Bulletin – December 2014 Global Economy Sovereign Debt Restructuring: Recent Issues and Reforms


Over the past decade, 14 countries have undertaken a total of 18 debt restructurings. Concerns surrounding some of these restructurings have led policymakers and capital market participants to review their policies and practices on restructuring sovereign debt. In particular, court rulings as a result of litigation against Argentina have raised fears that a small minority of creditors could block or frustrate a restructuring deal even when it has been agreed to by a supermajority of creditors. This article outlines the case for strengthening the current approach to debt restructuring and assesses recent proposals put forward by the International Monetary Fund (IMF), sovereign governments and capital market participants.


When a country is unable to service its debts, it is accepted practice in the international community that a country should negotiate with its creditors to restructure its debts, with the aim of returning debt to a sustainable level at the lowest cost to both the sovereign and its creditors.[1] For a country's creditors, a debt restructuring results in an upfront loss in their claims against the sovereign, but in cases where a country's economic conditions are likely to deteriorate further in the absence of a restructuring, a timely restructure may reduce the total magnitude of losses borne by creditors.

However, incentives also exist on both sides to delay a restructuring. A restructuring is an inherently costly exercise for a country to undertake as it may result in a sustained loss of access to capital markets. A country's political leaders may also be reluctant to undertake a restructuring for fear of the potential political consequences. Creditors may be reluctant to participate as they would realise losses on their claims and they may also be concerned that some creditors may receive preferential treatment by refusing to participate in any debt restructuring.

The current framework for restructuring sovereign debt is termed the ‘contractual market-based approach’. It relies on the use of collective action clauses (CACs) in sovereign bond contracts to increase creditor participation in a restructuring offer that has been negotiated by a sovereign and its creditors.[2] The clauses work by binding all creditors in a specific bond series to the decisions of a supermajority of creditors (usually 75 per cent) in that series. In effect, the supermajority of creditors exercise control over all creditors and this ensures that a small minority of creditors are unable to hold out and seek preferential treatment. An alternative framework that has been advocated at times by some academics and country authorities is a global legal mechanism to facilitate restructurings (termed the ‘statutory approach’). IMF staff suggested such a mechanism in the early 2000s – the Sovereign Debt Restructuring Mechanism (SDRM) – but IMF members rejected the proposal in favour of the contractual market-based approach.[3]

Although IMF members remain committed to the contractual market-based approach, the IMF concluded that experiences with sovereign debt restructurings over the past decade have exposed some weaknesses ( IMF 2013b). In particular:

  • the contractual market-based approach was becoming less effective in securing adequate creditor participation
  • debt restructurings were often occurring ‘too little, too late’, thus failing to restore debt sustainability.

Concerns over securing adequate creditor participation in a debt restructuring and the need to reduce incentives for creditors to hold out from a restructuring are also shared by capital market participants (ICMA 2013). As a result, policymakers – including the IMF, individual country authorities and industry groups representing capital market participants – have been discussing potential changes to the sovereign debt framework over the past two years or so. The key priority among these groups has been to enhance the current contractual market-based approach by strengthening clauses included in sovereign bond contracts. IMF staff have also made preliminary proposals to reform the IMF's lending framework to overcome concerns that debt restructurings were often occurring too little, too late.

Issues and Proposals around Holdout Creditors

For a debt restructuring to be effective in restoring debt sustainability, a sufficient majority of creditors must agree to a reduction in their claims. However, individual creditors may be reluctant to participate in a deal and so they will hold out in the hope of subsequently being able to recover their claims in full, or at the very least in amounts greater than that presented in the restructuring offer. This hope results in the collective action problem – while it is in the best interests of creditors as a whole to participate in the debt restructuring, from the perspective of each individual creditor the best outcome is if everyone else participates and they successfully hold out.

Holdout creditors can attempt to recover their claims by taking, or threatening to take, legal action in the courts of the country whose laws govern the relevant sovereign bond contract. Bond contracts are either governed by the domestic law of the sovereign or a pre-specified foreign legal system. Domestic-law bonds can be subject to retrospective changes that may affect the ability of a creditor to bring legal action against a sovereign. In contrast, foreign-law bonds are unlikely to be modified and holdout creditors holding foreign-law bonds have historically attempted to take legal action in the relevant foreign legal system (usually the United States or the United Kingdom).[4] In the extreme, distressed debt funds (often referred to as ‘vulture funds’) have used litigation as an investment strategy by purchasing distressed sovereign bonds on the secondary market at heavily discounted prices with the aim of litigating for the face value of the bonds following a country's default or a debt restructure.

Limitations of collective action clauses

The IMF and capital market participants have encouraged the widespread use of CACs in foreign-law bonds since the early 2000s (although CACs had widespread use in English law bonds prior to 2000, they had more limited use in New York law bonds). CACs have been successfully used to increase creditor participation in a number of debt restructurings to date. However, as most CACs are only binding across a single bond series, and countries usually borrow through multiple bond series, it is still possible for holdout creditors to accumulate a sufficiently large share to block the activation of the CAC in a particular series.[5] Although it may only be a specific series in which holdouts are successful, if holdout creditors are paid out in full, it may encourage creditors in future debt restructurings to hold out.

The 2012 Greek debt restructuring is a recent case where a small number of creditors were able to hold out from a restructuring even in the presence of CACs. Greece attempted to restructure €206 billion of debt, of which €21.6 billion was subject to foreign law. Of that amount, 50 per cent failed to achieve the share needed to activate the CACs or no attempt was made, resulting in around €6.5 billion or around one-third of foreign-law bonds not being restructured. In the end, Greece decided to pay out these holdouts in full, while the creditors who agreed to the debt restructuring had the value of their claims reduced by up to 75 per cent in net present value terms.

In Greece's case, these holdout creditors did not present a problem to an effective debt restructuring because the foreign-law bonds constituted only a minority of debt outstanding. However, for countries with a higher proportion of foreign-law bonds, the inability to bind all creditors to the decisions of the supermajority may present more significant problems. The Greek debt restructuring also highlighted the fact that most domestic-law bonds do not contain CACs. While the absence of CACs in domestic-law bonds can be overcome by changing domestic legislation (which Greece did by retrospectively inserting CACs), retrospective actions have the potential to lead investors to prefer foreign-law bonds, whose terms are more difficult to modify, and may undermine the functioning of domestic-law sovereign debt markets (IIF 2012).

The pari passu clause

While the possibility of holdout creditors in a single bond series has always been a potential problem with CACs, recent successful litigation against Argentina by holdout creditors on its New York law bonds based on the pari passu clause has heightened concerns. Future holdout creditors may succeed in having their claims paid out in full by preventing a country from making payments on its restructured debts.

The litigation against Argentina was based on the pari passu clause found in most sovereign bond contracts. Pari passu is a Latin phrase meaning ‘in equal step’ and is a promise by the borrower to ensure that a creditor's claim will rank equally with other creditors and not be subordinated in favour of another creditor. Despite the widespread use of the clause, the interpretation of the clause in a sovereign context is unclear and there is a rich academic literature debating this issue.[6] In a corporate context, pari passu means creditors rank equally in their claims on a firm's assets in the event of insolvency. However, unlike a corporation, a sovereign's assets are not available to be liquidated and there is no global mechanism currently available for creditors to take possession of a sovereign's assets.[7] Foreign sovereign immunity provisions in many countries may also limit the ability of creditors to gain information on where a foreign sovereign's assets are located and prevent the seizure of those assets by creditors.[8]

Prior to 2000, it was generally agreed that pari passu meant that the claims by creditors to a sovereign rank equally, but that this did not imply equal ranking in payment. This interpretation effectively meant that following a debt restructure, a sovereign was legally able to meet its repayment obligations to those who had restructured their debts, but it could stop or credibly threaten to stop servicing the debts of holdout creditors. The potential halting of payments to holdout creditors acted as a disincentive for them to hold out from a restructure. Holdout creditors could still litigate in the courts of the country whose laws governed the bond contract for payment.

However, recent decisions made by the New York District Court have strengthened an alternative interpretation of the clause. Under this interpretation (termed the rateable payment interpretation), the sovereign is required to make equal payment to all creditors in proportion to their claims.[9] The Court ruled that Argentina had violated the pari passu clause found in New York law governed bonds by failing to make payments to holdout creditors when it had made payments to holders of Argentina's restructured bonds, and also by passing laws that prevented Argentina from settling with holdout creditors. The Court ordered Argentina to make a ‘rateable payment’ to holdout creditors prior to, or at the same time as, making payments on restructured bonds. It is estimated that the accumulated principal and interest payments due to these holdout creditors is worth US$1.6 billion. To enforce the ruling, the Court also barred third parties from facilitating payment to Argentina's restructured creditors unless payment was also made to holdout creditors. In addition, to assist in enforcing the ruling, the Court allowed holdout creditors to subpoena third party banks to discover the location of Argentina's assets outside of the United States.

As Argentina was not willing to pay out holdout creditors, it was prevented from making payment on its restructured bonds and Argentina was placed on default by major rating agencies in July 2014 (though Argentina is attempting to make payments to restructured bondholders). There is speculation that some restructured creditors are considering accelerating the amounts due to them by demanding full payment to reduce payment uncertainty and counter holdout creditors (Scigliuzzo 2014).

Although developments in Argentina have had little impact on global bond yields to date (with volatility mostly restricted to Argentinian securities), it is unclear how courts will interpret the case in the future and what sort of precedent it creates for future debt restructurings. Some capital market participants argue that there are three special features of this case that potentially restrict its relevance as a precedent – that the specific wording of Argentina's pari passu clause lent itself to a rateable payment interpretation, that the court ruled that Argentina's actions violated the clause when it passed laws that effectively subordinated their claims, and that Argentina had waived sovereign immunity in its bond contracts. Analysis by the IMF also suggests that the rateable payments interpretation is unlikely to be adopted by English law courts (IMF 2014b). Nevertheless, until this becomes clear, the ruling increases the legal uncertainty around debt restructurings, which is likely to increase the incentives for creditors to hold out.

Proposed changes to sovereign bond contracts

The successful litigation against Argentina has concerned policymakers and many capital market participants. In response, the US Treasury convened a working group comprising country officials, multilateral institutions and academics to assess the implications of the litigation. Around the same time, the International Capital Markets Association (ICMA), the industry body that produces templates for bond contracts, undertook a review of its sovereign bond templates. These templates are used by sovereigns and their creditors as a basis for bond contract design, though the eventual terms are negotiated between a sovereign and its creditors. The informal working group and ICMA worked together closely and, reflecting these consultations, ICMA has recently published updated templates that clarify the interpretation of pari passu and provide for three aggregation options for sovereigns seeking to use CACs (ICMA 2014a, 2014b). To balance the rights of minority creditors, ICMA has also included provisions that sovereigns must disclose adequate information about their circumstances and the restructuring terms given to different creditor groups. To facilitate negotiations between sovereigns and creditors, creditor groups with at least 25 per cent of the vote are able to form committees to engage in negotiations with the sovereign.

To deal with holdout creditors that rely on pari passu clauses, ICMA's model clause has been changed to explicitly define pari passu to exclude the possibility of a sovereign having to make rateable payment to holdout creditors as a condition to meeting its obligations on restructured debts.[10] To provide a sovereign with flexibility in determining the best way to aggregate creditors' claims, and possibly allow for differentiation among them, the revised CAC allows for three types of voting procedures:

  1. The sovereign can choose an aggregation CAC, which would aggregate the claims across multiple bond series and thus bind all creditors in all bond series to the supermajority. Aggregation can be enforced in either of two ways:
    1. Using a ‘two-limb’ voting structure where at least half of the creditors in each bond series must accept the new terms and two-thirds of the total creditors agree to a restructure; or
    2. The sovereign can choose a ‘single-limb’ voting structure, where the claims would be aggregated for voting purposes when 75 per cent of total creditors agree to a restructure. This vote would bind all creditors across all bond series subject to the vote. As a safeguard, all affected creditors must be offered the same restructuring terms.
  2. Alternatively, the sovereign can choose the existing CAC (a supermajority of creditors in each individual bond series). The supermajority remains at 75 per cent and thus this option still retains the potential for a creditor who owns in excess of 25 per cent of a specific bond series to block that series from a restructuring and demand full payment.

The decision on whether to adopt ICMA's revised clauses will be made by sovereigns when they issue new debt. However, early signs are positive, with Kazakhstan issuing the first foreign-law bonds to include the clauses in early October 2014 with the clauses having little impact on pricing amid strong investor demand (Roy 2014). Mexico and Vietnam have also recently issued bonds containing ICMA's revised clauses, and innovatively Mexico has removed any ambiguity implied by the Latin phrase pari passu by replacing the Latin words with the English equivalent ‘equally’.

Supporting the adoption of ICMA's proposals, in October 2014, the IMF Board stated its intention to encourage its members to use these provisions in foreign-law bonds. G20 Leaders at the Brisbane Summit in November 2014 also called for their use and encouraged the international community and private sector to actively promote their use. Euro area countries are expected to continue with their current CAC model, which was implemented in January 2013 and includes an aggregation CAC with a two-limb voting structure.

The IMF estimates that there are around US$900 billion worth of foreign-law sovereign bonds outstanding, and around 29 per cent of these have maturities greater than 10 years. This means that it will take a number of years for the existing foreign-law sovereign debt stock to be completely replaced with bonds containing the new clauses, and holdout creditors could continue to frustrate debt restructurings for some time yet. The IMF has flagged some possible options if holdout creditors continue to be successful in frustrating future debt restructurings. For holdout creditors with New York law bonds, the United States could be encouraged to amend its Foreign Sovereign Immunities Act of 1976 to prevent rulings like that granted against Argentina from being granted by US courts in the future. A broader proposal could be for countries to undertake voluntary bond swaps to swap existing foreign-law bonds for bonds with the revised clauses.

Problems with Debt Restructurings Occurring ‘Too Little, Too Late’

The other weakness to the current restructuring framework is that recent restructurings have often occurred long after the time at which the debt was assessed as being unsustainable. And even where debt was restructured, in many cases debt sustainability was not restored, necessitating subsequent restructurings.

The IMF's experience with Greece's first assistance package in 2010, and its subsequent debt restructuring in 2012, was a pertinent example. Despite Greece receiving substantial financial assistance, the assistance program was unable to restore Greece's debt sustainability and an IMF evaluation of the program concluded that an ‘earlier debt restructuring could have eased the burden of adjustment on Greece and contributed to a less dramatic contraction in output’ (IMF 2013a). Other recent debt restructurings have also occurred long after the IMF had noted in its surveillance that debt was unsustainable, including Belize where debt was assessed to be unsustainable in 2005 but a restructure did not occur until 2007, and St Kitts and Nevis where debt was judged to be unsustainable in 2006 but a restructure did not occur until 2012 (IMF 2013b).

For the IMF, restructurings occurring too little, too late are a concern since the IMF effectively serves as a lender of last resort to countries experiencing debt distress. Delays to restructuring a country's debt mean that IMF resources are being used to meet the unsustainable debt repayments of a country – effectively bailing out private creditors. This presents two problems. First, it raises concerns around moral hazard to the extent that it may increase the willingness of private creditors to lend to countries with questionable debt sustainability in the knowledge that they have been repaid in the early stages of an assistance program in the past. Second, although debt may be assessed to be sustainable, due to the difficult and subjective nature of such an assessment, there is the potential for IMF funds to be used to meet unsustainable debt obligations, which is against IMF policies on lending.

In response, the IMF is currently considering proposals to reform the IMF's ‘exceptional access’ lending framework to address restructurings occurring too little, too late with the aim of reducing the cost of restructuring to the sovereign and its creditors as a whole (IMF 2014a). It is proposed that when the sustainability of a country's debt is uncertain (i.e. it cannot be determined to be sustainable or unsustainable with a high probability), the IMF would require a country to negotiate with its creditors to extend the maturity of its debts (termed a debt reprofiling) as a condition for IMF lending. The maturity of a country's debt would be extended by 1–3 years, buying time for a country to implement corrective policies, while resulting in only modest declines in the net present value of creditor claims because it does not reduce the undiscounted value of a country's debt stock. For the IMF, a reprofiling would preserve resources that would have otherwise gone to repay private creditors. In situations where debt was subsequently determined to be unsustainable, a full debt restructuring would then be required. The IMF staff also propose removing the systemic exemption clause that had allowed the IMF to lend to a country without a high probability of debt sustainability, if there was a high risk of systemic spillovers. Prior to the systemic exemption clause, if a country's debt was not sustainable with a high probability, the IMF required a country to undertake a debt restructuring that was sufficient to restore debt sustainability to a high probability.


An effective sovereign debt restructuring framework should allow for an orderly restructuring that is sufficient to restore debt sustainability at the lowest cost to a sovereign and its creditors. However, reforms to the framework should ensure that the potential consequences of a restructuring remain severe enough so that countries have adequate incentives to manage their debt sustainably and thereby not push up the cost of borrowing for sovereigns. To ensure an orderly debt restructuring, holdout creditors should not be able to credibly threaten to obtain preferential treatment relative to creditors that participate in the restructuring or be able to frustrate a restructuring deal that had been agreed to by a supermajority of creditors. The changes to sovereign bond templates by ICMA should lead to a smoother debt restructuring process by reducing the ability and incentives for individual creditors to hold out from a debt restructuring. As sovereigns still need to negotiate a restructuring deal with a supermajority of creditors, there should still be adequate incentives for a country to manage their debts prudently. However, with a large stock of foreign-law bonds still outstanding, it will take some time for these reforms to become effective and there is a risk that the holdout creditor problem could persist into the near future.


The author is from International Department. [*]

Over the past decade, 14 countries have undertaken a total of 18 debt restructurings. These countries are Argentina (2005, 2010), Belize (2007, 2013), Cameroon (2005), Dominica (2004), Dominican Republic (2005), Ecuador (2009), Greece (2012), Grenada (2005), Côte d'Ivoire (2010, 2012), Jamaica (2010, 2013), Nicaragua (2008), Paraguay (2004), Seychelles (2010), and St Kitts and Nevis (2012). [1]

The literature has used CACs to describe a number of contract terms designed to ease collective action problems. In this article, CACs refer to modification clauses that allow some percentage of creditors to approve a restructuring. [2]

See Richards, Flood and Gugiatti (2002) for a discussion of the SDRM proposal and the issues that were involved at the time. In September 2014, the United Nations General Assembly voted to establish a legal mechanism. However, a number of large developed countries voted against or abstained from the resolution (including Australia), and past initiatives to establish a legal framework have been unsuccessful. [3]

IMF (2014b) analysis suggests that foreign-law bonds constitute US$900 billion of sovereign debt (1½ per cent of global government debt), and it is estimated that 90 per cent of these are governed by New York or English Law. [4]

Each bond series is governed by a separate bond contract. Typically, each bond series has a different maturity date and coupon rate, which is specified in the term sheet of the contract. [5]

Complicating the interpretation has been the different formulations of the clause in sovereign bond contracts. [6]

This has not prevented holdout creditors from attempting to seize a country's assets for amounts due. In 2012, NML Capital, a holdout creditor in Argentina's debt restructuring, was successful in applying to Ghanaian Courts to impound the Argentine ship Libertad. It was later ruled by the UN International Tribunal for the Law of the Sea that the ship had sovereign immunity as it was a military vessel and Ghana released the ship in December 2012. [7]

Many sovereign issuers have waived sovereign immunity through clauses in their bond contracts and some countries have relaxed their sovereign immunity provisions so that they are waived for a sovereign's commercial activities. However, these waivers are unable to overcome a number of limitations including: non-commercial sovereign assets being immune; the problem of seizing a sovereign's assets when they are located outside of the governing law of the bond; and when assets are subject to another country's sovereign immunity provisions (Weidemaier 2014). [8]

Other courts have previously ruled in favour of the rateable payment interpretation. In 2000, a Belgian Court of Appeal ruled that the pari passu clause in Peru's New York law bonds should have the effect that a sovereign should render equal payments to all creditors in proportion to their claim. The Court also prevented Peru from making payments on restructured debt unless it also met its obligations to holdout creditors. [9]

The clause reads: ‘The Notes are the direct, unconditional and unsecured obligations of the Issuer and rank and will rank pari passu, without preference among themselves … Provided, however, that the Issuer shall have no obligation to effect equal or rateable payment(s) at any time with respect to any such other External Indebtedness and, in particular, shall have no obligation to pay other External Indebtedness at the same time or as a condition of paying sums due on the Notes and vice versa’ (ICMA 2014b). [10]


ICMA (International Capital Markets Association) (2013), ‘ICMA Sovereign Bond Consultation Paper’, December. Available at <http://www.icmagroup.org/assets/documents/Maket-Practice/Regulatory-Policy/Sovereign-Debt-Information/ICMA-Sovereign-Bond-Consultation-Paper-79801-5-863-v1-8-161213.pdf>.

ICMA (2014a), ‘Standard Aggregated Collective Action Clauses (“CACS”) for the Terms and Conditions of Sovereign Notes’, August. Available at <http://www.icmagroup.org/assets/documents/Resources/ICMA-Standard-CACs-August-2014.pdf>.

ICMA (2014b), ‘Standard Pari Passu Provision for the Terms and Conditions of Sovereign Notes’, August. Available at <http://www.icmagroup.org/assets/documents/Resources/ICMA-Standard-Pari-Passu-Provision-August-2014.pdf>.

IIF (Institute of International Finance) (2012), ‘Report of the Joint Committee on Strengthening the Framework for Sovereign Debt Crisis Prevention and Resolution,’ October. Available at <>.

IMF (International Monetary Fund) (2013a), ‘Greece: Ex Post Evaluation of Exceptional Access under the 2010 Stand-by Arrangement,’ IMF Country Report No 13/156. Available at <http://www.imf.org/external/pubs/ft/scr/2013/cr13156.pdf>.

IMF (2013b), ‘Sovereign Debt Restructuring – Recent Developments and Implications for the Fund's Legal and Policy Framework’, IMF Policy Paper, April. Available at <https://www.imf.org/external/np/pp/eng/2013/042613.pdf>.

IMF (2014a), ‘The Fund's Lending Framework and Sovereign Debt – Preliminary Considerations’, IMF Policy Paper, June. Available at <http://www.imf.org/external/np/pp/eng/2014/052214.pdf>.

IMF (2014b), ‘Strengthening the Contractual Framework to Address Collective Action Problems in Sovereign Debt Restructuring’, IMF Policy Paper, October. Available at <http://www.imf.org/external/np/pp/eng/2014/090214.pdf>.

Richards A, D Flood and M Gugiatti (2002), ‘Recent Proposals for Reform of Sovereign Debt Restructuring’, RBA Bulletin, August, pp 61–71.

Roy S (2014), ‘Kazakhstan Breaks 14-Year Duck’, International Financing Review, 11 October to 17 October, Issue 2054.

Scigliuzzo D (2014), ‘RPT-Acceleration Threatens to Make Argentina Holdout Crisis Messier’, Reuters, 13 August. Available at <http://www.reuters.com/article/2014/08/13/argentina-debt-acceleration-idUSL2N0QJ0UJ20140813>.

Weidemaier M (2014), ‘Sovereign Immunity and Sovereign Debt’, University of Illinois Law Review, 2014(1), pp 67–114.