated bonds have outperformed - investors have experienced much lower defaults historically on BB-rated bonds.

Click to enlarge

Examining the lowest ratings cohort in the bottom of the investment grade universe (BAA) versus the highest ratings cohort in the top of the speculative grade universe (BA) demonstrates that the increase in credit losses moving down in relative credit quality from Baa bonds to Ba bonds is minimal historically. This 57 basis point uptick in average annual credit losses is less than even the tightest spread differential between Baa and Ba bonds in the historical dataset, demonstrating that investors are always compensated for this incremental credit risk. The average spread differential of 180 basis points between Baa-rated bonds and Ba-rated bonds is triple the average incremental annual credit losses. The current differential of 263 basis points (Ba-rated corporate bonds have an option-adjusted spread of 468 basis points versus Baa-rated corporate bonds' OAS of 205) is close to the maximum credit loss differential in the past generation.

Click to enlarge

Investors over a long holding period have not historically been compensated on a loss-adjusted basis for moving down in credit risk between Ba-rated bonds and B-rated bonds. The 174 basis points in average spread pickup is less than the 236 basis point increase in average annual credit losses. The current 161 basis point spread differential may provide appropriate incremental compensation to investors over the near-term due to continued expectations of low near-term default rates, but is not sufficient compensation to offset the higher average annual credit losses expected on these bonds over a longer horizon.

While relative value between B-rated bonds and Ba-rated bonds may not be overly compelling in a historical basis, Caa-rated bonds look even more expensive in this historical context. On average, Caa-rated bonds require an additional 1,033 basis points of spread to offset the average annual credit losses, but the market is currently only compensating investors with an incremental spread of 324 basis points on a matched maturity basis. Markets have only provided investors with an appropriate risk premium for Caa-rated bonds during extreme market duress, which is why Caa-rated bonds have underperformed on a total return basis over this long sample period.

Ba-rated bonds provide superior returns over a long time horizon. Given the small amount of incremental default risk between Baa-rated bonds and Ba-rated bonds, why do these bonds trade at such a relatively wide spread? In the institutional market, there are occasionally forced sellers of fallen angels, bonds that migrate from investment grade to speculative grade. Higher capital charges on lower-rated securities, investment grade only mandates, and financial covenants that limit below investment grade bond holdings could lead to additional selling pressure on Ba-rated bonds that lower the relative price of these securities. If the speculative grade market more efficiently prices forward expected losses, then the regulatory frictions at the edge of the investment grade and speculative grade markets could be what is driving inefficient (and favorable) pricing of Ba-rated bonds.

Since market leverage is relatively inexpensive for large money managers compared to the cost for the typical retail investor, the ability to obtain leverage through these leveraged CEFs should be a benefit. While market yields are historically low, credit spreads remain relatively wide by historical standards, which should drive managers' ability to profitably capture this net spread. Given what we know about historical returns of double-BB rated bonds versus their lower rated and higher leveraged cohorts, it might make sense to add leverage outside of the underlying companies' balance sheets. This external leverage could be more easily paid down during market stress than could the outstanding debt of one of the underlying companies. A search for a double-BB rated leveraged fund yielded disappointing results as these funds are weighted towards single-B rated bonds. Below is a snapshot of some of the larger leveraged closed-end funds operating in the speculative grade corporate credit markets.

Click to enlarge

Four of these funds - AWF, BLW, NHS, and PHK- had allocations to double-BB and better rated securities of greater than 50%. While AWF is constituted of 74% corporate bonds, its top ten holdings include sovereigns from Argentina, the Dominican Republic, South Africa, and Brazil, as well as a corporate debt of a Russian bank and a Russian natural gas company. While these bonds may perform well in the future, if you are looking for domestic corporate credit risk, this may not be the correct fund for you. BLW has 30% of its holdings in mortgage-backed securities, which skews the average ratings higher, and may be a source of additional fund leverage via the reverse repo market. Again, if this is not a specific risk you are targeting, then this fund may not be appropriate. My negative thoughts on PHK have been documented, and largely surround the abnormally large market premium (68%) relative to the net assets of the fund. NHS may be the best bet for investors looking for externally leveraged exposure to BB-rated corporate credit. The fund trades at a more palatable 6% premium to net assets and has an indicated yield of 8.5%. The fund has posted a trailing 5-yr average total return of 10.5% through a tumultuous credit cycle.

A better bet for investors looking to add exposure to double-BB corporate bonds might be in a passively indexed ETF. In the equity markets, it has been demonstrated that active portfolio management does not appear to drive excess returns over the market return through lengthy time periods. If active money management does not provide alpha, then you should not trust high yield bond fund managers' ability to correctly time the market and appropriately switch between lower-rated credits before default risk rises. If you have a long-term allocation to high yield bonds in your investment portfolio, then you may want to examine passive bond funds with high allocations to BBs rather than the lower rated holdings of the closed end funds tabled above.

Below is a snapshot of three of the largest high yield corporate bond index funds, HYG, JNK, and PHB .

Click to enlarge

These funds provide three distinctive advantages to the closed end fund universe: 1) a higher long-run allocation to double-BBs, which has provided superior absolute and risk-adjusted returns over time; 2) higher levels of liquidity given the much greater trading volumes on these index funds, and 3) much lower expenses - the indexed funds have expense ratios of 0.4% - 0.5%, a full percentage point lower than the average expense ratio of the closed end funds. Given that the spread differential between BB and B bonds was 161 bps, over half of the increased risk you are taking by owning these lower rated bonds in the closed end funds goes into the manager's pocket. Expect passively indexed high yield funds to take share from actively managed funds prospectively akin to what we have seen in the equity markets. In the meanwhile, enjoy the advantages of investing in double-BB bonds for their superior risk-adjusted returns while keeping some of the management fees in your own pocket.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.