Reforming the International Monetary and Financial System

Comments: Private Sector Burden Sharing

Alexander Swoboda, and Peter Kenen
Published Date:
December 2000
  • ShareShare
Show Summary Details
Martin Wolf

Barry Eichengreen’s paper is exactly what one would expect from him—lucid, balanced, and thoroughly reasonable. I admire its virtues. I suspect that his principal conclusion—that private sector burden sharing is very difficult or, as Americans like to say, there is no “silver bullet”—is correct. For all that, his paper has flaws, partly of attitude, but more of substance. To assume, as Eichengreen does, that debt is owed by the public sector is to omit the most significant feature of the Asian crisis.

The complaint I have with its attitude is that it is altogether too reasonable. The starting point for serious discussion of the international financial system, as it affects emerging market economies, must be recognition of failure. The system has consistently failed to achieve what is most important: to deliver a sustained net flow of resources into capital-short economies. Far from achieving this, whenever significant current account deficits emerge, panic and disaster ensue. Over the past two decades, only flows to developed countries, above all the United States, have been crisis-free. This is perverse.

Before the last round of financial crises fade from memory, it is worth recalling the scale of the disaster. A system that forces countries to make current account adjustments of not far short of 20 percent of gross domestic product in two years, as it did in South Korea and Thailand, is guilty of malfunction. A system in which net lending by commercial banks can shift by 10 percent of precrisis GDP between 1996 and 1998, in the case of the five most afflicted Asian countries, is guilty of malfunction. A system that brings fear of worldwide recession out of a cloudless sky is, once again, guilty of malfunction.

The only sensible response to such events must be a mixture of rage and fear—rage that it did happen, fear that it may happen again. If outside observers do not express some passion, why should we expect the finance ministers of the Group of Seven to do more than respond with world-weary minimalism? After all, it is neither their economies nor their electors who are threatened.

Yet there is also an important matter of substance that Eichengreen ignores. This paper is essentially concerned with sovereign liabilities, explicit and implicit. The question it confronts is how to make possible the alteration of debt contracts that have been made by a sovereign—or behind which a sovereign stands—in the event of adverse circumstances. But this focus ignores what seems the most interesting and, in the long term, most significant feature of the Asian crisis, namely, that it was a crisis in the private sector.

As Eichengreen notes, there is a perfectly general problem with any procedure that forces alterations upon creditors. If creditors know in advance that this could happen—particularly that the global financial authorities, including the International Monetary Fund, are likely to help debtors reschedule or default—they are likely to run away still faster, thereby precipitating the crisis the world wishes to avoid. The danger is particularly great, Eichengreen concludes, if there are rules governing such modifications of terms. As he notes, “rules specifying the modalities and circumstances in which creditors will be bailed in (especially rules that subject investors to serious haircuts in order to redress the moral hazard problem) run the risk of precipitating additional crises.”

Yet this objection also applies to private bankruptcy procedures. This has not prevented countries from evolving “rules specifying the modalities and circumstances in which creditors will be bailed in” when private debtors are insolvent (or illiquid). So what is special about sovereign lending? Why do we persist with the myth that sovereigns do not default or are, at least, not supposed to do so (although, of course, they have defaulted, do default, and will, no doubt, continue to default)?

The explanation is that it is very difficult to devise bankruptcy procedures that they cannot manipulate to their own advantage. Creditors are unable to seize the assets of a sovereign, other than those held abroad. Since a sovereign’s principal asset is merely its ability to tax its citizens, this limitation appears self-evident. Only in very exceptional circumstances can creditors obtain a lien on this “asset.” This inability to construct acceptable bankruptcy procedures creates a contradiction in the sovereign debt market. The prices at which debt trades show that lenders often know default has a significant probability and that they are being rewarded for taking the risk, but there can, it is argued, be no institutional recognition of this fact, without greatly increasing its probability.

For these reasons, Eichengreen’s conclusion is akin to despair. Some things can be done to facilitate renegotiation of contracts: the inclusion of majority-voting and sharing clauses in bond contracts is one recommendation long supported by the author. But there is no question that the principal lesson a reader would draw from this paper is that it is already too late by the time crisis occurs. All that is left then is the messy case-by-case negotiation with which the world is so familiar.

I am not quite as despairing as Eichengreen is, for two reasons. First, there seems to me to be merit in introducing recognition into the market that sovereign borrowers will default—or require rescheduling. Too often these risks have been underappreciated and lending has been correspondingly excessive. Second, while the introduction of any bankruptcy procedure increases the risk of behavior likely to cause bankruptcy, borrowers still suffer serious costs from rescheduling or default in terms of lost access to credit markets. For both reasons, I still consider it possible, in principle, to develop a procedure that allows some international institution to recognize the reality of insolvency (or illiquidity) and require restructuring of sovereign liabilities.

The true obstacle to such a procedure is political. It would involve greater interference in financial markets and debtor country policies than the governments of the world are ever likely to grant. The conclusion I draw is that we will not find any rules-governed way of handling sovereign liquidity or insolvency crises, not because it is conceptually impossible, but rather because it is politically unacceptable. For this reason, I agree with Eichengreen’s somewhat pessimistic conclusions. But this does not mean it is impossible to do anything both to reduce the chances of economy-wide crises and improve upon current methods of handling them.

We do, after all, have perfectly good ways of handling insolvency in the private sector. It is called bankruptcy. In the East Asian case, governments were not heavily indebted. Their private sectors, including their banking systems, were. The obvious question then is why standard domestic insolvency procedures did not resolve the problem of private sector burden sharing.

There appear to be five answers to this question. First, the countries did not have well-developed bankruptcy procedures. Second, the governments were unwilling—or unable—to let much of their domestic entrepreneurial class disappear into bankruptcy. Third, so large a portion of the private sector was insolvent, following the collapse of the exchange rate and the excessive foreign currency indebtedness, that no conceivable bankruptcy procedure could have coped. If big enough, a private sector crisis is inevitably a nationwide one. Fourth, bankruptcy was concentrated in the financial system, which can never be allowed to collapse. Fifth, some of the external creditors were so powerful that governments could not risk offending them by letting bankruptcy procedures have the normal effects on their loans.

I would like to comment on these answers. First, it is possible for countries to develop bankruptcy procedures; it is not easy but it is possible and must be done with the greatest possible urgency. Second, if countries are unwilling to accept the risk of crisis-induced mass bankruptcy, they need to consider the case, and the conditions, for opening their private sectors to international borrowing with great care. Third, once the exchange rate crisis is over and the currency has, as it normally does, recovered from its low point, the scale of the bankruptcy should fall. This should make it easier to let a standard procedure work. Fourth, there may be, even so, too much bankruptcy for any case-by-case procedure to handle efficiently. It may be necessary, therefore, to force debt-equity conversion across a number of bankrupt private companies at once. The need for development of such emergency economy-wide bankruptcy procedures may be an important lesson from this crisis.

The final comment is that the financial system does indeed create special difficulties. Rightly or wrongly, there is now a consensus that no government can allow the core institutions of its financial system to collapse, even if depositors do benefit from limited insurance. This resistance is not just to the collapse of the financial system as a whole, but to the collapse of any big institution. I am not persuaded that this is absolutely necessary but we must accept the full implications of this consensus. It means that borrowing by core domestic financial institutions is potentially government borrowing. This borrowing should be regulated and controlled as such. To put the point bluntly, financial institutions are not private in any meaningful sense. The profits and a portion of the losses are private. Beyond that, losses fall on the public. So liabilities that are private ex ante become sovereign liabilities ex post. Once that has happened they are subject to all the difficulties with sovereign borrowing described by Eichengreen.

The conclusion I draw from all this is that precisely because there is such huge difficulty in managing sovereign foreign currency insolvency or illiquidity (which may have to include the foreign currency indebtedness of an only notionally private financial sector), there should be less of it. It is difficult to see why sovereigns need to borrow heavily in foreign currency. Let them either balance their budgets or borrow in domestic credit markets. That would have the advantage of stimulating development of the domestic credit markets. But most of the borrowing could be private. This would be facilitated by privatization of the bulk of a country’s infrastructure.

Most borrowing and lending would then be to, and from, private parties subject to normal bankruptcy procedures. Burden sharing or “bailing in the private sector” would be automatic. Difficulties would arise, therefore, only to the extent that government policies generated excessive foreign currency borrowing, followed by exchange rate collapse, or permitted guaranteed financial institutions to borrow freely abroad. The combination of floating exchange rates with tight regulation of core financial institutions should help avoid this.

If there were this separation between foreign currency borrowing by the private sector and the activities of the government, it would also be easier to provide the latter with foreign currency support during a crisis, without being accused of bailing out the former.

Creditors of bankrupt private borrowers would suffer losses. But international support would ward off mass private sector insolvencies generated by developments beyond the control of individual borrowers and lenders—in particular, by huge exchange rate declines, very deep recessions, or both.

My conclusions are four. First, I agree that there is no neat and acceptable solution to the problem of sovereign foreign currency debt crises, because the bankruptcy (or even illiquidity) of states cannot be made to work smoothly. But this is a reason for diminishing sovereign borrowing. Second, bankruptcy can and must be made to work where indebtedness is private. The biggest lacuna in Eichengreen’s paper is that it ignores this issue. Third, if foreign currency borrowing by financial institutions becomes sovereign ex post, it should be treated as potentially sovereign ex ante. These institutions should not be allowed to borrow freely abroad. And finally, if domestic insolvency procedures were the main way to ensure the bailing in of private sector lenders, it would be possible also to provide liquidity support to governments, in the form of supplementary foreign exchange reserves. This would not be a “bail out.” It would merely be a way of ensuring that the mistakes of some borrowers and lenders did not determine the fate of everyone engaged in the economy.

    Other Resources Citing This Publication