JEFFREY A. FRANKEL^*

The first three-fifths of Patrick Bolton’s paper surveys corporate bankruptcy practices in various major countries. Then the remaining two-fifths attempts to draw lessons for the SDRM initiative currently being pursued at the IMF and for competing proposals of other ways to ease debt restructuring in emerging-market crises. This well-written paper is a welcome contribution in two ways. In the first place, it is a very clear and concise exposition of bankruptcy practices. Most writings on the subject are impenetrable to nonspecialists, because they toss around terms like debtor-in-possession and cram-down without defining them, although there do exist a few other good papers that explain the basics. But they tend to focus exclusively on U.S. practice or, at most, U.S. and U.K. practice. So this comparative study is particularly useful, and would be even if it stopped at that. But there is a higher purpose.

SDRM: Background

The second, or higher, purpose is to offer some much-needed guidance in trying to figure out which of the competing proposals for an international version—a statutory approach such as the SDRM, a more market-based contracts approach such as debt exchanges and collective action clauses (CACs), or some combination of these—is preferable. Even for those who think they have the answer to that question, there are still a lot of details. There is hope that a history and comparative analysis of how different countries handle bankruptcy at the domestic corporate level can offer useful insights for these ongoing decisions at the global sovereign level.

The biggest division is between bankruptcy practice in the United States and practice in the United Kingdom and elsewhere. At first, one might think that this issue of competing domestic standards is like many others: accounting standards, regulations for issuing securities, rules governing intellectual property rights (IPR) or drug testing and approval, or phytosanitary standards. In these cases, and many others, the United States often tries to persuade multilateral forums to extend the approach that it follows domestically to the international level. The motive can be either conviction that its way represents free-market virtue, lazy ignorance of competing standards, or conscious desire to give its firms a leg up (either because the domestic standard intrinsically suits its comparative advantage in some way, or just because its firms have already adapted to it and will thus be a step ahead of foreign firms, which will have to adjust). But the case of international bankruptcy procedures is not necessarily the same. The paper makes it clear that the SDRM is rooted in the tradition of the United States, and is an attempt to extend it to the global sovereign level. And yet the U.S. Treasury has not been very supportive of First Deputy Managing Director Anne Krueger’s proposal first made in a speech in November 2001, partly because of opposition from the U.S. investor community, which remains strongly opposed. The investment community, as represented for example by the Institute for International Finance (IIF), is now supporting instead the wider use of CACs. This would be an extension of the U.K. system, as CACs currently thrive under London law, not New York law.

It is worth briefly reviewing the background, for anyone who may not have followed this issue.¹ Begin with what Jeffrey Sachs wrote in 1994 (and in his Princeton essay published in 1995). Any well-trained economist, when confronting a question like appropriate government intervention in debt crises of developing countries, will ask, “where is the market failure?” Sachs’ answer was that there is no international bankruptcy court to perform for the case of international debt difficulties (especially on the part of sovereign governments) the restructuring function that Chapter 11 fulfills for domestic U.S. corporate bankruptcy cases. Sachs proposed creating one. At the time, such proposals seemed hopelessly unrealistic.

As an indicator of how unrealistic the proposal was considered, I will quote from Barry Eichengreen’s 1999 IIE book (pp. 92-93), which surveys proposals for reform. He lists what he sees as the substantive drawbacks, and then goes on:

Above all there is the political question of whether the creditor countries would be prepared to vest such formidable powers in the hands of an international tribunal of officials, … Even the kind of limited scheme floated by the Canadian government … is patently unrealistic. And if it is unrealistic to think that the IMF or another international entity could be empowered to impose a standstill on payments, it is pure fantasy to suggest that the Fund could be given the power to impose settlement terms on debtors and creditors.

To be sure, after the East Asia crisis hit, U.S. Treasury Secretary Robert Rubin called for fundamental reform of the financial architecture, and British Prime Minister Tony Blair even called for a “new Bretton Woods” (as President Clinton would have if his economic advisers hadn’t stopped him). But the steps that were taken by the G-7, G-22, and other bodies really amounted to tinkering with the plumbing more than changing the entire architecture. Then, in November 2001, Anne Krueger (2001 a,b) proposed that the IMF host a sovereign Chapter 11, now called an SDRM. And in April 2002 she proposed a modified version (Krueger, 2002). At the IMF Annual Meetings at the end of September 2002, the major shareholders (G-10) approved IMF plans to draw up a detailed SDRM proposal, simultaneously with a more contract/market-oriented approach that envisions wider use of CACs.

The argument that the absence of an international bankruptcy procedure is the main identifiable market failure sounds plausible.

My own view is that the international financial community should either pursue this approach aggressively, to see whether it can be made to work, or, if not, should stop talking about reform of the financial architecture. I know of no other proposals that are both sensible in conception and sufficiently ambitious to merit the title “reform of the architecture.”

The U.S. Version of Corporate Bankruptcy Law

Patrick Bolton identifies three key elements of corporate bankruptcy and reorganization institutions, from the perspective of U.S. history in particular:

• (i) a stay on individual debt-collection efforts and possible suspension of debt repayments (to solve the collective action problem of a “rush to the courthouse”);

• (ii) debtor-in-possession financing, which receives seniority over existing claimants, with the objective of keeping the firm operating in the interim if there is long-term value that might be lost by premature dissolution of the firm; and

• (iii) delegation of negotiations to creditor committees, with ways of preventing a dissenting minority from blocking a settlement, but ways that are sharply limited in the case of the United States. Each of these functions has possible analogues in international sovereign debt problems, but the fit is perhaps the closest in the case of the third. New York law requires that creditors give 100 percent approval to restructuring proposals outside of bankruptcy court. Attempts to restructure in such cases as Russia and Peru are seen as having been hampered by a minority of holdout creditors, and the goal of all these proposals is to prevent that from happening.

I was interested to learn of the theory that the heavy reliance of U.S. corporate finance today on bonds (rather than bank loans) might be attributable to their treatment in bankruptcy, particularly under the 1978 Bankruptcy Act. Bondholders are relatively well protected: A reorganization plan must be approved by each of the creditor classes (a two-thirds majority by ownership, within each). U.K. bankruptcy law, in contrast, gives greater protection to bank creditors.

Other Countries’Versions

In a final section the author concludes that U.S. corporate bankruptcy law and practice appear to be the most relevant for a comparison with sovereign debt restructuring.

The leading alternative to the U.S. system of bankruptcy by statute is the U.K. bankruptcy-reorganization procedure, called “administrative receivership.” Relative to the U.S. system, the British system tends to be creditor-controlled, to restructure debt more quickly, to avoid debtor-in-possession financing, and to favor liquidation, which makes bankruptcy less attractive to debtors. The U.K. system shares with the United States a low level of court involvement, compared to others. But I come away with the conclusion that the United Kingdom is less relevant as an alternative model of what an SDRM would look like than as a reminder of the system under which CACs thrive—a market-oriented alternative to SDRM.

The paper also reviews systems in other countries, rich and poor. We learn, for example, that Japan offers strong protection to secured creditors, as did Germany, at least until recently; France and India provide strong protection for workers and other stakeholders.

Lessons for SDRM?

Three big questions characterize an SDRM. They are: (i) how much administrative involvement should there be?² (ii) how comprehensive should debt standstill be? and (iii) how much involvement should other stakeholders have?

There is a discussion, in Bolton’s paper of political economy, which is very welcome, as we economists too often ignore the politics in trying to design a system, or at most invoke some political constraints in ruling out some proposal. Bankruptcy procedures were very controversial in U.S. politics in the first half of the nineteenth century. Yet they eventually became well established. The instinctive response to the SDRM is similar to Eichengreen’s: “whatever its merits in theory, it will never be accepted in practice.” So the history of how bankruptcy proceedings developed domestically seems a promising source of insight into political barriers and how to overcome them.

Specifically, Bolton points to the exemptions or opt-out provisions that were granted to states to reduce resistance to a federal system. It sounds promising at first; “states’ rights” was the strategy that led all 13 former colonies to go along with the U.S. federal Constitution in the first place, and any attempts at global governance can be usefully informed by this precedent. But I fear the precedent may be less useful for an SDRM than at first appears.

In the first place, while farmers and other debtors were an important political constituency in nineteenth century America, developing countries unfortunately have little vote in the design of the international financial architecture. (I say this even though some large emerging-market countries have finally been given a seat in groups like the International Monetary and Financial Committee (IMFC), the G-20, and the Financial Stability Forum (FSF) and even though I think that on the trade side developing countries collectively for the first time do have some bargaining power in multilateral negotiations.) The greater obstacle that must be overcome as part of this, or any sensible, reform is more likely to be the U.S. Congress than debtor countries. The Congress is not in favor of giving up any sovereignty or approving any multilateral treaties of any sort these days. It is not likely to agree to amend the IMF Articles of Agreement in the way suggested, or to allow an international body to overrule U.S. securities laws and creditor protection.

In the second place, perhaps the most obviously bad aspect of U.S. bankruptcy law is precisely that some states retain provisions whereby residents declaring personal bankruptcy can keep such valuable property as expensive houses, and even racehorses. Currently, some top executives of scandal-ridden companies are reportedly engaged in crash projects to build luxury houses in Florida, as a way of safeguarding their wealth if they have to file for bankruptcy.

One place where I think I may disagree with the author is the subject of corporate debtors shopping for the most friendly jurisdiction. Bolton has suggested this may be a good feature of the U.S. system, which could be usefully adopted at the global sovereign level:

A form of jurisdiction shopping could be contemplated under an international bankruptcy procedure as well. To encourage courts to respond to the needs of the contracting parties it may be desirable to allow for jurisdiction shopping by, say, letting existing bankruptcy courts also handle cases of sovereign defaults. Thus, sovereigns and/or their creditors could file for bankruptcy protection in U.S. bankruptcy courts (say, New York or Delaware), in U.K. courts, or in an ad hoc sovereign debt restructuring body… (p. 51).

When it comes to most kinds of standards, say for accounting or securities, a possible drawback of a system that allows jurisdiction shopping is a “race to the bottom.” The “pro” argument is that the competition will discipline those jurisdictions, giving them an incentive to maintain standards that are user-friendly and low in cost. This is what Bolton has in mind. Perhaps this works for corporate bankruptcy law; perhaps Delaware is indeed the popular state in which to incorporate because the Delaware court delivers timely and efficient bankruptcy proceedings. But it doesn’t strike me as a good idea. A court in Delaware or New York or Indiana is likely to be heavily influenced by the economic interests of residents of the United States. If the other side in a dispute consists of residents of far-off countries, their interests may not receive a fair hearing, even in fact, let alone in terms of the perceptions of the citizens of those countries. Raffer (1990) proposed that the international bankruptcy court should be located in a neutral country, a land that is neither an active lender nor borrower. Marcus Miller (2002) considers that neutrality argues in favor of locating a bond restructuring forum in any G-7 country other than the United States or the United Kingdom.

Differences Between Corporate and Sovereign Contexts

The author points out some differences between corporate bankruptcy and sovereign debt restructuring. The most important one is that sovereign states are not liquidated and governments are not replaced, at least not directly by creditors, as part of default proceedings. Even though Walter Wriston’s famous statement that nations don’t go bankrupt is generally said to have been proven wrong (in 1982) soon after it was made, it is true that sovereigns do not go out of business.³ As footnote number three in Bolton’s paper mentions, Chapter 9 for municipalities may be a more relevant precedent for sovereign restructuring. I wonder why Chapter 9 is not cited more often in this context, and Chapter 11 less.

One would think that the knowledge that the debtor can’t ultimately be put out of business would shift bargaining power from the creditor to the debtor (as noted in Bolton’s paper). The standard countervailing aphorism to Wriston’s statement is, “If you owe your banker $1 million, you have quite a problem; if you owe your banker $1 billion, he has quite a problem.” This doesn’t seem to be the way it works in practice, however. Debtors in fact have the most to lose. For some reason, debtor countries almost never explicitly default, telling the world they have no intention of paying, despite the apparent advantages of doing so. Perhaps it is those on the creditor side of the table who in fact have the greater bargaining power, because they are going to return to nice warm homes, whatever happens, while those on the debtor side may experience economic hardship (or jail, in some recent cases of ministers from Argentina to Korea to Indonesia).

Bolton lists as a further difference that sovereigns do not have to protect themselves against creditors racing to grab the debtor’s assets. I have a question here. I remember that in the early 1980s, a constant fear was that a miffed creditor would get a court order and “attach assets”: a Brazilian ship in a U.S. port, an Argentine plane at a U.S. airport, or Mexican oil in a U.S. pipeline. The standard creditors’ scramble for assets seemed a real danger, from what the lawyers told us. Indeed, such fears were listed as one of only several possible incentives that prevented countries from declaring outright default. And yet it never happened, neither in the debt crisis of the 1980s, nor in those of the 1990s. Does anyone know why?

Opponents of an SDRM

When considering a proposed major change in law, it is standard to ask for testimony from those most affected. In the case of the SDRM, that would be the creditors and lenders. It is interesting that both groups are on record as being opposed.

Lenders, of course, view it as a recipe for making default more common, at their expense. They also object to letting the IMF have seniority (which it de facto already has).⁴

While the immediate goal is to make restructuring run more quickly and more smoothly, I would say that the larger goal is to avoid the need for sharp recessions as part of the adjustment process and to do it without a large increase in the size of IMF loans—ideally, with a decrease in the size of IMF loans relative to recent large packages—for moral hazard and political reasons. If the IMF subsidy component were indeed reduced, it does not strike me as implausible that the net effect might be negative on the lenders, at least in the short run. If so, the IIF is doing its job in representing the interests of the creditors.

A number of borrowing governments are also opposed, and this is a more interesting observation. The stated reason is that to make ex post restructuring easier, and make default more frequent would reduce the ex ante availability of financing, in particular by raising the risk premium. I am not convinced by this line of argument. True, the total amount of financing going from high capital/labor (K/L) countries to low K/L countries is currently already less than the first-best return-equating optimum, and in that sense a reduction in capital flows would have a first-order negative effect on global economic welfare. But given the severity of the recurrent crises, in terms of lost output, it seems to me that we have been living in a third-best world (or worse). I know that in the Michael Dooley (2000) view, recessions are a necessary device to assure creditors that debtors will not default lightly, thus making international finance possible. But surely we can do better than that.

A reduction in the amount of debt may actually be desirable, a move from third best to second best, if it can be done in a way that is not too distortionary. I am not like some who favor a gratuitous increase in exchange rate volatility to discourage foreign currency borrowing, and I think that capital controls are a complicated matter, more likely to be abused than to be used intelligently. Nevertheless, if a reduction in the volume of flows occurs as a side effect of a plan that substitutes an orderly restructuring for the recession mechanism, then I don’t see this as a clear drawback. In other words, maybe a moderate reduction in capital availability is not such a bad outcome. The goal in architecture design should be to get some of the benefits of international capital markets, without periodic collapses in real economic activity. Surely a system that depends on loss of reputation as a deterrent to default is better than a system that depends on collapses in economic activity to accomplish the deterrent, and this is true even if the deterrent is triggered only infrequently.

There are other counterarguments, such as political infeasibility and the fact that, as currently envisioned, the SDRM applies only to governments defaulting on international obligations, which was not the main problem in most recent crises. But these arguments do not suggest that trying to launch an SDRM would do any harm. At worst it would have no effect.

Why, then, are borrowers opposed? Is it possible that if the question were posed to the borrowers collectively, they would be more supportive? When they speak as individual countries, they are afraid of the stigma they would suffer if they favored the SDRM (i.e., they fear the international community would suspect they anticipate default). Do the emerging-market groupings have any voices in the international financial forums who are not first and foremost ministers for particular countries who worry that their views will come back to haunt their home countries?

Lenders opposed, borrowers opposed—that would seem to be everyone involved. Is the IMF then wrong to propose an SDRM, and are G-10 governments wrong to support it? Lenders and borrowers don’t in fact exhaust the list. Citizens of borrowing countries (workers, taxpayers) and of lending countries (exporters, taxpayers) are also on the list of stakeholders, and they stand to benefit a lot if a way can be found to resolve debt problems without collapse of the country’s economic activity.

One would think that “G-10” includes the Bush administration. Footnote 15 in Bolton’s paper says, “Interestingly, the SDRM is a political issue with some Republicans siding with ‘Main Street’ against ‘Wall Street…’.” The implication is that the Bush administration supports the SDRM, but I don’t think it is clear what the position of the administration is. Its rhetoric has been variable on the SDRM, as in the area of international finance more generally. John Taylor, U.S. Treasury Undersecretary for International Affairs, has been ambivalent, at best, or outright opposed. In October 2002, a few days after the IMF/World Bank Annual Meetings, then CEA Chairman Glen Hubbard said the SDRM might not be necessary if market players could construct their own private debt rescheduling forum, a remark that was interpreted as encouraging Wall Street opposition. There is no reason to think the private market could on its own achieve a debt rescheduling forum.

The Alternative of Market Mechanisms

Coming out of the crises of 1997-98 there appeared to have been widespread agreement that some improvement on the regime as it stood ten years ago was necessary. Bolton’s paper discusses two “contractual” approaches: exchange offers (Roubini), which work under New York law, and CACs (Eichengreen), which to date have only worked under London law. It is often argued that bonds, the dominant financing vehicle of the 1990s, couldn’t be restructured as easily as loans, the dominant vehicle of the preceding cycle, because there are so many disparate bondholders. But people forget how many banks there actually were in the 1982 debt crisis, how heterogeneous they were, and how difficult it was to keep them all on board with packages that required them to “voluntarily” roll over their loans.⁵ There were more similarities between the two cycles than most people realize. If the resolution of the debt crisis in the 1980s had been speedy or without large economic losses, then the shift in the composition of finance would be a key point. As it is, it never worked very well when judged by the ultimate test: the decade of lost growth. So any reforms that make restructuring work better would be desirable.

The proponents of exchange offers (including Roubini, for example, in a paper jointly authored with me) argue that they have been used effectively in some recent cases: Ukraine, Pakistan, and Ecuador. (And furthermore, CACs could have been used, as the instruments explicitly allowed for them, but an exchange offer was preferred.⁶) Thus, it is argued, no major changes are needed. Bolton responds that these were small countries for which “the IMF’s stated policy, that it would not consider any bailouts without some concessions from other creditors and bondholders, may well have been credible.” An exchange offer failed in Russia in 1998, precipitating the full-fledged Russian default/devaluation. The explanation was that Russia, for geopolitical reasons, was the quintessential moral hazard play, that investors thought they could do better than the exchange offer because the IMF would have to bail Russia out. But if that was indeed the problem, then the fact that the IMF and the G-7 did ultimately pull the plug must have had a salutary effect on future investor calculations of this sort, which could support the Roubini claim that exchange offers are good enough. Furthermore, geopolitics were also relevant in Pakistan and Ukraine. Nevertheless, I am not sure that we should not be satisfied with the way the current system is working.

The proposal to expand the use of CACs, as an important reform in the international financial architecture, has been around for quite awhile. Eichengreen and Portes (1995) and the Rey Report (Group of Ten, 1996) were proposing greater use of CACs even before the Mexican peso crisis of December 1994. Debtors have been said to be reluctant to ask for CACs, for the same reason they are reluctant to support an SDRM: the stigma. For quite awhile, the recommendation was that G-7 governments should take the lead and issue bonds with such contracts, and only their refusal was holding things up. But both the Canadian and U.K. governments have now issued bonds with CACs. Initially they did little to generate enthusiasm in emerging markets. But in April 2003, Mexico and Brazil announced new bond issues carrying CACs.

Bolton gives four drawbacks to contractual or market approaches: (i) what happens to any other bonds and claims, beyond the specific bonds covered by CACs and exchange offers; (ii) enforcement of priority; (iii) DIP financing; and (iv) the high debt burden that remains even after restructuring. Fischer (2002, pp. 31-35) lists drawbacks with the current system of market-based private sector involvement (PSI), including difficulty of enforcing “voluntary” restructuring, and finds that the most profound difficulty is the absence of an accepted framework, like the proposed SDRM, wherein a debtor, in extreme circumstances, can impose a payments suspension while working to restore viability.

The solution may be to pursue SDRM simultaneously with market alternatives such as CACs because it is compatible to have both in effect simultaneously; CACs are thought to be easier to put in place, so there is a case for starting with them; and having discussions over the SDRM in process creates an incentive for the private financial community to pursue CACs as a preferred alternative.