With issuers rated 'B' and lower making up an increasingly significant portion of the speculative-grade landscape, investors have expressed concerns about the prospects of borrowers falling into distress and eventually into bankruptcy. These investors, particularly collateralized loan obligation (CLO) managers, wonder what issues they could face when that happens.

U.S.-based marketing firm Acosta Inc.'s recent bankruptcy filing provides a valuable glimpse into a potentially concerning future for CLOs that hold on to their positions through bankruptcy. In Acosta's case, as many investors were unwinding their holdings to adhere to their investment strategies, mandates, and/or indentures, distressed funds were taking advantage of falling prices for the company's debt in the secondary market and stepped up their own purchases. We assume the distressed funds' strategy was to build up their positions to where they would have majority control of outstanding principal value of the two debt classes likely to vote in a bankruptcy restructuring (first-lien claims, Class 3: $2.14 billion; senior notes claims, Class 4: $840 million), and thus gain negotiating leverage.

The group of distressed investors used that position of leverage to control the pre-bankruptcy restructuring negotiations with the borrower, largely excluding the remaining investors. In the restructuring support agreement that initially emerged from those negotiations, the distressed funds disproportionately (that is, beyond their pro rata share) amassed both recovery potential and new money fees through the plan's preferred equity raise [1], debtor-in-possession (DIP) facilities, early tender consideration, rights offering oversubscription, preemptive rights, and general equity for debt swap [2].

Beyond that, and perhaps by design, the distressed group initially structured a deal that prevented the first-lien minority group (which included a significant number of CLO claims) from being able to participate in certain portions of the deal with higher returns. For example, the preferred equity direct investment raise and DIP funding would have provided significant ownership potential and fees, respectively.

The minority lenders had some leverage of their own, however. Under U.S. bankruptcy law, class approval of a plan of reorganization requires more than just a controlling interest as measured by the face value of each creditor's claims, but also a simple majority of the number of claims that vote. This concept is called numerosity. However, before we discuss numerosity, let us take a look at the negotiation timeline of Acosta's restructuring (Chart 1).

Chart 1

As the latter part of the timeline illustrates, bankruptcy class voting can get complicated and requires more than a majority or even a supermajority of creditors in terms of principal amount held.

Numerosity

Numerosity is a commonly used term in bankruptcy and other areas of law [3]. For class voting in bankruptcy proceedings, it refers to the portion of Subsection 1126(c) of the U.S. Bankruptcy Code that says for a plan to be accepted by a class, it must be "accepted by creditors ... that hold at least two-thirds in amount and more than one-half in number of the allowed claims of such class held by creditors ... that have accepted or rejected a plan." [4] [5]

This second part around the number of claims is the actual numerosity requirement. Further, the subsection is conjunctive, meaning both the amount and number requirements--in regard to those claimants that vote--must be met for the plan to be deemed accepted by the class. Therefore, voting members of a class holding either: 1) more than one-third in principal amount, or 2) more than half of the allowed claims can prevent the plan from being accepted by the class. [6]

The concept and requirements of numerosity get tested when one introduces closely affiliated creditors with multiple claims related to the same underlying fungible debt transaction (e.g., closely affiliated CLOs, all part of a single asset manager and each with its own claim against a debtor related to the same first-lien loan). This is common with broadly syndicated loans. In fact, the credit agreements for these loans usually include elaborate and standardized assignment [7] provisions to reduce friction and ease the process by which underwriting banks can sell a portion of their part of the loan on to investors (e.g., multiple CLOs under the same asset manager, each its own legal entity), who upon the execution of the assignment agreement, become lenders to the debtor with a right to payment from the debtor, which is facilitated by the agent bank.

This is what asset managers with multiple affiliated funds face in bankruptcy. Although it has not been widely tested in the courts, current case law favors respecting the separate nature of affiliated claimants. Courts have, so far, held that closely affiliated creditors have the right to vote multiple legitimate claims even if those claims originate from the same underlying debt transaction (such as a bank loan) because the wording in Subsection 1126(c) refers to "claims" and not "creditors" with respect to numerosity [8]. One important factor courts have looked to when making a determination on the legitimacy of these claims is whether each affiliate creditor has filed its own, separate, proof of claim. (A proof of claim is a document filed with the court by a creditor in the early stages of bankruptcy evidencing the creditor's right to payment from the debtor.) Likewise, courts have held that a single creditor can hold multiple claims arising from independent transactions (e.g., trade claims or general unsecured claims, such as a judgement). Moreover, they have said these claims remain independent when purchased by a single creditor. However, courts have not said the same about single creditors in regard to fungible [9] debt. Therefore, a single creditor buying multiple rights to payment related to fungible debt prior to bankruptcy (or the bar date) in a bankruptcy proceeding could fall short of the independent transaction as well as the separateness allowances and could likely see those rights to payment collapse into a single claim.

However, the waters get murkier still in either of two scenarios with even less case law to provide guidance: 1) creditors that appear to attempt to manufacture separateness of claims by divvying up its fungible debt right to payment among existing or newly-created affiliated entities, and 2) creditors that appear to attempt to manufacture multiple claims by having affiliated entities purchase debt in the secondary market from one another. In the first scenario, courts have not ruled directly on the issue, but have mentioned the scenario in more than one opinion as a contrast to what is permitted. The second scenario also has limited case law and is better addressed on a case-by-case basis [10] [11].

Acosta: The Deal And The New Deal

In Acosta's case, at the time of filing, there were four distressed funds holding up to 70% of the outstanding principal amount of the first-lien loans while, at the end of 2019, there were 18 asset management companies controlling upwards of 60 CLOs with a right to payment from Acosta. The CLOs, as they say, had the votes. The votes to force a renegotiation: After the minority first-lien lenders [12], which included those claims held by CLOs, were brought into the process and it was shown that their claims, collectively, would be more than half of the claims to vote in the first-lien class (Class 3), a renegotiation began. Given their ability to block the affirmative vote in one of the impaired classes, the minority first-lien group was able to extract, approximately, an additional $70 million in potential value for minority first-lien lenders (Chart 2).

Chart 2

Despite the various concessions listed above, the distressed lenders, because of the size of their holdings, were still able to heavily influence the process and receive a disproportionate amount of the return-generating portion of the deal. In terms of the common equity for debt swap, parties received their pro rata share. However, the distressed lenders retained exclusive rights to fund a $65 million direct investment convertible preferred equity raise. Beyond that, the distressed group served as backstop lenders on the preferred equity (also convertible) rights raise. CLO members of the minority group, by law, could not participate. Because of this, on a converted basis, the group of distressed funds will end up owning between approximately 74% and 82% of the company that emerged from bankruptcy. (These figures take into account the distressed funds' notes ownership.)

Yet, the minority group forced significant changes. It is likely this bankruptcy did not turn out how either of the major secured creditor groups envisioned. For CLO managers, it appears they were largely left out of the initial negotiations and fought to be included in a way that better reflected their holdings and power in bankruptcy.

The distressed funds took a haircut in renegotiations. However, because we do not know the exact price at which they bought their holdings or the true value of their equity interests, the economics of this transaction as it exited bankruptcy are unclear. However, they now own a significant portion of Acosta, which said in a Jan. 2, 2020, news release that it has the "strongest balance sheet in the industry."[13] They were able to do this in part by redirecting some of the recoveries (approximately 11% on the $840 million notes outstanding) to the unsecured noteholders' claims (distressed funds held up to 80% in principal amount)--an unusual approach, especially considering how low the recoveries were for first-lien lenders. Nevertheless, it is reasonable to deduce that members of the distressed funds group approached this transaction as a loan-to-own play. So, in that regard, it was a success for them.

Bankruptcy Reform

In the last six years, the American Bankruptcy Institute (ABI) has proposed changing bankruptcy laws to remove the ability of closely affiliated creditors to vote multiple claims. In a 2014 report [14], ABI recommended "one creditor, one vote." It argues closely affiliated creditors under common investment management may gain undue influence over whether a plan is accepted by a class, which can then be confirmed by a court even if other impaired classes reject the plan [15].

Others reason that while the numerosity principle for claims was meant to protect minority interests, it may have given minority holders in certain situations the power to dominate. Nevertheless, the dynamics are complex, because unlike a vote of shares in equity with each share being equal, the principal value of any one claim within a class can vary and may not be one-to-one. One can imagine the simplified scenario with four unevenly distributed claims that vote in a hypothetical "secured lending" class: a) 83% principal amount claim, b) 4% principal amount claim, c) 2% principal amount claim, and d) 6% principal amount claim. Any combination of claims b-d could potentially block the class's affirmative vote despite claim a being in favor.

Endnotes

[1] CLOs, by law, do not get credit for holding equity, and cannot participate in equity raises.

[2] Moreover, the rights offering was backstopped by the distressed group, which would earn the group of distressed funds substantial additional fees for the service.

[3] It can also refer to the threshold number of creditors needed to initiate an involuntary bankruptcy and the requisite number of plaintiffs to justify certification of a class in a class action proceeding.

[4] "Plan" in the subsection refers to a plan of reorganization.

[5] Approval of plans by a class are binding on all class members, including those that did not vote or voted to reject it.

[6] A plan can nevertheless be deemed accepted by an impaired class that originally rejected it via cram-down by the court.

[7] As opposed to participation, where one party transfers only a limited bundle of rights to another party and does not necessarily create privity of contract between transferee and borrower, assignment is the sale or transfer of a loan from an assignor to assignee, creating a credit relationship (lender-borrower) between the assignee and borrower.

[8] The Bankruptcy Code's current definition of "creditor" is an entity that has a claim against the debtor. So, in regard to fungible debt held by separate but closely affiliated entities, it may be arguable these claims do not originate from the same creditor.

[9] Fungibility has its own legal complexities, but in this context and for the purposes of this article, it refers to a single tranche of a company's debt originating from a single transaction, such as a single issuance of a bank loan.

[10] These transactions likely would need to be made at arm's length, in good faith, with some requisite number of days before the bar date.

[11] This analysis is covered in "The Pervasive Problem of Numerosity", Portfolio Media Inc.

[12] By principal amount.

[13] According the disclosure statement, the company would emerge with approximately $7 million in debt.

[14] "Commission to Study the Reform of Chapter 11", American Bankruptcy Institute.

[15] The court can confirm the reorganization plan over an impaired class's objection via cram-down.