A Commitment Rule for Insolvency Forum

I. Introduction

The Model Law on Cross-Border Insolvency (Model Law) was promulgated by the United Nations Commission on International Trade Law (UNCITRAL) in 1997. During its now more-than-a quarter century life, it has been adopted in more than 60 jurisdictions around the world and has played a major role in the improvement and successful management of insolvency proceedings with a cross-border element. Therefore, the Model Law is an exceptional achievement that the international insolvency community should celebrate. Indeed, it is not an exaggeration to say that the Model Law helped to foster today’s vibrant international insolvency community.

The Model Law is built on the idea of “modified universalism”. As such, it envisions the commencement of a main proceeding in a single jurisdiction even if non-main proceedings can also be opened and the laws of other jurisdictions can still be relevant for certain aspects of the proceeding. Once main and non-main proceedings are opened, the Model Law establishes a set of rules to facilitate cooperation and assistance for the successful management of the proceedings.

The adoption of modified universalism as a regulatory model to deal with cross-border insolvency is a sensible one. Indeed, in contrast to those favoring the adoption of a fragmented (or “territorialist”) approach, the existence of a centralized proceeding is a superior option.2 The type of cooperation and assistance facilitated by the Model Law significantly improves the efficiency and effectiveness of insolvency proceedings in cases where the debtor has assets, creditors, and operations in various jurisdictions.

Still, there is room for improvement. Specifically, the Model Law’s use of the debtor’s center of main interest (the COMI rule) to determine the proper forum for a foreign main proceeding3 is flawed. The COMI rule, as it currently exists under the Model Law, creates uncertainty and litigation costs and opens the door to opportunistic behavior by debtors. In a very real sense, the COMI rule presents the worst of all worlds by allowing debtors to manipulate forum at the time of filing while preventing them from committing to an efficient forum ahead of time, and then opening the door to expensive litigation regardless of where they do file.

This state of affairs undermines the legitimacy and efficacy of insolvency proceedings and prevents market negotiations that produce value for creditors and contribute to the effective reorganization of viable businesses. By stifling markets, the COMI rule hinders entrepreneurship, access to finance, and economic growth.

This Article proposes a better approach—one we call the “Commitment Rule”—for the choice of insolvency forum. In short, the Commitment Rule allows debtors to establish an advance commitment to a particularinsolvency forum. To make this commitment public and binding, the debtor must put it in their company’s constitution. This upfront and observable commitment eliminates uncertainty and opportunistic manipulation. It also allows companies to choose a more efficient insolvency forum that can benefit debtors, creditors, and society at large. Hence, the Commitment Rule can be particularly beneficial for jurisdictions without developed restructuring ecosystems, which often include emerging markets and developing economies, and therefore countries where the adoption of active policies to foster growth is most needed.

Moreover, the adoption of the Commitment Rule provides potential additional protection to vulnerable creditors and, in any event, leaves them no worse off than they are under the Model Law. The Commitment Rule thus presents a rare “win-win” legal reform requiring no major tradeoffs. It would reduce destructive forum shopping and minimize litigation costs while preserving beneficial former choice to promote the development and selection of efficient insolvency forums, which benefit all stakeholders. These improvements support the development of financial markets, entrepreneurial innovation, and economic growth more generally. Therefore, we urge UNCITRAL to consider adopting the Commitment Rule as part of the Model Law.

From here, we proceed as follows. In Section 2, we examine the harmful effects produced by the COMI rule. In Section 3, we present the Commitment Rule. In doing so, we also discuss implementation and present second-best alternatives. We also demonstrate that the Commitment Rule embodies the core pillars of modified universalism. We seek to improve modified universalism, not abolish it. In Section 4, we note the relationship between the Commitment Rule and the rules for determining the substantive law applicable to insolvency proceeding—which UNCITRAL is currently discussing. We also note that, in a true system of modified universalism, protectionist rules like the Gibbs rule should be abolished, but sensible rules like an international Butner principle—which preserves non-bankruptcy substantive rights and protects local tort victims and employees—should be preserved and reaffirmed. Section 5 concludes.

II. Harmful Economic Effects Generated by the COMI Rule

Despite its generally positive impact, the Model Law errs in its method for determining which country hosts the main insolvency proceeding. Under the Model Law, a proceeding qualifies as a foreign main proceeding if it takes place in the jurisdiction where the debtor has its center of main interests (COMI), which is generally the place of the debtor’s registered office unless it is shown that the central administration of the debtor is in a different location that is ascertainable by third parties.4 In our view, the use of COMI as the selection mechanism to determine the initiation of a foreign main proceeding presents various flaws that undermine the ability of insolvency law to facilitate the maximization of the returns to creditors, the effective reorganization of viable businesses, and the promotion of entrepreneurship, access to finance, and economic growth.5

The COMI rule encourages debtors to initiate insolvency proceedings in the place they can prove their COMI is located. Otherwise, even if a jurisdiction permits foreign debtors to initiate insolvency proceedings, as indeed various countries do,6 the debtor faces the risk that the proceeding or some aspects triggered by the proceeding, such as a moratorium or a discharge or modification of the terms of a debt, will not be recognized by other jurisdictions.7 But the debtor’s COMI may lie in a jurisdiction with an inefficient insolvency system.8 Or perhaps other jurisdictions will have special characteristics that make them more attractive to the debtor. These characteristics might be related to the legal environment, the financial markets—especially the markets for rescue financing—or other institutional foundations that might affect a firm dealing with financial distress.

By encouraging a debtor to ignore such things when it initiates insolvency proceedings, the COMI rule penalizes efficiency and subsidizes proceedings in a forum that may destroy value for creditors. This is costly to a debtor both at the time it initiates proceedings (because it destroys value) and at the time of borrowing (because lenders will price their loans on the basis that the debtor may initiate the proceeding in a forum with an inefficient insolvency system). The rule is, thus, antithetical to the core purpose of insolvency law. It fails to maximize the returns to creditors and imperils the effective reorganization of viable but financially distressed businesses.

The foreseeable costs of the COMI rule lead to an ex ante increase in the cost of credit, ultimately reducing firms’ access to finance and thereby stifling economic growth. There is a deeply unfortunate irony in all of this: The countries with inefficient insolvency frameworks often include emerging markets and developing economies. Therefore, the COMI rule would be more harmful in countries where adopting policies to reduce poverty and promoting growth is most needed.9

There is another troubling aspect of COMI. Ascertaining a company’s COMI is far from clear. This is especially true today as the world has become increasingly global, internationally connected, and technology driven. Indeed, many companies currently have assets, creditors, subsidiaries, offices, employees, and clients in many jurisdictions. In this context, determining the debtor’s COMI is not an easy task. This problem is particularly acute with certain new economy businesses, such as tech firms, crypto exchanges, and decentralized finance applications. With these businesses, it may be unclear whether a company even has a COMI.10 But even traditional companies today have complicated structures with multiple affiliates that make COMI hard to determine.11 And as global mobility increases, a company’s COMI may shift over time. Or a company may engage in a “COMI shift” to intentionally open up a new option for insolvency forum.

In such a world, market participants—such as lenders—can never be entirely sure about where a company’s future insolvency case will be administered. But these lenders are not helpless and unsophisticated. They will make the debtor pay for the uncertainty it faces. Lenders will rationally price their loans to account for all possible scenarios. That is, they will charge the debtor a fee merely because they know the debtor might later choose to initiate insolvency proceedings in an inefficient insolvency forum. For every option the debtor might have to choose an inefficient or creditor-unfriendly forum, there will be an undesirable increase in the cost of credit. Even worse, some market participants might just leave the market altogether because they cannot adequately price in the uncertainty about COMI. And the debtors, on the other side, will be helpless because—even if they never intend to choose an inefficient or creditor-unfriendly forum—they currently have no legal means by which to commit to good behavior. They cannot credibly signal their bona fides.

Thus, the current COMI rule discourages transactions that could potentially create jobs, wealth, and growth. The uncertainty it creates will increase the initial cost of borrowing, which in turn hampers economic growth. This is a cost that society in general bears because it will lead to an overall increase in the cost of credit for solvent firms. Put differently, the current COMI rule hinders access to finance and destroys jobs and wealth even if a situation of insolvency never arises.

To make things worse, countries around the world adopt different approaches when they implement the COMI rule. For example, one major question is timing.12 When is a COMI established for the purposes of determining the proper insolvency forum? Here the world is messy. While the United States and Singapore determine the COMI based on the date the application for recognition is filed, the United Kingdom uses the date of the filing of the foreign proceedings, and Australia uses the date of the hearing of the recognition application.13 Additionally, given the lack of a “universal court of appeal” the COMI rule is often interpreted differently, putting a greater or lower emphasis on the presumption of the registered office and the different factors that may rebut such presumption.14 Sorting out these complexities and ascertaining the “true” COMI inevitably results in litigation costs that destroy value at the expense of debtors, creditors, and society as a whole.15 Spending money on fighting over COMI creates no positive value. It is deadweight loss.

Finally, the current approach to determine how a main proceeding is initiated allows for opportunistic behavior by debtors. Indeed, the COMI rule imposes no restrictions on shifting COMI from one place to another, even on the eve of commencing a case. This ex post flexibility—which gives the debtor almost unlimited choice at the time of filing16 —is achieved at the expense of promoting predictability and creditor protection. Of course, this risk of opportunistic behavior can be priced by sophisticated lenders in the form of higher interest rates, requiring more collateral, reducing credit or, in the extreme, not extending credit at all. In other words, sophisticated lenders can protect themselves. The real costs of permitting those opportunistic changes of COMI are thus borne by borrowing companies and ultimately society at large. As noted above, with the current COMI rule, debtors cannot commit to choosing an efficient forum, and so these costs are unavoidable.

The situation is even worse for vulnerable creditors such as employees and tort claimants. Given the inability of these creditors to adjust the conditions of their claims, debtors can opportunistically move their COMI to a jurisdiction that would benefit the debtors or their managers—and sometimes their sophisticated lenders—at the expense of vulnerable creditors. If the sophisticated creditors could demand that debtors commit to an efficient forum, vulnerable creditors would often be protected as well. But with that commitment off the table, the sophisticated lenders simply charge more money and the vulnerable creditors are afforded no protection. As a result, the COMI rule fails to protect creditors, and particularly those who, in our view, deserve the most protection.17

III. Alternative Approaches to Determine the Insolvency Forum

As a result of the factors mentioned in Section 2, the Model Law’s COMI rule should be revisited.18 In our view, the COMI rule should be abolished and replaced by the approach suggested in Section 3.1. Alternatively, UNCITRAL should consider the adoption of the second-best solution proposed in Section 3.2. As summarized in the table included in Section 3.3, both approaches, and especially the preferred approach, provide a more desirable outcome than the current system embraced by UNCITRAL.

A. Preferred Approach: The Commitment Rule

i. Introduction

As one of us suggested in the 1990s, an alternative approach to determine the place where an insolvency proceeding will be initiated is to allow a debtor to commit to a chosen insolvency forum in its company constitution.19 This approach presents several advantages over the COMI rule. First, it provides more predictability about, and a binding commitment to, the place where an insolvency proceeding will take place. Since the debtor’s commitment to a certain insolvency forum would be identified in the company’s constitution, every market participant could know where a future insolvency proceeding would be initiated. No more having to guess where a company’s COMI might be. Second, this solution would reduce litigation costs once a debtor initiates an insolvency proceeding. No more wasting resources tussling over where a debtor’s COMI is. Finally, the choice of insolvency forum in the company’s constitution would allow debtors and creditors to have access to more attractive insolvency frameworks without having to worry about whether the procedure, or some aspects of the procedure, would be recognized by other jurisdictions. By allowing the company to commit to initiate any future insolvency proceeding in jurisdictions that can provide a more efficient insolvency regime, this approach would encourage lenders to extend credit at a lower cost. This lower cost would facilitate firms’ access to finance and the promotion of economic growth. Now the debtor can commit to good behavior to access lower-cost capital. It would also contribute to the maximization of the returns to creditors and the effective reorganization of viable but financially distressed businesses.

ii. Responding to concerns about the ex ante choice of insolvency forum

Some may contend that replacing the COMI rule with the Commitment Rule can lead to opportunistic behaviour by debtors. For example, the debtor could, so the objection goes, commit to a forum that puts its interests above that of its creditors. Or the debtor could, having initially selected one regime in its constitution, amend its constitution to select a different, more debtor-friendly, forum after borrowing funds. In both situations, by choosing a more debtor-friendly and less creditor-friendly regime, a debtor can benefit itself at the expense of the creditors.20 On the other extreme is the criticism that the Commitment Rule may hamper the change of insolvency forum even if debtors and creditors realize that a jurisdiction not initially chosen by the debtor can serve as a more attractive insolvency forum.21 Put differently, it could be argued that the “flexibility” to change the COMI (sometimes opportunistically, sometimes in the interest of debtors and creditors) currently allowed by the Model Law would be lost if the Commitment Rule is adopted. None of these arguments are persuasive if, as we urge UNCITRAL, the Commitment Rule is adopted with the safeguards and conditions suggested below.

First, and most importantly, debtors will not have incentives to choose an insolvency forum that is not attractive for sophisticated lenders. The key feature of the Commitment Rule is that the public commitment creates market constraints on debtor misbehavior. A debtor who chooses a creditor-unfriendly forum will be exposed to an increase in the cost of debt. It may even lose access to credit markets altogether. Therefore, the real risk of opportunistic behavior when initially choosing the insolvency forum only exists in the context of vulnerable creditors, such as tort claimants and employees, that do not have the ability, information, or bargaining power to adjust the conditions of their claims. We discuss the mitigation of this risk (which is no less present with COMI) below.

Second, the risk to vulnerable creditors is low and can be addressed through several mechanisms. For example, if countries seriously want to protect these creditors, a defined group of vulnerable creditors, such as tort claimants and employees, can be given a preferential treatment in the ranking of claims,22 and the lack of respect of this priority may serve as a cause for denying recognition, even on the basis of “public policy,” of any insolvency proceeding initiated by the debtor in a foreign jurisdiction.23 To be clear, there is nothing new about this concept. And the Model Law would not need to prescribe any rules on priorities. This is something entirely decided by a jurisdiction, especially given the sensitive nature of this topic and the policy considerations involved. Moreover, countries could internally interpret a violation of those priorities as a matter of public policy. In fact, this is likely the case in many jurisdictions already. Imagine, for example, a Chapter 15 case in the United States seeking recognition of a foreign proceeding that subordinate United States tort victims to a position below all other unsecured creditors. It is almost certain that a United States court would deny recognition on public policy grounds. Similarly, a United States proceeding that does not respect the priorities given to employees under French law is not likely to be recognized by a French court.

Still, UNCITRAL could clarify the point. It could provide that, for the purpose of recognition, lack of respect of the priority provided to a defined group of “vulnerable creditors,” whose identification would be exclusively decided by each country,24 would lead to the lack of recognition of the proceeding. Alternatively, countries can specify that a defined group of vulnerable creditors cannot be worse off, in terms of expected returns, compared to what they would receive if the procedure had been initiated in the debtor’s local jurisdiction. This latter approach would be more similar to the type of “value-enhancing forum shopping” often observed in the international insolvency practice. Again, these actions would not change the current law. They would simply add clarifications to what is already contained in the “public policy” exception to recognition.25

Third, some might argue that the Commitment Rule will force local creditors to bear the costs associated with initiating an insolvency proceeding in a foreign jurisdiction. This is a general criticism of universalism as a regulatory model to deal with cross-border insolvency.26 It is not specific to the Commitment Rule. Indeed, the exact criticism could be leveled on the COMI rule (if anything, with COMI it is a worse problem given the potential costs of litigation over whether the selected jurisdiction is in fact the debtor’s COMI). The worry that large companies would bring insolvency proceedings in a foreign jurisdiction cannot be a reason to maintain the status quo. Consider the long list of companies that have filed Chapter 11 proceedings despite unquestionably having a COMI elsewhere. And there are many other instances where companies have changed their COMI on the eve of initiating procedures. And still others file non-main proceedings in foreign jurisdictions. The Commitment Rule does not create the ability to forum shop. Forum shopping already exists. The Commitment Rule just brings more certainty and market discipline to the forum choice.

Fourth, as mentioned above, a critique of the Commitment Rule might be that the commitment could be changed. A company’s constitution can generally be changed by the shareholders and not by the creditors. The debtor might then opportunistically change the insolvency forum once it has obtained credit. It is unclear how much of a problem this really would be, at least in the context of sophisticated lenders. As a starting point, this strategy would only make sense for a debtor who never planned to return to the credit markets. Most debtors are repeat borrowers who pay a large cost for making such changes. More fundamentally, a lender can easily put in its lending agreement that any change in the location of a future insolvency without its consent would be an immediate event of default for the loan.

In any case, to address the problem of change, UNCITRAL might adopt a variety of rules. For instance, if UNCITRAL wants to favor predictability over flexibility, the Commitment Rule might include a requirement that debtors changing the forum in their constitution should provide notice to all the pre-existing creditors.27 Then, if no creditor (or if fewer than a certain percentage of creditors) objects within a reasonable period of time (perhaps three to four weeks), the forum change would then become effective.28 By adopting this approach, none of the company’s pre-existing creditors would be required to accept an insolvency forum that was not accepted at the moment of extending credit. As a result, this would be the most creditor-protective approach as well as the less flexible one.29

A more flexible approach could require approval of a majority or super-majority of the creditors.30 While this approach may avoid some holdout problems inherent in the previous approach, it might create other costs. Obtaining consent from the majority or super-majority of creditors can be costly. Additionally, some creditors may price their loans on the basis that they might eventually be required to be subject to an unwanted insolvency forum. Therefore, this approach can encourage sophisticated lenders to increase the cost of debt.

A third approach may consist of allowing the debtor to adopt specific amendment rules in its constitution when making the initial forum commitment.31 For example, these conditions might include: (i) requiring consent by, or not having the veto of, a minimum number or percentage of creditors; (ii) putting a delay on the effective date of a change in the constitution; (iii) allowing the change of the insolvency forum provided that it is approved by a particular individual (for example, an arbitrator, advisor, or designated independent director) or group of people (e.g., board of directors, company’s independent directors, external committee of legal and financial advisors). The debtors will write their amendment rules considering the demands and expectations of the credit markets. Again, sophisticated lenders take the adopted rules into account when they price their loans. Debtors, knowing this, will have incentives to choose a method that is trustworthy, value-enhancing, and protective of the interests of the creditors. As discussed above, vulnerable creditors will be protected through other—and more effective—mechanisms.

iii. Other practical considerations for the effective implementation of the Commitment Rule

The implementation of the Commitment Rule for the ex ante choice of insolvency forum requires certain changes by UNCITRAL and national legislators. First, the Model Law should include a default rule that would apply if, by any chance, companies have not chosen an insolvency forum in the constitution. Such a default rule may consist of the place of the registered office, which is the presumption of COMI established under the Model Law, and has been interpreted as a strong presumption in places such as the European Union.32 The choice of the registered office as the default rule can lead to more predictability and less litigation costs than other alternative options such as the debtor’s “center of main interests”.33 In any case, while such a rule would be necessary in the transition period to our preferred approach, it is likely that the choice of insolvency forum would become standard fare in all constitutions (at least for large companies with sophisticated creditors).

Second, the protections against the opportunistic change of insolvency forum mentioned in Section 3.2 will require some changes to the Companies Act (or equivalent statute)34 of the country adopting the approach. Namely, countries adopting the proposed approach should make sure that any amendment to the constitution not only complies with the current rules governing changes to the constitution (e.g., approval of a prerequisite majority of shareholders) but also with the rules potentially implemented to protect creditors if the amendment of the constitution affects the provision providing the insolvency forum.

Third, the Model Law should specify how to deal with any controversies potentially arising if the directors violate the commitment provision and initiate an insolvency proceeding in a forum not designated in the company’s constitution. One solution would be to require judicial review on whether the company is “eligible” to initiate an insolvency proceeding in their jurisdiction. If so, in addition to any eligibility requirement potentially imposed by a jurisdiction (e.g., proof of a “substantial connection”, as some jurisdictions require), courts should verify that the place of filing is indeed the insolvency forum chosen by the company in its constitution.35 This review could be conducted on the court’s own initiative at the beginning of the proceedings, especially in countries with efficient judicial systems. Alternatively, it could be conducted after initiation upon a motion or petition of party in interest, such as a creditor, shareholder, employee, or by a foreign representative of the jurisdiction designated in the company’s constitution as the insolvency forum if, for example, the creditors have initiated an involuntary insolvency petition in that jurisdiction.36 Either way, the cost of such a review would be trivial—the court simply has to look at the company’s constitution. Additionally, the Model Law should specify that, if a company initiates a proceeding in a forum contrary to a commitment in the company’s constitution, and that proceeding is not dismissed, the proceeding will not be recognized under the Model Law.

Finally, we note that our proposal to replace or at least revisit the COMI rule focuses on corporations, where the COMI rule does more harm than good and creates uncertainty due to the increasingly global and complex nature of many businesses. Nonetheless, if UNCITRAL abolishes the COMI rule, it would need to come up with a solution to deal with individuals. The analysis for individuals is entirely different. Corporations and individuals enter insolvency proceedings under distinct circumstances, and policy justifications and the goals of insolvency law are quite different for corporations and individuals. For instance, while many insolvency laws seek to provide a discharge of debts to honest but unfortunate individual debtors, this “fresh start policy” does not exist in many jurisdictions around the world—particularly in many emerging economies.37 The goal of corporate insolvency law, on the other hand, is more consistent across jurisdictions. Even though certain insolvency laws might be more attractive to debtors or creditors, most insolvency regimes seek to minimize the destruction of value, maximize returns to creditors, and achieve the reorganization of viable but financially distressed firms.

Therefore, allowing the choice of insolvency forum in the context of individuals can more easily contravene the “public policy” of a country. As a result, even if UNCITRAL decides to favor the choice of insolvency forum for individuals, the local jurisdiction can still deny recognition on the basis that it contravenes its local public policy. A rule based on the debtor’s habitual residence, as currently exists under the Model Law,38 would be more desirable and respectful of a country’s public policy. For that reason, if the COMI rule is abolished, the Model Law should—for individual debtors—adopt the debtor’s place of habitual residence as a non-rebuttable presumption in determining proper insolvency forum.

B. The Second-Best Approach: Ex Post Choice of Insolvency Forum

If UNCITRAL decides to keep the COMI rule, debtors should be allowed to initiate an insolvency proceeding in any jurisdiction that permits the initiation of insolvency proceedings by foreign companies.39 Additionally, and more importantly, the Model Law should establish that the place where the insolvency proceeding is initiated will be considered functionally equivalent to the debtor’s COMI for the purpose of the Model Law. Put differently, initiating an insolvency proceeding in the place of debtor’s COMI or in any other forum chosen by the debtor should trigger similar effects under the Model Law. To grant these functionally equivalent effects, however, the debtor needs to show that the place of filing is beneficial for the creditors as a whole. In the absence of evidence showing the beneficial effects of choosing a different insolvency forum, the debtor would still be allowed to initiate an insolvency proceeding if it is permitted by the laws of that jurisdiction, but it would be subject to the legal risks currently associated with initiating an insolvency proceeding in a place that it is not the debtor’s COMI.

This second-best solution improves the current regulatory framework for cross-border insolvency in several ways. First, it allows debtors and creditors to benefit from the choice of a more efficient insolvency forum.40 While this practice is already observed in the market,41 the adoption of the proposed solution in the Model Law would provide more certainty. There would be no more wasteful litigation over what is the “real” COMI. Second, if the debtor shows that the place of filing can be beneficial for the creditors as a whole, this solution may avoid some of the legal risks associated with choosing an insolvency forum that is not the debtor’s COMI. Third, COMI shifts can themselves require expenditures by the debtor. Our second-best solution eliminates the need for such costs.

Yet, it should be noted that this solution is inferior to the preferred approach suggested in Section 3.1. On the one hand, this solution can lead to litigation costs if, for example, there are disagreements about the ability of the place of filing to benefit the creditors as a whole. Still, litigating over whether the chosen forum makes the creditors better off is preferable to litigating over the debtor’s ability to meet the definition of COMI and over what precise day should be used to determine the debtor’s COMI. On the other hand, this option provides the debtor the ability to file in the forum of its COMI even when that forum is makes creditors worse off. As described above, sophisticated creditors will increase the debtor’s cost of borrowing to account for that inefficient possibility. Hence, while solution can improve the current regulatory framework for cross-border insolvency, it is second best. The Commitment Rule remains the optimal and first-best rule.

C. Comparing Approaches to Determine the Initiation of a Foreign Main Proceeding

An ideal rule to determine where a foreign main proceeding needs to be initiated should provide predictability, certainty, and the ability to choose a value-enhancing insolvency system. Additionally, it should minimize litigation costs and prevent the opportunistic choice of insolvency forum. The rule that performs best over these dimensions will best support financial markets and the promotion of firms’ ex ante access to finance. Based on these pillars, Table 1 compares the desirability of different systems for the choice of insolvency forum.

Table 1. Optimal rule for determining insolvency forum

Consistent with the views expressed in this article, the Commitment Rule provides the most desirable outcome. From the perspective of debtors, our preferred rule is superior to the rest of the approaches in every single aspect except for the flexibility to update the forum choice in light of changed circumstances. Still, flexibility can be good if it is used for value-enhancing forum shopping, and bad when it is used for opportunistic forum shopping. The COMI Rule does not promote the former and ostensibly prevents the latter.51 It allows for precisely the wrong kind of flexibility. Thus, with the Commitment Rule, UNCITRAL can promote beneficial forum shopping while preventing opportunistic forum shopping.

From the perspective of creditors, our preferred approach is the most protective one. Ex ante, sophisticated lenders are protected by the fact that they can choose not to extend credit—or require more collateral or a higher interest rate—to companies choosing a creditor-unfriendly insolvency forum. Moreover, unlike what happens with the COMI rule, any change of insolvency forum would require some form of approval from the creditors. Depending on the approach chosen by UNCITRAL, this approval can be express, or it can be implied by not challenging the change of insolvency forum or by accepting amendment rules potentially established in the company’s constitution.

IV. COMI, the Gibbs Rule and the Future of the Model Law on Cross-Border Insolvency

In a well-known case from 1890, the English Court of Appeal established that, unless a creditor voluntarily submits itself to the jurisdiction of a foreign court, a discharge or modification of a debt is only effective if it is undertaken in accordance with the law governing the contract.52 While the rule of the case might have been viewed as simply dealing with choice of law, it has been interpreted by English courts to provide that a creditor cannot be bound by a reorganization proceeding that takes place in a jurisdiction other than the one whose law governs the debt contract. This rule—known as the “Gibbs rule”—places harsh restrictions on forum choice. Some courts have gone as far as to hold that creditors with debt contracts subject to English law53 cannot be bound even by a moratorium triggered by a restructuring proceeding in other jurisdictions.54

Although the Gibbs rule has been criticized by many courts and commentators,55 it is still followed in various common law jurisdictions, including the United Kingdom and Hong Kong.56 Generally, the arguments in support of the Gibbs rule include its ability to provide predictability and legal certainty.57 Moreover, it is also mentioned that it can protect creditors from debtors opportunistically seeking to restructure their debts in a less attractive forum for lenders.58 Therefore, by providing a greater level of protection to creditors, the Gibbs rule can reduce borrowing costs for firms.

Despite these advantages, the Gibbs rule has a major flaw: it leads to an inefficient debt restructuring ex post. Indeed, debtors with multiple creditors may have debt contracts subject to different laws.59 Therefore, the Gibbs rule could prevent debtors from initiating a comprehensive debt restructuring in a single forum. Instead, unless creditors voluntarily submit to the forum of the foreign insolvency proceeding, debtors seeking to restructure their entire liabilities would be required to initiate parallel insolvency proceedings in all Gibbs-rule jurisdictions designated in the debtor’s various agreements. This fragmentated debt restructuring is more costly and can also lead to inefficient oversight and deployment of the debtor’s assets.

As mentioned in Section 1, we recognize the superiority of the policy option chosen by UNCITRAL to deal with cross-border insolvency—modified universalism—which is mainly based on the idea of centralization. Yet, we view predictability and access to finance as fundamental to the operation of an insolvency system. Therefore, we support the idea of allowing debtors and creditors to commit to the insolvency law applicable to a insolvency proceeding provided that the same law applies to the general body of creditors. In our view, any exceptions to the lex fori concursus (that is, the insolvency law of the jurisdiction where the proceeding has been opened) should be limited to and generally justified on the basis of clear market or contracting failures, such as those faced by employees or the negative externalities created by certain financial contracts. Allowing lenders to individually opt out of the general rules governing an insolvency proceeding by private contract with the debtor would undermine the collective nature that is at the core of insolvency proceedings.60

While the Commitment Rule favours a contractual solution to forum choice, the logic behind that rule does not support the Gibbs rule. It is worth noting that the contract in the Commitment Rule would—like a security interest—be publicly recorded and available. The public nature of this contract brings it squarely within the collective foundations of insolvency law.61 The same cannot, however, be said of the Gibbs rule.

Given that UNCITRAL is currently seeking to harmonize the law applicable to insolvency proceedings,62 the text eventually published by UNCITRAL should clearly establish that, as a general rule and subject to any necessary exceptions, the law applicable to the discharge or modification of debt contracts should be the lex fori concursus.63 This would allow a company to restructure all its liabilities in a single jurisdiction, which, under the Commitment Rule, would be the insolvency forum publicly announced in the company’s constitution. Under such a regime, jurisdictions would be able to enjoy any benefits traditionally associated with the Gibbs rule while creating a centralized insolvency forum that can lead to a more efficient outcome for debtors and creditors.

If the new rules on applicable law were not incorporated into the Model Law on Cross-Border Insolvency but just enacted as another model law, UNCITRAL may face the risk observed with other model laws, such as the Model Law on Enterprise Group Insolvency or the Model Law on Recognition and Enforcement of Insolvency Related Judgments, that have yet not been very successful in terms of implementation.64 As recognized by UNCITRAL, the new rules on applicable law are essential to achieve the key objectives of effective and efficient insolvency proceedings, such as promoting predictability and legal certainty while preventing opportunistic behaviour of debtors and creditors.65 As a result, the adoption of these new rules in the Model Law on Cross-Border Insolvency is justified given their importance for the consolidation of a robust and comprehensive system of cross-border insolvency.

To be clear, a rejection of the Gibbs rule and the adoption of a rule in favor of the lex fori concursus does not mean that a jurisdiction will surrender its substantive policy on local creditor priorities. To the extent the original opinion in Gibbs suggested that insolvency proceedings should respect substantive contract rights, it should be entirely uncontroversial. Very few jurisdictions today would recognize a foreign insolvency proceeding that nullified their citizens existing contract rights or security interests. Such proceedings would fall under the public policy exception to recognition. There is—or should be—a sort of international Butner principle providing that substantive rights come from the applicable non-insolvency law.66

The problem with Gibbs is that later courts interpreted it to provide direction about the content of insolvencylaw and even the choice of insolvency forum.67 That goes too far. By this reading, companies with debt contracts subject to the laws of a jurisdiction applying the rule in Gibbs are required to either conduct their entire restructuring proceeding in those jurisdictions or initiate separate parallel proceedings in various jurisdictions.

While it is sensible to allow companies to choose such complicated forum rules, the Gibbs rule mandates the bundling between substantive law and insolvency law (and insolvency forum). If widely adopted, Gibbs would thus impose large costs on multinational firms by necessitating fractured insolvency proceedings whenever the firm’s has two or more relationships governed by distinct substantive law.68 In a system of modified universalism, protectionist rules like the Gibbs rule should be abolished, but sensible rules like an international Butner principle—preserving and protecting substantive rights, such as those of local tort victims and employees, or even banks—should be preserved and reaffirmed through the public policy exception.

The Model Law on Cross-Border Insolvency should be revised to adopt both the Commitment Rule and the new rules on applicable law establishing that, with the exceptions that have been mentioned, any discharge or modification of debt contracts in insolvency proceedings will be subject to the lex fori concursus. Those supporting the Gibbs rule should not have any reasons to object. If the motivation behind the Gibbs rule is to promote predictability, creditor protection and access to finance, this is achieved through the combination of the Commitment Rule and its ability to determine the lex fori concursus. Given the improvements that the new rules on applicable law can bring, accompanied by the flexibility and opportunities provided by the Commitment Rule, jurisdictions that have already adopted the Model Law should welcome these new rules on cross-border insolvency and some of the jurisdictions that have traditionally be reluctant to the adoption of the Model Law will be more willing to implement a text that provides more protection and flexibility to their local debtors and creditors.

V. Conclusion

The Model Law has played a major role in the improvement and efficient management of cross-border insolvency cases. The principle of modified universalism and the cooperation and assistance promoted by the Model Law have contributed to the success of this instrument enacted by UNCITRAL. Therefore, any future reforms and developments in the space of cross-border insolvency should keep embracing these principles. But the Model Law does contain a flaw: the adoption of the COMI rule as the method to determine the place where a foreign main procedure should be initiated. The COMI rule can create uncertainty and litigation costs, and it can lead to opportunistic behavior of debtors vis-à-vis creditors. Moreover, it can undermine the ability of insolvency law to maximize the returns to creditors, to effectively reorganize of viable but financially distress businesses, and to ultimately promote entrepreneurship, access to finance and economic growth.

Therefore, UNCITRAL should abandon the COMI rule in favor of the Commitment Rule. In conjunction with this move, the (revised) Model Law should also deal with applicable law in insolvency proceedings, clearly specifying, among other aspects, that the discharge or modification of debts will be subject to the lex fori concursus. These reforms are consistent with the core principles of modified universalism and would significantly improve the efficiency of insolvency proceedings while also reducing opportunistic behaviour by debtors and creditors.

• 2This policy option has been generally supported in the literature. See, e.g., Lucian A. Bebchuk and Andrew T. Guzman, An Economic Analysis of Transnational Bankruptcies, 42 J.L. & Econ. 775 (1999); Jay L. Westbrook, Global Insolvency Proceedings for a Global Market: The Universalist System and the Choice of a Central Court, 96 Tex. L. Rev. 1473 (2018). For skepticism about universalist models, however, see Lynn M. LoPucki, Cooperation in International Bankruptcy: A Post-Universalist Approach 84 Cornell L. Rev. 696, 709–25, 728–32 (1999); Frederic Tung, Is International Bankruptcy Possible? 23 Mich. J. Int’l L. 31 (2001).
• 3UNCITRAL Model Law on Cross-Border Insolvency art. 2, 16, 17 (1997).
• 4See UNCITRAL Model Law on Cross-Border Insolvency art. 16(3). See also UNCITRAL, Model Law on Cross-Border Insolvency: Guide to Enactment and Interpretation, 70–71 (2014) https://perma.cc/3JPM-K5YN.  For a summary of the case law interpreting the COMI rule, see UNCITRAL, Digest of Case Law on the UNCITRAL Model Law on Cross-Border Insolvency, 39–42, https://perma.cc/7NTQ-6496.
• 5For early work criticizing the concept of COMI, see Robert K. Rasmussen, A New Approach to Transnational Insolvencies, 19 Michigan J. of Int’l L. 1 (1997). Emphasizing the harmful economic effects generated by the existence of the concept of COMI, see Aurelio Gurrea-Martinez, Reinventing Insolvency Law in Emerging Economies, 248–72, (Cambridge Univ. Press, 2024). Joshua C. Macey also laid out the case against COMI at a conference. See Maria Cecilia T. Sincangco & Kristina P. Castaneda, Strengthening Insolvency Systems in Asia and the Pacific: Post-Conference Booklet, Asian Development Bank 200 (Dec. 2022), https://perma.cc/X8DV-2W8R  (reporting Macey’s view that “COMI should not play a significant role in selecting a filing venue. Rather, the party should just have the market price in the ideal forum.”).
• 6These jurisdictions include the United States, the United Kingdom, and Singapore, provided that the debtor shows some forms of “connection” with the country. In the United States, this connection is generally shown if the debtor has property in the United States. See 11 U.S.C. § 109(a). To that end, the concept of property has been interpreted very broadly. See In re Global Ocean Carriers Ltd., 251 B.R. 31, 37–39 (Bankr. D. Del. 2000). In the United Kingdom, foreign companies can initiate insolvency proceedings if they show a “sufficient connection,” which can be found if, for example, the debtor has assets or creditors in the country or debt contracts subject to English law. See Van Gansewinkel Groep B.V. [2015] EWHC 2151, *19, ¶¶ 67–70. In Singapore, foreign companies can initiate insolvency proceedings if they show a “substantial connection,” which may include situations where the debtor: (i) has its COMI in Singapore; (ii) is carrying on business in Singapore or has a place of business in Singapore; (iii) has substantial assets in Singapore; (iv) has chosen Singapore law as the law governing a loan or other transactions; or (iv) has submitted to the jurisdiction of the Singapore Courts in the resolution of one or more disputes relating to a loan or other transactions. See Insolvency, Restructuring and Dissolution Act 2018 § 63(3), § 246(1)(d), and § 246(3). Other factors, such as the listing of securities in Singapore, can also show a substantial connection. See In Re PT MNC Investama TBK [2020] SGHC 149, *5, ¶ 13.
• 7This aspect, along with other weaknesses of the concept of COMI, has been highlighted previously. See Gurrea-Martinez, supra note 5, at 248–72.
• 8Insolvency systems can be inefficient in a number of different ways. For example, the system could have substantive rules that deter efficient negotiations by ceding too much power to single creditors, as the United States Bankruptcy Act did. Alternatively, it could give too much power to debtors, as was the case with the Bankruptcy Code when it first became effective. It could be biased towards liquidation or reorganization. It could also be inefficient if there is not a robust ecosystem to support the written legislation. On the last point, see Mark J. Roe & Michael Simkovic, Bankruptcy’s Turn to Market Power, 92 U. Chi. L. Rev. 1 (forthcoming 2025).
• 9Id.
• 10Recent cases—such as the collapses of FTX and Three Arrows—demonstrate the difficulties associated with determining the debtor’s COMI in the context of crypto exchanges. See In re FTX Cryptocurrency Exchange Collapse Litigation, 677 F.Supp.3d 1379 (J.P.M.L. 2023); In re Three Arrows Capital, Ltd., 649 B.R. 143 (Bankr. S.D.N.Y. 2023).
• 11Consider, for example, In re Latam Airlines Group S.A., 620 B.R. 722 (Bankr. S.D.N.Y. 2020). The parent holding company was located in Chile while the largest subsidiary with the largest proportion of employees and operations was located in Brazil. Id. at 731. Therefore, in the absence of exceptional circumstances showing that the central administration of the company is in another jurisdiction and that is ascertainable by third parties, one might expect the COMI to be in one of these jurisdictions. In the end, LATAM Airlines filed for bankruptcy in New York. Interestingly, when seeking recognition in Chile, the company argued that the COMI was in the United States and therefore the proceeding should be recognized as a foreign main proceeding.
• 12See Gurrea-Martinez, supra note 5, at 261.
• 13For a discussion of these different approaches, see Re: Zetta Jet Pte Ltd and others (Asia Aviation Holdings Pte Ltd, intervener) [2019] SGHC 53, 14–23.
• 14In some jurisdictions, such as the United Kingdom, Australia, and the European Union, the use of the registered office as the place of the debtor’s COMI has been understood as a relatively strong presumption. For the United Kingdom, see Re Stanford International Bank Ltd and others [2009] EWHC 1441 (Ch), ¶¶ 63, 70. For Australia, see Ackers v Saad Investments Company Limited (in official liquidation) [2010] FCA 1221, ¶¶ 45-57. For the European Union, see In re Eurofood IFSC Ltd (Case C-341/04) [2006] 1 Ch 508, ¶¶ 26–37. In other countries, such as Singapore and the United States, the presumption of the registered office can be more easily rebutted if it is shown that the debtor’s business decisions are made in another place that is ascertainable by third parties. For Singapore, see Re Zetta Jet Pte Ltd and others [2019] 4 SLR 1343, 1364–65 (HC), ¶¶ 76–81. For the United States, see In re Tri-Continental Exchange Ltd 349 BR 627, 633–35 (Bankr. E.D. Cal. 2006). See also Aurelio Gurrea-Martínez, The Implementation of the Model Law on Cross-Border Insolvency: International Divergences and Challenges Ahead, 21(3) Int’l Corp. Rescue 145 (2024) (analyzing how the concept of COMI and other key provisions of the Model Law have been implemented and interpreted differently around the world).
• 15These controversies and litigation costs have been observed in many cases where the COMI was disputed, as occurred in popular cases such as Eurofood, ¶¶ 2223 (where it was not clear if the COMI was located in Ireland or Italy), Stanford International Bank, ¶ 2 (where the US and Antiguan representatives each submitted that they were appointed in the “main proceeding”), or more recently LATAM (where some local creditors in Chile challenged that the debtor’s COMI was in the United States and, therefore, argued that the U.S. Ch. 11 reorganization procedure initiated by the company should not be recognized as a foreign main proceeding), Argent Energy (Emergency Application for Provisional Relief Pursuant to Sections 105(A) and 1519 of the Bankruptcy Code), In re Argent Energy Holdings, Inc., No. 16-20061 (Bankr. S.D. Tex. 2016) (where the foreign representative sought a Ch. 15 recognition of Canadian proceedings as foreign main proceedings, and a group of debenture holders objected, arguing, in part, that the debtors’ COMI was not Canada), In re Black Press (where the concept of COMI was litigated in the context of corporate group and the Delaware bankruptcy court ultimately denied the recognition of a Canadian proceeding as a “foreign main proceeding” with respect to the U.S. subsidiaries), or In Re Fullerton Capital Limited, [2024] SGHC 155, ¶ 1 (where the liquidators of the company submitted an application to the Singapore High Court for the recognition of the liquidation procedure as a foreign main proceeding, but a creditor challenged the application on the basis that the company’s COMI was not in the British Virgin Islands even if the company’s registered office was in that jurisdiction). Cases where the concept of COMI is not clear include virtually all cases in which a company initiates an insolvency proceeding in a jurisdiction where a company has its registered office but does not have significant operations (as often occurs in companies incorporated and initiating insolvency proceedings in offshore jurisdictions), or in cases where a company initiates an insolvency proceeding in a jurisdiction where it may have assets, creditors, and operations but it has a similar—or even greater—presence in other jurisdictions (as often occurs with many multinational companies). For that reason, it is not surprising that, as highlighted by Judge Allan Gropper (ret), “issues relating to the debtor’s centre of main interests have been litigated repeatedly in Chapter 15 cases”. See Allan L Gropper, Chapter 15 of the United States Bankruptcy Code , in K E Lindgren (ed), International Commercial Litigation and Dispute Resolution, Ross Parsons Centre of Com., Corp. and Tax Law, Publ’n Series, Sydney 2010 at 154 (cited in Ackers v Saad Investments Company Limited (in official liquidation) [2010] FCA 1221, 12–13, ¶ 41).
• 16Ted Janger and John Pottow make this point clear by noting that debtors can simply “migrate their COMIs” to forum shop. See Ted Janger & John Pottow, Cross-Border Insolvency Forum Shopping Naivete, Credit Slips (Dec. 20, 2023), https://perma.cc/PEW5-HEGH.  Inexplicably, they cite this as evidence in support of COMI and against our proposal. It is a mystery to us why unrestricted ex post debtor opportunism is superior to market-constrained ex ante debtor choice. But at least we all agree on the facts: COMI allows for manipulation at the time of filing and the Commitment Rule does not. Indeed, if contrary to our analysis, unfettered choice at the time of filing is indeed the optimal rule, our solution would allow a company to put this choice in its constitution. Alternatively, the general rule should just allow unfettered choice, without the pretense of COMI. We discuss this second-best approach below.
• 17One fascinating part of the debate on COMI is that its supporters are split into two groups making the exact opposite arguments. One group of COMI defenders argue that COMI is just fine because it cannot be manipulated to allow bad forum shopping. Another group argues that COMI is just fine because it can be manipulated to allow good forum shopping. Two supporters have taken the position—as cynical and disingenuous as it is—of arguing both points at the same time. See id. (arguing at one point that we need COMI because “forum shopping can also be bad” and then arguing that COMI is harmless because debtors “can migrate their COMIs or they can simply file a nonmain proceeding in a more attractive venue.”). The inconsistency and illogic of this argument is, at the very least, impressive.
• 18For different reasons but still advocating for the need to revisit the concept of COMI, see Jay L. Westbrook, Moss Fletcher Lecture (June 2023), https://perma.cc/5MRP-WZTG.
• 19See Rasmussen, supra note 5, at 4–5. For the purpose of this Article, the terms “company’s constitution,” “corporate charter,” or “articles of association” are used interchangeably. The terms “bankruptcy procedure” and “insolvency proceedings” are also used as synonyms.
• 20The fact that a country provides a debtor-friendly regime does not necessarily mean that the jurisdiction is less attractive to creditors. In fact, jurisdictions can be pro-debtor and pro-creditor, anti-debtor and anti-creditor, or somewhere in the middle. See Aurelio Gurrea-Martinez, The Myth of Debtor-Friendly or Creditor-Friendly Insolvency Systems: Evidence from a New Global Insolvency Index, 16–19 (Singapore Mgmt.Univ. Yong Pung How Sch. of L. Research Paper, Apr. 2023), https://perma.cc/M4AE-QQJP
• 21The optimal insolvency system for a debtor may change over time as forums and the debtor itself evolve. See Robert K. Rasmussen, Debtor’s Choice, A Menu Approach to Corporate Bankruptcy, 71 Tex. L. Rev. 51, 116–21 (1992); Alan Schwartz, A Contract Theory Approach to Business Bankruptcy, 101 Yale L.J. 1807, 1808–09 (1998).
• 22Employees generally have this preferential treatment in most jurisdictions. Tort claimants, however, only enjoy a preferential treatment in the ranking of claims in some jurisdictions (e.g., Spain). Indeed, given the generally poor treatment that tort creditors receive in most jurisdictions, it may well be the case that they will be unaffected by our proposal.
• 23See Gurrea-Martinez, supra note 5, at 266–67.
• 24Countries have the freedom to choose the type of creditors included under this category for the purpose of the Model Law. For example, countries more concerned with the protection of non-adjusting and weakly adjusted creditors in insolvency proceedings may decide to include employees, tort claimants, and tax authorities, and other jurisdictions (e.g., those that deal with non-adjusting and weakly adjusted creditors in a different manner) may even prescribe that no creditor would qualify as a vulnerable creditor for the purpose of cross-border insolvency.
• 25See Gurrea-Martinez, supra note 5, at 266–27.
• 26See, e.g., Lynn M. LoPucki, The Case for Cooperative Territoriality in International Bankruptcy, 98 Mich. L. Rev. 2216, 2248–50.
• 27Given that vulnerable creditors would always get priority, their involvement in the change of forum would not be needed.
• 28See Rasmussen, supra note 5, at 34.
• 29Requiring the debtor to provide notice to the creditors and allowing the change of insolvency forum if no creditor objects would be as protective as requiring individual consent from all the pre-existing creditors, as some authors have suggested. See Rasmussen, supra note 5; Aurelio Gurrea-Martinez, Insolvency Law in Emerging Markets 31–32 (Ibero-American Inst. for L. and Fin., Working Paper, Mar. 2020), https://perma.cc/6T4M-XRQJ.  However, our proposed approach would provide more flexibility if the debtor wants to change the insolvency forum in order to choose a more value-enhancing insolvency regime.
• 30For a suggestion of this approach, see Randall Thomas and Robert K. Rasmussen, Timing Matters: Promoting Forum Shopping by Insolvent Corporations, 94 Nw. Univ. L. Rev. 1357 (2000).
• 31Anthony J. Casey & Joshua C. Macey, Bankruptcy Shopping: Domestic Venue Races and Global Forum Wars, 37 Emory Bankr. Dev. L.J. 436, 500–01 (2001).
• 32In the European Union, the rules on cross-border insolvency create more predictability than those existing internationally given the existence of a more comprehensive and harmonized framework to deal with cross-border insolvency within the EU (the European Insolvency Regulation) as well as the existence of a single, superior court (the Court of Justice of the European Union) that facilitates a consistent and harmonized interpretation of the rules on cross-border insolvency within the EU. In the context of COMI, the EU’s approach, while still generating uncertainty and litigation costs and often preventing the choice of a more efficient insolvency system, is more predictable than the COMI rule. For example, in the EU, the presumption of the registered office is generally very strong. See In re Eurofood IFSC Ltd [2006], ¶¶ 26–37.
• 33See, e.g., Wolf-Georg Ringe, Forum Shopping Under the EU Insolvency Regulation 9 Eur. Bus. Org. L. Rev. 579 (2008).
• 34In the United States, this could be accomplished by a provision of federal law—likely in Chapter 15 of the Bankruptcy Code (11 U.S.C. §§ 1500, et seq). The law should make clear that it preempts any contrary state laws. Alternatively, it would have to be done by amendment to individual state corporations laws.
• 35Gurrea-Martinez, supra note 5, at 264–65.
• 36Id.
• 37In a survey conducted by the World Bank some years ago, it was found that more than half of the low- and middle-income countries do not have an insolvency framework for personal insolvency. See World Bank, Report on the Treatment of the Insolvency of Natural Persons (2014), https://perma.cc/SS4S-4FKQ.
• 38See Model Law on Cross-Border Insolvency art. 16(3).
• 39This solution is generally admitted. However, it entails some of the risks and practical challenges mentioned in Section 2.
• 40The choice of a value-enhancing insolvency forum has been largely supported by many courts, scholars and commentators. See, e.g., Randall Thomas & Robert K. Rasmussen, Timing Matters: Promoting Forum Shopping by Insolvent Corporations, 94 Nw. U. L. Rev. 1357, 1362 (2000); Horst Eidenmüller, Free Choice in International Company Insolvency Law in Europe, 6 Eur. Bus. Org. L. Rev. 423, 425–26 (2005); Wolf-Georg Ringe, Forum Shopping under the EU Insolvency Regulation, 9 Eur. Bus. Org. L. Rev. 579, 601–03 (2008); Sundaresh Menon, Chief Justice, Supreme Court of Singapore, Keynote Address at the 18th Annual Conference of the International Insolvency Institute (Sept. 25, 2018), https://perma.cc/BUH3-8V4L;  Kannan Ramesh, Justice, Supreme Court of Singapore, Address at the Texas International Law Journal Symposium (Feb. 6, 2021), https://perma.cc/BS6Q-QCPL;  Casey & Macey, supra note 31, at 474–82 (2021).
• 41This practice is particularly popular among companies from countries that do not have efficient insolvency frameworks as it typically occurs in emerging economies. Recent examples include Avianca, LATAM Airlines, and Philippines Airlines. Even though these companies are primarily based in Colombia, Brazil or Chile, and the Philippines, respectively, they all filed for bankruptcy in the United States.
• 42COMI is manipulable, which means there is considerable flexibility for the debtor to choose a forum. Still, the cost of manipulation is not zero. Thus, all three approaches allow choice. But COMI imposes meaningless transaction costs. For example, a COMI shift where a company creates an artificial structure to establish a new COMI is an economically meaningless transaction engage in to achieve the outcome that is free in the other two approaches.
• 43It would be most if litigation over whether creditors are better off were free and perfect. Because that is not likely the case, it is intermediate.
• 44The ex ante commitment and market discipline make the Commitment Rule far superior on this dimension.
• 45If COMI were strictly enforced and immutable, it would be “most” in this category. That is not the case. And if it were, that would reduce flexibility in the other rows.
• 46This is intermediate for two reasons. First, the opportunistic choice of COMI is still available. Second, while some protection is provided against opportunism through litigation about whether the chosen forum is beneficial to the creditors, that litigation will be imperfect and less protective than a clear commitment in the company’s constitution.
• 47Of course, if the debtor is allowed to change the forum according to the rules established in the company’s constitution, these adopted rules might be the most flexible. For example, the adopted rules might say the insolvency forum can be changed if it is approved by an arbitrator or by a majority of independent directors. But the debtor might alternatively choose something very inflexible, like “always Delaware.” The rule therefore does not guarantee ex post flexibility. But it does guarantee that the debtor will make the choice subject to market discipline.
• 48In practice, COMI is very manipulable, but it does create some transaction costs. See supra note 42.
• 49See supra note 46 on imperfect litigation. The only “most” rule here would be total choice at filing. But that would score least on protection against manipulation and ex ante predictability of forum.
• 50It is difficult to rank COMI and the Ex Post Choice Rule here. Both require ex post litigation. We have ranked COMI intermediate because COMI litigation has been fairly permissive. If courts take seriously the Ex Post Choice Rules requirement that “no creditor is worse off,” they will likely overrule the debtors choice more often than they do with COMI shifts. That is generally a good thing, but it does introduce some uncertainty about whether the debtor’s choice will be enforced. None of these issues arise with the Commitment Rule because a commitment in the company’s constitution is easily verifiable.
• 51In reality, it provides more flexibility than its text would suggest.
• 52Antony Gibbs & Sons v La Société Industrielle et Commerciale des Métaux (1890) LR 25 QBD 399, 406–08 (Eng.).
• 53The rule may hypothetically apply in any other jurisdiction governing the debt contract. Bakhshiyeva v. Sberbank of Russia [2018] EWCA (Civ) 2802 (UK).
• 54Chang Chin Fen v. Cosco Shipping (Qidong) Offshore Ltd [2021] CSOH 94 (Scot.).
• 55See, e.g., Ian Fletcher, Insolvency in Private International Law (Oxford Univ. Press, 2005), 2.127–2.129; Kannan Ramesh, The Gibbs Principle: A Tether on the Feet of Good Forum Shopping (2017) 29 Singapore Acad. L.J. 42, 46–48. See also Re Pacific Andes Resources Development Ltd [2016] SGHC 210, at [48] (Singapore); In re Agrokor DD 591 B.R. 163, 192, 196 (Bankr. S.D.N.Y. 2018).
• 56Bakhshiyeva v. Sberbank of Russia [2018] EWCA Civ 2802 (UK); Re Rare Earth Magnesium Technology Limited [2022] HKFCI 1686 (HK).
• 57For a summary of the debate on the pros and cons of the Gibbs rule, see for example Bruce Bell, Jo Fernandes & Tim Bennett, In Defence of “Gibbs”?, Latham & Watkins (June 20, 2023), https://perma.cc/YTB2-U2L4.
• 58Id.
• 59As bad as this problem is today, it could get worse. Few jurisdictions follow Gibbs, and those that do tend to have effective restructuring regime. But there is no guarantee that this state of affairs will remain constant. If more countries move to the Gibbs rule, the problems multiply.
• 60Justifying the existence of a mandatory and collective debt collection mechanism, see Thomas H. Jackson, The Logic and Limits of Bankruptcy Law 7–19 (2001).
• 61Advocating for a contractual solution that still preserves the collective nature that should characterized insolvency proceedings, see Robert K. Rasmussen, Debtor’s Choice: A Menu Approach to Corporate Bankruptcy, 71 Tex. L. Rev. 51 (1992).
• 62U.N. Secretariat, Applicable law in insolvency proceedings, Note by the Secretariat, 12-25, Int’l Trade Law, U.N. Doc. A/CN.9/WG.V/WP.187 (Feb. 7, 2023), https://perma.cc/MM5J-FS3H.
• 63So far, this seems to be the approach adopted by UNCITRAL. See id. at 12.
• 64To the best of our knowledge, none of these Model Laws have been adopted by any jurisdiction yet. Nonetheless, the adoption of the Model Law on Enterprise Group Insolvency in full, as well as the partial adoption of the Model Law on Recognition and Enforcement of Insolvency-Related Judgements, has been suggested in the United Kingdom. See Consultation Outcome Implementation of Two UNCITRAL Model Laws on Insolvency Consultation, The Insolvency Service (July 10, 2023), https://perma.cc/K5XK-UFK5.
• 65See U.N. Secretariat, Applicable Law in Insolvency Proceedings, Note by the Secretariat, 2–3, Int’l Trade Law, U.N. Doc. A/CN.9/WG.V/WP.187 (Feb. 7, 2023), https://perma.cc/SJ24-2KYC.
• 66Under the “Butner principle” existing in the United States, the rights created under non-insolvency law should be honored under insolvency law unless a specific insolvency policy or provision provides otherwise. See Butner v. United States, 440 U.S. 48 (1979).
• 67See, e.g., Compromising English Law Debts: Has the Rule in Gibbs Had Its Day?, Dentons (Nov. 27, 2018) https://perma.cc/VJG3-PPTW;  see also Bakhshiyeva v. Sberbank of Russia [2018] EWCA Civ 2802.
• 68The Gibbs rule has the additional flaw of allowing commitment to an insolvency forum to be done by private contract, a contract that can be executed even after other creditors have already made loans. A key component to the Commitment Rule is that it requires the ex ante commitment to be made publicly in the company’s constitution.