Copyright (c) 2026 the National Conference of Bankruptcy Judges
The Backstop Party
by
Bankruptcy exit financing has become a site of recurring conflict in the intra-class creditor skirmishes that now so often mark corporate reorganization. When creditors in an ad hoc group sufficiently large to accept a plan of reorganization on behalf of an impaired class agree to underwrite, or “backstop,” the debtor’s plan-contemplated capital raise, creditors who hold identical claims but are excluded from the backstop group cry foul that the Bankruptcy Code bars their unequal treatment. All-or-nothing arguments have proved unsatisfying, however, because, on the best reading of the Code, the merits depend on a matter of fact that is not readily ascertainable: is the debtor offering backstop parties a competitive risk-adjusted return for the risk they will bear, or something more?
This article reports the results of a study of the returns to equity rights offering backstop commitments in 49 bankruptcies since 2016. We find evidence that backstop parties face little risk and garner strong positive returns on their capital commitments. For example, the value of the securities sold in an offering exceeded the exercise price in 18 of the 19 deals for which an arm’s-length estimate of value is available—on average by a factor of 2. We estimate that, on average, backstop creditors realized 16 cents per dollar of eligible claim more than did similarly situated creditors who participated optimally in the offering but were not part of the ad hoc group. Although the small sample size warrants caution in interpretation, our results indicate that backstop parties in the typical case are compensated for more than their willingness to risk capital.
I. Introduction
Bankrupt companies often propose to raise new money from their existing creditors as part of a plan to emerge from Chapter 11. Most plans of reorganization oblige the debtor to deploy substantial amounts of cash as soon as a few days after the plan is confirmed. But cash is one thing that few companies in Chapter 11 have. Rights offerings are a frequent solution. In a rights offering, the plan distributes to impaired claim holders, in partial satisfaction of their claims, the right, exercisable on the plan’s effective date, to purchase securities newly issued at that time, typically at a price discount to the securities’ assumed value.[1] If all goes well, the rights holders exercise and the debtor uses the resulting influx of cash to do what the plan calls for.
There is no guarantee, however, that rights holders will exercise. To ensure that the emerging debtor will raise as much cash as it expects to need on the plan’s effective date, it is therefore standard practice to enlist one or more members of the impaired class(es) to “backstop” the offering—that is, to commit to buy any securities that rights holders decline to buy. The so-called backstop parties get one and sometimes two things in exchange for their commitment. They get a fee, paid in kind, and, in some deals, they get a carve-out or private placement of a fraction of the securities on offer.
This approach to capital raising has a commercial logic but also a potentially exploitable flaw. Existing creditors will often be the lowest-cost source of new capital.[2] Giving new-investment rights to those who are being forced to take a haircut, by allowing them in effect to fund cash distributions to senior creditors (and so “walk the walk”), can ameliorate valuation disputes. And firm-commitment underwriting of speculative capital raises, which is what a backstop commitment amounts to, is ubiquitous in corporate finance across time and space. The arrangement means, however, that two impaired creditors holding identical claims will face different all-in economics on account of the debtor’s emergence from its bankruptcy.[3]
In recent years, the exclusivity of backstop participation rights has spurred acrimonious litigation[4] and indignant commentary.[5] The nub of the issue is that backstop parties are not drawn randomly from the set of impaired claim holders. Usually, they are members of an ad hoc group that collectively holds enough claims to accept the debtor’s proposed plan of reorganization on behalf of the relevant class or classes of claims.[6] Critics thus see in the backstop agreement a means by which a coalition of creditors can use Chapter 11’s classified voting mechanism to appropriate, in combination with the debtor, value that belongs by right to a broader set of creditors.[7] These circumstances betray a possibility that backstop creditors might be getting not just different economics, but better economics, than similarly situated creditors not invited to the ad hoc group.[8] Defenders, meanwhile, emphasize the imperative of a deal. Exclusive dealing is not all that bad. It can yield benefits, too, including its tendency to unwind freeriding incentives.[9]
To students of corporate reorganization, the evolution of rights offerings is part of a larger story of the erosion of class solidarity norms. Disputes between creditor groups are as old as tradable corporate debt. Records of the earliest railroad reorganizations are replete with allegations that holders of one kind of instrument had colluded with company management to seize what rightfully belonged to holders of another kind of claim.[10] But for most of the last 150 years, creditor disputes were mainly hierarchical, pitting higher- against lower-priority creditors in a contest for enterprise value.[11] Recently, however, jockeying among creditors who hold similar or identical claims has become the dominant theme in restructuring practice. From uptier exchanges[12] to debtor-in-possession loans that are open only to a subset of first-lien lenders, the most remarkable practical innovations in recent years share a functional logic: they allow a group of creditors, with effective control of the governance rights of a debt facility, or class of claims, to provide new capital to the distressed debtor on terms not available to holders of the same claims who lack such control. They thus achieve a potentially superior blended return on investment. In this sense, debates about rights offerings are inevitably about more than rights offerings.
At the same time, arguments about rights offerings have a distinct legal tenor because a statute is the focal point of contestation.[13] The several theories that disaffected creditors advance boil down to a single idea embodied in the Bankruptcy Code’s requirement that plans of reorganization “provide the same treatment for each claim . . . of a particular class.”[14] Defenders and critics of standard rights-offering practices can each make out a plausible, but unsatisfyingly overbroad, argument from that language. On one hand, backstop arrangements are structured as contracts between the debtor and participating creditors. Subject to any objectors’ opportunity to be heard, debtors in possession are generally allowed to make binding deals.[15] Formally speaking, then, a plan of reorganization premised on a rights offering backstop simply honors a contract. The claims in the relevant impaired class or classes are treated identically; it’s just that holders of some such claims are also party to a separate financing contract. On the other hand, the formal structure of a plan ought not to obscure functional realities. If the debtor would not have allowed a backstop creditor to underwrite its capital raise on the agreed terms, but for the creditors’ claims—but for, more specifically, the creditors’ willingness to vote their claims in favor of the proposed plan—then, one might think, the backstop rights have been offered for the claims irrespective of what the plan proponents might insist.[16] To honor formalism in this context would practically negate the Code’s distributional logic.[17]
This article presents the results of an empirical study of backstop economics that we undertook with this legal dilemma in mind. Our motivation starts with the observation that one’s understanding of the equal-treatment rule may depend as much on one’s sense of the facts on the ground as on the strength of dueling canons of statutory interpretation. If the returns that backstop parties earn on their capital commitments are, in fact, commensurate with the risks they bear, then the returns should be understood as being forthe certainty they provide and not for their claims against the debtor.[18] That the backstop parties are also creditors would be, in a sense, accidental. But the opposite characterization would be justified if, in fact, returns on backstop commitments predictably exceed what the associated risk can justify. The debtor’s readiness to pay more than a competitive return would be puzzling unless the backstop parties’ claims also mattered. In short, rights offerings either are, or are not, a technique used to increase the recoveries of some class members relative to others. Proponents and critics of standard practice implicitly disagree about the answer.
To shed light on the matter, we study what we believe is the full set of Chapter 11 plans, confirmed between January 2016 and June 2024, that feature a backstopped rights offering of the debtor’s post-bankruptcy common equity. The full sample consists of 49 plans for which we find detailed information about the rights offering and backstop agreement. In addition, we find an arm’s-length estimate of the value of the equity in 19 instances (which we call the “market” subsample). To the best of our knowledge, this study represents the most comprehensive effort to assess the returns to backstop commitments, and the first to do so using realized economic outcomes rather than only plan-assumed values.
The results of our study are striking. Backstop parties receive fees that dwarf those paid to underwriters in similar types of capital raises. In the full sample of plans, the average fee (including the projected value of carve-outs) is estimated to be 20.1% of the capital to be raised. Analysis of the market subsample suggests that the actual value of fees is likely even more. The reason is that plan assumptions tend to understate realized equity value, on average by a factor of 1.5. Using arm’s-length prices, the average fee (including the realized value of any carve-out) is 25.1% of capital to be raised. To be sure, averages disguise substantial variation. Several cases in which backstop parties received enormous fees have an outsized influence; the leastgenerous backstop agreements appear almost stingy. Still, fees in most deals are multiples greater than what underwriters of initial public and seasoned equity offerings typically earn (from 3% to 7% of capital raised[19]), even though such offerings entail marketing and distribution work that has no analog in a Chapter 11 offering.
Backstop parties’ realized risk also appears to be low relative to what one might surmise from reviewing Chapter 11 disclosure statements. Mechanically this is a function of the tendency of plans to understate realized equity value and thus overstate the likelihood that the backstop will be called to fund at a loss. The average Chapter 11 plan sets the exercise price of the backstopped rights at a 25.3% discount to what the plan assumes the equity will be worth. In other words, using plan-assumed valuations, the average plan assumes that backstop parties will have to fund at a loss if the value of the debtor’s equity declines by more than a quarter between the disclosure statement hearing and the plan’s effective date. According to arm’s-length realized valuations, by contrast, the average plan sets the exercise price at approximately a 50% discount to value. On average, in other words, backstop parties are at risk only if the debtor’s value drops by half.
High fees and low risk together mean impressive returns for backstop commitments. Although we do not observe returns directly, we use realistic, conservative assumptions to estimate the magnitude of returns. In the full sample, we estimate that backstop parties receive an average of 21 cents per dollar of eligible claim held, compared to 8 cents for non-backstop creditors who participate in the offering and exercise their rights optimally. The difference is once again magnified in the market subsample. Using realized market prices, backstop parties receive, on average, 35 cents per dollar of eligible claims held, while similarly situated participating creditors receive 19 cents.
Our results suggest that backstop parties receive compensation for their capital commitments in excess of what competitive investors would demand for the associated risk. To be sure, our sample warrants caution. Nineteen cases are not a lot. Big losses for backstop parties in a couple of cases could change the picture. It is possible that the 30 cases in our sample for which no public mark is available are meaningfully different. But the high returns and low risk we observe are striking. Whatever weight one puts on the results, they should strengthen one’s conviction that backstop parties in general are being compensated both for their capital commitments and their willingness to vote claims in support of a proposed plan.
II. Institutional Background
A. New-Money Exit Financing
1. The Need for Cash
Cash shortage is the proximate cause of most Chapter 11 cases. During the first twenty years or so of the Bankruptcy Code, corporate debtors, freed from their ordinary debt-service obligations, could build up a cash balance by staying in bankruptcy. That changed around 2000, when amendments to the Uniform Commercial Code made it easier for lenders to secure liens on all a company’s productive assets, including cash,[20] and thus hold debtors on a tighter leash, even in Chapter 11.[21] Debtors today usually need external financing to get through bankruptcy and recapitalize the business. The lenders who provide the needed “debtor-in-possession” (or “DIP”) financing meter out liquidity, so that the company’s management has little cash beyond what is strictly needed to operate the business in Chapter 11 and to pay restructuring professionals. At the conclusion of a bankruptcy, therefore, emerging companies often must seek investment to move forward.
Two uses for new money frequently drive exit financing needs. First, a debtor may need to pay some claims in cash to get a plan confirmed. The Bankruptcy Code requires cash payment in full of administrative expenses and some priority unsecured claims.[22] DIP loans must be repaid, too, if the lenders do not wish to roll over their loans into a post-reorganization credit facility. A plan might also call for cash payment of senior claims about which the Code is agnostic. Doing so can resolve valuation disputes with wary creditors that might otherwise preclude consensual confirmation. Creditors in a relatively senior class who assert what others believe to be an unrealistically low value for the debtor’s business (or for the securities it might issue) can get in the way of a deal. A plan that pays their claims, in full or in part, with cash can dampen or eliminate the basis of dispute.
Second, a debtor may face a competitive need to make capital investments. Skimping on investments is a frequent tactic to manage the illiquidity that precedes most bankruptcy filings.[23] Corporate managers who hope to avoid bankruptcy can (at least) delay its reckoning by cutting costs that pay off only over time. They may put on hold everything from new product development to routine maintenance, even when the expected benefits of investment exceed the price tag. Thus, companies emerging from bankruptcy are apt to have more profitable capital investments to make than comparable companies that have not faced financial distress.
The sources of liquidity that a company will have upon emergence from Chapter 11 are its exit financing. A plan of reorganization must be “feasible” to be confirmed—shorthand for the Code’s requirement that the emerging debtor has to have financing sufficient to do what the plan provides and not need another bankruptcy filing in short order.[24] Consequently, debtors line up their exit financing during bankruptcy—indeed often before bankruptcy, through a restructuring support agreement—and incorporate it into the plan on which creditors will be asked to vote. Exit financing may be senior debt, including rolled-over DIP loans. In modern practice, however, debtors often want or need to raise cash by selling junior securities, especially equity.
2. Creditors as Low-Cost Capital Providers
A debtor’s existing creditors are often the best sources of new capital. They tend to be positioned better than third-party investors for two reasons. First, they possess asymmetric information about the debtor’s business.[25] Existing creditors have already invested in studying its risks. They have a working model of the enterprise and the cash flows it is likely to produce in various plausible scenarios. They do not need to repeat the effort, while third-party investors must underwrite from scratch. Because existing creditors have an informational edge, moreover—or more precisely because third parties expect that they do—third parties can be expected to draw adverse inferences about a debtor’s actual prospects if its existing creditors are disinclined to invest new money. In short, third-party investors hesitate to contribute because they understand themselves to be less informed than those who already have a stake. This can mean that existing creditors can afford to invest new money at a somewhat lower expected all-in return and are often prepared to do so on a faster timeline.
Second, new investments in a debtor’s business, if commercially justified, will often have positive spillover effects that enhance the value of existing claims. Positive-expected-value investments increase not only the value of the debtor company; they also increase the expected value of its outstanding debts. A third-party investor gets one thing in exchange for an investment: the agreed-upon return on capital. An existing creditor, by contrast, gets two things: the agreed-upon return on capital plus any expected appreciation in the value of its existing claims that is attributable to new money.[26] Consequently, existing creditors will often prove the best sources of new money even if third parties would also find an investment opportunity to be risk justified.
Historically, practical constraints have sometimes prevented creditors from investing in a debtor’s post-bankruptcy equity. Banks, for example, may be barred from using cash to buy equity. Taking equity in exchange for prepetition claims, as part of a bankruptcy process, is part and parcel of being in the lending business—investing fresh cash in equity positions less so. First-generation Collateralized Loan Obligations (CLOs) likewise seem not to have been allowed to buy equity. Now, though, most of the players in large-corporate Chapter 11s are ready to commit capital. Distressed and “special situations” funds with flexible mandates actively seek opportunities to deploy their capital and can hold large, illiquid positions. As their participation in the bankruptcy process has increased, rights offerings in bankruptcy have become more prevalent, allowing debtors to secure new capital while deleveraging.[27]
3. Rights Offering Dynamics
Consistent with this theory, debtors looking to raise funds upon emergence from Chapter 11 often do so through a rights offering, in which holders of claims in the fulcrum creditor class are given the right to purchase securities (frequently common stock) to be issued by the post-bankruptcy firm at a fixed price (typically at a discount to the assumed value of the security).[28] These rights are valuable options. A claimholder who, at the time the plan becomes effective, thinks that the value of the securities on an offering exceeds the strike price will exercise (if doing so is within its investment mandate), while a claimholder who sees it the other way around will not.
This optionality is costly to the debtor for the same reason that it is valuable to the claimholder. If unexpected economic difficulties arise, the rights offering may fail to generate the cash the debtor needs to emerge. To insure against this catastrophic prospect, debtors thus seek a backstop commitment, in other words, a promise from someone credible—usually a group of claimholders who are to receive rights under the plan—to buy any securities that the offerees decline to buy. Backstopping is thus akin to a “firm-commitment” underwriting in the public-offering context. Economically the transaction is equivalent to the debtor buying a put on its post-bankruptcy securities. If, as the debtor hopes, the value of the securities to be sold at emergence exceeds the strike price of the distributed rights, then the put will expire worthless, because the offerees will exercise their purchase rights. But if the strike price exceeds the value of the securities, then the debtor can force the backstop parties to buy.
Benign explanations for standard practice are thus available. If a debtor company wants to allow an entire class of claimants to participate in financing its post-bankruptcy future but also needs to ensure that it has enough cash available to do what its plan requires, then asking willing creditors to provide the backstop could be best for all constituents. Unlike an investment bank, the kinds of creditors who might backstop a Chapter 11 rights offering have already done their diligence on the company and have a clear view and confidence interval. Why pay for an investment bank to underwrite the offering when informed and engaged investors are standing at the ready?
At the same time, backstop agreements can be used to transfer value from excluded creditors to members of the ad hoc group who might provide the commitment. The debtor may enter into a backstop agreement with some of its existing creditors, not because they represent the most cost-effective underwriters, but as a form of covert distribution on their claims, effectively encouraging these creditors to support the plan even if its nominal treatment of their claims might be unsatisfactory. Understandably, those in charge of reorganizing the debtor’s capital structure may care little about creditor support beyond the minimum amount needed to confirm its plan—67% of the claims in a class by value and 51% by number—which the ad hoc group by design is able to deliver.
Whether a backstop arrangement primarily functions as a tool for enriching a subset of creditors depends on the price of the firm commitment relative to the benefits it provides to the debtor. Because a backstop serves as insurance against the undersubscription risk, its value is determined by the extent of that risk. If the fees exceed the benefit conferred upon the debtor, then they can be understood as juicing the returns to some creditors in a class, in exchange for their support for a plan of reorganization.
B. Legal Controversy
Common practice has sparked debate in and out of the courtroom. The present controversy over backstop terms could be said to begin with the confirmation, in 2017, of the plan of reorganization in the Peabody Energy Chapter 11 case.[29] With a large carve-out and generous backstop fees, Peabody’s plan epitomized the dynamics that excluded creditors would increasingly challenge.[30] Rights offering terms had been contested before, but the tone and frequency of disputes began to change in the years following Peabody,as it increasingly became clear that whatever intra-class solidarity norms had once limited the magnitude of differential treatment would no longer discipline non-pro rata deals.
Creditors seeking to throttle a plan’s confirmation on the ground that it contemplates a rights offering inappropriately favoring certain investors have pursued several distinct legal theories. In our opinion, the most vital of such theories turns on the creditor-specific right to equal treatment. Section 1123(a)(4) provides that all plans, to be confirmable, must “provide the same treatment for each claim … of a particular class” unless a creditor agrees to inferior treatment.[31] The logic of the argument, as the reader by now understands, is that a debtor’s offer of generous backstop fees to an exclusive subset of creditors is a way to indirectly give them superior treatment that the Bankruptcy Code directly prohibits. Disgruntled creditors have pursued other plausible theories, as well, including that backstop fees are “unreasonable” to an extent that precludes plan confirmation,[32] that high fees constitute impermissible vote buying, and that plans with exclusive offerings are not proposed in “good faith.”[33] In our view, though, the unequal-treatment line of argument is the most telling because it is the most parsimonious.[34]
In other, similar contexts, courts have required bankruptcy transactions that have a structural capacity to undermine distributional entitlements to survive a market check. The clearest comparison is to the Supreme Court’s decision in 203 N. LaSalle Street Partnership v. Bank of America, N.A.[35]In 203 N. LaSalle, the Court was asked whether it had been error for a bankruptcy court to confirm a plan of reorganization in which owners of the debtor’s equity were allowed to retain their equity, in exchange for the contribution of $6 million, despite an objecting class of unsecured claims receiving mere cents on the dollar. By its terms, the absolute priority rule allows equity investors to retain a stake in such situations as long as they do so not “on account of” their prepetition interests. The debtor argued that, in the event, the equity investors were to retain their stakes on account of their new contributions, not their prepetition relationship. Notwithstanding the textual logic of the argument, the Court balked. Because this so-called “new value corollary” to the absolute priority rule could be used to undermine altogether the hierarchical distributional logic of Chapter 11, going forward courts would have to satisfy themselves that the value of the post-bankruptcy interests being given to prepetition investors was subject to a market test.
Reorganization practice under § 363 has faced a similar, market-based discipline. Such sales may not implicate plan confirmation, but they do have the potential to steer value from creditors who by right should share in it to a favored buyer. On its face, the Code’s text does not impose any conditions on the debtor’s sale decision. Nevertheless, bankruptcy courts by local rule and custom have adopted processes to mitigate threats. In general, debtors who announce a plan to sell substantial assets outside the ordinary course of business must do so under a procedures order that envisions competition of some kind. A debtor who has a favored, “stalking horse” bidder in mind must test whether other investors are willing to offer superior terms.[36]
How exactly these precedents and the basic principle of § 1123(a)(4) should apply to the rights offering context is not a settled matter. The prospect that a particular backstop agreement could generate impermissible unequal treatment in the context of a particular case seems to have made some bankruptcy judges uneasy about approving contested arrangements. In the case of Pacific Drilling, for example, Judge Wiles was willing to approve the plan only after its terms were revised to eliminate a $100 million private placement and adjust the backstop fee structure.[37] Judge Goldblatt seems to have encouraged a compromise in the TPC Group bankruptcy.[38] Similarly, in the LATAMcase, Judge Garrity approved an exit finance package only after its terms were softened and the parties introduced valuation evidence tending to show that the terms were comparable to previously approved deals.[39]
Until very recently, no court had accepted an invitation to categorically condemn standard backstop practice.[40] Some courts, like the Eighth Circuit in Peabody, seem to have found the formal severance of a backstop agreement from a plan of reorganization decisive. Others, as we say, seem to have settled for a case-by-case examination of the propriety of a debtor’s proposed backstop fees. None of those instances accepted the direct translation of 203 North LaSalle to the exit-finance context.
In September, however, a district court reversed confirmation of ConvergeOne’s plan of reorganization on just that ground.[41] ConvergeOne had proposed a plan that featured a typical rights offering and backstop structure. The terms of the backstop arrangement were not extravagant by the standards of common practice. But on Judge Hanen’s reasoning, the market typicality or indeed reasonableness of backstop fees is immaterial.[42] Instead, he concluded, as in the new-value context that 203 North LaSalle confronted, that the Bankruptcy Code imposes a procedural obstacle to plan features that structurally threaten parties’ distributional entitlements. For a backstop fee to pass equal-treatment muster, it must be offered ratably to all members of an eligible class or else market tested.
III. Logic, Data, and Method
To better understand this legal controversy and, more generally, to shine light on contemporary backstop practice, we study recent Chapter 11 plans that feature an equity rights offering as part of the debtor’s emergence from bankruptcy. This part explains our aim in doing so and describes our data and analytical approach.
A. Logic
The legal significance of our study is premised on an underlying idea about how the Bankruptcy Code works. Our premise is that the answer to whether a creditor’s recovery under a plan of reorganization is “for” the claims it holds depends on a causal fact. If backstop parties receive value, nominally in exchange for their financing commitments, that the plan proponent would not have offered but for their status as claimholders, including their agreement to vote claims for the plan, then the treatment is for the claims. The value that backstop parties receive in exchange for their financing commitments is not for their claims, if the plan proponent would have offered it irrespective of their status as claimholders.
This premise poses an epistemic challenge. A plan proponent’s motivations cannot be directly observed, and our data do not—and cannot—prove them in any particular case. A substantial portion of the premium backstop parties receive may be attributable to structural features of the market, including the information advantages and spillover benefits of incumbent creditors.[43] Those advantages can give existing creditors quasi-monopoly power even in the absence of any voting leverage. At the same time, however, because the ad hoc group typically controls the votes needed to accept a plan on behalf of one or more classes, the Bankruptcy Code’s voting scheme itself gives the group market power in the provision of new capital and thus an opportunity to demand a supra-competitive return.[44] In practice, compensation will frequently reflect both channels—market structure and voting leverage—and mixed causation is fully consistent with the conclusion that some part of the premium that backstop parties receive is “for” their claims.
We therefore proceed on the view that the economics of backstop agreements give rise to reasonable inferences about tendencies, if not about a debtor’s motivations, in any single instance. Conceptually, the returns to backstop parties should mirror the returns that the seller of a put option on the debtor’s post-bankruptcy equity would receive. If the value of the securities to be sold exceeds their price (as of the plan’s effective date), then most, if not all, rights holders will exercise their rights, because to buy the security is free money. In this case, the issuer does not call on the backstop (except trivially, and in a way that benefits backstop parties, if some rights holders are unable to exercise due to financial frictions). The put expires worthless. If, by contrast, the price of the securities being sold exceeds their value (as of the plan’s effective date), then the rights holders will not exercise, because to buy the security would be to throw away money. In that case, the issuer calls on the backstop to fund at the agreed price. Just as with a conventional put, the seller of the option always gets a fee, but realizes a corresponding loss only if the value of the relevant security drops below the exercise price.
A backstop commitment’s similarity to a put means that a risk-justified fee is, in theory, calculable. The Black-Scholes option pricing paradigm says that the value of the option depends on four numbers: the strike price, the expected value of the underlying security, the expected volatility of prices, and the option’s duration. For any given Chapter 11 equity rights offering, two of these figures are known. The disclosure statement necessarily discloses the strike price (i.e., the price at which the equity is to be sold), and an express term in the backstop agreement limits the duration (i.e., the time between backstop commitment and the plan’s effective date). The expected value of the equity to be sold is unobserved, however—the plan-assumed value might be a good proxy but need not be—and likewise the expected volatility. This means that no mechanical calculation of the fair price for a backstop commitment is possible.
At the same time, the Black-Scholes framework is useful for developing an intuitive sense of the risk-justified return profile that an arm’s-length backstop party could expect. More specifically, realized prices in actual cases that are observable can help us calibrate a rough model of the risks that backstop parties face. This is because realized prices are useful albeit imperfect signals of expected prices and price volatility. If it turns out, for example, that the value of equity sold in Chapter 11 rights offerings is frequently less than the exercise price of the rights distributed to creditors pro rata, then either the expected prices are close to the strike price, or there is a lot of volatility; either way, a relatively high fee is to be expected. If, on the other hand, it turns out that the value of equity is always more than the exercise price—if, that is, the “put” always expires worthless—then a much smaller fee would be appropriate.
It should be clear by now that plan-assumed valuations have limited epistemic value. Their two defects relate directly to the Black-Scholes parameters. First, plan assumptions may be biased as well as imprecise. By choosing a higher or lower valuation, the plan proponent can increase or decrease the discount at which the debtor will issue new securities. Second, plans assume no volatility. On plan assumptions, the price of offered equity is always less than its value, and consequently, backstop parties are never called on to actually fund. This is not to say that plan assumptions are worthless. Plan assumptions are manipulable only within bounds. At some margin of inaccuracy, disaffected creditors will demand a valuation hearing at which they can contest the proponent’s formulations. Plan assumptions are not random. Still, one should much prefer realized prices where available.
Accordingly, an important part of our approach is to bring forth arm’s-length valuations. In many instances, unfortunately, no such valuation is available. Most companies emerge from Chapter 11 private, and the quality of a price signal diminishes as the interval between a company’s emergence and the first mark increases. But when a credible mark is available, it can help us develop Black-Scholes-inspired judgments about the fees paid to backstop parties.
B. Data Collection Methods
Sample Construction. We assembled a data set of Chapter 11 cases in which a plan of reorganization was confirmed between January 2016 and June 2024 and contemplated a backstopped rights offering. We were ecumenical in our efforts to compile a universe of cases. We started with a list that Debtwire publishes of cases in which the plan contemplates an equity offering. We supplemented with data from the Florida-UCLA-LoPucki Bankruptcy Research Database, which tracks every public company with debts in excess of $100 million, measured in 1980 dollars (about $314 million in current dollars), that has entered bankruptcy from the advent of the Code, in 1979, to the end of 2022. The Bankruptcy Research Database has, for each debtor, a dummy variable indicating that the debtor issued equity under a plan. Such plans subsume the kinds of plans we want to study and so allow us to cast a wide net. Because today, many large companies in Chapter 11 are private, we used LoPucki as a check on the completeness of the Debtwire sample. The Debtwire sample included all but one company indicated by LoPucki, allowing us to conclude that the Debtwire list was reasonably complete.
Some of the most recent Chapter 11 debtors we tracked informally through news services, included Debtwire, LFI, and Reorg Research. We thus were able to assemble a screening sample of cases that we believe is reasonably, if not perfectly, complete. The screening sample consisted of 134 confirmed plans. The associated disclosure statements were not readily available online or did not include the relevant backstop information in 14 instances. Where it was available, we read the disclosure statement manually and excluded cases in which the plan did not meet our requirements, either because it was a false positive (the plan did not in fact contemplate a rights offering) or because it did not offer common equity (which we ignore to simplify this analysis). After manual screening, we were left with a sample of 49 cases.
Coding Plan Terms. The terms of proposed exit financing are disclosed in a debtor’s disclosure statement. Often the terms are incorporated by reference to other documents, such as a restructuring support agreement or backstop agreement. The documents do not use uniform nomenclature. Words and phrases that are common across documents frequently signify slightly different concepts. Coding is thus a laborious process. Nevertheless, the disclosure statements we reviewed invariably provided enough information to allow us to calculate the economics of the deal using plan assumptions about the debtor’s post-bankruptcy enterprise or equity value.
For rights offerings, a number of variables are important: the amount of money the debtor proposes to raise, the amount of post-bankruptcy equity that rights holders are to be offered (pro rata), the allowed amount of the claims composing the class or classes that are to receive rights, the purchase price of the equity, and the fees paid to backstop parties (typically in post-bankruptcy equity). In some cases, we have also documented a percentage of the offered equity that is reserved exclusively to backstop parties, either directly or through an accompanying private placement.
We also observe the date on which commitments are made (the date on which the court approves the backstop or private placement agreement and/or the date of the hearing on the disclosure statement). From the docket, we are able to gather the “plan’s effective” date, which is the date, usually soon after plan confirmation, on which the exit financing transactions typically close.
Measuring Equity Value. To assess the returns to creditors who participate in a rights offering—backstop parties and otherwise—one has to compare the purchase price of the equity they receive with its value. Most debtors emerge from Chapter 11 without publicly traded stock. There is therefore no readily available, arm’s-length market assessment of the value of equity issued in those instances.
We looked for a reliable arm’s-length mark within a year of plan confirmation, using the first available mark indicating liquidity. Our objective was to measure the value investors could get for their equity at a time when they were able (or, in some cases, forced to) sell it for cash. There are three broad categories of mark. In a very few cases, the debtor emerges from bankruptcy with publicly traded debt. This is typically only the case when the debtor was a public company before bankruptcy and it turns out to be solvent, so that prebankruptcy equity is never canceled. Hertz is a recent example. In these cases, getting a mark is a trivial exercise using Factset, CRSP, or Compustat. In many other cases, the debtor emerges as a private company, but the equity soon begins to trade (after an initial public offering (IPO), in which case getting the mark is trivial, or in private transactions recorded by Bloomberg). Finally, there are cases in which the debtor emerges as a private company, we cannot observe trading, and then the company is acquired. In many of those cases, the consideration to shareholders is either a matter of public record or is reported in Pitchbook. Between Factset, CRSP, Compustat, Bloomberg, and Pitchbook, we were able to identify an arm’s-length estimate of the post-bankruptcy value of equity in 19 cases within our sample.
Sample Description. Ultimately, we were able to obtain the terms of the rights offering and backstop commitment for 49 confirmed plans. Those materials allow us to compute returns on backstop commitments assuming that the debtor’s post-reorganization equity is worth what the plan assumes it to be worth. This sample is similar to, although more comprehensive than, the samples other researchers have studied.[45] Table 1 reports basic characteristics of the companies and rights offerings we study. The companies span a range of industries. As one would expect from the corporate bankruptcy docket during our sample periods, they concentrate in the energy and retail sectors. The average offering sought to raise approximately $400 million of new capital. However, the range is enormous. The smallest offering sought to raise only $20 million, whereas the largest offering brought in $4 billion.
Table 1: Comparing the Samples
| Full Sample | Market Subsample | |||||||
| Industry | Transportation: 26% Oil & Gas: 24% Equipment Services: 14% Other: 36% | Oil & Gas: 32% Transportation: 26% Equipment Services: 16% Other: 26% | ||||||
| 25 | 50 | 75 | Mean | 25 | 50 | 75 | Mean | |
| Enterprise Value | $625 | $1,030 | $2,675 | $1,930 | $612 | $1,450 | $3,103 | $2,688 |
| Proceeds | $90 | $263 | $469 | $417 | $116 | $277 | $540 | $571 |
| Discount | 10% | 25% | 35% | 24% | 12% | 25% | 35% | 25% |
| Carve-Out Amount | 0% | 0% | 35% | 15% | 0% | 0% | 18% | 12% |
| Stated Fee | 6% | 8% | 10% | 9% | 6% | 8% | 10% | 8.9% |
Of the 49 plans in our sample, we find a reliable market price of the post-bankruptcy equity in 19 instances. We cannot be sure that what we call the “market” subsample includes plans—and companies—that are similar to the 30 for which no market price was available. The reader should thus be cautious in inferring that what goes for the market sample holds for all companies that conduct equity rights offerings. At the same time, Table 1 suggests similarity as far as it goes.
C. Data Limitations (and Corresponding Assumptions)
Our data have two important limitations that deserve emphasis. First, we do not observe the proportion of claims in the relevant class or classes that the backstop parties hold.[46] This restricts our ability to estimate the magnitude of the difference in treatment that backstop and non-backstop creditors receive from an exit financing package. Ideally one wants to compare investors’ blended return on two items—the purchase of equity, if any (the rights to do which may be distributed pro rata or may be privately placed with the backstop parties), and the commitment to purchase unsubscribed shares. Doing so is only possible, however, if one knows the fraction of the relevant claims that compose the backstop group.
To illustrate, suppose that a plan of reorganization offers a supra-competitive fee for the provision of a backstop commitment. If the class of claims from which the backstop parties are drawn is a fulcrum class, such that the claimholders are entitled to the residual value of the business, then by definition every dollar by which the backstop is overpriced reduces the value of the post-bankruptcy equity that they will purchase in the rights offering (or otherwise receive as a distribution on their claims). As returns to the backstop commitment go up, returns on the purchase of equity go down. The key point to notice is that the financial significance of the arrangement to the claimholders depends on their participation in the backstop. If one claimholder, C1, holds a small amount of the claims relative to the amount of exit financing it provides, then some other claimholders, C2 . . . Cn, must hold large amounts of claims relative to the exit financing they commit, such that C1’s blended return will exceed C2’s blended return. On the other hand, if every claimholder participates pro rata in the exit financing, then they all will have the same blended return, notwithstanding that the exit financing offers above-market returns. Put differently, backstop fees could not cause expropriation, even if overpriced relative to what a Black-Scholes calculation would indicate, if every member of a class were to backstop an offering pro rata. Because we can’t observe the size of the ad hoc group relative to the class (or classes) of creditors from which the group is drawn, we can’t directly measure how much better off, if at all, backstop parties are relative to similarly situated creditors who are excluded from the ad hoc group.
To deal with this limitation, we provide estimates of the relative returns to backstop and non-backstop creditors that would prevail at various sizes of the ad hoc group. Although our estimates are therefore only illustrative, however, we believe they are still valuable guides. We understand that ad hoc groups often close near, but just above, a threshold at which the Bankruptcy Code grants control rights. Debtors want to negotiate with creditors who, as a group, have the power to make things happen (or veto proposals that other creditors might like). Under the Code’s classified voting rules, the magic numbers for a class of financial claims are two-thirds (or one-third) of the claims in the class.[47] Thus, we offer return estimates that assume that the backstop parties hold those amounts of the relevant claims.[48]
A second limitation is that we do not observe whether, or to what extent, the emerging company calls upon backstop parties to purchase unsubscribed equity.[49] This is unfortunate because backstop parties’ returns depend on how much equity they (are obliged to) buy, as well as on the price at which they buy it and its economic value. To estimate returns, we therefore make two simplifying assumptions that will mechanically understate the actual difference in returns to backstop versus non-backstop creditors: (1) if the equity’s market price is below the strike price specified in the plan, we assume that no rights holders exercise their rights; by contrast, (2) if market price exceeds the strike price, we assume that all rights holders exercise their rights. In reality, some rights holders might exercise even when the mark on which we rely suggests that it would be a bad investment; and some rights holders cannot exercise, due to liquidity or other constraints, even when the mark suggests that exercising would have been the expected-wealth maximizing move. Compared to reality, then, our assumptions systematically understate backstop parties’ recoveries.[50]
IV. Results
A. The Full Sample
The full sample of 49 plans, for which we rely on plan assumptions of value, reveals some striking facts. Consistent with previous studies of smaller samples, we find generous fees relative to similar types of firm-commitment underwriting and accordingly strong (assumed) returns to the backstop parties’ capital commitments.[51]
Fees. Table 2 reports compensation that backstop parties received according to plan assumptions about the value of post-bankruptcy equity. The “stated fee” is the amount of the in-kind premium disclosed in a debtor’s disclosure statement, measured as a percentage of capital to be raised in the offering. The average stated fee is 9%—high relative to customary underwriting fees in similar contexts, but perhaps not alarmingly so. The stated fee mechanically understates economic value in most cases, however. Most rights offerings are premised on the notion that the newly issued securities will be sold for less than their value—that is, at a discount. Yet disclosure statements typically calculate the amount of the backstop premium as though it consisted of securities worth only the discounted offering price. In order to find the fee’s value according to the plan’s own assumptions, one has to back out the discount (and in some instances other adjustments). The average fee so calculated—what we call the “adjusted fee”—is 12.9%.
To get a complete picture of a backstop party’s compensation, one must add the value of a carve-out, where relevant. A carve-out is an exclusive allocation of securities for a price less than what they are assumed to be worth. (Of course, a carve-out becomes a net liability for the backstop parties if the value of the securities drops below the purchase price.) The difference between what backstop parties pay and the value of what they receive should be added to the adjusted fee. On average, using plan assumptions, the total compensation for backstop commitments is 20% of the capital to be raised.
Risk. Under plan assumptions, backstop parties face no risk of loss. That is a tautological conclusion because disclosure statements lack a notion of volatility. Instead, even though everyone knows that volatility is (of course) real, plans state a fixed estimate of value and then propose to offer newly issued equity at a discount to the stated value. On our assumptions about when rights holders rationally exercise their rights, the lack of volatility means that it is impossible for backstop parties to be called on to fund at a loss.
The conclusion is not only obviously unrealistic; it is unrealistic in an important way. If backstop parties really face no risk of loss, then they deserve no more than their transaction costs plus the risk-free rate of return for securities with comparable duration—for example, the yield on 3- or 6-month Treasury bills—on capital actually reserved for the transaction. This rate was in the range of 0–2% for most of the period we study, ranging up to approximately 5% in 2023–2024.[52] And the amount of capital that backstop parties would reserve in this counterfactual world would be very small, since rights holders by hypothesis would be sure to earn a positive return for exercising their rights. The fees that are risk-justified in the real world, in which risk is non-zero, will be greater. How much greater, though, is an open question.
Returns. Another way to measure backstop compensation is to compare the total returns, per dollar of rights offering-eligible claim held, that backstop parties receive from participating in a rights offering to the returns that similarly situated creditors who are excluded from the backstop group receive. This measure has the virtue of focusing attention on what the Bankruptcy Code directs attention toward, namely the putative equality with which a plan treats similar claims.
Two empirical obstacles preclude precise measurement of returns, however, and deserve comment. First, we do not observe how many non-backstop creditors exercise their rights. Instead, we make the simplifying assumption that they always exercise their rights optimally: they buy their full allotment of equity when its realized value turns out to be greater than the strike price and buy no equity (forcing the backstop parties to buy) when its value turns out to be less than the strike price. This assumption means that we are reporting what is actually a floor to backstop parties’ returns and a ceiling on non-backstop creditors’ returns.
Second, we do not observe the fraction of rights-offering eligible claims that members of the ad hoc group hold. This is a problem because returns to backstop parties are defined by the sum of two items—fees, broadly understood, and returns on the exercise of ratably distributed rights—divided by the amount of claims held. The fewer claims held, the more that fees dominate. Indeed, as the fraction of eligible claims held goes to zero, returns per claim dollar go to infinity. The more claims held, the bigger role for the exercise of ratably distributed rights. Indeed, as claims go to 1, returns to backstop and non-backstop creditors become identical. Consequently, we illustrate returns at two thresholds of ad hoc group size.
Table 3 reports our estimates of backstop parties’ incremental recoveries. At the 67% threshold, using plan assumptions, backstop parties on average recover approximately 21 cents per dollar of eligible claim held, compared to 8 cents per dollar for non-backstop creditors who exercise their rights optimally. We also illustrate returns for cases in which the ad hoc group members hold a mere blocking position in the eligible class.[53]
B. The Market Subsample
In the 19 cases for which we obtain arm’s-length valuations, the realized economics for backstop parties are better than what one would infer from Chapter 11 plan disclosures. The main reason seems to be that plan assumptions of the value of post-bankruptcy equity are systematically low relative to its realized value: on average, the market-implied value was 50% more than plans assumed.
Understatement of equity value drives two effects in the data. First, it means that fees are higher than what a naïve reading of disclosure documents would suggest. Because backstop parties are paid in kind, the value of newly issued equity determines their real compensation. Second, systematic understatement of post-bankruptcy equity value means that backstop parties are less likely to be called on to fund at a loss than plan documents would suggest. Holding volatility constant, the bigger the difference between the exercise price of the rights being distributed under the plan and the value of the underlying equity—i.e., the more “in the money” the rights are—the less risk the backstop parties face.
Fees. Table 4 reports the compensation paid to backstop parties in the market subsample. Panel A uses plan assumptions. On these assumptions, fees in the market subsample are similar to those in the full sample. The average stated fee in the market subsample is 8.9% (compared to 9.0% in the full sample). Adjusted fees are likewise similar (13.4% compared to 12.9%).
Panel B calculates fees using the realized, market-implied value of equity instead of plan assumptions. This story is somewhat different. Because the realized value of equity sold in rights offerings is much greater than the assumed value, on average, the realized value of (paid-in-kind) fees is also much greater. The average adjusted fee is 14.3% of capital to be raised. The realized value of carve-out rights is also much higher than what plans assume. On average, the average sum of adjusted fees and carve-out value is 25.1% of capital to be raised. (Median compensation is much lower, 15.4%, but still high compared to underwriting fees for similar offerings.)
Risk. Realized values indicate that backstop parties shoulder little risk of loss.[54] Again, the tendency of plans to understate the value of post-bankruptcy equity drives that result. The realized value of equity exceeded the plan value in 12 of 19 cases in our market subsample. In 6 of the remaining 7 cases, the size of the discount at which equity was sold in the rights offering nevertheless meant that backstop parties did not suffer losses. Rights holders who could participate would have been wise to do so; to the extent that some did not, the ensuing call for backstop funding would have enhanced rather than decreased the backstop parties’ recoveries. In only one case was the offering price for the post-bankruptcy equity less than its realized value, such that, if rights holders were wise, backstop parties would have been forced to fund at a loss.
Returns. Table 5 reports backstop parties’ incremental recoveries for the market subsample.[55] It reflects the same responses to empirical limitations as Table 3 does. To aid in comparison between the market subsample and the full sample, Panel A reports returns using plan-assumed valuations. The figures for the market subsample are again broadly like those for the full sample. Panel B reports returns using market-implied valuation. Returns to backstop parties and non-backstop creditors alike are significantly higher than plan assumptions would indicate: the value of the equity is, on average, more than plans indicate, so that the difference between its value and the (plan-fixed) price is correspondingly greater, too. On average, the incremental recovery to backstop parties is about the same as under plan assumptions, i.e., 16 cents per dollar of claim held. A few cases in which equity value vastly outstrips the offering price drive much of the disparity, however. In the median case, backstop parties receive 17 cents on the dollar while non-backstop creditors receive 12 cents.
V. Implications
Our summary conclusion is that Chapter 11 backstop parties typically get more compensation than the financial risks they bear would seem to justify. A couple of qualifications are important, but do not threaten the big picture. One is that deals differ. In a significant minority of cases, fees are roughly in line with what investment banks receive for underwriting equity offerings on a firm-commitment basis in similar contexts. Even if plans tend to understate the realized value of backstop compensation, the discrepancy between what ad hoc group members and similarly situated non-backstop creditors who participate in the rights offering recover on their claims may be quite small. Variation across deals is a fact. The other qualification is that the sample of companies with market-implied valuations is small and might not be randomly drawn from the full sample. Simply put, our estimates of realized fees and recoveries might not hold across the full sample of rights offerings. Still, compensation in the average case is remarkably high, even according to plan assumptions. Neither the variation we observe nor prudent skepticism about the weight that our precise estimates deserve undermines our general thesis.
What should one make of the conclusion that backstop parties often—usually—are compensated for their acquiescence in a debtor’s proposed plan of reorganization? This part looks at the question through two lenses. On first principles, the status quo is a mixed bag. There are costs to having similarly situated creditors scramble to be part of an exclusive coalition. But there are benefits, too, especially in a world in which non-pro rata restructurings are a part of common practice in out-of-court dealmaking. On the law, though, our results suggest a more straightforward conclusion. The Bankruptcy Code’s classified voting scheme is premised on the notion that the holders of claims in a single class stand in the same relationship to any proposed plan. Radical identity of interests is a farfetched dream, but the Code’s architecture makes little sense if debtors can induce requisite consents from creditors simply by giving side payments to the necessary majority of holders.
A. For Policy
Although the equal treatment of similarly situated creditors stands as a central tenet of U.S. bankruptcy law, it is far from clear that this formal commitment to parity always serves the system’s broader aims of efficiency or substantive fairness. Equality, in other words, may be foundational without being optimal. In a recent article, Ken Ayotte and Alex Huang pose a provocative thought experiment about a Wonka-esque reorganization mechanism.[56] In “golden ticket bankruptcy,” as they imagine it, a debtor that seeks judicial protection to reorganize chooses—and announces on Day One in open court—a lucky creditor who will take the business free and clear of its prepetition debts. Claims and interests held by other, rather unlucky investors are extinguished, and the company and its new owner emerge with a capital structure better tuned to the enterprise’s needs. A surprising insight is that on many dimensions “golden ticket bankruptcy” dominates the system that the United States, or indeed any jurisdiction’s, law prescribes. It is fast and cheap to administer, and it gives the ticket holder powerful incentives to make good use of the assets at hand.[57]
The point of the thought experiment, however, is not to lay groundwork for a statutory overhaul of the Bankruptcy Code. It is to illustrate the tradeoffs associated with any bankruptcy regime that honors non-bankruptcy distributional entitlements. There is a reason that no country has enacted “golden ticket bankruptcy,” namely that it introduces a variety of pre-bankruptcy problems. It increases the variance that creditors face. It encourages them to lobby for favored status. It chases away from the credit markets entirely those who believe they will not be favored (discouraging impersonal credit markets). It heightens the importance of the bankruptcy trigger. It gamifies the acquisition of control rights. In short, it warps corporate law and finance in myriad ways, including for companies that will never face a need for bankruptcy. Our point, though, is that inequitable distribution, even when extreme, is problematic—if at all—only insofar as it undermines independently significant non-bankruptcy goals. Yet accommodating those goals entails costs that make the reorganization of any particular business more cumbersome than it otherwise would be.
A high-fee backstop agreement is like a pale version of “the golden ticket bankruptcy” and poses analogous tradeoffs. Ad hoc group members do not get anything close to the entire value of the reorganizing business, but they get more than what the holders of legally identical claims receive. To weigh the soundness of such an arrangement on first principles means considering both how it can smooth a debtor’s path through bankruptcy, as well as the perverse incentives it creates for companies facing distress, but not in bankruptcy.
On one hand, a debtor’s pushing value toward a coalition of creditors who jointly have voting control should (in theory) and surely does (in practice) speed the reorganization process and tends to increase the company’s enterprise value going forward relative to a system of strict equality.[58] Defenders of the status quo say as much, and the limited academic research on the topic supports their claims.[59]
Ad hoc group members take concrete steps beyond voting their claims that tend to speed reorganization. Some or all “get restricted”—that is, they waive the ability to trade while they examine and negotiate (and thus receive material nonpublic information about the debtor) a plan for the reorganization. Together with the eventual restructuring support agreement that typically gets signed, the waiver allows debtors a stable group with which to bargain.[60] Part of any excess recovery that backstop parties receive may be a reward for active monitoring and innovation and for the implicit liquidity costs, as well as out-of-pocket expenses, of negotiating a deal—expenses that other creditors, particularly passive investors, like structured product or mutual fund managers, do not share.[61]
On the other hand, even limited favoritism can entail significant albeit indirect costs. The prospect of non-pro rata treatment increases the variance associated with holding syndicated loans and bonds.[62] It will also induce creditors to spend and trade in ways they think will either get themselves into the ad hoc group, avoid the excluded group, or both. Among other things, small institutions will tend to flee the scene. They are unlikely to hold a big enough percentage of an impaired class to plausibly threaten a blocking veto and therefore, are unappealing members of an ad hoc group. The same goes for creditors who lack a relationship with the debtor or large fellow creditors. In effect, the social and practical dynamics of coalition formation discourage the provision of impersonal credit. These effects are probably relatively minor when the incremental recovery for ad hoc group members is small and become larger as the increment widens. Thus, while a system that allows a modest premium for backstop parties might work fine, large premiums could cause real dislocations that bankruptcy courts, and indeed restructuring professionals more generally, may be ill-positioned to observe.
A comprehensive assessment of tradeoffs must consider, as well, the tendency of one or another bankruptcy rule to encourage or discourage out-of-court restructurings.[63] Requiring strictly ratable plans in bankruptcy might stimulate flight, on the margin, to “liability management exercises” that involve less judicial scrutiny.[64] In other words, the availability of non-pro rata recapitalization transactions outside bankruptcy bears on the optimal rule in Chapter 11.[65]
Our findings paint a controversial picture: backstop parties earning significant returns, transforming bankruptcy into what resembles a “golden ticket” system. In this version of Chapter 11, a select group of creditors walks away with the lion’s share, while others are left to watch from the sidelines. To be clear, our research does not imply that current rights offering deals are inherently inefficient; they may serve important purposes, such as securing commitments in volatile creditor landscapes. Yet the tradeoffs are delicate and complex.
B. For Law
In light of the Bankruptcy Code as it exists, however, our findings weigh in favor of a norm that the opportunity to backstop a rights offering be made open ratably to holders of eligible claims or else market tested. Our findings do not, of course, suggest that every backstop agreement negotiated under the current, generally permissive regime violates the Code’s distributive principles. Nevertheless, because fees and incremental recoveries are, on average, sufficiently large—and because sorting between reasonable and unreasonable fees is inherently challenging—courts, in our view, should take a prophylactic approach.
The courts have required market testing in other circumstances posing similar tradeoffs. The closest analog is the Supreme Court’s ruling in 203 N. LaSalle Street.[66] In that case, the debtor sought to cram down a plan of reorganization under which prepetition equity investors, in exchange for cash contributions, would retain their equity even though unsecured claims in a non-accepting class would not be paid in full. The debtor argued plausibly enough that the Bankruptcy Code prohibited confirmation in such a case only if the retention of equity was “on account of” the prepetition interests;[67] but here the debtor was insisting that it was for something else, that it was for the new-money commitment. The Code does not express a view about how widely the post-bankruptcy investment opportunity had to be offered. The bankruptcy court agreed and confirmed the plan. The Supreme Court, however, reversed for essentially epistemic reasons. When prepetition equity holders get an exclusive right to “buy” post-bankruptcy equity, it’s just too hard to know whether the reward is, in fact, on account of the new investment or not.
The absolute priority rule, however, is not the only analogous case. The out of the ordinary course sale of debtor assets is another.[68] The Code says little about the process a sale proponent must observe. It simply authorizes the debtor in possession to “use, sell, or lease” property outside the ordinary course “after notice and a hearing.”[69] Yet, when the going-concern sale became a frequent way to resolve financial distress, the bankruptcy courts that saw the highest volume of sales also noticed that a sale process could be used to steer value from disfavored creditors to a favored buyer (which could be an affiliate of an insider or favored creditors). They responded, not through case law, but with the promulgation of local rules and the observance of local customs that ameliorate a judge’s epistemic challenge.[70] Bankruptcy judges simply would not authorize a sale to a stalking-horse bidder favored by debtor management without getting comfort that other plausible bidders had been genuinely encouraged to offer more.
Rights offering backstops present bankruptcy courts with a now-familiar epistemic challenge: distinguishing between fair compensation for capital at risk and covert premiums granted in exchange for control. In our view, the high fees that backstop parties frequently receive—but also the variation in such fees and the difficulty judges face in telling how much risk a given offering faces—warrant a similar approach.
Judicial review, although formally available, is in practice limited by institutional and process constraints. The reality is that bankruptcy judges are poorly positioned to second-guess plan assumptions about value or to guess at the price volatility associated with a reorganizing company’s post-bankruptcy equity. Unfortunately, no amount of additional disclosure or standardization can cure the problem; the necessary calculations would continue to depend on the debtor’s assumptions and thus replicate their limitation.[71] Adversarial reckoning is fine as far as it goes, but to challenge a plan proponent’s projections is costly and risky. Rational creditors, even those who have good arguments, will usually prefer just to take a loss.[72]
These risks are magnified when the backstop participants are also members of a controlling ad hoc group capable of delivering (or blocking) plan acceptance by one or more impaired classes. In such cases, exclusive participation in the rights offering functions not merely as underwriting compensation, but also as a strategic inducement for plan support—a form of in-class value redistribution that occurs outside the reach of formal plan confirmation scrutiny. The resulting asymmetry undermines the premise that holders of like claims will receive like treatment.
To address these concerns, the courts might revisit whether exclusive backstop arrangements implicate the “equal treatment” requirement of § 1123(a)(4). A categorical approach—treating all rights offerings as either within or outside the scope of claim treatment—has proven analytically rigid and operationally problematic. If all rights offerings are deemed claim treatment, cross-class holdings would necessitate widespread bifurcation and invite holdouts. But if all such offerings are categorically excluded merely because they are inked through side agreements, then the equal treatment principle risks becoming hollow.
What is needed is a functional middle ground—an interpretive approach that looks not to the formal structure of the transaction, but to its broader economic effect. In our view, a market test offers precisely such a framework: a tractable mechanism for evaluating whether exclusive backstop terms in any particular case reflect market-clearing compensation for risk or instead mask a value transfer to pivotal claimants.[73] The Texas District Court’s recent decision in ConvergeOne,[74] like 203 N. LaSalle before it,seems to strike just such a middle ground. In our view, it is for that reason a model decision.
Conclusion
Creditors who agree to backstop Chapter 11 rights offerings appear to receive supra-compensatory fees for the risk they assume. This fact suggests, in turn, that debtors—not necessarily all debtors, but debtors on average—are paying for more than a financial commitment. They are paying for the backstop parties’ acquiescence in a plan.
From the perspective of a debtor company’s managers and advisors, a willingness to pay for an ad hoc group’s very real, Bankruptcy Code-conferred control rights is understandable. Often, the backstop parties can strategically delay or otherwise exacerbate the costs of a reorganization, and upon examination, super-compensatory fees represent little more than a transfer from some (minority) financial creditors to others who organize more effectively. In this sense, high backstop fees are just one manifestation of the intra-class conflicts that now permeate distress. In some cases, the superior returns that ad hoc group members receive may even be socially justified as compensation for their work identifying, facilitating and negotiating a viable path forward for the company.
At the same time, Chapter 11 announces a rule of equal treatment for identical claims. Whatever one thinks of such a rule as a matter of first principle, it is the rule, and indeed as a logical matter, it is crucial to the Code’s presupposition that supermajority consent is a reliable proxy for the economic interests of similarly situated creditors. Until now, each side of the debate about the propriety of exclusive backstop agreements has, for the most part, dealt with the equal treatment rule by simple assertion. One side asserts that backstop fees are treatment for the claims (and therefore impermissible); the other side asserts that they are not (and therefore unremarkable). Our findings suggest that, at least on average, the critics are right as a matter of fact.