We believe that ResCap Liquidating Trust (RLT) presents a highly attractive opportunity with an expected 3.2x MoM in the base case, generating a 80.2% IRR over the assumed 2 year wind-down of the trust. In a market that looks expensive, the ability to invest in an uncorrelated litigation play with an attractive IRR is especially appealing. The sources of value of the trust are non-litigation and litigation assets. While the value of the non-litigation assets is relatively certain, the high probability of successful litigation outcomes, in addition to the substantial potential recoveries, are at the core of why we believe RLT to be among the most attractive ideas we have come across. Unlike typical situations involving litigation, the contractual language and losses underlying RLT’s claims are unambiguous, a view which is supported by legal precedent in a substantially identical case involving ResCap in the Eighth Circuit Court of Appeals. Applying the most pessimistic assumptions within reason, we still expect to achieve a 1.8x MoM over 3 years, representing a 21.0% IRR. Forced selling by creditors who lack the mandate and/or patience to own units, combined with vague litigation disclosure provided by RLT, has created a significant dislocation. As non-litigation assets are monetized and lawsuits are settled, the units should significantly appreciate in value.

 

 

 

Prior to filing for Chapter 11, ResCap’s primary business model focused on securitizing and servicing residential mortgage loans. Securitization activities were primarily conducted by Residential Funding Company, LLC (RFC), a wholly owned subsidiary of GMAC-RFC Holding Company, LLC (see appendix for the organization chart). During the ordinary course of business, ResCap purchased mortgages from correspondent banks who typically directly originated these loans. ResCap pooled these mortgages together into a trust and issued various tranches of RMBS to investors. A portion of these securities, typically those with the highest rating (1-A tranches), were insured by monoline bond insurers for losses incurred. The below diagram illustrates ResCap’s role in securitizing residential mortgages.

 

 

 

In 2007, with the burst of the housing bubble and resulting poor performance of RMBS, various RMBS trusts and monolines filed lawsuits against ResCap, alleging breaches of contractual representations and warranties (R&W). With mounting R&W liabilities and an imminent bond maturity payment, ResCap filed for Chapter 11 on May 14, 2012 in the Southern District of New York. In the Chapter 11 process, the Global Settlement (docket 6066) that was reached detailed the allowed unsecured claims against ResCap. The settlement allowed $13.3B in total claims from the monolines, RMBS trusts, Senior Unsecured Notes, and private securities. The consequences of the Global Settlement to the monolines and RMBS trusts are shown below.

 

 

Pursuant to the Joint Chapter 11 Plan (docket 4153), the ResCap Liquidating Trust (RLT) was established to hold substantially all of ResCap’s assets after the conclusion of the Chapter 11 process. RLT would provide recoveries to unsecured creditors (now unitholders in RLT) by monetizing non-cash assets and making distributions in a timely manner. Based on recent disclosures, Paulson & Co. has acquired MBIA’s 33.5% stake in RLT (in addition to his original stake). We believe that Paulson’s ownership and active involvement via his board seat will help maximize RLT’s proceeds from litigation.

The liquidating trust broadly has two sources of value: non-litigation assets and litigation assets. The non-litigation assets are primarily comprised of government-insured mortgages and mortgage servicing rights. The litigation assets are comprised of expected settlement proceeds from lawsuits that RLT has filed against correspondent banks, asserting claims for breaches of contractual R&W and indemnification of losses stemming from the purchase of defective mortgages. In this write-up, we attempt to value RLT’s non-litigation and litigation assets.

 

 

Non-Litigation Assets

In conjunction with ResCap’s Chapter 11 proceedings, the debtors filed a Disclosure Statement (docket 4819) which provided a recovery analysis for creditors. The recovery analysis is based on an orderly wind-down of the assets in the estates using ResCap’s April 30, 2013 balance sheet. The book value and fair market value of these assets are disclosed, as well as the expected claims and administrative expenses associated with the estates. After using gross assets to pay expenses, satisfy claims, and fund other trusts, we arrive at the net non-litigation asset value available to RLT. The non-litigation asset pool is primarily comprised of cash, FHA / VA insured mortgages, mortgage servicing rights, and other assets. Since the FHA / VA mortgages (69% of total BV) are insured, MSR assets (15% of total BV) are valued by determinable fees, and restricted cash is 5% of BV, there is little uncertainty surrounding the valuation of 89% of total BV of this asset pool. In our base case, we’ve assumed that the Disclosure Statement accurately marks the fair value of the assets. Our bear and bull cases provide sensitivities around these valuations. We refer you to the appendix for our detailed set of assumptions. We believe the non-litigation assets are worth $7.65 per RLT unit.

 

 

Litigation Assets

ResCap’s Losses from R&W and Contractual Breaches

In effect, ResCap was a securitization “pipeline” that purchased mortgages underwritten by third party correspondent banks and sold them to a variety of ResCap sponsored, and non-ResCap sponsored trusts, in addition to various whole loan purchasers. As part of the securitization and sale processes, ResCap regularly, via its wholly-owned subsidiary Residential Funding Company, LLC (RFC), made certain Representations and Warranties (R&W) about the quality of the mortgages in the trusts. The numerous defects and compliance issues associated with these mortgages led RFC’s counterparties to assert that there had been material breaches of the contractual R&W, among other claims. These counterparties sued RFC to enforce their rights under the relevant contractual agreements.

There were two key constituencies that brought causes of action against RFC. The largest of these was the RMBS trust group, which comprised over 1,000 individual trusts. The trust group claimed that RFC had materially breached the loan-level R&W provided in the Master Loan Purchase Agreements (MLPA) and Pooling and Servicing Agreements (PSA). The trust group asserted that per these contracts, RFC was required to repurchase the defective mortgages at par.

As part of the mediation process in the bankruptcy, expert Frank Sillman was instructed to determine an appropriate claim amount for the RMBS trust group. According to the Sillman Examination (docket 5703), the RMBS trusts had estimated lifetime losses of $42.4-43.2B. Using statistical sampling and a thorough re-underwriting process, Sillman determined that the 2004-2007 trusts had a defect rate (in other words, loan-level material R&W breach rate) of 43.5%, and the pre-2004 trusts had a defect rate of 35.1%. Multiplying the estimated lifetime losses by the relevant defect rates, Sillman determined that the trusts would experience $18.1-18.5B of losses on defective loans. Based on ResCap’s historical repurchase experience, industry repurchase data, prior R&W settlements, and expert opinions on legal defenses, Sillman determined that a settlement factor of 41%-47% was appropriate. Bringing this all together, Sillman recommended an allowed claim for the RMBS trust group of $7.4-8.7B.

The second major constituency that sued RFC was the monoline insurers, comprised of MBIA, FGIC, Ambac, and Syncora. The monolines asserted that RFC had materially breached both the loan-level R&W provided in the MLPA and PSA, and the transaction-level R&W provided in the Insurance and Indemnity Agreements (I&I). In addition, the monolines asserted claims for material breach of contract, fraudulent inducement, and indemnification. Taken together, these claims required that RFC compensate the monolines for the entire lifetime insurance losses, legal expenses, and costs associated with the trusts they had insured, in addition to pre-judgment interest. In other words, there was no issue of loan-level breach rate here - upon proving material breach of contract or fraudulent inducement, the monolines could recover on all losses.

According to the Sillman Examination, the monolines had estimated lifetime insurance losses of $4.2-4.4B. After evaluating recent settlements of monoline claims, Sillman noted that:

“Settlement of monoline claims at a small discount, if any, to estimated bond losses, is within the range of recent settlements. In addition, the recent settlement after trial in Assured v. Flagstar (June 2013), reflected a roughly 15-20% premium over the total shortfall to the wrapped bonds” (Sillman Direct ¶ 73).

Since Sillman had not reviewed the legal basis for the claims for interest and indemnification of expenses, he determined that a settlement factor of 80%-100% was appropriate. Ultimately, Sillman recommended a total allowed claim for the monolines of $3.4-4.4B.

In the Global Settlement that was the foundation of the Chapter 11 Plan, the RMBS trust group received an allowed claim of $7.3B (17.1% of average expected losses) and the monolines received an allowed claim of $4.9B (116% of average expected insured losses). This Global Settlement crystallized the losses that ResCap incurred as a result of securitizing defective mortgages that breached R&W and contractual agreements with counterparties. The $12.2B of allowed RMBS trust and monoline claims represent the losses that serve as the underlying basis for RLT’s lawsuits against its former correspondent banks.

 

 

RLT’s Cases Against its Former Correspondent Banks

While most litigation possesses numerous uncertainties, RLT’s claims against its correspondent banks are robust as a result of the legal precedent set in substantially similar R&W cases, strong underlying contractual documents, and analysis set forth by Sillman in the bankruptcy process. Over the next 1-2 years, we believe that the vast majority of RLT’s lawsuits will be resolved via settlement, generating over $4.0B in recoveries for unitholders. We simply cannot envision a scenario in which RLT is not extremely successful in its litigation against correspondent banks. The high probability of successful litigation outcomes, in addition to the substantial potential recoveries, are at the core of why we believe RLT to be among the most attractive ideas we have come across.

At a high level, the correspondent banks made certain R&W to ResCap upon which ResCap, in turn, made R&W to the RMBS trusts and monolines. As discussed above, ResCap’s R&W to its customers resulted in numerous lawsuits and ultimately led to an allowed claim of $12.2B in the bankruptcy proceedings. RLT, as the successor to ResCap, is fully entitled to enforce its contractual rights by holding correspondent banks accountable for the R&W which they made to ResCap. In other words, RLT will seek to recover the $12.2B allowed claim, in addition to legal fees and expenses, from the correspondent banks.

Following the crystallization of the $12.2B allowed claim in the global settlement, between 12/12/13 and 1/15/14, RLT filed 71 substantially identical complaints primarily in the U.S. District Court of the District of Minnesota. Cumulatively, these complaints represent $31.5B of original principal balance in mortgages purchased by ResCap from correspondent banks. As we will discuss later, while ResCap certainly purchased mortgages from more than just the 71 defendants in its recent lawsuits, it is likely that these other correspondent banks are no longer going-concern entities. As such, our analysis uses the $31.5B of principal as the basis for computing legal recoveries.

Note: There are some cases that we don’t have access to: 1. Cases that have been served, but not filed; 2. Cases in Hennepin County District Court. Between these two buckets, there are 11 additional lawsuits that are not included in my analysis. One of these is against Countrywide (Residential Funding Company, LLC v. Countrywide Home Loans, Inc. No. 27-cv-14-2783, Hennepin Cty. Dist. Ct.). This has the potential to be among the largest of the cases. We expect additional lawsuits to be filed going forward and will provide updates accordingly.

 

The diagram below represents the contractual documents that served as the legal bedrock for the former ResCap value chain. As part of its securitizations, ResCap signed MLPA and PSA contracts with the RMBS trusts and monolines, along with I&I agreements with the monolines. In a similar fashion, the correspondent banks signed a Seller Contract with ResCap which incorporated the Client Guide.

 

 

The Client Guide is at the core of the RLT investment thesis. The document contains broad R&W made by correspondent banks to ResCap, a robust indemnification provision, and ResCap’s exclusive right to determine whether a loan is defective and has caused an “Event of Default”. In effect, the Client Guide is an even stronger contract than the I&I agreements ResCap signed with the monolines (let alone the MLPAs or PSAs), through which the monolines have historically recovered 80-120% of losses. During the boom years of the housing bubble, the Client Guide served ResCap and its correspondent banks well; the underwriters could sell as many mortgages to ResCap as they wished, and ResCap would simply turn around and securitize them. The exclusive right of ResCap to determine an event of default was meant to ensure that ResCap could occasionally reject a loan if it was defective. Clearly, the correspondent banks did not contemplate a situation in which ResCap would be in the position to demand repurchase of hundreds of poorly-performing mortgages. Now that ResCap’s relationships with correspondent banks are irrelevant, RLT can seek maximum recovery of losses via the indemnity provision and its sole authority to deem a loan as defective.

Pursuant to the Client Guide, the correspondent banks made a number of organizational and loan-level R&W to ResCap. An extensive list of the important R&W is presented in the appendix. The Client Guide makes unequivocally clear that a breach of any of the R&W constitutes an “Event of Default” under the contract (A208(3), see appendix). Moreover, any false representation made in the underwriting documentation is an “Event of Default,” whether or not the Client was aware of the misrepresentation (A208(4), see appendix). Most significantly, however, ResCap retains the sole and exclusive right to determine whether an “Event of Default” has occurred. Per section A210(A), “If GMAC-RFC determines that an Event of Default has occurred with respect to a specific Loan, the Client agrees to repurchase the Loan and its servicing.” The language here is unambiguous; ResCap, at its sole discretion, may determine whether there has been an “Event of Default” under the contract.

In general, the Client Guide provides that “Whenever any provision of this Client Guide contract requires GMAC-RFC to make a determination of fact or a decision to act, or to permit, approve or deny another party’s action, such determination or decision shall be made in GMAC-RFC’s sole and absolute discretion” (113(B)). While the Client Guide does provide the correspondent bank with the right to appeal ResCap’s determination of an “Event of Default”, ResCap “will in its sole discretion determine the validity of any appeal filed by the Client. If GMAC-RFC’s decision remains firm following an appeal, the Client shall repurchase the Loan and its servicing” (A201(H)). The Client Guide provides ResCap with a contracted-for exclusive right to determine whether an “Event of Default” has occurred, and this determination is not subject to judicial review. There is a plethora of judicial precedent affirming that courts shall not rewrite an unambiguous contract, which is exactly the case with the clear language of the ResCap Client Guide.

The Client Guide grants ResCap with a broad set of non-exclusive, cumulative remedies in case of an “Event of Default.” Aside from the basic requirement that ResCap send a letter notifying the correspondent bank of an “Event of Default,” there is no condition precedent to ResCap’s use of the contractual remedies. ResCap may “exercise any remedy outlined in this Client Guide or as allowed by law or in equity” to deal with a default, and is not limited to the repurchase protocol (A209(A)). With respect to the repurchase protocol, ResCap need not act within a defined time frame; per A210(A), “GMAC-RFC is not required to demand repurchase within any particular period of time, and may elect not to require immediate repurchase. However, any delay in making this demand does not constitute a waiver by GMAC-RFC of any of its rights or remedies.” In addition, ResCap can unilaterally determine that if “repurchase of a Loan and/ or the servicing is not appropriate…the Client shall pay GMAC-RFC all losses, costs and expenses incurred by GMAC-RFC…as a result of an Event of Default. This includes all reasonable attorneys’ fees and other costs and expenses incurred in connection with enforcement efforts under taken” (A210(A)).

Beyond the remedies discussed above, the Client Guide also provides ResCap with a broad indemnification provision that insulates it from any losses stemming from an “Event of Default.” Please refer to the appendix for the actual indemnification language. Specifically, ResCap is indemnified for claims asserted against it based on a breach of R&W made in reliance on the R&W given by the correspondent banks. The various costs that ResCap incurs in accordance with enforcing the contract are also to be reimbursed by the counterparty, including attorney’s fees.

Our reading of the contractual language is unequivocally supported by case precedent in a substantially identical cause of action. In Residential Funding Company, LLC v. Terrace Mortgage Company (09-cv-03455, U.S. District Court of the District of Minnesota), RFC sued Terrace on two counts: (1) one claim for breach of contract, and (2) one claim for indemnity for RFC’s legal fees and costs. RFC alleged that a number Terrace’s loans were defective and breached certain R&W, thereby constituting an “Event of Default” under the Client Guide. To enforce its contractual right to demand a repurchase of the loans and be reimbursed for expenses, RFC brought a lawsuit against Terrace. The District Court granted RFC’s motion for summary judgment, finding that “the plain meaning of the agreement allows for RFC’s right to demand that Terrace repurchase loans that RFC, in its sole determination, has concluded are non-conforming” (document 43, page 12). Terrace complained, without basis, that “allowing RFC to recover simply because it demanded repurchase would effectively allow RFC to force Terrace to take back any and every loan that Terrace ever sold to RFC, at RFC’s whim. RFC could require a repurchase of any loan, whether there was a default or not, regardless of its motives” (document 43, page 11). The court was not persuaded by this argument. The court noted that “Insofar as [Terrace’s] objection to the repurchase provisions is that there is no mechanism for judicial review of RFC’s determination that Terrace must repurchase a loan, a party to a contract is free to expressly provide for a forum for judicial review or to relinquish any right of action in the courts” (document 43, page 12). In other words, Terrace had contracted away its right to challenge ResCap’s determination of whether a loan is defective.

Unsatisfied with the outcome, Terrace appealed the District Court’s decision to the Eighth Circuit Court of Appeals. In its opinion, the appellate court affirmed the lower court’s ruling, determining that:

“If contract language is unambiguous, ‘courts should not rewrite, modify, or limit its effect by a strained construction’ (Travertine Corp. v. Lexington-Silverwood). Terrace asks us to do just that. The Client Guide gives Residential ‘sole discretion’ to determine whether an Event of Default has occurred, and Terrace agreed to buy back the loan if Residential determined as much. There is nothing ambiguous about this language. Minnesota law does not permit us to entertain Terrace’s strained reading of the contract” (document 75, page 6).

Like the lower court, the appellate court found that Terrace could not grant RFC the exclusive right to determine if an “Event of Default” has occurred, only to later ask for judicial review. Terrace again utilized a variety of unpersuasive arguments: 1. If RFC can demand repurchase “at its whim”, why include the several hundred other pages in the Client Guide?; 2. If RFC can demand repurchase “at its whim,” why did RFC act during litigation as if it needed to prove violations of the R&W in the Client Guide? The appeals court flatly rejected these arguments, noting that “Terrace identifies no ambiguity in the language of the contract which would permit us to look beyond its plain language” (document 75, page 7). While the contract clearly appears to favor RFC at present, the court noted that the agreement served the parties well for a number of years. Ultimately, the appellate court highlighted that Terrace willingly put itself in this situation; the court therefore “decline[d] to deem a voluntary, negotiated contract unconscionable merely because the parties’ relationship has soured” (document 75, page 8).

The decision of the appellate court confirms the prior discussion of the contractual language in the Client Guide. Given that all of RLT’s complaints are substantially similar and incorporate the Client Guide, the Eighth Circuit decision in RFC v. Terrace is broadly applicable to all of the lawsuits. We think the unambiguous language and precedent will make RLT’s lawsuits relatively clear-cut, and moreover, RLT will likely be granted summary judgment as there are no issues that warrant a trial. In the motions to dismiss, the defendants in these actions have complained that RLT has not specified which loans are defective, why they are defective, nor what losses resulted from these defects. These defendants are simply repeating the arguments dismissed by the District Court and the Eighth Circuit in RFC v. Terrace via summary judgment.

 

Litigation Process

In all of the Minnesota cases, RLT has submitted motions to transfer venue to the Southern District of New York. Under 28 U.S.C. §§ 1404 and 1412, it is appropriate to transfer venue “in the interest of justice or for the convenience of the parties.” Given that RLT is seeking to recover losses that were adjudicated in the ResCap bankruptcy, it would be far more efficient to administer these cases in the SDNY Bankruptcy Court under Judge Martin Glenn. As stated in the motion to transfer venue, Judge Glenn has “particular knowledge of the facts and circumstances underlying [RLT’s] claims that would promote the efficient resolution of those claims” (13-cv-03498, document 20, page 7). Given that the cases are substantially similar, a consolidation under Judge Glenn would also preclude the possibility of inconsistent rulings. The transfer of the lawsuits to Judge Glenn will serve to expedite settlements of these cases, thereby boosting the IRR of owning RESCU units. We do not believe that these cases will necessitate a trial given the precedent in RFC v. Terrace and Judge Glenn’s familiarity with the underlying losses that ResCap sustained. Ultimately, we believe that the vast majority of these cases will be settled in a 1-2 year time frame.

 

Litigation Recovery Analysis

As we discussed above, the $12.2B of allowed RMBS trust and monoline claims against ResCap serves as the basis for RLT’s lawsuits to be made-whole for the losses stemming from R&W breaches by the correspondent banks. While this might seem an appropriate starting point for assessing RLT’s likely recoveries, we do not know what proportion of correspondent banks are no longer going concerns. As such, our analysis uses the $31.5B of original principal balance referenced in the complaints against (still existing) correspondent banks as the basis for computing litigation recoveries. Please reference the appendix for the original principal balance for each correspondent bank that ResCap is suing.

In order to determine the estimated lifetime loss ratio for the pool of $31.5B of loans, we use the trust-level loss model constructed by Sillman. In the Sillman Examination, Annex A, current outstanding principal balances and expected lifetime losses are presented for 431 ResCap sponsored, non-monoline RMBS trusts. Using the ratio of original principal to current principal given in the defect rate analysis in Annex C, we can back into the original principal balance for the 431 ResCap trusts. Based upon a $58.6B estimated original pool balance, the average total loss ratio is expected to be 63.5%.

 

 

Applying the estimated total loss ratio from the Sillman data to our base and bull cases, we come up with ~$20.0B of losses from correspondent banks that RLT is presently suing. In our bear case, we use a conservative 43.5% loss ratio; we note that several of the monoline insurers experienced RMBS loss ratios as high as 30%+, even though they traditionally insured senior tranches. As such, we feel that a loss ratio of 43.5% represents a sufficiently adverse downside scenario.

Based on the allowed claims table presented earlier, we see that ResCap clients (both RMBS trusts and monolines) are expected to suffer total lifetime losses in the range of $46.6-47.5B. As part of the Global Settlement, these creditors received an allowed claim of $12.2B against ResCap. In other words, for every dollar of expected lifetime losses, ResCap’s creditors received 26 cents of allowed claim. Given that the Client Guide only permits ResCap to recover “all losses…incurred by [ResCap]…as a result of an Event of Default”, RLT cannot seek recoveries of more than 26% of the direct losses from correspondent bank loans (A210(A)). In effect, RLT’s recovery is capped by the 26% recovery on expected lifetime losses that ResCap’s clients received in the bankruptcy. Applying this 26% rate to the losses from defendant loans yields lawsuit claims by RLT of $5.2B in the base and bull cases, and $3.6B in the bear case.

The final adjustment we make is to multiply the lawsuit claims by a settlement factor. As was described in detail earlier, we believe the robust contractual language in the Client Guide and judicial precedent in RFC v. Terrace make RLT’s cases even stronger than a standard monoline suit. Based on recent monoline settlements, as highlighted in the Sillman Examination, a reasonable settlement factor is between 80% and 120% of losses. While we feel strongly that RLT’s claims are superior to monoline suits predicated on the language in an I&I agreement, we have made the conservative assumption that RLT will settle for 60% of losses in the bear case. In the base case, we use a settlement factor of 80%; while this is at the bottom of the recent range of monoline settlements, we believe that RLT could potentially take a discount to facilitate an expedient resolution of the litigation. Finally, in our bull case, we use the midpoint of the 80%-120% range and apply a settlement factor of 100%. Based on our interpretation that settlements seek to restore RLT to the pre-breach position, we do not foresee taxation on litigation proceeds. We refer you to the appendix for further details.

We believe that RLT will recover ~$4.2B or $42.69 per unit from litigation in the base case, an amount over 2.75x the current unit price. In our bear case, we still see a robust recovery of ~$2.1B or $21.81 per unit, despite using very conservative assumptions for the lifetime loss rate and settlement rate. In our bull case, we see RLT leveraging the strong contractual language in the Client Guide to secure a recovery of ~$5.2B or $53.68 per unit.

 

 

RESCU Valuation

Adding together the value per share of the non-litigation and litigation assets gives us our base case price of $50.34 per unit, representing a 3.2x MoM and an 80.2% IRR. This base case assumes the liquidation trust will wind-down in two years. Our bear case, which applies very conservative assumptions, still yields a 21% IRR assuming a 3 year wind-down. Note that we foresee little taxation on the wind-down of the trust per treatment of grantor trusts and associated distributions. Please see the appendix for details.

 

 

Considerations

• Delay in monetization of non-litigation assets
• Longer than expected resolution of lawsuits
• Inability to pay settlements by certain correspondent banks
• Execution of position entry and exit

 

Partings Thoughts

We believe that ResCap Liquidating Trust (RLT) offers compelling returns, even in a downside case that we view as relatively unlikely to occur. Forced selling by creditors who lack the mandate and/or patience to own units, combined with vague litigation disclosure provided by RLT, has created a significant dislocation. As non-litigation assets are monetized and lawsuits are settled, the units should significantly appreciate in value. We hope this write-up has been helpful in describing and quantifying the key value drivers.

 

Additional Info

We have created a Dropbox folder which contains the important bankruptcy, trust, and litigation documents. While we encourage you to explore the contents provided here, please refrain from re-distributing the link below and/or the contents of the folder.

bit.ly/1kD8RyR

 

• Disclosure of financial statements by 5/10/2014
• Announcements of settlements and/or judgments