Coming Soon To A Theater Near You: MBIA's $1 Billion World War Z

Frequent readers will recall that in the past, on several occasions, we expected that MBIA would rise due to two key catalysts: a massive short interest and the expectation that a BAC settlement would provide the company with much needed liquidity. That thesis played out earlier this year resulting in a stock price surge that also happened to be the company's 52 week high. However, now that we have moved away from the technicals and litigation catalysts, and looking purely at the fundamentals, it appears that MBIA has a new problem. One involving Zombies.

MBIA’s most recent issue is not new per se, as it dates back to 2003/2005 - as far as we can tell, however - the water looks like it is just starting to boil over and MBIA does not seem to have reserved for what appears to be a rather meaningful problem.  According to preliminary analyses, it does not appear that MBIA has ever really addressed this particular $1 billion of Zombie CLO insurance on their books, which looks to be coming due in short order (unlike their well-followed RMBS issues and long-dated future CMBS losses).

These freshly-surfacing problems stem from a particular pair of Zombie CLO’s – Zombie-I and Zombie-II (along with Zombie-III, illiquid/black box middle-market CLO’s).  While information is (intentionally?) difficult to gather, we have heard that MBIA would be lucky to recover much more than $400 million from the underlying insured Zombie assets over the next three years, which would leave them with a nearly $600 million loss on their $1 billion of exposure which would materially and adversely impact the company's liquidity.  And as it may take them a while to liquidate assets in a sure-to-be contentious intercreditor fight – their very own World War Z – MBIA may well have to part with the vast majority of the $1 billion in cash, before gathering some of the potential recovery.

The CLO’s in question are actually called Zohar CDO 2003-1, Zohar II 2005-1, and Zohar III, but instead of paying royalties to the Kabbalah by continuing to invoke reference to it here, henceforth we will refer to each of the distressed “Zohar” CLO’s as “Zombie” I, II, and III, respectively. They full structures are shown below:

Needless to say, even counting the expected $800 million ResCap settlement as a done deal, on top of all of their other structured finance issues, it appears that these short-dated Zombie CLO losses could present a serious problem for MBIA.


  • The Zombie CLO notes have recently been trading at distressed levels.  Just Tuesday, a BWIC with six pre-crisis original A+ rated CLO tranches saw five of the notes being offered between 95-101… and an originally AAA-rated Zombie tranche talked in the 60’s.
  • Recent Zombie ratings downgrades and Moody’s outright ratings withdrawals have been sources of deep noteholder concern, among other things, as regulated institution/bank holders are required to hold more capital against lower-rated assets.  Zombie noteholders also seem to be growing increasingly unhappy, as mature Zombie notes are not paying-down meaningfully and holders are further frustrated that they cannot easily gather useful information on the underlying assets, “loans” to small private companies controlled by the Zombie collateral manager.
  • Zombie noteholders appear to be organizing/forming an ad hoc group to strategize about their rights and potential remedies, as it appears that the Zombie collateral mostly comes from companies where the lender and the borrower are the effectively same person, with both sides of an underlying company’s ledger being controlled by the Zombie collateral manager – it appears that the Zombie “loans” are not really loans.
  • With a court ruling on 6/10/2013, MBIA lost their four-year battle with the Zombie collateral manager about getting their $120 million Zombie-I B-Note back; court documents provide some interesting insight into the conflict and history of the Zombie funds.
    • Extensive court documents describe the background to the lawsuit and legal proceedings in great detail – the irony of the situation is that it appears, in large part, that the case revolved around whether this B-Note had good value; the Zombie collateral manager successfully argued that there was no way that a rating agency would ever put an investment grade quality stamp on the shoddy note, which rating had been set as the trigger for handing the B-Note back to MBIA. Further, this B-Note seems now decidedly worthless; MBIA appears to have long since written this B-Note off (this B-Note issue is entirely distinct from the $1 billion insurance exposure described herein).
  • CEO Jay Brown forfeited 3 million MBIA shares in February (after failing to hit the five-year share price target) and sold 750,000 shares from his family trust in May at $14.50 after the Bank of America settlement-driven stock pop, leaving him owning only 700,000 shares, rather than the 4.5 million shares it appeared he controlled at the beginning of the year.
  • It is estimated that the maximum recoveries for MBIA’s insured notes are in the 40-45% range for Zombies-I & II  – Zombie-III was not wrapped by MBIA, but would be due recovery pari passu, as the three Zombie funds each extended loans to many of the same companies (i.e. a given Patriarch Partners portfolio company may well have pari passu loans from any combination of the three Zombie funds ).
    • Low $0.30’s total recovery estimates (on all Zombie loans combined) are based upon collateral information available from Intex reports, estimates of underlying companies’ debt held by the Zombie funds and anecdotal estimated orderly sale/liquidation values to be garnered from the monetization of the underlying small/middle-market private companies
    • Aside from MBIA, there may well be up to another $888mm of original AAA insurance wraps/CDS written against Zombie-I (up to $252mm, A3A tranche) and Zombie-III (up to $636mm, A1D/A1R/A1T tranches), as MBIA originally wrapped only 39% of Zombie-I’s original senior notes, along with all of Zombie-II’s original AAA notes
  • Widespread collateral defaults on/by 10/31/2015 seen unavoidable (when 75% of all Zombie loans mature, if noteholders do not accelerate beforehand) and would ripple through the three Zombie funds, putting all three into default, upon or immediately before the Zombie-I November 2015 senior notes’ legal final maturity date.  If this unfolds as accordingly, MBIA would need to pay Zombie-I insured noteholders $158mm November 2015 and then attempt to sell/liquidate all companies and/or collateral assets in order to garner their potential recovery; $871mm (less any pro rata Zombie-II recovery received) would then be due from MBIA to Zombie-II insured noteholders January 2017, unless the Zombie-II notes are paid-down by then (which now seems rather unlikely).
  • It appears that MBIA is entirely unreserved for the ~$600 million of potential losses on their Zombie insurance contracts and may have to come out-of-pocket for almost $1 billion over the next ~three years – at the Barclays financials conference just two weeks ago, MBIA explicitly stated that their $500 million of structured finance reserves are almost entirely intended to offset CMBS losses (as opposed to these Zombie High Yield Corporate CDO losses).
  • MBIA lists High Yield CDO #15 ($158mm insured) and High Yield CDO #23 ($871mm insured), apparently the Zombie deals in question, under the structured finance portfolio details on their website.
    • Other coverage of some of the relevant issues here can be seen in the 3/15/2013 issue of Asset-Backed Alert and in a 2011 multi-part Forbes online series

A little more detail on some of the key Zombie issues:

(i) Zombie CLO’s are not Paying-Down – Underlying “Loans” are not True Loans

  • With absolute power over the whole closed system, as the Zombie collateral manager also controls the equity in the Zombie CLO’s (so much for the purported benefit of the new risk retention rules) and is also the controlling shareholder (in some cases CEO) of nearly all borrowers of Zombie loans and is also the “Agent” on nearly all the Zombie “loans” >>> the “loans” are not really loans.  As both borrower and lender, any element of any “loan” can be amended at any time, as needed (maturity, covenants, interest rate, frequency of payment, etc.).  Taken to the extreme, we have heard stories of (a) Patriarch Partners companies being liquidated/having substantial assets sold and leaving in-place Zombie collateral loans to a then shell entity devoid of meaningful assets and (b) companies with existing Zombie collateral loans filing for bankruptcy, then selling-to and being repurchased-via additional new Zombie loans, with both the old and new Zombie loans continuing to be held on the books of the Zombie funds.  How can this be - who would let this type of circular arrangement without any checks-and-balances persist?
  • CLO liabilities typically begin to pay-down after the reinvestment period ends (as seen in the graphic below); as these Zombies are not really CLO’s/don’t really hold true senior secured loans, as detailed further below, the liabilities are not paying-down meaningfully.
    • 2003 vintage: of the 31 CLO’s issued in 2003, Zombie-I ranks 31st/31 in terms of repayment of the original AAA notes’ balance – 29/31 have entirely repaid their AAA’s and one deal has 32% of its AAA’s still outstanding  – 87% of the initial balance of Zombie-I’s original AAA notes remains outstanding (see Citigroup Structured Finance Group research and compiled Intex reports for vintage statistics on all U.S. CLO’s issued in each of 2003, 2005, and 2007)
    • 2005 vintage: of the 92 CLO’s issued in 2005, Zombie-II ranks 90th/92 in terms of repayment of the original AAA notes’ balance – 31/92 have entirely repaid their AAA’s and the average deal out of the 92 in the 2007 vintage has only 21% of its original AAA balance still outstanding – 87% of the initial balance of Zombie-II’s original AAA notes remains outstanding
    • 2007 vintage: of the 152 CLO’s issued in 2007, Zombie-III ranks 48th/152 in terms of repayment of the original AAA (first-priority tranches) notes’ balance.  100% of the deals which have paid down less of their AAA’s than Zombie-III, however, have reinvestment periods which ended/end after Zombie-III’s March 2012 reinvestment period end date (on average March 2014) – 91% of the initial balance of Zombie-III’s original AAA notes remains outstanding
  • An illustrative CLO lifespan of a normal vs non-paying down obligation:
  • As the Zombie collateral was mostly originated as 100%+ LTV “loans” to companies coming out of bankruptcy and/or in deep distress, additional “loans” (above and beyond the ~100% LTV acquisition price) were also made to these companies to fund working capital, operating losses, and in order to service the existing debt (essentially adding more principal, of or like a pay-in-kind debt arrangement) – therefore companies’ debt tends to grow, rather than gradually amortize, like most performing loans.
  • Of the 4,118 tranches from 719 U.S. broadly-syndicated CLO’s which Moody’s rated since 1996, no tranche originally rated Single-A or better has ever had a principal loss realized at maturity
    • Out of these same tranches, Moody’s (a) states that three investment grade tranches were involved in distressed exchanges (with the highest rated of these three exchanged tranches being a Single-A) and (b) estimates that exactly four investment grade tranches may be likely to suffer principal losses in the future (with the highest of these four tranches being rated Single-A)
    • AA or AAA impairment is literally unheard of for a real CLO
  • Collateral sales from controlled portfolio companies have also been few and far between, and the consensus seems to be that only a tiny fraction of the Zombie collateral loans outstanding could be reasonably refinanced.


(ii) Zombie Collateral “Loan” Maturities (and interest rates) May Have Been Manipulated

  • The majority of all collateral “loan” maturities have been amended to hit on the exact same day, 10/31/2015 – not by this date, but on the very same day (just a few weeks before the legal final maturity for Zombie-I) – again, by any standard, this too is unheard of.
    • 92% of Zombie-I collateral (@par) matures on 10/31/2015 ($481/525mm)
    • 74% of Zombie-II collateral (@par) matures on 10/31/2015 ($796/1,080mm)
    • 67% of Zombie-III collateral (@par) matures on 10/31/2015 ($689/1,033mm)   
    • All collateral information comes from or is derived from the Zombies’ latest Intex reports
  • CLO’s typically never reach their full legal final maturity, but this decidedly looks like the likely outcome here for Zombie-I (thereby beginning the default cascade), as there do not appear to be any meaningful efforts in process to monetize the collateral.
  • Interest rate spreads on Zombie collateral loans range widely from 0% to nearly 10%; the same obligors within the same Zombie fund may well have one loan with a 0% spread and another with a very high spread, which seems decidedly off-market.  It appears that interest rates are amended or deferred as need be, based upon company non/performance and the need to meet the Zombies’ interest coverage tests, in a given period, even though the bar has been extremely low for a while, with the all-in cost of liabilities on the Zombie funds hovering around a mere ~1% for the last four years. It is hard to imagine not being able to naturally service 1% interest rate liabilities with nearly any asset today – every fund manager in the world would kill to have this now-irreplaceable, super cheap pre-crisis financing.


(iii) Traditional CLO Ratings Process Repudiated – Moody’s Ratings Withdrawn; S&P Apparently Still Rating (albeit continually lower) Zombie Notes

  • S&P appears to still be rating the Zombies, while Moody’s has withdrawn all ratings.  “Moody’s has withdrawn the ratings because it believes it has insufficient or otherwise inadequate information to support the maintenance of the ratings on a going forward basis.” (2/26/2013 press release : Moody’s withdraws the ratings of CLO notes issued by Zohar CDO 2003-1 Limited, similar for Zombies-II and III)
  • Just four days before withdrawing their ratings entirely, Moody’s further downgraded the Zombie notes – key rationale was as follows: "(1) further credit deterioration of the underlying portfolio; (2) material increase in the proportion of debt obligations whose credit quality has been assessed through Moody's Credit Estimates ("CEs") that were not updated within the last 15 months; and (3) maturity amendments and extensions of the underlying loans that result in a slower paydown and an increase in the principal repayment uncertainty of the rated notes at the deal's legal final maturity.  The ratings on the Class A-1 Notes, the Class A-2 Notes, and the Class A-3 Notes reflect the financial guarantee insurance policy issued by MBIA Insurance Corporation..." (Moody’s: 2/22/2013 Moody’s downgrades the ratings of CLO notes issued by Zohar II 2005-1, Limited, similar for Zombies-I & III)
    • Moody’s continues further, referring to the rumored repudiation of the Zombie indentures’ governing parameters, as follows: "CLO notes’ performance may also be impacted by (1) the manager’s investment strategy and behavior and (2) divergence in legal interpretation of CLO documentation by different transactional parties due to embedded ambiguities."
  • Amazingly, while Moody’s was steadily downgrading the Zombie notes over the last four years, in their first Zombie-III ratings action (S&P does seem to have changed ratings on the insured notes of Zombies-I & II in the interim, solely based on the changes to MBIA’s ratings) in three years (nothing between 4/30/2010 and 7/31/2013), S&P cites some similar issues, but somehow continues to rate the Zombie notes: "The majority of obligors in Zohar III Ltd.'s collateral pool are not rated by Standard & Poor's, but instead are credit estimated for purposes of the CLO analysis based on the companies' financial statements that the collateral manager provided... According to our records, the top 10 obligors in the transaction represent more than 50% of the total assets… There are approximately eight obligors in the underlying pool for which we do not currently have either a rating or credit estimate based on current information. For these assets from obligors for which we have limited information, we treated the loans as defaulted and given little or no recovery credit." (S&P: 7/31/2013 Zohar III Ltd. Ratings Lowered on Six Classes)
  • Hence, the Zombie funds are really not like any other CLO’s because there does not appear to be any regular, comprehensive ongoing ratings agency oversight – this key CLO parameter seems to be uniquely missing here, which is most unusual, as nearly all CLO’s are managed very carefully to satisfy constant collateral quality testing, including maintaining minimum ratings requirements.


(iv) Reported Obligor and Industry Concentrations Pose Question Marks

  • All deals are reported to be performing (passing the all-important over-collateralization and interest coverage tests, etc.) – presumably so that the 2% (of reported assets) annual collateral management fees can continue to be paid.
  • It appears, however, that among other things, different borrowers/obligors are listed for the same companies and various obligor industry designations may be used for different loans to the same company, so as not to violate borrower and industry concentration rules (this can be rather difficult to piece together, as Zombies-II & III are anonymous pools/black box deals, with the borrowers only identified by random numbers/not by name).


(v) Interests of Zombie-III Noteholders May Diverge from Those of Zombie-I & II Holders

  • Zombie-III had $154mm of cash listed in its last Intex report, which would meaningfully add to Zombie-III noteholders’ recovery, if captured before it goes to fund losses and additional collateral management fees – using consistent assumptions as before, accelerating, stopping the collateral management fee leakage, and having that $154mm cash add to recovery dollar-for-dollar could be the difference between a 50% and 80% recovery for the Zombie-III senior-most noteholders.
  • While all Zombie recoveries would decline over time on account of the 2%/year collateral management fees, Zombie-I & II holders may be more inclined to try to maximize recovery in a longer, full collateral sale/liquidation process, as they do not have any easy cash to recover – in any event, there may well be a serious fight among all Zombie noteholders, as they largely share the same collateral.


(vi) 39 Small/Middle-Market Private Portfolio Companies Listed on the Patriarch Partners Website May be Difficult to Sell/Liquidate

  • The website lists 39 companies (some of which are essentially the same company by a few different names), which we believe contain nearly all of the Zombie funds’ collateral loans.
  • Nearly all of these small/middle-market companies seem to be controlled by affiliates of the collateral manager – furthermore, many have limited franchises and continue to operate in distress; these companies  may not weather the tumult of protracted uncertainty well (along with no ongoing funding), further limiting potential recoveries.
  • The following (also from the Patriarch Partners website) is not what one would expect to read about the portfolio of loans in a CLO:  "Although comprised of individual private equity investments, our portfolio of assets is managed as a global conglomerate, with a continued focus on synergies and ventures between companies that will enhance value.  The investment funds managed by Patriarch currently hold equity positions in more than 70 companies, approximately two-thirds of which are control positions, spanning almost a dozen industries."
  • We understand that nearly 100% of the capital for these 39 portfolio companies came from the Zombie funds, aside from a handful of true first-priority, third-party working capital facilities (asset-based loans/factoring lines, which will also tend to limit Zombie recoveries – a key question is therefore, how much of the meaningful (larger) companies’ equity is owned by each of the Zombie funds (vs. other/personal funds controlled by the collateral manager)?

* * *

And with all that said, we are left with a few other questions for some of the involved parties:


(vii) Key Questions for MBIA:

  1. Is the total remaining notional insured exposure to Zombies-I & II indeed approximately $1.03 billion? (if not, how much is it? – as of 6/30/2013)
  2. What amount of reserves (in dollars) do you currently hold for future losses on Zombies-I & II?  (as of 6/30/2013 – or put otherwise, how much does MBIA expect to recover from the Zombie collateral?)
  3. What are your internal ratings for High Yield CDO #15 and High Yield CDO #23?  Are they a part of your Below Investment Grade (BIG) category? (if so, why were they not listed on the last BIG report?)
  4. What is your plan for your Zombie exposure?
  5. What reaction did you have (if any) to the Zombie notes’ ratings downgrades over the last few years?
  6. Will MBIA Inc. (parent) support MBIA Corp. (structured finance subsidiary) if it is deemed to be undercapitalized/needs to raise capital?
  7. Which, if any, of the controlled companies in the Zombie portfolios have been sold in the last two years (or what meaningful debt has been refinanced)?  Given the strong credit markets, why is no new CLO deal in the works/why hasn’t there been any meaningful refinancing activity?
  8. How much does MBIA know about the Zombies’ largest obligors?
  9. What is the weighted average debt-to-trailing EBITDA (audited historical EBITDA) multiple for the collateral investments in each of the Zombie funds?
  10. What is the most recent financial information you have (balance sheet and income statement) for the companies in which the Zombie funds are invested?  If this information is not current, why not? (also what is the most recent audited financial information you have?)
  11. Do you know who the CEO and CFO are for each underlying company? (and for how long they have been in these roles?)
  12. Why has there never been any discussion of/disclosure about these troubled Zombie deals?
  13. Do you dispute any of the information contained herein?
  14. Do you have any other reaction to this report?


(viii) Key Questions for Moody’s and S&P:

  1. When is the last time you received the required comprehensive reporting information for the Zombie funds? 
  2. What is the most recent financial information you have (balance sheet and income statement) for the companies in which the Zombie funds are invested?  If this information is not current, why not? (also what is the most recent audited financial information you have?)
  3. How much do you know about the Zombies’ largest obligors?
  4. Do you know who the CEO and CFO are for each underlying company? (and for how long they have been in these roles?)
  5. Are you still rating the Zombie funds?  (if no, why not?)
  6. Are you still getting paid to rate the Zombie funds? (if no, through what date were the regular ratings payments current?)
  7. To what extent have you had similar ratings challenges/issues with other CLO’s?  How typical or unusual would you say the ratings process has been with the Zombie deals?
  8. To what extent will the information herein have an impact on your MBIA ratings?
  9. Do you dispute any of the information contained herein?

We look forward to hearing the responses.