Public Information Notices
Free Email Notification
On April 7, 2003, the Executive Board of the International Monetary Fund (IMF) continued its discussion on the design of collective action clauses (CACs), and promoting the inclusion of CACs in international sovereign bonds.1
The IMF's examination of CACs is part of its efforts to strengthen the framework for crisis resolution by developing and promoting effective tools to help make sovereign debt restructuring more orderly and predictable. In its Communiqué of September 28, 2002, the IMF's International Monetary and Financial Committee welcomed the ongoing dialogue in the Group of 10 and other fora with private creditors and emerging market sovereign issuers. It also encouraged the official community, the private sector, and sovereign debt issuers to continue working together to develop CACs, and to promote their early inclusion in international sovereign bond issues.
Notwithstanding the encouragement since 1996 by the official community of the use of CACs, and in particular of majority enforcement and majority restructuring provisions, progress toward their widespread inclusion in sovereign debt contracts has been limited. Over the past several months, however, there have been a number of important developments with respect to both the design of CACs and their inclusion in sovereign bonds governed by New York law.2
The Executive Board's discussion was based on a paper entitled "Collective Action Clauses: Recent Developments and Issues", which provided a comprehensive review of issues and developments in promoting the use of CACs in international sovereign bonds.
Executive Board Assessment
Executive Directors welcomed the opportunity to review progress toward the inclusion of CACs in sovereign debt instruments. They were encouraged by developments over the past year with respect to both the design of majority restructuring and majority enforcement provisions, and the incorporation of such clauses into bonds governed by New York law. In this regard, Directors lauded Mexico's recent successful issuance as well as efforts by other countries, including Egypt, Lebanon, and Qatar to include such collective action clauses in their bonds. They noted that the Mexico issuance has gone a long way toward resolving the "first mover" problem. There was broad agreement that the wider use of such clauses could contribute to more orderly and rapid agreement on debt restructurings, although a few Directors observed that without well-known ex ante presumptive limits on Fund lending CACs may not provide sufficient incentives to the sovereign and its creditors to expedite restructurings. Directors saw CACs as a useful element in the Fund's overall work to strengthen the framework for crisis resolution. A number of them considered that given the limitations of CACs, work in this area should proceed in parallel with the ongoing work on a possible statutory framework.
Given the outstanding stock of bonds that do not yet include CACs, Directors acknowledged that it will take some time before CACs are included in most international bonds. Moreover, given the contractual nature of CACs, they noted that any decision as to the inclusion and design of CACs will ultimately be made by the debtor and its creditors. Against this background, Directors reiterated that the Fund's most effective strategy is to promote the more widespread use of those types of provisions that already exist in many international sovereign bond contracts. They emphasized that the Fund should more proactively promote the use of CACs through its multilateral and bilateral surveillance. Directors also stressed that all member countries, both advanced and developing, should play their part in encouraging the use of CACs, and, in this regard, welcomed the intention of European Union member states to use contractual provisions based on the framework developed by the G-10 Working Group when issuing bonds under a foreign jurisdiction.
Most Directors noted that it is too early to reach a definitive view on the degree of standardization that should be sought in terms of the design of CACs, within and across jurisdictions. Several Directors agreed that it would not be advisable for the Fund at this early stage to promote or endorse a single set of model clauses as the exclusive benchmark. Several other Directors went further, noting that the Fund should refrain from endorsing any particular model of CACs as the exclusive benchmark, even at a later stage, and leave it up to the relevant parties to negotiate the details. Nonetheless, most Directors noted that, based on recent developments, a number of observations could be made regarding those features of CACs that would both facilitate a rapid and orderly restructuring process and have a reasonable probability of being acceptable to the market. Some Directors, however, were of the view that without a clearer endorsement of a set of model clauses by the Fund, efforts to promote them would be constrained. These Directors suggested that the G-10 Working Group proposal could serve as reflecting best practices.
Majority Restructuring Provisions
Directors noted that perhaps the most important feature of any CAC is the provision that enables a qualified majority to bind all bondholders within the same issue to the financial terms of a restructuring. This provision is found in bonds governed by English law, where the required majority is normally set at 75 percent, which Directors viewed as being a reasonable threshold. As regards the method used to calculate whether the 75 percent voting threshold has been met, Directors generally agreed that a 75 percent voting threshold based on the outstanding principal of the bond, rather than on the claims of bondholders present at a duly convened meeting, would be a reasonable approach. It was noted that such an approach was utilized by Mexico, and would also be consistent with the recommendations of the G-10 Working Group Report.
Directors also noted that Mexico's recent issuance had complemented such a provision with an expanded disenfranchisement provision that precludes bonds being voted where they are directly or indirectly owned or controlled by the sovereign issuer or its public sector instrumentalities. This provision was also consistent with the recommendations made by the G-10 Working Group. Most Directors were of the view that, taken together with the voting thresholds described above, this clause assisted in striking an appropriate balance between the objective of resolving collective action problems and protecting creditors' rights. A few Directors, however, preferred to limit the disenfranchisement exclusion only to bonds held by, or on behalf, of the issuer.
Majority Enforcement Provisions
Most Directors supported the use of clauses that are designed to limit the ability of a minority of bondholders to disrupt the restructuring process by enforcing their claims after a default and prior to a restructuring agreement. They noted that two of these provisions can already be found in bonds governed by English law and New York law: (i) an affirmative vote of a minimum percentage of bondholders is required to accelerate their claims after a default and (ii) a simple or qualified majority can reverse such an acceleration after the default on the originally scheduled payments has been cured. Directors generally viewed as reasonable the thresholds for these provisions that have already been generally accepted in bonds governed by New York law: a vote by 25 percent of outstanding principal is needed to accelerate the claims and a vote of more than 50 percent is needed to de-accelerate these claims.
Directors noted that a more difficult question relates to the use of provisions that confer the right to initiate litigation on behalf of all bondholders upon a bondholder representative who is only required to act if requested by the requisite percentage of bondholders. Moreover, these provisions ensure that the proceeds of any litigation are distributed by the representative on a pro rata basis among all bondholders. Many Directors were of the view that the use of trust deeds—or an equivalent legal structure—that can implement these provisions can play an important role in the restructuring process, and that the potential benefits may justify the limited financial cost of using them. Some Directors, however, felt that the financial costs of trusts may outweigh their benefits. Nevertheless, it was noted that this is an issue for issuers and investors to decide.
Directors agreed that, while it might be desirable for CACs to possess the features described above, differences in legal systems, market features, or country circumstances may necessitate some variation. For example, not all jurisdictions will be able to limit disruptive litigation through reliance on trust deeds and may need to implement other structures to achieve the same results. Similarly, reliance on the outstanding principal amount for purposes of calculating whether the voting threshold has been met may be appropriate in the New York market, which is dominated by institutional investors. However, continued reliance on the "quorum" approach—where the calculation is based on the vote of bondholders attending a meeting—may be a more practical method for the London market, where bonds are more widely held in the retail sector.
A number of Directors expressed the view that a representation or "engagement" provision could play a useful role in the restructuring process by giving a bondholder representative the authority to act as a channel of communication between bondholders and the sovereign debtor in the context of a crisis. They noted that, although such clauses would be novel, both the G-10 Working Group Report and the Industry Associations Draft have proposed variants of such a provision. Although it is too early to form a judgment as to its specific design features, most Directors were of the view that the provision should clearly avoid the establishment of a multiplicity of bondholder committees where a number of bond issues are being restructured. However, some Directors were not enthusiastic about engagement provisions, as it has not proven difficult so far to contact bondholders, and it would be inappropriate to determine beforehand the type of restructuring process that could take place years into the future, given the evolving nature of the practice. In any event, Directors agreed that market participants should be left to decide themselves the best approach to debtor/creditor engagement.
Directors supported improved data collection and dissemination, and emphasized that the Fund should be working with members to improve investor relations programs, data standards, and transparency. They considered that Fund country reports as well as other sources provide important information for creditors to make decisions about the economic situation of sovereign issuers, and that members' subscription to the Special Data Dissemination Standard (SDDS) would also be useful. Some Directors also supported the use of contractual covenants that require the provision of certain information. A number of others, however, were not in favor of such covenants as they are not directly related to the resolution of collective action problems, and because borrowers should be committed to transparency as a matter of principle. It was also felt that the issue of appropriate covenants related to financial data is one for the issuer and its creditors to consider. Directors also noted that some of the information identified by the Industry Associations Draft would either be confidential or would be owned or controlled by parties other than the sovereign.
Directors agreed that there are few reasons to expect that bonds containing CACs should trade at higher spreads than those that do not. They noted that there has been no systematic historical difference in pricing between bonds issued in London and those issued in New York; and that Mexico—as well as Egypt, Lebanon, and Qatar earlie —was able to issue a bond in New York containing clauses without paying a noticeable premium. A few Directors cautioned, however, that at least in the period before CACs become standard practice, it remains to be seen whether the same treatment will be accorded to other borrowers for incorporating CACs into their bond contracts.
Directors welcomed the proposals to continue several forms of outreach to encourage the use of CACs. First, they strongly encouraged the staff to hold a more active dialogue with emerging market issuers, with a view to encouraging the use of CACs in the New York market that are broadly in line with those issued by Mexico and recommended by the G-10 Working Group, as well as other markets, such as Germany, where CACs are not yet the norm. Directors agreed that progress by mature market economies in the use of CACs in international bond issuance would further strengthen these efforts. Second, as part of a more concerted effort to encourage the use of CACs, they encouraged the staff to hold an international seminar or workshops with key issuers and legal practitioners later this year on ways to promote CACs.
1 The Executive Board discussion followed that of June 27, 2002 in which the Executive Board discussed the design and effectiveness of CACs, and ways to encourage their greater use in sovereign bond contracts.
2 The inclusion of CACs in sovereign bond contracts has been long-standing market practice under English law.
IMF EXTERNAL RELATIONS DEPARTMENT