Brooklyn Law School
BrooklynWorks Faculty Scholarship
Virtual Territoriality Edward J. Janger Brooklyn Law School, [email protected]
Follow this and additional works at: http://brooklynworks.brooklaw.edu/faculty Part of the Bankruptcy Law Commons, Consumer Protection Law Commons, Other Law Commons, and the Secured Transactions Commons Recommended Citation 48 Colum. J. Transnat'l L. 401 (2009-2010)
This Article is brought to you for free and open access by BrooklynWorks. It has been accepted for inclusion in Faculty Scholarship by an authorized administrator of BrooklynWorks. For more information, please contact [email protected]
Articles Virtual Territoriality EDWARD J. JANGER*
Current efforts to harmonize the laws of secured credit and bankruptcy, and to centralize bankruptcy case administrationare predicated on the belief that regularizing debtor's rights and creditor's remedies will cause global business to flourish and will benefit both developed and less-developed countries. Certain and predictable remedies for creditors will facilitate lending and development. Coordinationamong courts will create opportunities to protect the going concern value of troubled businesses. The benefits that accompany such legal harmonization may, however, come at a price. Centralizing control of a bankruptcy case may create opportunitiesfor forum shopping, and local legal variation can be a product of local culture, consideredpolicy choices or innovation. As a result, excessive harmonization can sweep away cultural differences, produce political resistance and stifle efficient experimentationand evolution of the law. In this article Professor Janger explores whether choice-of-law rules might be used to address these concerns about harmonization and centralization of
David M. Barse Professor, Brooklyn Law School. The author would like to thank
Susan Block-Lieb, Hannah Buxbaum, Hon. Allan L. Gropper, Claire Kelly, Bob Wessels and participants in faculty workshops at the Rutgers School of Law Camden, Indiana University School of Law and Brooklyn Law School, as well as participants in the 2007 INSOL Academics Workshop in Capetown, South Africa, for comments on earlier drafts. Lisa Baldesweiler and Shannon Sneed provided excellent research assistance. The author also wishes to thank Dean Joan Wexler and the Dean's Research Fund for generous support of this project.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
bankruptcy law. He advocates a choice-of-law principle that he calls "virtual territoriality." Virtual territoriality envisions a global,procedurally centralized bankruptcy case that, to the extent possible, respects the entitlements created by the various jurisdictions where the debtor does business. Under this view, the procedural bankruptcy laws of the "home" country should govern the case, while the choice of substantive law should be determined by ordinary (nonbankruptcy) choice-of-law principles. This approach, he argues, can form the basisfor a cross-borderbankruptcy architecture that simultaneously (1) facilitates the administrationof a global bankruptcy, and (2) facilitates the internationalacceptance of rescue based domestic insolvency regimes. IN TRODU CTION .................................................................................
CHOICE-OF-LAW UNDER THE EC REGULATION AND IN THE UNCITRAL LEGISLATIVE GUIDE .......................................... 410 A. Choice-of-Law and Choice-of-Forum ........................... 411 B. Choice-of-Law for the Modified Universalist ............... 414 C. Modified Universalism in Practice-The EC R egulation ...................................................................... 4 15 D. Modified Universalism in Practice-The UNCITRAL Legislative Guide ...................................... 418
UNIVERSAL PROCEDURALISM AND VIRTUAL TERRITORIALITY .....................................................................
A. Universal Proceduralism and the Limits of H arm onization ............................................................... B. Centralizing and Decentralizing Choice-of-Law Ru les .............................................................................. C. Secured Credit, Bankruptcy Law and the Need for a Decentralizing Choice-of-Law Rule .............................. D. Bankruptcy, Secured Credit and the Need to Distinguish Procedure from Substance .......................... E. V irtual Territoriality ...................................................... 1. Single F irm ............................................................... 2. Corporate Group .......................................................
422 423 42 6 428 429 432 433 434
F. Disaggregating Substance and Procedure ...................... 435
VIRTUAL TERRITORIALITY IN ACTION ................................... 436
A. Collins and Aikman-Virtual Territoriality
O perationalized ............................................ 436 B. Federal Mogul-Virtual Territoriality Ignored ............. 439
CO N CLU SION .................................................................. 440
Current efforts to harmonize the laws of secured credit and bankruptcy are predicated on the belief that regularizing the law of debtor's rights and creditor's remedies will cause global business to flourish, and benefit both developed and less-developed countries. Certain and predictable remedies for creditors will facilitate lending and development,1 and coordination among courts will create oppor2 tunities to protect the going concern value of troubled businesses. These benefits that accompany legal harmonization efforts may, however, come at a price. Local legal variation can be a product of considered policy choices, culture or innovation. 3 As a result, excessive legal harmonization can sweep away cultural identity, produce political resistance and stifle efficient experimentation and evolution 4 of the law.
1. PHILIP R. WOOD, MAPS OF WORLD FINANCIAL LAW (Sweet & Maxwell 2008); PHILIP R. WOOD, COMPARATIVE LAW OF SECURITY INTERESTS AND TITLE FINANCE (Sweet & Maxwell 2007). 2. Jay Lawrence Westbrook, A Global Solution to MultinationalDelult, 98 MICH. L. REV. 2276, 2277 (2000) [hereinafter Westbrook, GlobalSolution]. 3. The interaction between formal legal systems and legal and non-legal culture can be quite complex in bankruptcy. Teresa A. Sullivan, Elizabeth Warren & Jay Lawrence
Westbrook, The Persistence qf Local Legal Culture: Twenty Years of Evidence .rom the Federal Bankruptcy Courts, 17 HARV. J.L. & PUB. POL'Y 801, 804 (1994). Different nations take different views of the purposes of the bankruptcy system. Some focus on liquidation, while others focus on rescue. Rizwaan Jameel Mokal & John Armour, The New UK
Corporate Rescue Procedure-TheAdministrator's Duty to Act Rationally, 1 INT'L CORP. RESCUE 136 (2004) (describing the development of "rescue" culture in the United Kingdom); see also ROBERT C. ELLICKSON, ORDER WITHOUT LAW: How NEIGHBORS SETTLE
DISPUTES (1991); Lisa Bernstein, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. LEGAL STUD. 115 (1992); Eric A. Feldman, The
Tuna Court: Law and Norms in the World's Premier Fish Market, 94 CAL. L. REV. 313 (2006). 4.
New State Ice Co. v. Liebmann, 285 U.S. 262, 310-11 (1932).
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
Over the last two decades, multilateral organizations such as the World Bank, the UN, the EU and the IMF have initiated projects that seek to harmonize procedures for cross-border bankruptcies and standardize the national laws that govern bankruptcy and secured credit. These related initiatives have produced a number of successful products in the last decade: The UNCITRAL Model Law on Cross Border Insolvencies (Model Law) 5 has now been enacted in the United States, as Chapter 15 of the U.S. Bankruptcy Code (Chapter 15),6 as well as in fourteen other countries; 7 the EC Regulation on Insolvency (EC Regulation) has been in force for several years; 8 the American Law Institute's Principles of Cooperation Among the NAFTA Countries (NAFTA Principles) articulates guidelines that have helped regularize cross-border practice among Canada, Mexico and the U.S. 9, and, most recently, UNCITRAL has promulgated its Legislative Guide on Insolvency Law (Legislative Guide),10 and is in the final stages of promulgating a similar legislative guide for the law of secured credit. 1 The nature of harmonization efforts, however, raises the concern that the proposed uniform rules may not be optimal, or even an improvement over the existing non-uniformity. Harmonization ef-
5. U.N. COMM'N ON INT'L TRADE LAW [UNCITRAL], MODEL LAW ON CROSSBORDER INSOLVENCY WITH GUIDE TO ENACTMENT, U.N. Sales No. E.99.V.3 (1997), available at http:/ www.uncitral.org/pdf/english/texts/insolven/insolvency-e.pdf [hereinafter UNCITRAL Model Law]; see also G.A. Res. 52/158, U.N. Doc. A/RES/52/158 (Jan. 30, 1998). 6.
11 U.S.C. § 1501 (2000).
7. The countries other than the United States which have adopted the Model Law are: Australia (2008), British Virgin Islands, overseas territory of the United Kingdom of Great Britain and Northern Ireland (2005), Colombia (2006), Eritrea (1998), Great Britain (2006), Japan (2000), Mexico (2000), Montenegro (2002), New Zealand (2006), Poland (2003), Republic of Korea (2006), Romania (2003), Serbia (2004) and South Africa (2000). UNCITRAL, Status of the 1997 Model Law on Cross-Border Insolvency,
visited May 5, 2010). 8. Council Regulation 1346/2000, 2000 O.J. (L 160) 1 (EC), available at http:/eurlex.europa.eu/LexUriServ/LexUriServ.do'?uri OJ:L:2000:160:0001:0018:en:PDF) [hereinafter EC Regulation]. 9.
AMERICAN LAW INSTITUTE,
PRINCIPLES OF COOPERATION AMONG THE
COUNTRIES (Juris Publishing 2003). 10. UNCITRAL Legislative Guide on Insolvency Law (2004), available at http:/ www.uncitral.org/uncitral/en/uncitral-texts/insolvency/2004Guide.html. 11. See UNCITRAL (Draft) Legislative Guide on Secured Credit, available at: http:// www.uncitral.org/pdf/english/texts/security-Ig/e/final-final.clean.0 1-07-09.pdf.
forts are prone to certain predictable types of political process failure, including capture and jurisdictional competition. 12 As such, the desirability and appropriate scope of such efforts turns on the nature of the interests seeking harmonization, the robustness of the institution coordinating the harmonization effort, and whether the rule subject to harmonization is of the sort likely to generate beneficial or pernicious jurisdictional competition. 13 In a recent article, I made a normative argument for limiting the harmonization of bankruptcy law to matters of procedure and, to a limited extent, choice-of-law.14 I call this vision of limited harmonization "universal proceduralism." Under universal proceduralism, the limited goal of harmonization would be to facilitate a single, coordinated, cross-border case while minimizing both the pressure to harmonize the substantive aspects of bankruptcy law and the stakes associated with forum choice. 15 I will rehearse those arguments here in somewhat abbreviated form. The focus of this article, however, is on fleshing out (though still in broad strokes) the choice-of-law principles that should govern such a regime. In this article I seek to develop a set of choice-of-law principles that would simultaneously (1) facilitate the administration of a global bankruptcy case as envisioned by the UNCITRAL Model Law, and (2) facilitate the acceptance of rescue-based, "modem," insolvency regimes, along lines envisioned by the UNCITRAL Legislative Guide. 16
12. Edward J. Janger, Predicting When the Unifrrm Law Process Will Fail: Article 9, Capture, and the Race to the Bottom, 83 IOWA L. REV. 569 (1998) [hereinafter Janger, Predicting]; Alan Schwartz & Robert Scott, The PoliticalEconomy (?'PrivateLegislatures, 143 U. PA. L. REV. 595 (1995); Paul B. Stephan, The Futility of Unification and Harmonizationin InternationalCommercialLaw, 39 VA. J. INT'L L. 743 (1999). 13. 1 explore the dynamics of legislative capture and jurisdictional competition in greater detail in a recent article, Edward J. Janger, Universal Proceduralism,32 BROOK. J. INT'L L. 819, 830 (2007) [hereinafter Janger, Proceduralism]. In particular, I distinguish beneficial jurisdictional competition, otherwise known as a "race to the top," from pernicious jurisdictional competition, otherwise known as "race to the bottom." See also Janger, Predicting,supra note 12, at 579. 14.
Janger, Proceduralism,supra note 13, at 831.
Id. at 834.
16. Later in this essay I will be quite critical of the choice-of-law provisions contained in the UNCITRAL Legislative Guide. See in/ra notes 25-72 and accompanying text. This should not be construed to reflect disagreement with other portions of the Legislative Guide. Indeed, the Legislative Guide showed remarkable sensitivity to the need for individual nations to adopt insolvency laws that are consistent with local legal systems and that reflect local policy preferences. Indeed, it is my hope that the choice-of-law framework articulated
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
Universal proceduralism is situated between the two main competing approaches to administering cross-border insolvency cases-modified universalism and cooperative territoriality. Modified universalism, advocated by Jay Westbrook among others, seeks, to the extent possible, to create one universal bankruptcy case, administered under the bankruptcy law of the jurisdiction where the debtor has its center of main interests (COMI). Courts of other "ancillary" jurisdictions participate to the limited extent of helping to gather as17 sets and to implement the decisions of the "main" forum. Cooperative territorialism, advocated principally by Lynn LoPucki, envisions multiple full scale bankruptcies administering local assets and local claims according to local law, but with the multi-
here will facilitate the modernization of insolvency laws along the lines envisioned in the Legislative Guide. 17. Jay Westbrook has written extensively on modified universalism. See Harold Burman & Jay Lawrence Westbrook, Introductory Note: United Nations Commission on International Trade Lawt: Model Law on Cross-BorderInsolvency, 36 I.L.M. 1386 (1997); Donald T. Trautman, Emmanuel Gaillard & Jay Lawrence Westbrook, Four Models br International Bankruptcy, 41 AM. J. COMP. L. 573 (1993); Jay Lawrence Westbrook, Avoidance of Pre-Bankruptcy Transactions in MultinationalBankruptcy Cases, 42 TEX. INT'L L.J. 899, 901 (2007); Jay Lawrence Westbrook, Chapter 15 and Discharge, 13 AM. BANKR. INST. L. REV. 503 (2005); Jay Lawrence Westbrook, Chapter 15 at Last, 79 AM. BANKR. L.J. 713 (2005); Jay Lawrence Westbrook, Universalism and Choice of Lalv, 23 PENN ST. INT'L L. REV. 625 (2005); Jay Lawrence Westbrook, International Judicial Negotiation, 38 TEX. INT'L L.J. 567 (2003); Jay Lawrence Westbrook, Multinational Enterprises in General Deliult: Chapter 15, the ALl Principles, and the EU Insolvency Regulation, 76 AM. BANKR. L.J. 1 (2002); Jay Lawrence Westbrook, Setting Global Standards Ibr Cross-Border Insolvency, 21 INT'L FIN. L. REV. 9 (2002); Jay Lawrence Westbrook, The TransnationalInsolvency Project of the American Law Institute, 17 CONN. J. INT'L L. 99 (2001); Jay Lawrence Westbrook, A Global Solution to Multinational Deflault, 98 MICH. L. REV. 2276 (2000); Jay Lawrence Westbrook, The Globalization qfInsolvency Reform, 1999 N.Z. L. REV. 401 (1999); Jay Lawrence Westbrook, Modeling International Bankruptcy, 1998-1999 ANN, SuRv. BANKR. L. 465; Jay Lawrence Westbrook, Universal Priorities,33 TEX. INT'L L.J. 27 (1998); Jay Lawrence Westbrook & Jacob S. Ziegel, The American Lav Institute NAFTA Insolvency Project, 23 Brook. J. INT'L L. 7 (1997); Jay Lawrence Westbrook, The Commission's Recommendations Concerning the Treatment of Bankruptcy Contracts, 5 AM. BANKR. INST. L. REV. 463 (1997); Jay Lawrence Westbrook, Creating International Insolvency Law, 70 AM. BANKR. L.J. 563 (1996); Jay Lawrence Westbrook, The Lessons ofMaxwell Communication, 64 FORDHAM L. REV. 2531 (1996); Jay Lawrence Westbrook, Developments in TransnationalBankruptcy, 39 ST. Louis U. L.J. 745 (1995); Jay Lawrence Westbrook, Comparative Insolvency Law, 28 TEx. INT'L L.J. 675 (1993) (reviewing HARRY RAJAK, A PRACTITIONER'S GUIDE To EUROPEAN CORPORATE INSOLVENCY LAW (1992)).
pie cases coordinated through communication among administrators. 18
The modified universalist vision animates the various multilateral initiatives, including the Model Law and the EC Regulation. Key to that vision is the idea of one "main" case that coordinates the global efforts of the debtor to reorganize. 19 For this regime to work, courts in multiple jurisdictions must recognize and extend comity (or defer) to orders issued by the court in the main jurisdiction. Courts should recognize the representatives of foreign bankruptcy cases, and respect the orders of the case pending at the debtor's COMI. The idea is to create a "universal" or "global" case that will allow for the orderly treatment of the debtor's assets and operations in all of the relevant jurisdictions. 20 The signature theme of modified universal21 ism is centralizing control of the bankruptcy in the "main" case. This centralization carries with it two consequences. First, changing the location of a debtor's main case may change the recoveries obtained by creditors, even those that never dealt with the debtor in its "home" country. Second, these outcome differences create incentives for parties, and indeed nations, to act opportunistically, respectively, by forum shopping and competing for debtors. This opportunity for gamesmanship created by centralization is the Catch-22 that leads LoPucki to reject modified universalism. 22 The modified universalists offer a number of responses: COMI choice is not entire-
18. LYNN M. LoPucKi, COURTING FAILURE (2005) [hereinafter LOPUCKI, COURTING FAILURE]; Lynn M. LoPucki, Cooperation in InternationalBankruptcy: A Post-Universalist Approach, 84 CORNELL L. REV. 696 (1999); Lynn M. LoPucki, Global and Out of Control?, 79 AM. BANKR. L.J. 79 (2005) [hereinafter LoPucki, Global]; Lynn M. LoPucki, Universalism Unravels, 79 AM. BANKR. L.J. 143 (2005); Lynn M. LoPucki, The Case Ibr Cooperative Territorialityin InternationalBankruptcy, 98 MICH. L. REv. 2216 (2000). 19. Samuel L. Bufford, Global Venue Controls Are Coming: A Reply to Professor LoPucki, 79 AM. BANKR. L.J. 105, 109 (2005) ("Universalism has developed a specialized
terminology. The dominant case in the 'home country' is called the 'main' case or proceeding. A case in any other country is called a 'secondary' or 'non-main' case or proceeding." (citations omitted)). 20. The emphasis is, however, an orderly treatment. The UNCITRAL Model Law and Chapter 15 of the U.S. Bankruptcy Code recognize that some cases may best be administered as dual (or multiple) plenary cases. Sections 1525 28 of the Bankruptcy Code specifically contemplate such cases and seek to facilitate coordination among courts.
U.S.C. §§ 1525-28. 21. Jay Lawrence Westbrook, Locating the Eye of the Financial Storm, 32 BROOK. J. INT'L L. 1019, 1022 (2007) [hereinafter Westbrook, FinancialStorm]. 22.
LoPucki, Universalism Unravels, supra note 18, at 143.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
ly manipulable; 23 courts can refuse to cooperate if fundamental local policies are implicated 24 and, most importantly, harmonization of national insolvency laws may ultimately eliminate these outcome differences. 25 In my view, these responses are each insufficient. While the concept of COMI constrains a debtor's forum choice, this has not prevented COMI from becoming a hotly contested issue in many cross-border cases. 26 The ability of courts to push back or refuse cooperation is not a solution, but simply a restatement of the problem. And, finally, for reasons I will discuss below, harmonization solves the forum shopping problem, but deprives nations of the power to implement their own policies about how various creditor constituen27 cies should be treated when a business fails. Instead of siding with LoPucki, and throwing in the towel on universalism, however, I seek to articulate a set of choice-of-law principles to help neutralize these linked concerns about jurisdictional competition and forum shopping. The principles articulated here suggest that there is a pragmatic and perhaps normative quid pro quo for realizing the modified universalist aspiration: In return for recognition and cooperation, the court handling the case pending at the debtor's COMI should treat assets and claims in a manner that, as closely as possible, matches the treatment that would have been accorded had the case been handled territorially (at least as a distributional floor). Instead of one case under one law or many cases under many laws, this essay advocates one case under many lawsgoverned under a principle of "best interests," or Pareto optimality. The choice-of-law principle that accompanies universal proceduralism can be described as "virtual territoriality." Virtual territoriality, like universalism and territoriality, is a metaphor rather than a legal rule, but it suggests a design principle for administering crossborder cases. Under virtual territoriality, in a cross-border case, a jurisdiction's bankruptcy procedure should, insofar as possible, be distinguished from that jurisdiction's law of substantive legal entitlements. The procedural bankruptcy laws of the "home" country should govern the case, but even in a case where all assets are admin-
23. Westbrook, Global Solution, supra note 2, at 2317; see also Westbrook, Financial Storm, supra note 21, at 1039. 24.
Westbrook, FinancialStorm, supra note 21, at 1032.
Westbrook, Global Solution, supra note 2, at 2318.
See infi-a notes 33 35 and accompanying text.
See infi-a notes 67 70 and accompanying text.
istered centrally, the choice of substantive law should be determined by ordinary (non-bankruptcy) choice-of-law principles. This distinction between bankruptcy substance and bankruptcy procedure is not, of course, tidy, either as a practical or a conceptual matter. Bankruptcy law, by its very nature, affects (and usually limits) the remedies available to creditors. 28 It therefore alters the practical value of substantive entitlements. Particularly in jurisdictions that seek to facilitate rescue (reorganization), the ability of a secured creditor to foreclose may be limited, and shareholders and creditors may be forced to share in the decision of how to dispose of a debtor's assets. 29 This does not mean, however, that one cannot, or should not, seek to distinguish substance from procedure. The essential question is functional rather than abstract: Whether (1) bankruptcy policy, or the exigencies of a bankruptcy case, demand that the law of the debtor's center of main interest should govern; or (2) reorganization efforts can accommodate a regime under which nonuniform choice-of-law principles may govern. The first section of this article will describe the choice-of-law regimes articulated in the EC Regulation and the UNCITRAL Legislative Guide, and show how both (with some important variations) are motivated by the modified universalist vision. It will then show how this has caused COMI choice to become a major fighting issue in cross-border cases. Second, the article will explain the reasons why a universal proceduralist approach to cross-border insolvency is preferable to either a universal or a territorial regime. Third, it will explore which bankruptcy rules must be governed by the law at the debtor's COMI for such a case to function, and which rules can be allowed to vary depending on national choices and private international law. Fourth, and finally, it will give two examples of how principles of virtual territoriality have already influenced the crossborder cases of Collins and Aikman 30 and FederalMogul.3 1 These cases offer excellent examples of how virtual territoriality might function in practice, and also illustrate a number of its positive attributes.
See, e.g., 11 U.S.C. §§ 524, 1141.
29. See, e.g., 11 U.S.C. §§ 362, 1129; see also Insolvency Act, 1986, c. 1, §§ 1 3 (U.K.), available at http://www.insolvency.gov.uk/insolvencyprofessionandlegislation/ legislation/uklinsolvencyact.pdf 30.
Re Collins & Aikman Europe,  EWHC (Ch) 1343.
In re Federal-Mogul Global, 330 B.R. 133 (2005).
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
CHOICE-OF-LAW UNDER THE EC REGULATION AND IN THE UNCITRAL LEGISLATIVE GUIDE
At the moment, determining a debtor's COMI is the most hotly litigated issue in international insolvency. In the European Union, no issue under the EC Regulation has been litigated as vehemently. In Daisytek,32 England and France wrestled for control of the debtor's case largely because a nervous director could not figure out with certainty whether a filing was necessary in France, and the French court misunderstood the British methods for publishing judicial opinions. In the Eurofoods33 case, an Irish registered financial subsidiary of Parmalat, an Italian food producer, in a case being administered out of Italy, was held to have its COMI in Ireland. 34 Similarly, under newly enacted Chapter 15 of the U.S. Bankruptcy Code, there are already three significant decisions on how to determine a debtor's COMI, including two conflicting cases in the Southern District of New York. 35 This should not be a surprise. While the Model Law does not address choice-of-law issues per se, and under Chapter 15, the consequences of COMI choice are as of yet untested, under the EC Regulation, U.S. law, and the regime envisioned by the Legislative Guide, the consequences are significant, and apparently worth fighting over. COMI choice matters. In this section, I will describe the choice-of-law rules articulated in the EC Regulation and the Legislative Guide. I will argue that both the EC Regulation and the Legislative Guide share the aspirations and language of modified universalism, and as a result, place much power in the hands of the forum court. This centralization increases the importance of COMI choice. In the rest of this essay, I
32. Re Daisytek-ISA Ltd.,  B.C.C. 562 (Ch); Cour d'appel [CA] [regional court of appeal] Versailles, Sept. 4, 2003,  B.C.C. 984 (Fr.) (Klempka v. ISA Daisytek SA); see also Gabriel Moss, Group Insolvency Choice of Forum and Lawt: The European Experience Under the Influence of English Pragmatism,32 BROOK. J. INT'L L. 1005, 101011 (2007). 33.
Re Eurofood IFSC, Ltd.,  B.C.C. 999, 1003 (S.C.) (Jr.).
34. Id.; see also Samuel L. Bufford, Center olMain Interests, InternationalInsolvency Case Venue, and Equality of Arms: The Eurolbod Decision of the European Court of Justice, 27 Nw. J. INT'L L. & Bus. 351, 374-75 (2007). 35. In re Bear Stearns High-Grade Structured Credit Strategies Master Fund, Ltd., 374 B.R. 122, 129-30 (Bankr. S.D.N.Y. 2007); In re SPhinX, 351 B.R. 103, 119-20 (Bankr. S.D.N.Y. 2006), aqf'd, 2007 WL 1965597 (S.D.N.Y. July 3, 2007); In re Tri-Continental Exchange Ltd., 349 B.R. 627, 639 (Bankr. E.D. Cal. 2006).
will question the wisdom of this approach, and suggest that a more modest, and preferable, approach is emerging on the ground. A.
Choice-of-Law and Choice-of-Forum
In a cross-border insolvency case, choice-of-law consists of two nested inquiries: First, "which jurisdiction will host the 'main proceeding?"' and second, "which questions of law will be determined conclusively by the law of the forum where the main case is pending, and which will be determined by ordinary choice-of-law 36
principles?" Most controversy has focused on the first of these inquirieshow to identify the "main proceeding." Under both the Model Law and the EC Regulation, the "main" proceeding is the case opened at a debtor's center of main interest or "COMI. ''37 Both statutes also agree that the center of main interest is presumed to be the jurisdiction where the debtor is incorporated. 38 Here, however, is where the agreement ends. There are a number of pressure points. First, how 39 strong is the presumption in favor of jurisdiction of incorporation? While it is agreed that the presumption can be rebutted by showing that the main executive functions, assets and/or operations are located elsewhere, how much proof is enough is unclear. 40 Second, conceptually, does a debtor only have one COMI or might a debtor have
36. Under Chapter 15, the first inquiry is mandated by 11 U.S.C. § 1517(b). The second is not discussed at all, and is presumably left to the choice-of-law rules of the forum court. By contrast, Article 4 of the EC Regulation spells out in detail when forum law should govern, and when private international law will determine the governing law. See notes 59-60 in/ra. 37.
EC Regulation, art. 3; 11 U.S.C. § 1517(b).
EC Regulation, art. 3; 11 U.S.C. § 1516(c).
39. Compare Re Eurofood IFSC, Ltd.,  B.C.C. 999, 1003 (S.C.) (Ir.) (applying a strong presumption in favor of jurisdiction of incorporation), with In re Bear Stearns HighGrade Structured Credit Strategies Master Fund, Ltd., 374 B.R. 122 (Bankr. S.D.N.Y. 2007) (refusing to recognize Cayman Island case as the "main" case for a Cayman Island registered corporation), and SPhinX, 351 B.R. at 117 (bankruptcy court has discretion in determining COMI). 40. Moss, supra note 33, at 1006; Council Regulation 1346/2000, Recital (13), 2000 O.J. (L 160) 1 (EC), available at http://aei.pitt.edu/952/01/insolvency-report-schmidt_ 1988.pdf.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
multiple COMIs? 41 Third, might a court have discretion in determining a debtor's COMI, as the Southern District of New York determined in the SPhinX case, 42 or is COMI choice non-discretionary, as another court in the Southern District decided in the Bear Stearns case?
This article focuses instead on which choice-of-law questions should be determined by forum choice, and which should be determined by ordinary choice-of-law principles. It seeks to delink choice-of-law from forum choice, and in so doing to reduce the stakes associated with determining the debtor's COMI. This, in turn, will reduce the incentives for parties to forum shop or to litigate forum choice. Choice-of-law is often conflated with choice-of-forum. This conflation is compounded by the fact that bankruptcy law is generally thought to be procedural rather than substantive. A forum is generally entitled to apply its own procedural rules, but will generally use choice-of-law principles (sometimes called "private international law") to determine the governing substantive rule of law. In the United States this is a rhetorical point and a normative bone of contention. Law and economics scholars such as Thomas Jackson, Douglas Baird and, more recently, Chuck Mooney argue that bankruptcy law is, and should be, merely a procedure for giving effect to nonbankruptcy entitlements in cases where a business is unable to pay all of its creditors. 44 Outside the U.S., this may be more accurately characterized as a descriptive point; in the absence of procedures for business rescue, insolvency law is little more than a procedure for winding up a failed business. To the extent that bankruptcy is merely procedure, choice-of-forum determines choice of procedure, and the inquiry is at an end. However, modem bankruptcy laws-in the United States and elsewhere-are not purely procedural. Many provisions of national 41.
In a world of multiple COMIs, a debtor may open its insolvency proceeding in any
one of a number ofjurisdictions; courts of other jurisdictions will be required to cooperate. See Westbrook, FinancialStorm, supra note 23. 42. SPhinX, 351 B.R. 103; In re Tri-Continental Exchange Ltd., 349 B.R. 627 (Bankr. E.D. Cal. 2006). 43. In re Bear Stearns High-Grade Structured Credit Strategies Master Fund, Ltd., 374 B.R. 122 (Bankr. S.D.N.Y. 2007). 44.
THOMAS H. JACKSON, THE LOGIC AND LIMITS OF BANKRUPTCY LAW (2001); James
Baird, A World Without Bankruptcy, 50 LAW & CONTEMP. PROBS. 173 (Spring 1987); Charles W. Mooney, Jr., A Normative Theory (#'BankruptcyLaw: Bankruptcy As (1s) Civil Procedure, 61 WASH. & LEE L. REV. 931 (2004).
insolvency laws are quite substantive. The rights of secured parties are often subject to modification in bankruptcy. While the monetary value of a secured creditor's claim is protected, 45 foreclosure is often delayed, 46 liens are frequently stripped down to the value of the collateral, 47 and cash collateral is used by the debtor to fund operations, 48 much to the frustration of secured creditors. Similarly, special governance rules are created by bankruptcy law, depriving shareholders of control over incumbent management, and forcing them to share control with a complex network of interested parties 49 under the supervision of a bankruptcy judge. Virtual territoriality seeks to take the substance/procedure distinction seriously, and, in so doing, to drain the heat from the debate over COMI choice. If forum choice does not guarantee that all of the provisions of the main jurisdiction's insolvency law will govern the case, then the importance of COMI choice recedes. In short, bankruptcy law might be disaggregated: While procedural rules should be conclusively determined by the law of the forum, governing law for substantive questions should, by contrast, be determined by reference to principles of private international law. This sets up a difficult task-determining the questions that will be governed by the law of the forum, sometimes referred to as lexforum concursus, and those that might potentially be governed by the law of another jurisdiction under the forum's non-bankruptcy choice-of-law rules, sometimes referred to as lex situs. 50 To frame the question more functionally, it is obviously easiest to administer a cross-border bankruptcy case under one unified law. However, if there are important reasons not to do this, then one must ask when uniformity is essential to an orderly bankruptcy procedure, and when non-uniformity can be tolerated. Which rules must be centralized (through forum choice), and when can the process tolerate a decentralized approach to governing law (through the use of private international law)? When should the forum state's interest in enforcing its bankruptcy policies serve as a trump over ordinary (though perhaps unpredictable) choice-of-law principles? Before offering my
11 U.S.C. §§ 361, 506.
Id. §§ 362, 363.
Id. §§ 506(d), 1125.
Id. § 363.
Id. §§ 363, 1107.
Legislative Guide, supra note 10, at 5.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
own suggestions, it seems worthwhile to survey the existing approaches. B.
Choice-of-Lawfor the Modified Universalist
For the modified universalist, centralization of the bankruptcy case is the goal, and to the extent possible, forum choice should carry with it the full panoply of the forum court's bankruptcy rules. In a recent article, 51 Jay Westbrook articulates the view that modified universalist choice-of-law rules should place significant power in the hands of the forum court. As he puts it: The close integration among bankruptcy rules and policies in each jurisdiction applies to the big four of bankruptcy policy: control, priority, avoidance, and reorganization policy. In a system of universalism each of these four elements should be governed by the law of the main proceeding. Under modified univer-
salism, such centralization should be the goal, although not always the result. 52 Under Westbrook's view, "centralization" is "the goal," and COMI choice has significant consequences. It determines for whose benefit the estate is administered, the effect of having a perfected security interest 53 and the priority of claims and preference and fraudulent conveyance law. These are among the most controversial elements of a bankruptcy case, and national laws differ considerably. For example, in England, a case was, historically, administered by an administrator for the benefit of the secured creditors. 54 By contrast, in the United States, the estate is administered by a trustee or incumbent manage51.
Westbrook, FinancialStorm, supra note 23.
Id. at 1021-22 (emphasis added).
In using the word "priority" Westbrook could mean a number of different things.
He could be referring only to the ranking of various types of unsecured claims. He could
also be referring to the rights of the secured party against the bankruptcy estate, and finally he could also include the mechanism for establishing priority among various secured parties. Based on other writings, in particular his criticism of the decision in In re Treco, 240 F.3d 148 (2d Cir. 2001), 1 will infer that he includes the first two, but not the third. Westbrook, FinancialStorm, supra note 23, at 1024. 54.
VANESSA FINCH, CORPORATE INSOLVENCY LAW: PERSPECTIVES AND PRINCIPLES
(Cambridge 2002). The United Kingdom's Enterprise Act of 2002 has, more recently, limited the power of secured creditors in bankruptcy.
ment for the benefit of the creditors. 55 Similarly, among the most common reasons that creditors will fight cooperation of a local proceeding with a main proceeding is the fear of different priority treatment for secured claims. In one case, for example, a U.S. Court declined to send assets to the Bahamas because a secured creditor's claims were subordinate to the (in that case overwhelming) costs of administration of the estate. 56 Avoidance has been a common subject of litigation as well. Indeed, it was the key issue in the Maxwell case, where the U.S. and U.K. courts agreed to apply U.K. avoidance rules 57 rather than U.S. rules.
Modified Universalism in Practice-TheEC Regulation
Only one multinational document currently in force addresses choice-of-law in cross-border bankruptcies. The EC Regulation on insolvency governs choice-of-law and choice-of-forum questions for the bankruptcies of entities that cross borders within the EU. Under the EC Regulation, choice-of-law is tightly linked to choice-offorum, and to the EC Regulation's rules on forum choice. Under Article 3(1), a company may only open a bankruptcy proceeding in the country that is its COMI. 58 Once a proceeding has been opened at a
11 U.S.C. §§ 1106, 1107.
Treco, 240 F.3d 148.
57. In re Maxwell Commc'n Corp., 170 B.R. 800, 818 (Bankr. S.D.N.Y. 1994), q/"d, 186 B.R. 807, 822-23 (S.D.N.Y. 1995), qa/'d, 93 F.3d 1036, 1051 (2d Cir. 1996). As a historical footnote here, the court appointed expert in Maxwell was Professor Westbrook. Judge Brozman, the bankruptcy judge, agreed with Westbrook that U.K. law should govern the preference claims at issue. However, Westbrook took the view that U.K. law should apply because the United Kingdom was the COMI of the corporate debtors, while Judge Brozman concluded that U.K. law should govern because the transactions themselves were "located" in the United Kingdom. In this regard, the Brozman decision could be characterized as an early example of virtual territoriality. 58.
Article 3 of the EC Regulation states in pertinent part:
1.The courts of the Member State within the territory of which the centre of a debtor's main interests is situated shall have jurisdiction to open insolvency proceedings. In the case of a company or legal person, the place of the registered office shall be presumed to be the centre of its main interests in the absence of proof to the contrary. 2. Where the centre of a debtor's main interests is situated within the territory of a Member State, the courts of another Member State shall have jurisdiction to open insolvency proceedings against that debtor only if he possesses an establishment within the territory of that other Member State. The effects of those proceedings shall be restricted to the assets of the debtor situated in the territory of the latter Member State.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
debtor's COMI, other "secondary" proceedings can be opened, but, under Article 3(3), these secondary proceedings must be "winding up" or liquidation proceedings. On its face, this structure appears to create a strong bias in favor of centralizing the debtor's case around the case pending at the debtor's COMI. The choice-of-law provisions gesture even further in the direction of centralization. Article 4 of the EC Regulation states: Save as otherwise provided in this Regulation, the law applicable to insolvency proceedings and their effects shall be that of the Member State within the territory of which such proceedings are opened, hereafter referred to as the "State of the opening of proceed59 ings." Article 4 of the EC Regulation then goes on to list the legal matters that will be determined by the law of the forum court. The list is long, and covers the full panoply of bankruptcy procedures, but also includes a long list of rules that might more accurately be characterized as substantive: (2) The law of the State of the opening of proceedings shall determine the conditions for the opening of those proceedings, their conduct and their closure. It shall determine in particular: (a) against which debtors insolvency proceedings may be brought on account of their capacity; (b) the assets which form part of the estate and the treatment of assets acquired by or devolving on the debtor after the opening of the insolvency proceedings; (c) the respective powers of the debtor and the liquidator; (d) the conditions under which set-offs may be invoked; (e) the effects of insolvency proceedings on current contracts to which the debtor is party;
3. Where insolvency proceedings have been opened under paragraph 1, any proceedings opened subsequently under paragraph 2 shall be secondary proceedings. These latter proceedings must be winding-up proceedings. 59. EC Regulation, art. 4(1).
(f) the effects of the insolvency proceedings on proceedings brought by individual creditors, with the exception of lawsuits pending; (g) the claims which are to be lodged against the debtor's estate and the treatment of claims arising after the opening of insolvency proceedings; (h) the rules governing the lodging, verification and admission of claims; (i) the rules governing the distribution of proceeds from the realisation of assets, the ranking of claims and the rights of creditors who have obtained partial satisfaction after the opening of insolvency proceedings by virtue of a right in rem or through a set-off; () the conditions for and the effects of closure of insolvency proceedings, in particular by composition; (k) creditors' rights after the closure of insolvency proceedings; (1) who is to bear the costs and expenses incurred in the insolvency proceedings; (m) the rules relating to the voidness, voidability or unenforceability of legal acts detrimental to all the creditors. 60 If one were only to read Article 4, one might mistakenly think that the "modified universalist" dream had been realized. The reality is more ambiguous, and turns on the role of "secondaries" in the structure envisioned by the Regulation. First, Article 3 of the EC Regulation is silent about what law should apply in a secondary proceeding. 61 Presumably, however, a court in a secondary would apply its own law to deal with claims of creditors that could be asserted against assets of the debtor located in the secondary jurisdiction. In other words, the case pending at the debtor's COMI is "universal" but the EC Regulation appears to anticipate that the secondaries will be "territorial." Second, the EC Regulation requires that a secondary be a liquidation or "winding up" proceeding. It is difficult to imagine a company engaging in a going concern reorganization based at the COMI, while simultaneously liquidating in the 60. Id. art. 4(2). 61.
Id. art. 3.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
secondaries. In sum, the centralizing, reorganization-facilitating "yin" of Article 4 runs headlong into the territorial, liquidationoriented "yang" of Article 3, and one is left asking whether the drafters of the Regulation were schizophrenic, unable to agree, or confused. The EC Regulation appears to construct a game of choice-oflaw chicken. Universal principles apply in the main case to all assets that can be administered without the benefit of secondaries, but territorial principles control when a secondary must be opened. How might this be operationalized? Where liquidations are involved, the process should work smoothly, with the main case handling most of the assets, but secondaries mopping up with territorial liquidations where necessary. Rescue, however, is more problematic. Opening a secondary would appear to cripple any firm's chance of reorganizing, at least if key assets are located in the secondary jurisdiction. Furthermore, Article 4 requires the main case to apply its own law to questions of priority. 62 On the one hand, this centralizing principle might make reorganization more straightforward. However, any time a creditor thinks it might do better under the law of another63jurisdiction, it has the incentive to force the opening of a secondary. D.
Modified Universalism in Practice-The UNCITRAL Legislative Guide
The Legislative Guide is less schizophrenic about its universalism, but this stems in part from the fact that it is merely a guide and not intended to be enacted as a statute. The Legislative Guide, as a tactical matter, generally takes a permissive rather than a prescriptive approach. 64 Even its prescriptions are phrased as recommendations. The recommendations relating to choice-of-law are numbers thirty through thirty-four. 65 On its face, the Legislative Guide hews 62.
Id. art. 4(2)(i).
63. As the discussion of Collins and Aikman, in/ha Part ILA, will show, Article 4 may also limit the ability of the debtor to respond to these forum shopping threats by bargaining with the uncooperative creditors. 64. Susan Block-Lieb & Terrence Halliday, Harmonization and Modernization in UNCITRAL's Legislative Guide on Insolvency Law, 42 TEX. INT'L L.J. 475, 507 (2007). 65. Recommendations 30-31 of the Legislative Guide draw the distinction between the legal questions that are to be determined by forum choice and those that are to be determined by private international law. They provide in pertinent part: 30. The law applicable to the validity and effectiveness of rights and claims
existing at the time of the commencement of insolvency proceedings should be
to the procedure/substance distinction. The commentary to those recommendations begins by drawing a distinction between the bankruptcy law and non-bankruptcy law that will be applied in an international case: In the context of cross-border insolvency, it is essential to distinguish between the creation of rights and claims under the law designated as applicable law (whether domestic or foreign substantive law) in accordance with the conflict of laws rules of the forum and the insolvency effects on those rights and 66 claims. With regard to the status and validity of non-bankruptcy claims, the Legislative Guide suggests that the forum state should apply its own private international law (choice-of-law rules) to determine which law applies. For example, for contract disputes the law of the situs of the contract will determine the rights of the parties. 67 Similarly, the law of the jurisdiction where property is located 68 will generally determine the status of claims against the property. The Legislative Guide draws a distinction, however, between the validity of the claim and the treatment it receives in bankruptcy. It is here that the Legislative Guide is at its most "universal," stating that once the validity of the claim has been determined under the relevant national law, the priority and treatment of the claim should be dealt with under the bankruptcy law of the forum (lex fori concursus): determined by the private international law rules of the State in which insolvency proceedings are commenced.... 3 1. The insolvency law of the State in which insolvency proceedings are commenced
commencement, conduct, administration and conclusion of those insolvency proceedings and their effects. These may include .... (list set forth in text in/ha Part I.B) UNCITRAL Legislative Guide, supra note 10, at 72 73. 66.
UNCITRAL Legislative Guide, supra note 10, at 68, para. 82.
67. "It is typical under general conflict of laws rules, for example, that the law governing the contract will determine if a contractual claim exists against the insolvent debtor and the amount of that claim; that the lex rei sitae will determine if a security interest in immovable assets has been created in favour of a specific creditor, and so on." Id. 68. One large exception is intangible property and mobile goods, which, at least under U.S. law are deemed located where the debtor is located. Rules for perfection of security interests in tangible property in the U.S. are also governed by the law of the jurisdiction where the debtor is located. However, the effect of perfection is governed by the law of the jurisdiction where the collateral is located. U.C.C. §§ 9-301 318.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
It is quite typical that the law of the State in which insolvency proceedings are commenced, the lex fori concursus, will govern the commencement, conduct, administration and conclusion of those proceedings. This would generally include, for example, determining the debtors that may be subject to the insolvency law; the parties that may apply for commencement of insolvency proceedings and the eligibility tests to be met; the effects of commencement, including the scope of application of a stay; the organization of the administration of the estate; the powers and functions of the participants; rules on admissibility of claims; priority and ranking of claims; and rules on distribution. 69 Recommendation 31 contains an extensive list of matters that "may" be determined by choice-of-forum: The insolvency law of the State in which insolvency proceedings are commenced (lex fori concursus)
should apply to all aspects of the commencement, conduct, administration and conclusion of those insolvency proceedings and their effects. These may include, for example: (a) Identification of the debtors that may be subject to insolvency proceedings; (b) Determination of when insolvency proceedings can be commenced and the type of proceeding that can be commenced, the party that can apply for commencement and whether the commencement criteria should differ depending upon the party applying for commencement; (c) Constitution and scope of the insolvency estate; (d) Protection and preservation of the insolvency estate; (e) Use or disposal of assets; 0 Proposal, approval, confirmation and implementation of a plan of reorganization;
UNCITRAL Legislative Guide, supra note 10, at 68.
(g) Avoidance of certain transactions that could be prejudicial to certain parties; (h) Treatment of contracts; (i) Set-off; () Treatment of secured creditors; (k) Rights and obligations of the debtor; () Duties and functions of the insolvency representative; (M) Functions of the creditors and creditor committee; (n) Treatment of claims; (o) Ranking of claims; (p) Costs and expenses relating to the insolvency proceedings; (q) Distribution of proceeds; (r) Conclusion of the proceedings; and 70 (s) Discharge. This list of matters to be determined by choice-of-forum is most decidedly not limited to the purely procedural elements of the home court's bankruptcy law. While the list is permissive, rather than prescriptive, it is quite long and, again, includes the full panoply of bankruptcy issues from scope of the stay, nature of the estate, avoidance powers, priority, treatment of secured claims, plan confirmation to scope of the bankruptcy discharge. A jurisdiction whose bankruptcy law sought to govern all of these matters would be making a strong statement in favor of centralized control over bankruptcy cases filed in that country. It would also be claiming for itself the power to allocate and reallocate rights among creditor constituencies in ways that differ from the allocations that would apply outside bankruptcy or in a purely territorial bankruptcy case. The universalism of the Legislative Guide is not unalloyed. While envisioning a maximal universal case, the comments to Recommendation 31 acknowledge that many jurisdictions might choose a narrower list, and might except: (1) settlement of payments and regulation of financial markets; (2) labor law; (3) security interests and (4) avoidance powers. 71 The first two exceptions are not particu-
Id. at 73.
Id. at 72.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
larly troubling, even for a committed universalist. It would be problematic for a foreign bankruptcy proceeding to interfere with the settlement of payment transactions, and the law governing employees. The second two would appear to be more radical departures from the modified universalist vision. By excluding security interests (and by implication property rights more generally), the Legislative Guide makes a significant concession to political reality. 72 Indeed, the inclusion of both security interests and avoidance powers on the suggested list of topics that are appropriately governed by forum choice, followed by a concession that they might be omitted, suggests that the universalist aspiration of some of the Guide's drafters may have been tempered by the territorial concerns of others. Nonetheless, the list of exceptions in the legislative guide is far shorter than the list in the EC Regulation. In short, both the Model Law and the EC Regulation stop short of full-scale realization of the modified universalist vision, but choice-of-forum matters a great deal, especially under the EC Regulation where the choice-of-law rules are mandatory. The Legislative Guide is less prescriptive, but if a large number of jurisdictions were to follow its recommendations there would be a significant substantive component to forum choice under that regime as well. If the goal of these provisions is to pave the way toward greater universalism, then one can expect courts and legislatures to take an expansive view of the effect of forum choice. I believe, however, that this decision to take an expansive view of lex forum concursus is both a tactical and a normative mistake. If this is the approach followed worldwide, one can expect to continue to see extensive litigation over the location of a debtor's COMI. Moreover, my concern is that outside the EU, the approach may overstress the fragile cross-border architecture, and perhaps endanger the widespread harmonization even of procedural rules like the Model Law. II.
UNIVERSAL PROCEDURALISM AND VIRTUAL TERRITORIALITY
While the benefits of uniformity and predictability extolled by universalists and the proponents of the Legislative Guide are laudable, they may be overstated. The recent history of uniform lawmak-
72. In doing so, as I will discuss later, the Legislative Guide leaves open the possibility for a universal proceduralist approach.
ing in the U.S. 73 shows that harmonization efforts are prone to certain predictable process failures that make them no more than second-best lawmaking institutions. Uniform and model laws are not self executing. Therefore, harmonizers who hope for widespread adoption of their product must anticipate the possibility of interest group activity in adopting legislatures, and they must anticipate jurisdictional competition. Accordingly, the scope of harmonization efforts should be limited to the provisions where uniformity is most necessary. 74 This view has profound implications for choice-of-law rules generally, and for the law of bankruptcy and secured credit in particular. It suggests the choice of "decentralizing" choice-of-law rules that do not allow repeat players to externalize the impact of their home coun75 try law. In this section, I will explain both my reservations about uniform law making generally, and the desirability of decentralizing choice-of-law rules. I will then discuss the implications of those principles for choice-of-law in cross-border insolvencies. A.
UniversalProceduralismand the Limits of Harmonization
In my view, international lawmaking efforts relating to insolvency should limit their aspirations toward harmonization to the few 73.
Janger, Predicting,supra note 12; Edward J. Janger, The Locus of Lawmaking, 74
AM. BANKR. L.J. 97 (2000). Here, I am referring to the recent efforts by the American Law Institute and the National Conference of Commissioners on Uniform State Laws to promulgate revised versions of Articles 1, 2, 3 and 4 of the Uniform Commercial Code. While Article 1 is gaining an increasing number of adoptions, a number of provisions have been uniformly rejected. Revised Article 2 has not been adopted anywhere, and appears unlikely to gain widespread acceptance. Revised Articles 3 and 4 were amended in 2002. Currently, Arkansas, Kentucky, Minnesota, New Mexico, Nevada, South Carolina and Texas have been the only states to adopt the revised article. See Amendments to Articles 3 and 4 of the UCC, available at http://iww.nccusl.org/Update/uniormactjctsheets/uniormactsIs,-ucca3.asp. These problematic legislative trajectories are mild compared to the history of the Uniform Computer Information Transactions Act or UCITA, which was initially intended to become part of the Uniform Commercial Code as Article 2B, but proved so controversial that the American Law Institute withdrew its support. UCITA has been adopted by two states, Virginia and Maryland, but many more states have adopted so-called "bomb shelter" statutes that instruct courts not to give effect to the provisions of UCITA in those states. See infia note 92 and accompanying text. 74.
Janger, Predicting,supra, note 12, at 593; Janger, Proceduralism,supra note 13, at
845. 75. Ted Janger, The Public Choice (#'Law in Sqftware Transactions: Jurisdictional Competition and the Dim Prospects of Unilbrmity, 26 BROOK. J. INT'L L. 187, 190 (2000).
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
procedural rules necessary to administer a case comprising all of the debtor's assets and operations and a set of choice-of-law principles that would limit the effect of choice-of-forum on substantive entitlements. These limits on the universalist aspiration are driven by two related concerns: (1) the effect of anticipated capture and anticipated jurisdictional competition on efforts to harmonize bankruptcy law and procedure; and (2) the effect of jurisdictional competition on courts and legislatures. The first set of concerns derives from reservations about harmonization efforts generally. Alan Schwartz and Bob Scott 76, and 1,77 have raised concerns about "private legislatures" in the domestic context, and Paul Stephan 78 has raised similar concerns in connection with lawmaking efforts by UNCITRAL and UNIDROIT in the commercial law area. One of the goals of UNCITRAL when it drafts a model law is to obtain widespread adoption. 79 As a practical matter, there are two prerequisites to such widespread adoption: (1) strong support from a group or groups; and (2) lack of significant opposition. Harmonizers as distant in time as Homer Kripke have recognized that without an interest group to spearhead adoption, uniform laws (domestic or national) are likely to fall victim to the inertia of the status quo. 80 This need for support makes uniform lawmaking processes prone to capture. And, even where law reform efforts are not spearheaded by a dominant interest group, they must respond to the threats of various groups to oppose the statute when it is promulgated. 81 The model law drafters must anticipate the possibility that individual legislatures in enacting states may be captured. The harmonizers must, therefore, act preemptively to co-opt any group capable of threatening opposition at the local level. Harmonization ef76.
Schwartz & Scott, supra note 12, at 607.
Janger, Predicting,supranote 12, at 623 24.
Stephan, supra note 12, at 756 61.
UNCITRAL Legislative Guide, supra note 11.
80. Homer Kripke, The Principles Underlying the Draftingof the Unilbrm Commercial Code, 1962 U. ILL. L.F. 321, 327 (1962). As Kripke noted, "[t]he draftsmen and the members of the sponsor organizations knew that to draft a dead-letter bill would accomplish nothing. The Code had to be enacted .. " Id. He went on: The Code would have been a sitting duck target for any determined special interest or combination of special interests who chose to attack one or more features of the bill persistently. Thus, it was important not to arouse the opposition of banks or finance companies, warehouse companies, railroads, or other private trade groups. Id. 81.
Janger, Predicting,supra note 12, at 584.
forts are also prone to the effects of pernicious jurisdictional competition. Jurisdictions compete with each other for commercial activity. Sometimes this competition is beneficial, leading to a race to the top. 82 Other times this competition leads to a race to the bottom. In either case, harmonization can be problematic. Where jurisdictional competition is likely to lead to a race to the top, harmonization may hasten the adoption of an efficient rule, but it can also short circuit or prevent the experimentation that can lead to the formulation of that rule. Where a race to the bottom is concerned, the problems are even greater. Harmonizers may be forced to anticipate the results of jurisdictional competition, and may encourage the adoption of an inefficient rule. 83 Because of these related concerns, I have argued at length elsewhere that the aspirations of harmonization ought to be 84 limited, for the most part, to procedural rules. In this regard, the UNCITRAL Model Law on Cross Border Insolvencies, now enacted in the United States as Chapter 15 of the U.S. Bankruptcy Code provides an excellent example of procedural harmonization.8 5 I have reservations, however, about ongoing efforts to harmonize the laws of secured credit and bankruptcy, at least to the extent that they seek harmonization for its own sake. 86 Instead, I advocate a uniform choice-of-law principle that will provide predictability, without mandating uniformity. In the next section, I argue that predictability can be achieved while minimizing the effects of forum shopping, by adopting, where possible, decentralizing choice-oflaw rules.
82. Lucian Bebchuk, Federalism and the Corporation: The Desirable Limits on State Competition in CorporateLaw, 105 HARV. L. REv. 1435 (1992).
Janger, Predicting,supranote 12, at 827-31.
Janger, Proceduralism,supra note 13, at 821 22.
John A. E. Pottow, Procedural Jncrementalism:
A Model fi)r International
Bankruptcy, 45 VA. J. INT'L L. 935, 992 (2005) ("[The Model Law's] gentle incrementalism regarding indirect, non-core areas of the law likely assuaged some hesitant, territorialisminclined states skeptical about universalism's benefits, and perhaps even tricked (to their paternalistic betterment) some troglodyte states prejudiced against universalism altogether."); Janger, Proceduralism,supra note 13, at 824, 832. 86. The UNCITRAL Legislative Guide on Insolvency may be immune from these reservations. That document might be better described as a template for "standardization" rather than "harmonization." The drafters of the Legislative Guide recognized that many aspects of bankruptcy law may reflect considered policy choices. In these instances transparency and predictability may be the most that can be hoped for, and may indeed be the most that is desirable.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
Centralizingand DecentralizingChoice-of-Law Rules
A frequently overlooked aspect of choice-of-law is that the choice-of-law rules can have a significant impact on forum shopping, and hence, jurisdictional competition. In a world characterized by enterprises and transactions that span multiple jurisdictions, choice-oflaw rules come in two flavors, "centralizing" and "decentralizing." A centralizing choice-of-law rule allows a transaction or enterprise to be governed by the law of one jurisdiction. A decentralizing choiceof-law rule creates the possibility that an enterprise or transaction may be governed by the law of more than one jurisdiction. To the extent that uniformity and predictability are desired, centralizing choice-of-law rules are helpful. By contrast, to the extent that forum shopping or a race to the bottom are of concern, centralizing choiceof-law rules are likely to exacerbate the problem by increasing the benefits conferred by a forum shop, and increasing the benefit that can be offered by a jurisdiction seeking to attract business (or litigation). Two examples of centralizing choice-of-law rules demonstrate their effect on jurisdictional competition: " In Marquette NationalBank of Minneapolis v. Firstof Omaha Service Corp.,87 the U.S. Supreme Court declared that in consumer credit transactions, the law governing a loan was the law of the jurisdiction where the lender was located.88 The effect of this decision is not hard to discern. All of the major credit card banks have now located themselves in jurisdictions with favorable laws on such important questions as usury and late fees. Citibank's credit card bank is located in South Dakota. MBNA and many others have incorporated their credit card banks in Delaware. * The Uniform Computer Information Transactions Act, known as "UCITA," 89 adopts a centralizing choice-of-law rule for mass market software licenses. The law of the jurisdiction of the licensor controls. This is in contrast to the rule followed in Europe for mass market software transactions, where the
439 U.S. 299 (1978).
Id. at 299-300.
89. Uniform Computer Information Transactions Act (1999), available at http://www. law.upenn.edu/blil/archives/ulc/ucita/ucita200.ht-n.
law of the jurisdiction of the purchaser/licensee controls. 90 This rule was very important to the large sellers of software, who took the view that they did not want to have to master the licensing law of multiple jurisdictions. 9 1 The rule was also extremely controversial, drawing intense criticism from, among other groups, librarians. While UCITA has only been adopted in two states, Virginia (the home state for America Online) and the neighboring state of Maryland, intriguingly, four states have adopted so-called "bomb shelter" amendments to the UCC that are intended to deny effect to UCITA 92 in those jurisdictions. Both the Marquette case and the UCITA experience illustrate a familiar public choice insight. Where a transaction type is characterized by relatively small numbers of repeat players on one side of the transaction and a large number of one-shot players on the other, the repeat players may seek a choice-of-law rule that allows them to govern their transactions with the law of their home state. This has considerable advantages in terms of efficiency and predictability for the repeat player, but it also creates considerable opportunities for small states to attract business with legal rules that advantage the repeat player at the expense of more diffuse customers in other states or nations. 93 A "decentralizing" choice-of-law rule, by contrast, limits the ability of jurisdictions to compete for business by offering advantageous legal treatment. The cost of a decentralizing choice-of-law rule is the loss of uniformity, and the added cost of doing business across multiple legal regimes. Whether one should choose a centralizing or decentralizing rule turns on two inquiries: (1) Is this a rule on which jurisdictional competition is likely to lead to a race to the top or a race to the bottom, and (2) if it is a rule where a race to the bottom is expected, then how do the costs of the inefficient rule compare to the costs of nonuniformity? The calculus is dynamic-once businesses have been attracted by a favorable legal rule, the businesses have an incentive to
90. Francois Dessemontet, The European Approach to E-Commerce and Licensing, 26 BROOK. J. INT'L L. 59, 72-73 (2000). 91. Id.; see also Ted Janger, The Public Choice of Choice of Law in So/hiare Transactions: JurisdictionalCompetition and the Dim Prospects frlr Uniformity, 26 BROOK. J. INT'LL. 187 (2000). 92. American Library Association, UCITA and Related Legislation in Your State, available at http://www.ala.org/ala/aboutala/offices/wo/woissues/copyrightb/ucita/ states.cfm. 93.
Bebchuk, supranote 82.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
ask for more, in return for a commitment to stay, and the host jurisdiction has an incentive to grant it. 94 For example, Delaware, South Dakota and other jurisdictions that initially attracted large credit card banks by lifting their usury rules have gone on to adopt asset backed securitization facilitation acts that seek to grant favorable bankruptcy treatment to financing transactions entered into by credit card banks. 95 To the extent that these legal rules are efficient, the positive effect of jurisdictional competition is magnified. To the extent that the legal rules are the product of special interest lobbying and impose costs on citizens of other jurisdictions or on disorganized groups, that effect too is magnified. C.
Secured Credit,Bankruptcy Law and the Need for a DecentralizingChoice-of-Law Rule
Thus, the key determinant in choosing between a centralizing or decentralizing choice-of-law rule is whether a particular rule or enactment is likely to generate a race to the top or a race to the bottom. 96 If a race to the top is likely, then a centralizing rule is in order. Capture is also a concern, but, at least in this context, the stakes of capture are tightly linked to the stakes of jurisdictional competition. If a race to the bottom is a concern, then the costs of pernicious competition must be weighed against the costs of non-uniformity. As Bebchuk has argued, a race to the bottom is likely where a rule allows one constituency of a firm to increase the risk faced by another constituency without having to pay, or where a rule allows the citizens of one state to externalize risk to the citizens of another state. 97 In the corporate law context, rules of the first sort are rules that interfere with transparency or that entrench incumbent management. Secured credit operates in this fashion to allow equity holders and secured lenders to join forces and place uncompensated risk on
See LAURA JOFFE NUMEROFF, IF You GIVE A MOUSE A COOKIE 2 (1985)("he's going
to ask for a glass of milk."). 95. Edward J. Janger, The Death qf Secured Lending, 25 CARDOzO L. REV. 1759, 1760-61 (2004) [hereinafter Janger, Death ofSecured Lending] (listing statutes). 96. While the focus of this article is bankruptcy law, the close interrelationship between bankruptcy law and the law of secured credit, coupled with the ongoing efforts to harmonize the law of secured credit, has caused me to discuss both areas together. 97. Lucian Arye Bebchuk, Federalism and the Corporation: The DesirableLimits on State Competition in CorporateLaw, 105 HARV. L. REV. 1435 (1992).
non-adjusting creditors such as employees and small suppliers, and on non-consensual creditors such as tort claimants. 98 In the case of bankruptcy, the dynamics are more complicated. Lynn LoPucki has argued forcefully, but controversially, that in the bankruptcy context Delaware and the Southern District of New York have competed for Chapter 11 business by offering bankruptcy rules that advantage incumbent management and senior lenders at the expense of others. 99 While I think that LoPucki's concerns about U.S. judges are overstated, I agree that forum shopping in bankruptcy does place undesirable stresses on the bankruptcy system. Particularly in the international context, there may be an incentive for local legislatures to adopt rules that are friendly to incumbent management or particular groups of powerful creditors, in order to encourage debtors to arrange their affairs so that a particular country will be an appropriate bankruptcy venue. If this is the case, it counsels strongly for a "decentralizing" choice-of-law rule for both bankruptcy law and secured credit. D.
Bankruptcy, Secured Credit and the Need to Distinguish Procedurefrom Substance
Note that this concern about jurisdictional competition applies only to rules that can be described as distributional, or substantive, and that allow for the externalization of costs. In the absence of concern that states may be externalizing costs, or that incumbent management of the debtor may be seeking to advantage or disadvantage a particular constituency, there is no reason to distrust the beneficial effects of competition. Thus, to the extent that one can identify "procedural" rules and distinguish them from rules that are "substantive" and prone to pernicious competition, then it is possible to seek the benefits of a centralizing rule without fearing its costs. This is where "virtual territoriality" comes in. To the extent that bankruptcy procedure can be distinguished from bankruptcy substance, the costs of forum shopping and pernicious jurisdictional competition can be avoided. The dangers of centralized procedure can be ameliorated by the deployment of a decentralizing choice-of-law rule for substantive questions. The key principle of virtual territoriality is that the choice
98. Bebchuk, supra note 82; Edward J. Janger, Predicting, supra note 12; Lynn M. LoPucki, The Unsecured CreditorsBargain, 80 VA. L. REV. 1887 (1994). 99.
LoPUCKI, COURTING FAILURE, supra note 18.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
between a centralizing versus a decentralizing choice-of-law rule falls along the procedure/substance axis. Making such a distinction is not always easy, but that does not mean that it can never be done. Indeed, and perhaps unintentionally, it appears that the drafters of Revised Article 9 may have stumbled first onto the principle of virtual territoriality when they drafted the choice-of-law rules for that statute. While the law of secured credit is by its very nature substantive, it has some procedural components. Article 9 establishes and effectuates the property rights of secured parties-a distributive task. 100 However, it also sets forth procedures for putting third parties on notice, and for operating filing offices and so forth.10 1 This is precisely the line that the revisers of Article 9 of the Uniform Commercial Code (UCC) followed when they centralized the procedures for publicizing personal property security interests. One of the principal changes effectuated by Revised Article 9 of the UCC was to change the choice-of-law rule that governed tangible collateral. Under Former Article 9, a secured creditor perfected its security interest in tangible collateral by filing a financing statement in the jurisdiction where the collateral was located (a decentralizing rule). 10 2 For multi-state enterprises this created a need for multiple filings in multiple jurisdictions. The effect was to increase the cost of completing a transaction, and, worse, it created multiple opportunities for error. 10 3 Revised Article 9 changed that rule, and now provides that to perfect a security interest in tangible collateral, the secured party must file a financing statement in the jurisdiction where the debtor is located. 10 4 Moreover, for a corporate debtor this was deemed to be the jurisdiction of incorporation. 10 5 Procedurally, one filing in Delaware replaced many filings throughout the country. Substantively, however, an opportunity for abuse was created. If this choice-of-law rule applied to all aspects of a secured credit transaction, it would turn Article 9 into a lever for jurisdictional competition. One juris-
U.C.C. § 9-322 (2000).
For example, the rules about where to file a financial statement merely establish
predictable procedure. Id. § 9-501. 102.
Former U.C.C. §§ 9-103.
103. Lynn M. LoPucki, Why the Debtor's State o!Incorporation Should Be the Proper PlacefirArticle 9 Filing: A Systems Analysis, 79 MINN. L. REV. 577 (1995). 104.
U.C.C. § 9-301.
Id. § 9-307.
diction might compete for corporate charters by enacting a particularly favorable rule for secured parties. The creation of a uniform filing rule might lead to the creation of non-uniform priority rules. Whether intentionally or unintentionally, the drafters of Revised Article 9 effectuated a rather ingenious splitting of the choiceof-law rules between the law governing perfection of a security interest, and, for tangible collateral, the law governing the effect of perfection. 10 6 Choice-of-law for procedure was divorced from choiceof-law for substance. The filing procedure could then be safely centralized. Unfortunately, by comparison, the drafters of Revised Article 9 had it easy. For bankruptcy law, the line between substance and procedure is not as simple to draw. Because the substance of bankruptcy law can lead to forum shopping, a decentralizing choice-oflaw rule is in order. However, to maximize the value of a multinational enterprise in general default, a unified coordinated procedure is also necessary. 10 7 While bankruptcy law is largely procedural, providing a procedure for winding up a business, it also contains some major substantive rules limiting the rights of secured parties upon default, establishing priority among unsecured creditors and other things. Bankruptcy rules must therefore be divided into three categories: procedural rules that do not have distributive consequences, distributive rules that are not essential to effectuating efficient going concern reorganizations and, finally, rules that have some distributive consequences but are essential to effectuating efficient going concern reorganizations. On its face, this tripartite categorization of bankruptcy rules would appear to doom an effort to construct a workable disaggregation of bankruptcy rules along the substance/procedure line. If there are distributive rules that are essential to effectuating going concern reorganizations, then a certain amount of pernicious forum shopping is going to be inevitable. However, at least as a conceptual matter, there is a way around this problem. A unified, universal bankruptcy
106. Id. § 9-301. This same bifurcation was not possible for "intangible" collateral such as accounts and general intangibles. Not surprisingly, Delaware has enacted at least one non-uniform statute that, to facilitate securitization, potentially alters the perfection rules for sales of promissory notes and payment intangibles. Janger, Death o!SecuredLending,supra note 95, at 1760-61. 107. Procedural centralization is not, in and of itself, troubling. Procedure is less likely to be subject to pernicious jurisdictional competition. Instead, jurisdictions that compete on the basis of procedure will seek to move faster, coordinate better and administer assets more efficiently.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
procedure only really matters where value can be created by centralization. It is a common exercise in U.S. business reorganizations to compare the value of a company's assets when liquidated to the value of the company if reorganized as a going concern.10 8 It is this "going concern surplus" that justifies the limitations that Chapter 11 places on the non-bankruptcy rights of creditors. Under U.S. law, for example, a plan of reorganization cannot be confirmed unless the debtor establishes that the distribution to creditors under the plan will be more valuable to the creditors than their share of the liquidated com109 pany. It is this same "best interests" or "Pareto" principle coupled with the existence of a going concern surplus that saves virtual territoriality. In a cross-border case, the debtor should not be able to reorganize unless the creditors in each country receive at least what they would have received had the entity been liquidated in a territorial case. E.
The universal proceduralist model for cross-border bankruptcy cases seeks to capture the benefits of coordination in a single main case, without triggering the dynamics of jurisdictional competition. The mechanism combines (1) a mandatory rule for choice-of-forum; (2) a centralizing choice-of-law rule for procedures that allow for the coordination of a bankruptcy case across borders and (3) a decentralizing choice-of-law rule for distributive/substantive rules of law that affect creditor entitlements (up to the territorial liquidation threshold). The centralizing procedure generates the benefits associated with a single coordinated case. Forum shopping is neutralized, however, through a mandatory rule for forum choice (such as COMI) and a predictable decentralizing choice-of-law rule. I refer to this scheme as "virtual territoriality" because it will frequently mean that one court will have to administer a case that is governed under the law of several jurisdictions at once. To understand it, it is necessary to describe it first for a single firm, and then for a corporate group.
11 U.S.C. § 1129(a)(7) (2005).
Where a single firm has assets and operations in multiple jurisdictions, one must ask which questions are governed by the law of the bankruptcy forum state, and which are governed by the law of the jurisdiction where the assets or the creditors are located. Imagine a company with its center of main interest in country A, but with significant tangible assets located in jurisdiction B (encumbered by liens held by creditors also located in jurisdiction B), and significant contracts entered into with counterparties in jurisdiction C (who received significant payments shortly before the bankruptcy). The creditors in jurisdiction B and the counterparties in jurisdiction C all knew that they were dealing with an entity located (with a COMI) in jurisdiction A. Further imagine that the debtor seeks to reorganize by filing a bankruptcy case in jurisdiction A (its home court). While the courts of A may provide a procedure for reorganizing, and reorganization may be in the best interests of all of A's creditors, if the debtor wishes to gather the assets in jurisdiction B it will need to have the orders of the proceeding in jurisdiction A recognized by the courts of B. Similarly, if the debtor wishes to bind the parties to contracts in jurisdiction C to the orders issued by the court in jurisdiction A, or recover preferential payments made to creditors in jurisdiction C, it will have to do so with the cooperation of the courts of jurisdiction C. A modified universalist would take the view that all of the debtor's assets should be administered in the courts of jurisdiction A. The validity of the property and contract claims would be determined using the law of the jurisdiction dictated by private international law, but the treatment of those rights in bankruptcy would again be determined under the bankruptcy law of the jurisdiction A, the forum state. A territorialist, by contrast, would take the view that assets should be administered wherever located. The validity of claims and the disposition of assets should be determined under the law of the jurisdiction where the assets are located, and private international law should determine which jurisdiction's laws should determine the validity of claims. To put it another way, the contract claims and property rights are to be administered in the courts of the situs jurisdiction under the law of the situs jurisdiction. The universal proceduralist would administer both the assets and the contracts in the court of the debtor's COMI, but would apply the principles of virtual territoriality to deal with both validity of contract claims and property rights, and their treatment in bankruptcy.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
Control over the case and decisionmaking would be governed by forum law, but issues such as priority and avoidance would be determined under the law of the situs (of the assets or the claim). The courts in jurisdiction A would seek to apply, to the extent possible, the same law that a court in B would apply with regard to assets located in jurisdiction B, and the law that C would apply to contract disputes and avoidance actions for transactions "sited" in jurisdiction C. In all likelihood, the law of the jurisdiction where the assets are located will govern the tangible property, and the law of the jurisdiction that is the situs of the contract will govern the contractual rights. 2.
Where corporate groups are involved, virtual territoriality could work the same way. The cases could be administratively consolidated in the jurisdiction that was the COMI for the group. However, the COMI court would administer the cases of the various subsidiaries under the law of the COMI for the subsidiary. So, if the COMI of the debtor was jurisdiction A, but it had subsidiaries located in jurisdictions B and C, the court in A would be charged with, as much as possible, administering the cases of the subsidiaries as if they had been filed in the COMI of the subsidiary. This is merely an extension of the principles articulated for a single firm. The main difference is that control rights with regard to the assets might be different across national boundaries. As I have noted elsewhere, this may seem awkward on its face, but even under current law, courts are not unfamiliar with simultaneously administering multiple cases for members of a corporate group whose cases have been administratively consolidated. Furthermore, where it is efficient to do so, the court in the debtor's COMI would still be free to allow the debtor to open up a secondary case in a secondary jurisdiction to allow it to administer the assets of the subsidiary separately.110 Whether or not to do so would be determined according to the exigencies of the case, as determined in the home court, rather than through far flung fights over recognition in multiple jurisdictions.
110. Unlike under the EC Regulation, however, the "secondary" proceeding would not have to be a "winding up" or "liquidation."
DisaggregatingSubstance and Procedure
Westbrook takes the position that "control, priority, avoidance and reorganization policy" should be determined by the law of the jurisdiction where the "main proceeding" is pending.1 11 He acknowledges that "lex fori" or the law of the forum will determine much of what goes on in a bankruptcy case. Private international law, or ordinary choice-of-law principles, will be used to determine the status of a claim, but virtually everything about how a claim will 1 be treated will be determined by the law of the chosen forum. 12 Virtual territoriality envisions a much more limited role for the law of the forum, and a much more substantial role for principles of private international law. Virtually the only thing that should be determined by the law of the forum is control. Where a single entity is involved, control will likely also carry with it governance, but that is coincidental, as the Model Law determines that the main case must be the entity's COMI, and under non-bankruptcy law, the real seat of the corporation would usually determine the governance rules that apply to the assets. Similarly, with regard to priority, the question should be answered not by reference to the choice-of-forum but by reference to ordinary principles of choice-of-law. If the situs of a contract is located in a particular country, then the priority that would be accorded in that country ought to be granted to the creditor. Avoidance can similarly be handled by using the law of the jurisdiction most tightly related to the transaction. As a first cut, the line might be drawn as follows. Essential procedures, to be determined by law of forum might include scope of the estate, scope of the automatic stay and scope of discharge (subject to best interests). Substantive provisions, where applicable law would be determined by private international law, might include claims allowance, claim priority, avoidance and plan (a single entity will use the plan procedures of the COMI, but a group will have to handle national subsidiaries according to principles available under law of the COMI of the subsidiary). Thus, where a cross-border case is dealing with a single entity, the situations where modified universalism and universal proceduralism will diverge will be rather small, except in the cases where (as in consumer credit or mass market sales) the debtor is doing business in many jurisdictions with individuals who do not perceive themselves to be engaged in cross-border
Westbrook, FinancialStorm, supra note 22, at 1021 22.
Id. at 1022.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
transactions (buying software or widgets at the shopping mall). However, where corporate groups are involved the divergences between the two approaches may become more significant. A main court may find itself administering a subsidiary under the law of an entirely different jurisdiction. This may seem improbable, but it is not much more difficult than what happens when U.S. bankruptcy courts administer cases that have been administratively but not substantively consolidated. When the EC Regulation and the provisions of the Legislative Guide are evaluated in light of the virtual territorialist approach, they give far too great a role to lex concursus.113 The Legislative Guide allows forum choice to determine governance and the priority of claims, and excludes only avoidance actions from the powers of the home court. The EC Regulation goes even further. The treatment of secured creditors and the avoidance actions that are available are determined by the law of the jurisdiction where the case is opened. If the goal of bankruptcy choice-of-law rules should be to distinguish substantive from procedural rules, the EC Regulation and the Legislative Guide fail. 11.
VIRTUAL TERRITORIALITY IN ACTION
While neither the EC Regulation nor the Legislative Guide follow either a universal proceduralist or virtual territorialist approach on paper, this does not mean that virtual territoriality does not have some descriptive power when one observes how recent crossborder cases have actually been conducted. In this section, I look at two cross-border bankruptcies, the Collins and Aikman 114 bankruptcy and the FederalMogul 115 bankruptcy, to show that, even now, the principles of virtual territoriality can be seen operating in U.S. and EU cases. A.
Collins and Aikman- Virtual TerritorialityOperationalized
Collins and Aikman provides an example of how virtual territoriality can operate, even under the EC Regulation, at least in the
See supra note 65 and accompanying text.
Collins & Aikman Europe SA, Re,  EWHC (Ch) 1343 (Eng.).
In re Federal-Mogul Global, Inc., 330 B.R. 133 (D. Del. 2005).
hands of a pragmatically oriented common law judge.1 16 Collins and Aikman was a manufacturer of automobile parts. When it filed for bankruptcy, it opened up plenary cases in the U.S., Canada and the U.K. to deal separately with its operations in North America and the 7 EU.
The European case provides an example of how a "virtual territorial" approach can pave the way toward a successful reorganization.1 18 The goal of the cases opened in the U.K. was to find a buyer for the company's European operations. Once a buyer was found, there was concern that opening secondary cases would delay and perhaps block the sale. However, German and Spanish creditors took the view that application of local priorities would have yielded a better distribution than that available under British law.' 19 The administrators of the U.K. case did not want to complicate the case by opening secondary cases in Germany or Spain, so they promised to grant the equivalent of local treatment to the Spanish and German trade creditors.120 However, they were then faced with the problem of figuring out how to deliver on that promise within the strictures of the EC Regulation. 121 As noted above, Article 4 of the EC Regulation
Collins & Aikman,  EWHC 1343.
Id. para. 8. The court noted:
The joint administrators were of the view that the opening of such secondary proceedings and the appointment of local officeholders would have been likely to have impeded the achievement of the purposes of the administration by
making it difficult to continue to trade the businesses, fund the administrations and conduct sales processes on a group-wide basis. To avoid such secondary proceedings oral assurances were given by or on behalf of the joint administrators to creditors at creditors' meetings and creditors' committees' meetings that if there
were no secondary proceedings in the relevant
jurisdiction then their respective financial positions as creditors under the relevant local law would as far as possible be respected in the English administration. Id. 121.
Id. para. 10. The court noted:
By April 2006, then, the joint administrators had a problem before them. They had given assurances which they wished to honour, assurances given not only with a view to the benefit of creditors generally but assurances which had conduced to achievement of that benefit. Those assurances had included performance by the joint administrators of differing provisions, country by country, as to local law as to, for example, the preferences to be given to
particular classes of creditors and the subordination or not of inter-company indebtedness, provisions which were different from the applicable provisions of English law, the law of the main proceedings.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
mandates the application of U.K. priorities to the U.K. case. Therefore, the administrator was forced to engage in some creative lawyering to convince the U.K. judges to allow the debtor to honor the deal. 122 The administrator appealed to the British court, and was able to convince the judge (stripped to the essentials) that under U.K. law, where officers of the court make promises that are in the best interest of the estate but deviate from the U.K. priority scheme, and where the honor of the administrator can be upheld only by enforcing those promises, the court may allow the administrator to honor those promises.
Thus, in Collins and Aikman, one company was able to implement the principle of virtual territoriality, but doing so was in tension with the structure set forth in the EC Regulation and required some sophisticated common law lawyering to achieve. As Gabriel Moss, one of the lawyers in Collins and Aikman, noted: If sufficient flexibility can be found in other European laws where main proceedings are opened, Collins & Aikman will be an obvious model for the way to harmonize the need for centralization and simplicity, on the one hand, and the respecting of local priorities, on 124 the other. Under current law, however, such flexibility is by no means guaranteed. One might well imagine a civil law judge or a judge in a jurisdiction with territorial bankruptcy law refusing to implement the structure envisioned by the debtor in Collins and Aikman. They might instead have required the opening of secondaries. The resulting delay and winding up required by the Regulation might then have precluded a going concern sale of the debtor's assets.
Moss, supra note 32, at 1017 18.
123. Collins & Aikman,  EWHC 1343, para. 15. The court cited the rule of Ex parte James, In re Condon, (1874) 9 Ch App. 609 [1874-80], and described it as follows: "'This elusive and difficult principle is based on morality. At the centre of the principle is that if an officer of the court is under an obligation of conscience, then the court will direct the officer to fulfill that obligation."' Collins & Aikman,  EWHC 1343, para. 15 (quoting MCPHERSON, THE LAW OF COMPANY LIQUIDATION (4th ed. 1999)). 124.
Moss, supra note 33, at 1018.
FederalMogul- Virtual TerritorialityIgnored
By contrast, in the case of FederalMogul, 125 a failure to follow the virtual territorial approach nearly derailed the reorganization. When Federal Mogul, another U.S. automobile parts supplier, filed for bankruptcy, its goal was to resolve its outstanding asbestos claims. Cases were opened in the U.S. and the U.K., and early in the case a basic settlement was reached under which the asbestos claimants against the U.S. and the U.K. entities would receive the same proportional payout. Later in the case, however, it became apparent to the parties that they had missed a key issue. The capital structures of the U.S. and U.K. companies were substantially different. In the U.S., virtually all of the company's assets were liened by secured creditors. As a result, the payout to unsecured creditors, including the asbestos claimants, would have been relatively small. By contrast, the capital structure of the U.K. subsidiaries was relatively free of secured debt. As a result, the unsecured claimants against those entities would have received a considerably more generous distribution than their U.S. counterparts. This led the initial settlement to crumble. The case was only put back on track when the settlement was adjusted to reflect the different territorial payouts to the claimants against the various entities. Both Collins and Aikman 126 and Federal Mogul 127 illustrate the importance of respecting the territorial entitlements of creditors, even in a universal bankruptcy case. Collins and Aikman further demonstrates the value of being able to achieve one's territoriality "virtually" through choice-of-law. To the extent that a universal main case can administer assets in a manner that mirrors the treatment that would be received in a purely domestic case, the less resistance local creditors will have to the results in the main case. Furthermore, where secondary or ancillary cases must nonetheless be opened, the ancillary courts will have fewer qualms about cooperating, and fewer domestic pressures not to cooperate. At the same time, by enhancing the scope of the "main" case, the virtues of coordination can be better obtained. In other words, it is possible that a
125. In re Federal-Mogul Global, Inc., 330 B.R. 133 (2005). An excellent retelling of the history of the Federal Mogul case, by one of the participants, can be found in the proceedings of the 2007 Meeting of the National Conference of Bankruptcy Judges. Mark Andrews, The FederalMogul Restructuring,Program Materials, 81 st Annual Meeting of the National Conference of Bankruptcy Judges 14-5 (2007) (on file with author). 126. Collins & Aikman,  EWHC 1343. 127.
Federal-Mogul,330 B.R. 133.
COLUMBIA JOURNAL OF TRANSNATIONAL LAW
decentralizing choice-of-law rule for "substantive" questions will enhance the ability of the main case to centralize and coordinate the "procedural" aspects of the case. CONCLUSION
In sum, under U.S. bankruptcy law, a prerequisite to the confirmation of a plan of reorganization is the requirement that the plan must satisfy the so-called "best interests" test. Creditors must receive at least as much in the reorganization as they would have received in liquidation. 12 8 Since there is no such thing as "global" bankruptcy law, there is no such requirement in cross-border cases. However, both Collins and Aikman and FederalMogul illustrate that Pareto optimality operates as a practical constraint on a cross-border case as well. The entitlements that a creditor would have received under local law, in the absence of a going concern reorganization, provide a practical floor to distributions in cross-border cases. Because territorial liquidation is the default result in the absence of a successful consensual reorganization, local creditors can, as a practical matter, always oppose local recognition of orders issued in the main jurisdiction, whenever the distribution in the cross-border case is less than the alternative "territorial" distribution. Virtual territoriality embraces this principle respecting territorial entitlements by seeking to limit the scope of lex concursus and by seeking to ensure that the distribution to creditors in a cross-border case mirrors or exceeds the distribution they would receive in a territorial case. Virtual territoriality also has implications for efforts to harmonize bankruptcy procedure and to standardize bankruptcy law. Specifically, the broad scope given to lex forum concursus in the proposed choice-of-law provisions in the Legislative Guide may be a tactical error, with consequences not just for the Guide but also for the future adoption of the Model Law. The reality of international insolvency reform is that many countries view both the Model Law and the Legislative Guide as instruments of U.S. (and to an increasing extent U.K.) bankruptcy imperialism. By enhancing the central force of the "main" case, the Model Law, it is feared, will subject 129 creditors to the long reach of U.S. and U.K. bankruptcy courts. This fear is enhanced when the choice of the main forum carries with it choice of rules with distributive content. In other words, adoption 128.
11 U.S.C. § 1129(a).
LoPucki, Global,supra note 18.
of the approach to choice-of-law suggested in the Legislative Guide may actually endanger widespread adoption of the Model Law, and may harm the chances for evolution of global norms for coordinating bankruptcy cases. By contrast, the approach advocated here should considerably reduce the stakes associated with COMI choice and the costs of deference to the main case. The more the court in the main case focuses solely on coordination and value maximization, and does little to alter territorial distributional priorities, the more likely it is that the benefits of coordination will be achieved.