Energy Complex Credit Agency Arbitrage: Inefficiency In CCC Rated Credit Risk

by: Dallas Salazar


I propose a thesis; that there is significant and widespread inefficiency in CCC rated Energy Complex credit.

I further propose that the individual investor is at a meaningful advantage in arbitraging this inefficiency.

I provide nine examples of Energy Complex names pricing at inefficiency; via both data and fundamental contextual breakout.

Financial Engineering evolution has created a dynamic where individual investors can use modern Capital Structure, Stress, and Game Theory to achieve outsized returns.

It has been argued that, "credit ratings may provide information on the quality of a firm beyond other publicly available information…rating agencies may receive significant company information that is not public" [Kisgen, Darren J., 2006, Credit Ratings and Capital Structure, The Journal of Finance]. It has also been argued that, "credit ratings are significant predictors of yield to maturity beyond the information contained in publicly available financial variables". Finally, that, "ratings apparently provide additional information to the market" [Ederington, Louis H., Jess B. Yawitz, and Brian E. Roberts, 1987, The Informational Content of Bond Ratings, The Journal of Financial Research]. I'd argue that at one point in time, this was true.

However, today, at least in the stressed space [and I'm referring within this note strictly to the Energy Complex] I'd argue that the opposite is true; that investors willing to put in the "human analysis"-time have a significant advantage over ratings agencies. This thesis becomes important, if ultimately proven out, in that investors can arbitrage this analysis disparity. Investors can achieve outsized performance [i.e. alpha] via a consistent front running of credit agency upgrades or downgrades [or both]. Investors can also be reactive to both, after the fact. Of course, this would only apply to upgrades and/or downgrades taking place if undeserving and as the result of not completely indicative measures of credit-risk.

Given the complexity of modern Energy Complex capital structures and given the complexity of the "new normal" global-commodity environment [and further, its interconnected dependence], I don't believe that credit agencies can keep up with individual investors in assessing total credit-risk. I believe that recent financial engineering and financial innovation evolutions [again, in the Energy Complex] have embellished this fact; or, put another way, have exacerbated this disadvantage facing the credit agencies. No longer are simple credit-risk indications like bond pricing, leverage multiples, etc., perfectly reliable in and amongst themselves. Some would argue that they never were.

I'd argue that, at the very least, this is true in the here and now. In fact, I'd go a step further into saying that in most cases these metrics become a distraction, when assessed separately of total Capital Structure and Game Theory, in that they're relied upon too heavily in the "new normal" environment.

With that as the backdrop of this note, I'd like to breakout several credit upgrade candidates while at the same time breaking down several "false-candidates". All candidates assessed and/or in breakout/breakdown are currently rated CCC +/-.

The Bond Market is [in most cases] Reliable as The New Credit Agency:

Given the efficiency of the bond market, I do believe that the bond market is reliable as "The New Credit Agency". For a bond isn't going to price into, say, the high-90's [referring to "cents on the dollar"] with ample credit-risk, right? Again, this is true in most instances. What I mean by this, in calling the bond market "The New Credit Agency", is that I do believe, in most cases, that bond pricing is an excellent starting point for determining credit-risk.

A large part of this has to do with the bond market's collective agility, as well as the bond market's collective "special situations" trading-buildout [dating back to 2008-2009]. This has reduced the liquidity issues that plagued bond trading in decades prior, especially in the stressed instances. These were formally noted by Patel, Evans, and Burnett in their stating that, "liquidity affects whether speculative-grade bonds experience abnormal positive or negative returns" [Patel, Jayen, Dorla Evans, and John Burnett, 1998, Junk Bond Behavior with Daily Returns and Business Cycles, The Journal of Financial Research]. I quote and source Patel, Evans, and Burnett only in that I want to contextualize why I believe pricing to be much more and nearly perfectly efficient in today's bond markets as opposed to bond markets prior.

However - and this is where I refer again to the caveat of "most" - modern capital structures and "Stress Theory" [my phrasing, not the academia's] do allow for instances where bonds trade at PAR [referring to "100 cents on the dollar"] while the obligated enterprise is still extremely stressed. Again, this is where the individual investor has his or her advantage. The individual investor can apply modern Capital Structure, Game, and Stress Theory when performing "human analysis" of the enterprise's credit-risk. Still, bond pricing is an excellent starting point.

But, Read the Bond Book - Not Just the Title:

That said, make sure to read the "Bond Book" and not just the "Title". This is to say, take the bond pricing and then determine the "why". This can be done secondarily via using capital structure analysis and then finally using Game Theory [at least this is the order in which I deploy the analysis]. Kamakura CEO Donald van Deventer eloquently summarized why the latter, Game Theory, is so important in deployment via a recent note he published via Seeking Alpha:

" The credit spread calculation assumes that bonds of different maturities and coupons have different cash flows in all scenarios. This is of course false in the default scenarios: all bonds of the same seniority have identical cash flows upon default: remaining interest payments are zero and the principal that will be paid is the recovery amount; and the payment date is the date [or series of dates] that recovery payments are made after the bankruptcy is resolved in court."

I point specifically to van Deventer's calling out that bonds become "class-based" [my phrasing - not van Deventer's] in default-risk scenarios [i.e. scenarios in which credit-risk is extremely high]. Effectively, when default is appearing to be imminent or of high probability, coupon and duration become nearly meaningless on a bond by bond basis; as the bonds become traded on a simple price-basis. The bonds, if in similar seniority classes [and/or collateralization classes] and if without some unique feature, become vehicles of recovery-value rather than vehicles of cash flow. The recovery-value is reflected in the points in which the bonds trade [i.e. the price of the bond].

To provide some clarity and to provide a real-world, real-time example, I'll offer Chesapeake Energy's (NYSE:CHK) junior bond pricing [i.e. bonds that are not of the "first" recovery class]:

Unless there are unique, one-off features or factors, all the bonds in the above data-visual should trade relatively similar in recovery value [again, referring to pricing]. You can see clearly, this is true.

But, given that Chesapeake Energy's long-term viability is still largely in question [8.5X Q3/2016 net debt/EBITDA, 7.6X Q4/2016 net debt/EBITDA, 6.3X EQ1/2017 net debt/EBITDA; ALL figures adjusted for ATLAS proprietary modeling] - one wouldn't expect these bonds to trade at PAR. However, if one understands that Chesapeake Energy management has made clear [in statement and, more importantly, in execution] that it is willing to engage in financial engineering to 1) reduce these debt obligations [as is the case in debt/equity swaps executed, in exchanges executed, in tender offers executed, etc.] and/or 2) extend duration risk [via refinancing] regardless of how punitive to the immediate-term financials and/or other parties in the capital structure, [READ: at the expense of fully encumbering the enterprise, increasing interest expense, dilution, shrinking the total enterprise, etc.] one realizes why the bonds can be fully-valued.

The "second class" of the capital structure [currently the junior bonds exampled above], in this instance, can leverage the creation of a "first class" above it [eventually this will come to be a massive, catastrophic stress provider to the "second class"] and the dilution of the "third class" below it to monetize at PAR. Effectively, this Game Theory-based realization [explained at nauseam in prior notes on Chesapeake Energy] allows for the bonds to be fully valued. Importantly, however, this does NOT indicate that Chesapeake Energy is of little or no credit-risk; as is implied by the collective bond pricing alone. These bonds, ultimately, are of much, much greater credit-risk than is implied.

I don't believe credit agencies will regularly and consistently perform this level of due diligence [in applying Game Theory upon Capital Structure and Stress Theory], so I believe that there will be arbitrage available upon any credit upgrade to Chesapeake Energy. Many in the financial community expect an upgrade to be imminent. The arbitrage referenced, in this instance, will be via the common equity in my humble opinion [and later via the junior debt].

Other, similar examples in which there exists a secondary and tertiary analysis-arbitrage, in my opinion, are:

  • California Resources Corp. (NYSEMKT:CRC): 9.2X Q3/2016 net debt/EBITDA, 11.4X Q4/2016 net debt/EBITDA, 8.7X EQ1/2017 net debt/EBITDA
  • Hornbeck Offshore Services (NYSE:HOS): 15.9X Q3/2016 net debt/EBITDA, 19.0X Q4/2016 net debt/EBITDA, 24.5X EQ1/2017 net debt/EBITDA
  • Northern Oil & Gas (NYSEMKT:NOG): 9.1X Q3/2016 net debt/EBITDA, 9.2X Q4/2016 net debt/EBITDA, 7.7X EQ1/2017 net debt/EBITDA
  • PetroQuest Energy (NYSE:PQ): 19.8X Q3/2016 net debt/EBITDA, 18.3X Q4/2016 net debt/EBITDA, 15.3X EQ1/2017 net debt/EBITDA

As contextual reference points [for the bond pricing exampled so far], I provide examples of two similarly stressed Energy Complex names:

  • Vanguard Natural Resources (NYSE:VNR): 9.8X Q3/2016 net debt/EBITDA, 11.6X Q4/2016 net debt/EBITDA, 18.4X EQ1/2017 net debt/EBITDA
  • EXCO Resources (NYSE:XCO): 71.2X Q3/2016 net debt/EBITDA, 36.6X Q4/2016 net debt/EBITDA, EQ1/2017 net debt/EBITDA is N/A

There is some argument in the case of Vanguard Natural Resources as to the ultimate recovery value of its Second Lien notes [referring to our "second class"] which explains the "seniority" [i.e. collateralization] driven pricing differential. I note this as a special situation to explain the irregularity. Given that its recovery value is expected to be substantially impaired, I included it within this data-set.

The Arbitrage Isn't Always Negative, However:

As stated in the section title, the arbitrage in "human analysis" isn't always negative, however. Take Comstock Resources (NYSE:CRK) as an example:

Comstock, while stressed the point of nearing default T24M, has executed masterfully into what I've called [in other forums] a future MBA-case study in stress management. Comstock is currently within single points of an enterprise value increase of force-converting substantially 40% of its debt to equity. These swap transactions have been handsomely rewarded so far into this current commodity crisis. Comstock, if you assume the force-conversion is executed, at just a 4% increase to enterprise value will be showing a 28% improvement in leverage ratio [based on a $3.00-$3.75 NYMEX gas price]. The stressed E&P also has multiple other levers to pull in destressing efforts, should it come to that.

In saying this, Comstock is an example of a firm in which is rated CCC+ but which shows to have bonds trading at PAR and substantial destressing in-hand/of high visibility. Comstock is an example of a firm, in my opinion due to a lack of attention to detail and due to the complexity of its situation, which is being mispriced [to risk] by credit agencies. The current analysis-arbitrage available is in Comstock equity, as the bonds are substantially trading at PAR.

Clearly, the bond market - Capital Structure and Game Theory inclusive [the junior bonds above, the 7.75% NT REDCONV 01/04/2019 USD 1000 and the 9.50% NT REDCONV 15/06/2020 USD 1000 issues, traded as low as 7 cents on the dollar on the lows] - has determined Comstock to be of higher credit quality [i.e. of lower credit-risk] than is being indicated by its rating. Again, with the arbitrage having been taken in the bonds - this leaves only Comstock equity to be traded.

Another example of a positive analysis-arbitrage to be traded would be Eclipse Resources (NYSE:ECR):

Contextualizing the Thesis:

Indiscriminate risk assessment is as dangerous as it is irresponsible. Frankly, I think it borderline negligent. That said, one understands that in a volume-based business - such as providing credit ratings - that inefficiencies are created as a result of maximizing monetization-utility. I believe this to be the case of credit rating agencies in the stressed space of the Energy Complex; that inefficiencies are created.

I should think it's been clearly exampled that the bond market, when one applies Capital Structure, Game, and Stress Theory, agrees with me. If this weren't the case, we wouldn't see the efficient, risk-curve compliant bond pricing that we see in the Scatter Plot below - which aggregates the above bond data:

What we would see is indiscriminate "bunching" or "clustering" of pricing [i.e. data-points]. The fact is, we do NOT. This fact is made more apparent and with emphasis when excluding outliers Vanguard Natural Resources and EXCO Resources from the Scatter Plot presentation:

In this Scatter Plot, we see the same efficient, risk-curve compliant [all factors included] bond pricing.

I believe in presentation of this thesis that investors now have a starting point for arbitraging credit agency analysis inefficiency.

Disclosure: I am/we are short CHK.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: All data sourced via: ATLAS Consulting, Kamakura Corp., TRACE, FINRA, Market Axess, Exchange Data International, and/or public filings via EDGAR

Editor's Note: This article covers one or more stocks trading at less than $1 per share and/or with less than a $100 million market cap. Please be aware of the risks associated with these stocks.