(story to be published in tomorrow's issue, March 29)
The authors are from the University of California at Berkeley and World Bank, respectively. None of the opinions expressed necessarily represent the official views of the World Bank, its Executive Directors, or any other official body. For comments and suggestions, we would like to thank, without implicating, Stijn Claessens, Stanley Fischer, Francesco Luna, Pradeep Mitra, Andriy Storozhuk, and Andrew Vorkink. New developments in the negotiations surrounding the debts of some of the countries we consider are occurring as we write; this draft was completed in late-February 2000.
Moral hazard is increasingly seen as a problem in international financial markets, and private-sector burden sharing is increasingly seen as the solution. Investors must bear at least some of the costs of crises, the argument goes, if they are not to disregard the risks of lending. Ensuring that they do not escape all losses as a result of support provided by the international financial institutions (IFIs) - seeing that investors, rather than being "bailed out," are "bailed in" - is central to any strategy for limiting moral hazard. The failure of the IFIs to articulate a coherent approach to this problem thus provides ammunition to those who insist that they are part of the problem rather than part of the solution. While the critics have a point, neither is it feasible to simply let the moral hazard problem fester. Hence, the IFIs have been groping toward a strategy for addressing it, most recently in their policies toward Pakistan, Ecuador, Romania and Ukraine. In all four cases, the focus of their efforts is bonds, not bank loans like those which created such problems in Thailand in Korea in 1997. Our assessment is that the IFIs' efforts to condition official assistance on "private sector participation" - specifically on the willingness of investors to roll over maturing loans, provide new money, or restructure existing debts - have been less than a success. The reason is not hard to see. A new strategy built on statements of intent that does not also change the underlying payoffs will not be taken at face value. Because it is not credible, it will not change the strategies of market participants. They it will not change the results of their interaction with the multilaterals and the debtor. Finally, the current approach to private-sector burden sharing has a number of negative side effects. In some cases it has destroyed the rudimentary seniority structure that debtors and creditors have sought to create. In others it has substituted for impending debt-servicing problems the prospect of an even more severe crisis in the not-too distant future - it has traded bad today for worse tomorrow. And it has undermined the credibility of the IFIs by committing the International Monetary Fund and the World Bank to policies that are not credible or time consistent, and on which they are unlikely to follow through. For all these reasons, we believe that the approach taken in these test cases is misguided. The IFIs would do better to devote less attention to ad hoc efforts to bail in captive investors in countries like Pakistan, Ecuador, Romania and Ukraine. They would do better to concentrate on a forward-looking solution of more general applicability.
Burden sharing in principle
Efforts at achieving greater private-sector burden sharing are motivated by the perception that IMF programmes starting with Mexico have let investors off the hook and are a source of moral hazard. In addition, because the Fund is almost always paid back, these payoffs are effectively transfers from the taxpayers in the crisis country to international investors. On both efficiency and equity grounds, then, the status quo is unacceptable.
Burden sharing in practice
To date, efforts to encourage private-sector burden sharing have attempted to involve the private sector in re-financing or rolling over existing obligations on an ad hoc basis. The obligations have been bonds; the debtors have been states. The four test cases have been Pakistan, Ecuador, Romanian and Ukraine. Pakistan incurred sanctions as a result of its nuclear tests and a loss of investor confidence as a result of its military coup, while Ecuador has suffered a very severe recession and continuing political unrest, including a short-lived military takeover. Ukraine and Romania, in contrast, have relatively light debt loads, as is typical of the transition economies, but sharp repayment peaks that will create liquidity problems unless their maturing debts are rolled forward. Romania has managed to meet its obligations to date, while Ukraine is in the process of restructuring.
ROMANIA
Romania faced bullet repayments on two obligations in May and June 1999 - one a Samurai bond of $460 million, the other a eurobond of $245 million. Both issues, contracted in 1996, were distributed among German and Japanese retail investors. As a precondition for the IMF standby negotiated in the spring and summer of 1999, the Fund required the country to roll over 80 percent of that debt. Predictably, this demand came to nothing. Faced with this reality, the IMF changed course, conditioning its programme instead on the country raising $600 million of new money (roughly equal to the amount of the original rollover request). In the event, Romania managed to raise only $130 million. The vast majority, $108 million, came in form of a "club loan" from a consortium of commercial banks already active in the country. This outcome was reminiscent of the debt crisis of the 1980s. Then too the IMF had conditioned its programmes on the creditors first agreeing (in principle) to ante up additional funds. So long as success hinged on the efforts of a small, cohesive group of creditors who feared that their solvency would be threatened by a failure to provide additional resources, this strategy could work. But once the banks had provisioned against losses, they refused to participate further. The danger that this would hold IMF agreements hostage, at considerable cost for the crisis country, was what led the Fund to adopt its new policy of lending into tranches, breaking the link between its disbursements and new private lending. In a sense, then the "innovation" of the Romanian programme was a step back to the old, untenable state of affairs, allowing a large, loose collection of creditors, retail bondholders among them, to hold official money hostage. Romania then used its reserves to redeem both the eurobond and the Samurai bond as they matured without IFI support, before Bank and Fund programmes were formally in place. Reserves fell to their low point immediately following these repayments. They then recovered (reflecting the beneficial balance of payments effects of a large currency depreciation in the spring and renewed efforts at structural reform). By October, reserves exceeded the Fund's target by some $300 million. Despite the fact that the condition of raising $600 million of new finance was not met, in the summer of 1999 the IMF decided to disburse the first tranche of its Standby Agreement with Romania anyway. It did so despite the country having paid off its creditors, contrary to the spirit of the bail-in strategy. However, the Fund limited its first disbursement to the minimum access to quota possible and set as a condition for the release of further tranches that new money be raised to the tune of the missing $470 million. The irony was that Romania's need for new money was hardly pressing. It had managed to retire the debt which caused the problem and was over-performing on external targets like the current account deficit and currency reserves. But not raising new money would have jeopardised the standby, which was important for market confidence. The Romanian authorities, together with two investment banks, therefore set off on a road show to prepare the markets for the issue of a new eurobond and possibly a Samurai. Romania was the only country possessing a non-investment grade (B- and B3) rating with the temerity to approach the markets. Investor response was predictably tepid. Failing to attract much interest in Europe, the road show, by the time it reached the U.S., had become a ghost show. Tapping the Samurai market disappeared from the agenda. The IMF responded by reducing the sum required prior to first review to a more modest $100 million but without foreswearing the option of again raising the issue before releasing further tranches. Nevertheless, the gesture was widely read as giving in to the fact that Romania could not raise new funds. The only conditions imposed were the amount of new money and that it have a maturity of no less than two years.
IMPLICATIONS
This review of the four test cases of the new strategy for achieving greater private-sector burden sharing has a number of implications. Requiring countries seeking IMF assistance to first raise new money is unrealistic, given the palpable reluctance of investors who do not already have a stake in the crisis country to lend into uncertain conditions. Encouraging countries to suspend payments as a way of driving their bondholders to the bargaining table will not work so long as there is no bargaining table to be driven to. The result, mote likely than not, will be formal declarations of default, leading to an extended period of messy negotiations and lost capital-market access. And given the high costs of default, it is simply not time consistent for the IFIs to plan to stand aside if the markets refuse to roll over maturing claims, restructure problem debts or provide new money. The fact that default and restructuring are so painful and costly for the country, creating an incentive to disburse even if investors fail to comply, places the IFIs in the position of having to back down on their previous conditionality, which undermines their credibility. And since investors are aware of these facts, their behaviour is unlikely to be modified by the IFIs' less-than-credible statements of intent. The equilibrium in the game between the IFIs and the markets will remain fundamentally unchanged. Among the undesirable effects of the case-by-case approach has been the destruction of the rudimentary seniority structure that existed in the markets in question. Another unintended consequence of this approach is, ironically, moral hazard itself. Only treating new debtor as senior to existing debt can entice new lending once a country has made it onto the bail-in list. This guarantee requires the explicit or implicit backing of IFIs if it is to carry credibility in the institutional vacuum just created. It thus runs counter to the stated principle of making private sector loans less, and not more, dependent on official financing. A second source of moral hazard is that investors will recognize that the subjects of the bail-in exercise are all small countries whose debt problems are unlikely to have systemic repercussions. They will therefore prefer to lend to large countries that are less likely to be subject of the bail-in experiment. By lending to such countries, they will then contribute to the very systemic risks that everyone agrees must be avoided. Bail-in candidates, for their part, will find it hard to finance legitimate investments. They will have to live under the threat of investors trying to withdraw at the first sign of trouble, which carries the potential of triggering or accelerating crises. A final and especially regrettable casualty of these attempts to secure "private sector involvement" has been the effort to change the legal provisions of loan agreements. In all of the cases considered here, the IFIs and G7 countries failed to stake out a position as advocates of new contractual provisions. In no instance were legal or institutional innovations that might help to resolve future difficulties part of the restructuring agreement. In particular, collective action clauses were not mentioned in discussions of how new money was to be assembled. This is unfortunate because only if the legal provisions and institutional arrangements governing sovereign borrowing are changed will efforts to secure greater private sector involvement in crisis resolution bear fruit. Without them, restructuring will remain unacceptably costly and painful for the country concerned. It will not be possible for the IMF and the World Bank to stand aside if restructuring negotiations fail. And since investors are aware of this fact, they have no incentive to modify their behaviour when presented with IMF statements to the contrary. Hence, the current approach to encouraging private-sector involvement is doomed to fail.
CONCLUSION
Nothing we have written questions the need to address the moral hazard created by multilateral assistance to crisis countries. Our analysis does, however, challenge the wisdom of the case-by-case approach taken in Pakistan, Ecuador, Romania and Ukraine. Requiring countries seeking IMF assistance to first raise new money is unrealistic, given the reluctance of investors who do not already have a position in the crisis country to lend into uncertain conditions. Demanding that creditors roll over their maturing claims as a condition for multilateral assistance may be slightly more realistic, given incumbent investors' stake in the country, but still must overcome formidable collective action problems when the creditors are bondholders. Encouraging countries to suspend payments as a way of driving the bondholders to the bargaining table will be disastrous so long as there is no bargaining table to which to be driven. The result will be formal declarations of default and an extended period of messy negotiation and lost capital-market access. Given the fact that default and restructuring are so painful and costly, it is simply not time consistent for the IFIs to plan to stand aside if the markets refuse to roll over maturing claims, restructure problem debts or provide new money. Because these realities create an incentive to disburse even if investors fail to comply, the IFIs are then placed in the position of having to back down on their previous conditionality, which undermines their credibility. And since investors are aware of these facts, their behaviour is unlikely to be modified by the IFIs' less-than-credible statements of intent. The equilibrium in the game between the IFIs and the markets will consequently remain unchanged. This approach to "bailing in the private sector" will not work. Fortunately, there is an alternative: introducing collective-action clauses into loan agreements. Under present institutional arrangements, restructuring is unappealing except under the most extraordinary circumstances. Collective-action provisions would make it feasible to pursue this alternative. Debtors and creditors could decide when restructuring was desirable; it would no longer be necessary for the issue to be forced by the IFIs. Limited IMF lending into tranches to countries engaged in good-faith negotiations would become feasible. No longer would the only alternatives be paying off creditors in full with official funds or enduring a costly, extended interruption to market access. This, and not ad hoc efforts to bail in the private sector, is the forward-looking solution to the moral hazard problem.
Pentru alte știri, analize, articole și informații din business în timp real urmărește Ziarul Financiar pe WhatsApp Channels
