Printer Friendly


INTRODUCTION                                                     1600
    A. Distortion From Other Bondholders                         1606
       1. Voting Distortions: Self-interest, Holding Out, and
          Misjudgment                                            1606
       2. Free Riding                                            1607
    B. Distortion from Issuers                                   1608
       1. Sticks: Exit Consents                                  1609
       2. Sticks: Differential Consideration                     1610
       3. Carrots: Terms of the New Bonds                        1610
       4. Carrots: Consent Fees and Vote-Buying                  1610
    C. Implications and Correctives                              1611
II. REGULATORY CONSTRAINTS                                       1611
    A. Stockholders                                              1612
    B. Bondholders--Federal Law                                  1614
       1. Amendment: TIA Section 316                             1615
       2. Exchange Offers                                        1619
    C. Bondholders--Contract Law                                 1620
    D. Commentary                                                1621
III. THE NEW RESTRUCTURING                                       1625
    A. The Dysfunctional Workout                                 1627
       1. Holdouts                                               1628
       2. Business Frictions                                     1629
    B. Workouts Versus Bankruptcy: Comparative Costs             1630
    C. The Shift to Workouts                                     1634
       1. Volume                                                 1634
       2. The Empty Creditor Problem                             1635
    D. The New Practice                                          1636
       1. New Data: Distressed Exchange Offers, 2010-2016        1637
       2. Institutional and Household Bondholders                1640
    E. Explaining the Changes                                    1641
       1. Registration Exemption                                 1641
       2. Creditor Control in Chapter 11                         1642
    F. Summary                                                   1645
IV. RETAIN OR REPEAL?                                            1646
    A. The Rise and Fall of the Broad Reading of Section 316(b)  1649
    B. The Case for Outright Repeal of Section 316(b)            1655
       1. Federal Mandate                                        1655
       2. Contracting Practice                                   1656
          a. Drafting Out from Under Marblegate                  1657
          b. Unanimous Action Expanded & Exit Consents
             Permitted                                           1659
       3. Conclusion                                             1663
    C. CACs--the Worst Case                                      1663
       1. The Drafting Task in the Wake of Repeal                1663
       2. The Worst Case                                         1664
    D. Backstop Policing: An Intercreditor Duty of Good Faith    1665
       1. Intercreditor Good Faith Cases                         1665
       2. The Disappearance of Intercreditor Good Faith Cases
          in the United States                                   1668
       3. Scope                                                  1671
       4. Conclusion                                             1673
V CONCLUSION                                                     1673


Bond workouts are famously dysfunctional. When a company is in financial distress, its stockholders and bondholders have every reason to negotiate a restructuring (or "workout") of its obligations to produce a sustainable capital structure and avoid the costs of a bankruptcy. The reality is different. Bondholders hold out and free ride in response to restructuring offers from distressed debtors. Debtors respond with coercive inducements and procedural maneuvers. The result is a destabilizing and potentially toxic mix of creditor opportunism and debtor coercion that can derail the workout process, forcing a bankruptcy restructuring.

Bond workouts occupy a space governed by neither corporate law nor bankruptcy law, regimes designed to bring unruly investors together to settle matters by majority vote. In contrast, the law actually stands in the way of majoritarian decisionmaking with bond workouts. The primary governing law is the Trust Indenture Act of 1939 (TIA), a hoary New Deal securities law that mandates terms in the contracts governing publicly issued bonds. (1) Section 316(b) of the TIA prohibits majority-vote amendments of the payment terms of bonds, foreclosing workout by direct contractual amendment. (2)

But the TIA leaves open a second route to restructuring--the exchange offer, in which the debtor offers to exchange new, scaled-down bonds for the original bonds. (3) Exchange offers are intrinsically susceptible to disruption by holdout bondholders and coercive tactics by issuers. Few of the process protections accorded by corporate and securities law to stockholders receiving tender and exchange offers apply to bond exchanges. (4) There is no judicial oversight of the restructuring process, as would be the case in bankruptcy. Nor does contract law provide in the way of protection against distorted bargaining in a financial context like this one.

The TIA itself bears much of the responsibility for the empty doctrinal toolbox. The TIA was a New Deal reaction to the excesses of a Depression-era out-of-court restructuring market in which insider equity holders and their favored creditors siphoned value away from bondholders. (5) The statute's drafters wanted restructurings to proceed in bankruptcy under judicial and administrative oversight so as to prevent process abuses. (6) They largely succeeded. The TIA's very success in shifting restructuring practice into bankruptcy resulted in the atrophy of the federal equity doctrine that policed earlier restructurings. (7)

In recent years, however, the picture has changed quietly but markedly. Workouts have started to work. A substantial portion--around one-fifth of restructuring activity--has shifted from bankruptcy court to out-of-court workouts effected through exchange offers made only to large institutional investors. (8) The shift resulted from a temporary external shock--the brief disappearance in 2008-2009 of debtor-in-possession financing for bankrupt companies. (9) But the altered pattern persists.

Coercive tactics figure more prominently than ever in the new workouts. Ugly facts and court challenges result. Thus confronted, but possessing no obvious doctrinal tool, courts in the Southern District of New York--the near exclusive forum for bond litigation--responded by adopting a new reading of TIA section 316(b) that would give courts broad power to police workouts. (10) The Second Circuit Court of Appeals subsequently reversed the leading Southern District case, likely returning the law of bond workouts to its earlier posture. (11) Even so, the episode was the biggest jolt to the normally staid world of bond contracting since the leveraged takeovers and buyouts of the 1980s. While we are sympathetic to the policing impulse behind the broad reading of section 316(b), we think the broad reading went much too far. In a context where fact-sensitive policing is needed, the broad reading imposed bright-line mandates that overrode terms in bond contracts and threatened to choke off the new workouts altogether.

This Article exploits the bond market's reaction to these decisions to reassess a longstanding debate in corporate finance regarding the desirability of TIA section 316(b). Section 316(b) has attracted intense criticism in the past, with calls for its amendment or repeal because of its untoward effects on the workout process and tendency to push restructuring into costly bankruptcy. (12) Section 316(b) also has been staunchly defended on the ground that mom-and-pop bondholders need protection from strong-arm tactics. (13)

We draw on a pair of original, hand-collected data sets to show that many of the empirical assumptions made in the debate over section 316(b) no longer hold true. First, we show that workouts are more tractable than thought heretofore. But for the recent judicial intervention in the Southern District, the markets have learned to live with section 316(b), denuding the case for repeal of any urgency. (14) Workouts generally succeed, so there is no serious transaction cost problem stemming from the TIA; when a company goes straight into bankruptcy, there tend to be independent motivations. (15) Second, we show that workout by majority amendment will not systematically disadvantage bondholders. Indeed, the recent turn to secured creditor control of bankruptcy proceedings makes workouts all the more attractive to unsecured bondholders. Third, we show that bond workouts are more coercive than previously thought in some respects, but also less coercive in others.

Based on this empirical background we cautiously argue for the repeal of section 316(b). (16) Section 316(b) no longer does much work, even as it prevents bondholders and bond issuers from realizing their preferences regarding modes of restructuring and voting rules. We do not know what contracting equilibrium would obtain in the wake of repeal, but think that the matter is best left to the market. It follows that repeal should be complete and prospective. We recognize, however, that markets are imperfect and that a free-contracting regime may result in abuses. Accordingly, we argue that a repeal of section 316(b) should be accompanied by the resuscitation of a long forgotten, but still valid, equity doctrine of intercreditor good faith duties, which presents a more fact-sensitive and targeted tool for policing overreaching in bond workouts than the Southern District's broad reading of section 316(b). (17)

This Article makes several contributions to the scholarly literature on corporate restructuring. First, the Article is the only comprehensive treatment of bond workouts. Section 316(b) is a central topic in the law of corporate finance, yet there has been no thorough inquiry into the practice pattern. We go beyond anecdotal evidence to develop a working empirical picture while simultaneously explaining the development of the applicable law against the background of a theoretical discussion of group decisionmaking by investors.

Second, the Article shows that there has been a marked change in the world of debt restructuring, and that a cognizable part of restructuring activity has moved outside of bankruptcy. We explain why the shift is occurring, looking to securities law compliance practice and incentive realignment in the wake of secured creditor control of Chapter 11 proceedings. We draw on an original data set to provide a first glimpse of the new workouts. The descriptive data show that contemporary workouts are flexibly structured and tend to succeed where those attempted before 2008 tended to fail. Specifically, we show that the holdout problem assumed by the previous literature has diminished in salience and that the position of small bondholders, to the extent they still exist, also looks different because they are simply ignored in contemporary exchange offers, which are made only to large institutional investors.

Our third contribution to the literature stems from a second original data set that collects the process terms of contracts governing bonds issued under the Rule 144A exemption and thus not subject to the TIA. The data offer a glimpse at the preferences of bond issuers and bondholders, again upsetting settled assumptions. We show that contracts issued prior to the recent judicial opinions tend to adhere to the broad outlines of the section 316(b) regime, but do introduce some significant modifications. Contracts issued after the recent cases show a new pattern, one group carrying on as before, and another affirmatively rejecting the Southern District's broad reading of the TIA. The contracts also take the surprising step of affirmatively sanctioning a coercive device, the exit consent, utilized in exchange offers.

Finally, we play at legal archeology and rediscover a doctrinal tool better suited to the policing task than the Southern District's broad reading of the TIA--the intercreditor duty of good faith, an equitable tool that became irrelevant following the TIA's passage. It sits on the books unremembered, but it is amenable to revival under the contractual duty of good faith. The recent turn to workouts points to the importance of reconsidering this doctrine. Indeed, it likely will become essential in the event section 316(b) is repealed.

The Article has four parts. The first three look at practice and law along parallel tracks, with the last part bringing them together. Part I lays out the bargaining framework in bond workouts, explaining the array of distortionary incentives and devices that come to bear and showing that, in theory, majoritarian amendment is the least distorted framework. Part II reviews the legal background, looking at the TIA, at other provisions of the federal securities laws, and at contract law, and comparing the treatment of collective decisionmaking by stockholders. It shows that there are precious few legal constraints other than section 316(b), setting the stage for a game of creditor opportunism and debtor coercion in connection with largely unregulated exchange offers. Part III presents the first empirical profile of new workouts, showing how it differs from the traditional picture of dysfunction. Part IV begins by describing the rise and fall of the broad reading of section 316(b), looking carefully at the facts of the cases. The analysis reveals an overbroad and unpredictable standard likely to chill workouts garnering supermajority consent. Part IV then turns to the ultimate policy question: what to do with section 316(b). We recommend outright repeal, but drawing on our empirical evidence of drafting practice, warn that the contract drafters' responses could be incomplete. We suggest that the intercreditor duty of good faith, once pulled out of the doctrinal wardrobe and given a good dusting, would provide an effective solution to any resulting problems.


When corporate borrowers cannot pay, they seek to scale down (or "restructure") their financial obligations. In the United States, this tends to occur in one of two venues. First, restructuring can take place in bankruptcy court, under Chapter 11 of the Bankruptcy Code. Alternatively, the restructuring can take place out of court in what is known as a "workout." (18) A workout is simply a contractually concluded modification of debt effected either by amendment of the terms of the existing debt or an exchange of the existing debt for new obligations. (19)

Bankruptcy can be set in motion by unilateral action by the debtor. By staying enforcement of debt contracts, (20) bankruptcy forces creditors to the negotiating table. Workouts call for more in the way of creditor cooperation. For a company with a large number of individual, uncoordinated creditors-such as tort claimants, trade creditors, and tax authorities--it may not even be worth trying. Outside of bankruptcy, each individual creditor has a veto, at least in regard to its own debt, while inside bankruptcy creditors are grouped into classes within which a majority can bind a minority to a restructuring and in some situations a majority of a single class can bind other classes to a restructuring. (21)

Financial debt can be more tractable. Where there are multiple creditors under the same debt instrument--principally bondholders and syndicated lenders--the contract can provide for majoritarian amendment. Such provisions are known as "collective action clauses" (CACs) when they condition amendment of terms on a majority (or supermajority) creditor vote that binds dissenters. (22) There is, however, a more preclusive alternative. Under a "unanimous action clause" (UAC), (23) each creditor must individually consent to an amendment of terms of its own obligation, (24) preventing majoritarian amendment.

A UAC does not necessarily prevent out-of-court restructuring, however. There is another route: the debtor firm can offer new, scaled-down obligations in exchange for the old obligations, which are not amended directly, but are instead canceled when they are returned to the debtor. (25) A workout by exchange offer still presupposes creditor cooperation, for each creditor retains the choice of whether to exchange, and there is no legal mechanism outside of bankruptcy by which exchange can be compelled.

The concessions bound up in a workout, whether by direct amendment under a CAC or by an exchange offer, may enhance the creditors' collective interest. Publicly traded bonds present the classic case. They trade at a discount to face value when the issuer gets into distress, reflecting the possibility of default and bankruptcy. Bankruptcy entails added costs: direct costs of administration and indirect costs due to the proceeding's destabilizing effect on the company's customers, suppliers, and other constituents. A negotiated reduction of the company's debt burden potentially avoids these costs, keeping the company out of bankruptcy by refitting it with a sustainable debt load. The cost avoidance adds value to the company, making it possible that the bonds will be worth more net of the concessions.

This all sounds nice and neat, but the playing field is bumpy, ridden by problems of distorted consent-giving. This Part lays these problems out, applying a powerful theoretical analysis articulated by Professor Zohar Goshen, which we refer to as the "efficiency account." Goshen's basic proposition is this: when corporate investors make collective decisions impacting their investments' value, the best available process is a binding simple-majority vote. (26) Such a voting process must satisfy a further condition: the investors must vote sincerely, which means that they seek the outcome that maximizes the investment's value as a whole rather than seeking to maximize their own individual returns. (27)

There is also a corollary proposition: any other process for effecting a collective investor decision is presumptively infirm. The corollary does not follow because majority decisionmaking is possessed of some magical property that assures first-best results. No such template exists. Instead, the point follows from negative implication--all other processes carry a higher risk of distortionary influence.

With bonds, some distortions come from within the investor group, as when the consenting bondholders have private agendas and vote strategically, or, alternatively, hold out for a side deal. Other distortions come from the bond issuer, which can inject coercive elements into the decisionmaking process. Section A describes distortions from within the bondholder group, while Section B turns to coercive tactics employed by bond issuers. Section C describes negative implications for successful out-of-court restructuring.

A. Distortion From Other Bondholders

1. Voting Distortions: Self-interest, Holding Out, and Misjudgment

Some voting distortions arise because of bondholder self-interest, holdout strategies, or simple misjudgment. To understand how this works, assume that Company ABC has outstanding $100 million 7% unsecured bonds--more properly known as "notes" or "debentures"--due in two years. (28) ABC has not yet defaulted on its interest payments, but is experiencing severe business difficulties, and default is a possibility. ABC does not expect to be in a position to refinance the 7% bonds when they come due. The 7% bonds are trading for $30, a deep discount from their $100 face value. Assume that the bond contract contains a CAC permitting amendment, but only by approval of a 90% supermajority of the bondholders. ABC has proposed a series of amendments to scale down the bonds' interest rate to 5%, reduce their principal amount (a "haircut") from $100 to $75, and extend their duration by three years. A bondholder will be better off having consented so long as the bonds emerge trading for more than $40. In fact, the amendments will cause the bonds to trade at $50, and the deal allocates all surplus value created by the shift to a sustainable capital structure to the bondholders. Consider the following three scenarios:

(1) More than 10% of the bondholders either (a) have a more significant interest in Company XYZ, which competes with Company ABC and will be injured because the restructuring will make ABC stronger, (b) also own ABC stock which will lose value to the bondholder favorable surplus allocation, (c) hold a freely assignable put option on their bonds that will allow them to sell the bonds at an above-market price at a future date, or (d) hold credit default swaps on the bond that will pay off in aggregate more than the face amount of the bond if the issuer defaults. They vote no and the beneficial deal is lost. The problem here is self-interested, strategic voting activated by a conflict of interest.

(2) Two hedge funds, both of which understand this to be the best deal available, each own 5.1% of the bonds and refuse to consent unless they receive extra consideration on the side, from either the other bondholders or ABC. No such consideration forthcoming, they vote no, and the deal is lost. The hedge funds are voting strategically and self-interestedly, but there is no conflict of interest as regards the transaction. The hedge funds are holding out to extract disproportionate consideration. (29)

(3) The bondholders possess heterogeneous views about the amendments. Although it is the best deal available, more than 10% misjudge the situation, voting no because they believe that the surplus has been allocated to the equity. Although they are voting sincerely, they are still holding out, and their misjudgment kills the deal. (30)

The magnitude of each of these three problems, conflicted voting, holding out, and misjudgment, diminishes as the approval threshold decreases to a simple majority. If we could identify and disqualify conflicted voters and holdouts without incurring collateral costs, we should do so, for they detract from the collective good. We should at the same time distinguish sincere misjudgment from conflicts and holding out. Misjudgments about transaction quality are an inevitable incident of contracting under imperfect information against an uncertain future. One can ameliorate, but not eliminate, the problem by disclosing fully regarding the debtor's business prospects.

2. Free Riding

Let us now bring a UAC into the fact pattern. The UAC blocks majority or super-majority amendment of payment terms. It follows that ABC can only restructure by closing an exchange offer.

Exchange offers work only if enough bondholders accept them. Assume ABC authorizes $75 million 5% unsecured bonds due in seven years and offers to exchange $75 face value of the new bonds for each $100 of old ones. If only a few bondholders refuse to exchange, the new bonds will trade for more than $40. But supermajority acceptance will be necessary in order for the deal to make sense. An exchange of 51% of the old bonds for the scaled-down bonds will not achieve a reduction in the debt load of a magnitude sufficient to avoid bankruptcy. Nor would the new bonds, with their reduced financial rights, trade for more than $40. The offer accordingly will be conditioned on a 90% supermajority tender threshold.

The supermajority minimum tender creates the same potential for disruption from conflicts, holding out, and misjudgment as did the 90% supermajority vote. In addition, the shift from collective voting to bilateral contracting between ABC and individual bondholder-offerees opens up an additional distortionary possibility. A nontendering bondholder cannot have its bonds amended. If the offer succeeds, there will emerge two groups of bondholders, one holding the old bonds and the other the new bonds. Holding out can make sense, even absent a side payment, because there is a potential free ride at the expense of the majority that tenders and takes the scaled-down rights. (31) If the exchange offer closes and the issuer emerges from financial distress, the original, unexchanged bond will be worth more than the scaled-down new bond given to exchanging bondholders. Furthermore, if the issuer emerges in stronger financial condition, the unamended bond is worth more ex post than ex ante. (32) Add all of this up, and a successful restructuring through an exchange offer that is effectuated by less than 100% of the bondholders effects a wealth transfer from the cooperative bondholders to the uncooperative bondholders. The holdouts are able to free ride off of the concessions made by the exchanging bondholders. If enough bondholders try to free ride, however, then the minimum tender threshold will not be reached, and the exchange offer will fail. All other things else equal, amendment by a binding majority vote works better than does an exchange offer, because a majority vote leaves the bondholders in a single group with scaled-down rights, cutting off the free ride.

B. Distortion from Issuers

Now let us shift over to the other side and view the transaction from Company ABC's perspective, make the deal a bad one, and see what ABC can do to coerce the bondholders into taking it anyway.

Exchange offers are inherently coercive because they threaten the liquidity of the old bonds. The liquidity of a bond is a function of how widely it is held. A successful exchange inherently reduces the number of bondholders of the old bonds, and thus their liquidity, resulting in a loss of market value. Indeed, a listed issue can be delisted. (33) This threat encourages bondholders to accept the exchange offer even without any deliberate coercion by issuers.

Issuers rarely rely on the implicit coercion of illiquidity alone, however. An exchange offer is a species of tender offer, and the tender offer form invites coercion because (1) it forces each member of the offeree group to act individually rather than by a collective decisionmaking procedure, and (2) it splits the offerees into two groups, permitting an offer to be structured so that opponents who refuse to consent are left in a worse position. Issuers take advantage of these properties, deploying both substantive and procedural sticks and carrots to encourage acceptance.

1. Sticks: Exit Consents

Bond contracts contain a variety of provisions designed to protect the right to payment. Many of these protective provisions in bondholder contracts are open to majority amendment even when the payment terms are subject to a UAC. (34) Issuers frequently condition bondholders' tenders in exchange offers on the execution of a consent (an "exit consent") to amend or remove these provisions. Business covenants can usually be amended or eliminated by a simple majority vote under the bond contract and, in the case of a secured bond, collateral can usually be stripped with a two-thirds vote. (35) Thus, in an exit consent transaction, a bondholder is invited to exchange the old bond for a new one, but is allowed to do so only after first consenting to an amendment of the terms of the old bond. This move distorts the bondholder's choice. Even if the holder would reject the offer based on an appraisal of its value, the holder might nonetheless accept to avoid being stuck with an old bond with diminished rights and no liquidity in the event the other bondholders accept and the offer succeeds.

Note that this tactic is not injurious per se. The stripping of rights by the exiting bondholders lowers the free ride payoff from refusing to tender, discouraging self-interested holding out within the investor group and making it more likely that a fair offer succeeds. Given a fair offer, the exiting bondholders have every reason to consent to the amendment. Once the bondholder thus decides to cooperate with the issuer, the bondholder will see anything that lowers the value of old bonds left in circulation after the offer's conclusion as adding value to the new bonds. (36)

2. Sticks: Differential Consideration

Issuers also can use procedural machinations to coerce acceptance of exchange offers. Absent regulation, an issue may keep an offer open only a short time, so as to discourage a coordinated response by bondholders. Higher consideration can be offered for earlier tenders toward the same end. Finally, disfavored bondholders can be excluded altogether from the set of offerees, facilitating side deals with self-interested bondholders.

3. Carrots: Terms of the New Bonds

Attractive terms can be included in the new bonds at the expense of the old bonds. Suppose Company ABC offers to exchange the $100 7% old bond due in two years for a $75 face value 8% new bond due in seven years with a junior lien on its property. The 100 basis point interest step up adds a little sweetener without erasing the fact that the new bonds carry lesser financial rights. The lien does even more. Should bankruptcy follow for ABC despite a successful exchange offer, the new bonds will therein rank prior to the old bonds, making the lien a stick as well as a carrot. The same could be accomplished by inserting a subordination provision into the old bonds via the exit consents, so that the old bonds would be explicitly subordinated to the new bonds.

4. Carrots: Consent Fees and Vote-Buying

Coercion can also occur in connection with a majority bondholder vote. The issuer can skew preferences by paying a consideration, known as a "consent fee," to those voting its way. The consent fee, like the exit consent, splits the bondholders into two groups and leaves the nonconsenting bondholders in a worse position. Given a 51% CAC, the issuer can pay a majority of bondholders to approve an amendment that makes the bonds as a whole less valuable but leaves the payees better off net of the payment. If the bondholders cannot coordinate to resist, the issuer can even induce an amendment that leaves each consenting bondholder less well-off, but still willing to vote yes and take the money for fear that a simple majority of other bondholders will do likewise.

C. Implications and Correctives

Exchange offers put bondholders in an unstable situation. They are forced to choose even as the value of the outcome depends on choices made by other members of the group, as well as the issuer's prospects. Holding out and taking a free ride may look attractive, even given exit consents. But, it is a dangerous game. If enough bondholders refuse to tender and the minimum condition is not met, then the offer fails. If a successful offer would have averted bankruptcy, everybody might be worse off. Exit consents make this result less likely, but they too have a dark side. Suppose a successful restructuring does indeed create a surplus, but the issuer has structured the terms of the new bonds such that the entire surplus redounds to the benefit of its stockholders. Here, considered judgment counsels holding out. Unfortunately, given an exit consent, refusing to tender on the merits invites punishment in the form of impaired terms in the event the other members of the group buckle and accept. It is a game without an equilibrium solution as the bondholders choose between holding out and a high payoff, and cooperation and a lower payoff, all against the threat of failure and a still lower payoff for everybody.

Meanwhile, all of the issuer "sticks" just described admit of a simple corrective. In order for group consents to be collected without coercion, each member must be allowed to register its preference without consequences tied to the outcome. This takes us back to the theoretical baseline--the best way to get the investor group from here to there is by simple majority vote conducted without side payments by the issuer. Such a vote also minimizes problems arising from holdouts, free riders, conflicts, and misjudgment. But problems will remain, particularly as regards the latter two. Conflicts that are difficult to detect can occur in large segments of a voting population. Sound judgment depends on (but is not guaranteed by) complete information, a commodity not necessarily forthcoming from transactional proponents.


If corporate and securities law followed the efficiency account articulated in Part I, all collective decisionmaking by security holders, whether stockholders or bondholders, would be subject to a norm of sincere voting and a blanket prohibition of coercive tactics. But that's not how it works. There is a pattern. First comes a duty-driven state law base under corporate and contract law. Under this, absent a fiduciary or contractual duty to be solicitous of the interests of the corporate issuer or the other securities holders in a group, nothing prohibits self-interested voting. Nor, absent a standalone duty, is issuer coercion prohibited. The base is modified by a hodgepodge of provisions in federal securities law pursuing the goal of undistorted investor choice. These constraints tend to take the form of bright-line rules.

The rules covering stockholders and bondholders differ markedly in their details, even as both follow the same pattern. The comparison is instructive. Generally, stockholders are better protected than are bondholders because corporate law contains an overlay of fiduciary duty where contract law does not. Stockholders are also more likely to benefit from the regulatory solicitude of the Securities and Exchange Commission (SEC). But there is also a formative federal intervention on the bondholder side. Under section 316(b) of the Trust Indenture Act, payment terms in bond contracts may not be amended directly. (37) Workouts accordingly proceed only by means of exchange offer. Ironically, both self-interested voting and issuer coercion come to the forefront of the practice as a result. Courts have refused invitations to invoke the contractual duty of good faith against them.

Section A outlines the treatment of stockholder decisionmaking. Section B moves up the right side of the balance sheet to bonds, focusing on federal law, in particular TIA section 316(b). Section C then analyzes the contractual duty of good faith as applied to bonds. Section D explains the overall pattern. It shows that investor self-interest and issuer coercion present targets ill-suited to control under open-ended common law standards. No easily drawn lines distinguish "proper" from "improper" self-interest or coercion. The efficiency account, even as it works on the whole, provides no assistance.

A. Stockholders

Corporate law's voting defaults are majoritarian, (38) minimizing frictions from holdouts and other problems related to supermajority thresholds. But self-interested shareholder voting is not prohibited. Shareholders do not owe one another fiduciary duties and private agendas do not lead to disqualification. The only exception to the rule of self-interest addresses vote-buying, which breaks the overall regulatory pattern. (39) But only outright exchanges of cash for proxies clearly traverse the prohibition. Other arrangements, such as side deals that inject an element of self-interest into the tally, can pass if fully disclosed. (40)

Fiduciary scrutiny is triggered when a shareholder has voting control of the company. (41) But the inquiry does not focus on the vote itself. It takes controlled boardroom action leading to unequal outcomes rather than self-interested voting per se to trigger scrutiny. (42) That said, self-interested shareholder voting does arouse a response at a secondary level--the votes of a fiduciary seeking a shareholder ratification to shield a self-dealing transaction from fiduciary review are dropped from the tally. (43)

The federal securities laws, in contrast, do seek to ensure undistorted consent-giving, although their coverage is intermittent. The primary contribution concerns information, the full production of which is mandated by the periodic disclosure system and the proxy rules. (44) There is also process regulation of third-party tender offers, which are subject to the same coercive tactics employed in distressed debt exchanges. (45) The Williams Act, (46) added in 1968 to the Securities and Exchange Act of 1934 ('34 Act), subjects tender offers to a package of bright-line rules designed to minimize coercion. The rules control an offer's timing and require equal treatment of all members of the stockholder group. (47)

State corporate law also inhibits coercive tactics of tender offerors, but as an incident of fiduciary constraint of management defensive tactics. (48) Coercion is thus deterred indirectly because it gives management a justification for defensive barriers. Tender offers by majority shareholders also come in for special constraint against coercion, again as an incident of a fiduciary duty--in this case the controlling shareholder's duty. (49) Under this regime, a majority of the minority must accept the offer, the independent directors of the target must get the chance to engage a banker and pronounce on price fairness, and the offeror must commit to go forward with a cash-out merger at the same price and abjure retributive treatment of the holdouts. (50)

B. Bondholders--Federal Law

With bondholders, we reverse order and begin with the federal overlay. The reversal follows from the magnitude of the federal intervention, which restricts the contracting space. The TIA mandates terms in the contracts, called "trust indentures," that govern publicly issued bonds, including terms facilitating workouts. (51) The Bankruptcy Code goes on to block contract enforcement (52) and channel restructuring into a judicially supervised process.

1. Amendment: TIA Section 316

Section 316 of the TIA addresses bondholder waivers and amendments under trust indentures, seeking to prevent distorted decisionmaking by taking the decision itself off of the table. Subsection (a) contains two provisions. One of them constrains majoritarian forgiveness of interest defaults by allowing only a payment moratorium not exceeding three years based on a seventy-five percent bondholder majority. (53) The second provision is a limited prohibition against self-interested voting, requiring that votes of the issuer and anyone controlling, controlled by, or under common control with the issuer be disregarded. Subsection (b) applies to amendments, providing as follows:
Notwithstanding any other provision of the indenture to be qualified,
the right of any holder of any indenture security to receive payment of
the principal of and interest on such indenture security, on or after
the respective due dates expressed in such indenture security, or to
institute suit for the enforcement of any such payment on or after such
respective dates, shall not be impaired or affected without the consent
of such holder, except as to a postponement of an interest payment
consented to as provided in paragraph (2) of subsection (a) of this
section.... (54)

The section prevents majorities from binding minorities to amendments of terms implicating "the right to receive payment," sometimes called the "core" terms of the indenture. It does not prevent unilateral consent giving by individual bondholders. If an issue is held by a large number of bondholders and ninety-nine percent consent to remove a core term, the reduction still goes through, but only as to the consenting bondholders. The section, in effect, imposes a mandatory UAC that covers some but not all terms in the indenture. As to "non-core" terms there is no prohibition under the TIA, and trust indentures tend to cover them with simple majority CACs. Despite the different treatment of "core" and "non-core" terms, the TIA does not define what indenture terms are part of the core "right to receive payment." The promises to pay principal and interest, including payment dates and currency of payment, are clearly in the "core," as they constitute the heart of bondholders' right to receive payment--the what and when. The location of the line separating "core" from "non-core" terms has recently become a matter of interpretive dispute, particularly because the words "impair" and "affect" imply a further prohibitive reach to TIA section 316(b). We will take up these matters in Part IV.

Whatever its reach, section 316(b)'s blunt mandate against majority vote amendment is surprising. Indeed, the section makes no sense when viewed through the lens of Part I's efficiency account, which concludes that amendment by majority vote is the least distorted context for out-of-court restructuring. (55) For an explanation, we need to look to the historical context.

The House and Senate reports accompanying the TIA offer the same (verbatim) statement of purpose for section 316(b): "Evasion of judicial scrutiny of the fairness of debt-readjustment plans is prevented by this prohibition." (56) "Evasion" occurred when payment terms were amended in out-of-court workouts. The reasons for concern are set out in the SEC's famous report on protective and reorganization committees, supervised by an all-star team of William O. Douglas, Abe Fortas, and Jerome Frank--among these three, two would become SEC Chairmen, two would become Supreme Court justices, and one would become a Second Circuit judge.

The SEC Report detailed the recent appearance of CACs covering payment terms in new bonds issued in connection with workouts in the real estate sector. The express purpose of the provisions was to substitute faster, cheaper workouts for bankruptcy proceedings. (57) The Report took a close look at the real estate bond indentures (58) along with CACs in bond documentation in Canada and Great Britain, (59) accurately stating the policy case in their favor. (60)

For the SEC reporters, the problem was not that CACs were intrinsically distortionary, but that they would exacerbate distortionary influences in the then-prevailing institutional context. The reporters had nothing against majoritarian concession-making, provided that it was exercised on a fair playing field.

The federal bankruptcy regime had only included corporate reorganization since 1934, (61) when Congress added section 77B to the Bankruptcy Act. (62) Section 77B had dual purposes that stood in tension with one another. (63) The first purpose was facilitative--creditor majorities now, for the first time, could bind creditor minorities to reorganization plans that impaired the minorities' contract rights, making reorganization easier to accomplish. (64) The second purpose was to assure a playing field that would be protective of the interests of smaller bondholders and other security holders, an assurance to be effectuated through judicial process oversight and judicial application of substantive standards. (66)

Section 77B succeeded in facilitation but failed in protection, despite judicial supervision. The corporate insiders and investment bankers who had been stage-managing nonbankruptcy receiverships in the decades before 1934 transitioned to bankruptcy reorganization without missing a beat, continuing to use the process vehicle of protective committees to control every important aspect of bankruptcy proceedings. (66) Unfortunately, protective committees were not very protective of their participating bondholders. (67) There resulted out-of-control agency costs on the part of those responsible for framing restructuring plans and strategic distortion of consent-giving processes. The SEC Report devoted hundreds of pages to detailing the abuses. (68) A proliferation of CACs would have facilitated the protective committees' survival. Section 316(b)'s CAC prohibition followed in 1939.

Significantly, the TIA's progenitors knew both that their scheme implicated a tradeoff in the form of additional bankruptcy costs (69) and that CACs prevailed in other systems. (70) They also knew that the buyers of new bond issues tended to be institutional investors and that mom-and-pop bondholders had more or less disappeared from the Depression-era market's buy side. (71) They traded all of this in pursuit of the ideal of a system in which investors make undistorted choices, an ideal they thought realizable only given some sort of supervision by an omniscient, neutral administrator.

2. Exchange Offers

Section 316(b)'s protective purpose was never realized due to a critical omission. The TIA does not constrain exchange offers, which emerged as the exclusive vehicle for out-of-court workouts after 1939, complete with almost every distortionary feature described in Part I--holdouts, exit consents, differential consideration, sweeteners, and consent fees. Federal law imposes only a single anticoercive restraint, going to timing. The Williams Act (72) requires the issuer to hold open both offer and any exit consent for twenty days and an additional ten days in case of an extension. (73) But the Williams Act's other process protections do not apply to offers for debt securities. (74) Therefore, exchange offers accordingly do not have to be made to all holders (75) and can address only a limited group of bondholders. Nor is the issuer required to pay the same and highest consideration under the offer to all tendering bondholders. (76) Instead, the offer can be structured to pay more for early tenders, hustling the bondholders to accede.

C. Bondholders--Contract Law

State contract law adds little in the way of supplemental protection, apart from leaving the drafters of trust indentures the option of including explicit terms that foreclose the possibility of exchange offers and coercive processes. Contract's supplemental policing doctrine, the implied duty of good faith and fair dealing, (77) turns a blind eye to coercive tactics in the corporate finance context.

The good faith axiom generally receives only lip service from the courts in the context of bonds. The courts proceed from an assumption that the parties to these contracts are sophisticated and can bargain for the terms they want. (78) They accordingly have held that the implied duty of good faith derives directly from the language of the indenture. It follows that the good faith duty can be implied only when directly supported by an express term (79) and cannot provide bondholders with rights inconsistent with the indenture's express terms. (80) Thus formulated, the duty loses its gap-filling quality. This approach makes life simple and predictable for courts and comports with the courts' general approach to interpreting financial instruments. (81)

The leading case taking this approach in connection with a debt restructuring is Katz v. Oak Industries, (82) in which Delaware Chancellor William T. Allen rejected a good faith challenge to an exit consent attached to an exchange offer and directed to lifting business covenants. The device violated no express terms of the contract. (83) Its coercive character was acknowledged, but not found to traverse any applicable norm--the bondholders were deemed to have a free choice between participating and holding out. (84) Nor did the Chancellor see any problem with the issuer taking actions to benefit shareholders at the expense of creditors--that, after all, is what directors are supposed to do. (85)

Chancellor Allen also is responsible for the leading decision on bondholder vote-buying, Kass v. Eastern Air Lines, Inc. (86) This case concerned a thirty-five-dollar-per-bond payment in exchange for a consent to waive a dividend covenant. (87) The plaintiff, invoking a general public policy against vote-buying, claimed a breach of the good faith duty stemming from the fact that the payment went only to consenting bondholders rather than going to the group on an equal basis. (88) Chancellor Allen rejected the argument, seeing no reason to doubt that the payments lay within the expectations of the parties to the contract. (89)

D. Commentary

A comparison of Part Fs efficiency analysis and this Part's sketch of the regulatory framework raises two questions. First, why do we have this incomplete, patchwork response to the distortion problem? Second, why is there a treatment differential between stockholders and bondholders? Stockholders enjoy full application of the Williams Act and a prohibition against vote-for-cash trading, and minority shareholders enjoy special anticoercion rules that go beyond those imposed by the Williams Act. Bondholders get none of these benefits. This section suggests some answers to these two questions.

Note, first, that there is an across-the-board reluctance to imply protective, common law duties to assure undistorted consent-giving. The big statutory interventions--the TIA and the Williams Act--are the legislative results of moments in history in which their respective subject matters emerged as front-line policy concerns. In both cases, bright-line federal mandates supplemented the base of state law without influencing its normative coloration.

At the same time, the logic of Part I's efficiency account has not filtered into state law's normative framework. Without this policy baseline, there is little on which case law might build. Self-interested voting raises no hackles, absent a contrary duty on the voter's part. Were it otherwise, there would be a difficult problem of sorting proper from improper self-interested motivation. No theory presents itself other than Part I's efficiency analysis, which signals a regime of sincere voting, an ideal result difficult to realize in the real world due to problems of verification. Unless a given voter has a duty to disclose, there may be no way to ascertain his or her motivations. The prohibition of votes-for-cash applied to common stock emerges more as an exception to the rule than as a normative base point susceptible to expansion.

Coercive setups are easier to identify than voter self-interest. But, absent an independent duty not to coerce, it once again is difficult to formulate a norm that draws a line between the proper and the improper. The efficiency analysis again provides no help in drawing the line. It signals that both tender and exchange offers should be banned outright, a result that makes no sense either as regards stock, as to which cash tender offers play an agency cost reductive role, or bonds, as to which the TIA makes the exchange offer the sole mode for out-of-court restructuring. When a court is asked to intervene against a coercive exit consent, nothing precludes the possibility that doing so would inhibit the closing of a beneficial deal. As we saw in Part I, coercion is not objectionable per se in practice, and can have a useful instrumental aspect when holdouts are present.

Such intervention is duty driven, and contract law offers very little with which to fill the gap. This is not only a function of the ad hoc barriers developed in the case law to application of the good faith duty in financial contexts. There also are structural impediments. Coercion per se invalidates a contract only if it amounts to duress, which presupposes an improper threat and no reasonable alternative, (90) extreme conditions not present in bond workouts. Coercive tactics and hard bargaining do figure into unconscionability avoidance. (91) But, with the exception of one outlier case from a half century ago that invoked the interest of helpless mom-and-pop bondholders in invalidating a bond contract term as adhesive, (92) the bundle of notions bound up in unconscionability have no traction in this big money context. Finally, the TIA itself defuses any sense of fact-driven urgency. As Congress has already intervened here to ensure protection, no further exertion in the form of an implied contractual duty is required.

Bankruptcy law provides an instructive contrast. Though bankruptcy is an equitably driven process explicitly devoted to creditor protection, it does relatively little regarding self-interested voting and coercive kickers. Under Chapter 11, (1) a reorganization plan's proponent must be in good faith (93) and (2) the votes of creditors that are not solicited or cast in good faith may be designated by the court (that is, cast as the court sees fit). (94) There is little case law interpreting these mandates, and neither provision has been read expansively. Thus, claimants protesting coercive "death trap" reorganization plans that penalize nonconsenting creditors have succeeded in invalidating them for failure to meet the "fair and equitable standard," (95) but not as bad faith coercion. (96) The ban on bad-faith voting applies only when a creditor casts its vote not to maximize the return on its claim, but rather to increase the value of other investments; (97) or when a creditor has engaged in "obstructive tactics ... [and] hold-up techniques" to exact better treatment for its individual claim, rather than for all similarly situated creditors with claims in the same classes Merely amassing a blocking position for the purposes of preventing plan confirmation will not, by itself, result in designation. (99)

We note that a court inclined to take self-interested voting or bond issuer coercion seriously could break with the foregoing pattern by making two moves. First, process would be distinguished from substance. In the leading bondholder good faith cases, the plaintiffs sought added substantive rights, and the courts refused, hesitating to disturb settled allocations of risk and return. Objections to workouts ask for considerably less in the way of intervention, going not to the risk-return allocation but to the bargaining framework for its modification. As such, a good faith claim falls into territory already problematized in contract law, which admits substantive review of modifications of executory contracts. (100) Second, the court would intervene discretely so as to conform to the pattern of exception by rule. A different result in Katz called only for a rule against exit consents. Kass asked only for a cash-for-votes ban like that already in place for common stock. (101)

Unfortunately, there is a powerful policy reason to refrain from such a process-based application of the good faith duty to bond workouts. Workouts already are hard to do because TIA section 316(b) blocks majoritarian amendment of payment terms; courts do not want to make things even harder. Exchange offers are seen as likely to fail because they prompt holding out. In order to minimize the holdout problem, issuers attach high minimum tender conditions to the offer, which make success less likely. When an issuer also resorts to coercive exit consents and consent payments, it is in effect retaliating against holdouts in the service of a deal that just might make everyone better off. A different result in Katz, prohibiting exit consents, would make workouts still harder to do, implying cognizable opportunity costs. The same goes for vote-buying. Return to the consent payment sanctioned in Kass, which concerned an amendment that relaxed a business covenant. (102) Going concern modifications like these succeed without cash consideration only given cooperatively disposed lenders. "Relational" lenders, such as banks in syndicated loans and insurance companies in classic private placements, have the requisite cooperative incentives, (103) which bondholders do not. Payment is expected when bond issuers seek waivers. Vote-buying, although suspicious when considered in the abstract, imports useful flexibility in this arm's length context.

This policy case against judicial intervention to inhibit issuer coercion stems from a generally accepted picture of uncooperative bondholders and dysfunctional exchange offers, a picture that in turn informs a longstanding policy case for repeal of section 316(b). Part III shows that the practice picture has changed in recent years. Today, exchange offers tend to succeed, so the case for judicial intervention looks different.


Professor Mark Roe initiated a policy case against section 316(b) in an article published in 1987. (104) Roe saw section 316(b) as a source of avoidable costs. Workout by exchange offer was just too hard to do. Uncertainty and opportunism were combining to prevent value-enhancing exchange offers from closing where direct amendment would have succeeded, causing the agency costs of debt and the cost of bankruptcy to run to excess. (105) Bondholders who might otherwise support a deal of uncertain value withheld consent because they did not wish to be victimized by free-riding holdouts. (106) Exit consents ameliorated the holdout problem without solving it. (107)

Professor Victor Brudney countered a few years later, arguing for the status quo from what amounted to a polar opposite position. Like Roe, he assumed that holdouts tend to cause exchange to fail, but drew very different policy inferences. (108) Brudney insisted on a process platform offering undistorted choice and found that all proposals for majoritarian amendment came up short under that standard. (109) Meanwhile, the contracting platform was tilted against the bondholders, (110) who labored under a collective action problem and had no opportunity to negotiate with issuers over what amounted to a take-it-or-leave-it offer. Issuers behaved strategically, timing their offers to suit their own agendas. Information asymmetries were intrinsic and irremediable. (111) Brudney also challenged the notion that restructuring failure necessarily triggers excess bankruptcy costs, asserting that in many cases the issuer was deteriorating so severely as to land in Chapter 11 even in the event of a successful workout. (112) Exchange offers had no significant cost reductive effect in such cases, and simply served to weaken the rights of those classes making concessions. (113)

Roe, Brudney, and subsequent participants in this discussion work from a common bundle of assumptions. In the generally accepted picture, section 316(b) makes out-of-court restructuring dysfunctional by foreclosing direct majority amendment of payment terms. Exchange offers come with 80% to 90% minimum tender conditions because bondholders insist that the holdout possibilities be minimized. (114) This makes the offers likely to fail, prompting issuers to respond with coercive ploys like exit consents. Even so, failure is likely. It is a lose-lose outcome, with bankruptcy following and a surfeit of additional costs that make everyone worse off. With direct amendment by a two-thirds or three-quarters majority, none of this would happen.

This Part shows that the practice has changed materially in recent years. First, bankruptcy itself has become cheaper with the proliferation of prepackaged, or "prepac," bankruptcy during the 1990s and the shift to creditor control after the turn of the century. Second, and more importantly, out-of-court restructuring activity has increased markedly during the past decade. Moreover, restructurings now tend to succeed--holdouts are much less of a problem. Today's exchange offers are negotiated deals--the take-it-or-leave-it ad in the paper has disappeared. At the same time, issuer coercion is more salient than ever. Where formerly it was merely common, now it is ubiquitous. We call these "the new workouts."

Section A reconstructs the old picture of workouts, drawing on financial economic studies of datasets dating from before 2008. The studies provide support for both the Roe and Brudney positions, variously showing a high failure rate, issuer coercion, and holdout behavior but also detailing a long list of business reasons why a well-informed, uncoerced, and sincere bondholder might refuse to tender.

Section B considers comparative costs. It shows that workouts are, indeed, the cheaper mode of proceeding, but explains that, quite apart from the section 316(b) barrier, circumstances often close off the out-of-court route. Discussion then turns to the cost of bankruptcy, detailing the proliferation of prepacs and a more recent trend toward faster, cheaper disposition in regular bankruptcies. The negative cost implications of a failed restructuring have dropped substantially.

Section C looks at workout volume. The uptick in out-of-court activity began as the result of an external shock, when financing sources dried up in 2008, but the change in the pattern has persisted. The shift occurred in the teeth of predictions that workouts would disappear altogether due to the proliferation of credit default swap protection, the purchasers of which have everything to gain from a bankruptcy filing.

We then turn, in Section D, to contemporary restructuring practice. We draw on a hand-collected data set of exchange offers made by SEC reporting companies from 2011 to 2016 to show that most workouts close. None of this goes to say that there are no holdouts or that holdouts no longer complicate matters. It just recharacterizes them as a secondary concern to issuer coercion.

Section E pursues parallel explanations for the change, narrow and broad. The narrow explanation concerns the federal securities laws: a shift in the basis for the exchange offers' exemption from section 3(a)(9) of the Securities Act of 1933 ('33 Act) to the Rule 144A exemption under the 1934 Act makes the field more receptive to negotiation. The broad explanation concerns the interplay between restructuring out of court and in bankruptcy. Previously, a bright line separated the alternatives. The choice between the two lay in the debtor, and debtor discretion was the rule, even inside Chapter 11. Now, secured creditors tend to call the shots inside bankruptcy. These creditors also loom large in out-of-court negotiations. The result is a graying of the line between in and out of court, with process choices emerging as incidents of a single, overall negotiating process.

A. The Dysfunctional Workout

There is considerable support for the assertion that workouts are dysfunctional. The empirical literature shows that restructurings tend to fail, with holdouts figuring prominently in the account. It also shows that bondholder opportunism prompts issuer coercion, which in turn enhances the chance of success. But the downward spiral of opportunism and coercion is not the only salient factor. Business fundamentals also figure prominently when workouts fail.

The studies, which cover the 1980s and 1990s, tend to show that half or less of attempted exchange offers succeeded, (115) although depending on the set of deals examined, success rates varied from 27%, (116) to as high as 75%. (117) Putting aside the 75% outlier, the failure rate implies that holdouts, taken together with ancillary creditor coordination problems, impose an opportunity cost. (118) But it does so without foreclosing other, business-based explanations.

1. Holdouts

The studies attempt to confirm a causal connection between holdouts and deal failure indirectly, by establishing a causal connection between the holdout threat and the exit consent. The studies proceed from the premise that a coercive device with the purpose and effect of targeting holdouts is arguably justified, whereas a coercive device with the purpose and effect of cramming down a bad transaction is not. The studies compare offers made with exit consents to offers made without. Each study contains a different respective percentage of offers that include exit consents, 28%, (119) 56%, (120) or 33% (121) of the offers include exit consents. The studies show, intuitively, that exit consents increase the number of tenders and the success rate. (122) They compare the two sets of offers on the assumption that exit consents will tend to show up where holdouts are more of a problem, using proxies to indicate holdouts. (123)

2. Business Frictions

A bondholder considering whether to accept an exchange offer worries about more than strategic holdouts. Some offers are bad on their own. A good offer meets two conditions. First, it cuts deeply enough into the bondholders' contract rights to yield a sustainable capital structure, (124) and, second, it must not cut so deeply as to allocate to the common stockholders too much of the surplus arising from the achievement of sustainability. Threading this needle is not easy.

As Brudney points out, a workout can make the bondholders worse off if the issuer limps into bankruptcy anyway, for the old bond would entail a bankruptcy claim in the original face amount. Bond analysts cite post-exchange bankruptcy as a significant risk factor for bondholders taking exchanges. The studies confirm that creditors are less likely to support proposals as distress becomes more severe. (125) Yet, according to the studies, bankruptcy eventually happens about half of the time even when the offer succeeds--depending on the study, in 52%, (126) 59%, (127) or 46% (128) of the cases. And, as Brudney also notes, the bankruptcy rate implies a need to make a corresponding adjustment to the back-of-the-envelope estimate of the deadweight costs of failed workouts, for bankruptcy costs are not opportunity costs to the extent they were going to be incurred in all events. (129) Instead, the only issue is the time differential of when those costs are incurred.

Other business factors also figure into the causation picture. Assets matter--it has been found that a workout is more likely to succeed if a firm's assets are intangible because the value of intangibles tends to erode in bankruptcy. (130) Capital structure complexity also merits attention. The smaller the number of classes of creditors, the more likely a successful restructuring. (131) Financial creditors are not the only pertinent players--a large and uncooperative population of trade creditors can force a company into bankruptcy. (132) The presence of a large bank lender also makes a difference, but the particular effect depends on the case. (133)

B. Workouts Versus Bankruptcy: Comparative Costs

We now turn to a cost comparison between out-of-court restructuring and bankruptcy. This confirms the point that constraints on the out-of-court route imply opportunity costs, but also shows that the cost differential has decreased substantially since Professor Roe problematized section 316(b) in 1987.

Exchange offers, when they do work, yield better cost numbers than do bankruptcies. Bondholder recovery rates are higher in out-of-court restructurings. (134) The deals close relatively quickly and without much additional cost. One study averages the cost of an exchange offer at 0.6% of pre-exchange asset book value, another at 2.5%. (135) In contrast, results of major studies of the direct costs of bankruptcy average out at 6.5% of book value. (136) Indirect costs of bankruptcy, while difficult to measure, are thought to be higher still, at 10% or more. (137) A caveat still should be noted. Indirect cost incurrence tends to be concentrated during the period after distress sets in, but prior to bankruptcy. (138) It follows that some of these costs also befall distressed firms that succeed in closing a workout.

Even with the caveat, the cost and recovery numbers suggest that workouts would predominate over bankruptcies in a perfect world. Real world frictions prevent that, quite apart from bargaining instability and regulatory constraints. If the debtor's problems are due to tort, tax, trade debt, or a complicated capital structure, then exchanges have little to offer. Moreover, if the debtor lacks a viable business model, a liquidation makes more sense than a restructuring, so there is no point to an exchange. Chapter 11 then offers a high-octane federal sales power that facilitates liquidation. (139) Severe liquidity problems also can make bankruptcy the only alternative. (140) Finally, bankruptcy has special advantages for some companies, like disaffirmance of onerous contracts, (141) and the nationwide service of process. (142)

Managers may balk at the suggestion of an early exchange offer because an exchange offer takes a distressed but non-defaulting company across the line to default in the eyes of the credit rating agencies. (143) An exchange offer announcement tends to prompt a drop in the price of the company's common stock, even though economic theory holds that the offeror is trying to unlock a surplus. (144) Apparently, the offer by itself imports negative informational content. The bonds themselves also are likely to lose value in the wake of an announcement, (145) even as they can be expected to increase in value if the offer succeeds. (146) There is also a negative tax result at some companies: the difference between the face value of the old bond and the principal amount for which it is exchanged amounts to cancellation of indebtedness income. (147)

Meanwhile, bankruptcy has gotten cheaper, reducing the policy stakes surrounding section 316(b). Prepac bankruptcy processes, (148) which have proliferated during the past quarter century, ask the court to confirm the amount in a workout approved by at least two-thirds of the amount and over one-half of the number of each impaired class of creditors. The entire prepac process can, in theory, result in plan confirmation in twenty-eight days and a plan going effective in forty-two days. (149) Prepacs are more expensive and time consuming than workouts, but not by all that much. Studies variously find average (median) direct costs of prepacs at 2.8% (2.4%) (150) and 1.8% (1.4%) (151) of prebankruptcy assets, putting them between workouts and full-dress bankruptcies on the cost scale, but much closer to workouts. Prepacs, while slightly more expensive, also offer powers not available in workouts: amendment of core payment terms based on majority consent; redemption of nonredeemable debt; cure and reinstatement of accelerated debts without creditor consent; the ability to get new, superpriority financing; and the asset sale power.

Time is an important factor--cost goes up as debtors linger in Chapter 11. (152) Classic studies from before the turn of this century put the average duration of a bankruptcy reorganization proceeding at 2.5 years, (153) 2.3 years, (154) and 2.2 years. (155) But the most recent set of numbers, covering 1981 to 2013, shows a notable reduction to a median of 1.04 years. Prepacs, which are much quicker, have a lot to do with this, averaging 0.34 years over the same period, while regular bankruptcies take an average of 1.53 years. (156) In 1990, prepacs made up 3.3% of public company bankruptcy filings; in 2012, they made up 33.3%. (157)

Regular bankruptcies, taken by themselves, have also gotten faster. Since 2000, performance terms attached to debtor-in-possession financing agreements (DIP loans) shifted power from the debtor's managers to its creditors. (158) Where once debtors made endless motions for extensions of plan exclusivity, there are now frequently timetables for asset sales or plan confirmation and other performance metrics. (159) We also have more dispositions by sale of going concern assets as compared to conventional reorganization by renegotiated capital structure. (160) In 1990, the average non-prepac took 2.21 years, a duration that shortened to 0.71 years for the class of 2012. (161) As yet, no exhaustive new study of bankruptcy costs covers these recent years. We predict that when such a study is conducted, it will show a continued substantial reduction.

We note one further factor: debt contracting has itself evolved so as to reduce the risk of default. The debt incurred during the private equity wave of 2003-2007 came to maturity without a hint of an insolvency crisis. The terms of the deals had something to do with this--many omit financial covenants and contain "pay-in-kind" (PIK) interest terms allowing distressed managers to postpone cash payments. (162) Efficiency enhancements in Chapter 11 practice also surely played a role. But, contrary to expectations, out-of-court restructurings also figured in.

C. The Shift to Workouts

1. Volume

Restructuring practice shifted abruptly in 2008 due to a temporary absence of financing. It is impossible to reorganize a firm in Chapter 11 without adequate financing for ongoing operations. This financing comes from DIP loans, but the DIP lending market contracted drastically during the credit crunch of 2008--precisely at the moment when demand for financing spiked.

Without a viable bankruptcy option, distressed borrowers turned to exchange offers. (163) The number of distressed exchanges effected in 2008 was double that of any year since 1984, the first year when data are available, and the total amount exchanged ($30.3 billion) exceeded twice the total amount exchanged in all years from 1984-2007. (164) The implicated debt issues comprised 59% of issues in default at the time, compared with 10% of issues in default from 1984-2007. (165) Although DIP financing only disappeared temporarily, the percentage of workouts remained large in 2009 and 2010, at 18.5% and 36%, respectively. (166) For these two years there was a tax explanation--a temporary deferral of taxation of cancellation of indebtedness income until 2014. (167) The tax deferral ended after 2010, and in 2011 exchange defaults constituted a more normal 9.5% of total defaults. (168) In 2012 and 2013, however, the number of restructuring exchange offers picked up again to 21% of issues. (169) Overall, debt exchanges made up 29% of all issuer defaults during 2008-2013, compared to 10% for 1984-2007. (170)

Chapter 11's continued dominance is not surprising. As we have noted, restructuring by exchange makes sense for only a subset of debtors. We are not able to say what percentage of debtors would meet this description. The point is merely that, with around one-fifth of all defaulting public company debtors turning to exchanges and only a subset of the group well-suited to the exchange alternative in the first place, workouts now vie with bankruptcy as a restructuring option within the subset. This is an important change.

2. The Empty Creditor Problem

The workout uptick is doubly surprising in view of a lengthening list of inhibiting frictions. A so-called "empty creditor" problem has spilled over from the proliferation of credit default swaps (CDSs). (171) In a CDS, a protection seller promises to pay a protection buyer a certain sum in the event of a defined credit event on a reference asset in exchange for a periodic fee. When a bondholder purchases CDS protection, the CDS hedges the bondholder's default risk with the protection seller's promise to pay in the event of a default. Significantly, the International Swaps & Derivative Association's (ISDA) Master Agreement defines credit events to include payment defaults and bankruptcies but not voluntary debt exchanges. (172) It follows that a bondholder with swap protection has no incentive to tender into an exchange offer, good or bad. As between a successful restructuring paying, say, 75 cents on the dollar, and a bankruptcy proceeding resulting in the CDS protection seller paying 100 cents (or more (173)) under the swap, the bondholder wants the latter. It follows that, where a substantial subset of a distressed issuer's bondholders are paying for swap protection, an exchange offer with a supermajority tender condition should be impossible to bring about. In fact, an ex ante subset of protected bondholders is not even a necessary prerequisite. Financial punters who do not hold the bonds buy credit protection as a way of betting on a default. Once a distressed issuer puts out an exchange offer, such a "naked" protection buyer has an incentive to buy the bonds in order to prevent their being tendered.

So goes the story, and it makes a great deal of sense. But the supporting evidence was entirely anecdotal until recently. This shortcoming was sharply highlighted by a spokesman for the ISDA, which controls the standard swap contract documentation, determines when a credit event triggering a swap has occurred, and manages the process of settling up. (174) The spokesman also pointed to the upsurge in successful workouts (175) as contrary evidence and set out some arithmetic demonstrating that the naked swap arbitrage play just hypothesized really does not work. (176)

A couple of empirical studies have appeared, reaching conflicting results. One took a sample of distressed exchanges from 2006 to 2011 and compared bondholder participation rates as between those issuers whose debt CDS contracts had been written, and those whose debt was uncovered. The study found a twenty-nine-point reduction in participation in the former group. (177) The other study took a sample from 2008 to 2009, and found that the factors already identified in the literature drove the results (178) with no difference resulting from the availability of swap protection.

The salience of empty creditors in distressed exchanges remains an unresolved question. We do note that none of our empirical findings undercut the claim that empty creditors prevent restructurings, for it is possible that issuers and their advisors screen for them in advance, directing capital structures with extensive CDS coverage to prepac bankruptcy. If, as seems likely, CDS coverage does limit the class of companies suited to out-of-court restructuring, the rise in the numbers of restructurings is doubly impressive.

D. The New Practice

The transactional profile of workouts also has changed. This Section shows that the prevailing picture of dysfunctionality no longer holds, drawing on hand-collected data on workouts commenced since 2010. The data show a notable increase in flexibility and success. We also note a shift in the position of household bondholders: now that issuers rely on the Rule 144A exemption from registration, bondholders with assets under $100 million are entirely excluded from the process.

1. New Data: Distressed Exchange Offers, 2010-2016

None of the empirical surveys of workout practice look solely at the period since the shift of 2008. To fill this gap, we collected data from a search of EDGAR Form 8-K files, using "exchange offer" and "indenture" as required terms and "consent solicitation" as an optional term. (179) The search covered the period from January 1, 2010 to June 30, 2016. We narrowed the results to a group of forty-six exchange offers made by distressed issuers. (180) The offers in the data set all entail either a reduction of principal amount, an extension of time without an increase of interest, or a reduction of interest, and frequently a combination of the three. We do not claim the data set to be complete. With EDGAR as the mode of data collection, distressed issuers owned by private equity partnerships and not reporting publicly are not included.

The terms of the workouts in the data set span a wide range of possibilities: 50% offer new debt, 17.4% offer new debt and cash, 19.6% offer new debt and equity (preferred or common stock), 6.5% offer new debt, equity and cash, and 6.5% offer all equity. Within the subset offering debt only, the new bonds extend duration in 91.3% of the cases, reduce the interest rate in 34.8%, increase the interest rate in 47.8%, and implicate a principal haircut in 56.5%. The average (median) proposed haircut is 43.5% (32.7%). The average (median) number of days between the announcement of the offer and its completion or termination is 65.4 (42). Stated as a portion of a year, the average duration is .18, which compares favorably with prepac bankruptcy's average duration of .34. Workouts are almost twice as fast on average.

The new data also displace the old picture in which the holdout problem always leads to an 8o%-90% minimum tender condition that causes around one-half of the offers to go on to fail. Our data yield a more flexible and successful picture.

Minimum tender requirements take a wide range. The 90% minimum is the modal approach. In this sample, seventeen of forty-six offers (40%) specify a supermajority minimum tender (90% or more in sixteen cases, 85% in one case). In eight cases (17.4%) the minimum drops to either a simple, two-thirds, or three-quarters, majority. One offer split the difference, specifying 90% for a senior issue and a simple majority for junior issues.

Fifteen offers (33.3%) had no minimum tender requirement. There were an additional five offers (10.9%) without a minimum, but as to which either a supermajority of bondholders signed onto a formal support agreement with the issuer or such a high subscription rate (99%) resulted as to implicate a high level of organized support. In other words, no minimum tender condition was needed because the necessary support already had been secured. In the final tally, only one-half of the offers proceeded on condition of 90% supermajority support.

Once the issuer makes the offer, things get even more flexible. Only 40% of the offers closed with subscriptions having met a stated minimum condition. In 17.4% of the cases, a stated minimum was waived in connection with the closing, while 32.6% of the offers were completed with no minimum having been stated. Extensions of time also were common, occurring in twenty-two (47.8%) of the cases, often multiple times. Minimum tender conditions were relaxed as the offer was extended in five cases. Only five of the offers were sweetened.

The new data also dispel the image of failure. Overall, 87.5% of the offers closed. Two of the closings occurred in respect of a prepac bankruptcy solicitation run concomitantly with the out-of-court offer, where the two-thirds of dollar amount tender threshold for the prepac was met but the exchange offer's higher threshold was not met. If those two cases are omitted from the success column, the proportion goes down to 83%. Either way, the success rate is higher than in the previous studies of distressed offers--only the outlier study registering 75% (181) even comes close.

There were six complete failures. They are worth a closer look. Bankruptcy followed the offer's failure in five of the six cases, suggesting that the issuer was too far gone to be an appropriate company for a workout. Indeed, in two cases, the issuers were so close to bankruptcy's door as to make the exchange offer a nonevent, yielding tenders of only 7.2% of the bonds in one case (Dynegy) and only 5.1% in the other case (Colt). As to the other four cases, the tenders came close to the minimum in only one (Angiotech), which involved two issues with minimums of 90% and 98% and tenders of 76% and 85%, respectively. In the first of the other two cases (Penson), subscriptions of 51% and 70% failed against a minimum of 95%; in the second (Goodrich), 62% came in against a minimum of 95%. The implication is that exchange offers fail on the merits, rather than due to decisional instability caused by holdouts, at least in this sample.

Increased coercion accompanies the picture of increased success. Consent solicitations stripping covenants from the old bonds accompanied 82.6% of the offers in our sample, much in excess of the 33% to 46.6% range found in earlier studies. (182) The thing that needs explaining, then, is not why the issuers included exit consents, but why nine issuers omitted them. In four of those nine cases, the explanation is easy. Two were exchanges with no minimum condition subject a cap on the number of new bonds, with proration to occur in the event of oversubscription. In both cases, the maximum number of old bonds to be accommodated fell short of the number of consents necessary to amend the indentures. Two were exchanges concluded pursuant to side deals without a further solicitation of non-signatory bondholders. All of the remaining five offers fit the general profile. Significantly, all of them failed to close, confirming the effectiveness of coercive tactics.

Coercion did not stop with exit consents. There was an "upstream" feature in 59.6% of the offers, with the new bond offering junior secured status in exchange for an unsecured old bond (65.2% of the debt-only offers had this feature). As noted above, secured status has a negative effect on the value of untendered old bonds and discourages holdouts.

There is also timing coercion. Fifty percent of the offers took advantage of Williams Act exclusion of debt exchanges from the operation of the all-holders rule to offer better terms for acceptances received by an early tender date. Typically, this comes in the form of a new bond with a principal amount $20 to $50 higher than on the new bond received for late tenders. There were also five cases of vote-buying in the form of a fee (typically $2.50) paid separately for the exit consent.

Additionally, there is concrete evidence of a factor that, at least nominally, counterbalances coercion--negotiation. Many issuers prenegotiate the offer with large bondholders. A formal "restrcturing support agreement" between the issuer and (usually) a majority of the bondholders preceded the offer in 45.8% of the cases. We suspect that the number of negotiated deals was actually higher. Issuers hire advisors whose job it is to manage relationships with the large bondholders. The fact that no formal agreement is reached does not imply that discussions did not proceed in the ordinary course.

Finally, the new data confirm the older picture in one important respect: eventual bankruptcy remains salient. Across the entire sample, 39% of the issuers eventually went into bankruptcy reorganization or liquidation; 13% were acquired by other companies; and 48% remained as independent operating companies. Although the percentage is lower than in the earlier studies, our sample incudes some recent closing dates. In the subset of workouts commenced before January 1, 2016, 43% went bankrupt. We expect the bankruptcy rate to continue to rise over time: three of the standing companies remain in severe distress as of this writing.

2. Institutional and Household Bondholders

Another change in the workout pattern should be noted. This one concerns the treatment of household bondholders, but not their numbers. With bonds, institutions rose to the fore in the bondholder population more than six decades ago. Figure 1 draws on the Federal Reserve Board's Statistical Release Z.1 to break out the percentage of bonds held by households since 1945. A notable shift to institutions occurred at the beginning of the period--by 1953 household holdings had fallen from 31% to 11%. Since then the household share has risen and fallen with surprising volatility, but has never exceeded an upper 20% (1976 and 1994) and never fallen below a low of 5% (1984). In 2015, the figure was the same 11% as in 1953 (which may itself be surprisingly large to some).

Today the question concerning the bondholder population goes less to institutions versus households than to status under the federal securities laws. Today's issuers rely on the Rule 144A exemption from registration for the securities offered in the exchange, an exemption requiring that only "qualified institutional buyers" (QIBs) be solicited. (183) QIBs are defined as substantial institutions or trust funds that own at least $100 million of securities on behalf of unrelated parties. (184) The offer process starts with the identification and precertification of the QIBs in the bondholder group. (185) Once the offer goes forward, any mom-and-pop bondholders will not even receive it. They are, in effect, written off as statutory holdouts. It is a complete reversal of the old picture of the coerced, small investor. Rather than being coerced, the mom-and-pops are ignored as too small to matter.

E. Explaining the Changes

Clearly, something other than the temporary disappearance of DIP financing in 2008-2009 and the 2009-2010 tax deferral is at work. We offer two explanations, one keyed to the federal securities laws, and the other to the power allocations within Chapter 11.

1. Registration Exemption

The first explanation lies in the fact that there has been a shift in securities law compliance strategy. The contemporary bond workout took shape in the early 1980s, when investment bank Drexel Burnham Lambert promoted exchanges that made use of the registration exemption in 1933 Act section 3(a)(9). (186) Section 3(a)(9), however, places unwelcome constraints on the debtor's investment bankers and other financial advisors. Under section 3(a)(9), financial advisors must be paid for their advice and may not be paid success fees. (187) They are also prohibited from soliciting consents or advising offerees, even though they may negotiate with the bondholders, deliver fairness opinions, and circulate disclosed information. (188) Old school section 3(a)(9) exchange offers were, as a result, less likely to be fully negotiated. (189)

Today, Rule 144A provides a cheaper, more user-friendly alternative. Rule 144A allows for exchanges without constraining the conduct of investment bankers and other advisors engaged by the issuer. The bankers line up the bondholders and work together with law firms and specialist restructuring advisors to draft and negotiate restructuring support agreements. (190) Hands-on management by restructuring specialists is complemented by enhanced price discovery, which facilitates evaluation of the value on offer in the exchange, (191) The result is that the restructuring market has increasingly come to resemble the mergers and acquisitions market, a space in which intermediaries guide managers and investors toward closing. The day of the unnegotiated exchange offer has passed.

2. Creditor Control in Chapter 11

The second explanation lies in the power shift to the creditor side in Chapter 11. Chapter 11 has increasingly become a creditor-controlled process, specifically a process controlled by the senior secured lender, which is often also the DIP financier. (192) Whereas public firms filing for bankruptcy in the 1980s and 1990s often had significant unencumbered assets, most public firms that file for bankruptcy now have few, if any, assets not subject to liens. (193) The lack of unencumbered assets gives the secured creditor a pervasive veto over the bankrupt company's business decisions--even use of cash generated by operations falls into the zone of creditor consent. (194) In addition, an amendment to the Bankruptcy Code in 2005 limited the time during which the debtor has the exclusive right to propose a plan, thereby diminishing one of management's key negotiating advantages: the threat of interminable delay. (195)

Secured creditors usually agree to act as DIP lenders, and DIP financing agreements present a second, even more potent power source. DIP loans come with not only superpriority status, but also highly invasive promises and conditions--inter alia, detailed budgets, timelines, and the right to approve the appointment of a "Chief Restructuring Officer." (196) DIP lenders often use their control to buy the company, either by forcing a sale of the debtor's assets (at which the lender has the advantage of credit bidding) or pushing through a cramdown restructuring in which the lender ends up with a controlling stake in the equity. (197) Such processes effectively sidestep many of the key protections for other, more junior creditor constituencies in Chapter 11, even while complying with the literal terms of the Bankruptcy Code. (198) Chapter 11, then, often becomes a vehicle for the senior secured creditor to carry out a foreclosure sale, either of specific assets or of the entire firm, often to itself. Debtors have little wiggle room, evasive maneuvers having come to naught in the bankruptcy courts. (199)

Increased secured creditor control implies incentive adjustments for the other members of the cast of characters. Certainly, bankruptcy is less attractive than heretofore for the debtor's managers. (200) The same will be the case for its unsecured financial creditors. Their interests often do not favor asset sales and other fast track strategies, making them more amenable to an out-of-court settlement, even a flawed one, so long as it provides a significant reduction in bankruptcy costs and averts the possibility of a settlement skewed to the interests of the secured class. This does not imply that an out-of-court composition somehow imports unilateral power to evade the secured creditor and injure its interests; (201) it merely shows that the other players see things differently. Where bankruptcy once offered a comfort zone, today's managers have every reason to avoid it. Meanwhile, bondholder calculations will have shifted in the direction of cooperation.

Indeed, the presence of dominant secured creditors is directly evidenced by the proliferation of second liens in the new bonds in our data sample. Restructuring support agreements are ubiquitous in bankruptcy as well as in workouts. They set out constraints and timetables similar to those in DIP financing agreements, but go further, lining up support for a stated outcome. (202) Nothing stops an issuer and a secured creditor from beginning the negotiating process before a bankruptcy filing. An exchange offer can figure in at this point, posing a cost-effective alternative to a prepac filing. What matters is less the venue of composition than the negotiating framework, which proceeds in much the same mode whether the context is a workout, a prepac, or a full-dress Chapter 11.

These, then, are the new bond workouts: exchange offers made to QIBs as an incident of the new politics of Chapter 11 bankruptcy.

F. Summary

Our review of restructuring practice has important implications for the policy discussion directed to section 316(b). The recent developments undercut the polar positions of both Professors Roe and Brudney.

As to Roe, any opportunity costs stemming from distortions related to section 316(b)'s restructuring barrier have been ameliorated substantially. Holdouts no longer systematically prevent deals from closing. This is not just a matter of across-the-board issuer coercion. Many offers proceed successfully without a 90% tender condition. When substantial tenders come in but do not reach 90%, the issuer tends to waive the condition and close anyway. If bondholder retribution in the wake of such bait-and-switch maneuvers were a problem, issuers would be forced to make minimum tender conditions unwaivable so as to make their offers credible. In cases where the out-of-court route is unfeasible or undesirable, bankruptcy affords a faster, cheaper alternative than before. To the extent that section 316(b) has a potential to do harm, the markets have figured out how to minimize it.

The problems that concerned Brudney203 also have lost salience. Mom-and-pop bondholders already had largely disappeared from the holding group at the time he wrote. (204) Any disabling coordination problems on the part of bondholding institutions are much diminished as well, to the extent they exist at all.

To say that the markets can live with section 316(b) does not also say that they would be better off with it than without it. Even as the force has gone out of the case for outright repeal, it also has gone out of the case for retention. Coercive tactics are more salient than ever. There are also new factors in the policy calculus. To the extent that secured creditor control (along with restructuring support agreements that follow from it) is skewing bankruptcy into undesirable directions, the new workouts' negotiated aspect, rather than being an ameliorating factor, becomes subsumed in a larger problem. But that is a problem for bankruptcy courts to address.


The new workout is a hardball game. Dissatisfied bondholders have pushed back, persuading courts in the Southern District of New York to adopt a reading of TIA section 316(b) that extends its prohibitive reach beyond direct amendment of payment terms to any change effected in a restructuring negatively impacting the value of the bonds. This broad reading of section 316(b) disrupted practice assumptions going back three-quarters of a century and triggered more litigation. Uncertainty resulted for bond counsel regarding opinion letters on the validity and enforceability of trust indentures, prompting representatives of twenty-eight prominent law firms to issue a joint interpretive statement. (205) Industry representatives even slipped a clause retroactively overruling the cases into an (unenacted) appropriations bill. (206) But the status quo finally was restored when, in January 2017, a panel of the Second Circuit reinstated the traditional reading of section 316(b) in a 2-1 decision over a strong dissent. (207) The ruling consigns the broad reading to history unless a contrary decision in another circuit keeps it alive.

The sturm und drang surrounding the broad reading's rise and fall awakens the dormant policy question regarding repeal of section 316(b). The broad reading threw a wrench into today's workout machinery, making section 316(b) more counterproductive than ever. At the same time, the facts of the cases that inspired the new interpretation cast an unflattering light on the distortionary tactics that drive the new restructuring, suggesting a need for a revised approach. The issue respecting repeal can be expected to persist even assuming, as seems likely, that the Second Circuit panel opinion definitively reestablishes the traditional reading. The repeal question returns in a policy context very different from that prevailing at the time Professors Roe and Brudney debated the matter decades ago. As we saw in Part III, bankruptcy is less costly, the dysfunctional restructuring is a thing of the past, and negotiated workouts now present a viable alternative. There may be power imbalances on the fact patterns, but bondholder collective action problems no longer contribute to them. Now that the broad reading is off the table, it would seem that issuers and bondholders can continue to live with section 316(b) without excess discomfiture. But, at the same time, coercive tactics have assumed greater salience. Maybe it is time to put CACs back on the process table.

The literature describes two contrasting means to that end--qualified repeal with continuing SEC control and outright repeal that leaves the future process regime in the hands of indenture drafters. The primary point of difference lies in the observer's view of the efficacy of bond contracting.

Professor Roe makes the case for qualified repeal. He projects that contractual adjustment will be slow and rigid, with issuers resisting CACs due to worries about negative bond market signals. (208) In addition, self-interested bondholder voting will remain a problem under CAC indentures--separate lending relationships between issuers and bondholders will result in side payments. (209) He accordingly suggests that the SEC exercise its exemptive power under TIA section 304210 to allow indentures to include two-thirds majority CACs on the condition that the provisions bar coercive transactions and disqualify conflicted votes. (211)

Professor Marcel Kahan counters with a case for outright repeal. (212) For Kahan, there is no justification for any mandated terms in bond contracts. Between good information diffusion, (213) robust pricing, (214) and almost universal institutional holding, bondholders can protect themselves. (215) He does not deny that coercive devices play an important role in workouts, but takes the position that coercion by itself does not justify a preclusive mandate such as that embodied in section 316(b). (216)

This Part explores these alternatives, taking a position in the middle. That is, we favor outright repeal but enter a doctrinal caveat--outright repeal would work best against a background threat of judicial intervention. Standard trust indentures do address process matters related to amendment and exchange offers. Indenture drafters also adjust indenture forms in response to external shocks. But the process is indeed slow and the results are variable--some indentures do a better job of addressing process contingencies than do others. Meanwhile, depending on the drafting, a CAC indenture can expand the list of coercive possibilities. We think the drafters will be more likely to focus on the process issues and treatment alternatives with a judicial officer on patrol.

Section A takes a brief look at the broad reading of section 316(b), laying out the facts of the leading cases and our critical analysis of the reading's application. We also take note of its merits, describing circumstances where its application could be defended as commonsense policing.

We turn to the repeal question in Section B. We explain our doubts about further SEC involvement in workouts and go on to show that trust indenture drafters are more responsive to the stresses of restructuring than some observers have conceded heretofore. Indentures contain an elaborate set of process rules operating against the background of section 316(b). We surveyed the terms of two sets of indentures governing bonds issued under 144A, one set issued in 2011 and the second in 2016. The findings are surprising: first, UACs cover considerably wider ground than required under section 316(b), and, second, a majority of the contracts explicitly sanction exit consents. We also show that indenture drafters adjust to background shocks, in this case, the broad reading--a subset of the 2016 indentures draft out from under section 316(b). That said, Section C goes on to predict that process problems will persist in the wake of repeal. To achieve an even playing field, one needs indentures containing completely drafted process rules. Our dataset does not support a prediction that the drafters will respond comprehensively to the process questions that would arise in the wake of section 316(b)'s repeal.

Section D lays out our recommendation. The contractual good faith duty, as we have seen, only rarely comes to bear against bond issuers, amounting to considerably less than a threat. Given this, we project that in the event section 316(b) repeal resulted in anything short of comprehensive revision of trust indentures, that CACs could be combined with exchange offers and exit consents with devastatingly coercive effect. Further, given repeal, exit consents would lose the expedient attractiveness they currently enjoy as a weapon against holdouts. We project that pressure for judicial intervention would build up accordingly, and describe a robust basis for justifying it--a revived intercreditor duty of good faith.

A. The Rise and Fall of the Broad Reading of Section 316(b)

In a 1999 case called Federated Strategic Income Fund v. Mechala Grp. Jamaica Ltd., (217) a court in the Southern District of New York disregarded the traditional reading of TIA section 316(b), substituting a broader interpretation that considerably extended the section's prohibitive scope. (218) Under the traditional understanding, the reader takes the statutory language--"the right... to receive payment of the principal of and interest on such... security... shall not be impaired or affected without the consent" of the holder--interpolates the contract's promise to pay as the "right," and interprets "impair" and "affect" to refer to amendment or waiver of the promise. (219) It follows that section 316(b) only applies to amendments and waivers of payment terms. A thin case law confirmed the reading,220 which guided practice in financial markets and law firms. Under the broad reading, the impairment of the right to be paid encompasses any unconsented action under the trust indenture undertaken in connection with a restructuring that compromises the issuer's ability to pay the bonds, even if the bond is not amended. "Impair" and "affect" are, under this interpretation, any unconsented changes that make it less likely that a bondholder will be repaid.

No one paid much attention to Mechala until the Southern District of New York decided a trio of workout cases in 2014 and 2015: Marblegate Asset Mgmt. v. Educ. Mgmt. Corp., (221) Meehancombs Glob. Credit Opportunities Funds, LP v. Caesars Entm't Corp. (Caesars 7), (222) and BOKF, N.A. v. Caesars Entm't Corp. (Caesars II). (223) The court found that all three restructurings potentially violated the TIA, relying on the broad reading. Then, in January 2017, a panel of the Second Circuit reversed the Marblegate ruling. (224) The reversal consigns the broad reading to history unless continued litigation in the case or a contrary decision in another circuit keeps it alive. Whatever the outcome, the distorted playing field in out-of-court restructurings returns to the forefront as an unresolved policy problem.

The broad reading implicated two issues. The first concerned the TIA and its legislative history. The Southern District drew purposive inferences from the legislative history (225) in support of the broad reading, while the Second Circuit's reinstatement of the narrow reading followed from a closer look at the historical evidence. (226) The appellate court had much the better of the argument: there is specific evidence that Congress intended to bar only unconsented amendment of payment terms. The second issue concerned the substantive merits: whether the broad reading improved workout practice by ushering in a constructive regime of judicial policing. The courts of the Southern District clearly thought so. The Second Circuit, however, expressed no views on the matter, reinstating the traditional reading solely by reference to the legislative history. Our own view is mixed, but ultimately negative. The broad reading did facilitate intervention against egregious overreaching in workouts. But it also overreached, disrupting settled allocations of risk without targeting the problem of distorted choice at its source.

To see the problems, one only needs to look at Marblegate's facts. The case concerned Education Management Corporation (EDMC), a distressed for-profit education provider that could not file for Chapter 11 without losing its eligibility for federal student aid revenue, which would have precluded any reorganization attempt. (227) The company's operating subsidiary had a $1 billion secured bank loan and around $200 million of unsecured notes issued under a TIA indenture, both issues guaranteed by the parent. (228) Significantly, the parent's guaranty of the unsecured notes was a "tag along," which meant that the guaranty had been created in connection with the bank loan and could be terminated at the bank's option. (229) EDMC, the bank, and 81% of the noteholders agreed to a negotiated composition that allocated the bank 55 cents on the dollar, the unsecured notes 33 cents on the dollar, and for all intents and purposes wiped out the common stock.230 But, because the deal implicated the indenture's payment terms and bankruptcy was not an option, unanimous consent was necessary. (231) A single hedge fund holdout blocked the deal. The dealmakers accordingly resorted to a backup plan, under which the bank foreclosed on the subsidiary's assets and lifted the parent guaranty, and then consented to the transfer of the assets to a new, unencumbered subsidiary in exchange for the same deal that the holdout earlier had rejected. (232) The difference was that this time the holdout had no choice in the matter and, as a claimant against a subsidiary (the assets of which had been stripped), was left with nothing. (233)

Undaunted, the holdout sought an injunction, which, had it been granted, would have brought down the company. (234) The Southern District refused to grant the injunction but found in favor of the holdout on the likelihood of success on the merits. (235) The transaction neither amended nor waived any term of the note or the indenture and so did not create an issue under the traditional reading of section 316(b). But the Court invoked the broad reading to find a violation: the minority bondholder had been "disinherited" without its consent in connection with a "restructuring." (236) The holdout, said the court, had been offered a choice between 33 cents on the dollar and zero, which is no choice at all, justifying an override of the terms of the contract under section 316(b). (237) It did not matter to the court that the guaranty, as a tag-along, was a contingent right whose contingency had failed, and had been judicially reconferred on the bondholders ex post.

Ironically, there was nothing normatively troubling about the unconsented cramdown. The term overridden was not "just boilerplate." We surveyed 109 trust indentures (49 executed in 2011 before Marblegate, and 60 executed thereafter in 2016) and found that most (82%), but not all, of the indentures containing guaranties (76%) took steps to include the guaranty in their UACs, even though the TIA does not necessarily require inclusion. (238) The inclusion makes sense--guaranties can figure critically in the structure of promises creating a given issue's borrowing base. At the same time, 98% of the guaranties had release provisions and accordingly contingent status, even as most of them could not be amended. Only 28% of the guaranties contained a tag-along condition, implying specific focus, deal by deal. The override of guaranty release in Marblegate raised a serious policy question accordingly.

Furthermore, there was no statutory basis for justifying the contract override on the facts of the case. Even if Congress, back in 1939, had expressed an intent to force workouts into bankruptcy, bankruptcy was not a viable alternative for this case's debtor, EDMC. (239) Nor did the process employed in the case raise any red flags. The deal, based on negotiation, had garnered supermajority bondholder support, and the cramdown foreclosure was resorted to only as a last ditch move against a holdout. (240) From a policy perspective, there was nothing about which to object.

The broad reading, in sum, was overly broad and fact insensitive. But, at the same time, it had its merits when deployed in the right case, as Caesars I and II demonstrate. Caesars involved aggressive debt-equity management in the wake of a private equity buyout. Caesars I concerned pre-buyout bonds issued by an operating subsidiary and guarantied by the parent. (241) Following losses, the parent started an asset-stripping program, moving assets from the operating subsidiary to a new unencumbered subsidiary. (242) The indenture erected two barriers to the program, first the parent guaranty, and, second, a successor obligor clause triggered by sales of substantially all assets. (243) But both terms were susceptible to majority amendment. (244) The parent divided the bondholders into a 51% in-group with which it apparently had other lending relationships, and a 49% out-group. With the bonds trading for 48 cents on the dollar, the parent redeemed those held by the in-group for 100 cents on the dollar, collecting exit consents from the in-group that lifted the guaranty and reset the successor obligor clause so that the subsidiary's denuded asset base now comprised all of its assets within its meaning. (245) The out-group sued, and won easily under the broad reading--the guaranty removal amounted to an unconsented restructuring. (246)

Caesars II concerned the operating subsidiary's post-buyout bonds, also guaranteed by the parent. This guaranty was conditioned on the subsidiary being "wholly-owned." (247) Looking to get the guaranty lifted, the subsidiary issued its own near-worthless common stock, first, to a friendly bank lending group, (248) and, second, to its own directors and officers under a ginned-up equity compensation plan. (249) When this second group of bondholders sued under 316(b), the company took the position that no "restructuring" within the meaning of the broad reading had occurred--the subsidiary was solvent and no obligations were being scaled down. (250) The court held that impairment for 316(b) purposes still could obtain and referred the matter over for fact finding on the question whether this really was a restructuring. (251)

Caesars showcases the good side of the broad reading, for between the Caesars I vote-buying and the Caesars II strategic equity placements, there was ample cause for judicial intervention. The demonstration of fit is unsurprising, for we have seen in this Article's Parts I, II, and III that distorted consent-giving in workouts continues to be a cognizable problem, a problem that remains unaddressed by conventional policing tools due to the courts' longstanding habit of refusing to apply contract good faith to financial contracts.

While Marblegate left unresolved some questions about the border between core and non-core bondholder rights (e.g., is collateral part of the right to payment?), it made clear that section 316(b) is to be read narrowly. Even so, section 316(b) still presents an obstacle for workouts, leading to calls for its repeal, as discussed in the next section.

B. The Case for Outright Repeal of Section 316(b)

1. Federal Mandate

Professor Roe would condition repeal on the institution of a protective layer of federal law, proposing that the SEC by rule promulgate a broad standard directed to any and all distortionary influences. As a theoretical proposition, the suggestion has everything to recommend it, for the suggestion it amounts to a mandate embodying the policy bottom line described in Part I. But we worry about implementation. Marblegate shows that single-minded pursuit of undistorted choice can have perverse effects. Although existing law is far from perfect, its overall pattern is instructive--rules, not standards, and reticence respecting the advisability of pushing toward the theoretical ideal in arm's length contexts. Conflicts are hard to smoke out, and it would take an invasive disclosure regime to get them on the table. And then there is the flood of suits. An SEC rule would presumably fall within the existing private right of action under the TIA.252 As such, it would not be subject to the lawsuit baffler that is universal in trust indentures--a "no-action provision" requiring that a group holding at least 25% of the bonds coalesce to pursue a contract claim when the trustee declines to do so. (253)

We also question whether it makes sense to leave the process regime's terms up to the SEC. The agency has been out of the bondholder protection business for decades and has a lot of other things on its plate. Once the SEC puts through a 66.67% CAC alternative, one doubts that it would monitor investor preferences regarding further particulars on a going concern basis even as experience accumulated in the marketplace.

Finally, and most importantly, the proposal follows from theory and makes no reference to the preferences of bondholders. We will see below that it is not safe to assume that they would agree to a ban on issuer coercion, or indeed, that they would prefer a CAC regime in the first place.

2. Contracting Practice

Trust indentures are famously unresponsive to change. This is partly because the TIA locks in a number of their provisions. Even outside of the TIA's purview, standardization (wherein lies a part of the value of these contracts) inhibits innovation. For policy purposes, trust indentures tend to be written off as expressions of bondholder preferences.

We think that is a mistake, at least with regard to the practice of respecting indenture terms governing amendments and waivers. To support our point, we collected a sample of indentures governing new issues of bonds issued during the second quarter of 2011, well before the Marblegate decision, and the second quarter of 2016, after understanding of Marblegate and its implications had a chance to diffuse through the bond market. Our sample is restricted to offerings pursuant to the 144A registration exemption because 144A indentures do not need to be qualified under the TIA. Some issuers scrupulously conform to the TIA nonetheless--it is common in the 144A market to give purchasers a concomitant option to exchange their 144A indentures for registered bonds offering bondholders ex post TIA qualification for their indentures. Other issues are "144A for life" and, once their transfer restrictions lift after six months, trade in a QIB market. The territory thus is open to heterogeneous responses to developments under the TIA--nothing prevents a drafter of a "144A for life" issue from including an across-the-board CAC. Our sample is culled from EDGAR's Form 8-K files, pursuant to a request requiring "144A," "indenture," and "notes" and the relevant item and exhibit designations. (254) The 2011 sample includes 49 indentures and the 2016 sample includes 59.

Figure 2 describes basic terms of the transactions in the data set. In 2011, two-thirds of the deals included registration rights, a percentage that declined to 36% in 2016, reflecting the diminishing relevance of trading restrictions in the 144A market. Most of the bonds were guarantied by other entities in the issuer's corporate group. Most of the bonds also were straight senior debt, but there were large subsets of secured and convertible bonds along with a smaller subset of subordinated issues.
Figure 2: Core Terms of Rule 144A Indentures

              2011  2016

rights        67%   36%
Guarantied    69%   81%
Secured       10%   20%
Subordinated   8%    7%
Convertible   25%   17%

Note: Table made from bar graph.

a. Drafting Out from Under Marblegate

Bond counsel reacted negatively to Marblegate and responded in "144A for life" issues by contracting out from under section 316(b), changing a drafting pattern dating back at least to the TIA.

Figure 3 depicts the appearance of amendment provisions. In 2011, 100% of the issues contained a classic trust indenture UAC operating as a proviso to a CAC. The CAC permits amendment of any terms by a simple majority but is qualified by a limitation prohibiting changes to core terms without the consent of each bondholder affected thereby. (255) In 2016, the classic UAC proportion declined to 93.2%. Four of the indentures in the 2016 group drop the unanimous consent requirement for core terms and substitute supermajority provisions (90% in three, 75% in one). The shift suggests resistance to Marblegate in "144A for life" issues, with the choice of a 90% CAC in three out of four further suggesting that the relaxation of the unanimity requirement specifically targets holdouts.
Figure 3: Core Terms of Rule 144A Indentures

                2011  2016

UAC             100%   93%
extension        92%   88%
Exit consent
permission       67%   79%
316 language    100%   81%
TIA controls     88%   42%
blocked                27%

Note: Table made from bar graph.

The implication of resistance becomes stronger when we look at section 316(b) clauses. Trust indentures customarily include a term that repeats the language of section 316(b) in addition to including an explicit UAC. In 2011, 100% of the indentures contained a provision that replicated section 316(b) without substantive variation. In 2016, that number had dropped to 81%.

The section 316(b) clause has become the platform for getting out from under the broad reading, and is utilized as such in 17% of the 2016 indentures. It is a less drastic approach than that taken in the four CAC indentures. The drafter rephrases the section 316 clause to narrow the scope of the "right." Thus do we see a prohibition of impairment of "the contractual right to bring suit." (256) This indicates an intent to block broader readings emanating not from the contract but from the TIA. The "contractual" right is further defined as a right "to bring suit" and a right against application of unconsented amendment, more concretely blocking interpolation of a transcendent right to be paid. Another drafter simply states an affirmative "right to bring suit" for principal and interest, (257) eliminating the separate right "to receive payment" entirely.

Eight of the indentures proceed this way. Another three leave the section 316(b) clause out entirely. An additional four contain a clause that blocks application of the TIA even as they also contain a classic section 316(b) clause. (258) The apparent objective is to block the broad reading of the clause by blocking application of the statute. By one means or another, 23.7% of the indentures block the broad reading.

There is also a more drastic decline in inclusion of a standard backstop clause specifying that the TIA controls the indenture whatever else the indenture says--down from 88% of the 2011 indentures to 42% of the 2016 indentures.

b. Unanimous Action Expanded & Exit Consents Permitted

The belt-and-suspenders drafting approach--explicit UAC plus 316(b) clause--has important implications for section 316(b) repeal. If repeal occurred tomorrow, the 316(b) clauses would still be in the indentures. Even if repeal had the effect of removing 316(b) clauses from existing bonds, the separate UACs would remain, thereby limiting the legislative shift to majority consent to future indentures.

Three additional drafting patterns should be noted, these appearing in common in both 2011 and 2016. The drafters effectively endorse the current framework of restructuring in three ways: they extend the UAC's coverage beyond the mandate of 316(b), leave the section 316(a) mandate as is, and sanction exit consents.

The UAC extensions go beyond the traditional core payment terms to pick up redemption terms, guaranties, conversion provisions, and subordination language. (259) Amendment terms on liens split the difference, permitting security to be released on a two-thirds or three-quarters vote. Further extensions appear frequently but not consistently--48.6% of the indentures extend the UAC to "priority" and 21.1% pick up their poison put provision. In effect, the drafters of these indentures are broadening the set of "core" terms. (260)

Movement toward bondholder protection is not thorough-going, however. The drafters might also have extended the reach of TIA section 316(a), which disqualifies votes of bonds held by the issuer and its affiliates. (261) The indentures in our sample track 316(a) closely, where they might have defined additional classes of prohibited self-interest. For example, consent payments also could be barred in specified situations, and empty voting could be addressed. This is probably a considered result, for a more expansive reading of section 316(a) directed to exit consents has been mooted in the law journals (262) and also litigated without much success. (263)

Most indentures also sanction exit consents, confounding the expectations of academics. Professors Coffee and Klein, writing in the wake of Katz v. Oak Industries, suggested that all it would take is a little nudge and indenture drafters would wake up and shut down exit consents by requiring that amendments and exchange offer tenders be unbundled. (264) But the opposite occurred. Express sanction of amendments by exiting bondholders is widespread without being ubiquitous--67.3% of the 2011 indentures and 79.6% of the 2016 indentures include them.

The bond issuers' interest in assuring the validity of exit consents is clear enough. Significantly, it concerns more than the reservation of freedom of action in distress situations. The primary value of an exit consent lies on the upside, when the issuer of a noncallable bond seeks to pay it down early. The means to that end is a premium cash repurchase tender offer. An exit consent tie in gives the offer a kick. The result is additional flexibility for the issuer, which gets to shave a few basis points off the interest rate on the bond in exchange for it being noncallable, while simultaneously keeping open a paydown option. (265)

Bondholder acquiescence presents a bit of a puzzle, for bondholders are supposed to be the victims here. Why would they sanction their own coercion? Indeed, the preferences suggested appear contradictory--in favor of both unanimous action (retarding workouts) and coercive exchange offers (encouraging workouts).

Here is our attempt to make the pattern intelligible. The bondholders do not trust one another to vote sincerely on amendments of core terms and are jealous of their holdout privilege, hence the UAC expansion. At the same time, they understand that holding out can lead to problems and want to give issuers a reasonable shot at an out-of-court restructuring. Restating, "Make me an offer and I'll look at it, but don't try to tell me what to do." In addition, bondholders are happy to pick up a little added yield when an issuer makes a premium cash tender offer and have no trouble handing an issuer a stick to help such a deal along.

If that is a fair characterization, follow-up questions arise: can we go a step farther and conclude that bondholders prefer unanimous action; that they are unconcerned about coercive issuer tactics; and that, left to their own devices, they would leave workout practice as is? We can suggest these possibilities but cannot go farther. We do not know how large of a role path dependency plays in bond indenture drafting.

It also can be projected that, given a blank slate, bondholders would herd to CACs. The UAC versus CAC question has been exhaustively discussed in the sovereign bond market since the spectacular Argentine default of 2001. (266) Historically, sovereign debt issues effected in New York were covered by UACs (despite the inapplicability of the TIA to sovereign bond issues), while issues effected in London were governed by CACs. (267) Argentina had a significant amount of debt outstanding under UACs. (268) Default led to a fractious workout by exchange offer. Related litigation finally settled in 2016. (269) The public side of the international financial community (which has never been able to come up with a sovereign bankruptcy regime) blamed the mess on the UACs. CACs became part of a policy push for orderly restructuring. (270) Sovereign borrowers were pressured to insist on CACs, (271) and a widespread shift of preferences followed. (272) Two points emerge: first, bondholder preferences regarding CACs and UACs are mutable, and, second, boilerplate amendment terms can indeed change in response to events.

It should be noted that our data also suggest a counterprojection. Caesars and Marblegate did not precipitate a widespread shift to CACs, implying that there is no pent-up demand for collective action. If there were, it is not implausible to suggest that a cascade would have occurred in the 144A market in Marblegate's wake.

3. Conclusion

We see, then, that present arrangements could reflect considered preferences and would be replicated unchanged in future contracts. It is also possible that preferences would adjust if Congress opened the door for CACs, and that the shock might break the drafting pattern in other respects as well. There is no certain prediction. What we can say is that bond contracting is sensitive to workout process questions and does register the preferences of issuers and bondholders, and that as between bond contract drafters and SEC staffers, the former are better equipped to accomplish the task.

C. CACs--the Worst Case

1. The Drafting Task in the Wake of Repeal

Outright repeal of section 316(b) would clean the slate, raising a series of questions for indenture drafters. First comes the CAC versus UAC choice. A choice for CAC raises a subsidiary question about acceptance percentages, not only the number set for payment terms, but whether to incorporate percentage differentials for different categories of contract term. There would also be a critical additional question: whether to bar exit consents; indeed, whether to bar exchange offers altogether. Arguably, with a 66.67% or 75% CAC, exchange offers no longer would serve a legitimate purpose in workout contexts. A prohibition would close off a whole avenue of coercive possibilities. Finally, conflicted bondholder voting and vote-buying (whether an offer to all holders or selected holders) could and should be reconsidered. (273)

The adjustment process could be slow and messy. The Marblegate drafting adjustment presents a case in point. If we assume, as seems sensible, that preferences lie against the broad reading, the fact that 37.5% of the 144A for life indentures in the sample opted out one year to eighteen months after the decision does not bespeak highly focused drafting. Why not all of them? Perhaps preferences differ from issue to issue. More likely, there is both inattention and considered avoidance of change (also known as path dependence). Alternatively, let us assume that bondholders generally are happy to sanction exit consents, an assumption that seems reasonable in view of the fact that 68% and 80% of the indentures in the samples did just that. The same question arises.

The lack of focus follows from the deals' underwritten character. The bondholders are represented at the drafting table only by a proxy--counsel for the underwriter--whose client is interested in bondholder protection only to the extent that it imports marketability. There is no marching order to assure that the lender's perspective informs all fine points, as would be the case in a negotiated bank loan. Risk allocation conventions evolve, but with incomplete coverage. Imperfections creep in.

2. The Worst Case

Let us imagine that section 316(b) is repealed, and indenture drafters respond by inserting 75% CACs, but do not follow up by blocking exchange offers and exit consents. Instead, they follow the practice that evolved with European sovereign bond issues and raise the threshold percentage for covenant amendments from 51% to 75%. The operative notion is that the increase in the threshold makes exit consents less user-friendly for the issuer, making direct amendment more attractive than an exchange offer.

This sets the stage for the worst case scenario, illustrated by a decision that arose under the law of England and Wales. Assenagon Asset Mgmt. S.A. v. Irish Bank Resolution Corp. (274) involved an exchange offer featuring new bonds with a face value of [euro]20 and an exit consent that amended the principal amount of the old bonds from [euro]100 to [euro]0.01. (275) The issuer was an insolvent bank subjected to a government takeover, and the offer derived from a government policy decision. There was disputed evidence that the bonds had been trading for [euro]20, but no evidence of negotiation. Presumably, none was needed, given the high-octane coercive possibilities that result when a drafter combines an across-the-board CAC without simultaneously barring exchange offers and exit consents. 92% of the bondholders knuckled under.

There is a notable similarity to Marblegate, the choice between [euro]20 and [euro]0.01 being no different from the choice between $33 and $0. But the cases are otherwise sharply distinguishable. Marblegate was a negotiated deal amongst creditors who had nowhere else to go and acute vulnerability to holdouts due to a UAC. Assenagon was a take-it-or-leave-it offer embodying the issuer's notion of an appropriate price where a CAC otherwise would have facilitated amendment without coercion.

The moral of the story is that a CAC regime, even as it nominally serves the purpose of reducing distortion, opens up wildly coercive possibilities unless the drafting is thorough-going. Current drafting practice promises responsiveness but not completeness, inserting an element of risk.

D. Backstop Policing: An Intercreditor Duty of Good Faith

Given repeal of section 316(b) and substitution of CACs, exit consents would lose their expedient attractiveness and emerge as an instrument of pure coercion. Absent a contractual prohibition, the judiciary would be called on to intervene. In this section we show how that can be done, looking to the intercreditor duty of good faith.

The intercreditor duty responds to a good faith claim asserted by one bondholder in response to an action taken by other bondholders. The posture is different from the limited good faith duty described in Part II, which involves claims asserted in suits by bondholders against issuers. The duty is articulated in a largely forgotten line of cases antedating the TLA, cases showing strong grounds for a robust implied duty in workout contexts. Not coincidentally, this duty led to an outcome in the plaintiff's favor in Assenagon. It also would have provided an alternative basis for blocking the Caesars I amendment, an approach which would not also have implied invalidation for the Marblegate restructuring.

An intercreditor duty would do more than just provide a basis for the right result in an extreme case. It would also improve the drafting environment. Most observers assume that the burden to draft an explicit contract provision should fall on the party asserting the right, on the sensible ground that this is the party with the correct incentive and possibly superior information. Trust indentures are different, however, because the deals are underwritten, meaning that the bondholders themselves are not represented in the drafting. Instead, to the extent their interests are represented, it is by the underwriters' counsel, but the underwriters' interests are limited to the initial marketability of the bonds and do not perfectly align with those of the bondholders, particularly if the bonds were originally investment grade, such that a restructuring was not anticipated. Incentives on the bondholder side are diffuse, and influence is indirect. To get a contract-forcing result, it makes much more sense to put the burden on the issuer, at least so far as concerns process terms in the wake of section 316(b)'s repeal.

1. Intercreditor Good Faith Cases

A bondholder duty constraining self-interested unilateral or majority action not in the best interests of the bondholders as a group can be seen in a pair of nineteenth-century Supreme Court cases. In the 1874 case of Jackson v. Ludeling, the Court was faced with the problem of an individual bondholder who foreclosed on the collateral securing the bond and artificially depressed the price in the foreclosure sale so that his associate could purchase a $2 million railroad for $50,000. (276) The Court noted that the foreclosing bondholder was "not a partner with [the other bondholders], nor strictly a tenant in common, but the relation into which he introduced himself by his purchase [of the bonds] imposed on him some duties." (277) According to the Court, it "was a duty which he owed to the other bondholders not to destroy [the mortgage's] value. When two or more persons have a common interest in a security, equity will not allow one... to impair its worth to the others. Community of interest involves mutual obligation." (278) The Court commented similarly in Shaw v. Railroad Co.:
If there are differences of opinion among the bondholders as to what
their interests require, it is not improper that [the indenture
trustee] should be governed by the voice of the majority, acting in
good faith and without collusion, if what they ask is not inconsistent
with the provisions of his trust. (279)

An analogous approach can be found in an 1896 Second Circuit Court of Appeals case, Hackettstown Nat'l Bank v. D.G. Yuengling Brewing Co. (280) In Hackettstown, the bond issuer--the famous brewery--had defaulted on its interest payments. (281) The company's principal shareholder arranged for a friendly party to buy enough of the bonds to amass (together with bonds held by other Yuengling family members) the three-quarters majority necessary to amend the bonds' indenture to defer the interest payments. (282) The principal shareholder personally agreed to repay the friendly party's investment with interest. (283) A minority bondholder sued over the indenture amendment. The Second Circuit nullified the amendment, noting that a vote "made collusively ... for the purpose of defeating the remedy of the minority, and not in the exercise of an honest discretion in the general interest, is not a consent within the meaning of the indenture." (284) The court's ruling was based on a "community of interest" that "creates mutual obligation, and imposes upon all persons occupying that position the duty of acting in good faith toward the interests of their associates." (285)

In the nineteenth century, then, majority rule was always subject to a requirement of good faith and a prohibition on collusion with the debtor. It is important to note that the courts did not think of this as an imposition of fiduciary duty, even as they analogized to partners. It was a situation-specific duty, springing not from the overall relationship, but from the vested power to vote or enforce. While a fiduciary-like duty of self-abnegation did result, it was situational and without further implications for the relationship. (286)

The good faith constraint on majority bondholder voting power retains vitality in English case law, perhaps because there is no statutory equivalent to the TIA and bonds are governed by across-the-board CACs. (287) Assenagon is an exemplar of its application. A nonexchanging minority bondholder, whose [euro]17 million were forcibly redeemed for a mere [euro]170, brought suit challenging the validity of the majority's exit consents. (288)

The High Court of Justice applied the intercreditor duty:
The exit consent is, quite simply a coercive threat... [the] only
function [of which] is the intimidation of a potential minority....
This form of coercion is ... entirely at variance with the purposes for
which majorities in a class are given power to bind minorities....
[O]ppression of a minority is of the essence of exit consents of this
kind, and it is precisely that at which the principles restraining the
abusive exercise of powers to bind minorities are aimed. (289)

Assenagon is cut from the same cloth as the nineteenth-century American intercreditor cases, but carries them a step into new territory. In those old cases, the voting infirmity arose in respect of bondholder conflicts of interest, independent of issuer action. In Assenagon, the conflict stemmed from bondholder responses to issuer coercion. Indeed, it delivered the good faith constraint at the issuer's doorstep by indirection, nominally in conflict with the result in Katz v. Oak Industries. The High Court acknowledged the point, noting that the exit consent is an invitation from the issuer for the "majority to levy against the minority." (290) But the duty under consideration was that of a bondholder, not an issuer, so "it is no answer for [the majority] to say that it is the issuer which has required or invited them to do so." (291)

2. The Disappearance of Intercreditor Good Faith Cases in the United States

The idea that express contract rights given to majority bondholders must be equitably exercised was black letter law on the eve of the enactment of the TIA. (292) An evolutionary dead end followed. Why did the line of cases die out? First and foremost, the TIA had the effect of shifting a substantial portion of restructuring activity into bankruptcy, where it occurred under judicial supervision and with express good faith duties. As a result, intercreditor duties slipped out of the collective memory, so much so that they were not cited in the cases of Marblegate and Caesars.

Second, intercreditor duties simultaneously lost their framework of application. The duty was articulated in equity receiverships in which federal courts applied federal equity jurisprudence, a body of law that looks much like a federal common law, in umpiring corporate restructuring. While federal equity receiverships are still possible, they have largely been supplanted by bankruptcy proceedings. Moreover, the merger of law and equity in the 1938 Federal Rules of Civil Procedure and the Supreme Court's 1938 holding in Erie Railroad Co. v. Tompkins that "there is no general federal common law" (293) rendered further confusion about the role and continued viability of federal equity jurisprudence. As a result, the old equity receivership cases endure as homeless precedents; strictly speaking, binding no one. (294) But that does not deprive them of persuasive status--the late twentieth century good faith duty can be seen picking up legal notions that formerly took shape on the courts' equitable side. Meanwhile, out-of-court restructuring is returning to real world salience. Looking back to pre-Depression case law provides sensible doctrinal guidance.

Third, as the bond market shifted from mom-and-pop bondholders to institutional investors, courts likely became less concerned about intercreditor voting conflicts. Consider Aladdin Hotel Co. v. Bloom, (295) which generations of corporate finance textbooks have used to illustrate the evils that can befall minority bondholders in the absence of the TIA. (296)

The story of Aladdin Hotel begins with the bankruptcy of the Armour Building Company, which in 1925 built a sixteen-floor Italian Romanesque building in downtown Kansas City, Missouri, called the Aladdin Hotel (now a Holiday Inn). (297) The Armour Building Company found itself in bankruptcy in the 1930s and was reorganized as the Aladdin Hotel Company, with its prepetition bondholders receiving both Aladdin's newly issued bonds and equity. (298) As in Hackettstown, Aladdin's majority equityholders also held a majority of the bond issue. (299) When the new bonds came due, the Aladdin Hotel Company sought to extend the bonds' maturity date by ten years; this amendment had the knock-on effect of decreasing the interest rate from the postmaturity rate of 8% to the prematurity rate of 5%. (300) The indenture had a two-thirds majority CAC (301) and the TIA did not apply because the bonds had been issued in 1938. (302) Critically, the members of the bondholder majority who approved the amendment also owned a majority of the issuer's common stock. (303) A minority bondholder challenger won in the district court, but lost in the Eighth Circuit.

Although Aladdin was not precisely framed in intercreditor terms--the Eighth Circuit posed it as an equitable inquiry--the case has intercreditor characteristics. Because the indenture explicitly permitted majority amendment, the plain language of the indenture gave the plaintiff no case at law. The only attack the plaintiff had was an equitable claim about the amendment's unfairness. The Eighth Circuit, however, proved blind to the operative incentives, proclaiming that "it is inconceivable that the [equityholders] should deliberately act to the prejudice or detriment of the bondholders when they held and owned some 72 per cent of the entire outstanding bond issue." (303) Although this language is extreme, the case should not be read as a wholesale repudiation of the earlier intercreditor cases. The court noted that the alleged conflict of interest would not have mattered anyway: the plaintiff had purchased the bonds after the amendment's completion, and equitable rights were not assignable under Missouri law. Thus, the plaintiff's purchase included only the bond's express legal rights. (305)

So, while Aladdin denied the plaintiff a good faith remedy, it kept open the possibility of such relief for other plaintiffs at the time of the amendment. Back in the 1950s, good faith was still understood as an equitable matter requiring clean hands from the party seeking relief; in those days, distressed debt investors apparently did not meet the requirement. (306) Marblegate's contrastingly decorous treatment of the holdout plaintiff demonstrates how attitudes toward distressed debt investing have changed.

Aladdin, then, does recognize an intercreditor good faith duty; it simply reaches that duty through an equity framework. This is common for the older precedents, while the modern approach focuses on a contract law framework of an implied covenant of good faith and fair dealing. The older equitable intercreditor good faith duty could be readily adapted, however, to the modern implied covenant framework.

3. Scope

We now address the intercreditor duty's scope of application in exchange offer contexts. As a good faith duty, it would operate as a default while explicit process rulemaking would be left to the drafters. The duty would easily replicate the result in Caesars I, because it plainly constrains giving consent in exchange for selective redemptions. The question is whether the duty does a great deal more than that: as articulated in Assenagon, the duty could be read to imply a categorical prohibition of exit consents. Given section 316(b), this would throw a wrench into out-of-court restructuring, at least under indentures not explicitly permitting exit consents. (307) Contrariwise, categorical prohibition of exit consents would make perfect sense in a CAC regime.

There is a further question regarding Marblegate: whether the duty, responding to self-interested voting in Caesars I, also applies to the Marblegate bank lender's self-interested enforcement action. We think a solid line can be drawn.

Marblegate, in which an unsecured bondholder successfully challenged an enforcement action undertaken by a secured bank lender, involved two issues of debt. The court's ruling imposed a duty on creditors under one instrument to refrain from imposing what amounted to a restructuring on the creditors under the other. The intercreditor good faith cases, in contrast, arise from disputes among creditors under common instruments. We see no reason to extend intercreditor good faith to creditors under multiple instruments because creditors under different instruments are necessarily competitors who have not impliedly bound themselves to share a recovery from what might be a limited pool of assets.

Some gray areas complicate the line-drawing exercise. Consider conflicts between different classes of creditors under the same instrument, such as the senior-subordinate structure created in securitizations and multi-facility syndicated loans. We think these should fall within the zone of good faith because such creditors have bound themselves to share a recovery from a potentially limited pool. Although they have structured the sharing as something other than pro rata, the particular allocation is irrelevant; these creditors should still have a duty to each other to maximize the overall recovery. The same outcome should hold for creditors under separate instruments who have bound themselves together through an intercreditor agreement because they have voluntarily moved from the zone of self-interest into the zone of mutual obligation.

Now consider situations involving creditors under separate instruments who are connected not through an intercreditor agreement, but through either (1) a senior-subordinate structure in which bondholders under one instrument acquire senior status by taking an assignment of the junior claim, or (2) a tag-along right, such as a tag-along guaranty or negative pledge clause. We do not think that intercreditor good faith duties should apply. Creditors on a junior note necessarily have a voluntary potential conflict with those on a senior note. Likewise, tag-along rights can appear solely in one instrument; the parties to the other instrument might not even be aware of the tag-along rights. Other doctrines, such as tortious interference with contract, might be better suited to policing any manipulative behavior rather than a general duty of good faith.

4. Conclusion

Looking forward to a repeal of section 316(b), we believe that the intercreditor good faith duty, taken together with the good faith duty's limited constraints against bond issuers, presents a viable framework for policing coercion and opportunism in restructuring. It also would focus the drafters' attention. And, as a default, the rule would work better in the long run than an open-ended federal antidistortion mandate.


The practice picture respecting workouts has changed dramatically, but not so much as to solve a number of longstanding problems. There is no perfect regulatory solution, leaving us in a zone of imperfections and tradeoffs. But we draw five particular conclusions.

First, the expansive reading of section 316(b)--although laudably motivated--was overly broad and followed from an incomplete review of the legislative history. The Second Circuit was amply justified in reversing Marblegate. Despite this, stepped up judicial policing under the good faith rubric on an intercreditor basis would be useful and could respect contractual risk allocations.

Second, increased workout activity, taken together with changes in bankruptcy practice, denude the case for repeal of section 316(b) of policy urgency. Markets have shown that they can muddle through with the status quo.

Third, section 316(b) can be repealed prospectively without undue risk to the bondholder interest. Take-it-or-leave-it exchange offers no longer present a serious problem, for issuers now negotiate solely with bondholders who know how to evaluate deals and say no. Even the section's bankruptcy-forcing purpose now rings hollow given the reality of secured creditor control, which comes at the expense of unsecured bondholders.

Fourth, any repeal should be outright, leaving the matter to the drafters and the market. The theory signals a CAC regime with ancillary process protections, but it is not clear that investors prefer a theoretically correct regime. Moreover, their preferences will be dynamic in time, evolving in response to events. Also, a theoretically correct regime would be difficult to implement due to the verification problem respecting bondholder conflicts. Tradeoffs would have to be made, a matter best left to the contracting parties.

Finally, a backstop default regime incorporating an intercreditor good faith duty would assist the transition process in the wake of repeal. Such a regime could focus the issuers' attention and spur indenture forms in the direction of a complete set of process rules.

The restructuring business has changed in recent decades, such that its New Deal-derived legal paradigms no longer fit market realities. It is time to consider a reform of restructuring law that facilitates fair and efficient workouts.

WILLIAM W. BRATTON ([dagger]) & ADAM J. LEVITIN ([double dagger])

([dagger]) Nicholas F. Gallicchio Professor of Law and Co-Director, Institute for Law & Economics, University of Pennsylvania Law School.

([double dagger]) Agnew N. Williams Research Professor and Professor of Law, Georgetown University Law Center. The article has benefitted from presentations at the Georgetown University Law Center Faculty Workshop, the University of Pennsylvania Law School Institute for Law & Economics Corporate Roundtable, the Goethe/Penn Conference on Law & Finance, the Villanova Business Law Colloquium, the Columbia-Duke-NYU Black Holes Conference, and the University of Pennsylvania Law Review Symposium. The authors wish to thank Philip Abelson, Adam Badawi, Martin Bienenstock, Tony Casey, Jill Fisch, Jesse Fried, Whitman Holt, Henry Hu, Marcel Kahan, David Kershaw, Frank Partnoy, Robert Rasmussen, Mark Roe, Gabriel Rauterberg, Richard Schifter, and Eric Talley, for their comments on earlier drafts, and Joshua Kaufman, Emil Kleinhaus, and participants in the 2016 Financial Lawyers Seminar for sharing their practice experience, and Ari Dropkin and the Georgetown University Law Library staff for research assistance.

(1) 15 U.S.C. [section] 77bbb (1939).

(2) See infra text accompanying note 54.

(3) Section 316(b) accomplishes this by omitting to mention exchange offers.

(4) Stockholder protections are described in section II.A., infra; bondholder protections are described in section II.B., infra.

(5) See infra notes 57-58 and accompanying text (describing concerns raised in SEC report regarding pre-TIA practices).

(6) See infra note 56 and accompanying text (summarizing House and Senate reports accompanying enactment of section 316(b)).

(7) See infra text accompanying note 293.

(8) See infra text accompanying notes 164-70 (reviewing uptick in workout activity).

(9) See infra text accompanying note 167.

(10) Marblegate Asset Mgmt., LLC v. Educ. Mgmt. Corp., 111 F. Supp. 3d 542 (S.D.N.Y. 2015), rev'd, 846 F.3d 1 (2d Cir. 2017).

(11) Marblegate Asset Mgmt., LLC v. Educ. Mgmt. Corp., 846 F.3d 1 (2d Cir. 2017).

(12) See infra text accompanying notes 104-06 (discussing viability of repeal).

(13) See Victor Brudney, Corporate Bondholders and Debtor Opportunism: In Bad Times and Good, 105 HARV. L. REV. 1821,1834-35 (1992) (describing the structural disabilities of dispersed bondholders).

(14) See infra Section III.D.1.

(15) See infra text accompanying notes 192-193 (discussing creditor control).

(16) See infra Section IV.B.

(17) See infra Section IV.D.

(18) Workout, BLACK'S LAW DICTIONARY (10th ed., 2014) ("A debtor's agreement, usu[ally] negotiated with a creditor or creditors out of court, to reduce or discharge the debt").


(20) 11 U.S.C. [section] 362(a) (2012).

(21) See 11 U.S.C. [section][section] 1126(b) (2012) (requiring only a single consenting class, not counting votes of insiders, for cramdown confirmation). In some situations in bankruptcy, a debtor can bind creditors without any of them consenting. See 11 U.S.C. [section] 1124(2) (2012) (treating as unimpaired creditors whose debts are deaccelerated, cured, and reinstated); see also ADAM J. LEVITIN, BUSINESS BANKRUPTCY: FINANCIAL RESTRUCTURING AND MODERN COMMERCIAL MARKETS 542-43 (2015) (discussing "cramup").

(22) See Collective Action Clause, NASDAQ.COM, [] (last visited Apr. 4, 2018) ("CAC allows bondholders to agree on debt restructuring even when some bondholders are against restructuring as long as majority agrees.").

(23) See William W. Bratton & G. Mitu Gulati, Sovereign Debt Reform and the Best Interest of Creditors, 57 VAND. L. REV. 1, 3-4 (2004) ("[UACs] condition amendment of the bond contracts' key payment terms on unanimous bondholder consent.").

(24) Technically, the reference to unanimity is a misnomer. UACs require a given bondholder to consent before an amendment can be binding; they do not prevent a majority from making nonbinding concessions. The appellation does make sense as a practical matter, since UACs generally condition across-the-board application on unanimity, making workout by direct amendment unfeasible.

(25) See Bratton & Gulati, supra note 23, at 21 (defining workout by exchange).

(26) See Zohar Goshen, Controlling Strategic Voting: Property Rule or Liability Rule?, 70 S. CAL. L. REV. 741, 749 (1997) ("[W]henever an individual's action has implications for the group, that action should be approved by a simple majority vote and decisions should be binding on the entire group.").

(27) Id. at 745-46; see also Edward J. Janger & Adam J. Levitin, One Dollar, One Vote: Mark-to-Market Governance in Bankruptcy, 104 IOWA L. REV. (forthcoming 2019) (discussing the problem of insincere voting in bankruptcy).

(28) For the sake of consistency, we will refer to debt securities as "bonds" whether the credit instrument is denominated as a "bond," "debenture," or "note," and we will refer to the investors in all such instruments as "bondholders."

(29) Goshen, supra note 26, at 755.

(30) Id. at 756.

(31) See id. at 757 ("Therefore, if as a result of the reorganization the value of the original bonds is greater than the value of the new bonds... the rational bondholder will prefer to sit back and watch the other bondholders offer their bonds while she takes a free ride.").

(32) See id. at 785-86 (discussing the risks and benefits that a bondholder may face when refusing to tender prior to closing).

(33) See N.Y. Stock Exch., NYSE Listed Company Manual [section] 801.00 (2018), [] (last visited Apr. 4, 2018) ("Securities admitted to the list may be... removed from the list at any time that a company falls below certain... listing criteria.").

(34) See John C. Coffee, Jr. & William A. Klein, Bondholder Coercion: The Problem of Constrained Choice in Debt Tender Offers and Recapitalizations, 58 U. CHI. L. REV. 1207,1224-25 (1991) (noting that bondholders can eliminate protective covenants).

(35) See infra text accompanying note 211.

(36) See Goshen, supra note 26, at 785 (explaining why bondholders will vote for an amendment that will ostensibly decrease bond value).

(37) See 15 U.S.C. [section] 77ppp(b) (1990) ("The right of any holder of any indenture security to receive payment of the principal of and interest on such indenture security... shall not be impaired or affected without the consent of such holder.").

(38) See, e.g., DEL. CODE. ANN. tit. 8, [section] 216(2) (2018) ("In all matters other than the election of directors, the affirmative vote of the majority of shares present in person or represented by proxy at the meeting and entitled to vote on the subject matter shall be the act of the stockholders."). It is noted that supermajority provisions governing charter and bylaw amendments are nonetheless very common. See e.g., What's (Really) Hot: A Quick Score Analysis of 2015's Real Governance Trends, INSTITUTIONAL SHAREHOLDER SERVICES, [] (last visited Apr. 4, 2018) (questioning the "common wisdom" that it is rare for a supermajority vote to be required to amend corporate governing documents).

(39) See, e.g., Chew v. Inverness Mgmt. Corp., 352 A.2d 426,430 (Del. Ch. 1976) (disallowing votes cast in a board election by shareholders who obtained options to purchase shares in exchange for consideration).

(40) The exception has narrowed over time. See Schreiber v. Carney, 447 A.2d 17, 17 (Del. Ch. 1982) (permitting a loan inducing a vote). What was once an open-ended standard generally directed to voting for consideration recently has taken on rule-like characteristics. See Portnoy v. Cryo-Cell Int'l, Inc., 940 A.2d 43, 68 (Del. Ch. 2008) (rejecting an intrinsic fairness test in favor of a distinction between improper and permissible inducement for voting).

(41) See infra note 42 and accompanying text (describing the prevailing case where a controlling shareholder fiduciary duty was imposed).

(42) The classic case is Sinclair Oil Corp. v. Levien, 280 A.2d 717 (Del. 1971), where the Delaware Supreme Court held that board control implies a duty against "self-dealing," defined as taking something to the exclusion or detriment of the minority shareholders. Id. at 720-21.

(43) See Kahn v. M&F Worldwide Corp., 88 A.3d 635, 653-654 (Del. 2014) (holding that approval of a majority of the minority shareholders was required in a cash-out merger); Lewis v. Vogelstein, 699 A.2d 327, 335-36 (Del. 1997) (requiring a majority of disinterested shareholders to approve a director self-dealing transaction).

(44) 17 C.F.R. [section][section] 240.14a-1, 14b-2 (1970). The federal securities laws now also mandate sincere voting on the part of institutional intermediaries. See Proxy Voting by Investment Advisors, 17 C.F.R. [section] 275.206(4)-6 (2003) (requiring investment advisers to adopt policies reasonably designed to ensure that the adviser votes proxies in the best interest of clients). Significantly, this innovation follows not from a revision of shareholder voting norms, but from a revision of the norms governing the relationship between fund managers and their beneficiaries. It thus follows from fiduciary duty and conforms to the overall pattern.

(45) Goshen, supra note 26, at 766-68.

(46) See Williams Act, Pub. L. No. 90-439, 82 Stat. 454 (1968) (codified at 15 U.S.C. [section][section] 78m(d), 78m(e), 78n(d)-(f) (2012)).

(47) 17 C.F.R. [section][section] 240.13e-4(f)(8)(i)-(ii),240.14d-10(a)(1)-(2) (2017).

(48) See, e.g., Unitrin, Inc. v. Amer. Gen. Corp., 651 A.2d 1361, 1356 (Del. 1995) ("The Court of Chancery should have directed its enhanced scrutiny: first, upon whether the Repurchase Program the Unitrin Board implemented was draconian, by being either preclusive or coercive and... if it was not... upon whether it was within a range of reasonable responses to the threat"); Unocal Corp. v. Mesa Petroleum Co., 493 A.2d 946, 949 (Del. 1985) ("[W]e are satisfied that the [defensive] device Unocal adopted is reasonable in relation to the threat posed, and that the board acted in the proper exercise of sound business judgment.").

(49) The Delaware courts backed into this result after the Delaware Supreme Court, as a matter of statutory construction, barred fiduciary review of minority shareholder cash-out mergers in cases where the majority holds ninety percent or more of the shares. See Glassman v. Unocal Exploration Corp., 777 A.2d 242, 248 (Del. 2001) ("[A]bsent fraud or illegality, appraisal is the exclusive remedy available to a minority stockholder who objects to a short-form merger."). Majority shareholders holding less than ninety percent thereafter evaded direct fiduciary scrutiny of their mergers by conducting antecedent tender offers to bring themselves up to the ninety percent threshold. The Delaware courts thereafter adjusted with an anticoercion rule applied to the antecedent tender offer.

(50) In re Pure Res., Inc. S'holder Litig., 808 A.2d 421, 445 (Del. Ch. 2002).

(51) Bonds involve two separate contracts: a note, which is a promise from the bond issuer to repay the bond, and an indenture. The indenture creates a common enforcement mechanism for all of the bonds through an entity called an "indenture trustee," who is to represent the interests of the dispersed bondholders. Although there need not be any actual trust corpus, the term is a holdover from older practice when bonds were generally secured and the collateral was held in trust. As a result, the indentures are often called "trust indentures," hence the name "Trust Indenture Act." The term "indenture" refers to the contract itself--an indenture is merely a contract written with a primitive antifraud device consisting of two counterparts of the contract written on the same sheet of paper or parchment, which would then be cut in two, so as to divide the counterparts. The cut would be made with a set of zigzagged indents, hence the name indenture. The idea was that the two counterparts would have to fit together like Little Orphan Annie's locket or the Passover afikomen, which would protect against fraudulent documents. Obviously, such devices are not in use today, but the term has persisted.

(52) See 11 U.S.C. [section] 362 (2012) (barring legal claims and other means of contract enforcement against individuals and entities who file for bankruptcy, except for those actions explicitly permitted under the statute).

(53) 15 U.S.C. [section] 77ppp(a) (1990). Subsection (a) also provides that bondholder majorities must be permitted to direct enforcement proceedings conducted by the indenture trustee.

(54) 15 U.S. C. [section] 77ppp(b) (1990).

(55) See supra Part LA.

(56) s. REP. NO. 75-1619, at 19 (1939); S. REP. NO. 76-248, at 26 (1939); H.R. REP. NO. 76-1016, at 56 (1939). There also was a purpose to synchronize the TIA's regime of mandatory terms with state-based legal regimes that required an unconditional promise to pay in order to import negotiability. See, e.g., ARTHUR W. SELOVER, THE NEGOTIABLE INSTRUMENTS LAW 1 (1900) (identifying the purpose of the Negotiable Instruments Law as a "codification of the principle rules of law governing negotiable instruments"). Section 316's rights-based language derives from this concern. The formulation came from contemporary trust indentures and was designed to assure negotiability. AM. B. FOUND., COMMENTARIES ON MODEL DEBENTURE INDENTURE PROVISIONS 1965: MODEL INDENTURE PROVISIONS ALL REGISTERED ISSUES 1967 AND CERTAIN NEGOTIABLE PROVISIONS WHICH MAY BE INCORPORATED IN A PARTICULAR INCORPORATING INDENTURE 234 (1971). CACs applicable to payment terms were thought to undercut negotiability by interjecting uncertainty as to sum. But opinion was mixed. See De Forest Billyou, Corporate Mortgage Bonds and Majority Clauses, 57 YALE L.J. 595, 600-02 (1948) (arguing that a five-year postponement of principal and interest based on a three-quarters vote would not run afoul of the law); see also Robert T. Swaine, Reorganization of Corporations: Certain Developments Over the Last Decade, 27 COLUM. L. REV. 901, 927 (1927) (opining that no action clauses precluding individual suit presented "a serious question whether they do not destroy the negotiability of the obligations affected by them").

(57) SEE 3 SECURITIES AND EXCHANGE COMM'N, REPORT ON THE STUDY AND INVESTIGATION OF THE WORK, ACTIVITIES, PERSONNEL AND FUNCTIONS OF PROTECTIVE AND REORGANIZATION COMMITTEES 227 (1936) ("By virtue of the[se provisions'] voluntary features, the expense and delay of foreclosure or other court proceedings are to be avoided.").


(59) Id. at 137-43.

(60) See id. at 147-48. ("To assume that the holder of a single bond out of a large issue is entided to block a constructive program favored by the overwhelming majority of his fellow investors, is to cling to a conception of rights which has no application to the modern methods of issuing and distributing securities.").

(61) For a description of the pre-1934 regime of equity receivership, see DAVID A. SKEEL, JR., DEBT'S DOMINION: A HISTORY OF BANKRUPTCY LAW IN AMERICA 56-69 (2003); Stephen J. Lubben, Railroad Receiverships and Modern Bankruptcy Theory, 89 CORNELL L. REV. 1420,1441-43 (2004).

(62) Act of June 7,1934, [section] 77B, 48 Stat. 911 (repealed 1938).

(63) See E. Merrick Dodd, Jr., Reorganization Through Bankruptcy: A Remedy for What?, 10 J. NAT'LASS'N REF. BANKR. 46,46 (1936) (describing the competing interests Congress sought to accommodate in 1934).

(64) See Developments in the Law: Reorganization under Section 77B of the Bankruptcy Act: 1934-1936, 49 HARV. L. REV. 1111, 1185 (1936) ("One of the chief purposes of enacting Section 77B undoubtedly was to overcome the necessity of holding a judicial sale and paying dissenting creditors the value of their claims in cash--a technique apparently necessary in 'consent' receiverships, but often a bar to carrying out an effective reorganization. Section 77B has met these difficulties in part by providing that a plan accepted by 'creditors holding two thirds in amount of the claims of each class' binds all in the class when confirmed by the court's decree.").

(65) Act of June 7,1934, [section] 77B, 48 Stat. 911, 920-22 (repealed 1938); see also Dodd, supra note 63, at 56-61 (describing the objectives of small claimants and showing how the section 77B process was supposed to satisfy those objectives, but predicting that judges would continue to rule in favor of insiders and creditor majorities).

(66) See Richard W. Jennings, Mr. Justice Douglas: His Influence on Corporate and Securities Regulation, 73 YALE L.J. 920, 936 (1964) ("With a plan of reorganization prepared in advance by the lawyers for the Committee and the necessary consents in hand, courts generally could be prevailed upon to confirm the plan against attacks of unfairness on the theory that the creditors and shareholders had freely consented and were in the best position to protect their own interests. The whole affair thus became a 'lawless' operation masquerading under a facade of legitimacy. The larger question was whether the machinery of government was capable of making any lasting and effective reforms in this traditionally chaotic field.").

(67) The effect was ironic given the inclusion in section 77B of a provision allowing courts to disregard prebankruptcy protective committees' main tool for binding creditors: the depository agreement. Act of June 7,1934, [section] 77B, 48 Stat. 911,915 (repealed 1938).

(68) SEE, E.G., 1 SECURITIES AND EXCHANGE COMM'N, REPORT ON THE STUDY AND INVESTIGATION OF THE WORK, ACTIVITIES, PERSONNEL AND FUNCTIONS OF PROTECTIVE AND REORGANIZATION COMMITTEES 506-96 (1937) (describing coercive tactics employed against bondholders); id. at 863-83 (describing insider and underwriter control and rent-seeking motivations on the part of their opponents).

(69) See supra note 57 and accompanying text (showing an understanding that CACs save the time and expenses associated with bankruptcy hearings).

(70) See supra text accompanying note 59 (demonstrating knowledge of the widespread use of CACs in Canada and Great Britain).

(71) See Trust Indenture Act of 1939: Hearing on H.R. 2191 & H.R. 5220 Before the Subcomm. on Interstate and Foreign Commerce, 76th Cong. 76-77 (1939) (statement of John K. Starkweather, Chairman of the Federal Legislation Committee of the Investment Bankers' Association of America) ("I would say today it probably is a very safe assumption to say that from 50 to 75 percent of the desirable issues are going into the hands of insurance companies or banks."). In 1933, however, Jerome Frank, one of the drafters of the SEC report, infamously wrote, "Courts of equity have a tradition of aiding the helpless, such as infants, idiots and drunkards. The average security holder in a corporate reorganization is of like kind." Jerome Frank, Reflections on Corporate Reorganizations, 19 VA. L. REV. 541, 569 (1933).

(72) Pub. L. No. 90-439, 82 Stat. 454 (1968) (codified at 15 U.S.C. [section][section] 78m(d), 78111(e), 78n(d)-(f) (1982)). The Williams Act requires that the offer stay open twenty days, 17 C.F.R. [section] 240.146-1 (2008), that tenders may be withdrawn at any time prior to the offer closing, that the offer go to all holders, 17 C.F.R. [section][section] 240.13e-4(f)(8)(i), 240.14d-10(a)(1) (2008), and that all tenders be paid the highest consideration on offer. 17 C.F.R. [section][section] 240.13e-4(f)(8)(ii), 240.14d-10(a)(2) (2008).

(73) Rule 14e-1 under section 14(e) of the 1934 Act requires that tender offers be held open for twenty business days, and an additional ten business days from the date of a change in terms. 17 C.F.R. [section] 240.146-1 (2008). In a case where the exchange offer carries an exit consent, this requirement prevents the issuer from putting through the contract amendment ahead of closing the exchange offer.

(74) See 15 U.S.C. [section][section] 78m(e), 78n(d)(1), 240.13e-4(a)(2) (2008) (defining issuer tender offer as an offer for equity securities); 15 U.S.C. [section] 240.14d-1(a) (2008) (remitting tender offers to debt securities to Regulation 14E); E.H.I., Inc. v. Ins. Co. of N. Am., 652 F.2d 310,313-15 (3d Cir. 1981) (addressing 17 C.F.R. [section] 240.14d-1); Royce de R. Barondes, An Economic Analysis of the Potential for Coercion in Consent Solicitations for Bonds, 63 FORDHAM L. REV. 749,762-65 (1994) ("However, the requirements that an offer be made to all holders, that the highest consideration paid in the tender offer be paid to all holders who tender and that partial offers be prorated generally would not apply to debt tender offers"). The upshot is that an offer of straight debt is subject to Regulation 14E, and Rules 14e-1, 14e-2, and 146-3--the twenty-day rule and the antifraud rules. However, exchange offers paying cash or debt convertible into equity must comply with the full-dress requirements in Rule 136-4.

(75) 17 C.F.R. [section][section] 240.13e-4(f)(8)(i), 240.14d-10(a)(1) (1999).

(76) See 17 C.F.R. [section][section] 240.13e-4(f)(8)(ii), 240.14d-10(a)(2) (1999) (discussing security holders' rights).

(77) RESTATEMENT (SECOND) OF CONTRACTS, [section] 205 (AM. LAW INST. 1981).

(78) See, e.g., Sharon Steel Corp. v. Chase Manhattan Bank, N.A., 691 F.2d 1039, 1049 (2d Cir. 1982) ("Contract language is thus the starting point in the search for meaning... Sharon's literalist approach simply proves too much."); Katz v. Oak Indus., Inc., 508 A.3d 873, 879 (Del. Ch. 1986) ("Arrangements among a corporation, the underwriters of its debt, trustees under its indentures and sometimes ultimate investors are typically thoroughly negotiated and massively documented... The terms of the contractual relationships agreed to and not broad concepts such as fairness define the corporation's obligation to its bondholders").

(79) When the issuer acts in accordance with the indenture's express provisions, good faith claims are foreclosed. See, e.g., Aaron Ferer & Sons Ltd. v. Chase Manhattan Bank, Nat. Ass'n, 731 F.2d 112, 126 (2d Cir. 1984) ("All of Williams & Glyn's claims are barred by its release, knowingly and voluntarily given."); Banco Urquijo, S.A. v. Signet Bank, 861 F. Supp. 1220, 1249 (M.D. Pa. 1994) ("Moreover, a fiduciary relationship does not arise from a lender participation agreement unless the agreement expressly provides for such a relationship."); Banque Arabe et Internationale d'Investissement v. Md. Nat'l Bank, 819 F. Supp. 1282,1296 (S.D.N.Y. 1993) ("[T]his Court has been unable to find, and the plaintiff has not identified, any section of the Participation Agreement or related documents which creates such a [fiduciary] duty."); Banco Espanol de Credito v. Sec. Pac. Nat'l Bank, 763 F. Supp. 36, 45 (S.D.N.Y. 1991) ("In the case of arm's length transactions between large financial institutions, no fiduciary relationship exists unless one was created in the agreement.").

(80) See e.g., Broad v. Rockwell Int'l Corp., 642 F.2d 929, 957 (5th Cir. 1981) (en banc) ("We note first that this implied covenant of good faith and fair dealing cannot give the holders of Debentures any rights inconsistent with those explicitly set out in the Indenture."); Metro. Life Ins. Co. v. RJR Nabisco, Inc., 716 F. Supp. 1504,1517 (S.D.N.Y. 1989) (refusing to impose an "unbounded and one-sided" duty to the terms of an indenture); Gardner & Florence Call Cowells Found, v. Empire, Inc., 589 F. Supp. 669, 673 (S.D.N.Y. 1984), vacated on procedural grounds, 754 F.2d 478 (2d Cir. 1985) (finding no breach of an implied covenant since the contractual rights of the Indenture were not violated); Cincinnati SMSA Ltd. P'ship v. Cincinnati Bell Cellular Sys. Co., 708 A.3d 989, 992 (Del. 1998) (determining a bondholder must show from express terms that particular implied term would have been included if parties had negotiated for it).

(81) The Rule of Explicitness, which "prevents a senior lienholder from obtaining post-petition interest under a subordination... agreement unless there is language in the agreement that is 'precise, explicit, and unambiguous,'" is one mechanism bankruptcy courts utilize to ensure parties have predictability in the interpretation of their financial agreements. John C. Murray, Enforceability of Intercreditor Agreements in Bankruptcy, 19 PRAC. REAL EST. L. 27, 29 (2003).

(82) 508 A.2d 873 (Del. Ch. 1986).

(83) Id. at 881.

(84) Id. at 881-82.

(85) Id. at 879.

(86) 1986 WL 13008 (Del. Ch. Nov. 14, 1986), aff'd, 518 A.2d. 983 (Del. 1986); see also Drage v. Santa Fe Pac. Corp., 1995 WL 396370 (Ohio Ct. App. July 3,1995). Corporate law's anti-vote buying prohibition does not carry over to senior securities. New York, for example, relaxes a statutory prohibition against vote-buying to permit the votes of preferred stockholders to be bought, provided the offer to purchase is made to all holders and left open for twenty days. N.Y. BUS. CORP. L. [section] 609(e) (LexisNexis 1963). The same would seem to follow for bonds, which in any event lie outside the corporate law pale.

(87) Kass, 1986 WL 13008 at *1.

(88) Id.

(89) Id. at *4-5.

(90) RESTATEMENT (SECOND) OF CONTRACTS, [section] 175(1) (AM. LAW INST. 1981).

(91) Oren Bar-Gill & Omri Ben-Shahar, Credible Coercion, 85 TEX. L. REV. 717, 757-58 (2005).

(92) Van Gemert v. Boeing, 520 F.2d 1373 (2d Cir. 1975).

(93) 11 U.S.C. [section] 1129(a)(3) (2012).

(94) 11 U.S.C. [section] 1126(e) (2012).

(95) 11 U.S.C. [section] 1129(b) (2012).

(96) Compare In re MCorp Fin. Inc. 137 B.R. 219, 236 (Bankr. S.D. Tex. 1992) (holding that a death trap provision violated the "fair and equitable" requirement for cramdown plans) and In re Allegheny Int'l, Inc., 118 B.R. 282, 304 n.15 (Bankr. W.D. Pa. 1990) (finding that a death trap provision was not fair and equitable because it discriminated against classes of creditors who voted against the plan), with In re Zenith Elecs. Corp., 241 B.R. 92,105-06 (Bankr. D. Del. 1999) (holding a death trap plan permissible if it does not violate "fair and equitable" requirement) and In re Drexel Burnham Lambert Grp., Inc., 138 B.R. 714, 715 (Bankr. S.D.N.Y. 1992) (approving death trap plan), aff'd, 140 B.R. 346 (S.D.N.Y. 1992).

(97) H.R. REP. NO. 95-595, at 411 (1977) (explaining that section 1126(e) is intended to overturn Aladdin Hotel Co. v. Bloom, in which the majority bondholders voted for an amendment because of the benefit to their equity interest in the debtor firm).

(98) 7 COLLIER ON BANKRUPTCY [paragraph] 1126.o6[2] (16th ed. 2011); see also Young v. Higbee Co., 324 U.S. 204, 213 (1945) ("[Creditors] cannot avail themselves of the statutory privilege of litigating for the interest of a class and then shake off their self-assumed responsibilities to others by a simple announcement that henceforth they will trade in the rights of others for their own aggrandizement.").

(99) See Figter Ltd. v. Teachers Ins. & Annuity Ass'n of Am., 118 F.3d 635, 639 (9th Cir. 1997) (stressing that "the mere fact that a creditor has purchased additional claims for the purpose of protecting his own existing claim does not demonstrate bad faith or an ulterior motive").

(100) See RESTATEMENT (SECOND) OF CONTRACTS, [section] 89(a) (AM. LAW INST. 1981) (calling for changed circumstances and fairness review).

(101) Kass v. E. Air Lines, Inc., 1986 WL 13008, at *3 (Del. Ch. Nov. 14,1986).

(102) Id. at *1.

(103) See generally William W. Bratton, Bond and Loan Covenants, Theory and Practice, 11 CAP. MKTS. L.J. 461, 477-81 (2016) ("Bank loans and private placements tend to involve a small number of lenders able to coordinate renegotiation of the contract terms with the borrower in the event of a covenant default. Indeed, renegotiation is expected.").

(104) See Mark J. Roe, The Voting Prohibition in Bond Workouts, 97 YALE L.J. 232, 263 (1987) ("As a protective measure, section 316 is incomplete.").

(105) Id. at 243.

(106) Id. at 239.

(107) Roe argues (1) that many bonds are without significant covenants to strip away, (2) that exit consents are a factor only in a debt exchange and not in a deal involving cash or stock, and (3) that the courts might eventually suppress them. See id. at 250, 256 (noting that majoritarian amendment of payment terms no longer impaired negotiability).

(108) See Brudney, supra note 13, at 1877 ("[R]ules could be fashioned to encourage, or even require, public bondholders to organize so that they might deal with the debtor in the same way as a sole lender would.").

(109) See id. at 1828 (evaluating the costs and benefits of strategic behavior).

(110) Id. at 1832-34.

(111) Id. at 1853-54.

(112) Id. at 1862-63.

(113) Id. at 1860.

(114) See Roe, supra note 104, at 236-37 n.11.

(115) See Julian R. Franks & Walter N. Torous, A Comparison of Financial Recontracting in Distressed Exchanges and Chapter 11 Reorganizations, 35 J. FIN. ECON. 349, 351 (1994) (tracking 161 distressed firms from 1983-1988 and finding that 76 restructured out of court and 78 went into bankruptcy); Stuart C. Gilson, Kose John & Larry H.P. Lang, Troubled Debt Restructurings: An Empirical Study of Private Reorganization of Firms in Default, 27 J. FIN. ECON. 315, 328 (1990) (studying distressed exchange offers from 1978-1987 and finding that 52.7% of the firms went into bankruptcy); Lewis S. Peterson, Who's Being Greedy? A Theoretical Examination of Holdouts and Coercion in Debt Tender and Exchange Offers, 103 YALE L.J. 505, 506 (1993) (citing a distressed sample from 1990-1992, of which 50% of the total debt volume was restructured out of court). Salomon Brothers kept running totals, showing one-half or less completed on an annual basis, 1991-93. See Peterson, supra note 115, at 506 n.6.

(116) See Sris Chatterjee, Upinder S. Dhillon & Gabriel G. Ramirez, Coercive Tender and Exchange Offers in Distressed High-Yield Debt Restructurings: An Empirical Analysis, 38 J. FIN. ECON. 333, 339-40 (1995) (studying 46 exchange announcements by 40 firms, from 1989-1992).

(117) See Christopher James, Bank Debt Restructurings and the Composition of Exchange Offers in Financial Distress, 51 J. FIN. 711, 716-18 (1996) (studying 68 proposed exchange offers from 1980-1990).

(118) See Edith S. Hotchkiss, Kose John, Robert M. Mooradian & Karin S. Thorburn, Bankruptcy and the Resolution of Financial Distress, in 2 HANDBOOK OF CORPORATE FINANCE: EMPIRICAL CORPORATE FINANCE 253 (B. Espen Eckbo ed., 2008) ("The general conclusion from much of this literature is that absent holdout problems and other coordination problems, private debt restructurings such as exchange offers provide a lower cost restructuring mechanism than formal bankruptcy.").

(119) Peterson, supra note 115, at 526.

(120) See Kenneth Daniels & Gabriel G. Ramirez, Debt Restructurings, Holdouts, and Exit Consents, 3 J. FIN. STABILITY 1, 2 (2007) (analyzing a sample of exchange and tender offers).

(121) Chatterjee et al., supra note 116, at 339.

(122) See id. at 349; (noting the effect of coercion on the increased number of tenders); Peterson, supra note 115, at 526 (noting that 65% of offers with exit consent offers close while 59% of offers without consent offers close).

(123) See Chatterjee et al., supra note 116, at 346-47 (associating senior debt with exit consents); Daniels & Ramirez, supra note 120, at 8 (associating exit consents with larger and riskier issuers). The studies are marred by over inclusion because they include cash tender offers as well as exchange offers. See, e.g., Chatterjee et al., supra note 116, at 343 (noting that the type of debt targeted, the type of a firm's financial claims, and the type of securities offered in the workout influence the severity of holdouts); Daniels & Ramirez, supra note 120, at 10, 15 (reasoning that exit consents are sought when holdout issues are not mitigated by offering bondholders more senior claims); Peterson, supra note 115, at 525 (describing the relationship among holdouts, non-consent tender offers, and exchange offers). Unsurprisingly, cash tender offers succeed more often. See id. ("We can determine the frequency of holdout problems by examining the values of non-tendered debt before and after non-consent tender and exchange offers").

(124) Patrick McGeever, An Introduction to Distressed Debt Exchanges, AAM (Feb. 18, 2016), [] (noting that the threatened increase in distressed debt exchanges has resulted in fewer "investors willing to buy unsecured debt given the lack of confidence in the future capital structure of the investment.").

(125) Hotchkiss et al., supra note 118, at 252 ("The fact that senior creditors are willing to give up a greater fraction of the firm to junior claimants in a workout suggests that on average firms attempting workouts may be less severely financially insolvent than bankrupt firms.").

(126) Chatterjee et al., supra note 116, at 352.

(127) Paul Asquith, Robert Gertner & David Scharfstein, Anatomy of Financial Distress: An Examination of Junk-Bond Issuers, 109 Q.J. ECON. 625, 640 (1994).

(128) Edward I. Altman & Brenda Karlin, The Re-Emergence of Distressed Exchanges in Corporate Restructurings, 5 J. CREDIT RISK 43, 51 (2009).

(129) This was Brudney's point. Note that additional factors will come to bear on the resulting deduction from the opportunity cost ledger. Consider the following possibility: maybe management put a low-ball haircut into the offer due to fear of holdouts; had management been free to set the haircut at the level needed for a sustainable capital structure, Chapter 11 could have been avoided.

(130) See Gilson et al., supra note 115, at 316 ("Financial distress is more likely to be resolved through private renegotiation when more of the firm's assets are intangible.").

(131) See id. at 322 (discussing factors that make a restructuring feasible).

(132) See Chatterjee et al., supra note 116, at 343 (noting that firms with heterogeneous claimants must deal with wider disagreement and are therefore harder to restructure).

(133) Banks tend to be secured, and a fully secured lender has little to fear regarding impairment of its interest in bankruptcy and so will be disinclined to make concessions outside of bankruptcy. See James, supra note 117, at 712 (noting that banks tend to only make concessions when firms are in severe financial distress). At the same time, under-secured banks are more likely to make concessions out of court precisely because their recovery in a foreclosure--the alternative to concessions--would be limited. When an under-secured bank is willing to negotiate, the bank's negotiation facilitates a parallel bondholder workout. See id. (finding that exchanges accompanied by bank concessions entail larger haircuts and junior claims, and are more likely to succeed). Banks tend to make larger investments in information retrieval, so bank participation also imports a reduction in information asymmetry which is facilitative. See Hotchkiss et al., supra note 118, at 250 ("[B]ecause banks are better informed than public debtholders, reducing potential information asymmetries, it is easier for firms with banks as dominant creditors to renegotiate their debt.").

(134) See Altman & Karlin, supra note 128, at 50 tbl.2 (showing an average recovery rate, 1984-2009, on exchanges of 50.8 cents on the dollar for out-of-court restructurings compared to 37.5 cents on the dollar for other defaults); see also Hotchkiss et al., supra note 118, at 269 (noting that restructured firms are still typically highly leveraged after Chapter 11 as compared to firms that complete out-of-court restructurings).

(135) Altman & Karlin, supra note 128, at 29.

(136) Id. at 44.

(137) Id. at 48.

(138) Id. at 49.

(139) See 11 U.S.C. [section] 363(f) (2012) (allowing the trustee to sell a property free and clear during bankruptcy if certain conditions are met).

(140) Hotchkiss et. al., supra note 118, at 250.

(141) 11 U.S.C. [section][section] 365, 1110,1113,1114 (2012).

(142) FED. R. BANKR. P. 7004.

(143) Albeit in a separate subcategory. McGeever, supra note 124.

(144) See Chatterjee et al., supra note 116, at 352 (showing a negative announcement period return of-3.28%); Gilson et al., supra note 115, at 342 (reporting a return of -1.6% for firms whose exchange offers succeed and -6.3% for firms whose offers go on to fail).

(145) See Chatterjee et al., supra note 116, at 17 (showing an announcement negative return of 0.98% for private workouts and -0.54% for public workouts).

(146) Peterson, supra note 115, at 528-30.

(147) I.R.C. [section] 108 (West 2018). To the extent that the company has taxable income and lacks loss carryovers to offset it, this tax tab has a cognizable downside. See Altman & Karlin, supra note 128, at 44. The downside is exacerbated by the fact that the same haircut, if effected in Chapter 11, creates no taxable income. I.R.C. [section] 108(a)(1)(a) (West 2018).

(148) Chapter 11 debtors are generally prohibited from soliciting consents on a Chapter 11 plan prior to the court approving a plan disclosure statement. There is an important exception to this rule, however: if the consents are solicited prior to the filing for bankruptcy, then no disclosure statement is required. 11 U.S.C. [section][section] 1125(b), 1125(g), 1126(b) (2012). This facilitates a much faster bankruptcy process in two ways. First, the plan can be filed with the bankruptcy petition, enabling a plan confirmation hearing in as little as twenty-eight days. FED. R. BANKR. P. 2002(b), 3020(b)(2). Second, because there is no requirement of a court-approved disclosure statement, there is less ability for dissident creditors to hold up the process. The solicitation of votes on a prepac must still comply with any relevant nonbankruptcy law, such as federal or state securities laws, but these disclosure regimes only create ex post liability. In contrast, bankruptcy has a merits-based disclosure regime that allows for the injunction of the dissemination of a disclosure statement that has not been preapproved by the court, which in turn enables holdup objections. 11 U.S.C. [section] 1125(b), (g) (2012).

(149) Faster timetables have been approved in court. See e.g., Andriana Georgallas, Chapter 22 Commencement to Confirmation in Just 6 Days: Exploring Roust Corporation, WEIL BANKRUPTCY BLOG, [] (discussing the recent trend of quick confirmation of prepacking plans, including In re Roust Corp., which was approved in six days, and In re Blue Bird Body Co., which was confirmed within 24 hours).

(150) Brian L. Betker, The Administrative Costs of Debt Restructurings: Some Recent Evidence, 26 FIN. MGMT. 56, 62 (1997).

(151) Elizabeth Tashjian, Ronald C. Lease & John J. McConnell, Prepacks: An Empirical Analysis of Prepackaged Bankruptcies, 40 J. FIN. ECON. 135,143 (1996).

(152) Hotchkiss et al., supra note 118, at 262.

(153) Lawrence A. Weiss, Bankruptcy Resolution: Direct Costs and Violation of Priority of Claims, in CORPORATE BANKRUPTCY: ECONOMIC AND LEGAL PERSPECTIVES 260, 263 (Jagdeep S. Bhandari & Lawrence A. Weiss eds., 1996)

(154) Arturo Bris, Ivo Welch & Ning Zhu, The Costs of Bankruptcy: Chapter 7 Liquidation versus Chapter 11 Reorganization, 61 J. FIN. 1253,1272 (2006).

(155) See Franks & Torous, supra note 115, at 354 (showing that average Chapter 11 reorganization took twenty-seven months).

(156) Edward I. Altman & Brenda J. Kuehne, Defaults and Returns in the High-Yield Bond and Distressed Debt Market: Review and Outlook, in CONTEMPORARY CHALLENGES IN RISK MANAGEMENT: DEALING WITH RISK, UNCERTAINTY AND THE UNKNOWN 203, 212 (T. Andersen & Stanley Mayes eds., 2014).

(157) Edward I. Altman, The Role of Distressed Debt Markets, Hedge Funds and Recent Trends in Bankruptcy on the Outcomes of Chapter 11 Reorganizations, 22 AM. BANKR. INST. L. REV. 75, 92 fig.7 (2014). The prepac percentage varies widely. In this century, the low was 3.4% (2001) and the high was 76.9% (2013). The average since 1988 is 11.6%. Id.

(158) Altman & Karlin, supra note 128, at 52.

(159) Since late 2005, debtors also face a hard cap on the period in which they have the exclusive right to propose a plan. See Bankruptcy Abuse Prevention and Consumer Protection Act of 2005, Pub. L. No. 109-8, [section] 411 (codified at 11 U.S.C. [section] 1121(d)) (prohibiting extensions of the period in which parties can file plans).

(160) See infra text accompanying note 194.

(161) Altman, supra note 157, at 92 fig.7.

(162) See Bratton, supra note 103, at 479 ("Cov[enant]-lite loans resemble junk bonds, containing restrictive covenants but no maintenance tests.").

(163) Altman & Karlin, supra note 128, at 46 (describing the "resurgence in the incidence of distressed exchanges").

(164) Id.

(165) Id.

(166) Altman & Kuehne, supra note 156, at 226 tbl.8.7.

(167) See I.R.C. [section] 108(i)(1)(A)-(B) (West 2018) (describing qualification for deferral and ratable inclusion of business indebtedness income).

(168) Altman & Kuehne, supra note 156, at 226.

(169) Id.

(170) Id. at 225; Altman & Karlin, supra note 128, at 46.

(171) See Bernard Black & Henry Hu, Equity and Debt Decoupling and Empty Voting II: Importance and Extensions, 156 U. PA. L. REV. 625, 728-35 (2008) (examining and explaining the rise of the "empty creditor problem").

(172) At least since ISDA's "Big Bang" Protocol of 2009. See Helen Haworth, A Guide to Credit Events and Auctions, CREDIT SUISSE 5-8 (Jan. 12, 2011) (noting that "hard credit events," like bankruptcy or failure to pay, trigger CDS obligations but that "[v]oluntary debt restructurings may or may not trigger CDS contracts").

(173) The difference is amount paid under the swap plus the value of the defaulted bond.

(174) David Mengle, ISDA Research Notes: The Empty Creditor Hypothesis ISDA 1 (2009) (noting that "the evidence available thus far, in the form of restructuring choices by distressed firms as well as in market practices surrounding credit derivatives, does not appear favorable to the empty creditor hypothesis").

(175) Id. at 9 (arguing that "the correlation statistics" relating the number of defaults to restructurings as a percentage of defaults "would not appear to support the empty creditor hypothesis").

(176) Id. at 10 (noting that "it is not clear, however, how an overhedging strategy might be exploited systematically in a way that could distort credit markets or the bankruptcy process").

(177) Andras Danis, Do Empty Creditors Matter? Evidence from Distressed Exchange Offers 16-18 (Institute of Economics, Centre for Economic and Regional Studies, Working Paper No. 1334, Apr. 24, 2013) (describing data showing that "the participation rate is on average 29.1 percentage points lower if the issuer is a reference entity in the CDS market").

(178) Mascia Bedendo, Lara Cathcart & Lina El-Jahel, Distressed Debt Restructuring in the Presence of Credit Default Swaps, 48 J. MONEY, CREDIT, AND BANKING 165, 167 (Feb. 2016) ("Contrary to the empty creditor argument, we do not find evidence that companies whose bondholders might be insured via CDSs are more likely to restructure their debt in court. In fact, the restructuring choice is driven by essentially the same variables in both reference entities and nonreference entities.").

(179) We ran the search through the Wharton Research Data Services SEC Analytics Suite.

(180) We excluded offers paying all cash at a premium over market price and offers that shifted obligations to different entities within a corporate group without modifying core financial terms. This left us with forty-four issuers. We added two issuers whose exchange offers have been the subject of litigation--Education Management Corp. and Chesapeake Energy Corp.--and worked the details up with press releases.

(181) See supra note 117 and accompanying text.

(182) See supra notes 115-16 and accompanying text.

(183) See Anna Pinedo & Remmelt Reigersman, Debt Repurchases & Exchanges, MORRISON & FOERSTER 1, 47 (Nov. 8, 2012), [] (noting that "[p]rivate exchange offers conducted pursuant to Section 4(2)... [m]ay not constitute a 'general solicitation' and must be made only to 'sophisticated investors,' usually qualified institutional buyers (QIBs).").

(184) See 17 C.F.R. [section] 230.144A(a)(i)-(iv) (2013) (defining QIBs).

(185) See Pinedo & Reigersman, supra note 183, at 48 (noting that parties engaging in "private exchange offers" should "[i]dentify and pre-certify (QIB, accredited investor status) investors").

(186) See Altman & Karlin, supra note 128, at 44 (noting that "The first instances of distressed exchanges in the modern high-yield bond era were the so-called 3(a)9. exchanges championed by Drexel Burnham Lambert in the 1980s.").

(187) See, e.g., Section 3(a)(9) Exchange Offers, MORRISON & FOERSTER 9 (June 5, 2009), [] (noting that "the SEC Staff has indicated through various no-action letters that a financial adviser may receive a fixed fee for its services, not contingent upon the success of the exchange, plus reasonable expenses related to the exchange").

(188) See id. at 6 (listing the "impermissible activities" for third parties such as financial advisors).

(189) See Robert M. Mooradian & Harley E. Ryan, Jr., Financial Distress: Section 3(a)(9) Compared to Investment Bank Managed Exchange Offers, 78 J. BUS. 1593, 1596-97 (2005) (explaining that investment bankers make a difference because they deal directly with the bondholder offerees).

(190) Altman, supra note 157, at 78.

(191) Id.; see also Press Release, Cumulus, Cumulus Enters Restructuring Support Agreement to Reduce Debt by Over $1 Billion (Nov. 29, 2017) (on file with University of Pennsylvania Law Review) (describing restructuring support agreement and closing with mention of a law firm, a financial advisor, and a restructuring advisor).

(192) See Douglas G. Baird & Robert K. Rasmussen, Chapter 11 at Twilight, 56 STAN. L. REV. 673, 675 (2003) (noting that "if the experience of large businesses leaving Chapter 11 in 2002 is any guide, those at the helm do the bidding of the creditors throughout the case"); Douglas G. Baird & Robert K. Rasmussen, The End of Bankruptcy, 55 STAN. L. REV. 751, 784 (2002) (asserting that "[t]he revolving credit facility, installed as the firm begins to have trouble making debt payments, also gives the lender who runs it the ability to control the firm inside of Chapter 11 as well as out"); David A. Skeel, Jr., Creditors' Ball: The "New" New Corporate Governance in Chapter 11,152 U. PA. L. REV. 917, 919 (2003) (noting that "lenders are using [DIP] loan contracts to influence corporate governance in bankruptcy"); cf. Lynn M. LoPucki, The Nature of the Bankrupt Firm: A Response to Baird and Rasmussen's The End of Bankruptcy, 56 STAN. L. REV. 645, 666 (2003) (arguing that "boards of directors rather than DIP lenders control most large, public firms in reorganization").

(193) See Kenneth M. Ayotte & Edward R. Morrison, Creditor Control and Conflict in Chapter 11, 1 J. LEGAL ANALYSIS 511, 523 (2009) (noting the large percentage of firms whose assets are "encumbered by liens"). A revision of Article 9 of the Uniform Commercial Code that went into effect in 2001 facilitated this change. Prior to 2001, creditors had jurisdictionally spotty and uncertain rights to take and maintain a security interest in deposit accounts. Since 2001, the Uniform Commercial Code has expressly permitted security interests in deposit accounts. See U.C.C. [section] 9-104 (listing the ways in which secured parties can control deposit accounts).

(194) The cash is likely to be the proceeds of a secured creditor's collateral and thus also subject to the lien. U.C.C. [section] 9-315; see also 11 U.S.C. [section] 552(b)(1) (2012) (providing that postpetition, a security interest "extends to such proceeds, products, offspring, or profits acquired by the estate after the commencement of the case"). Bankruptcy law prohibits the debtor from using this "cash collateral" unless it either provides the secured creditor with "adequate protection" of its interest in the collateral, something that is usually impossible given the debtor's lack of unencumbered assets, or obtains the consent of the secured creditor for the use of the cash. See, e.g., 11 U.S.C. 363(c)(2) (2012) (prohibiting trustee from using the cash collateral except in two narrow situations); see also 11 U.S.C. [section] 361(1) (2012) (noting that "adequate protection may be provided by requiring the trustee to make a cash payment or periodic cash payments").

(195) The Bankruptcy Abuse Prevention and Consumer Protection Act of 2005, codified in the Bankruptcy Code at 11 U.S.C. [section] 1121(d)(2)(A), sets limits on the time during which the debtor can file plans.

(196) See, e.g., LEVITIN, supra note 21, at 797-99 (noting that "[t]he right to appoint a chief restructuring officer is a striking measure of DIP lender power"); Sris Chatterjee, Upinder S. Dhillon & Gabriel G. Ramirez, Debtor-in-Possession Financing, 28 J. BANKING & FIN. 3097, 3107 (2004) (concluding that "DIP loans consistently include restrictive covenants that are similar to covenants of bank loans, presumably the type of pre-petition loans held by lenders"); see also Sreedhar T. Bharath, Venkatesh Panchapagesan & Ingrid M. Werner, The Changing Nature of Chapter 11 20 (Fisher Coll. of Bus., Working Paper No. 2008-03-003, Nov. 2010) (finding a 65% increase in management turnover rates in bankruptcy between 1990 and the early 2000s, with a 200% increase in the turnover rate for managers with significant equity holdings).

(197) An increasing percentage of plan distributions appear to comply with the absolute priority rule, a marker of diminished managerial power in bankruptcy. Compare Ayotte & Morrison, supra note 193, at 523 (showing APR violations in 12% of Chapter 11 plans, and payments to equityholders in 18% of plans for cases filed in 2001) and Vedran Capkun & Lawrence A. Weiss, Bankruptcy Resolution and the Restoration of Priority of Claims 12 (HEC Paris Research Paper No. ACC-2016-1160, 2016) (reporting APR violations in 37% of recent cases), with Lawrence A. Weiss, Bankruptcy Resolution: Direct Costs and Enforcement of Claims, 2 J. LEGAL ECON. 79, 85 (1992) (citing APR violations in 78% of 37 cases studied) and Julian R. Franks & Walter N. Torous, An Empirical Investigation of U.S. Firms in Reorganization, 44 J. FIN. 747, 754 (1989) (finding APR violations in 78% of cases in the 1980s).

(198) Neither the best interest test, which guarantees that creditors receive at least as much value in Chapter 11 as in a Chapter 7 liquidation, nor the absolute priority rule, which protects unsecured creditors from value being diverted to equity, matter because creditors receive their liquidation rights, and nothing more. See 11 U.S.C. [section] 1129(a)(7) (2012) (stating the best interest test); 11 U.S.C. [section] 1129(b)(2)(A)(ii)-(b)(2)(C) (2012) (stating the absolute priority rule). Likewise, plan feasibility, as stated in 11 U.S.C. [section] H29(a)(n), does not matter because in this new world, the Chapter 11 plan does not promise to pay junior creditors anything. And the Chapter 11 good faith requirement, as stated in 11 U.S.C. [section] 1129(a)(3), has little bearing on such a plan.

(199) These include "new value" plans that would let old equity holders retain the equity in the reorganized company in exchange for a non-market-tested contribution of new value. Under these "gift plans," seniors agree to pay off out-of-the-money equity in order to facilitate a quick plan confirmation, and restrictions on credit bidding at asset sales are designed to prevent senior secured lenders from purchasing assets. Courts, however, have held that these ploys violate various provisions of the Bankruptcy Code. See RadLAX Gateway Hotel, LLC v. Amalgamated Bank, 132 S. Ct. 2065, 2070-73 (2013) (prohibiting restriction of credit bidding under 11 U.S.C. [section] 1129(b)(2)(A)(1)(iii)); Bank of Am. Nat'l Tr. & Sav. Ass'n v. 203 N. LaSalle St. P'ship, 526 U.S. 434, 458 (1999) (prohibiting cramdown confirmation of new value plans that violate absolute priority when there is not a market test); In re DBSD Nor. Am. Inc., 634 F.3d 79, 93-101 (2d Cir. 2016) (prohibiting plan that would have violated absolute priority rule); In re Armstrong World Indus., 432 F.3d 507, 514-15 (3d Cir. 2005) (refusing to confirm plan that violated absolute priority rule).

(200) See Barry E. Adler, Vedran Capkun & Lawrence A. Weiss, Value Destruction in the New Era of Chapter 11, 29 J. L. ECON. & ORG. 461, 479 (2013) (showing that the shift to creditor control may cause management to "aggressively" delay filing).

(201) Such a case would exist where there is no secured creditor or where the secured creditor has a limited claim or lien. The unsecured creditors would likely take the downsides of DIP-financed Chapter 11 into account in evaluating the present giveups bound up in a workout designed to return the issuer to sustainability.

(202) See generally Douglas G. Baird, Bankruptcy's Quiet Revolution (Univ. of Chi. Coase-Sandor Inst, for Law & Econ., Research Paper No. 755, 2016) (detailing the "quiet revolution" of restructuring support agreements and their impact on modern bankruptcy).

(203) See, e.g., Brudney, supra note 13, at 1826-27 (describing the systematic difficulties that public bondholders face in bankruptcy).

(204) See supra Figure 1 (showing decrease in household bondholdings).


(206) See Liz Moyer, Wall Street's Debt Restructuring Fight Heads to Washington, N.Y. TIMES: DEALBOOK (Dec. 7, 2015), (describing how proposed amendment to Trust Indenture Act would "hand a victory to Apollo Global Management"); Matt Jarzemsky, Caesars Takes Aim at Law Aiding Creditors, WALL ST. J. (Dec. 6, 2015, 9:52 PM), (noting how Apollo Global Management supported legislation to amend the TIA in order to "gut" lawsuits brought by bondholders of Caesars Entertainment). We note that we both were signatories to a legal scholar's letter to Congressional leadership opposing the proposed amendment of the TIA through the appropriations process.

(207) See Marblegate Asset Mgmt., LLC v. Educ. Mgmt. Fin. Corp., 846 F.3d 1,3 (2d Cir. 2017) (holding that "[s]ection 316(b) prohibits only non-consensual amendments to an indenture's core payment terms").

(208) See Roe, supra note 104, at 273-74 (noting that "[s]ignalling and agency costs" are two reasons why managers would avoid "majority action clauses").

(209) See id. at 270 (noting that "even the votes now permitted can be structured to distort bondholder choice by offering assenting bondholders side payments").

(210) Trust Indenture Act section 304(d), amended by 15 U.S.C. [section] 77ddd(d) (2012), provides that "the Commission may, by rules or regulations upon its own motion... exempt conditionally or unconditionally any person, security or transaction, or any class or classes of persons, registration statements, indentures, securities, or transactions, from any one or more of the provisions of this subchapter, if and to the extent that such exemption is necessary or appropriate in the public interest and consistent with the protection of investors and the purposes fairly intended by this subchapter." The section's legislative history also allows this on a literal read, although the examples stated therein are of a considerably lesser magnitude than the exemption Roe proposes. The reports and testimony variously describe the amendments to section 304(d) as facilitating adjustments to new developments in debt security design and underwriting, citing in particular the appearance of securitization. See Trust Indenture Reform: Hearing on H.R. 1786 Before the Subcomm. on Telecomms. and Fin. of the Comm. on Energy and Commerce, 101" Cong. 28 (1989) (statement of David S. Ruder, Chairman, Securities and Exchange Commission), microformed on Sup. Docs. No. Y 4.En 2/3:101-48 (U.S. Gov't Printing Office) (supporting the Trust Indenture Reform Act of 1989 on the theory that "Changes in the types of debt securities, in the methods of public financing, and in the character of relations between obligors and their financial intermediaries have created situations not anticipated when the act was passed in 1939.").

(211) See Mark J. Roe, Commentary, The Trust Indenture Act of 1939 in Congress and the Courts in 2016: Bringing the SEC Back to the Table, 129 HARV. L. REV. F. 360, 372 (2016) (recommending that the SEC "exempt bond indentures from section 316(b) if they (1) provide for a binding vote on payment terms approved by two-thirds of the bondholders, without the vote of any conflicted bondholder, and (2) bar coercive transactions such as exit-consent exchange offers"). This is the second iteration of the proposal. In the 1987 paper, Roe suggested an SEC rulemaking pursuant to which core indenture terms could be amended by a two-thirds vote and exit consents would be prohibited; in addition, conflicted voting would be addressed through the interpolation of the Bankruptcy Code's rules on voting disqualification, and second, by self-reporting of insider status by the bondholders. See Roe, supra note 104, at 270-71 (proposing a "new Rule 316").

(212) See Marcel Kahan, The Qualified Case Against Mandatory Terms in Bonds, 89 Nw. U. L. REV. 565, 622 (1995) (concluding that "no mandatory prohibition should be imposed.").

(213) See id. at 603 (describing how "the collective action problems for bondholders are not as great as they are for shareholders").

(214) See id. at 579 (noting that "studies of informational efficiency in the bond market establish that at least some legal terms are priced").

(215) Kahan cited his own study of covenant amendments, which found that less than half of the solicitations succeeded as originally proposed; that in 42% of the cases bondholder resistance resulted in improved terms; and that 17% of the solicitations failed altogether. Id. at 604 (citing Marcel Kahan & Bruce Tuckman, Do Bondholders Lose from Junk Bond Covenant Changes?, 66 J. BUS. 499 (1993)).

(216) See id. at 605 (noting that the coercion argument "lacks any plausibility in the bondholder context").

(217) See No. 99-10517,1999 WL 993648, at *1 (S.D.N.Y. Nov. 2,1999) (discussing a restructuring effected by an exchange offer and exit consents that moved assets out of obligor entities and lifted guaranties of the bonds).

(218) Id. at *7.

(219) See generally N.Y. BAKER BOTTS, LLP, ET AL., OPINION WHITE PAPER ON RECENT JUDICIAL OPINIONS RELATING TO TIA SECTION 316(B) (Apr. 25, 2016), available at [] (providing guidance in the interpretation of section 316(b)).

(220) See, e.g., YRC Worldwide Inc. v. Deutsche Bank Tr. Co. Am., No. 10-2106, 2010 WL 2680336, at *6 (D. Kan. July 1, 2010) (agreeing with other courts that section 316(b) applies to holders' legal, not practical, rights); In re Nw. Corp., 313 B.R. 595, 600 (Bankr. D. Del. 2004) (holding that the Trust Indenture Act of 1939 protects "the holder's legal rights and not the holder's practical rights to the principal and interest itself"); see also Upic & Co. v. Kinder-Care Learning Ctrs., Inc., 793 F. Supp. 448, 455-57 (S.D.N.Y. 1992) (opining that section 316(b) guarantees the procedural right to sue on the promise but does not "alter the substance of a noteholder's right to payment of principal and interest under the Indenture and, in particular, cannot 'override' the Indenture's subordination provisions").

(221) 75 F. Supp. 3d 592 (S.D.N.Y. 2014), rev'd, 846 F.3d 1 (2d Cir. 2017).

(222) See 80 F. Supp. 3d 507, 516 (S.D.N.Y. 2015) (upholding the broad interpretation of section 316(b) using Marblegate reasoning).

(223) See 144 F. Supp. 3d 459, 473 (S.D.N.Y. 2015) (noting that section 316(b) must be interpreted in accordance with "the purpose underlying the provision--allow[ing] for corporate flexibility while protecting minority bondholders").

(224) See Marblegate Asset Mgmt., LLC v. Educ. Mgmt. Fin. Corp., 846 F.3d 1, 17 (2d Cir. 2017) (reversing and remanding the judgment of the district court).

(225) See id. at 8 (noting that the District Court understood the legislative history to reveal "Congress's broad intent to prohibit an out-of-court debt restructuring that has the purpose and effect of eliminating any possibility of receiving payment under their notes."). The court fairly characterized the language of section 316(b) as ambiguous. A cursory look at the legislative history limited to the "evasion of judicial scrutiny" line from the committee reports followed. See supra text accompanying note 56 (discussing House and Senate reports containing "evasion of judicial scrutiny" language). Effectuation of the purpose, said the court, meant reading the section to protect "the ability, and not merely the formal right, to receive payment in some circumstances." See Marblegate, 846 F.3d at 5 (summarizing district court's analysis of section 316(b) when deciding whether to grant preliminary injunction).

(226) See id. at 9 ("Based on our review of the legislative history of Section 316(b), we conclude that Congress did not intend the broad reading that Marblegate urges and the District Court embraced."). The appellate panel drilled down into the legislative history and came up with a very different story. Yes, Congress wanted to push restructuring into bankruptcy. But Congress only took a limited step in pursuing the goal, to wit, prohibiting "non-consensual amendments of core payment terms (that is, the amount of principal and interest owed, and the date of maturity.)." Id. at 7. The Court reviewed the famous SEC Report, see supra text accompanying note 58, and found no mention of an objective to require unanimous consent to all out-of-court restructurings. The Court went on to consider the congressional committee process, focusing on the testimony of SEC chair William O. Douglas in 1938 and the assistant director of the SEC's reorganization division, Edmund Burke, Jr., in 1939. This was the pay dirt. Douglas had stressed that the section covered only amendment and waiver of payment terms, and Burke has noted that the barrier went only to an unconsented "variation from [the] contract." Marblegate, 846 F.3d at 12.

(227) See Marblegate Asset Mgmt. v. Educ. Mgmt. Corp., 75 F. Supp. 3d 592, 595 (S.D.N.Y. 2014) (noting that EDMC depended on this source of revenue because it accounted for 78.6% of its revenues).

(228) See id. at 597 (explaining that EDMC's $1,553 billion in outstanding debt "consists of $1,305 billion in secured debt, divided between $220 million drawn from a revolving credit facility and $1,085 billion in term loans, and $217 million in unsecured loans").

(229) See id. at 597-98 (discussing release of the guarantor).

(230) See id. at 600-01 (detailing the specific requirements of the restructuring).

(231) See id. (noting that restructuring could only proceed with the consent of 100% of the creditors and Plaintiffs' decision not to participate in the Exchange Offer).

(232) See id. at 601-02 (describing the "Intercompany Sale" procedure).

(233) See id. at 601-02 (quoting the Exchange Offering Circular, which clearly spelled out the financial consequences of not tendering all notes).

(234) See id. at 603 (describing scope of relief sought by plaintiffs).

(235) See id. at 605-11 (denying Plaintiffs' motion for a preliminary injunction but explaining why "Plaintiffs have demonstrated a likelihood of success on the merits").

(236) See id. at 615 ("But where a debt reorganization that seeks to involuntarily disinherit the dissenting minority is brought about by a majority vote, that violates the fundamental purpose of the Trust Indenture Act.").

(237) See id. at 616 (arguing that "Plaintiffs may have been warned that modifications were possible, but they were not told that they could be forced to accept a wholesale abandonment of their right to receive payment.").

(238) We note, however, that the TIA defines "obligor" to include "guarantor," suggesting that if the right to payment is from an obligor, it includes a guaranty of that payment as well. 15 U.S.C. [section] 77ccc(i2) (2012). TIA section 316(b) does not use the term "obligor," however. 15 U.S.C. [section] 77ppp(b) (2012).

(239) See Marblegate Asset Mgmt. v. Educ. Mgmt. Corp., 75 F. Supp. 3d 592, 596 (S.D.N.Y. 2014) (noting that "an institution loses its eligibility for Title IV funds if it, or a controlling affiliate, files for bankruptcy").

(240) See id. at 601 (noting the broad support for the restricting plan).

(241) See MeehanCombs Glob. Credit Opportunities Master Fund, LP v. Caesars Entm't Corp., 80 F. Supp. 3d 507, 510 (S.D.N.Y. 2015) (describing the notes issued by Ceasers Entertainment Operating Company, a Ceasers' subsidiary, prior to Ceasers' acquisition in a leveraged buyout).

(242) See id. at 511 (describing the details of the parent's attempt to transfer assets while retaining company debt).

(243) See Complaint at 1 5, MeehanCombs Glob. Credit Opportunities Master Fund, LP v. Caesars Entm't Corp., 80 F. Supp. 3d 507 (S.D.N.Y. 2015) (No. 14-7091) (noting two of the Note Purchase and Support Agreement's "critical provisions": "(1) the Guarantees by CEC for payment of principal and interest, and (2) a covenant in the governing Indentures that would prevent Caesars from transferring all or substantially all of CEOC's assets").

(244) See id. (noting that the Favored Noteholders were required to amend the agreement).

(245) See id. at [paragraph] 72 (noting that this price was "more than a 100% premium over market").

(246) See Caesars I, 80 F. Supp. 3d 507, 516 (S.D.N.Y. 2015) (holding that "the Complaint's plausible allegations that the August 2014 Transaction stripped plaintiffs of the valuable CEC Guarantees leaving them with an empty right to assert a payment default from an insolvent issuer are sufficient to state a claim under section 316(b)").

(247) See BOKF, N.A. v. Caesars Entm't Corp., 144 F. Supp. 3d 459, 463 (S.D.N.Y. 2015) (noting that "[a]ccording to CEC, because CEOC was no longer a wholly owned subsidiary, the Guarantee was automatically terminated under section 12.02(c)(i) of the Indentures.").

(248) See id. (noting that "CEC sold five percent of CEOC's common stock to certain institutional investors").

(249) See id. at 464 (reciting that "CEC authorized the CEOC Board to adopt a 2014 stock performance incentive plan, which enabled CEOC to grant shares of CEOC stock to its directors and officers").

(250) See id. at 473 (characterizing CEC's view as that "plaintiffs must establish a restructuring of their particular debt" to prove impairment).

(251) See id. at 465-466 (holding that 316(b) protects noteholders against impairment of the right to receive payment); see also id. at 475 (permitting "limited discovery" to "determine whether a restructuring occurred").

(252) The action dates back to Zeffiro v. First Pa. Banking & Tr. Co., 623 F.2d 290 (3d Cir. 1980), which found that the Trust Indenture Act grants a cause of action to injured investors. Id. at 292.

(253) See Brian Trust, New York State's Highest Court Clarifies Scope of 'No-Action' Clause Under Trust Indenture, MAYER BROWN (June 18, 2014), [] (describing operation and purpose of no-action clauses).

(254) The search was run through the Wharton Research Data Services SEC Analytics Suite.

(255) A simple majority is employed across the board. A half century ago the rule of thumb was two-thirds. See AMERICAN BAR FOUNDATION, supra note 56, at 305-06 (including a two-thirds CAC and commenting that the two-thirds threshold is "required generally").

(256) Realogy Group LLC, Indenture (Exhibit to Form 8-K) [section] 6.07 (June 3, 2016), available at [].

(257) PVH Corp., Indenture (Exhibit to Form 8-K) [section] 6.07 (June 20, 2016), available at [https ://].

(258) Still four more employ this clause in addition to employing the other techniques.

(259) The correspondences are not quite perfect, more likely indicating sloppy use of forms than intentional omission. For example, out of twenty-two convertible issues, one omits to put the provisions under the UAC. The numbers respecting guaranties are less thorough--88.2% in 2011, 77% in 2016--and conceivably could support an inference of issuer pushback.

(260) The parameters have widened since the drafters of the Model Indenture reported in 1971. Their amendment clause includes payment terms only, with the note mentioning practice extension only to redemption, conversion, and liens. See AMERICAN BAR FOUNDATION, supra note 56, at 305-09.

(261) See supra text accompanying notes 54-55.

(262) Professors Coffee and Klein suggested that bonds submitted with exit consents could be deemed "owned" or "controlled" by the bond issuer, running afoul of both the trust indenture's voting provisions and TIA section 316(a). See Coffee & Klein, supra note 34, at 1257 (explaining that "at the time the amendment is voted upon, the obligor does not 'own' the bonds; they are still owned by the person who has tendered them"). The idea was to read "owned" in 316(a) broadly to include control, and it suggested an economic reality test looking to beneficial interest, not just legal ownership, thereby sweeping exit consents into the prohibition. See id. at 1258 (noting that "[o]ne could, of course, torture the interpretation of 'ownership' to produce a rule allowing the use of exit consents for benign, but not for pernicious, purposes").

(263) The theory that the issuer owns bonds tendered with exit consents proved persuasive in a UK case, Assenagon Asset Mgmt. S.A. v. Irish Bank Resolution Corp., [2012] EWHC 2090 (Ch) (Eng). The indenture provision in question provided that "[n]either the Issuer nor any Subsidiary shall be entitled to vote at any meeting in respect of Notes beneficially held by it for its account." Id. at [paragraph] 16. Taking a formalist approach, the High Court of Justice ruled that by the time the votes garnered under the exit consents were cast at a later bondholders' meeting, the bonds were beneficially held by the issuer because the contract of sale contained a specific performance provision and a refusal to deliver the bond would not be adequately compensated by monetary damages. Id. at [paragraph][paragraph] 62-67.

The Caesars I plaintiffs tried this out, arguing that the issuer either owned or controlled the bonds redeemed from the favored bondholders and thus could not vote those bonds. Meehancombs Glob. Credit Opportunity Funds, LP v. Caesars Entm't Corp., 80 F. Supp. 3d 507, 516-17 (S.D.N.Y. 2015). The court rejected the theory, taking the same formalist tack as did the earlier Assenagon court. See id. at 517 (holding that Plaintiff's "control allegations... are insufficient" because their complaint "does not allege that the Participating Noteholders were anything other than unaffiliated, independent third parties that entered into an arm's length transaction to provide their consents."). But this time the background was different, for the consents were delivered before the issuer took ownership of the bonds. See id. (noting that "the August 2014 Transaction was structured so that the Favored Noteholders' consents were given before the notes were sold to Caesars, [so] MeehanCombs does not allege ownership."). Hence, there were no "owned" votes; nor were the votes under issuer "control," for the voters were "unaffiliated, independent third parties" acting at arm's length. Id.

(264) See Coffee & Klein, supra note 34, at 1257 (asserting that "[i]t would be possible, of course, to devise language that would clearly foreclose or limit the exit consent procedure"). Coffee and Klein also suggested that the contractual duty of good faith, long somnolent in the area of relationships between issuers and bondholders and within groups of bondholders, could come to bear effectively if only courts would wake up and bring it to bear. See id. at 1258 (recommending that courts "generate a rule consistent with good results by invoking one of the two judicial constructs described below: the implied covenant of good faith and fair dealing or the requirement of good faith voting").

(265) See BRATTON, supra note 19, at 351-53 (explaining that issuers pay for redemption rights with higher interest rate).

(266) See Bratton & Gulati, supra note 23, at 4 (noting that after Argentina's experience, "the Treasury projects that sovereign bondholders will willingly exchange their UAC bonds for CAC bonds, ameliorating the coordination problems").

(267) See id. at 7, 12 (noting "the persistent use for more than a century of UACs in New York-based sovereign debt issues and CACs in London-based debt issues").

(268) See id. at 20-21 (noting that "[o]f Argentina's $111.8 billion of foreign bonds, 89 percent were issued pursuant to debt contracts containing unanimous action clauses").

(269) Argentina effected a take-it-or-leave-it exchange offer implicating a substantial haircut, but left outstanding a sizable population of holdout bonds. A several-year standoff followed as the holdouts, led by hedge funds, pursued their U.S. law remedies where they could, finally vindicating their right to payment in the U.S. Court of Appeals for the Second Circuit in 2012. See NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 250 (2d Cir. 2012) (holding that "an equal treatment provision in the bonds bars Argentina from discriminating against plaintiffs' bonds in favor of bonds issued in connection with the restructurings and that Argentina violated that provision by ranking its payment obligations on the defaulted debt below its obligations to the holders of its restructured debt"). Only in 2016 (when execution of the federal court judgment threatened to disrupt Argentina's payments to its cooperative creditors) did Argentina come to the table and settle, giving the holdouts the last laugh. But by then, the contracting practice had changed substantially, with supermajority CACs becoming the sovereign debt standard. See, e.g., Alexandra Stevenson, How Argentina Settled a Billion Dollar Dispute with Hedge Funds, N.Y. TIMES (Apr. 25, 2016) (summarizing highlights of the negotiation between Argentina and its debtholders); see also Josh Macfarlane, Recent Development, Could Collective Action Clauses Have Saved Argentina's Presidential Plane?, HARV. INTL L.J., Apr. 1, 2016, (noting that Argentina might have avoided its problems "had Argentina's sovereign bonds included a novel contract provision known as a Collective Action Clause").

(270) See Bratton & Gulati, supra note 23, at 46 (noting that "majority action enforcement clauses," alongside CACs, "make[] restructuring easier by relaxing the unanimity requirement and makes it harder for maverick creditors, in search of preferential payments, to bring disruptive litigation").

(271) The borrowers had assumed that, given issuance in New York, bondholders' marked preference for UACs entailed a beneficial give back in the form of a lower interest rate.

(272) Ben Emons, Collective Action Clauses: No Panacea for Sovereign Debt Restructurings, PIMCO, Oct. 2010, at 1, 1 (describing CAC diffusion after the Argentinian default).

(273) We received a report from a practitioner that language barring selective offers of consideration has appeared in recent indentures. We did not run across such a clause in our own sample, however.

(274) [2012] EWHC (Ch) 2090.

(275) Id. at [paragraph] 32. The amendment inserted a right to redeem for [euro]0.01. Procedurally, Assenagon was on a Part 8 claim, the England and Wales equivalent to a motion for summary judgment in which there is no material dispute as to facts. See Summary judgment: A Quick Guide, THOMSON REUTERS PRACTICAL LAW, (noting that "[s]ummary judgment is available in Part 8 proceedings although the nature of such proceedings may mean that it is not necessary") (last visited Mar. 14, 2018).

(276) 88 U.S. (21 Wall.) 616, 621 (1874).

(277) Id. at 622.

(278) Id.

(279) Shaw v. R.R. Co., 100 U.S. 605, 612 (1880).

(280) 74 F. 110(2d Cir. 1896).

(281) Id. at 110.

(282) Id. at 111.

(283) Id.

(284) Id. at 112.

(285) Id. While duty qualified the principle of majority rule, it did not prevent majorities from binding minorities. See, e.g., First Nat'l Bank of Cleveland v. Shedd, 121 U.S. 74, 86-87 (1887) (applying Shaw v. R.R. Co.); Shaw v. R.R. Co., 100 U.S. 605, 612 (1880) (noting that "it is not improper that [minorities] should be governed by the voice of the majority"); In re Schommer, 112 F.2d 311, 314-315 (7th Cir. 1940) (following the approach laid out in Shaw v. R.R. Co.); see also Crosthwaith v. Moline Plow Co., 298 F. 466, 469 (S.D.N.Y. 1924) (denying minority bondholder's suit in contravention of collective action clause to challenge equity reorganization). The duty went together with a requirement that the benefits of the majority's decision be shared by all the holders of the same security. See id. (noting that "it would be most inequitable to allow a small minority of bondholders, or a comparatively insignificant number of creditors, in the absence of even any pretense of fraud or unfairness, to defeat the wishes of the overwhelming majority of those associated with them in the benefits of their common security").

(286) See Charles H. Haines, Jr., Comment, Modification of Corporate Mortgages and Trust Indentures, 38 MICH. L. REV. 63, 67 (1939) ("But granted an exercise within the explicit power of the majority, still the courts will inquire into the manner and circumstances of its exercise before holding the minority bound by their irrevocable assent to the alterations. The approach of the courts is that the majority are in a fiduciary relationship to the minority, with a power in trust to be used only for the common good of all.").

(287) The English principle derives from partnership and majority-minority shareholder cases. See Blisset v. Daniel [1853] EWHC (Ch) 10 Hare 493, 523-24 (describing duty in partnership); In re Westbourne Galleries [1973] AC 360 (HL) 381 (evaluating conduct of shareholders in limited company); O'Neill v. Phillips [1999] 1 WLR 1092 (HL) 1098-1101 (discussing duties owed by shareholders in limited company); Allen v. Gold Reefs of West Africa [1900] EWHC (Ch) 656, 671 (same). It imposes a good faith limitation on the power of a creditor majority to bind a minority. See, e.g., Brit. Am. Nickel Corp. v. M. J. O'Brien [1927] AC 369 (PC) 371 (noting that majority shareholder action is subject to the general principle that "power given must be exercised for the purpose of benefiting the class as a whole, and not merely individual members"); Redwood Masterfund Ltd. v. TD Bank Europe Ltd. [2006] 1 BCLC 149 (holding that majoritarian amendment of the terms of a syndicated loan facility was not in bad faith). Any votes not cast in good faith are to be disregarded. In re Empire Mining Co., [1890] EWHC 44 (Ch) 403; In re Wedgwood Coal & Iron Co., [1877] EWHC 6 (Ch) 627.

(288) Assenagon Asset Mgmt. S.A. v. Irish Bank Resolution Corp., [2012] EWHC (Ch) 2090, [paragraph] 37.

(289) Id. [paragraph][paragraph] 40-47.

(290) Id. [paragraph] 85.

(291) Id. As noted in Part II, the intercreditor good faith tradition also lives on within bankruptcy, which provides that any votes not cast in good faith in Chapter 11 may be "designated" and cast in the judge's discretion. 11 U.S.C. [section] 1126(e) (1988). Although this is not a blanket bar against self-interested voting, it covers much the same territory as did the intercreditor cases.

(292) Hackettstown was featured in leading corporate finance texts in the 1930s and was favorably cited in congressional debates on the Trust Indenture Act. See Coffee & Klein, supra note 34, at 1262 n.158 ("The case was featured in a leading treatise on bonds published in 1937... [and] also figured in prominent legal corporate finance texts of the 1930s"); Roe, supra note 104, at 252 n.54 ("The Hackettstown case figured in prominent legal corporate finance texts of the 1930s."). The Hackettstown case received extended treatment in a treatise written the year after the opinion came out. See EDWARD LYMANN SHORT, THE LAW OF RAILROAD BONDS AND MORTGAGES IN THE UNITED STATES OF AMERICA: WITH ILLUSTRATIVE CASES FROM ENGLISH AND COLONIAL COURTS 50-51 (1897); see also Sage v. Cent. R.R. Co., 99 U.S. 334, 341 (1879) (holding that an agreement was reasonable even though it prevented a "small minority of bondholders from forcing unreasonable and inequitable concessions from the majority"); In re Georgian Hotel Corp., 82 F.2d 917, 919 (7th Cir. 1936) (rejecting the proposition that section 77B is unconstitutional and void in so far as it empowers the bankruptcy court to require minority nonconsenting creditors of a particular class to accept any major modification of their original contractual relation); Vogelstein v. Ath. Mining Co., 192 S.W. 760, 763 (Mo. Ct. App. 1917) (enforcing a provision that gave the owners of two-thirds of the bonds at issue the power to modify the terms of the deed trust).

(293) 304 U.S. 64, 78 (1938).

(294) We do not venture a theory here for the dormancy of intercreditor good faith cases in England and Wales in the decades prior to Assenagon.

(295) 200 F.2d 627 (8th Cir. 1953).

(296) See, e.g., BRATTON, supra note 19, 548-52 (introducing Aladdin Hotel with a discussion of problems facing shareholders in workout situations); LEVITIN, supra note 21, at 157-61; MARK J. ROE & FREDERICK TUNG, BANKRUPTCY AND CORPORATE REORGANIZATION: LEGAL AND FINANCIAL MATERIALS 470-74 (4th ed. 2016). Curiously, despite its academic canonization, Aladdin Hotel received no notice in the legal press when it was first decided. The case has only been cited a few times since in other decisions, and the scholarly literature did not start citing to it until the 1980s.

(297) DEP'T OF INTERIOR, National Register of Historic Places Inventory: Nomination Form 2-3 (June 24, 1983), [].

(298) Record at 16-17, Aladdin Hotel Co. v. Bloom, No. 14,645 (8th Cir. pre-trial stipulations July 16,1952) (on file with authors).

(299) See Aladdin Hotel, 200 F.2d at 628 (noting that equityholders also held 72% of the bonds).

(300) See Record at 91, 95, 99,103,107,111,115, Aladdin Hotel Co. v. Bloom, No. 14,645 (8th Cir. pre-trial stipulations July 16, 1952) (on file with authors) ("Aladdin Hotel Company... promises to pay the registered holder hereof... the sum of FIVE HUNDRED DOLLARS ($500.00), on the 1st day of September, 1948... and to pay interest thereon from date to maturity at the rate of five per cent (5%) per annum, payable only out of net income of the Company... and to pay interest at the rate of eight per cent (8%) per annum from maturity until paid.").

(301) See Aladdin Hotel, 200 F.2d at 628 ("[T]he bonds and deed of trust contained provision empowering the bondholders of not less than two-thirds principal amount of the bonds, by agreement with the Hotel

Company to modify and extend the date of payment of said bonds provided such extension affected all bonds alike.").

(302) See id. (recounting that the Aladdin Hotel Company "executed and delivered a series of 647 bonds" on September 1,1938).

(303) See id. at 628-29 (noting that the defendants owned a majority of the Hotel Company's stock while also being the owners and holders of more than two-thirds of the bonds' principal amount).

(304) Id. at 630.

(305) One wonders if there might have been more going on in Aladdin Hotel. The plaintiff, Mrs. Bloom, was not some little old widow who had put her retirement savings into the Aladdin bonds. Instead, her two brothers were corporate raiders who owned, among other things, Trans World Airlines (at the time still headquartered in Kansas City). Mrs. Bloom apparently bought five of her seven bonds from her brothers. See Appellant's Brief at 3, 19, Aladdin Hotel Co. v. Josephine Loeb Bloom, No. 14,645 (8th Cir. filed Aug. 11, 1952) (on file with authors); Reply Brief for the Appellant at 9, Aladdin Hotel Co. v. Josephine Loeb Bloom, No. 14,645 (on file with authors). The other bonds she had purchased for ten cents on the dollar. Appellant's Brief at 3, Aladdin Hotel Co. v. Josephine Loeb Bloom, No. 14,645 (8th Cir. filed Aug. 11, 1952) (on file with authors). The point of the repeated mention of these facts in Appellant's brief is that Mrs. Bloom was serving as a front for her brothers and was not bringing the suit in good faith.

(306) The same approach can be seen in the case cited by Aladdin Hotel: Monticello Bldg. Corp. v. Monticello Inv. Co., 330 Mo. 1128 (1932). Monticello held that a bondholder for a defaulted mortgage was obligated "to show his own good faith and the equities of his own position" before getting a court-appointed receiver for the mortgaged property. Id. at 1142.

(307) The solution to the problem, in the present context, would be to bring forward the nineteenth-century cases but leave Assenagon behind. This would define a bad-faith action as a self-interested action not in the best interests of the bondholders as a group, but not as an action undertaken for self-preservation and not otherwise for private gain in response to an issuer's coercive offer.

Caption: Figure 1: Household Sector Bondholdings, 1945-2015

Please Note: Illustration(s) are not available due to copyright restrictions.
COPYRIGHT 2018 University of Pennsylvania, Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Symposium on the Fortieth Anniversary of the 1978 Bankruptcy Code
Author:Bratton, William W.; Levitin, Adam J.
Publication:University of Pennsylvania Law Review
Date:Jun 1, 2018

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |