Executive Summary

The recent annual meetings of the IMF and World Bank highlighted two proposals to reform the process for restructuring international sovereign debt: collective action clauses (CACs) and a Sovereign Debt Restructuring Mechanism (SDRM), the so-called statutory approach to reform. Neither the IMF nor its critics believe that these measures would be sufficient to minimize the likelihood of sovereign debt crises or to ensure that they are efficiently resolved. Nonetheless the proposals have attracted widespread attention and deserve careful analysis. The Shadow Committees believe that the IMF proposals go both too far and not far enough. They go too far with respect to the immediate reforms suggested for the sovereign debt resolution process. We favor a more gradual approach that begins by strengthening existing contractual means for resolving debt problems. They fail to go far enough with regard to reforming the IMF's policies that give rise to incentives to postpone the recognition and resolution of unsustainable debt.

A central objective of recent reform proposals is to alleviate conflicts among creditors that often arise during the debt renegotiation process. In particular, holdouts by a minority of creditors can delay debt resolution and prevent sovereign debtors from regaining access to international credit sources. The Shadow Committees believe that this problem can be addressed by the general adoption of CACs, which would impose majority rule on bond creditors. The Committees further believe that this goal can be achieved by voluntary actions on the part of creditors and debtors supplemented by the use of incentives for creditors and debtors to adopt CACs. At this time, the Committees do not endorse the adoption of the statutory approach to debt resolution. However, in the event that the use of CACs alone proves inadequate after a period of trial, a statutory approach may have to be reconsidered.

Guiding Principles

The ideal debt restructuring reform would seek to achieve the following objectives:

1. discourage countries from overborrowing and creditors from overlending;
2. reduce the likelihood of sovereign debt crises by reducing moral hazard, inter alia by ensuring that creditors, who were compensated ex ante to bear risk, share appropriately in the ex post losses;
3. when debt burdens become unsustainable, reduce the recognition lag during which economic conditions deteriorate to the disadvantage of the debtor country and creditors before the restructuring process is begun;
4. once the debt restructuring process has begun, ensure a prompt resolution; and
5. reach a resolution that will enable the country to regain access to stable capital flows.

SDRM and innovations in CACs mainly address the fourth criterion, but don't adequately address the others. The likelihood of sovereign debt crises and delays in recognition depend on several additional factors, including importantly, IMF lending that supports unsustainable policies. Attention to improved debt restructuring mechanisms should not distract attention from the importance of reducing the recognition lag and the role the IMF has played in unintentionally enabling undesirable delays in recognizing and dealing with unsustainable debt.

Creditor Coordination Problems

One of the central objectives of the IMF proposals is to deal with collective action problems among creditors that arise in the debt renegotiation process. This is often characterized as the holdout or rogue creditor problem, in which a minority of creditors delay resolution until their demands are met, to the detriment of other creditors and the debtor. The CACs proposal deals with this problem directly by binding all bondholders to the will of a super-majority.

CACs would serve as an alternative to existing ex post mechanisms (e.g., exit consents, defined below) used by sovereigns to encourage creditors to share in the outcome of the renegotiation process. One advantage of CACs over existing mechanisms is a potentially beneficial impact on IMF lending; by making the renegotiation process more predictable, CACs may also reduce pressure on the IMF to support unsustainable sovereign borrowers.

Another advantage of CACs is the elimination of potential legal uncertainties and delays. Current legal uncertainties may limit the usefulness of exit consents. Exit consents are contractual amendments to existing bond contracts accepted by a simple majority of bondholders who have agreed to exchange those bonds for renegotiated debt. The attraction of holding out as a minority bondholder is reduced if the old debt now contains undesirable contractual amendments. There is, however, some uncertainty about which amendments are permissible. CACs would avoid those legal uncertainties.

The Shadow Committees believe that the impact on the cost of issuing debt from including CACs in bonds is unlikely to be significant. Furthermore, the adoption of CACs would not foreclose other useful innovations in private contracting that could facilitate debt resolution.

The Case for Encouraging CACs

CACs are not a new idea. In fact, they already exist in international bonds issued in the United Kingdom, Luxembourg and Japan. If CACs are attractive, why are they not generally adopted voluntarily? To the extent that the widespread use of CACs would reduce the prospects for an IMF bailout (as we have argued they might), neither borrowers nor debtors may want to adopt them. If bonds containing CACs are more costly to issue, an argument often made by sovereign debtors and creditors, two other reasons may be relevant. First, the benefits of a more orderly renegotiation process may extend beyond the issuing debtor and its creditors to include reduced cross-country contagion. Those benefits cannot be captured by the issuing debtor or its creditors and therefore might not encourage a country to adopt beneficial CACs. Second, national leaders facing short-term political pressures might be unwilling to trade-off slightly higher current debt service costs for lower prospective renegotiation costs.

This suggests that there may be a benefit from official encouragement of the adoption of CACs. Indeed, some have suggested that adoption be made mandatory. For example, one mechanism could involve amendment to the Articles of Agreement of the IMF to require appropriate changes in the law of each member nation (a more modest version of the IMF's proposal to amend the Articles of Agreement to establish an SDRM). A more moderate approach would be to require that countries that benefit from IMF support adopt CACs, or to provide access to lower-rate multilateral financing for countries that adopt CACs voluntarily. It would also be desirable to remove the existing customary impediments to CACs in the U.S. Contrary to popular belief, CACs are permissible in sovereign debt contracts under the U.S. Trust Indenture Act of 1939. Official encouragement of the adoption of CACs in sovereign debt contracts issued under U.S. law may be useful in overcoming customary resistance to CACs.

Do We Need the Statutory Approach?

If CACs are adopted for all new issues of debt, there still remains a serious problem regarding the outstanding stock of debt that does not contain CACs. Transitional issues, however, should not derail desirable long-run reforms. One of the stated motivations for the IMF's SDRM proposal is that it would solve the transitional problem by encompassing new debts and preexisting debts within the same resolution mechanism. But there are other approaches that would also accomplish that objective. For example, an alternative possibility would be to swap non-CAC for CAC debt. Various possible G7 initiatives could encourage, or even subsidize, such swaps if the transition problem were deemed sufficiently important.

Another rationale sometimes offered for SDRM is the need to insulate sovereigns from adverse court judgments during the renegotiation process. This concern is overstated. Recent episodes indicate that sovereigns are able to protect themselves from such judgments, with the possible exception of transactions in the clearing and settlement process. But this problem could be remedied more directly by legislation to protect sovereign debtors from attachment during the clearing and settlement process, comparable to the protection of wire transfers found in the U.S. Uniform Commercial Code for firms in bankruptcy.

Advocates of the SDRM argue that a key advantage of the statutory approach is the ability to coordinate the resolution of many different debts (bonds, bank loans, swaps, etc.) through statutory rules that define the relative power of creditors (e.g., veto power of creditor classes over restructuring plans, as proposed by the IMF). Advocates see this as a more orderly alternative to the raw bargaining that takes place among sovereigns and creditors during a renegotiation.

It is not clear whether the statutory approach would hasten the renegotiation process. For example, the vesting of veto power over restructuring plans in different classes of debt holders could produce a slower process in comparison to the result produced by a debt swap organized by the sovereign in which the relative positions of creditors are determined by the sovereign's judgment of what would work best. It is also far from clear where it would be best to vest the oversight authority over the proposed statutory process.

Given those uncertainties, and given the limited experience with renegotiation of international bonds over the past four years, it is too early to conclude that the statutory approach is warranted. Unlike efforts to increase the use of CACs, there may be important irreversibilities in establishing the statutory approach, as doing so might foreclose desirable alternatives from evolving within the contractual approach. On balance, we believe that it would be best to leave the statutory approach on hold for the time being, preserving it as an option to consider if the strengthening of the contractual approach proves inadequate.

Additional Reforms

Is more needed? We believe so, because we remain skeptical that either our proposal for strengthening CACs or a statutory process would address the important problem of the delayed recognition of unsustainability. That problem has more to do with incentives to delay on the part of sovereigns (who have short-term political motives) and creditors (who see the option of receiving an IMF bailout as a basis for resisting renegotiation). Thus, the G7 and the International Monetary and Financial Committee of the IMF should focus more effort on establishing proper incentives within the IMF to ensure that IMF loans do not delay debt resolution.

Indeed, part of the reason for uncertainty about the efficacy of the contractual approach today is that IMF actions in the past have made it very hard to observe how creditors and sovereigns would behave in a world where they really were left alone to renegotiate. A further advantage of IMF reform would be that it would help us determine whether a statutory approach was truly needed in future.

Other reforms outside the IMF may also be desirable for altering creditors' incentives to act in ways that would reduce capital flow volatility in emerging market countries. For example, a proposed reform to the Basel Accord would aggravate the existing distortion in international capital regulation that encourages shorter maturity lending to emerging market economies, thus exacerbating capital flow volatility. On the contrary, the Basel Accord should be modified to remove that distortion. (See Latin American Shadow Financial Regulatory Committee Statement No. 2, April 2001).

Conclusion

Given the rapid pace of innovation in which market participants are developing new approaches to dealing with the inefficiencies in the debt renegotiation process, we believe that an incremental approach is best. Consequently, CACs should be adopted. But premature adoption of the SDRM might be counterproductive or foreclose other beneficial adaptations.