Emory Bankruptcy Developments Journal

Bankruptcy’s Endowment Effect
Anthony J. Casey  Professor of Law and Mark Claster Mamolen Teaching Scholar, The University of Chicago Law School. I thank Elizabeth Kiernan, Alex Weber, and Stephanie Xiao for excellent research assistance and The Richard Weil Faculty Research Fund and The Paul H. Leffmann Fund for generous research support.

Abstract

In this Essay, I respond to Professor Markell’s analysis of the recent controversy over the cramdown interest rate applied in corporate bankruptcies. I argue that the main source of controversy is a misperception that pervades much of bankruptcy law and scholarship. Namely, courts and scholars commonly assign undue importance to preserving creditors’ nonbankruptcy endowments, which is inconsistent with foundational bankruptcy policy.

I make the case here that the guiding principle for optimal bankruptcy design should not be the preservation of nonbankruptcy rights but rather should be the minimization of opportunistic behavior that reduces the net value of a firm. Applying this principle to the question of the cramdown interest rate, I show that an optimal rule—properly focused on the minimization of opportunistic behavior—supports a cramdown interest rate based on the prevailing market rates for similar loans. Along the way I also show that this approach is consistent with the Bankruptcy Code and the theoretical principles (although not the ultimate conclusion) that Professor Markell has advocated.

Introduction

The endowment effect runs strong in corporate bankruptcy scholarship. Scholars commonly make the mistake of assuming that because a creditor is endowed with a right outside bankruptcy, that creditor must therefore be entitled to maintain the same right inside bankruptcy.  1See, e.g., Oliver Hart, Firms, Contracts, and Financial Structure 160 (Oxford 1995); Barry E. Adler & Ian Ayres, A Dilution Mechanism for Valuing Corporations in Bankruptcy, 111 Yale L.J. 83, 88 ̶ 90 (2001); Barry E. Adler & George Triantis, Debt Priority and Options in Bankruptcy (Mar. 2, 2016) (unpublished manuscript) (on file with authors); Walter J. Blum, Commentary, The “New Directions” for Priority Rights in Bankruptcy Reorganizations, 67 Harv. L. Rev. 1367, 1374 (1982) [hereinafter Blum, New Directions]; Thomas H. Jackson, Bankruptcy, Non-Bankruptcy Entitlements, and the Creditors’ Bargain, 91 Yale L.J. 857, 857–58 (1982). These scholars often assert that this result is required by the foundational theory of bankruptcy—it is not—and then defend policy proposals as scrupulously protecting creditors’ entitlements.  2See Adler & Ayres, supra note 1, at 88 ̶ 90; Adler & Triantis, supra note 1; Jackson, supra note 1, at 857 ̶ 58.

Despite its popularity in academic scholarship, this idea of sacred endowments is an untenable position that misunderstands the fundamental principles of bankruptcy. Corporate bankruptcy is, at its core, a system that alters nonbankruptcy endowments according to a hypothetical bargain—hypothetical because it is not the one the parties actually entered into—that we assume all creditors of a firm would have entered into if bargaining were costless.  3Jackson, supra note 1, at 860. The entire point of that hypothetical bargain is to suspend and alter some nonbankruptcy endowments to protect other endowments that maximize the value of the bankruptcy estate  4I use the term “estate” in a nontechnical sense throughout this Article to refer to the collective interests of all stakeholders, not just one class of creditors. and the firm as a whole.  5Anthony J. Casey & Aziz Z. Huq, The Article III Problem in Bankruptcy, 82 U. Chi. L. Rev. 1155, 1194 (2014) (“Bankruptcy’s proper goal is rather best understood as one of limiting certain private rights to protect others.”). Indeed, if every party retained every nonbankruptcy endowment, the Bankruptcy Code (the “Code”) would have no provisions at all.

Of course, altering nonbankruptcy endowments can impose costs. Foremost among those costs is the risk of opportunistic behavior.  6See Jackson, supra note 1, at 869. We do not want to create incentives for parties to pursue (or avoid) a bankruptcy filing for the sole purpose of transferring (or avoiding the transfer of) value between stakeholders. Such opportunistic maneuvering is costly for the estate as a whole. Thus, optimal bankruptcy policy will be designed to achieve its estate-maximizing purpose in part by minimizing opportunistic bankruptcy behavior that destroys firm value. Protecting nonbankruptcy endowments can, in many cases, be a means to that end. But protecting those endowments is not, as many scholars appear to believe, an end in and of itself.  7I have made this point elsewhere. See Anthony J. Casey & Edward R. Morrison, Beyond Options (2016) (unpublished manuscript) (on file with authors); see also Douglas G. Baird, Priority Matters, 165 U. Pa. L. Rev. (forthcoming 2016) (manuscript at 47 ̶ 48) (on file with the University of Pennsylvania Law Review) [hereinafter Baird, Priority Matters].

The unwarranted focus on nonbankruptcy endowments is a mistake that bankruptcy law scholars commonly make, including scholars arguing for absolute priority  8See Adler & Ayres, supra note 1, at 88 ̶ 90; Adler & Triantis, supra note 1; Jackson, supra note 1, at 857 ̶ 58. as well as scholars arguing against it.  9Including me. Anthony J. Casey, The Creditors’ Bargain and Option-Preservation Priority in Chapter 11, 78 U. Chi. L. Rev. 759, 778 (2011); see also Melissa Jacoby & Edward Janger, Ice Cube Bonds: Allocating the Price of Process in Chapter 11 Bankruptcy, 123 Yale L.J. 862, 916, 918, 922 (2013); cf. Omer Tene, Revisiting the Creditors’ Bargain: The Entitlement to the Going-Concern Surplus in Corporate Bankruptcy, 19 Bankr. Dev. J. 287, 396 (2003). Courts and lawyers are no different. The recent and ongoing dispute over cramdown in In re MPM Silicones, LLC (“Momentive”) provides a salient example.  10See In re MPM Silicones, LLC (Momentive), No. 14-22503-rdd, 2014 WL 4436335 (Bankr. S.D.N.Y. Sept. 9, 2014), aff’d, 531 B.R. 321 (S.D.N.Y. 2015), appeal docketed, No. 15–1771 (2d Cir. June 1, 2015). The dispute in Momentive was about what interest rate the senior creditors would get in a chapter 11 cramdown.  11There was a separate dispute about make-whole payments that also focused on creditor entitlements, but the outcome there turned in large part on whether the state-law entitlements existed in the first place. See Bruce A. Markell, “Shoot the. . .”: Holes in Make-Whole Premiums, 36 Bankr. L. Letter No. 6 (Thomson Reuters, Saint Paul, Minn.), May 2016, at 3–4 [hereinafter Markell, Make-Whole Premiums]. The bankruptcy court ultimately decided the case by importing a creditor-endowment framework from consumer bankruptcy law.  12Momentive, 2014 WL 4436335, at *23–32. That framework—the “Till formula” or “Till interest rate”—comes from the plurality opinion in Till v. SCS Credit Corp13541 U.S. 465, 478 ̶ 80 (2004). Eight of the nine Supreme Court Justices framed Till as an endowment case.  14Justice Thomas is the exception. See Till, 541 U.S. at 473–74, 489, 491–92 (Thomas, J., concurring) (focusing on statutory language rather than nonbankruptcy entitlements). Indeed, the plurality and dissenting opinions in Till focused almost exclusively on ex ante endowments. The same was true of the bankruptcy court’s opinion in Momentive. In each case, considerations about a properly designed bankruptcy system were conspicuously absent from the discussion.  15The purpose of chapter 13 may be different enough to justify this approach in Till, but not in chapter 11 cases like Momentive. I discuss this briefly below, infra note 95. But my primary inquiry here is into the bankruptcy court’s move to import Till’s analysis into chapter 11 corporate bankruptcies. Momentive, 2014 WL 4436335, at *23–32. In chapter 13 cases, we might worry about different things for policy reasons unrelated to corporate reorganization. Chapter 13 deals with individuals’ and society’s interest in providing opportunities for a fresh start. See H.R. Rep. No. 95-595, at 117 (1977). Chapter 13 also exists in a nonmarket environment—there is no going-concern sale, no DIP lender, and no market for exit financing. For individuals, there is simply the break-up of the assets (chapter 7) or the option to cram down a loan (chapter 13). Conditional upon being in chapter 13, the debtor is trying to keep things together without market support. We might not want to require an interest rate that makes it impossible to do that.

The result of all of this focus on endowments is that corporate bankruptcy scholarship and precedent are often bogged down in a back-and-forth about who is entitled to what, when the relevant question is what rule makes the most sense. The first part of Professor Markell’s article on Till and Momentive provides a refreshing alternative. Instead of focusing on senior creditors’ nonbankruptcy entitlements, Professor Markell marshals history and precedent to suggest three basic doctrinal principles that courts should follow in applying priority rules in chapter 11 cases: (1) “don’t pay too little”; (2) “don’t pay too much”; and (3) “don’t expect precision.” In doing so, he exposes the fallacies in commonly held notions (partially embraced by the Till dissent) that a creditor is somehow entitled to get exactly what it had before bankruptcy or to be made “subjectively indifferent between present foreclosure and future payment.”  16Bruce A. Markell, Fair Equivalents and Market Prices: Bankruptcy Cramdown Interest Rates, 33 Emory Bankr. Dev. J. 91, 114 (2016) [hereinafter Markell, Fair Equivalents] (quoting Till, 541 U.S. at 476 n.14 (2004)). Professor Markell shows that there are no statutory or historical grounds for such arguments.

The Markell principles (as I will call them) not only have the support of history and caselaw, but they also make for good bankruptcy policy. But when Professor Markell attempts to operationalize his principles, bankruptcy’s endowment effect sneaks back in. “Too much” and “too little” are defined for Professor Markell—just as they were for eight Justices on the Till Court—by reference to what the creditors are entitled to outside of bankruptcy. Here I must part ways with his analysis. Neither the history that Professor Markell presents  17Id. at 95–103. nor the statute that controls cramdown  1811 U.S.C. § 1129(b) (2012). requires this definition. Rather, Professor Markell convincingly shows us that the history leaves doubt about (and creates wiggle room for) how one should define entitlements. And the statute—as well as recent precedent interpreting it—suggests an altogether different inquiry, which has little to do with nonbankruptcy entitlements and much to do with implementing a coherent foundational theory of corporate reorganization. As such, I suggest a fourth and preeminent guiding principle that is supported by the statute and history: “do what makes sense.” And on that front, the court in Momentive got things exactly wrong.

I proceed in four parts. In Part I, I briefly review the applicable statutory provisions, Professor Markell’s historical insights on those statutes, and the import of recent Supreme Court cases interpreting them. I suggest, similarly to Professor Markell, that these foundations provide only general and uncertain guidance. But that guidance does foreclose certain forms of cramdown, including that which was used in Till (and imported to Momentive). I show, therefore, that, on the one hand, the application of the Till formula in Momentive is inconsistent with the Markell principles. On the other hand, the foundational guidance does not foreclose a cramdown system that is consistent with and advances general policies of a properly functioning corporate bankruptcy system. The inferences one can, and should, draw support embracing a method and framework of cramdown that best advance the foundational theory of corporate bankruptcy and reorganization.

In Part II, I review and examine that foundational theory and explain how it relates to the cramdown process. The real question that emerges from the analysis is not who is entitled to what, but rather, how should one design a bankruptcy system that produces a coherent set of incentives and outcomes. I suggest that while properly designed bankruptcy law should not, and does not, require any special attention to senior creditor entitlements or to making a senior creditor subjectively indifferent between foreclosure and cramdown, the law does require a system that minimizes opportunistic behavior that would destroy firm value. That is, the bankruptcy system should minimize the extent to which it creates value-destroying opportunities for stakeholders to exploit the bankruptcy process to capture value from other stakeholders.

In Part III, I derive the optimal cramdown rule. A coherent cramdown system will not only prevent creditors from destroying value by opportunistically opposing a plan, but it will also prevent debtors from destroying value by opportunistically proposing cramdown. If a creditor can insist on foreclosing on an asset that has going concern value, it will make threats to extract rents. Those threats can lead to bargaining failures that destroy estate value. Similarly, if debtors get consistently better rates of interest in cramdown, they will flood courts with chapter 11 cases that never should have been filed. Moreover, debtors who otherwise would have filed will be artificially drawn toward proposing inefficient cramdown plans. They will even, as the debtors in Momentive did, make threats of or propose inefficient plans for the sole purpose of extracting rents from creditors.  19According to the debtor’s plan in Momentive, Momentive would pay cash to the First Lien Notes and the 1.5 Lien Notes in an amount equal to their face value, including accrued interest, if they agreed to the plan and dropped certain objections. Disclosure Statement for Joint Chapter 11 Plan of Reorganization for Momentive Performance Materials Inc. & Its Affiliated Debtors at 35–36, In re MPM Silicones, LLC, No. 14-22503-rdd (Bankr. S.D.N.Y. Sept. 9, 2014), ECF No. 516, 2014 WL 4255110, at *35–36. If the note holders rejected this proposed treatment, however, Momentive would pay the note holders over time at an interest rate crafted according to Till. Id. See generally Markell, Fair Equivalents, supra note 16, at 125 (discussing the ramifications of Momentive’s proposed plan). The result was that the noteholders would receive lower value if they objected to the plan. Id.

Neither of these outcomes serves any valid purpose for corporate reorganization. Both can be avoided by a system that allows cramdown but tests its rate against the market alternatives that face a debtor at that time in the real world. I show that a focus on real-world market rates—and not intrinsic values—is necessary to properly align the incentives of those making the decision to pursue the cramdown option rather than the other options available in the market. In Part IV, I present examples and apply this framework to the specific facts of Momentive, showing how the focus on real-world market rates better aligns incentives and is more consistent with bankruptcy’s fundamental purpose.

I. History, Precedent, and Statutory Authority

Many of the debates about entitlements are beyond the scope of this Article. There is no question that the absolute priority rule, in some form, is the current law of the land.  20The absolute priority rule prevents equity holders from retaining or realizing any value in a reorganized debtor company unless creditors consent or are paid in full. See 11 U.S.C. § 1129(b); Bruce A. Markell, Owners, Auctions, and Absolute Priority in Bankruptcy, 44 Stan. L. Rev. 69, 72 & n.17 (1991). It is codified in chapter 11 through the words “fair and equitable.”  2111 U.S.C. § 1129(b)(1). That much is uncontroversial.  22 See generally Baird, Priority Matters, supra note 7, manuscript at 4. Accordingly, Professor Markell opens his discussion about cramdown with an informative and persuasive history of the statutory requirement that a plan be “fair and equitable.”  23Markell, Fair Equivalents, supra note 16, at 95–103 (outlining the caselaw and legislative history behind the “fair and equitable” standard). From this history, he abstracts three general principles that form the core of absolute priority (and therefore govern cramdown): “don’t pay too little”; “don’t pay too much”; and “don’t expect precision.”  24Id. at 103–04.

These principles combine, Professor Markell tells us, into a standard of “fair equivalence.”  25Id. at 104. So for cramdown, a creditor is entitled to a fair equivalent of its prepetition interest. That standard, of course, requires valuation of both the interest rate and the thing that purports to be its fair equivalent. The former is usually easy; the latter is often difficult. Recognizing this difficulty, Professor Markell invokes the “don’t expect precision” principle to justify application of the Till formula. This is a mistake.

The Till formula applied to chapter 11 ironically gets everything backward. It is a formula that provides a precise but inequivalent payout, thus violating the Markell principles, while also violating the clear directives of the statute and distorting the bankruptcy process. Thus, while the Markell principles are sound in theory, the application here is not. To show this, I first examine the relevant statute and then turn to the Markell principles.

A. The Statutory Provisions: § 1129(b)(1) and § 1129(b)(2)

The statutory language is as good a place to start as any. The relevant provisions for cramming down a secured creditor are §§ 1129(b)(1) and 1129(b)(2)(A) of the Code. The first of these provisions requires that any cramdown plan must be “fair and equitable” with respect to the objecting class.  2611 U.S.C. § 1129(b)(1). The second provision adds that the words “fair and equitable”—whatever else they may mean—at least include a specific set of requirements, which the provision lists.  27The specific language states: “[T]he condition that a plan be fair and equitable with respect to a class includes the following requirements . . . .” Id. § 1129(b)(2). For cramming down a secured creditor, the plan is required to provide the creditor with one of these three things: (1) retention of its liens along with deferred cash payments “of a value, as of the effective date of the plan, of at least the value” of the creditor’s interest in the property subject to the lien; (2) the right to credit bid in a sale of the property subject to the lien; or (3) the “indubitable equivalent” of the creditor’s claims. See id. § 1129(b)(2)(A).

Professor Markell notes that the list of requirements in § 1129(b)(2) is not exhaustive.  28Markell, Fair Equivalents, supra note 16, at 130. Other things might also be required for a plan to be fair and equitable.  29Id. The non-exhaustive nature of § 1129(b)(2) is a crucial point that courts and scholars often ignore. It is not enough that a plan meet the requirements of § 1129(b)(2). It must also meet all other requirements that are implied by § 1129(b)(1)’s invocation of the words “fair and equitable.”  30See In re Castleton Plaza, LP, 707 F.3d 821, 823 (7th Cir. 2013) (noting that compliance with § 1129(b)(2) was not all that was required to satisfy absolute priority). Because that phrase has a long history, the list of implied requirements can be identified by looking at the application of § 1129(b) over the last century. Thus, for example, a plan cannot be fair and equitable if it overpays a class of creditors.  31Markell, Fair Equivalents, supra note 16, at 131.

But this analysis does not mean we can ignore § 1129(b)(2). In a sense, though, that is what Momentive does. Professor Markell seems to justify this outcome in Momentive by calling the requirements “examples.”  32Id. at 130. But such justification cannot withstand scrutiny. The statute plainly calls them “requirements” for a plan to qualify as “fair and equitable.”  3311 U.S.C. § 1129(b)(2) (“[T]he condition that a plan be fair and equitable with respect to a class includes the following requirements . . . .”).

Moreover, the Supreme Court has made it clear that it does, in fact, view the provisions of § 1129(b)(2) as requirements—most recently in RadLAX Gateway Hotel, LLC v. Amalgamated Bank34132 S. Ct. 2065, 2069–70 (2012). In that case, a unanimous Court stated, “A Chapter 11 plan confirmed over the objection of a ‘class of secured claims’ must meet one of the three requirements in order to be deemed ‘fair and equitable’ with respect to the nonconsenting creditor’s claim. The plan must provide [one of (i), (ii), or (iii)].”  35Id. This interpretation is consistent with the statutory language. It is also consistent with the history of the Code and the concerns of the drafters. River East Plaza, LLC v. Geneva Leasing Ass’n (In re River East Plaza, LLC), 669 F.3d 826, 828–29 (7th Cir. 2012); see also Douglas G. Baird, Remembering Pine Gate, 38 J. Marshall L. Rev. 5, 9 (2004) [hereinafter Baird, Pine Gate].

Prior to RadLax, the Court’s opinion in Bank of America National Trust & Savings Association v. 203 North LaSalle Street Partnership expressed the same premise,  36526 U.S. 434, 437 ̶ 38 (1999). noting that a plan can be “fair and equitable” only if it complies with § 1129(b)(2).  37Id. at 441. The Court then went even further, suggesting that compliance with the requirements of § 1129(b)(2) was not only mandatory, but the validity of such compliance also must be market tested38Id. at 458. A mere judicial sense that the statute was being followed was not enough; a market test showing it to be true was required.  39Id. at 436.

These interpretations are consistent with the Court’s general tendency—for better or worse—to restrict the discretion of bankruptcy judges in altering the rights of creditors.  40See Douglas G. Baird & Anthony J. Casey, Bankruptcy Step Zero, 2012 Sup. Ct. Rev. 203, 220. To be sure, bankruptcy judges can alter those rights, but only according to very specific rules. This lesson has been repeated often in the Court’s bankruptcy jurisprudence.  41I have explored these lessons more fully elsewhere:Together these cases start to form a pattern. On certain questions, the market and Congress are the only competent arbiters. To the extent Congress is unclear in addressing those questions, the Court will view it as providing as narrow a rule as is reasonably consistent with the Code’s language.Id. at 221; see also Casey & Huq, supra note 5, at 1215–16.

The best way to understand § 1129(b)(2), then, is as a floor for what will qualify as fair and equitable. That floor cannot be ignored.

None of the analysis so far contradicts the Markell principles. Indeed, to say that § 1129(b)(2) sets a floor for payment merely restates the first principle: “don’t pay too little.” The long-recognized implied ceiling in § 1129(b)(1) provides the second principle: “don’t pay too much.”  42Markell, Fair Equivalents, supra note 16, at 104. And Professor Markell’s historical analysis of “fair and equitable” suggests the third principle: “don’t expect precision.” But if we want to use the principles in an actual case, we are still left to decipher what exactly is “too little” or “too much.”

B. History and the Markell Principles

The statute cannot get us much further. For cramming down a secured creditor, the relevant language merely reframes the too-little question as one of whether “a deferred cash payment [is] . . . of a value, as of the . . . date of the plan of at least the value of” the creditor’s interest.  4311 U.S.C. § 1129(b)(2)(A)(i)(II) (2012) (“[T]hat each holder of a claim of such class receive on account of such claim deferred cash payments totaling at least the allowed amount of such claim, of a value, as of the effective date of the plan, of at least the value of such holder’s interest in the estate’s interest in such property.”). But how does one answer that question?

We might turn to caselaw, but Supreme Court precedent does not provide controlling authority on how to answer the question. This is true, despite Till, for two reasons. First, Till did not garner a majority opinion. It was a four-member plurality not joined by Justice Thomas.  44541 U.S. 465 (2004) (plurality opinion). This point was somewhat irrelevant for the bankruptcy court in Momentive. See Momentive, No. 14-22503-rdd, 2014 WL 4436335, at *24 (Bankr. S.D.N.Y. Sept. 9, 2014), aff’d, 531 B.R. 321 (S.D.N.Y. 2015), appeal docketed, No. 15-1771 (2d Cir. June 1, 2015). It was obviously bound by decisions of the Second Circuit, and the Second Circuit had adopted a similar approach to the Till plurality in an earlier case. Id. (citing GMAC v. Valenti (In re Valenti), 105 F.3d 55, 63 (2d Cir. 1997)). Thus, statutory text and first principles aside, the bankruptcy court was bound to at least apply the Till rule to chapter 13 cases. Id. Second, Till was construing § 1325 and not § 1129, and the footnotes in the plurality opinion provide conflicting guidance as to whether the reasoning of the case is applicable to chapter 11 cases.  45See 541 U.S. at 476 n.14; Markell, Fair Equivalents, supra note 16, at 110–13. It is also worth pausing to note that Justice Thomas’s concurrence was based on a close reading of the statutory text of § 1325(a)(5)(B)(ii) that is not easily imported to § 1129(b)(2)(A)(i)(II).  46Section 1325 deals with “the value, as of the effective date of the plan, of property to be distributed under the plan.” 11 U.S.C. § 1325(a)(4). Justice Thomas notes that the value of the cash to be distributed was its face value, regardless of the risk of nondistribution. Till, 541 U.S. at 485–86 (Thomas, J., concurring). He went on to note that if the “property distributed” was “a note (i.e., a promise to pay),” then the value of that note “necessarily includes” a risk premium. Id. at 488–89. Section 1129(b)(2)(A)(i)(II) talks of “deferred cash payments . . . of a value, as of the effective date of the plan, of at least the value of such holder’s interest in the estate’s interest in such property.” The question for Justice Thomas, then, is whether “of a value” modifies “payments” or “deferred cash payments.” The former would make the provision analogous to the case in Till; the latter would look more like Justice Thomas’s hypothetical “notes.” Put another way, “deferred cash payments” are promises to pay, and valuing them today (like valuing notes) implies a risk adjustment in a way that valuing payments or property (that happen to be coming in the future) does not. For my eye, the natural and plain reading is that “value” modifies the entire clause, “deferred cash payments.” In short, even if the plurality reasoning in Till applied to § 1129(b)(2)(A)(i)(II), it is likely that Justice Thomas’s reasoning does not. So for chapter 11, that would give a fifth vote to the dissent. Does that matter? Probably not. But it does make the case that Till somehow provides binding authority even more dubious.

Next, we can turn to history and apply the Markell principles.  47The historical support for the reasoning is set forth in Professor Markell’s piece. Markell, Fair Equivalents, supra note 16, at 95–103. I do not restate it here. Through those principles, we can accept errors as long as we try to get things right. The problem with the Till formula is that it does not even try to get things right.

To see why, we must first note that concepts like “too little” and “too much” must be anchored to some baseline concept. The baseline for the Justices in Till (and for Professor Markell and the judge in Momentive) was nonbankruptcy entitlements. For them, the debate about too much and too little was about comparing what the creditors and other stakeholders would have received outside of bankruptcy. Interestingly, in Till, each side succeeded in showing that the other side was getting the entitlements wrong. Everyone was achieving a form of false precision at the expense of paying too little or too much. Thus, everyone was violating at least two of the Markell principles.

The dissent in Till, in search of precision, suggested that the creditors should receive a stream of payments subject to the prebankruptcy interest rate.  48Till, 541 U.S. at 491 (Scalia, J., dissenting). But that rate will generally overpay creditors. True, if the creditors had foreclosed and liquidated the assets outside of bankruptcy, they would have been able to reinvest the cash proceeds in a new loan. But the sale and reinvestment would have imposed significant transaction costs in the form of finding a buyer for the assets and finding a new borrower. And the market rate for such a loan may have changed since the time of the original transaction. Moreover, a firm coming out of a successful bankruptcy (if the system works at all) should be less risky than a firm that has just defaulted and is on the brink of filing bankruptcy. That reduced risk should justify a lower interest rate.

That being said, the plurality’s approach in Till performs no better. Following the plurality view does not produce imprecise but fairly equivalent payouts. Instead, the Till plurality directs a court to use a formula that produces a fairly precise rate (prime plus one to three percent is a relatively narrow range) with a payout that is, in most cases, certainly not equivalent to creditors’ prepetition entitlements.  49Id. at 491–92 (Thomas, J., concurring) (noting that the plurality started with “a rate we know is too low” and that the formula “will systematically undercompensate”). Indeed, because the formula starts with the prime rate and allows for limited deviations, it will underestimate the value of creditors’ entitlements by significant margins.  50Id. at 504 (Scalia, J., dissenting) (noting that the plurality’s rate in Till was off by an order of magnitude).

As a result, neither of these precise methods  51Markell, Fair Equivalents, supra note 16, at 103–04. Precision implies both exactness and accuracy. The Till rate is precise in the sense that it achieves relative certainty and accurately underestimates the interest rate. The three Markell principles, taken together, suggest that “don’t expect precision” means you do not need exactness, and unbiased inaccuracy is okay. We should not sacrifice fairness by introducing bias in the name of exactitude or mathematical certainty. Inaccurate equivalency is better than accurate inequivalency. is consistent with the Markell principles. They do not produce a fair equivalent of creditors’ prepetition interests in the estate. As Professor Markell shows, a fair equivalent “cannot be made by the use of any mathematical formula.”  52Id. at 105 (quoting Grp. of Inst. Inv’rs, v. Chi., Milwaukee, Saint Paul & Pac. R.R., 318 U.S. 523, 565–66 (1943)). Rather, the Markell principles call for an “estimate” that does not have “mathematical certitude.”  53Id. at 98 (quoting Consol. Rock Prods. Co. v. Du Bois, 312 U.S. 510, 526 (1941)). Fair equivalence does not imply “an illusion of certainty,” but instead requires “equitable equivalence.”  54Id. at 98–99 (quoting Grp. of Inst. Inv’rs, at 565). The takeaway from all of this analysis is that the law requires the creditor to be given something that is a fair estimate of its prepetition interest. A fair estimate, by definition, can deviate from actual value, but it cannot deviate in a biased (i.e., unfair) way. Thus, an estimate that precisely and consistently understates creditors’ prepetition interests would not be a fair estimate.

That is, however, exactly what the Till interest rate does. And so Momentive, in adopting the Till formula for chapter 11, violates the Markell principles. It pays too little, and not because it is imprecise. Rather, the underpayment is a direct result of the attempt for precision. The rate paid is precise and precisely too low.

II. Of Entitlements and Opportunistic Behavior

The story so far is that the law requires fair equivalence and not precision, but the Till rate gives us the opposite. The disconnect stems directly from the Till Court’s misguided focus on entitlements. The law’s tolerance for less-than-precise valuation (as the Markell principles demonstrate) provides a signal that the bankruptcy purpose behind the cramdown rules is not to ensure that entitlements are protected and treated like property rights—that would require precision—but rather to create a system that facilitates appropriate incentives. In this Part, I argue that such a system is not only appropriate from a policy perspective, but also consistent with the statute and the Markell principles.

To do so, I first review the foundational theory of bankruptcy to show that a well-designed bankruptcy system need not focus on the full preservation of nonbankruptcy entitlements. Then, I explain the key features of a system focused on minimizing opportunistic behavior and discuss how that approach is consistent with the statute and the principles behind it.

A. The Creditors’ Bargain Theory and the Lure of Creditor Entitlements

The academic focus on preserving nonbankruptcy entitlements goes at least as far back as Thomas Jackson’s seminal 1982 article on the matter.  55See Jackson, supra note 1, at 857. It perhaps goes further back. See Walter J. Blum, Full Priority and Full Compensation in Corporate Reorganizations: A Reappraisal, 25 U. Chi. L. Rev. 417, 437–39 (1958); Blum, New Directions, supra note 1, at 1368–69. These entitlements were central to Jackson’s discussion and model of his creditors’ bargain theory.  56See Jackson, supra note 1, at 858. But today, it has become clear that the theory does not itself require the preservation of these entitlements.

The core of the creditors’ bargain theory justifies the existence of corporate bankruptcy based on the assumption that creditors cannot collectively reach an ex ante agreement over how to deal with financial distress.  57See id. at 862–63. Without that agreement, they cannot bind themselves to refrain from behavior that might destroy (or fail to maximize) the value of an estate. The result is a collective-action problem that cannot be solved by private contracting and, therefore, justifies a mandatory bankruptcy regime.  58Id. at 862.

That regime, the theory suggests, should be the one that mimics the bargain that creditors would have entered into if they were capable of doing so.  59See id. at 867–68. As a result, we should expect the bargain to be the one that maximizes the value of the estate in bankruptcy. Of course, that alone is not enough. There are many ways to maximize the value of a bankruptcy estate that might have side effects that hinder other goals. Strictly speaking, we should expect the parties in a hypothetical costless bargain to enter an agreement that maximizes the average expected value of the estate across all states of the world60By “all states of the world,” I mean to encompass the various contingencies that might play out both in and out of bankruptcy. The parties, ex ante, will desire to maximize the overall expected value of the firm based on their probabilistic expectations about which state of the world will ultimately result. If a rule produces a benefit in one state of the world, the value of the benefit must be discounted for any costs imposed in another state of the world (weighted by the probabilities that each state will actually come to exist). Because parties are generally allowed (and assumed) to contract for efficient outcomes, this is usually translated into the principle that bankruptcy law should tinker with nonbankruptcy rights as little as possible.

An important nuance, however, lies buried in this translation. We can confidently say that the contracts that parties enter into outside of bankruptcy indicate their contracting preferences in those states of the world. These contracts do not, however, tell us anything about what the parties would contract for with regard to their rights and entitlements in bankruptcy. In the world we live in—with its mandatory bankruptcy rules—we have no idea what bankruptcy terms the parties would freely choose in the absence of those rules. Moreover, in a world where bankruptcy law is necessary because we have assumed that no efficient contract can be drafted to govern in the bankruptcy state of the world, it would be strange to determine the bankruptcy rules by looking to the inefficient-by-assumption contracts that a subset of the relevant parties did in fact draft.

What is left, then, of the respect for nonbankruptcy entitlements? The answer lies in the dynamic effect that rules can have across different states of the world. We should be wary of rules in bankruptcy that either destroy value for firms not in bankruptcy or that create incentives for parties to opportunistically shift from one state to another.

For example, a rule that converts all loans to an interest-free regime in bankruptcy would have two deleterious effects. First, it would cause healthy debtors to feign distress and file for bankruptcy when there is no value to be gained from the filing. Second, it would reduce the willingness of creditors to make loans to even healthy firms with low bankruptcy risk. Contrast that outcome with a bankruptcy rule that guarantees a creditor its foreclosure value but awards all additional value created by the bankruptcy process to the debtor.  61Casey, supra note 9, at 796; see also Comm’n to Study the Reform of Chapter 11, Am. Bankr. Inst., Final Report and Recommendations 75 (2014). Such a rule only encourages the debtor to file for bankruptcy when there is value to be created by the bankruptcy, and it has no impact on the rights the creditor enjoys in other states of the world.

The takeaway is that bankruptcy law should minimize its interference with entitlements outside bankruptcy. That principle is much different from the idea that bankruptcy law must preserve entitlements inside bankruptcy. This distinction is often lost. The original exposition of the creditors’ bargain theory itself makes the explicit assumption that the absolute priority of secured creditors’ rights brings aggregate efficiencies to the estate and should therefore be respected. But nothing about the underlying theory justifies that assumption. Neither does any real world evidence.  62Baird, Priority Matters, supra note 7, manuscript at 32 (“It is also worth noting that we do not see anything as simple as the absolute priority rule in an analogous environment in which free contracting is permitted.”). Indeed, all one can really say about the market for (and aggregate efficiencies of) secured creditors’ rights is that parties view them as valuable in the states of the world that do not trigger bankruptcy law.  63I have stressed this point in prior articles. Casey, supra note 9, at 772–73; Casey & Morrison, supra note 7; see also Baird, Priority Matters, supra note 7, manuscript at 22 (“The debate about relative and absolute priority must therefore take place in an empirical vacuum.”).

B. The Creditors’ Bargain and Minimizing Opportunistic Behavior

Despite the lack of theoretical or empirical grounding, the assumption that the creditors’ bargain theory requires the protection of nonbankruptcy entitlements has spawned a vast literature championing the protection of creditors’ nonbankruptcy entitlements in bankruptcy for its own sake. This detour is unfortunate and unnecessary, and it leads to unhinged debates about entitlements—debates that lack a strong connection to considerations of the proper design of bankruptcy policy.

Much clarity can be gained going forward by dropping the entitlement framework altogether and focusing instead on the incentives and distortions to incentives that bankruptcy law creates when it assigns rights to stakeholders. Thus, the goal should not be to preserve or protect entitlements, but instead to minimize opportunistic behavior.  64The idea that bankruptcy law can and should change nonbankruptcy endowments is consistent with the Supreme Court’s holding in United States v. Butner. 440 U.S. 48, 54 (1979). Butner is sometimes cited for the principle that bankruptcy law should leave state law undisturbed. That application is not quite right. The case merely says that in the absence of a bankruptcy statute, courts should look to state law. Id. at 57–58. But here, there is a statute (11 U.S.C. § 1129) and so we look to it and not to state law. Indeed, in Butner the Court explicitly said that the constitutional bankruptcy power “would clearly encompass a federal statute” defining a creditor’s interest in estate property. Id. at 54. The Court also noted that a nonbankruptcy state interest could be altered by bankruptcy law if a federal (that is, bankruptcy) interest required that result. Id. at 55.

All of this is to say that we want bankruptcy rules to be efficient in operation and in effect. Deviations from entitlements are only relevant to the extent that they distort behavior and create inefficiencies within the bankruptcy system. Lots of parties are over- or undercompensated in bankruptcy (depending on your baseline). There is no question about that. The question is whether they are over- or undercompensated in a way that distorts the process or distorts behavior in other contexts outside of bankruptcy.

This framing is consistent with the structure of § 1129(b), with its creditor bidding and indubitable equivalent requirements, and with the cases interpreting that section.  65See, e.g., RadLAX Gateway Hotel, LLC v. Amalgamated Bank, 132 S. Ct. 2065, 2069 ̶ 70 (2012); Bank of Am. Nat’l Tr. & Sav. Ass’n v. 203 N. LaSalle St. P’ship, 526 U.S. 434, 457–58 (1999). See generally Baird & Casey, supra note 39, at 208–10. For example, § 1129(b)(2)(A)(ii)—as interpreted by the Supreme Court in RadLAX—requires that a debtor wanting to sell an asset that is subject to a lien must provide the creditor the opportunity to credit bid in the sale.  66132 S. Ct. at 2069 ̶ 70. This framing of § 1129(b) limits the debtor’s ability to opportunistically sell the good into a depressed market. The result is that if a debtor attempts to sell into a depressed market, the creditor will take the asset by credit bid. The creditor, on the other hand, has limited opportunities to extract value because it cannot demand the right to buy the asset under any circumstances.

Note that neither the Code nor the Court requires any proof about efficiencies in the market or about intrinsic values. Instead, they force the debtor to make (and the creditors to live with) a choice that is cabined by the realities of the market for the assets. The debtor can give the creditor a cash equivalent, a cramdown payment plan (which I have suggested must be equal to market loans), or the ability to buy the asset in a market sale. The effect is that the debtor is choosing between cash and the things that cash can buy on the market. Faced with that choice, the debtor has no incentive to act opportunistically.  67These rules are also consistent with historical cases of opportunistic behavior that were on the minds of the drafters of the Code. See In re Pine Gate Assocs., Ltd., No. B75-4345A, 1976 U.S. Dist. LEXIS 17366 (N.D. Ga. Oct. 14, 1976); In re Pine Gate Assocs., Ltd., 12 Collier Bankr. Cas. (MB) 607 (Bankr. N.D. Ga. Mar. 4, 1977); Baird, Pine Gate, supra note 34, at 5.

Similarly, § 1129(b)(2)(A)(iii) provides that when a debtor wants to give the creditor something other than a cramdown loan or cash, it must meet the highest standard of “indubitable equivalence.”  6811 U.S.C. § 1129(b)(2)(A)(iii) (2012). This limit is set at a very high level because the debtor’s rejection of the market-tested options is highly suspect. As Judge Posner noted in a recent case, “[t]he debtor’s only motive for substitution of collateral in such a case is that the substitute collateral is likely to be worth less than the existing collateral.”  69In re River East Plaza, LLC, 669 F.3d 826, 831–32 (7th Cir. 2012). And so the Code requires a high burden of equivalence to limit the debtor’s ability to opportunistically choose alternative compensation.  70See id. at 832 (“[B]ecause of the different risk profiles of the two forms of collateral, they are not equivalents, and there is no reason why the choice between them should be made for the creditor by the debtor.”).

Finally, the Court’s opinion in 203 North LaSalle, with its implied new-value exception, also prevents debtors from opportunistically buying into the equity of a reorganized firm without a market test.  71See 526 U.S. 434, 457–58 (1999). That market test goes a long way toward eliminating a debtor’s ability to use the inside investment (or threat of an inside investment) to opportunistically extract value from a creditor. Again, there is no suggestion that the test is secondary to nonmarket evidence about intrinsic entitlements.

The opportunistic-behavior framework is also consistent with other core provisions of the Code. In many instances, the Code alters nonbankruptcy rights with the goal of minimizing opportunistic behavior. In others, the Code preserves nonbankruptcy rights with that same goal. For example, the automatic stay broadly prohibits most creditors from exercising their nonbankruptcy enforcement rights.  72See 11 U.S.C. § 362(a). This alteration of rights is intended to prohibit creditors from using those rights to gain advantage over other creditors in a way that destroys estate value.  733 Collier on Bankruptcy ¶ 362.03 (Alan N. Resnick & Henry J. Sommer eds. 16th ed.) (“[T]he stay provides creditors with protection by preventing the dismemberment of a debtor’s assets by individual creditors levying on the property. This promotes the bankruptcy goal of equality of distribution.”). The rule also prevents creditors from using threats of opportunistic enforcement to extract rents.  74Id. (“[The stay] protects property that may be necessary for the debtor’s fresh start and . . . provides breathing space to permit the debtor to focus on rehabilitation or reorganization.”). It is an alteration of nonbankruptcy rights for the purpose of maximizing estate value.  75Id. ¶ 362.03[2] (“The stay prevents this piecemeal liquidation, offering the chance to maximize the value of the business.”).

The automatic stay, without more, however, would create an opportunity for the debtor to use stalling tactics to extract value from creditors. The ability to stop enforcement when time is of the essence can be a powerful threat.  76See generally Chrysler LLC v. Plastech Engineered Prods., Inc. (In re Plastech Engineered Prods., Inc.), 382 B.R. 90, 107–08 (Bankr. E.D. Mich. 2008) (struggling with competing claims about whether the invocation of, or the motion to lift, the automatic stay was opportunistic). A more general example would be when a debtor uses the automatic stay to hold seasonal or perishable goods hostage. To partially alleviate that concern, bankruptcy law provides for things like adequate protection and other safety valves that allow the judge to lift the automatic stay for cause.  77See 11 U.S.C § 362(d).

To be sure, none of these provisions is precise. There is a lot of fudging, and the Code and the courts will get things wrong in many close cases. But the goal is clear: to provide a system that maximizes the estate value while minimizing the instances of opportunistic behavior. Sometimes this will require the protection of nonbankruptcy entitlements; sometimes it will require the destruction of those entitlements.

In the abstract, this theory tells us two things: we should (1) view as more suspect rules that transfer value in bankruptcy that otherwise would not have been transferred; and (2) view as less suspect rules that distribute value that would not otherwise have existed. Thus, the creditors’ bargain theory has little to tell us about distributions of the going concern value that bankruptcy has created or saved (the bankruptcy surplus), but it has much to tell us about making sure parties receive at least the value they would have received in nonbankruptcy proceedings. The former has only a small impact on nonbankruptcy values and the decision to enter bankruptcy,  78It may be wise to give at least some of the bankruptcy surplus to the person or entity that makes the filing decision to ensure that value-creating filings do occur. But beyond that, the distributional implications are unclear. See Alan Schwartz, A Normative Theory of Business Bankruptcy, 91 Va. L. Rev. 1199, 1239 (2005). Compare Barry E. Adler, Illinois ABI Symposium on Chapter 11 Reform: Priority in Going-Concern Surplus, 2015 U. Ill. L. Rev. Online 811, 814 (2015), and Adler & Triantis, supra note 1, at 19–20, with Douglas G. Baird & Donald S. Bernstein, Absolute Priority, Valuation Uncertainty, and the Reorganization Bargain, 115 Yale L.J. 1930, 1937 (2006), and Casey, supra note 9, at 778, with Jacoby & Janger, supra note 9, at 916, 918, 922, and Tene, supra note 9, at 395–96, with Baird, Priority Matters, supra note 7. while the latter has an enormous impact on those things.  79Allowing debtors (or junior creditors) to capture foreclosure value from senior creditors by entering bankruptcy would have at least three major effects: (1) debtors will spend resources to enter bankruptcy even where it is not value maximizing; (2) creditors will spend resources to prevent bankruptcy even when the bankruptcy would be value maximizing; and (3) the cost and availability of capital for debtors outside of bankruptcy will be adversely impacted because the risk for creditors will be higher.

In the next part, I will show that for cramdown, this approach demands the use of the actual market rates (net of transaction cost) for similar exit loans as they exist at the time of confirmation.

III. Cramdown

Moving from the abstract to the concrete, we can now derive the optimal rule for cramdown using the Markell principles viewed in connection with our goal of minimizing opportunistic behavior.

A. “Don’t Pay Too Much”

An obvious question from the above is whether cramdown, or some form of it, is reconcilable with the fundamental theory that bankruptcy should maximize the value of the estate while minimizing instances of opportunistic behavior. It is.  80See George G. Triantis, Mitigating the Collective Action Problem of Debt Enforcement Through Bankruptcy Law: Bill C-22 and Its Shadow, 20 Can. Bus. L. Rev. 242, 252 (1992) (“The cram down is one way to address the problem of holdouts by classes of secured lenders.”). Without cramdown, in fact, the collective-action problem would be insurmountable. A secured creditor might, in the absence of a Coasean bargain  81The Coase Theorem posits that where there are completely competitive markets with no transaction costs, parties in a transaction will always reach the most efficient outcome, regardless of how property rights are allocated. See generally R.H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1 (1960). with the debtor, seek to foreclose on an asset even when doing so destroys estate value. There are three primary explanations for why this bargaining failure might occur: (1) irrational or idiosyncratic preferences;  82The creditor may also have artificial reasons for foreclosing on assets even when doing so destroys value. Such reasons often stem from external forces such as regulatory pressure from tax and accounting rules. See, e.g., In re RTJJ, Inc., No. 11-32050, 2013 WL 462003, at *15 (Bankr. W.D.N.C. Feb. 6, 2013) (“First National recognizes this. Community One does not seem to care. Under pressure from federal regulators, Community One seeks to rid itself of this nonperforming loan, at any cost. Its aims are noneconomic—at least as to this Debtor—and are destructive.”). (2) asymmetric information (the creditor does not know the true value of the estate as reorganized but the debtor does); or (3) opportunistic posturing (the creditor may be threatening foreclosure in hopes of extracting more value).

The bargaining problem is important regardless of which reason is driving it. And the dynamics of the negotiations are likely to be such that a debtor cannot distinguish between them anyway. The creditor who is posturing might want to appear irrational or idiosyncratic to make the threat of foreclosure credible. In any event, if the debtor and creditor cannot reach a bargain, the debtor will have two remaining options: (1) go to the market to refinance the asset (and give cash to the secured creditor); or (2) allow the foreclosure.

There is a risk that significant value will be destroyed if transaction costs for the first option are high, which is likely to be the case. Providing the market with clear information may be prohibitively costly or impossible, and the bargaining with outside lenders might fail for the same reasons as the bargaining with the secured creditor. The outsiders will have even less information about the loan than the insider. And if, as is often true in these cases, it is only feasible to provide full information to one potential lender, that lender may try to extract rents by posturing, or it may have its own irrational or idiosyncratic preferences. The final result is that if these transaction costs are high enough, resorting to the market might not be worth it at all.

To correct for these problems, the Code allows the debtor to choose whether the creditor will (1) get the right to foreclose on the property;  83The debtor could also sell the asset subject to credit bidding. But, assuming the creditor is undersecured, that alternative amounts to the same thing as foreclosure. (2) get paid in cash; or (3) get a note or some other property that is of an equivalent value.  84See 11 U.S.C. § 1129(b)(2) (2012); In re River East Plaza, LLC, 669 F.3d 826, 828–33 (7th Cir. 2012) (summarizing the options). Giving this choice to the debtor prevents bargaining failure. Chiefly, the creditor can no longer opportunistically demand a higher interest rate simply because the debtor would incur transaction costs if it tried to go to the market to get a new loan. The creditor cannot idiosyncratically raise its interest rate above market and then demand that the debtor pay the higher rate or risk foreclosure. And the creditor cannot demand the right to take its asset away simply because it has bad information and overestimates the risk of leaving the asset with the debtor.

Instead, the debtor has the choice to keep the asset within the estate if the asset is more valuable there. Moreover, the debtor can force the creditor to continue to finance the asset at a market rate, thereby avoiding any transaction costs that it might otherwise incur on the market. This option of the debtor prevents the creditor from opportunistically profiting from the mere existence of these transaction costs.  85It also prevents the creditor from incidentally profiting from those costs. But that is not the rule’s primary purpose.

Going back to the creditors’ bargain theory, we would expect a creditor, ex ante, to enter into a pact to accept cramdown when that cramdown is expected to give the creditor something of the same market value as the cash value of its foreclosure rights. If the debtor with going-concern value has to choose between (a) giving the creditor $100 in cash (or allowing a foreclosure, which amounts to almost the same thing) and destroying the firm, or (b) giving the creditor $100 in a promise and saving the firm, we want the debtor to choose (b) every time. And this choice is likely to arise quite often. If the debtor has to pay off the creditor, there are (as Professor Markell and the judge in Momentive point out)  86See Momentive, No. 14-22503-rdd, 2014 WL 4436335, at *25–29 (Bankr. S.D.N.Y. Sept. 9, 2014), aff’d, 531 B.R. 321 (S.D.N.Y. 2015), appeal pending, No. 15-1771 (2d Cir. June 1, 2015); Markell, Fair Equivalents, supra note 16, at 120. transaction costs associated with going out into the market to obtain financing. The debtor and the lender have to find each other. The debtor has to convince the lender that it is viable and has a good business plan. Then the debtor has to bargain with the lender (in a market that might be imperfect) to achieve the best rate. And all of this has to be done quickly. The costs of doing so might deplete the cash of the debtor. As those costs add up, the outside loan might end up costing $110 rather than $100.

Everyone is better off if the estate keeps the cash, saves the transaction costs, and gives the old creditor a promise of $100. At that point, we have created a surplus (in savings) of $10 that can be divided among the stakeholders. Bankruptcy law does not trust the parties to reach this outcome on their own. The senior creditor might, as we have suggested, have idiosyncratic preferences or, more likely, it might play a hard bargaining game to try to extract the surplus for itself. And while the law should not really be concerned with who gets the surplus, it should be concerned about whether bargaining failure destroys value of the estate. And so we allow cramdown, which is merely the “don’t pay too much” principle.

B. “Don’t Pay Too Little”

There is, however, a flipside to all of this. If debtors get a consistently better rate in cramdown than they would on the market, the rate would function as a subsidy for cramdowns that artificially moves demand to those loans, even though market loans are better. This subsidy creates two particular distortions that destroy firm value. The first among these distortions is to filing incentives. If debtors can lower interest rates below market simply by filing bankruptcy, all firms will have an incentive to go through bankruptcy after taking on a loan, and whenever they are considering refinancing. Conversely, creditors will have an incentive to expend resources to prevent such bankruptcies. The second distortion is to the decision a debtor must make while in bankruptcy between cramming down a loan, allowing foreclosure, or refinancing. If cramming down a loan produces a below-market rate, a debtor will favor doing so when the market forces would otherwise suggest refinancing or submitting to foreclosure.

These distortions from normal market forces are not justified by any theory of corporate reorganization. To solve these problems, bankruptcy law should (and does) place the debtor in a position of choosing between options based on how those options benefit the stakeholders as a whole rather than based on opportunities to capture value from one stakeholder. Bankruptcy law, therefore, seeks to limit the instances of, and incentives for, a debtor to engage in opportunistic behavior by requiring the debtor to pay market rates when it opts to cram down a loan. This only works if the market rate is taken from the real market. If the court allows an intrinsic value that is below market, or a rough estimate of a below-market rate (like the Till rate), the debtor will still act opportunistically.

This is merely the “don’t pay too little” principle.

C. “Don’t Expect Precision”

Of course, determining the market rate for any exit loan is difficult. There will be noise and inaccuracy. Evidence can be hard to find and expert testimony can be unreliable. But that inaccuracy does not mean we should give up. An imprecise estimate is better than no estimate or one known to be wrong. As the saying goes, don’t let the perfect be the enemy of the good. This is merely the “don’t expect precision” principle.

Oddly, the market-rate approach is, nonetheless, rejected (by Professor Markell and the judge in Momentive) in part because it is too imprecise. Professor Markell provides a lengthy discussion about why inefficient markets will not reflect intrinsic values.  87Markell, Fair Equivalents, supra note 16, at 117–20. The judge in Momentive focused specifically on the fact that market rates might include transaction costs or profits that are not reflective of actual intrinsic values.  88Momentive, 2014 WL 4436335, at *29. If these arguments are about precision and error, we can reject them outright; Markell’s principles establish that precision is not required.

A more nuanced argument can, however, be built around this point and on the foundations of bankruptcy’s endowment effect: the absence of an efficient market means that the market rate is not estimating the intrinsic value of the asset (here the stream of payments) at all. The lack of equivalence with intrinsic value is problematic from an endowment perspective—creditors are not getting exactly what they are entitled to. And so, if the point of equitable equivalence is to preserve nonbankruptcy rights just for the sake of preserving them, we might have to give a creditor a payment with the same intrinsic value as its prepetition interest in the estate. And inefficient market rates do not do that. At its core, then, an endowment view of bankruptcy requires that fair equivalence means equivalence “of intrinsic value,” which requires that any market estimate be an efficient-market estimate.

But these problems only exist for the endowment perspective. If, as I have argued above, the purpose of fair equivalence is not to protect endowments but to align incentives, then fair equivalence only requires that creditors get (or more accurately, that debtors pay) the actual market rate for the cramdown loan, regardless of whether that rate is efficient. Incentives are set and altered by actual (and imperfect) markets that exist in the real world and not by perfectly efficient markets that exist in economic models. We therefore need only concern ourselves with those actual markets.

The inefficient-market critique is, in this way, a red herring. To give the creditor a fair equivalent of its interest in the cramdown context—to not pay too much or too little—is to require the debtor to pay a rate equal to the prevailing market rate (net of transaction costs) for similar exit loans. Payments on those terms will, by definition, also be equivalent to the stream of payments that the creditor could buy on the real market (not the market we wished existed) with the cash value of its interest (ignoring transaction costs).

As a final objection to the inefficient-market critique, it is odd to suggest (as Professor Markell and the judge in Momentive do) that cash is worth anything less than what it can buy on the market. After all, the value of money is entirely based on what it can buy (in the real world). If $100 in cash can buy a stream of payments secured by an asset with certain terms at a 10% interest rate, then regardless of the market inefficiencies (indeed, because of the market efficiencies) that cash is worth exactly that stream of payments. Giving someone a stream of payments with the same exact terms but a less-than-10% interest rate is not equivalent in any sense. This would be true even if the 10% being paid on the market were all supracompetitive “profit” (which seems to be the concern of the judge in Momentive).  89Id. at *26.

The counterargument is hard to grasp. The judge in Momentive seems to argue that if the market is mispricing the loan, the stream of payments with a 10% interest rate is a better deal than the cash because the market is mispriced.  90Id. at *29. The lender who makes a market loan is, therefore, being overpaid and getting a windfall, and the crammed-down lender would also get a windfall if that rate were used. That logic is not obviously correct. At best, it would only be true if the crammed-down secured creditor could not access the inefficient market with its cash (again, cash is worth what it can buy).

But even so, let us say there really is overpayment here. It still does not matter. The reality is that giving (or not giving) the creditor the exact equivalent of the cash is not the thing that should concern bankruptcy law. It is only the incentive effects on the debtor that the law should care about for cramdown. And from that perspective it is only the actual market rate—not the intrinsic value of the investment—that has any importance.

D. “Do What Makes Sense”

To summarize, if a cramdown loan from the secured creditor would be cheaper than a market loan—because of transaction costs—we want the debtor to choose to keep that loan. If, however, the debtor can get a better rate from a well-informed outsider, then it should take that loan and pay off the creditor. And if the cost of a market loan—excluding transaction costs—is still greater than the value of the collateral to the enterprise, then we want the debtor to allow for foreclosure.

While the Till formula does not achieve this outcome, a cramdown rule that requires the debtor to pay the secured creditor the actual market rate, net of transaction costs, does. Such a rule is not, and need not be, the same as requiring the debtor to give the creditor the intrinsic value of the creditor’s interest.

For cramdown to work, the debtor needs to be in a position where it cannot gain opportunistic advantage by choosing a less-efficient reorganization plan. We want the debtor to choose the efficient outcome and then force it upon the creditor. The law limits opportunistic creditor behavior by prohibiting it from demanding too much—that is, so much that it would destroy value of the estate. The law limits opportunistic debtor behavior by prohibiting it from paying too little—that is, so little that it would destroy value of the estate.

And the law should not worry about getting intrinsic value wrong as long as it sets the incentives well enough to produce a sensible set of bankruptcy decisions.

IV. Examples & Momentive

A. Examples

Some examples can add clarity to the insights set forth above.

1. An Example without Cramdown

Imagine a debtor in bankruptcy has an asset that is subject to a lien. The asset is more valuable in the hands of the debtor than it is when put to other uses. In other words, the debtor in possession of the asset has going-concern value. The secured creditor has threatened foreclosure, which led to the debtor’s bankruptcy filing. The debtor is considering cramming down a loan or looking for financing on the market to preserve its going-concern value.

Assume also that the loan is a good business decision if the interest rate is 10%. But if it costs anything more than 10%, the loan is not worth it; the debtor would be better off selling the asset or letting the creditor foreclose. The loan is worth the price of 10%, but not a penny more.

Now let us assume that the actual market rate for that loan would be 10% if transaction costs were zero. That means the debtor will not be able to find financing in a world where there are any transaction costs. Those transaction costs will either be borne directly by the debtor or show up as a higher interest rate for the loan. Either way, the debtor will not borrow the money. Finally, assume that the secured creditor is like any other lender and would also make the loan at 10% if it had full information and bargaining did not break down.

With those assumptions, bargaining failure between the secured creditor and the debtor will mean that the debtor loses the asset and the going-concern value is destroyed. Without cramdown, there is nothing the debtor can do. The secured creditor acting with bad information or bluffing refuses to offer 10% financing, and the asset is taken out of the estate.

2. An Example with Cramdown at the Market Rate

With these assumptions, cramdown at the market rate fixes things. The best outcome occurs if the debtor can cram down a loan on the secured creditor at 10%. As long as the secured creditor has the same cost of capital as the market lenders who would loan at 10%, then it should welcome the chance to have the 10% loan imposed upon it without incurring transaction costs.  91There is the issue of legal fees associated with the bankruptcy. But those are sunk costs to the creditor at this point. The debtor, in contrast, is choosing between the market rate (10% plus transaction costs) and the cramdown rate (10%), and has every incentive to choose the efficient rate.

Of course, sometimes the market loan will be better than the one the secured creditor can offer. Cramdown at market rate gets those cases right as well. Assume that the secured creditor cannot lend at the market rate. The secured creditor is, itself, in distress, and is no longer competitive with the market.  92This is also assuming that the secured creditor cannot easily sell the loan after the reorganization. But this is a consistent assumption because we have already assumed high market transaction costs. The secured creditor will lose value if it is forced to make a loan at 10%. Here, the secured creditor can settle with the debtor and offer to pay the debtor money not to cram down the loan. If that settlement process does not break down, it will produce the optimal bankruptcy outcome.  93If the market rate and the transaction costs are common knowledge, then the bargaining process should be smooth. But those might be bold assumptions. If the transaction costs of going to the market are lower than the harm to the secured creditor, then the debtor will take the offer. If the transaction costs are higher, it will not. The debtor will only choose cramdown to avoid the transaction costs that hurt the estate as a whole.

Note that when the creditor pays to avoid cramdown, the creditor is getting less than its prebankruptcy entitlement. It is paying to avoid being forced to make an unprofitable loan. From an entitlement perspective, therefore, this outcome is bad. But from an opportunistic-behavior perspective, it is a good outcome. Because the debtor cannot get better than the market rate in cramdown, it does not have an incentive to use cramdown to extract rents. Rather, the parties are just bargaining over who pays the transaction costs.

To be sure, this outcome is imprecise. Sometimes the secured creditor will be forced to make an inefficient loan. For example, if the secured creditor has a higher cost of capital and would charge 11% for the loan, but the market charges 10% (plus transaction costs that are less than 1%), it is not, strictly speaking, the right allocation of capital for the secured creditor to be loaning at 10%.

In a perfect world, we want a rule that does not cram down the loan in this instance. But there is no perfect way to distinguish between, on the one hand, an actual high-cost inside lender and, on the other hand, a lender that is either pretending to be, or taking measures to become, a high-cost lender to gain bargaining leverage or avoid cramdown in other instances. Cramdown, then, is a little overbroad. But, if the rate gets things right in most cases, allowing for a little imprecision is nonetheless consistent with maximizing the value of the bankruptcy estate in general. And we should expect that creditors will be close to market-rate lenders in most cases.

3. An Example with Cramdown at a Non-Market Rate

Things look very different if the debtor can cram down the secured creditor at a below-market rate. Keeping the market rate at 10%, let us now assume a cramdown rate of 5%. To be concrete, the parties have full information that a willing and able market lender would charge 10% to provide a new loan secured by the same collateral and under the same terms, but the debtor is able to cram down the loan on the secured creditor at 5%.  94This scenario is essentially what happened in Momentive. 2014 WL 4436335, at *29. There, the market lender was willing to make the loan at around 6.25%, and the cramdown rate ended up being around 4%. Id. at *32.

The debtor stands to gain 5% over the market (on top of saving transaction costs) by cramming down the secured creditor. That gives the debtor an incentive to opt for cramdown even in cases where the secured creditor’s cost of capital is greatly in excess of the market. But the problems go deeper than that. Imagine that the debtor’s going-concern value is such that the loan is not profitable for the debtor at any rate higher than 6%. Now we have a firm that would never survive in the market, forcing its creditors to subsidize its existence with a below-market loan. This produces a reallocation of capital in the economy unsupported by any bankruptcy purpose.  95There might be reasons to support such a reallocation in chapter 13, where we might want to make sure that lenders are not earning supracompetitive profits on the backs of individuals who are denied a fresh start. We might think that bankruptcy is the entry point of last resort to protect against a lending market that is taking advantage of unsophisticated borrowers. Or, we might simply not want to doom individual debtors to failure with interest rates that cannot possibly be paid.But none of those concerns is present in chapter 11. Chapter 11 is intended to maximize the value of the estate. See 7 Collier on Bankruptcy, supra note 73, ¶ 1100.01. The way to accomplish that goal is to make sure that viable firms are reorganized and nonviable firms are liquidated. See id. (“Chapter 11 of the Bankruptcy Code provides an opportunity for a debtor to reorganize its business or financial affairs or to engage in an orderly liquidation of its property either as a going concern or otherwise.”). And along the way, we want to make sure the process does not create avenues for stakeholders to opportunistically transfer value from one stakeholder to another. Even if capital markets are inefficient, it would be a drastic move (not supported by any history of, or any provision in, the Code) to enlist bankruptcy as the means of compelling banks to subsidize reorganized firms to correct for a perceived inefficiency in market-wide lending rates.

Moreover, the below-market cramdown rate creates an incentive for healthy and failed firms alike to file for bankruptcy to refinance their existing loans. And those incentives of debtors, in turn, create incentives for lenders to account for the risk on the front end by charging large origination fees, raising interest rates, or simply refraining from making loans in the first place.

Finally, the below-market cramdown rates provide the debtor with the incentive and opportunity to hold up creditors and use the threat of cramdown to extract value. In our example, a cramdown at 5% is a below-cost loan from the secured creditor. The debtor may offer a choice to the creditor: pay me your cost or accept the cramdown. With a market cramdown rate, the most the debtor can extract from the creditor is the difference between the creditor’s cost of capital and the general cost of capital on the market. And that difference will be zero in expectation. In our below-market example, the difference is the 5% interest rate in expectation. This number can be significant. In Momentive, it was a difference that amounted to potentially $200 million.  96Markell, Fair Equivalents, supra note 16, at 123 (noting that some estimates put the differential in value in Momentive at $200 million). And in Momentive, the debtors did, in fact, use a threat of cramdown to extract value from the secured creditors.  97Disclosure Statement for Joint Chapter 11 Plan of Reorganization for Momentive Performance Materials Inc. & Its Affiliated Debtors, supra note 19, at 35–36. Through their “death-trap” voting provisions, they offered cash payment in exchange for the creditors waiving their right to object to the plan.  98See id. The particular objection there had to do with make-whole payments, which are controversial.  99For further discussion of make-whole premiums, see generally Markell, Make-Whole Premiums, supra note 11, at 3–4 (providing background information on how make-whole premiums operate, and how courts, including the court in Momentive, have either allowed or disallowed them). But if Momentive is affirmed, there is no reason to think that the use of threats in the future will be limited to demands to waive controversial objections, as opposed to broader demands to waive standard procedural rights.

These examples demonstrate that cramdown equivalence geared to actual markets focuses debtors’ choices away from opportunistic behavior, while below-market cramdown rates provide opportunities for value extraction. Of course, these examples do assume that the market rate can be ascertained. And that is a big assumption. But here, the Markell principles can do a great deal of work. The market rate need not be known with mathematical certainty for it to serve its purpose. As long as it is fairly equivalent to the actual market in expectation, it will cabin the spectrum of opportunistic behavior. A rough, unbiased estimate is neither too low nor too high. What is too low is a rate that is known with certainty to be $200 million below the market rate.

B. Momentive

The outcome and application of Till to Momentive is, therefore, puzzling. The bankruptcy court was essentially throwing away information.  100Momentive, No. 14-22503-rdd, 2014 WL 4436335, at *25–29 (Bankr. S.D.N.Y. Sept. 9, 2014), aff’d, 531 B.R. 321 (S.D.N.Y. 2015), appeal pending, No. 15-1771 (2d Cir. June 1, 2015). The difficulty of estimating the market was not an issue at all because there existed two objective measures of the market for exit loans. The first was the rate of the loan commitment the debtor had lined up to fund the potential cash-out of the senior creditors if they took the offer in the death-trap agreement.  101Id. at *11. The second was the market price of the notes when the court approved the cramdown rate.  102See Reply Brief of Appellant Wilmington Trust, N.A., as Indenture Trustee for the 1.5 Lien Notes at 17, Momentive Performance Materials, Inc. v. BOKF, NA (In re MPM Silicones, L.L.C.), Nos. 15-1682, 15-1824 (2d Cir. Mar. 30, 2016), 2016 WL 1380968, at *17. This evidence suggested that the cramdown rate was significantly below the market rate for that type of exit loan.

Despite this clear evidence of the market value of an exit loan, the court applied the Till formula to justify a much lower rate based on a theory that the market rate might improperly include profit and transaction costs, to which the secured creditor was not entitled.  103Momentive, 2014 WL 4436335, at *25. Implicit in my analysis above is the point that the court was right about transaction costs. But there is no way the transaction costs for a loan come anywhere near the underpayment discount in Momentive, which by some evidence was 14% of the entire loan value, and by other evidence was nearly $200 million.  104See Reply Brief of Appellant Wilmington Trust, N.A., as Indenture Trustee for the 1.5 Lien Notes, supra note 102, at 6 (stating loan value was reduced by over 14% by court’s ruling); Markell, Fair Equivalents, supra note 16, at 123 (finding estimates at $200 million).

As for profits, the court was wrong to exclude them.  105Momentive, 2014 WL 4436335, at *29. The idea that part of the market rate included profits is only relevant for entitlement frameworks. From an opportunistic behavior framework, however, the amount of profit is irrelevant. The debtor simply compares the cramdown rate with the market rate in choosing whether to act opportunistically. It does not care whether those rates include profits; it only cares which rate is lower.

As noted, the court’s discounted rate opens the door to attempts by the debtor to extract value by demanding procedural waivers. It also will likely change the ex ante bargain in the general market between senior creditors on the one hand and debtors and junior creditors on the other.

The ability of senior creditors to adjust ex ante rates in future loans means that the direct wealth transfers created by the Momentive ruling will be unimportant.  106Justice Thomas noted this in his Till concurrence. 541 U.S. 465, 488 (2004) (Thomas, J., concurring). But the procedural distortions cannot be so easily erased. Every estate, as a whole, will be worse off because the stakeholders cannot commit ex ante to refrain from this opportunistic behavior. The best they can do is pay each other in advance for the option to act opportunistically even when everyone is made worse off by the existence of that option. This inability to commit to optimal behavior is precisely the type of ex ante contracting problem that bankruptcy law is designed to prevent and why Momentive is inconsistent with fundamental bankruptcy principles, with the statute, and with the Markell principles.

Conclusion

Professor Markell moves the analysis of cramdown, absolute priority, and corporate bankruptcy forward a great deal with his insightful analysis and three simple principles. But, as I suggested at the beginning, there is a fourth principle that should also govern in these matters: “do what makes sense.” That is to say, the rules should further good bankruptcy policy.

Of course, courts are not supposed to simply make up new rules just to achieve pragmatic ends. Fortunately, though, sense and good policy require nothing of the sort here. The statute and the history are well crafted. They require equivalence, and equivalence, properly defined, gets the right result.

The Till rate, however, is not the equivalent of anything. It is not a market rate. Nor does it have anything to do with intrinsic value. The Till rate, therefore, fails both the endowment and the opportunistic-behavior approaches. In contrast, the market rate—while certainly imprecise—is consistent with the statute and is, by any metric that considers the proper purpose of corporate bankruptcy, a fair equivalent that aligns incentives and reduces opportunistic behavior. It is, therefore, neither too high nor too low to achieve bankruptcy’s proper purpose.

The market rate also frees courts from unnecessary concerns about market efficiency. While market rates may incidentally include profits that exist because of inefficient markets, there is nothing wrong with that from a policy or statutory perspective. There is no reason for courts to attempt to estimate intrinsic value in perfect markets when imperfect-market equivalence satisfies both the statute and the purpose behind it. Those attempts, after all, led the Momentive court to stretch the notion of equivalence so far as to allow for a rate that is known to be tens or hundreds of millions of dollars below market value. Courts should instead focus on ensuring that debtors pay actual market rates to reduce the risk of opportunistic behavior. Doing so furthers the ultimate purpose of maximizing the value of the estate in a way that intrinsic-value methods do not.

Footnotes

Professor of Law and Mark Claster Mamolen Teaching Scholar, The University of Chicago Law School. I thank Elizabeth Kiernan, Alex Weber, and Stephanie Xiao for excellent research assistance and The Richard Weil Faculty Research Fund and The Paul H. Leffmann Fund for generous research support.

1See, e.g., Oliver Hart, Firms, Contracts, and Financial Structure 160 (Oxford 1995); Barry E. Adler & Ian Ayres, A Dilution Mechanism for Valuing Corporations in Bankruptcy, 111 Yale L.J. 83, 88 ̶ 90 (2001); Barry E. Adler & George Triantis, Debt Priority and Options in Bankruptcy (Mar. 2, 2016) (unpublished manuscript) (on file with authors); Walter J. Blum, Commentary, The “New Directions” for Priority Rights in Bankruptcy Reorganizations, 67 Harv. L. Rev. 1367, 1374 (1982) [hereinafter Blum, New Directions]; Thomas H. Jackson, Bankruptcy, Non-Bankruptcy Entitlements, and the Creditors’ Bargain, 91 Yale L.J. 857, 857–58 (1982).

2See Adler & Ayres, supra note 1, at 88 ̶ 90; Adler & Triantis, supra note 1; Jackson, supra note 1, at 857 ̶ 58.

3Jackson, supra note 1, at 860.

4I use the term “estate” in a nontechnical sense throughout this Article to refer to the collective interests of all stakeholders, not just one class of creditors.

5Anthony J. Casey & Aziz Z. Huq, The Article III Problem in Bankruptcy, 82 U. Chi. L. Rev. 1155, 1194 (2014) (“Bankruptcy’s proper goal is rather best understood as one of limiting certain private rights to protect others.”).

6See Jackson, supra note 1, at 869.

7I have made this point elsewhere. See Anthony J. Casey & Edward R. Morrison, Beyond Options (2016) (unpublished manuscript) (on file with authors); see also Douglas G. Baird, Priority Matters, 165 U. Pa. L. Rev. (forthcoming 2016) (manuscript at 47 ̶ 48) (on file with the University of Pennsylvania Law Review) [hereinafter Baird, Priority Matters].

8See Adler & Ayres, supra note 1, at 88 ̶ 90; Adler & Triantis, supra note 1; Jackson, supra note 1, at 857 ̶ 58.

9Including me. Anthony J. Casey, The Creditors’ Bargain and Option-Preservation Priority in Chapter 11, 78 U. Chi. L. Rev. 759, 778 (2011); see also Melissa Jacoby & Edward Janger, Ice Cube Bonds: Allocating the Price of Process in Chapter 11 Bankruptcy, 123 Yale L.J. 862, 916, 918, 922 (2013); cf. Omer Tene, Revisiting the Creditors’ Bargain: The Entitlement to the Going-Concern Surplus in Corporate Bankruptcy, 19 Bankr. Dev. J. 287, 396 (2003).

10See In re MPM Silicones, LLC (Momentive), No. 14-22503-rdd, 2014 WL 4436335 (Bankr. S.D.N.Y. Sept. 9, 2014), aff’d, 531 B.R. 321 (S.D.N.Y. 2015), appeal docketed, No. 15–1771 (2d Cir. June 1, 2015).

11There was a separate dispute about make-whole payments that also focused on creditor entitlements, but the outcome there turned in large part on whether the state-law entitlements existed in the first place. See Bruce A. Markell, “Shoot the. . .”: Holes in Make-Whole Premiums, 36 Bankr. L. Letter No. 6 (Thomson Reuters, Saint Paul, Minn.), May 2016, at 3–4 [hereinafter Markell, Make-Whole Premiums].

12Momentive, 2014 WL 4436335, at *23–32.

13541 U.S. 465, 478 ̶ 80 (2004).

14Justice Thomas is the exception. See Till, 541 U.S. at 473–74, 489, 491–92 (Thomas, J., concurring) (focusing on statutory language rather than nonbankruptcy entitlements).

15The purpose of chapter 13 may be different enough to justify this approach in Till, but not in chapter 11 cases like Momentive. I discuss this briefly below, infra note 95. But my primary inquiry here is into the bankruptcy court’s move to import Till’s analysis into chapter 11 corporate bankruptcies. Momentive, 2014 WL 4436335, at *23–32. In chapter 13 cases, we might worry about different things for policy reasons unrelated to corporate reorganization. Chapter 13 deals with individuals’ and society’s interest in providing opportunities for a fresh start. See H.R. Rep. No. 95-595, at 117 (1977). Chapter 13 also exists in a nonmarket environment—there is no going-concern sale, no DIP lender, and no market for exit financing. For individuals, there is simply the break-up of the assets (chapter 7) or the option to cram down a loan (chapter 13). Conditional upon being in chapter 13, the debtor is trying to keep things together without market support. We might not want to require an interest rate that makes it impossible to do that.

16Bruce A. Markell, Fair Equivalents and Market Prices: Bankruptcy Cramdown Interest Rates, 33 Emory Bankr. Dev. J. 91, 114 (2016) [hereinafter Markell, Fair Equivalents] (quoting Till, 541 U.S. at 476 n.14 (2004)).

17Id. at 95–103.

1811 U.S.C. § 1129(b) (2012).

19According to the debtor’s plan in Momentive, Momentive would pay cash to the First Lien Notes and the 1.5 Lien Notes in an amount equal to their face value, including accrued interest, if they agreed to the plan and dropped certain objections. Disclosure Statement for Joint Chapter 11 Plan of Reorganization for Momentive Performance Materials Inc. & Its Affiliated Debtors at 35–36, In re MPM Silicones, LLC, No. 14-22503-rdd (Bankr. S.D.N.Y. Sept. 9, 2014), ECF No. 516, 2014 WL 4255110, at *35–36. If the note holders rejected this proposed treatment, however, Momentive would pay the note holders over time at an interest rate crafted according to Till. Id. See generally Markell, Fair Equivalents, supra note 16, at 125 (discussing the ramifications of Momentive’s proposed plan). The result was that the noteholders would receive lower value if they objected to the plan. Id.

20The absolute priority rule prevents equity holders from retaining or realizing any value in a reorganized debtor company unless creditors consent or are paid in full. See 11 U.S.C. § 1129(b); Bruce A. Markell, Owners, Auctions, and Absolute Priority in Bankruptcy, 44 Stan. L. Rev. 69, 72 & n.17 (1991).

2111 U.S.C. § 1129(b)(1).

22 See generally Baird, Priority Matters, supra note 7, manuscript at 4.

23Markell, Fair Equivalents, supra note 16, at 95–103 (outlining the caselaw and legislative history behind the “fair and equitable” standard).

24Id. at 103–04.

25Id. at 104.

2611 U.S.C. § 1129(b)(1).

27The specific language states: “[T]he condition that a plan be fair and equitable with respect to a class includes the following requirements . . . .” Id. § 1129(b)(2). For cramming down a secured creditor, the plan is required to provide the creditor with one of these three things: (1) retention of its liens along with deferred cash payments “of a value, as of the effective date of the plan, of at least the value” of the creditor’s interest in the property subject to the lien; (2) the right to credit bid in a sale of the property subject to the lien; or (3) the “indubitable equivalent” of the creditor’s claims. See id. § 1129(b)(2)(A).

28Markell, Fair Equivalents, supra note 16, at 130.

29Id.

30See In re Castleton Plaza, LP, 707 F.3d 821, 823 (7th Cir. 2013) (noting that compliance with § 1129(b)(2) was not all that was required to satisfy absolute priority).

31Markell, Fair Equivalents, supra note 16, at 131.

32Id. at 130.

3311 U.S.C. § 1129(b)(2) (“[T]he condition that a plan be fair and equitable with respect to a class includes the following requirements . . . .”).

34132 S. Ct. 2065, 2069–70 (2012).

35Id. This interpretation is consistent with the statutory language. It is also consistent with the history of the Code and the concerns of the drafters. River East Plaza, LLC v. Geneva Leasing Ass’n (In re River East Plaza, LLC), 669 F.3d 826, 828–29 (7th Cir. 2012); see also Douglas G. Baird, Remembering Pine Gate, 38 J. Marshall L. Rev. 5, 9 (2004) [hereinafter Baird, Pine Gate].

36526 U.S. 434, 437 ̶ 38 (1999).

37Id. at 441.

38Id. at 458.

39Id. at 436.

40See Douglas G. Baird & Anthony J. Casey, Bankruptcy Step Zero, 2012 Sup. Ct. Rev. 203, 220.

41I have explored these lessons more fully elsewhere:Together these cases start to form a pattern. On certain questions, the market and Congress are the only competent arbiters. To the extent Congress is unclear in addressing those questions, the Court will view it as providing as narrow a rule as is reasonably consistent with the Code’s language.Id. at 221; see also Casey & Huq, supra note 5, at 1215–16.

42Markell, Fair Equivalents, supra note 16, at 104.

4311 U.S.C. § 1129(b)(2)(A)(i)(II) (2012) (“[T]hat each holder of a claim of such class receive on account of such claim deferred cash payments totaling at least the allowed amount of such claim, of a value, as of the effective date of the plan, of at least the value of such holder’s interest in the estate’s interest in such property.”).

44541 U.S. 465 (2004) (plurality opinion). This point was somewhat irrelevant for the bankruptcy court in Momentive. See Momentive, No. 14-22503-rdd, 2014 WL 4436335, at *24 (Bankr. S.D.N.Y. Sept. 9, 2014), aff’d, 531 B.R. 321 (S.D.N.Y. 2015), appeal docketed, No. 15-1771 (2d Cir. June 1, 2015). It was obviously bound by decisions of the Second Circuit, and the Second Circuit had adopted a similar approach to the Till plurality in an earlier case. Id. (citing GMAC v. Valenti (In re Valenti), 105 F.3d 55, 63 (2d Cir. 1997)). Thus, statutory text and first principles aside, the bankruptcy court was bound to at least apply the Till rule to chapter 13 cases. Id.

45See 541 U.S. at 476 n.14; Markell, Fair Equivalents, supra note 16, at 110–13.

46Section 1325 deals with “the value, as of the effective date of the plan, of property to be distributed under the plan.” 11 U.S.C. § 1325(a)(4). Justice Thomas notes that the value of the cash to be distributed was its face value, regardless of the risk of nondistribution. Till, 541 U.S. at 485–86 (Thomas, J., concurring). He went on to note that if the “property distributed” was “a note (i.e., a promise to pay),” then the value of that note “necessarily includes” a risk premium. Id. at 488–89. Section 1129(b)(2)(A)(i)(II) talks of “deferred cash payments . . . of a value, as of the effective date of the plan, of at least the value of such holder’s interest in the estate’s interest in such property.” The question for Justice Thomas, then, is whether “of a value” modifies “payments” or “deferred cash payments.” The former would make the provision analogous to the case in Till; the latter would look more like Justice Thomas’s hypothetical “notes.” Put another way, “deferred cash payments” are promises to pay, and valuing them today (like valuing notes) implies a risk adjustment in a way that valuing payments or property (that happen to be coming in the future) does not. For my eye, the natural and plain reading is that “value” modifies the entire clause, “deferred cash payments.” In short, even if the plurality reasoning in Till applied to § 1129(b)(2)(A)(i)(II), it is likely that Justice Thomas’s reasoning does not. So for chapter 11, that would give a fifth vote to the dissent. Does that matter? Probably not. But it does make the case that Till somehow provides binding authority even more dubious.

47The historical support for the reasoning is set forth in Professor Markell’s piece. Markell, Fair Equivalents, supra note 16, at 95–103. I do not restate it here.

48Till, 541 U.S. at 491 (Scalia, J., dissenting).

49Id. at 491–92 (Thomas, J., concurring) (noting that the plurality started with “a rate we know is too low” and that the formula “will systematically undercompensate”).

50Id. at 504 (Scalia, J., dissenting) (noting that the plurality’s rate in Till was off by an order of magnitude).

51Markell, Fair Equivalents, supra note 16, at 103–04. Precision implies both exactness and accuracy. The Till rate is precise in the sense that it achieves relative certainty and accurately underestimates the interest rate. The three Markell principles, taken together, suggest that “don’t expect precision” means you do not need exactness, and unbiased inaccuracy is okay. We should not sacrifice fairness by introducing bias in the name of exactitude or mathematical certainty. Inaccurate equivalency is better than accurate inequivalency.

52Id. at 105 (quoting Grp. of Inst. Inv’rs, v. Chi., Milwaukee, Saint Paul & Pac. R.R., 318 U.S. 523, 565–66 (1943)).

53Id. at 98 (quoting Consol. Rock Prods. Co. v. Du Bois, 312 U.S. 510, 526 (1941)).

54Id. at 98–99 (quoting Grp. of Inst. Inv’rs, at 565).

55See Jackson, supra note 1, at 857. It perhaps goes further back. See Walter J. Blum, Full Priority and Full Compensation in Corporate Reorganizations: A Reappraisal, 25 U. Chi. L. Rev. 417, 437–39 (1958); Blum, New Directions, supra note 1, at 1368–69.

56See Jackson, supra note 1, at 858.

57See id. at 862–63.

58Id. at 862.

59See id. at 867–68.

60By “all states of the world,” I mean to encompass the various contingencies that might play out both in and out of bankruptcy. The parties, ex ante, will desire to maximize the overall expected value of the firm based on their probabilistic expectations about which state of the world will ultimately result. If a rule produces a benefit in one state of the world, the value of the benefit must be discounted for any costs imposed in another state of the world (weighted by the probabilities that each state will actually come to exist).

61Casey, supra note 9, at 796; see also Comm’n to Study the Reform of Chapter 11, Am. Bankr. Inst., Final Report and Recommendations 75 (2014).

62Baird, Priority Matters, supra note 7, manuscript at 32 (“It is also worth noting that we do not see anything as simple as the absolute priority rule in an analogous environment in which free contracting is permitted.”).

63I have stressed this point in prior articles. Casey, supra note 9, at 772–73; Casey & Morrison, supra note 7; see also Baird, Priority Matters, supra note 7, manuscript at 22 (“The debate about relative and absolute priority must therefore take place in an empirical vacuum.”).

64The idea that bankruptcy law can and should change nonbankruptcy endowments is consistent with the Supreme Court’s holding in United States v. Butner. 440 U.S. 48, 54 (1979). Butner is sometimes cited for the principle that bankruptcy law should leave state law undisturbed. That application is not quite right. The case merely says that in the absence of a bankruptcy statute, courts should look to state law. Id. at 57–58. But here, there is a statute (11 U.S.C. § 1129) and so we look to it and not to state law. Indeed, in Butner the Court explicitly said that the constitutional bankruptcy power “would clearly encompass a federal statute” defining a creditor’s interest in estate property. Id. at 54. The Court also noted that a nonbankruptcy state interest could be altered by bankruptcy law if a federal (that is, bankruptcy) interest required that result. Id. at 55.

65See, e.g., RadLAX Gateway Hotel, LLC v. Amalgamated Bank, 132 S. Ct. 2065, 2069 ̶ 70 (2012); Bank of Am. Nat’l Tr. & Sav. Ass’n v. 203 N. LaSalle St. P’ship, 526 U.S. 434, 457–58 (1999). See generally Baird & Casey, supra note 39, at 208–10.

66132 S. Ct. at 2069 ̶ 70.

67These rules are also consistent with historical cases of opportunistic behavior that were on the minds of the drafters of the Code. See In re Pine Gate Assocs., Ltd., No. B75-4345A, 1976 U.S. Dist. LEXIS 17366 (N.D. Ga. Oct. 14, 1976); In re Pine Gate Assocs., Ltd., 12 Collier Bankr. Cas. (MB) 607 (Bankr. N.D. Ga. Mar. 4, 1977); Baird, Pine Gate, supra note 34, at 5.

6811 U.S.C. § 1129(b)(2)(A)(iii) (2012).

69In re River East Plaza, LLC, 669 F.3d 826, 831–32 (7th Cir. 2012).

70See id. at 832 (“[B]ecause of the different risk profiles of the two forms of collateral, they are not equivalents, and there is no reason why the choice between them should be made for the creditor by the debtor.”).

71See 526 U.S. 434, 457–58 (1999).

72See 11 U.S.C. § 362(a).

733 Collier on Bankruptcy ¶ 362.03 (Alan N. Resnick & Henry J. Sommer eds. 16th ed.) (“[T]he stay provides creditors with protection by preventing the dismemberment of a debtor’s assets by individual creditors levying on the property. This promotes the bankruptcy goal of equality of distribution.”).

74Id. (“[The stay] protects property that may be necessary for the debtor’s fresh start and . . . provides breathing space to permit the debtor to focus on rehabilitation or reorganization.”).

75Id. ¶ 362.03[2] (“The stay prevents this piecemeal liquidation, offering the chance to maximize the value of the business.”).

76See generally Chrysler LLC v. Plastech Engineered Prods., Inc. (In re Plastech Engineered Prods., Inc.), 382 B.R. 90, 107–08 (Bankr. E.D. Mich. 2008) (struggling with competing claims about whether the invocation of, or the motion to lift, the automatic stay was opportunistic). A more general example would be when a debtor uses the automatic stay to hold seasonal or perishable goods hostage.

77See 11 U.S.C § 362(d).

78It may be wise to give at least some of the bankruptcy surplus to the person or entity that makes the filing decision to ensure that value-creating filings do occur. But beyond that, the distributional implications are unclear. See Alan Schwartz, A Normative Theory of Business Bankruptcy, 91 Va. L. Rev. 1199, 1239 (2005). Compare Barry E. Adler, Illinois ABI Symposium on Chapter 11 Reform: Priority in Going-Concern Surplus, 2015 U. Ill. L. Rev. Online 811, 814 (2015), and Adler & Triantis, supra note 1, at 19–20, with Douglas G. Baird & Donald S. Bernstein, Absolute Priority, Valuation Uncertainty, and the Reorganization Bargain, 115 Yale L.J. 1930, 1937 (2006), and Casey, supra note 9, at 778, with Jacoby & Janger, supra note 9, at 916, 918, 922, and Tene, supra note 9, at 395–96, with Baird, Priority Matters, supra note 7.

79Allowing debtors (or junior creditors) to capture foreclosure value from senior creditors by entering bankruptcy would have at least three major effects: (1) debtors will spend resources to enter bankruptcy even where it is not value maximizing; (2) creditors will spend resources to prevent bankruptcy even when the bankruptcy would be value maximizing; and (3) the cost and availability of capital for debtors outside of bankruptcy will be adversely impacted because the risk for creditors will be higher.

80See George G. Triantis, Mitigating the Collective Action Problem of Debt Enforcement Through Bankruptcy Law: Bill C-22 and Its Shadow, 20 Can. Bus. L. Rev. 242, 252 (1992) (“The cram down is one way to address the problem of holdouts by classes of secured lenders.”).

81The Coase Theorem posits that where there are completely competitive markets with no transaction costs, parties in a transaction will always reach the most efficient outcome, regardless of how property rights are allocated. See generally R.H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1 (1960).

82The creditor may also have artificial reasons for foreclosing on assets even when doing so destroys value. Such reasons often stem from external forces such as regulatory pressure from tax and accounting rules. See, e.g., In re RTJJ, Inc., No. 11-32050, 2013 WL 462003, at *15 (Bankr. W.D.N.C. Feb. 6, 2013) (“First National recognizes this. Community One does not seem to care. Under pressure from federal regulators, Community One seeks to rid itself of this nonperforming loan, at any cost. Its aims are noneconomic—at least as to this Debtor—and are destructive.”).

83The debtor could also sell the asset subject to credit bidding. But, assuming the creditor is undersecured, that alternative amounts to the same thing as foreclosure.

84See 11 U.S.C. § 1129(b)(2) (2012); In re River East Plaza, LLC, 669 F.3d 826, 828–33 (7th Cir. 2012) (summarizing the options).

85It also prevents the creditor from incidentally profiting from those costs. But that is not the rule’s primary purpose.

86See Momentive, No. 14-22503-rdd, 2014 WL 4436335, at *25–29 (Bankr. S.D.N.Y. Sept. 9, 2014), aff’d, 531 B.R. 321 (S.D.N.Y. 2015), appeal pending, No. 15-1771 (2d Cir. June 1, 2015); Markell, Fair Equivalents, supra note 16, at 120.

87Markell, Fair Equivalents, supra note 16, at 117–20.

88Momentive, 2014 WL 4436335, at *29.

89Id. at *26.

90Id. at *29.

91There is the issue of legal fees associated with the bankruptcy. But those are sunk costs to the creditor at this point.

92This is also assuming that the secured creditor cannot easily sell the loan after the reorganization. But this is a consistent assumption because we have already assumed high market transaction costs.

93If the market rate and the transaction costs are common knowledge, then the bargaining process should be smooth. But those might be bold assumptions.

94This scenario is essentially what happened in Momentive. 2014 WL 4436335, at *29. There, the market lender was willing to make the loan at around 6.25%, and the cramdown rate ended up being around 4%. Id. at *32.

95There might be reasons to support such a reallocation in chapter 13, where we might want to make sure that lenders are not earning supracompetitive profits on the backs of individuals who are denied a fresh start. We might think that bankruptcy is the entry point of last resort to protect against a lending market that is taking advantage of unsophisticated borrowers. Or, we might simply not want to doom individual debtors to failure with interest rates that cannot possibly be paid.But none of those concerns is present in chapter 11. Chapter 11 is intended to maximize the value of the estate. See 7 Collier on Bankruptcy, supra note 73, ¶ 1100.01. The way to accomplish that goal is to make sure that viable firms are reorganized and nonviable firms are liquidated. See id. (“Chapter 11 of the Bankruptcy Code provides an opportunity for a debtor to reorganize its business or financial affairs or to engage in an orderly liquidation of its property either as a going concern or otherwise.”). And along the way, we want to make sure the process does not create avenues for stakeholders to opportunistically transfer value from one stakeholder to another.

96Markell, Fair Equivalents, supra note 16, at 123 (noting that some estimates put the differential in value in Momentive at $200 million).

97Disclosure Statement for Joint Chapter 11 Plan of Reorganization for Momentive Performance Materials Inc. & Its Affiliated Debtors, supra note 19, at 35–36.

98See id.

99For further discussion of make-whole premiums, see generally Markell, Make-Whole Premiums, supra note 11, at 3–4 (providing background information on how make-whole premiums operate, and how courts, including the court in Momentive, have either allowed or disallowed them).

100Momentive, No. 14-22503-rdd, 2014 WL 4436335, at *25–29 (Bankr. S.D.N.Y. Sept. 9, 2014), aff’d, 531 B.R. 321 (S.D.N.Y. 2015), appeal pending, No. 15-1771 (2d Cir. June 1, 2015).

101Id. at *11.

102See Reply Brief of Appellant Wilmington Trust, N.A., as Indenture Trustee for the 1.5 Lien Notes at 17, Momentive Performance Materials, Inc. v. BOKF, NA (In re MPM Silicones, L.L.C.), Nos. 15-1682, 15-1824 (2d Cir. Mar. 30, 2016), 2016 WL 1380968, at *17.

103Momentive, 2014 WL 4436335, at *25.

104See Reply Brief of Appellant Wilmington Trust, N.A., as Indenture Trustee for the 1.5 Lien Notes, supra note 102, at 6 (stating loan value was reduced by over 14% by court’s ruling); Markell, Fair Equivalents, supra note 16, at 123 (finding estimates at $200 million).

105Momentive, 2014 WL 4436335, at *29.

106Justice Thomas noted this in his Till concurrence. 541 U.S. 465, 488 (2004) (Thomas, J., concurring).