Speech

Emerging Market Debt. What is the Problem? Speech by Kenneth Rogoff, Economic Counselor and Director, Research Department, IMF

January 22, 2003

1. Private debt flows to emerging markets are remarkable for their unpleasant side effects—wild booms, spectacular crashes, overindebtedness, excessive reliance on short-term and foreign-currency denominated debt, and protracted stagnation following a debt crisis. Emerging economies' governments need and wish to borrow—often for all the right reasons. However, they sometimes borrow more than is good for their citizens in the long run, and they are sometimes willing to take on excessive risk to save on interest costs. On the investor side, there is often a reluctance to hold instruments that would provide for more flexibility and risk sharing, such as GDP-indexed bonds, domestic equity, and local currency debt—in part, because of poor policy credibility and weak domestic institutions. The result is an excessive reliance on "dangerous" forms of debt, such as foreign-currency denominated debt and short-term debt, which aggravate the pain of crises when they occur. Support by the International Financial Institutions (IFIs) can help reduce the cost of crises in the affected countries, but in solvency crises lending by IFIs cannot address all the problems. Moreover, the availability of IFI support gives a false sense of security to investors, which magnifies booms in the run-up to the crisis (as in Russia in 1998).

2. There are many different factors and distortions that contribute to this nasty side of sovereign credit markets. I'd like to highlight four of them:

• Weak property rights and uncertain debt recovery processes. By their very nature, sovereign debt contracts are not enforceable in the same way that corporate debt contracts usually are. Moreover, sovereign debt typically lacks an enforceable seniority structure, which is an extremely useful device for private financial institutions.


• Political economy problems and short horizons of government in debtor countries. Typically, these generate a tendency to overborrow and assume excessive risks.


• Collective action problems among creditors, particularly in bond markets. This may be a problem in its own right, but as I shall argue below, it could also be viewed as a consequence of the first two problems.


• Finally, moral hazard resulting from the presence of crisis lending by the IFIs and creditor countries.

3. There is an obvious tension between the first two of the problems I have listed in paragraph 2. above. The lack of well-defined property rights tends to keep lending to emerging economies too low, while short time horizons, political economy factors and institutional weaknesses sometimes induce country governments to want to borrow too much. In fact, much of the time emerging countries seem to be "credit starved," and willing to accept debt placements at either extremely high cost or extremely high risk.

4. Painful debt crises can be viewed as the outgrowth of the tension. In the absence of better institutions and enforceable rights, the high cost of debt crises equilibrates emerging market countries' thirst for debt flows and investors' reluctance to lend. By making defaults extremely costly, risky forms of debt—particularly those that are prone to collective action problems—provide a measure of confidence to international investors. Without an explicit seniority structure, there is an incentive for investors to look for forms of debt that are the hardest to restructure (such as short-term debt), a subject explored in a recent paper by Patrick Bolton and Olivier Jeanne of the IMF Research Department. This enables sovereign borrowing in large amounts, often beyond what is socially desirable. In other words, default costs provide a punishment that in some sense substitutes for effective property rights at the international level. The problem is that it is a very poor substitute. Risky debt structures may be the market's response to the fundamental lack of property rights and the excess demand for sovereign credit, but they are a very inefficient response.

5. This is where the idea of an international insolvency regime comes in. There is a long and distinguished history of proposals to build such a regime (which I reviewed in a recent paper with Jeromin Zettelmeyer in the IMF Staff Papers) of which the one we are discussing today is certainly the most fully developed. International bankruptcy proposals are often viewed as sweepingly ambitious. But in some sense they are modest in taking the root problems of sovereign debt markets—the tendency to overborrow and the lack of enforceable property rights—largely as a given. Instead, they focus on mitigating the costs of debt crises and debt restructuring for both sides, but particularly for the debtor, by reducing the collective action problems that are at the heart of these costs. This will make the system less inefficient ex post—after things have gone wrong. But improving our framework for international debt workouts can also make the international financial system more efficient ex ante. For countries that have credible policies and that do not attempt to overborrow, reducing the length and cost of crises will mean lower capital costs, because investors can expect higher recovery values if thing do go wrong. It is no more likely to encourage bad habits among responsible governments than a new lung cancer treatment would encourage non-smokers to start smoking. In contrast, countries that do not have credible policies are likely to see their capital costs go up if creditors believe that default will now be less painful. They will end up not being able to borrow as much. That is not necessarily a bad thing, either.

6. An additional way in which an international solvency regime can help improve matters is through its effect on international financial institutions. Because there is currently no institutional mechanism for dealing with deep debt crises, there is sometimes pressure on international institutions to engage in large scale crisis lending even when such policies may be inappropriate—crises that are really more about solvency than liquidity. A system whereby those crises can be resolved in a fast and transparent way, with lower costs for borrowers and creditors alike, would enable IFIs to better resist pressures, and thus reduce any moral hazard associated with large scale crisis lending. While this is a by-product of most insolvency proposals, and not their main rationale, it may be an important by-product nonetheless.

7. An international insolvency regime can provide other efficiencies as well, beyond helping to more efficiently sort out borrowers. At the top of my list would be the introduction of a more explicit seniority structure in privately held international debt. At present, there is no such seniority structure, although markets often do make assumptions about the seniority of various international debt instruments, and countries occasionally do restructure debts selectively. To my knowledge, selective defaults have not yet occurred among different international law instruments, but Russia, for example, defaulted on some domestic-law bonds that were held in part by foreign investors while staying current on its Eurobonds—contrary to market expectations judging from the collapse in Russian international bond prices in 1998. Also, debtor countries (and G7 authorities) at times approach their bank creditors for a coordinated rollover of debts rather than trying with the more disperse and less coercible bondholders. This creates a bias toward debt that is hard to restructure, so that when things go wrong for a country, they can go wrong really quickly. Potentially, an SDRM could provide a legal framework that helps define an explicit seniority structure in international borrowing, improve transparency, and remove perverse incentives. A senior claim would be restructured through the SDRM only after all claims junior to it have been extinguished. The current proposal does not include this feature, but I surmise that such an extension would remain well within the scope of the institution currently envisaged.

8. Should a sovereign insolvency regime also cover domestic-law debts? This is a difficult question. For countries with well-developed legal systems, it may not be necessary. However, many countries have poor legal institutions—either because they are subject to interference by the executive branch, or because they are weak or corrupt—and these may be better off with a broad-reaching international institution. Regardless of where one stands in this debate, however, one thing is clear: even an institution that narrowly covers external debt may be helpful to domestic debt markets. Where domestic debt is significant, external creditors will not agree to a restructuring deal unless it specifies how domestic debt will be treated. Thus, the restructuring of both types of debt will be closely related, and domestic debt restructuring may gain in transparency and equity from the existences of a well-established mechanism for external debt.

9. Summarizing: I have identified four problems in international financial markets, Poor property rights—particularly on recourses available to creditors when defaults occur; a tendency to overborrow by developing countries; difficulties to implement collective action; and creditor moral hazard that may result from lending by the IFIs. Of these, the more fundamental ones are the first two. Of course, no workout procedure proposed to date addresses all of these problems, and any practical proposal will have its flaws. An international insolvency regime such as the SDRM will not remove all distortions from international capital markets, but it may go further in mitigating their consequences than is obvious at first. And, in principle, it could be developed to go even further, by addressing some property right issues more directly, such as the lack of explicit seniority among private creditors.