Sovereign Debt Restructuring Mechanism -- One Year Later, Address by Anne O. Krueger, First Deputy Managing Director, IMF

December 10, 2002

Sovereign Debt Restructuring Mechanism-One Year Later
By Anne O. Krueger
First Deputy Managing Director
International Monetary Fund
Presented at the European Commission
Brussels, Belgium
December 10, 2002

I. Introduction

Ladies and Gentlemen: Thank you for this opportunity to discuss the IMF's proposal for a sovereign debt restructuring mechanism (SDRM). The proposal, initially presented a year ago, has stimulated a number of interesting reactions. I was pleased to find evidence of support for a statutory approach—which the SDRM is—from an association representing foreign bondholders. Their statement said:

"During the autumn of last year, a conference of jurists and public men of various countries ... [discussed] ... the possibility of international agreements upon the principles of law which should determine the liability of sovereign states and foreign subjects in their relations to one another. ... There can be no question as to the advantage that would result from such an agreement."

That quote is from the annual report of the Corporation of Foreign Bondholders. I only wish the report was dated 2002 and not 1874. It goes to show that the debate over how best to carry out sovereign debt restructurings has been with us for a very long time.

The question of how to handle situations of unsustainable sovereign debt acquired increased urgency in the 1980s, when private capital flows-mostly lending from private banks to sovereigns-had grown enormously relative to official flows. Then, in the 1990s, private flows-bonds as well as bank lending-exploded. As I will describe later, this diversity in debt instruments has added to the challenge of how to tackle situations where sovereign debt has become unsustainable.

Today, I want to focus on three interrelated issues. The first is the implications—for the debtor and the creditors—of unsustainable sovereign debts. The second is a brief reminder of where the proposals for sovereign debt restructuring stand. I want to emphasize that the proposal is a "work in progress". We are engaging with many in discussion of the proposal, and it is reshaped in response to these discussions. And the third is how the implementation of these proposals would affect the volume and terms of capital flows and functioning of international capital markets.

II. Unsustainable Debt Burdens

Let me turn first to unsustainable debt burdens.

Private capital flows can greatly benefit source and destination countries. Lenders receive higher returns and can diversify their portfolios. Borrowers are enabled to invest more than domestic savings alone will allow. (There are, of course, major benefits in technology transfer, learning, and other areas of development as well, particularly from foreign direct investment, but these are not central to my concerns today).

Capital flows to sovereigns are a significant part of these flows and can finance necessary infrastructure and other public sector investments. When returns on these investments are high, economic growth accelerates, and debt-servicing is readily financed.

For most countries, most of the time, that is the end of the story. Occasionally, however, sovereign debt becomes unsustainable. And I want to focus for a moment on what that means, because that's the key to the SDRM proposal.

A sovereign debt is sustainable when the sovereign can, with reasonable policies, service the debt to an extent that the future debt-to-GNP ratio will ultimately stabilize or fall. It is unsustainable when, under any realistic set of policies and circumstances that can be envisaged, the debt-to-GDP ratio (or debt-to-export ratio in some cases) will rise without limit.

In the latter circumstance, debt will ultimately have to be restructured with a reduced net present value (NPV) relative to its face value. There are very important points to be made about these circumstances when they do arise:

  • First, once debt is unsustainable, the true NPV of the stock of debt is below its face value as creditors cannot be repaid in full. Further borrowing by the sovereign dilutes the value of claims of existing creditors.

  • Second, as debt-to-GDP mounts, real interest rates in the debtor country will rise. The sovereign may try to increase taxes or take other measures to service debt; but all of these measures are growth-reducing. Growth falters, and at some point real GDP itself can start falling, thereby increasing the debt-to-GDP ratio still further. A rising real interest rate at the same time increases interest costs of the debt.

  • Third, in circumstances where the real rate of growth of the economy plus the primary surplus as a percent of GDP is less than the interest payments as a percent of GDP, the debt-to-GDP ratio will grow indefinitely.

  • Fourth, as debt service rises, lenders see that debt sustainability is increasingly improbable, and at some point the supply of funds dries up, perhaps abruptly.

  • Fifth, and importantly, when debt is clearly unsustainable, there comes a point at which the true NPV of the debtor's primary surplus can be larger if debt is restructured because the economy's growth prospects can increase and real interest rates can fall. This is the so-called debt overhang and it is in these circumstances that an orderly and prompt restructuring can create value for both creditors and debtor.

This last point is crucial-restructuring can increase returns to all parties in cases where debt is unsustainable. But the circumstances in which this is the case are highly limited. Nothing being proposed is designed to lead to restructuring when debt is sustainable—the process is far too costly and painful. Indeed, it is so painful that sovereigns typically put off the day of reckoning beyond the point when there are any reasonable prospects of the situation correcting itself.

It is for these extreme circumstances that there are proposals on the table to allow a more orderly resolution of crisis arising from, or accompanied by, an unsustainable debt burden. The proposals would contribute to both crisis prevention and crisis resolution.

They would contribute to crisis prevention because private markets would lend less to countries with already high debt-to-GDP ratios. The proposals will also reinforce the fact that the official sector is not waiting on the sidelines to bail out imprudent creditors; this should help prevent crises by discouraging overlending and overborrowing.

In addition, the proposals would make crisis resolution more orderly and less costly, by providing incentives for countries to face up to their problems promptly, and in those cases where resolution is necessary there would be less debt to deal with.

The proposals would not, however, make restructuring an easy option, as the economic dislocation implied by even an orderly workout could be considerable (and there is no intention to reduce the incentives to service debt when it's a feasible proposition).

Let me now briefly review where the discussion and proposals stand.

III. Where We Stand

As you know, there are two broad classes of proposals on the table. One is to introduce collective action clauses (CACs), which are already accepted practice in some markets, more universally in sovereign debt contracts. The other is for a statutory framework—the SDRM—under which broader debt restructuring might occur.

The proposals are intended to ameliorate delays in restructuring unsustainable debt by addressing problems that currently arise in those circumstances. Currently, a creditor who holds out can scuttle an agreement acceptable to the majority and quite possibly obtain better terms for himself. That also serves as a disincentive for other creditors to organize. In short, there is a lack of incentives to resolve the collective action problem.

Developments in capital markets have amplified these difficulties. Debt restructuring was difficult enough in the 1980s, when you could bring together the holders of the majority of a country's debt by getting representatives of 15 to 20 banks around a table. Even so, restructuring took a long time, and growth did not resume for many highly-indebted countries until after the inauguration in the late 1980s of the Brady Plan, which could be described as an informal kind of debt restructuring mechanism.

As already mentioned, things are even more complicated now. Over the 1990s, countries have increasingly borrowed by issuing bonds. As a result, the share of bonds in outstanding public external debt owed to private creditors has nearly quadrupled to about 60 percent in 2000. Diversification in the sources of finance is a healthy development, but it can become a hindrance when negotiating a restructuring.

This is because bondholders can be even more difficult to coordinate than bank creditors. Unlike bankers, bondholders often do not have long-term relationships with the debtor. Bondholders also have greater incentives to sue delinquent creditors, because unlike banks they do not have to share the proceeds of litigation. The situation is further complicated by the growing variety of debt instruments and derivatives in play. Because of all these factors, it is difficult to get everybody in the same room and hammer out an agreement that everyone accepts as a fair solution.

The two proposals-CACs and SDRM-would help bring about faster and more orderly restructuring. Let me try and describe briefly how each would work.

CACs apply to individual bond issues. They would permit a specified super-majority of holders of the bond issue to agree to a restructuring that would be binding on all holders of that issue-that's what the clause does. That would then prevent hold-outs in individual bond issues, thereby facilitating any needed restructuring. A registry of holders, or trustee arrangements, could accelerate the process. The use of such CACs would be an improvement over the current system and the IMF is committed to promoting their use among its member countries.

The SDRM proposal goes further than CACs and could complement it nicely. It provides a mechanism which, when activated, would enable creditors and debtors to negotiate a restructuring, aggregating across instruments, and ratifying an agreement binding on all by a specified super-majority. As with a domestic insolvency law, it would aggregate claims for voting purposes and could apply to all existing claims. An independent and centralized dispute resolution forum would be established to verify claims, insure the integrity of the voting process, and adjudicate disputes that might arise. By providing the locus and secretariat for these activities, this forum would enable a smoother and quicker negotiation than now seems feasible.

Both approaches were endorsed at the Annual Meetings of the World Bank and the IMF by the International Monetary and Financial Committee, the IMFC. The IMFC noted that CACs and SDRM are complementary proposals and asked for further progress on both. At the Fund, we are supporting both approaches and have been instructed to develop a concrete proposal on SDRM for consideration at the Spring Meeting of the IMFC.

Over the past year, we have had extensive discussions with the creditor community, and have listened carefully to the concerns that have been expressed regarding the objectives and proposed design of the SDRM. These concerns appear to fall into four different — but related — categories.

  • The first appears to be directed at moral hazard: to the extent that the SDRM reduces the cost of restructurings, will it not increase their frequency?

  • The second set of concerns revolves around the impact on creditor rights: will not the SDRM significantly enhance the legal leverage of the debtor vis a vis its creditors?

  • Third, there are concerns regarding the role of the Fund: if the SDRM provides greater legal authority to the Fund, will it not use it to protect its own interests as a creditor?

  • Finally, there have been concerns regarding the impact of the mechanism on bondholders, and specifically whether the mechanism would reduce bondholders' recovery values in the event of a restructuring-and even that the impact of the mechanism would be to subordinate investors holding international bonds to domestic investors and official bilateral creditors.

We believe that these are important issues. We also believe that they are addressed in the most recent proposal that will be discussed by the Fund's Executive Board within the next two weeks, a proposal that has been developed in close consultation with market participants and members of the legal and judicial professions.

When this proposal is published — which should be by the end of the year — I hope that all stakeholders — including those within the creditor community — will recognize that the contemplated framework is designed in manner that will provide them with a number of benefits. At a minimum, I think they will realize that the SDRM is designed simply to facilitate an agreement between creditors and debtors — it is not to override such an agreement or to impose one.

The proposal is relatively detailed and specific and I would overwhelm you if I described all of it to you today. But let me identify some of the key principles that we have relied upon when shaping it.

First, the mechanism should only be used to restructure debt that is judged unsustainable.

The purpose of the SDRM is to reduce the costs of restructuring debt that creditors and sovereign debtors recognize will need to be restructured anyway. It should neither increase the likelihood of restructuring nor encourage defaults. In circumstances where creditors are of the view that the activation of the mechanism was not justified, the mechanism will enable them to vote to terminate it as soon as they are able to organize such a vote...

Second, any interference with contractual relations will be limited to those measures that are needed to resolve the most important collective action problems.

The principal feature of the mechanism is that it would allow a sovereign and a qualified majority of creditors to reach an agreement that would then be made binding on all creditors that are subject to the restructuring, paying due regard to seniority among claims and the diversity of creditor interest. Giving creditors the ability to make this decision on an aggregated basis will not shift the legal leverage from the creditors to the debtor; rather it would increase the leverage of creditors over potential holdouts and free riders, enabling an agreement to be secured more rapidly.

During the period between activation and the reaching of a restructuring agreement, we are not proposing that there be an automatic stay on creditor enforcement or a general suspension of contractual provisions. In short, the SDRM will not provide a debtor in default with the type of legal protection that one finds in the corporate insolvency context. Ideally, sovereigns with unsustainable debt will actually use the SDRM before they default — which is when there is still the greatest amount of value to be preserved but where collective action problems are the most acute. In these cases, the SDRM would be activated after the debtor has negotiated an agreement with a qualified majority — say 75 percent — of its creditors and would simply be used to make the agreement binding on the rest. In this sense, the closest analogy is the "pre-packaged" reorganization framework that is used so effectively in the corporate context. Again, we are trying to facilitate an agreement between the debtor and its creditors — not enhance the debtor's legal leverage.

The framework will be designed in a manner that promotes greater transparency in the restructuring process.

The mechanism will establish procedures that enable creditors to have adequate access to information regarding the debtor's general situation, including its treatment of all creditors — even those that may not be subject to the mechanism. This will require the sovereign to provide relevant information when the mechanism is activated and also when an agreement is proposed. This is a critical element of the framework: experience demonstrates that creditors will only be willing to make a decision in a timely manner if they enough information to enable them to conclude that their decision will be an informed one

The mechanism will encourage early and active creditor participation during the restructuring process.

In addition to providing for a creditor vote on the terms of a restructuring, the framework will enable creditors to play an active role at earlier stages in the process, including through the formation of creditors' committees. We recognize that such committees can play an invaluable role not only in the negotiations with the debtor, but also in resolving inter-creditor issues.

The integrity of the decision making process under the mechanism will be safeguarded by an efficient and impartial dispute resolution process.

Since the mechanism will aggregate diverse claims for voting and restructuring purposes, disputes are likely to arise as to the validity and value of these claims. Resolving these disputes in a fair, impartial and expeditious manner is critical to the success of the restructuring exercise.

Finally, the formal role of the Fund under the SDRM will be limited.

Although the SDRM would be established through an amendment of the Fund's Articles of Agreement, we are not proposing that it give the existing organs of the Fund any new legal powers. Drawing on the approaches that have been used in other organizations —and with the benefit of considerable consultation with experts in this area — we are confident that the Sovereign Debt Dispute Resolution Forum will be independent —and be perceived as being independent from the Fund's Executive Board.

Of course, the Fund will use its existing financial powers to create incentives for an appropriate use of the mechanism. For example, our newly revised lending into arrears policy will be implemented in a manner to ensure that the debtor engages in a collaborative dialogue with creditors before making a proposal.

In the final analysis, however, the framework should be designed to catalyze early and effective dialogue between the debtor and creditors-it should not increase the role of the Fund in this dialogue.

Let me turn now to the question of investors' recovery values. The SDRM is intended to address those hopefully rare circumstances in which debt is unsustainable, and where—as a practical matter—investors will not be able to collect the full contractual value of their claims. In such cases there is typically a window of opportunity between the time when it becomes apparent that debt is unsustainable and the onset of a full-blown crisis. Time is the friend of neither a debtor nor its creditors in such circumstances. The current system introduces delays that lead to costs borne by debtors and creditors that are unduly large. SDRM is intended to help reduce these costs. If by improving the mechanisms for restructuring we can bring forward the time when a debtor initiates the restructuring of unsustainable debts, and by addressing market failures associated with collective action difficulties, there is a realistic prospect of helping to limit economic dislocation and preserve the economic value of assets. This is in the interests of creditors and debtors alike.

Concerns have also been raised that the possible exclusion from the mechanism of the claims of Paris Club and domestic creditors would have the effect of subordinating the claims of investors holding international bonds. Let me underscore that the exclusion of certain claims from the mechanism does not imply that they would not need to be restructured. In fact in most cases it is likely that both domestic debt and the claims of Paris Club creditors would need to be covered by a restructuring both to bring the debt payments to a sustainable level, and to achieve intercreditor equity. The mechanism is designed in a fashion to facilitate the coordination of restructurings by various groups of creditors, whether within the mechanism, or in parallel with the mechanism. Finally, I would point out that a final decision has not been made to exclude Paris Club claims from the SDRM. We understand that this is an important issue for investors and the official community is exploring whether it would be feasible to include these claims under the mechanism, perhaps as a separate class.

IV. Imagining a World with SDRM

Let me now turn and ask what difference these proposals make to international capital markets.

Let me begin with the impact on capital flows. A more orderly framework for debt restructuring will provide investors greater incentives to differentiate between risks, thereby making it easier for countries with sound economic policies to attract capital. (Clearly, SDRM is not an issue for countries with an investment-grade rating.) Countries with weaker policies may initially find it more difficult, but they will quickly realize that better policies are needed to attract capital. So what would happen over time is that the incentives for countries to follow sound policies will increase. And this, in turn, should increase the attractiveness of emerging markets as an asset class and therefore the flows of capital to them. And, while policies are weak, lenders would have greater incentives to limit their lending, thus providing stronger and earlier incentives for policy reforms.

Next, what about the impact of SDRM on the cost of borrowing for emerging markets? Pursuing the line of argument just made, the borrowing costs for countries with sound economic policies should fall as it would pay investors to be more active than in the past in differentiating among countries. So countries with sound policies would have a larger pool of foreign capital to draw on and better terms on which to borrow. Moreover, for a number of reasons, there should be over time a reduction in the borrowing costs for the asset class as a whole.

First, if more countries start to follow sound policies-because there would be increased incentives to do so-the increased supply of capital to emerging markets as a whole would lower borrowing costs. Second, SDRM should increase investor recovery rates by shortening the negotiation process and putting in place an efficient workout procedure in which a lot of power is given to the creditors collectively; this in turn would lead to a reduction in borrowing costs. Finally, with the SDRM in effect, the expected size of the restructuring—the haircut, if you will—should also be smaller; that too should increase recovery rates and lower borrowing costs.

What I've said so far is theory. What empirical evidence can we bring to bear on the impact on the volume and terms of capital flows? One obvious source of evidence is the rich history of national bankruptcy laws from which the SDRM proposal drew some of its inspiration.

U.S. bankruptcy procedures, for instance, were an outgrowth of attempts to deal with numerous railroad failures in the 1850s at a time when there were no formal bankruptcy institutions. It was obvious that liquidating the railroad, and giving each creditor a piece of the track, was not a solution that would get anyone anywhere.

So procedures evolved that tried to maintain a railroad's going-concern value during the restructuring process. This required, among other things, new funds to keep the railroads running and pay suppliers; and this has since become known as debtor-in-possession financing.

A second source of evidence is the anecdotal and more systematic empirical evidence on the impact of collective action clauses on borrowing costs. CACs have already existed in British trust-deed bonds, and there is no evidence that they raise borrowing costs. Likewise, several data and news services that report in detail on each new bond issue never explain the pricing of a bond in terms of the presence or absence of collective action clauses. More systematic econometric investigations, including a study presented at the IMF's annual research conference last month, have also failed to uncover any systematic impact of collective action clauses on borrowing costs for the asset class as a whole.

V. Conclusion

Let me now conclude. Unsustainable debts have to be restructured, one way or the other. The only question is at what cost. Adam Smith wrote that "when it becomes necessary for a state to declare itself bankrupt ... a fair, open and avowed bankruptcy is always the measure which is both least dishonorable to the debtor, and least hurtful to the creditor."

A more orderly process will be to almost everyone's benefit. The fact is that both the debtor country and its creditors stand to gain from a restructuring of unsustainable debts before the country has exhausted its reserves and condemned itself to a deeper economic downturn than necessary. At present, the threat of a disorderly workout means that the value of creditor claims falls more sharply on the secondary market when a country gets into trouble than it would likely do in a more predictable environment. A framework that allows creditors to preserve better the value of their claims and debtors to minimize output losses during the restructuring period helps both creditors and debtors. I have no doubt that they will increasingly come around to that view.

As I noted earlier, SDRM is part of the effort for better crisis prevention and resolution. No single instrument, SDRM or any other, is suitable for all crises. The SDRM is one of a set of instruments and policies that, taken collectively, will enable the IMF better serve all our member countries. So in addition to the work on SDRM, we will continue our work in these other areas as well, including finding ways to reward member countries that are pursuing sound economic policies.

Thank you very much.


Public Affairs    Media Relations
E-mail: E-mail:
Fax: 202-623-6278 Phone: 202-623-7100