You Never Give Me Your Money? Sovereign Debt Crises, Collective Action Problems, and IMF Lending

We review the impact of the global financial crisis, and its spillovers into the sovereign sector of the euro area, on the international “rules of the game” for dealing with sovereign debt crises. These rules rest on two main pillars. The most important is the IMF’s lending framework (policies, financing facilities, and financial resources), which is designed to support macroeconomic adjustment packages based on the key notion of public debt sustainability. The complementary pillar is represented by such contractual provisions as Collective Action Clauses (CACs) in sovereign bonds, which aim to facilitate coordination among private creditors in order to contain the costs of a debt default or restructuring. We analyze the most significant changes (and their consequences) prompted by the recent crises to the Fund’s lending framework, not only in terms of additional financial resources, new financing facilities (including precautionary ones), and cooperation with euro-area institutions, but also as regards the criteria governing exceptional access to the Fund’s financial resources. We highlight a crucial innovation to these criteria, namely that, for the first time, they now explicitly take account of the risk of international systemic spillovers. Finally, we discuss how the recent crises have provided new political support for a broader dissemination of CACs in euro-area sovereign bonds. Importantly, in the first case involving an advanced economy, CACs were activated in the debt exchange undertaken by Greece in Spring 2012.


We review the impact of the global financial crisis, and its spillovers into the sovereign sector of the euro area, on the international “rules of the game” for dealing with sovereign debt crises. These rules rest on two main pillars. The most important is the IMF’s lending framework (policies, financing facilities, and financial resources), which is designed to support macroeconomic adjustment packages based on the key notion of public debt sustainability. The complementary pillar is represented by such contractual provisions as Collective Action Clauses (CACs) in sovereign bonds, which aim to facilitate coordination among private creditors in order to contain the costs of a debt default or restructuring. We analyze the most significant changes (and their consequences) prompted by the recent crises to the Fund’s lending framework, not only in terms of additional financial resources, new financing facilities (including precautionary ones), and cooperation with euro-area institutions, but also as regards the criteria governing exceptional access to the Fund’s financial resources. We highlight a crucial innovation to these criteria, namely that, for the first time, they now explicitly take account of the risk of international systemic spillovers. Finally, we discuss how the recent crises have provided new political support for a broader dissemination of CACs in euro-area sovereign bonds. Importantly, in the first case involving an advanced economy, CACs were activated in the debt exchange undertaken by Greece in Spring 2012.

You never give me your money – You only give me your funny paper And in the middle of negotiations – You break down

(The Beatles, 1969)

I. Introduction

Sovereign debt crises have been brought back to centre stage as a result of the problems faced since 2010 by some countries belonging to the European Union’s Economic and Monetary Union (EMU or the euro area). It is notable that these problems have occurred despite the presence of institutional frameworks explicitly aimed at fostering fiscal discipline, which clearly failed to ensure effective surveillance and rule enforcement.

The international rules of the game for managing and resolving sovereign debt crises1 have evolved significantly over the years under the pressure of events, but also reflecting changes in the structure and sophistication of international financial markets. One development that greatly contributed to shaping the policy debate in academia and in official circles was the shift of sovereign borrowing away from syndicated bank loans towards bonds in the late 1980s. This shift had a profound impact on the relationship between sovereign debtors and their private creditors, most notably by amplifying the potential for so-called collective action problems, whereby individually rational actions lead to sub-optimal outcomes from a collective standpoint. Because of coordination failures, risks of litigation and holdout behavior on the part of (some) creditors, collective action problems may not only raise the costs of solving a sovereign debt crisis but can also increase its very likelihood.

The string of crises in some key emerging market economies since the second half of the 1990s prompted the international community to devote a great deal of effort to strengthening crisis prevention and resolution, including by addressing these collective action problems. One of the main objectives at the time was to minimize the economic costs of sovereign crises as well as the use of official financial resources (bailouts); much emphasis was placed on the need to facilitate so-called Private Sector Involvement (PSI), i.e. a reduction (usually in terms of net present value) of private creditors’ claims, so as to share with the sovereign debtor the burden of adjusting an unsustainable debt position.

The cumulative outcome of these efforts was the framework of rules and tools for crisis management and resolution prevailing before the recent global and regional crises. At the risk of some over-simplification, this framework consisted of two main pillars: (a) the Fund’s lending policies2 and financing facilities, which could count on a broadly adequate volume of financial resources; and (b) the inclusion of Collective Action Clauses (CACs) in the sovereign bonds issued in key financial jurisdictions, with a view to facilitating PSI and making debt restructuring more orderly. To be sure, sovereign debtors and creditors could also rely on a set of “Principles” for fair debt restructuring, first developed in November 2004, by which a number of international investor associations and emerging market sovereign debtors voluntarily abide.3

The outbreak of the global financial crisis in 2007-08 and its subsequent spillover into the sovereign sector of some euro-area countries posed unprecedented threats to global stability, ushering in major changes in the above framework. Some of these changes had seemed implausible until then, but were needed in order to deal adequately with new distinct challenges. Importantly, the existing rules and tools for crisis management and resolution were tested for countries and under circumstances very different from those envisaged at the time of their inception in early 2000s.

Against this background, this paper intends to inform the current policy debate in two main ways: by reviewing in sufficient detail how and why the two pillars for dealing with sovereign debt crises have evolved since the mid-1990s; and by analyzing the impact and consequences that the recent crises have had on them. In our concluding remarks, we also highlight some open issues and challenges regarding the ways to address sovereign debt crises and their systemic dimensions in the future, and we emphasize the importance of crisis prevention, most notably for systemically important countries.

Our main findings are the following.

We identify two main phases in the evolution of the Fund’s lending framework by focusing, to a very large extent, on cases of exceptional access to Fund financing (i.e. above the normal limits). The first phase spans the period from the Mexican crisis of 1994-95 to, while the second begins with the reforms approved since March 2009 in response to the global crisis.

This first phase featured steps to move from an initial regime that allowed greater case-by-case discretion in the use of the Fund’s financial resources to a new, more systematic and rule-based one, designed to make Fund lending more predictable while preserving some flexibility. In admittedly rather simplistic (and somewhat abused) terms, this can be described as a move from constructive ambiguity to constrained discretion. The initial ambiguity was mostly related to the “exceptional circumstances clause” (ECC), as it left unspecified the circumstances under which it could be invoked to provide large-scale loans.

A key driver of policy changes in this first phase was the repeated reliance on the ECC to justify cases of exceptional access to Fund financing, especially starting from the Mexican crisis. This more frequent use of a clause originally intended to remain of an extraordinary nature created a perception that the Fund’s lending behavior was insufficiently systematic. This perception, in turn, raised two concerns among the Fund’s major shareholders and in academia. On the one hand, it was feared that the proliferation of large-scale official loans could exacerbate moral hazard problems on both the creditor and debtor sides, thereby potentially increasing the likelihood of new crises in the future. On the other hand, the perceived low predictability of the Fund’s lending decisions per se was judged to be a potentially dangerous source of market uncertainty and contagion.

Several important measures were taken to overcome these problems. In September 2000, the Prague Framework, endorsed by the Fund’s International Monetary and Financial Committee (IMFC), explicitly stated that IMF resources had to remain limited, and that extraordinary access to them should be exceptional. The Action Plan adopted by the G-7 countries in April 2002 was a definite breakthrough, as it operationalized the above two concepts by paving the way for such key reforms as the Fund’s Exceptional Access Policy (EAP) in Capital Account Crises approved in 2003, and the official sector’s push for the voluntary inclusion of CACs in international sovereign bonds (which was eventually sparked in February 2003 by Mexico). It was underscored that these components were complementary and mutually reinforcing, so that both had to be in place to deliver their intended benefits in full.

The common thread running through these reforms can be summarized in two words: greater predictability. The reforms were driven by the belief that the expectations of both Fund members and private market participants had to be better directed in order to foster market discipline and, ultimately, crisis prevention. One notable outcome was the significant reduction of the margins for discretion embedded in the (previously unconstrained) exceptional circumstances clause.

A new phase of Fund lending began with the comprehensive reforms approved in March 2009, which were triggered by the need to address effectively the specific challenges posed by the global crisis and to meet the fast-growing demand for Fund loans. These reforms introduced inter alia two key changes. First, new facilities of a “precautionary” type were created to help contain contagion risks for countries with broadly sound fundamentals and policies. Second, the rules governing the EAP were systematized and reinforced, notably by clarifying that the exceptional circumstances clause could be invoked only when all of the policy’s four substantive criteria were satisfied. These changes dramatically constrained (and perhaps virtually removed) the scope for discretion allowed by this clause, thereby reinforcing the presumption that decisions on IMF lending would strictly follow a rule-based process in the future.

Within this new phase of Fund lending, the (first) Greek program of May 2010 represented a landmark moment. Two consequential developments stand out, both of which were prompted by unprecedented threats of a systemic crisis in the euro area and attendant global spillovers. First, given that the country’s financing needs proved extremely large by historical standards, the resources committed by the IMF had to be complemented by even larger bilateral contributions from other euro-area member countries. Accordingly, for the first time ever in the Fund’s history, program conditionality was devised, negotiated, and monitored by a Troika composed of the IMF, the European Commission, and the European Central Bank. Second, the key EAP criterion on debt sustainability (i.e. that public debt should be sustainable with high probability in the medium term as a necessary condition for a country to receive large-scale IMF loans) was amended to consider exceptional access explicitly in cases of a high risk of international systemic spillovers.

This amendment, which we label the systemic clause, is a distinct development. It has strengthened the Fund’s ability to address situations of systemic stress with heightened uncertainty in a timely fashion, by means of large-scale loans to all countries that can originate significant cross-border spillovers. At the same time, however, the revised criterion might imply a lower “evidentiary bar” on debt sustainability, given that Fund financing can now be granted even in circumstances where significant uncertainties make it difficult to state categorically that a country’s public debt is with high probability sustainable in the medium term. In any event, the more rigorous level of scrutiny required by requests for exceptional access to Fund financing is still preserved by the fact that such requests will continue to be always evaluated in the context of all EAP criteria and relevant risk mitigating procedures of the IMF, most notably those concerning program design and conditionality.

As far as CACs are concerned, the sovereign debt crisis in the euro area has brought about two major developments. First, in March 2011, the European Council decided to include standardized and identical CACs in all new euro-area sovereign bonds with maturity above one year. This measure, set to come into effect from January 1, 2013, is expected to provide decisive momentum for broader diffusion of CACs in the coming years. Second, the activation of CACs during the successful debt exchange undertaken by Greece in March-April 2012 has shown clearly that they still represent a useful (if not indispensable) tool for sovereign debt restructuring.

Notwithstanding these major developments, the question of whether CACs will make a material difference in crisis resolution remains uncertain at this juncture. For one, CACs have long been recognized as tools that can help address only a specific aspect of a debt restructuring, namely the creditor holdout problem. The use of CACs in the Greek debt exchange presents some specific features that might not be applicable to more general cases. CACs could indeed be inserted retroactively in most of the sovereign debt eligible for exchange, as it was governed by Greek law. Moreover, a creditor coordination role was played by the Institute of International Finance.

Uncertainty about the actual impact of CACs stems from two major concerns, already noted in the early 2000s: (a) the transition problem, that is, the speed at which CACs could spread into the outstanding stock of sovereign bonds; and (b) the aggregation problem, i.e. the fact that CACs usually bind creditors within a single bond issue but not across multiple issues. Most fundamentally, given the higher potential for contagion and systemic risks (including from today’s stronger bank-sovereign linkages) that characterize global financial markets, a safe and serviceable activation of CACs appears to require more demanding preconditions than in the past, most notably in terms of stronger capital and liquidity requirements and appropriate resolution mechanisms for financial institutions; and larger official financial pools (firewalls), including at the regional level. The relationship between CACs and credit derivatives (notably their legal features) also needs to be more thoroughly investigated.

This paper is organized as follows. Section II illustrates the problems attached to a sovereign debt crisis and the reasons underlying the quest for new mechanisms to ensure more orderly and efficient processes of debt restructuring. Section III reviews the debates that have led to the current contractual approach to the resolution of sovereign debt crises, along with a discussion of how CACs have actually been adopted by sovereign issuers. Section IV illustrates developments in the IMF’s lending framework since the Mexican crisis of 1994-95, notably those pertaining to the exceptional access policy. The impact of the global crisis on the two pillars of the framework for dealing with sovereign debt crises is analyzed in Section V. Finally, Section VI presents some concluding remarks and briefly touches upon what we consider the main open issues to be addressed for a better collective governance of systemic fiscal and financial risks.

II. Sovereign Debt Crises and the Collective Action Problem

Sovereign debt crises are a recurrent feature of the international financial landscape, and there have been literally hundreds of them since the mid-fourteenth century (Reinhart and Rogoff, 2009). The recent crisis-driven rise in public debt-to-GDP ratios in G-20 advanced countries (from 80 percent in 2007 to close to 120 percent expected in 2012; IMF, 2012c) is a stark reminder that sovereign debt might eventually undermine macro-financial stability if governments and markets alike fail to keep public finances in check.

The economic literature has long tried to explain the behavior of sovereign debtors, particularly the reasons why they generally choose to honor their financial obligations. Indeed, one distinct feature of sovereign debt is the inherent problem of moral hazard associated with borrowing without collateral (attachable assets).4 Another key fact is that countries have historically shown a different degree of debt tolerance, i.e. a tendency to default at very different levels of sovereign debt (Reinhart et al., 2003), which highlights the role of reputational and institutional factors in supporting the perceived sustainability of public finances in the medium term (see Giordano and Tommasino, 2009; Balcerowicz, 2010; and Visco, 2010).

While economic theory acknowledges a number of legal and reputational penalties, the single most important factor that makes sovereign borrowers strive to service their debts is the severe loss of economic output that usually follows a default. To a significant extent, this loss stems from the impact of a default on the domestic financial system, most notably on banks. Since the latter are generally important holders of sovereign bonds, government insolvency will usually significantly affect banks’ balance sheets and, in the worst-case scenarios, their solvency. Contagion effects and sovereign-bank inter-linkages may further increase the costs of a default and spread them internationally. All these arguments have played a major role in the debate on how to solve the euro-area debt crisis, greatly shaping the attendant policy strategy and actions.

From a theoretical viewpoint, extremely costly defaults have been interpreted as the endogenous response for a sovereign debt market to exist, as any attempts to dilute creditor rights and make restructurings easier would hinder the functioning of the market itself. In other terms, costly defaults may be viewed as an equilibrium outcome or a “punishment that in some sense substitutes for effective property rights at the international level” (Rogoff, 2003).

Although sovereign debtors have clear incentives to repay their debts, defaults and restructurings do actually occur.

In general terms, the management and resolution of sovereign debt crises normally implies a collective effort to combine, in varying doses that depend on country-specific circumstances, three different factors: domestic policy adjustment, official financing (including from the IMF), and, when necessary, Private Sector Involvement (PSI). The latter is usually pursued through a pre-emptive debt restructuring (i.e. launched before an outright default), effected through an exchange offer to creditors that often involves a reduction (haircut) in the net present value of sovereign debt (not necessarily in its face value). Domestic adjustment, i.e. the adoption of restrictive macroeconomic policies, first and foremost on the fiscal side, is an obviously pivotal factor to rein in an unsustainable debt burden. However, it typically takes time to bear fruit, and too large or rapid an adjustment may have unacceptable costs from an economic, social, and political point of view. Official financing thus comes to play a key role, as it aims to help smooth the adjustment needed to place a country on a sustainable path towards internal and external balance. If policy adjustment and official financing were to prove insufficient to restore debt sustainability in the medium term, PSI would unavoidably become a necessary ingredient of crisis resolution. And PSI would be facilitated, inter alia, by using such legal tools as Collective Action Clauses.

A. The Collective Action Problem

Against this background, the resolution of an unsustainable sovereign debt is a typically complex and often painful process: “there is one lesson [from the history of sovereign debt restructurings of the modern era] that is inescapable. A sovereign debt crisis can be a painful experience for both the debtor and its creditors; a mismanaged sovereign debt crisis can be a catastrophically painful experience” (Buchheit and Gulati, 2010).

This process is shaped by the behavior of (and the incentives to) the three players involved: the sovereign debtor, its private sector creditors, and the official sector (including the IMF). These players have different objectives and responsibilities. Importantly, a key feature of a sovereign default or restructuring is that its consequences can be amplified by collective action problems arising from information asymmetries between lenders and borrowers, as well as from co-ordination failures among creditors, notably bondholders (see Haldane et al., 2005, for a theoretical model that analyses the merits of CACs).

It should be noted that collective action problems can be a consequential factor, not only in the context of a sovereign debt default or restructuring, but also (and perhaps more disruptively) as the immediate cause of a crisis. According to some interpretations, the Mexican crisis of 1994-95 was a first-hand demonstration of the disruptive power of collective action problems posed by dispersed holders of sovereign debt: in the case of Mexico, these problems materialized in the form of a flight from short-term debt that triggered a currency and banking crisis (Bi et al., 2011, p. 6). It is therefore no wonder that, in the face of the Mexican crisis, fears that lack of creditor coordination could trigger a crisis ultimately led to a number of policy proposals to reform the international financial architecture.

Collective action problems pertain to both private and sovereign debtors, but they are generally more acute for the latter. The analogy between the two cases is rooted in the fact that the collective actions problems affecting a distressed firm are remarkably similar to the ones faced by a sovereign borrower in financial trouble. The parallel is drawn because, much like a corporation, a defaulting government faces the risks of a creditor grab race, problems in obtaining new money, and potentially serious obstacles to a debt restructuring induced by the threat of holdout by particular classes of creditors.

In a national context, bankruptcy arrangements, such as Chapters 11 and Chapter 9 of the U.S. Bankruptcy Code, have been developed to address collective action problems and to provide court assistance for the orderly treatment of the debts owed by insolvent private corporations and, in some cases, local governments. By contrast, although the idea of applying corporate bankruptcy principles to sovereigns goes back as far as Adam Smith (Rogoff and Zettlemeier, 2002), similar mechanisms are not available for the orderly resolution of sovereign debt crises.

This absence largely explains the variety of approaches to sovereign debt restructuring adopted over time to deal with country-specific circumstances. More worryingly, this vacuum can lead to market equilibria where costly defaults, as noted before, are a very poor substitute for stronger property rights, and collective action problems can create situations where both the country and its creditors face even larger economic losses. The need to avoid (or lessen) the costs associated with a default or even a disorderly (non co-operative) restructuring is thus the major economic rationale underpinning efforts to develop predictable procedures for resolving unsustainable sovereign debts in an orderly and cost-efficient way.

B. Bonded Debt and Litigation Risk

From an empirical viewpoint, the desirability of such procedures for sovereign debt restructurings was made evident, inter alia, by a major shift in the composition of emerging markets’ creditors, which started in the early 1980s. During those years, when syndicated bank loans were the vehicle to finance emerging market economies, in case of a crisis the debt workout procedures of creditor committees such as the London Club (commercial banks) were able to ensure a degree of co-ordination among creditors participating in the loan syndicate.

The Brady Plan of 1989 solved the sovereign debt crises of the 1980s by converting bank claims into securitized instruments (Brady bonds). The task of financing sovereigns in emerging market economies shifted from a few international banks providing syndicated loans to a large number of bondholders worldwide. In principle, this greater dispersion of bondholders could be expected to reinforce the resilience of the international financial system, by making it less likely that liquidity or solvency problems of individual debtors would give rise to systemic consequences.

At the same time, however, the expansion of international bond markets in the 1990s made creditor-debtor relations more complex and crisis resolution more difficult, as information asymmetries and collective action problems increased: for example, because bondholders lacked both the practical apparatus and the disciplining device of a lending syndicate or the London Club to enforce co-operative solutions.

Accordingly, the rise of bond financing brought with it heightened risks of litigation. The discretion of a creditor to accelerate its bond following a default, or to commence a lawsuit and attach the (sovereign) borrower’s assets, could dramatically reduce the other bondholders’ options for dealing with this event. Creditor grab races might thus become the rational, instinctive sauve qui peut response of some bondholders (Buchheit and Gulati, 2002).

Historically, litigation risks have not been regarded as very serious by sovereign borrowers, as they typically have few assets to be attached and some of them benefit from sovereign immunities (Group of Ten, 1996, p. 10). In the cases of Pakistan (1999), Ecuador (2000), Ukraine (2000), and Uruguay (2003), creditor litigation essentially did not materialize during the restructuring process; as noted by the IMF, the limited role played by holdout creditors was one of the critical factors that permitted an orderly settlement (IMF, 2002b and 2012a).

This notwithstanding, litigation risks cannot be dismissed in light of some key changes that have taken place in the financial and legal environment in the last two decades (IMF, 2004a). As noted before, the switch of international sovereign financing from syndicated bank loans to the bond market precludes recourse to traditional collective negotiating frameworks such as the banks’ steering committees. Moreover, a more litigation-prone environment may have also been favored by legal developments that have eroded sovereign debtors’ defenses against litigation, exposing them to aggressive judiciary enforcement actions.

Until about the middle of the twentieth century, the doctrine of sovereign immunity endorsed by the law of most countries, including the U.S. and the U.K., prevented a sovereign from being sued in foreign courts without that sovereign’s consent (Mauro and Yafeh, 2003). However, in 1985 the ruling in Allied Bank vs. Costa Rica set an important precedent for creditors enforcing their claims through the court system, as it demonstrated that, at least under New York law, creditor litigation against a sovereign was a legal alternative to a restructuring.

Another landmark moment was the Elliot vs. Peru ruling of September 2000, which stated the principle that payment streams between sovereign debtors and foreign creditors, made through the international payments system, may well constitute assets to target for satisfying the claim of a litigating creditor. This was a landmark case as it showed that a holdout strategy could be successful despite the lack of attachable physical assets. Most importantly, it conveyed the perception of a heightened risk of litigation, which could discourage creditors as a whole from participating in an exchange offer because the very threat of legal actions by a few holdouts could impair the sovereign debtor’s commitment to resume payments in the post-restructuring stage. The renewed relevance of litigation risk has been confirmed by the many court cases that followed Argentina’s sovereign default in December 2001.

The risk of litigation is at the root of one of the most consequential collective action problems, the creditor holdout problem. This refers to a situation in which each creditor is better off if he or she individually holds out for full repayment of the original claim while the other creditors make debt relief concessions to the debtor. Holdout bondholders (also known as maverick or rogue creditors) would exploit the power to veto a restructuring and pursue individual legal actions to seize the sovereign’s assets. In essence, by raising the risk of a lengthy and confrontational legal process for the debtor, holdout behavior can constitute a major barrier to orderly sovereign debt restructuring.

The relevance of creditor coordination problems and holdouts is questioned by some, however, as neither the absence of a formal statutory mechanism nor the lack of CACs appeared to be major impediments to restructurings. As noted earlier, even with uncoordinated atomistic creditors and the possibility of successful holdout litigation, all major bond restructurings in the last decade have achieved very high levels of creditor participation (over 90 percent) and have not led to significant litigation (perhaps with the only exception of Argentina’s external debt restructuring of 20055).

In other words, sovereign bonds with unanimity (rather than collective action) clauses do not necessarily preclude a debt restructuring; the issuers of such bonds could still count on debt exchanges as the basic technology to coordinate their negotiations with international bondholders in case of need (Roubini and Setser, 2004).

Past cases of successful debt exchanges without CACs can be ascribed to two factors (Bi et al., 2011): first, countries may still make an exchange offer that is sufficiently attractive to discourage litigation; second, these offers can be complemented by legal mechanisms (such as minimum participation thresholds or exit consents6) designed to discourage litigation and mitigate possible coordination failures. All told, the available evidence would seem to indicate either that the coordination problem was overstated (in the sense that it was in the power of a debtor country to avoid it, by making a reasonable exchange offer), or that successful legal design could ensure an orderly restructuring even in the absence of CACs.

Finally, it is interesting to recall that even in the case of Mexico in 1995, which brought the creditor coordination problem dramatically to the fore, the crisis did not involve its New York law bonds (without CACs), but rather domestic bonds (Tesobonos) governed by Mexican law. It has thus been observed that Mexico, like Greece in 2012, could have enacted a law to restructure the Tesobonos by fiat. Accordingly, the international push for CACs prompted by the Mexican crisis can be interpreted as a move to “expand the policy options for the next sovereign crisis” (Gelpern and Gulati, 2011, p. 4). These considerations will be elaborated on further in Sections III, V, and VI.

C. The Role of Official Financing in a Sovereign Debt Crisis

IMF-led rescue packages have usually represented the main device to address sovereign financial crises, building on the catalytic role that Fund financing is expected to play for both other official creditors and private investors.

In a wide variety of cases, official bailouts have been broadly successful in limiting the economic and social costs of a crisis and the potential for systemic effects, thus helping to preserve the stability of the international financial system. Arguably, Fund financing can help address sovereign debt difficulties and related market failures that may emerge at different stages:

  • - in a crisis prevention setting, when public debt is approaching excessive levels but problems have not yet materialized, by clarifying ex ante the conditions under which countries can (if illiquid) or cannot (if insolvent) expect to receive financial support. Knowledge of these conditions should help shape private investors’ expectations and foster market discipline (bond market vigilantes) by reinforcing the pressure on sovereign debtors to pursue prudent policies. More recently, Fund financing can also play a preventive role also through its new precautionary facilities introduced in the face of the global crisis (the Flexible Credit Line and the Precautionary and Liquidity Line), whose prequalification criteria have been designed to reward countries with sound policies and fundamentals;

  • - in a crisis management situation, where the country gets into trouble and loses market access, and thus policy adjustment is all the more necessary to restore fiscal and/or external sustainability and avoid default. Fund financing (especially when front-loaded) can smooth out the adjustment burden and play the above-mentioned “catalytic” role for complementary financing; and

  • - in a crisis resolution context, where countries may be in need of financing in order to contain the economic losses attached to a pre-emptive debt restructuring or an outright default. Although the Fund is not directly involved in debtor-creditor negotiations, it may offer a potentially convenient means for obviating specific types of market failure that may emerge at that stage, namely: (a) the absence of a legally enforceable seniority structure for sovereign debt, which weakens the incentives of the private sector to maintain its exposure or provide the country with new financing; and (b) information asymmetries between the sovereign debtor and its private creditors, which may negatively affect debt restructuring negotiations. The IMF can address these failures mainly through its Lending-Into-Arrears (LIA) policy, which also purports to enhance the credibility of the debtor country’s adjustment program (see Section IV).

However, IMF-led rescue packages can have their own drawbacks, and mitigating these problems is part of the collective solution of sovereign debt crises.

Perhaps the most celebrated of such problems is the risk of creditor and debtor moral hazard (see, for example, Dell’Ariccia et al., 2002; Spadafora, 2003; European Central Bank, 2005; and Hagan, 2005). Furthermore, in so far as Fund financing were not to be perceived as following a rule-based approach, it could give rise to uncertainty about the availability of official resources in a crisis, which, in turn, might spark contagion and affect international capital flows (Taylor, 2007).

As amply discussed in Section IV, in early 2000 these concerns prompted the official community to consider a more rule-based approach to discipline large-scale IMF loans, while keeping the flexibility needed to deal with unexpected challenges. This approach was expressly meant to be part of a strengthened framework for crisis prevention and resolution and to complement the design of new contractual or statutory mechanisms for an orderly restructuring of sovereign debt (Eichengreen, 2003).

III. The Contractual Approach to Sovereign Debt Restructuring

The contractual approach is rooted in the idea that inter-creditor relations are governed by incomplete contracts, notably regarding what to do in the case of sovereign debt restructuring (which may or may not be associated with an outright default). The lack of clear contractual rules on how to address this event is at the root of the collective action problems discussed before, which usually make creditor coordination so difficult and sovereign debt restructurings so unpredictable and generally costly. Accordingly, the contractual approach is centered on legal provisions, known as Collective Action Clauses (CACs), which help specify the procedures to be followed in the event of a debt restructuring.

More precisely, these clauses are designed to facilitate a voluntary, timely, and orderly debt restructuring when a sovereign faces an unsustainable debt burden, while at the same time protecting essential creditor rights and avoiding disruptive litigation. One of their most prominent features is to allow a qualifying majority of bondholders to agree on modifying the so-called Reserved Matters of a bond contract (essentially its payment terms), and to make these changes binding on dissenting bondholders (holdouts), who might otherwise veto a restructuring and hold out for preferential treatment.

Collective action clauses are set against Unanimous-Action Clauses (UACs), where any change in a bond contract requires the consent of each holder. They can be broadly classified into three main groups: (a) majority representation clauses, to foster early dialogue, coordination, and communication between creditors and a sovereign facing debt problems; (b) majority amendment clauses, to allow creditors and debtors to reach an agreement on the terms of a restructuring and to prevent a minority of bondholders from obstructing the process; and (c) majority enforcement clauses, to avoid disruptive legal action by individual creditors that would hamper an ongoing workout, while protecting the interests of the creditors as a whole.

In the mid-1990s, when the debate on sovereign debt restructuring resumed, CACs were routinely included in debt securities issued under U.K. law and in a few minor jurisdictions. According to some commentators (Buchheit and Gulati, 2002, p. 9), the absence of CACs in other major jurisdictions, particularly in the New York market (the largest one for sovereign bonds), was mainly a matter of market convention rather than of fundamental differences in the legal approach.

Until 2003 the absence of CACs in sovereign bonds issued under New York law could be explained by what is called drafting momentum, i.e. a lack of experience or inertia on the part of the lawyers tasked with drafting the bond contract, which implied that “the last corporate bond indenture [became] the model for the next sovereign bond indenture”. Overall, these considerations help explain why the debate on CACs in the early 2000s was largely focused “on convincing the U.S. investor community that the use of CACs did not represent a threat to their interest” (Gray, 2004).

A. Key Steps towards the Adoption of the Contractual Approach

Modern proposals to improve creditor coordination and develop a formal restructuring process for sovereign debt go back to the late 1970s. Up to the early 1990s, the emphasis was mostly on the creditor holdout problem, but after the Mexican crisis of December 1994-January 1995 two additional concerns emerged forcefully: addressing the risk of self-fulfilling debt crises, and containing the potential for moral hazard posed by large IMF-led bailouts (Rogoff and Zettlemeier, 2002).

Rogoff and Zettelmeyer (2002) attribute to Oechsli (1981), who suggested “an established procedural framework for debt renegotiation”, the credit for the first modern proposal of this kind, one that explicitly addressed the problem of creditor co-ordination while invoking the analogy with the Chapter 11 of the U.S. Bankruptcy Code.

With an influential lecture delivered in April 1995, Sachs was the precursor, in a radical form, of what would later come to be known as Private Sector Involvement in crisis resolution (Sachs, 1995). He called for an extensive overhaul of the IMF activities in helping sovereigns in financial distress, “so that the IMF plays a role far more like an international bankruptcy court and far like the lender of last resort to member countries”. Sachs’ call had an immediate and widespread resonance; most importantly, it helped shape the discussion on this topic held at the consequential Halifax Summit of G-7 countries in June 1995, which in turn paved the way for an important report of the G-10 on the resolution of sovereign liquidity crises.

The Report of the G-10 Working Party on the Resolution of Sovereign Liquidity Crises (the Rey Report, Group of Ten, 1996), while endorsing the fundamental principle that terms and conditions of debt contracts are to be met in full and on time, laid the foundations of a new (incremental) approach for involving private creditors in sovereign crisis resolution. It was recognized that, as an action of last resort and “in certain exceptional cases, the suspension of debt payments [i.e. a moratorium] may be part of the crisis resolution process” (p. 21).

Importantly, the Rey Report was fully cognizant of the creditor holdout problem, and thus underlined the potential role of CACs as a means to avoid disruptive behavior by dissident creditors (p. 3). A “market-led process” was encouraged “for inclusion in sovereign debt instruments of contractual provisions that facilitate consultation and co-operation between debtors and their private creditors, as well as within the creditor community, in the event of a crisis” (p. 1).

After the 1997-98 Asian crises, which mainly involved private debt, in the late 1990s a second wave of sovereign debt crises, involving such issuers as Russia, Ecuador, Pakistan and Ukraine, brought back to center stage the proposals to improve crisis resolution and triggered renewed interest in the use of CACs. Although these cases of debt restructuring did not pose particular problems of litigation, they stimulated the official sector to speed up the implementation of the framework devised in the Rey Report.

The official sector’s strategy to promote the diffusion of CACs began to take shape during the meeting of the IMF’s International Monetary and Financial Committee (IMFC) of September 2000 in Prague. By drawing on the conclusions of the Rey Report, the official community endorsed a set of principles for a new approach to the prevention and resolution of sovereign crises, which would come to be known as the Prague Framework (IMF, 2000).

Part of those principles had a direct bearing on IMF lending, which is discussed more extensively in Section IV. In particular, the IMFC underlined that Fund resources were limited (especially in comparison with the rising volume of international capital flows) and that extraordinary access to these resources had to be of an exceptional nature; furthermore, to contain the risk of moral hazard, neither creditors nor debtors should expect protection by the official sector from adverse outcomes.

More importantly for the themes discussed in this Section, a debt restructuring was deemed to be warranted, in the context of a Fund-supported program, when “the early restoration of full market access on terms consistent with medium-term external sustainability may be judged as unrealistic”; it was then underscored that in certain extreme cases, where voluntary and market-based solutions prove unavailable, “a temporary payments suspension or standstill may be unavoidable”. In this case, the Fund could resort to its lending-into-arrears policy. The IMFC also agreed that the operational framework for private sector involvement should rely to the largest possible extent on voluntary and market-oriented solutions.

Despite the clarity of this theoretical setting, the second Turkish crisis of February 2001 made it evident that there were still serious difficulties in involving private creditors in crisis resolution. Aware of those hurdles, the IMF’s then First Deputy Managing Director Anne Krueger pointed out that the Prague Framework could not ensure an orderly and cooperative debt workout even with the inclusion of standstills and lending into arrears as part of the process for crisis resolution. Specifically, the framework in question appeared unable to prevent holdout creditors from disrupting a restructuring, given that debtor countries were lacking legal protection from court action by those creditors.

Ms. Krueger’s considerations were at the root of her proposal for a statutory Sovereign Debt Restructuring Mechanism (SDRM), which was first made public in a speech delivered in November 2001 (Krueger, 2001).

The SDRM’s main objective was to provide strong incentives for debtors and their creditors to reach cooperative agreements on their own and achieve an orderly debt restructuring; the mechanism was intended to apply not only to bonds but to the whole of a sovereign borrower’s external debt (i.e. governed by foreign law), including instruments such as bank loans. Moreover, it would allow a single vote to restructure several debt instruments by aggregating votes across multiple issues (for more details see Krueger, 2001 and 2002).

For a while, the statutory and contractual approaches were both judged to be possible solutions to usher in orderly sovereign debt restructurings. In April 2002 the IMFC encouraged the IMF to pursue a twin-track approach to further develop both the SDRM and the contractual approach founded on CACs, underlining that they were meant to be “complementary and self-reinforcing”. In practice, however, the debate polarized around these two extreme views. The U.S., a few major emerging market debtors, and representatives of the financial industry lined up in favor of a market-oriented approach centered on CACs; on the other side, many others were in favor of the SDRM.

Despite several revisions of the initial proposal of November 2001 aimed at reducing the role of the IMF, the lack of consensus around the SDRM became clear at the Spring Meetings of April 2003, when the IMFC recognized that it was “not feasible … to move forward to establish the SDRM”.

The demise of the SDRM was only partly due to the fact that it would have required an amendment to the Fund’s Articles of Agreement. Perhaps the key reason for its defeat was the substantial transfer of sovereignty that it would have entailed, notably in terms of setting up an international court with authority over the handling of sovereign debt (including a statutory basis for suspending legal procedures against a country). Concerns about such a loss of sovereignty were shared by many emerging market countries (Group of Twenty-Four, 2003). Nevertheless, Krueger’s proposal is credited with having dramatically changed the terms of the discussion, giving impetus to the adoption of CACs as a more acceptable alternative to a statutory mechanism (Cohen and Portes, 2006; Quarles, 2010).

Official calls for a broader use of CACs remained essentially disregarded until the aftermath of the Argentine crisis of December 2001, as the bulk of sovereign bonds continued to be issued without such provisions. A breakthrough was made in April 2002, when the G-7 countries adopted an action plan promoting, inter alia, the use of CACs in international bonds, as the key component of a market-based approach to sovereign debt restructuring (Group of Seven, 2002).

The then U.S. Under Secretary of Treasury for International Affairs John Taylor, who came to play a crucial role in promoting CACs (see Taylor, 2007, Chapter IV), noted that these clauses could be freely determined by borrowers and lenders on their own terms; he also proposed that the legal features of such clauses conform to several essential guidelines (Taylor, 2002a and 2002b). Taylor’s proposal was in the tradition of Eichengreen and Portes (1995), but went further, first by suggesting new clauses on how to organize the debt restructuring process (initiation clause and engagement clause), in addition to the standard majority action clause; and second, by addressing more explicitly the problem of incentives to implement CACs.

Importantly, according to Taylor, the dissemination of CACs would have positive effects also on IMF lending: by describing what happens if a country faces a debt default, they would strictly complement the rules that were being introduced to discipline large-scale IMF loans (the exceptional access policy – see Section IV). This, as a result, “added much more predictability to the way bond markets work and has enabled the IMF to be more rule-like or systematic” (Taylor, 2006).

In June 2002 a Working Group on Contractual Clauses was set up by the G-10 Ministers and Governors, with the aim of defining a template of contractual provisions to make the resolution of debt crises more orderly (for more details see Group of Ten, 2002; Elderson and Perassi, 2003; and Group of Ten, 2004). At the same time, some trade associations in the U.S. private sector, grouped into the Gang of Seven, also developed their own proposals on model clauses.

B. The Dissemination of CACs until the Euro-Area Debt Crisis

The reaction of both sovereign issuers and institutional investors to the official sector’s renewed call for CACs was initially far from enthusiastic. When the use of these clauses was first suggested by the G-10, the private sector was broadly satisfied with the status quo and did not favor major innovations, major losses on sovereign securities not yet having materialized. Besides, in the view of many market participants, bonds represented an arm’s-length source of finance, and the obligation to repay them had to be regarded almost as sacred by the debtor. A further reason for the reluctance of private investors to accept the official sector’s views was “the suspicion that this intervention may tilt the balance too much in favor of the debtor” (Group of Ten, 1996, p. 12). Bondholders also feared that easing a restructuring would encourage strategic (opportunistic) defaults, on the grounds of a parallel drawn with domestic bankruptcy law, where steps to shield creditor interests provide a debtor with the potential to seek protection through bankruptcy even though it has the resources to pay.

On the other hand, sovereign borrowers shared the opposite worry (IMF, 2001, p. 14). They feared that the activation of formal negotiations would strengthen the degree of organization and cooperation among creditors, thereby increasing their leverage during the restructuring process. Moreover, they were concerned that if CACs were interpreted as signaling a lessened commitment to honor contractual obligations, they could raise external financing costs.

In this regard, economic theory does not provide univocal guidance on whether and how CACs affect the pricing of bonds, as in principle both upward and downward effects are possible. On the one hand, the view that CACs would push up the cost of external financing was mainly justified with reference to the debtor moral hazard that could be induced by these clauses: the presence of CACs would not only make it easier to restructure sovereign debt, it would also signal that borrowers could be less willing to honor their obligations to bondholders, thus increasing the probability of observing an actual restructuring. On the other hand, others have argued that CACs could actually decrease bond financing costs, insofar as they reduce information asymmetries (and the attendant lemon problems) between parties and contain the risk of a disorderly grab-race for the sovereign’s assets, ultimately increasing the expected recovery value of defaulted debt.

In the face of this theoretical debate, the empirical analysis has tried to shed light on the relationship between CACs and sovereign bond pricing, mainly (and prior to 2003) by comparing U.K.-style international bonds (which have long incorporated CACs) with equivalent U.S.-style instruments (without CACs). Overall, after surveying the available empirical studies on the effects of CACs, Häseler (2009) concludes that sovereign bonds with CACs do not generally carry higher yields than those without. This conclusion is in line with some more recent findings by Bradley and Gulati (2012) on bonds governed by foreign law, while Bardozzetti and Dottori (2012) find that the inclusion of CACs lowers most yields for bonds of sovereign issuers in the middle of the rating scale.

The official sector’s push for a broader use of CACs started to bear fruit in February 2003, when Mexico launched an unprecedented $1 billion bond issuance containing CACs on the U.S. market. Mexico’s move took the markets by surprise since in the months prior to the issue some country officials had expressed skepticism about the possibility of adopting CACs. According to Gray (2004), the move was motivated by concerns about the approval of the SDRM and was a means to pre-empt other issuers from setting different market norms with an unattractive CAC language.

Despite fears about a possible increase in spreads, the issue paid no discernible yield premium. As a result, the Mexican move helped to strengthen the commitment to promote the adoption of CACs at the IMFC meeting of April 2003. On that occasion, the strategy gained conclusive momentum when the Finance Ministers of EU countries announced that they would include CACs in the documentation of bonds issued under foreign jurisdiction (European Council, 2003).

Following Mexico, a growing number of sovereign debtors (from both advanced and emerging economies) started to include CACs in their new bonds issued under New York law. As for advanced economies, Italy took the lead by introducing CACs in the documentation of all its New York governed global bonds starting from June 2003.

It is important to document how the diffusion of CACs proceeded in the subsequent years. This is not straightforward, given the lack of comprehensive data collection systems specifically targeted at ascertaining the types of clauses included in sovereign bonds. What we know today is the outcome of ad hoc research on individually-constructed bond databases (see among the others, Drage and Hovaguimian, 2004; Gugiatti and Richards, 2004; and Bradley et al.,). According to these studies, while until 2002, the overwhelming fraction of sovereign bonds issued under New York law did not include CACs, by the end of 2003 nearly 70 percent (by value) of all new sovereign bonds issued under the New York governing law contained these clauses; this fraction had risen to more than 90 percent by 2005, and the most recent estimates point to the same number for 2010 as well (Bradley and Gulati, 2012, p. 13).

These findings suggest that, on the eve of the recent crises in the euro area, the share of sovereign bonds issued internationally and containing CACs had possibly more than doubled since 2003. Yet, despite this progress in the case of sovereign bonds governed by foreign law, it appears that the actual inclusion of collective action clauses in sovereign bonds governed by domestic law has generally been far from widespread. Furthermore, according to BIS data on government debt securities, the international share of such securities seems to have fallen over the last decade, partly reflecting the increasing importance of local currency bond markets for emerging market economies. This issue will be discussed further in Section V.

IV. The IMF’s Lending Framework: the First Phase, from the Mid-1990s to 2008

After illustrating the CAC-based pillar of the international framework for dealing with sovereign debt crises, we now review the reforms that have defined since the mid-1990s the most important pillar, represented by the IMF lending framework. We discuss these reforms together with the circumstances that led to their adoption and to key lending decisions. We believe that this historical perspective will help towards a better understanding of the impact of the global crisis on that framework.

Our review begins in the mid-1990s for two main reasons: (a) the 1994-95 Mexican crisis marked a turning point as regards the provision of large-scale Fund financing,7 the repeated use of the exceptional circumstances clause (ECC) and, more generally, the attention paid to the role of creditor coordination problems in sovereign debt crises; (b) the crises of the second half of the 1990s led to the establishment of new IMF financing facilities, explicitly aimed at addressing capital account crises and contagion risks.

The Fund entered that decade with a lending framework comprising a set of policies and a toolkit of financing facilities with different purposes and financial terms. These elements were supported by a de facto status of preferred creditor, which aimed to insulate the IMF from the risk of non-repayment by a member, as well as from possible litigation with private creditors in the context of a sovereign debt restructuring.8 Access to the Fund’s financial resources was disciplined by ordinary access limits and the ECC, which allowed these limits to be surpassed in unspecified extraordinary situations.

Importantly, the Fund’s toolkit also comprised the Lending into Arrears - LIA policy, which governs Fund lending to countries with payments arrears vis-à-vis their external creditors (mainly commercial banks at that time). Broadly speaking, the LIA policy specifies the conditions under which these countries can expect to receive loans from the Fund, and aims to limit the economic losses associated with the resolution of unsustainable debts while maintaining adequate safeguards for IMF resources.9

The framework evolved markedly over the following years, largely in response to major crises, which normally act as a powerful catalyst for change. Discussions have revolved mainly around such key aspects as the type of balance-of-payments (BoP) need under which a country can request the Fund’s assistance (actual versus potential); the definition of a norm for access by country members to Fund resources (access limits); and the special criteria and procedures for access beyond these limits (the exceptional access policy).

A. Balance-of-Payments Need, Access Limits, and the Exceptional Circumstances Clause

The notion of need is expressly contemplated in the IMF’s Articles of Agreement and is central to all forms of Fund lending. The identification of a balance-of-payments need represents at the same time a prerequisite for and an upper bound to Fund financing, as the latter should always be justified by a need and should never exceed its extent (IMF, 1994).

While attempts to clarify the notion of need date back to the mid-1970s, the IMF has been reluctant to establish specific guidelines in regard, reflecting a concern that such a codification could create unnecessary inflexibility.

Interestingly, in 1994 the Executive Board examined three different notions of need that could be applied under the Articles of Agreements: present need, i.e. experienced at the time when Fund financing is requested; prospective need, which is “expected to arise in the future as a result, inter alia, of the implementation of an economic program”; and potential need, which “may, but is not expected to, arise during the period of the arrangement” (IMF, 1994, p. 2).

These notions were held to be equally important for justifying loans made under all Fund arrangements. Consistent with this view, new IMF-supported programs of a precautionary type began to proliferate in the 1990s (see Chart 1).10

Chart 1 -
Chart 1 -

IMF Lending Arrangements (*), Credit Outstanding, and Undrawn Balances Under GRA

(SDR millions)

Citation: IMF Working Papers 2013, 020; 10.5089/9781475578713.001.A001

(*) Red dots indicate precautionary arrangements in which the beneficiary country did not make any drawings during the life of the arrangement. Source: authors’ elaborations on IMF data.

Normal access limits to IMF resources play the important role of making members confident about the scale of possible financing while preserving Fund resources and their revolving character. They have existed since the creation of the Fund, and have been altered over time.11 Since October 1992, there have been two types of normal limits: annual and cumulative. They were initially set at 68 percent and 300 percent of each member country’s quota in the Fund, respectively. In October 1994, the annual limit was raised to 100 per cent to support the transition economies while the cumulative limit remained unchanged. In March 2009, both limits were doubled as part of a major overhaul of the Fund’s lending and conditionality framework (see Section V).

Normal limits set important thresholds beyond which access decisions are subject to greater scrutiny through the Exceptional Access Policy (see the next Section). In order to increase the scope and flexibility of Fund lending, however, the possibility to “approve access in excess of … access limits in exceptional circumstances” was introduced (IMF, 1984). This provision became known as the exceptional circumstances clause; it was originally included in the decision to establish, in February 1979, the Supplementary Financing Facility (SFF), the Fund’s first effort to supplement its own resources with borrowing from other official creditors for its regular arrangements (Boughton, 2012, p. 752).

It is important to note that during those years the IMF’s Executive Board refrained from defining the notion of exceptional circumstances in any precise manner, on the grounds that members’ financial needs might vary and that the Fund should retain the flexibility to provide timely exceptional access in unforeseeable situations. For example, in 1983 the Board was opposed to singling out the impairment of the international monetary systems as a criterion for invoking the ECC, because it might have implied special treatment for larger countries (IMF, 1983). In any event, it was underscored that any use of the ECC was expected to be very rare (IMF, 2004b, p. 11).

Under the Enlarged Access Policy in place between January 1981 and November 1992, the ECC became almost moot (Boughton, 2001, p. 879), as the above policy was able to meet most borrowers’ financing needs within the normal limits. During the sixteen years following its establishment, the ECC was invoked only three times. However, the 1994-95 Mexican crisis was a turning point, as in February 1995 the ECC was invoked to enlarge the Fund’s stand-by arrangement for that country. Subsequently, it was invoked eleven more times for seven other countries through the end of 1999 in order to address the wave of financial crises that occurred in the second half of the 1990s (Boughton, 2012, p. 752).

This repeated reliance on a clause that had been intended to remain of an exceptional nature became a source of concern. On the one hand, there was a fear that the proliferation of large-scale IMF loans could exacerbate moral hazard problems on both the creditor and debtor sides, thereby increasing the likelihood of new crises in the future. On the other hand, the margins of discretion allowed by the unconstrained ECC could indeed leave potential borrowers in doubt as to whether the IMF would be willing to lend to them should they get into trouble.12 While this ambiguity may have reflected the intent to foster prudent macroeconomic policies and avoid moral hazard, at the same time it led to a perception that the Fund’s lending behavior was not sufficiently systematic.13

According to Taylor (2006), the Fund’s responses to the crises that occurred from 1994-95 (Mexico) to 1998 (Russia) lacked adequate predictability, and this factor could per se account for the contagion observed at the time.14 Importantly, the same author traced this insufficient predictability to the absence of a clear framework defining how the IMF should operate in exceptional situations that call for financing above normal limits.

B. The Exceptional Access Policy in Capital Account Crises (2002-2003)

Under the pressure of events, two new high-access facilities were created in the midst of the Asian crises of 1997-98, embodying considerable departures from previous Fund practice. The Contingent Credit Line (CCL) was in principle the most innovative of these, given its precautionary nature, as it was meant to avert crises through a form of ex ante conditionality (i.e. pre-qualification criteria to identify eligible countries). Yet, the CCL remained unattractive to its potential users, mostly because of fears about the possible stigma, and was left to expire in 2003.

The Supplemental Reserve Facility (SRF), instead, aimed to address actual capital account crises. It required a strict circumstance test to be met, as it could be available only when a member was experiencing “exceptional balance of payments difficulties due to a large short-term financing need resulting from a sudden and disruptive loss of market confidence reflected in pressure on the capital account and the member’s reserves” (IMF, 2003a). The SRF was based on penalty rates and had only a short-term horizon; it could be provided if there was “a reasonable expectation that the implementation of strong adjustment policies and adequate financing will result, within a short period of time, in an early correction of such difficulties”. Importantly, the facility was not subject to access limits and could thus be used to grant large-scale loans without invoking the ECC.

As a result of these developments, the rules followed by the Fund in its access policy became somewhat unclear. Were ordinary limits meant to apply to all or only to some facilities? Should capital account crises be handled through the SRF only or through a varying mix of facilities, depending on circumstances? But the most challenging issues were posed by the high frequency, since mid-1990s, of cases of exceptional access, triggered by capital account crises, which required the activation of the ECC. In this regard, it was noted that “a policy which rests on too-frequent general claims of exceptional circumstances is not satisfactory” (IMF, 2002c, p. 11).

For all these reasons, it became apparent that the prevailing crisis management regime needed to be strengthened through a framework and new rules to clarify, guide, and constrain access decisions in order to ensure that Fund financing in excess of normal limits remained truly exceptional.

A landmark step in addressing these issues was the new Exceptional Access Policy in Capital Account Crises (EAP). This was formally developed in the period September 2002-February 2003 (IMF, 2002c, 2003a, 2003b, and 2003c), along the lines envisaged in the G-7 Action Plan of April 2002.

Building on the principles laid down by the Prague Framework of September 2000, notably that Fund resources had to be limited and that extraordinary access to them should be exceptional, the Executive Board, while keeping the existing normal access limits unchanged, agreed on a set of substantive criteria to be met (and decision-making procedures to be followed) when exceptional access under the ECC was requested by a member country experiencing pressures on its capital account.

The following four criteria for exceptional access were established:

  • Criterion 1.0 (Exceptional BoP pressures) – The member is experiencing exceptional balance-of-payments pressures on the capital account resulting in a need for Fund financing that cannot be met within the normal limits;

  • Criterion 2.0 (Debt sustainability in the medium term) – A rigorous and systematic analysis indicates that there is a high probability that debt will remain sustainable;

  • Criterion 3.0 (Expectation of re-entry to capital market) – The member has good prospects of regaining access to private capital markets within the time Fund resources would be outstanding, so that the Fund’s financing would provide a bridge;

  • Criterion 4.0 (Strong program design and implementation) – The policy program of the member country provides a reasonably strong prospect of success, including not only the member’s adjustment plans but also its institutional and political capacity to deliver that adjustment.

Given the emphasis that the Articles place on the temporary nature of Fund assistance and the revolving character of its resources, the criterion on debt sustainability acquires particular relevance. Although debt sustainability had already been the inspiring principle for all types of IMF lending, a key merit of the new policy was to give more visibility to rigorous, country-specific Debt Sustainability Analyses (DSAs) as a crucial input for the decision on whether exceptional access was justified or a debt restructuring was warranted instead. Such analyses were due to receive continued methodological refinements in subsequent years, and they have rightly become a standard feature of the decision-making process that supports IMF lending (IMF, b).

Another merit of the new policy was to clarify what should be done when a formal rescheduling or restructuring of debt (i.e. the field of the LIA policy) was warranted in order to restore the viability of the balance of payments. It was openly recognized that in such instances the four substantive criteria of the EAP would generally not be met and the conditions for using the SRF would not apply (IMF, 2003a, p. 20). As a result, the Board agreed that Fund lending to a member undertaking a restructuring of its sovereign debt “would ordinarily be expected to be within normal access limits” (IMF, 2003c). This principle was rooted in the idea that, without any prospect of regaining access to private market financing in the short term, Fund resources could be committed into an unsustainable debt situation only until the process of restructuring was completed.

Importantly, however, the Board also recognized that there could be rare cases in which exceptional access might be warranted despite a looming debt restructuring. Such circumstances could arise when the country was not in a position to make significant repayments (net repurchases) to the Fund and there were large scheduled repurchase obligations. Other exceptional circumstances that could justify Fund support, despite the risks of adding to the debt burden of an already over-indebted sovereign, were when additional financing would support the member’s strategy for limiting economic disruption associated with a debt restructuring.

Against this background, it is crucial to underscore that, in order to achieve one of its key objectives (namely to help shape the expectations of both Fund members and private market participants on the availability of Fund financing), the EAP needed to provide “as much clarity as possible in defining cases where exceptional access in financial crises was appropriate, and where it was not” (IMF, 2002c, p. 13, emphasis added).

In other terms, clarifying the circumstances under which exceptional access to Fund resources would not be provided (thus implicitly pointing to the need for a debt restructuring) was meant to be as important as the goal of establishing the criteria that would justify the provision of large-scale Fund loans. In turn, making markets well aware about the availability (or lack) of such loans, as well as about their conditions and limits, was seen as instrumental in fostering market discipline ex ante by strengthening “incentives for prudent risk management by policy makers and financial markets alike” (IMF, 2002f). (These considerations will be further expanded in Section IV)

Be that as it may, the new lending regime had to be made credible with coherent actions, as it ultimately rested on the ability to avoid exceptions, which could be perceived as permanent changes in the rules. Most prophetically, in one of the earlier reviews of the new policy, IMF staff considered that while capital account crises were “likely to be intermittent”, the related decisions on access were “expected to remain among the most important and difficult the Fund will make” (IMF, 2004b, p. 7).

C. Earlier Tests of the Exceptional Access Policy

Although all exceptional access cases between 1995 and 2002 related to capital account crises, during the discussions that had led to the approval of the EAP Fund staff had already pointed to the possibility that other circumstances could arise that might require exceptional access, but for which the four substantive criteria would not be relevant (IMF, 2003a, p. 5). As a result, the Executive Board agreed that, as a bare minimum, the procedural requirements of the new policy had to be applied anyhow, “even when the member is not experiencing a capital account crisis” (IMF, 2003b).

In late 2003, the new lending framework was indeed tested under conditions other than those that had characterized previous capital account cases. Two major emerging market countries, Argentina and Brazil, approached the Fund to seek financial support while already having in place high-access arrangements that had been agreed before the entry into force of the new policy. These countries’ requests, which entailed the prolongation (Argentina) or augmentation (Brazil) of access beyond normal limits, were assessed against the four substantive criteria for exceptional access, and neither request met all of them.

Argentina’s case was the most difficult one. Pending the final approval of the EAP, in January 2003 a transitional Stand-By Arrangement (SBA), with exceptional access of about 150 percent of quota, had been provided under the ECC; it was followed by a new 3-year SBA arrangement approved in September 2003 that granted Argentina exceptional access of over 420 percent of quota, despite the presence of payment arrears to private creditors and the apparent need to restructure its sovereign debt after the default of December 2001. As a result, the country failed to meet the second and third criteria of the EAP. Since Argentina’s debt was judged to be unsustainable, this did not come as a surprise. As noted earlier, in establishing the EAP in 2002-03, it had been recognized that countries in need of a debt restructuring would not meet all of the criteria. In these cases, the EAP dictated that access would normally be expected to be within the access limits, while at the same time recognizing that there could be rare circumstances warranting exceptional access.

Both the January and September SBAs for Argentina were predicated under the LIA policy;15 the latter loan was approved in light of the room for exceptional access in a debt restructuring context provided by the new EAP.

The case of Brazil was different, as the country had a reasonably sustainable debt profile and continued access to financial markets, and it was not experiencing actual BoP pressures (thus failing to meet the first EAP criterion). However, Brazil needed to repay a substantial portion of IMF credit outstanding in the next months.

In that context, Fund staff argued that exceptional financing was warranted in order to cope better with the country’s scheduled repayment obligations to the Fund (which could hamper the restoration of market confidence) and facilitate Brazil’s exit from official financial support. The extended program entailed additional financing of 150 per cent of quota, for a total cumulative amount of about 340 percent (IMF, 2003g).

In sum, the requests of Argentina and Brazil were consequential as they clearly brought to the fore the issue of how to apply an access framework explicitly designed for capital account crises in cases in which a country was not experiencing such pressure, in particular where (a) the member’s BoP need was related to a pre-existing large exposure to the Fund, or (b) the BoP need was potential rather than actual (i.e. in a precautionary setting).

These rather challenging issues were addressed by the Fund in subsequent EAP reviews (March 2004 and May 2005) by building on the overarching principle that the Fund must retain discretion to lend in emergencies, while acknowledging that this discretion had to remain subject to constraints.

The Executive Board agreed that the enhanced decision-making procedure for exceptional access would continue to apply in the rare instances where a need for exceptional access could arise in circumstances other than a capital account crisis. However, the request would be judged only in light of the four substantive criteria and its approval “would not necessarily be conditioned on meeting these criteria” (IMF, 2005, p. 15). Accordingly, in non-capital account crises, access to Fund resources could still be granted on the basis of the flexibility allowed by the exceptional circumstances clause (IMF, 2004b and 2005).

The above discussions paved the way for the approval, in 2005, of two high-access arrangements for Turkey (May) and Uruguay (June). These countries were facing substantial BoP needs, but neither of them was experiencing pressures on its capital account. As the first EAP criterion was not met, both countries were granted exceptional access to Fund resources under the ECC.

D. Greater Predictability as the Common Thread of Reforms

Before moving on to examine the impact of the global crisis on the Fund’s lending framework, it is important to take stock and underscore what in our view was the key common thread running through the reforms discussed so far.

This thread can be summarized in two words: greater predictability. In essence, the most important objective of these reforms (notably the newly-established exceptional access policy, the revisions of the lending into arrears policy,16 and the promotion of CACs) was to make the official sector’s behaviour more predictable and rule-based than before, with a view to ultimately limiting the use of official financing in the resolution of sovereign debt crises. To be sure, this search for greater predictability was not an end in itself; rather, it was driven by the belief that the expectations of both Fund members and private market participants needed to be better shaped in order to foster market discipline and thus help prevent crises.

As regards IMF lending, the development of more clearly and narrowly defined conditions for exceptional access was instrumental in better distinguishing between (a) cases where financial assistance in excess of normal limits was warranted to support a country’s macroeconomic adjustment, and (b) cases where large-scale loans would not be provided and should not be expected, for example (and perhaps most significantly)17 because a sovereign debt was unsustainable and a restructuring needed. Evidently, the latter cases are the most important for enforcing a more rule-based approach to sovereign debt crises that fosters market discipline and prevents unjustified market expectations (and risk of moral hazard) about an unwarranted deployment of sizeable Fund resources. In this regard, the EAP was very explicit in aiming “to reduce the risk that Fund members or market participants would expect exceptional access to be available in situations where a member’s debt burden was clearly unsustainable” (IMF, 2003a, p. 4).

It is therefore no wonder that, according to Fund rules, the exceptional access policy is normally not applicable in situations where the sustainability of a sovereign debt is under question. In this context, as noted before, greater predictability as to when exceptional Fund financing will not be available is indeed meant to reinforce market discipline ex ante, on the assumption that, when a sovereign debt is on an unsustainable path, markets will request higher risk premia that should force early corrective actions in order to prevent a full-blown crisis. In other terms, EAP rules are ideally expected to work as a force by which a tool for crisis management turns out to promote crisis prevention.

The discussions that led to the adoption of the 2002-03 EAP also offer a vivid illustration of the trade-offs faced by any policy that defines access to Fund resources (and, more generally, to official financing). On the one hand, ensuring greater predictability of Fund lending in the eyes of private market participants was regarded as quintessential, as it could reduce market uncertainty, foster market discipline, and help involve the private sector in cases where sovereign debt was unsustainable and a restructuring appeared appropriate. On the other hand, in some circumstances such predictability was deemed to carry risks of moral hazard, which would instead point to the need to retain some “constructive ambiguity” in the Fund’s lending similar to the lender of last resort function of central banks.

Greater predictability was also a central feature of the Fund’s LIA policy. By making Fund financing available under certain conditions, despite the presence of sovereign arrears, the policy aims to facilitate the process for the resolution of an unsustainable sovereign debt, also making it more timely and predictable, most notably by requiring a sovereign to make a good faith effort to reach a collaborative agreement with creditors. In turn, the awareness that Fund financing could be deployed would provide incentives to accelerate an orderly debt restructuring.

It is also worth noting that ensuring greater predictability of the process for sovereign debt restructuring (by reducing the risk of litigation) was also an explicit objective of the contractual approach based on CACs. In common with the alternative approach based on the Sovereign Debt Restructuring Mechanism (SDRM), CACs recognize that an early approach to creditors is a necessary feature of a crisis resolution framework that aims to contain the cost of debt restructuring for all parties concerned (IMF, 2002b).

Finally, it is important to recall the emphasis that the G-7 Action Plan of 2002 placed on the fact that the IMF’s exceptional access policy and mechanisms (such as CACs) to facilitate an orderly debt restructuring were strictly complementary and mutually reinforcing. A rule-based reform of the policy governing large-scale IMF loans was inseparably linked to a reform of the process for sovereign debt restructuring (Taylor, 2007, p. 110). Indeed, the very existence of ex ante mechanisms to mitigate the costs of sovereign defaults or debt restructurings was meant to be a key prerequisite for the IMF to be more predictable in its lending behavior and to avoid the risk of moral hazard attached to its loans. In the same vein, insofar as CACs facilitate crisis resolution, they may limit the risk of observing a full-blown crisis in the first place and thus help to limit the need for Fund intervention. By contrast, without a process for the orderly resolution of an unsustainable sovereign debt, the Fund’s goal of adhering to limits or rules would face a typical time inconsistency problem and would thus not be credible.18

All in all, the two main policy reforms ushered in by the G-7 Action Plan have been credited as being among the key reasons for the absence of crises in emerging market economies since 2002, as opposed to the string of crises and widespread contagion that afflicted the years from the 1994-95 Mexican crisis to the Uruguayan crisis of 2002 (Taylor, 2007).

At the same time, these reforms were incomplete in a number of respects and contained several important gaps, of which perhaps the most notable related to the differential treatment in the EAP between capital account and non-capital account crises. Pressure to fill this and other gaps (e.g. the treatment of actual versus potential BoP needs) lost considerable impetus during the years that followed, when the demand for IMF financing fell to very low levels (see Chart 1). These problems were eventually addressed only in the context of the crisis-driven reforms of March 2009 (see Section V).

V. The Impact of the Global Crisis on the International Framework for Dealing with Sovereign Debt Crises

The global crisis that first broke out in the summer of 2007 and its spillover into the sovereign sector of some euro-area countries have had an unprecedented impact on all the components of the existing international framework for dealing with sovereign debt crises. While the crisis affected all of the Fund’s core functions, its most significant impact was on the lending framework, providing a strong case for re-assessing the latter’s adequacy and effectiveness.

To begin, the crisis originated at the very heart of the global financial system; contagion and systemic risks reached exceptional levels, also affecting countries with reasonably sound fundamentals and policies (including many emerging market economies). The IMF began to lend again through its traditional facilities, but the crisis exposed gaps in the toolkit, particularly for large and frontloaded financing, while heightened contagion risks underscored the need for new precautionary tools.

In order to improve its ability to respond to the different and evolving financing needs of its membership, in March 2009 the Fund approved a major overhaul of its lending framework (IMF, 2009c).

Given the systemic nature of the crisis and the consequent expectation of a fast-growing demand for Fund loans, such an overhaul needed to be supported by a sizeable increase in the Fund’s financial firepower, so as to boost confidence in its ability to intervene effectively. To this end, in a key complementary move in April, the G-20 leaders announced a tripling of IMF financial resources (from $250 billion to $750 billion) and an extraordinary SDR allocation of $250 billion (Group of Twenty, 2009).

Overall, we believe that these reforms marked the beginning of a new phase of Fund lending, one in which systemic risks, contagion, and international spillovers would play a much greater role than in the past across all of the institution’s core functions. The reforms were necessary in order to address the distinctive challenges posed by the global crisis more effectively with new tools.

In the same vein, the debt problems experienced by some euro-area countries since 2010 also drew renewed attention to the use of CACs as a tool to help address an unsustainable debt burden. Importantly, in the end these clauses were activated by Greece during a debt exchange undertaken in Spring 2012.

The impact of the global financial crisis and the subsequent euro-area sovereign debt crises on the above two pillars is reviewed in detail in the rest of this Section.

A. Overhaul and Systematization of the Fund’s Lending Toolkit

The reforms adopted by the Fund since March 2009 were very comprehensive (IMF, 2009a and 2009b). They included a rationalization of its existing lending facilities (some of which were eliminated), the doubling of ordinary access limits (unchanged since 1994) to 200 per cent of quota annually and 600 percent of quota cumulatively, changes in the facilities’ lending terms (cost and maturity structures), and the formalization of High-Access Precautionary Stand-by Arrangements (HAPAs) to improve the effectiveness of this key facility as a crisis prevention instrument. Changes were made also in the Fund’s conditionality, to make it more focused and tailored to country-specific fundamentals and policies.

A landmark innovation was the introduction of a new facility of a precautionary nature, the Flexible Credit Line (FCL), reserved to countries with very strong fundamentals and policies. In September 2010, it was complemented by the Precautionary Credit Line (PCL), which was subsequently replaced by the Precautionary and Liquidity Line (PLL) in December 2011. These two facilities were intended to provide effective crisis prevention to members with sound fundamentals, policies, and institutional policy frameworks but moderate vulnerabilities that would not meet the FCL’s higher qualification standards.

The most distinguishing feature of these new precautionary facilities is the absence (in the case of the FCL) or the focused nature (for PCL/PLL) of traditional ex post conditionality, which has been replaced (FCL) or complemented (PCL/PLL) by a number of pre-defined qualification criteria (ex ante conditionality). One of their key objectives is to reduce the perceived stigma of borrowing from the IMF, and to encourage countries to ask for assistance before they face a full-blown crisis.

The innovations to the Fund’s lending toolkit have been matched by equally important changes to the rules governing its exceptional access policy. These changes mostly aim to (a) resolve two asymmetries in the treatment of capital versus non-capital account crises, and (b) broaden the scope of the policy so as to cover both actual and potential BoP needs.

The first asymmetry had already been discussed during the review of the EAP. It related to the fact that while in cases of capital account crisis the policy’s four substantive criteria had to be met to justify access above the normal limits, such criteria only needed to be assessed (but not necessarily satisfied) for non-capital account crises. A concern emerged that this kind of flexibility could have generated a perception that access decisions in non-capital account crises were “ad hoc and unpredictable” (IMF, 2009a). In turn, this insufficient clarity might affect the credibility of the EAP, as use of the exceptional circumstances clause to approve non-capital account exceptional access requests was “perceived (incorrectly) as constituting an exception to the framework” (IMF, a, p. 19). Moreover, it was noted that the overall EAP framework had the ironic effect of constraining exceptional access in those cases (capital account crises) where it might have been most appropriate, while allowing greater flexibility in other cases.

The second asymmetry pertained to the debt sustainability criterion of the EAP. The original formulation of this criterion effectively precluded exceptional access when the debt position, at the time of the member’s request for financing, was judged to be unsustainable “even if sustainability could be restored through policy adjustment and/or debt restructuring” (IMF, 2009a, Box 5, p. 24). The same restriction was not binding in cases of non-capital account crisis, as the debt sustainability criterion did not have to be satisfied. In other terms, this criterion lacked the forward-looking perspective needed to factor into the Debt Sustainability Analyses any credible strategy to address the debt situation and restore sustainability. Moreover, the debt sustainability criterion was unclear in a number of respects (e.g. it did not specify whether both public and private debt had to be assessed), pointing to a need to clarify the scope and methods of Debt Sustainability Analyses.

Against this background, in March 2009 the first two substantive criteria for exceptional access in capital account crises were amended as follows (IMF, 2010b, pp. 426-427):

  • Criterion 1.1 – The member is experiencing or has the potential to experience exceptional balance of payments pressures on the current account or the capital account, resulting in a need for Fund financing that cannot be met within the normal limits.

  • Criterion 2.1 – A rigorous and systematic analysis indicates that there is a high probability that the member’s public debt is sustainable in the medium term. Debt sustainability for these purposes will be evaluated on a forward-looking basis and may take into account, inter alia, the intended restructuring of debt to restore sustainability. This criterion applies only to public (domestic and external) debt. However, the analysis of such public debt sustainability will incorporate any potential contingent liabilities of the government, including those potentially arising from private external indebtedness.

It is important to underscore that the inclusion of the potential BoP need in Criterion 1 allowed only some types of precautionary lending to be brought under the umbrella of the EAP. The FCL was not subject to the EAP itself, given its rigorous qualification requirements (ex ante conditionality), and its procedures that were substantively similar to those under the EAP, particularly as regards to the Board’s early involvement (IMF, 2009d). Conversely, arrangements under the PCL (and its successor, the PLL) were subject to the EAP (where relevant).

For its part, Criterion 2 contained the important additional specifications that (a) the member’s public debt (both domestic and external) should be sustainable in the medium term; and (b) sustainability assessments should take both planned debt restructurings into account (something that underscores the links between the EAP and LIA policy) and potential contingent liabilities of the government.

B. The Greek Crisis, Contagion, and Systemic Risk

In the twelve months following the March 2009 reforms, several financing arrangements were approved by the Fund under the revised lending framework, including a few of a precautionary nature. While these arrangements generally did not raise particular problems, the revised EAP received its baptism of fire with the Greek crisis of 2010, which was mainly triggered by idiosyncratic factors, i.e. a negative fiscal surprise against the backdrop of a number of fundamental macroeconomic imbalances (see, among others, Bini Smaghi, 2010 and 2011a, for a discussion of the origins of the sovereign debt crisis in Europe).

Within the new phase of Fund lending that began in March 2009, the Greek program of May 2010 represented a landmark moment. Two consequential developments stand out, both prompted by unprecedented threats of a systemic crisis in the euro area and attendant global spillovers.

First, the financing needs of Greece turned out to be extremely large by historical standards (especially in terms of Fund quotas), reflecting among other things large debt rollovers and systemic contagion risks given strong economic and financial integration within the euro area. Accordingly, the already sizeable resources committed by the Fund for the country (€30 billion, equivalent to an unprecedented 3,200 per cent of quota19) had to be supplemented by even larger bilateral contributions (€80 billion) from other euro-area member countries in the form of a single loan agreement signed by the European Commission on their behalf with the Greek authorities. For the first time ever in the Fund’s history, program conditionality for Greece was devised, negotiated, and monitored by a Troika composed of the IMF, the European Commission, and the European Central Bank. The joint rescue package of €110 billion was approved on May 9.

On the very same day that the Fund approved the Greek program, the Economic and Financial Affairs’ Council of the European Union agreed to establish a new regional financial firewall (the European Financial Stability Facility - EFSF, replaced by the permanent European Stability Mechanism in October 2012) to support joint EU-IMF rescue packages for euro-area countries.

Second, the EAP had to be amended further to factor in the threat of systemic risk to the euro-area financial system as a whole. This amendment was introduced during the same Board discussion that approved the Stand-by Arrangement requested by Greece.

When the compliance of Greece’s request with the EAP criteria was assessed, Fund staff could only argue that, on balance, the country’s debt was “sustainable over the medium term, but the significant uncertainties around this [made] it difficult to state categorically that this [was] the case with a high probability”. This notwithstanding, Fund support at the proposed level was deemed to be justified given the “high risk of international systemic spillover effects” (IMF, 2010a, p. 20). The Executive Board unanimously supported the proposed arrangement and also agreed that this approach to assessing debt sustainability would be followed in all cases with a high risk of systemic spillovers.

As a result, the second substantive criterion of the Fund’s exceptional access policy was modified as follows (IMF, 2010b, pp. 426-427):

  • Criterion 2.2 – A rigorous and systematic analysis indicates that there is a high probability that the member’s public debt is sustainable in the medium term. However, in instances where there are significant uncertainties that make it difficult to state categorically that there is a high probability that the debt is sustainable over this period, exceptional access would be justified if there is a high risk of international systemic spillovers. Debt sustainability for these purposes will be evaluated on a forward-looking basis and may take into account, inter alia, the intended restructuring of debt to restore sustainability. This criterion applies only to public (domestic and external) debt. However, the analysis of such public debt sustainability will incorporate any potential contingent liabilities of the government, including those potentially arising from private external indebtedness.

We label the added reference to a “high risk of international systemic spillovers” in the above criterion the systemic clause. Other joint EU-IMF rescue packages with exceptional access were subsequently justified by referring to this clause: for Ireland (December 16, 2010), Portugal (May 20, 2011), and once again for Greece (March 15, 2012). In all these cases, uncertainties about the future public debt paths made it difficult to state categorically that these countries’ debts were sustainable with a high probability. As in the first Greek program, the total size of these packages was impressive (€85 billion for Ireland, €78 billion for Portugal, and over €170 billion for Greece); the IMF committed resources equal to one third of the total for Ireland and Portugal (equivalent to about 2,300 per cent of these countries’ quotas), and to three elevenths for Greece (2,200 per cent of quota). In all cases, financial support was granted under an Extended Fund Facility (EFF) rather than an SBA, given the more structural nature of the required adjustments.

C. Exceptional Access Policy and International Systemic Spillovers

As noted earlier, the response to the global crisis included changes to the rules governing the exceptional access policy that are as significant as the innovations introduced in the Fund’s toolkit of financing facilities. These new rules clarified that: (a) the EAP would cover actual and potential BoP needs relating to both the current and capital accounts; and (b) the exceptional circumstances clause could be invoked only when the four substantive criteria of the EAP were met.

This reinforced the presumption that all large-scale lending decisions of the Fund would abide by the EAP’s criteria without exceptions in the future. These amendments have dramatically constrained (and perhaps virtually eliminated) the scope for discretion previously allowed by the exceptional circumstances clause, which had often been used in the past to justify large-scale loans in situations where debt sustainability could not be systematically and rigorously assessed or where the other EAP criteria could not be applicable in full or in part.

The systemic clause introduced in May 2010 is a most notable development, as it has strengthened the Fund’s ability to provide large-scale financing to address in a timely manner situations of systemic stress with heightened uncertainty. The clause in question is meant to apply to all countries that can originate significant cross-border spillovers, including those that do not necessarily meet the commonly-accepted definition of systemic countries based on their size and interconnectedness.

To put the systemic clause into perspective, as well as to appreciate indirectly how material the threat of systemic disruptions must have appeared in the Spring of 2010, it is instructive to recall that systemic risks had already been examined during the first discussion on the EAP in September 2002. Then, it had been observed that regional and systemic implications were often cited as a potential justification for exceptional access, thus envisaging the possibility of a systemic criterion (IMF, 2002c, p. 14).

This notwithstanding, the view was that “it would be inappropriate to make the systemic criterion a necessary or a sufficient condition for providing exceptional access” (p. 14). The reason was that such a criterion could create a bias towards larger countries gaining access to higher amounts of Fund financing; it was feared this would be potentially inconsistent with uniformity of treatment under the Articles of Agreement, and might also magnify perceived moral hazard risks with regard to large emerging market members.20

Furthermore, it was explicitly stated that “a systemic test (however specified) should not be used to justify exceptional access in circumstances where the other criteria were not met” (p. 14). Specifically, in the case of a high debt burden that could not be made sustainable in the medium term through program adjustment, it would be inappropriate to justify exceptional access solely by the fact that a crisis could have systemic effects.21

However, the discussion recognized, somewhat predicting the 2010 systemic clause, that the Fund should be prepared to provide access above the normal limits in cases where the member’s problems had regional or systemic implications, but only when all criteria of the EAP were met (IMF, 2003c).

To be sure, the systemic criterion discussed in 2002 appears to be stronger than the current systemic clause, which we underscore is neither a necessary nor a sufficient condition to justify exceptional access. Besides, Fund financing that could be granted under the systemic clause needs to satisfy the other three substantive criteria of the EAP. This notwithstanding, it is somewhat ironic that the systemic clause was eventually introduced in 2010 to allow exceptional access for non-large advanced countries, as opposed to large emerging market economies, which were arguably meant to be the main beneficiaries of a possible systemic criterion.

While the systemic clause is certainly a distinctive and well-timed development, it has raised concerns that the evidentiary bar on debt sustainability might be lower when dealing with risks of systemic spillovers (IMF, 2012b, p. 14). By allowing Fund financing also when it is difficult to state categorically that a country’s public debt is sustainable in the medium term with high probability, there may be a risk that the systemic clause might unintentionally conceal, in the eyes of private market participants, the necessary emphasis on debt sustainability as one of the key prerequisites for access to IMF resources, especially for large-scale loans. In other words, there may be a risk that markets perceive this clause as one that keeps up the expectations of Fund-supported financial packages for systemically relevant countries. In turn, this perception, in the extreme, might negatively affect market discipline and the incentives of private investors to make a timely and accurate assessment of sovereign credit risks.

While pertinent in theory, the actual relevance of these concerns remains uncertain, as a number of important factors significantly mitigate them. Even under the systemic clause, the typically more rigorous level of scrutiny required by requests for exceptional access to Fund resources will still be assured by the need to satisfy the EAP criteria and consider all relevant policies and risk mitigating procedures of the IMF, most notably those pertaining to program design and conditionality. In practice, the latter elements have played an especially important role in recent Fund programs for euro-area countries, which were generally accompanied by ambitious and frontloaded fiscal adjustments and stringent structural conditionality. The number and depth of structural conditions has indeed increased in euro-area programs compared with previous programs, including in areas outside the Fund’s core areas of responsibility such as labor and product markets (IMF, 2012b). These developments should be meant as necessary exceptions to the broader effort to streamline Fund conditionality.

D. New Impetus for Collective Action Clauses

The euro-area crisis has brought new impetus to CACs, in terms of both their widespread inclusion in sovereign bonds and their actual use as an instrument of crisis resolution. Two developments stand out.

First, in March 2011, the European Council decided to include standardized and identical CACs in all new euro-area sovereign bonds with maturity above one year (European Council, 2010 and 2011). Following the approval in February 2012 of the Treaty Establishing the European Stability Mechanism, this inclusion has been brought forward to January 2013 rather than July 2013 as originally foreseen. This move is expected to provide decisive momentum to a broader diffusion of CACs in the coming years.

Second, the activation of CACs during the successful debt exchange undertaken by Greece in March-April 2012 points to the relevance of these legal instruments as a useful (if not necessary) component of a sovereign debt restructuring (before Greece, CACs had been used in a few cases involving countries such as Uruguay in 2003 and Belize in 2006; see Bradley and Gulati, 2012).

Yet, despite these developments, the question whether CACs can make a material difference in crisis resolution in the future remains uncertain at this juncture. For one, the use of CACs in the Greek debt exchange presents some specific features that might not be applicable to more general cases. CACs could indeed be inserted retroactively in most of the sovereign debt eligible for exchange, as it was governed by Greek law. This move, enacted through an act of parliament in February 2012, has been debated by some market participants.22

Uncertainty regarding the role of CACs stems from two major concerns, already noted in the early 2000s: (a) the transition problem, that is, the speed at which CACs could spread to the outstanding stock of sovereign bonds; and (b) the aggregation problem, i.e. the fact that CACs usually bind creditors within a single bond issue but not across multiple issues.

Regarding the first concern, we have already noted that the actual inclusion of CACs across the entire spectrum of sovereign debt securities was likely to be less than satisfactory on the eve of the recent sovereign debt crises in the euro area (see Section III). The European Council’s decision to include standardized and identical CACs is certainly welcome, but it only pertains to net flows, and it will therefore take several years before CACs are fully in place. Furthermore, according to BIS data, euro-area countries currently account for about one fifth of global government debt issued domestically and one fourth of sovereign debt issued both domestically and internationally. Therefore, even assuming that CACs are eventually included in all EMU sovereign bonds outstanding, large amounts of government debt securities across the world will continue to be without such clauses.

Regarding the second issue, little experience has been gained to date as to how to include the aggregation provision in a master agreement governed by a single law and subject to a single jurisdiction. Furthermore, it is fair to say that aggregating instruments governed by different laws or subject to different jurisdictions remains a key difficulty. Here, again, it is hard to imagine that any effective solutions can be found without intensified cooperation at the official level. As the IMF has noted (2003d and 2003e), in the event of a dispute regarding the application or interpretation of such clauses, there would be a risk that holders of different bond issues could find themselves in different courts, which in turn could provide different interpretations of the provision in question. One way to defuse this risk would be to include standardized aggregation clauses in the governing law of all jurisdictions where sovereign debtors can issue their securities. This is likely to be the solution chosen by European countries. But such a strategy may have an impact on the geography of international financial markets if similar aggregation clauses are not embedded in the governing law of other jurisdictions.

The statutory SDRM would have provided an expedient solution to both these key drawbacks of a CAC-based strategy (see Section III). However, the political concerns that led to its demise in 2002 remain and will continue to play a role in the foreseeable future.

It is important to underscore that the prevalence of CACs over the SDRM has implicitly pointed to the role of the Fund’s LIA policy. CACs do not usually address the issue of providing access to interim financing (also known as debtor-in-possession financing in a corporate context) to a sovereign undertaking a debt restructuring.23 Interim finance can be a crucial determinant of a successful restructuring; this issue was intended to be addressed by the SDRM and it currently remains within the scope of the Fund’s LIA policy.

Most fundamentally, it has long been recognized that CACs help address only one of the market failures that complicate a sovereign debt restructuring: the risk of holdout litigation. These clauses are therefore widely recognized as tool to facilitate a debt restructuring (by making exchange offers work better), but not one that can radically transform the restructuring process (Roubini and Setser, 2004, p. 348). More recently, it has been pointed out that the euro-area debt crisis does not involve a creditor coordination problem, and so the emphasis on the role of CACs is unclear (Gelpern and Gulati, 2011). In the case of Greece, a creditor coordination role was played by the Institute of International Finance (IIF, 2012). Finally, the very concept of an orderly debt restructuring, and its use as a preventive tool, is forcefully disputed, while recognizing that it may become inevitable as a last resort to solve a clearly unsustainable debt (Bini Smaghi, 2011b).

All in all, we believe that, despite their limitations, a more widespread diffusion of CACs is welcome, at the very least because virtually nobody doubts that they can be useful tools in a debt restructuring. Their recent activation by Greece points once again to their relevance. Moreover, given some discussions triggered by Greece’s retroactive inclusion of CACs (IIF, 2012), the European Council’s decision to include such clauses in European bonds takes on importance as a signal of a proactive rather than a reactive strategy.

Furthermore, and setting aside standardization, we do not underestimate what is perhaps the key novel element in the recent European revival of CACs, namely the required inclusion of an aggregation clause. It has been argued that the aggregate vote across different sovereign bond issuances resembles the process that was to have been put in place through the SDRM (Roubini and Setser, 2004, p. 349).

VI. Concluding Remarks

The global financial crisis that started in mid-2007 and its spillover into the sovereign sector of some euro area countries have had an unprecedented impact on the prevailing rules of the game for dealing with sovereign debt crises: the IMF’s lending framework and the use of CACs in sovereign bonds.

Following the string of crises in emerging market economies of the late 1990s, these pillars had been refined or developed to reflect the main objectives of the 2002 G-7 Action Plan: to counter market perceptions that official sector bailouts were the only available solution to a sovereign debt crisis; and to make debt restructurings more orderly and cost-effective, so that they could represent a credible Plan B. In essence, the outcome of those efforts was a framework for the management and resolution of sovereign debt crises in emerging market countries (the main clients of the IMF at the time) rather than in advanced economies, let alone those belonging to a currency union featuring very strong bank-sovereign linkages.

The recent crises originated at the very heart of the global financial system, provoking unparalleled risks of global financial meltdown and spillovers on a worldwide scale. In addition, their management was further complicated by the initial lack of effective intervention tools in the European camp. Such tools had to be developed hand in hand with the complex build-up of a more robust and credible system for regional macroeconomic surveillance. Thus, these crises posed distinct and unprecedented challenges, which required major adaptations to the existing management and resolution framework in order to tackle systemic risks effectively.

As far as Fund financing is concerned, the design of a comprehensive lending framework, with the right mix of constraints and discretion, is by its very nature a complex endeavor, one that requires laborious learning and adaptation processes. We have underscored the comprehensive overhaul of the Fund’s lending toolkit since March 2009; we have also emphasized how the Greek program of May 2010 was another landmark moment in terms of: cooperation between the Fund and European institutions; (large) size of euro-area countries’ financing needs and related burden sharing; program design (conditionality) and monitoring. We have also taken special care to illustrate the equally significant impact of the crises on the rules governing exceptional access to Fund resources, in order to determine their implications for the key objective of ensuring greater predictability set by the international community in the early 2000s.

On this latter point, we believe that the net effects of recent EAP changes appear somewhat difficult to gauge in a forward-looking perspective. On the one hand, the reforms adopted in March 2009 have certainly strengthened the rule-based nature of the EAP, by resolving the asymmetric treatment of capital versus non-capital account crises and by clarifying that both actual and potential BoP needs are covered under the policy. On the other hand, the introduction of a systemic clause to justify exceptional access to the Fund’s resources in case of a “high risk of international systemic spillovers” has broadened the scope of application of the EAP, but at the same time may have to some degree expanded the room for discretion.

In any case, we wish to emphasize that the introduction of a systemic clause is a consequential development. In our view, new lending rules to address explicitly the risks of systemic crises are a necessary (and desirable) response by the Fund to the evolution of the global financial landscape, where the greater size, interconnectedness, and sophistication of international capital markets amplify the potential for pure contagion and international spillovers (including those stemming from countries that do not necessarily meet the commonly-accepted definition of systemically-important economies on the basis of size).

We also believe that the growing relevance of cross-border interdependencies has magnified problems of collective action and time inconsistency deriving from the higher potential for systemic crises at the global and regional levels. Recent experiences have dramatically confirmed the fact that, like coordination failures among creditors, systemic factors do raise the costs (if not the likelihood) of sovereign debt crises, as they may more easily affect other countries via contagion. As a result, while it would be appropriate (and rational) to strictly enforce market discipline and institutional rules ex ante (i.e. before a crisis), and thus let countries default if their debts are truly unsustainable, this option may not appear preferable in practice ex post, once contagion and risks to systemic stability triggered by the above threats of default materialize.

In principle, one can argue that the combined effects of a higher potential for systemic crises (including those originated in non-large countries), on the one hand, and the introduction of a systemic clause, on the other, may increase the risk that, under new egregious failures in crisis prevention, the Fund might be faced with more cases of exceptional access to its resources than before. This scenario has a number of important consequences for the Fund, including in terms of: the analytical requirements for surveillance and lending; the adequacy of financial resources; and burden-sharing and cooperation with regional financial arrangements.

While each of these dimensions would require a study of its own, in what follows (and at the risk of oversimplifying) we sketch some of what we consider to be the main elements needing closer attention going forward.

Analytical Requirements for Surveillance and Crisis Management

The possibility of granting exceptional access to Fund resources under a systemic clause needs to be matched by improved analytical capabilities to identify and monitor systemic risks, so as to provide stronger underpinnings for such critical decisions. To begin with, the notion of systemic spillovers should be clarified and given greater operational content, also to assist program design and ensure evenhandedness (IMF, 2012b). This task also raises difficult questions, such as how to distinguish ex ante cases of genuine systemic threats from others; where the dividing line should be drawn; and how to better prevent unjustified large-scale Fund loans. On the other hand, waiting for systemic spillovers to materialize may result in sub-optimal crisis responses ex post, given that timely intervention is of the essence in these cases.

Adequacy of Official Financial Resources (“Firewalls”)

Perhaps one of the most salient questions posed by the recent debt crises concerns the availability and adequacy of financial resources effectively (i.e. credibly) deployable by the official sector (firewalls).

Given the obviously larger financing needs of systemically relevant economies, and in the light of the much stronger inter-linkages between the sovereign and banking sectors, there is now an imperative need to ensure a more permanent expansion of these firewalls, although the appropriate extent of this expansion remains to be clarified. As a bare minimum, however, it is necessary to devise more predictable means for quickly mobilizing additional resources in case of systemic emergencies.

Two among the many lessons that can be learned from the European experience offer hints as to how to address these issues. First, the official sector’s response to incipient crises was generally more reactive than proactive, partly reflecting region-specific circumstances such as incomplete political union and the complexity of existing European decision-making processes (Bini Smaghi, 2011c). Second, policy responses that were untimely, piecemeal, and not appropriately sequenced helped to aggravate market reactions and self-fulfilling dynamics; they also allowed markets to game the official sector into providing more money ex post than seemed sufficient ex ante (Draghi, 2011).

The IMF continues to play a clearly pivotal role in crisis management and resolution, given its analytical expertise, financial resources, and legal tools to mobilize funds from other parties in case of systemic crises (for example by borrowing money from other official creditors via bilateral or multilateral agreements such as the New Arrangements to Borrow - NAB).

However, the financing needs of euro-area countries currently under a program have proved too large to be addressed by drawing only on the Fund’s resources, which had to be combined with even larger contributions from European partners.

Thus, while during the crisis of the 1990s a key question in many observers’ minds was whether the Fund would be willing to lend to emerging market economies, today’s question is more like whether the IMF will actually be able to assist countries that have larger financing needs than in the past and pose potential systemic risks, while at the same time containing the related credit and liquidity risks for the institution. It is also important to emphasize that an adequate lending capacity for the IMF is vital not only for crisis management purposes, but also in a crisis prevention perspective, as it can boost the very credibility of Fund-led adjustment packages and precautionary arrangements.

Regular increases in the Fund’s quotas represent the natural tool to deal with cases of exceptional access; over the years, however, IMF shareholders have typically resisted increases in its permanent resources,24 and the Fund’s size has tended to shrink relative to the world economy or the size of international capital flows. This trend has been only partly countered with the 2009 decision to triple Fund resources (via bilateral loans to be subsequently folded in the NAB) and the 2010 decision to double IMF quotas with a rollback of the NAB.

Recent progress in boosting official firewalls has consisted in temporary bilateral contributions to the Fund and structural increases in the resources of regional financing arrangements (RFAs). As regards the Fund, a new wave of bilateral borrowing agreements for almost $460 billion was agreed in June 2012, and some of them were signed in October 2012. In Asia, the size of the Chiang Mai Initiative Multilateralization (CMIM) was doubled to $240 billion in May 2012. On Europe’s side, the European Stability Mechanism (ESM) has become operational since October 2012, providing a sizeable financial backstop of €700 billion for euro-area countries. Equally importantly, in October 2012 the ECB announced its decision on new Outright Monetary Transactions. Many of these developments were unthinkable until the outbreak of the global crisis in 2007.

These developments suggest that regional financing mechanisms and bilateral contributions will continue to play an important role in the future as well, as instruments for complementing IMF quotas. Henceforward, it will be crucial to ensure that these mechanisms can be deployed to support IMF lending arrangements in an effective and consistent manner.

Crisis Resolution and Collective Action Clauses

While recognizing that their role is limited to containing a specific risk of the restructuring process, namely the creditor holdout problem, CACs still represent a potentially cost-effective device for facilitating the restructuring of unsustainable sovereign debts, one that would not pose the politically sensitive constraints on countries’ sovereignty attached to the alternative statutory approaches. Thus, they will continue to represent a useful (and perhaps in some cases necessary) crisis resolution tool for the foreseeable future.

The further dissemination of CACs in sovereign bonds will receive a significant impulse from the recent decision to include standardized and identical clauses in all new euro-area sovereign bonds with maturities above one year.

On the other hand, in the light of the unprecedented systemic risks experienced during the recent crises, it is fair to argue that no predefined framework for debtor-creditor relations (CACs or even the SDRM) would be sufficient alone to address these risks satisfactorily and to mitigate the economic costs associated with the possible default of a systemically important country.

Uncertainty as to whether CACs would make a material difference in crisis resolution is compounded at this juncture by the transition and aggregation problems discussed earlier. Moreover, while the recent experience of Greece shows that, in specific circumstances, CACs can be imposed retroactively in the context of a bond exchange, issues of transition and aggregation would surely be addressed more smoothly if CACs were included in a proactive manner, without waiting for a restructuring to become necessary.

Other important questions are whether standardized CACs will be used as widely as possible across the existing variety of sovereign debt securities, and whether similarly standardized aggregation clauses will be legally acknowledged in the relevant jurisdictions. Certainly, there seems to be ample scope for the official sector to encourage the diffusion of standardized CACs and to adjust national legal settings accordingly.

More fundamentally, on crisis resolution, we believe that the most novel element evidenced by the euro-area debt crisis is the heightened risk of systemic effects attached to a debt restructuring in a highly integrated region with particularly strong inter-linkages between the sovereign and financial sectors. In addition to such inter-linkages (which may also have grown in emerging market countries), another major factor to consider when assessing the viability of an orderly debt restructuring is the potential implications of today’s massive use of credit default swaps (which were much less available during the crises of the late 1990s). Indeed, the possibility that all the above elements could have a bearing on the effectiveness of CACs had already been clearly identified before the recent crises.25

The benefits expected from a more widespread availability of CACs should be reassessed against this higher potential for systemic risks. The latter may reinforce the well-known limits of CACs to work as a credible Plan B to escape massive financial bailouts by the official sector (Leipold, 2011; Skeel, 2012).

In an environment with a higher potential for contagion and spillover effects, additional preconditions appear necessary for CACs to be activated in a safe and serviceable mode. Most notable among such preconditions are stronger capital and liquidity frameworks for financial institutions and the availability of adequate regional and global firewalls to contain contagion risks stemming from stronger bank-sovereign linkages at both the national and cross-border level. Furthermore, a more thorough assessment needs to be made of the possible consequences for a debt restructuring of today’s widespread use of credit derivatives and their legal implications.

The above arguments on the limited role to be played by CACs alone in a major debt crisis buttress some of the conclusions that had already been drawn following the crises of the 1990s, namely that better processes for restructuring international sovereign bonds would not reduce the need for official lending as a key supporting component. Again, the novel element now is that the IMF is more likely than before to have to share its role of crisis manager with regional institutions and financing arrangements.

Finally, it is important to underscore that, even when a debt restructuring is warranted on the basis of rigorous debt sustainability analysis, macroeconomic adjustment (notably fiscal consolidation, often significant and frontloaded) will remain a necessary component of a debt resolution strategy supported by official financing.

Crisis Prevention is Key

The previous considerations lead us to our principal conclusion, namely that a strengthening of crisis prevention capabilities in all their aspects (analytical, institutional, and financial) should henceforth be given distinct priority. This prominence stems, first and foremost, from the potentially very large economic costs of a disorderly restructuring (or even default) of a systemic country, as well as from uncertainty concerning the effectiveness (and at times the very availability) of the existing tools for crisis management and (above all) crisis resolution. Progress in the area of sovereign crisis resolution is currently linked to the prospective adoption of CACs by euro-area countries, as there appears to be virtually no room for developing alternative statutory mechanisms. More importantly, while underscoring the relevance of their recent sizeable increases, the financial adequacy of both multilateral and regional firewalls might come into question for countries whose size or interconnectedness makes them to be perceived as “too big to save”. As they face market pressures and higher refinancing costs for their sovereign debt, these countries may argue that the best available (and desirable) strategy is through domestic adjustment without official financial support.

The key question, therefore, boils down to how to improve sovereign crisis prevention in a highly interconnected world facing a higher potential for systemic risks. Sound public finances and fiscal discipline are key to this end. Fundamental progress would arguably be made in this direction if threats of sovereign default were made more credible, so as to foster effective (i.e. timely and gradual) market discipline ex ante. Pre-crisis market failure to discriminate among euro-area sovereigns by applying different spreads is evidence that the no-bailout clause of European treaties was not credible in the private sector’s eyes.

Effective market discipline in the field of sovereign finances is ultimately rooted in two basic prerequisites: consistent and timely assessment of sovereign risks by the private sector, and fiscal and banking transparency. While recent initiatives to improve transparency are most welcome (IMF-FSB, 2009; IMF, 2011, pp. 73-78; CGFS, 2011), they may not be sufficient alone to enforce more discipline, as markets should also be provided with appropriate incentives to use available information.

In terms of policy responses, reinforcing the credibility of sovereign default threats requires an effort to contain their economic cost and ensure that it is borne mainly by the debtor country and its private creditors. This calls for effective ex ante debt restructuring procedures (such as CACs) and a well-designed crisis management framework (including firewalls) to reduce two key drivers of market volatility: uncertainty and unpredictability. The above elements would all convey to markets the message that defaults and restructurings are in fact possible (Visco, 2010 and 2011). As previously noted, these elements of crisis resolution can be instrumental in fostering crisis prevention.

One of the key lessons to be drawn from the euro-area debt crisis is that the two-way linkages between the sovereign and banking sectors can have consequences for the effectiveness of any framework for crisis prevention, management, and resolution. The very possibility that a sovereign needs to intervene and financially support domestic banks creates the potential for sizeable public contingent liabilities. In turn, as domestic banks are traditionally important holders of sovereign bonds, sovereign risks affect negatively the cost and availability of bank funding, while a sovereign debt restructuring or default may impact the very stability of the banking sector. In both cases, sovereign default or perhaps even disorderly restructuring could easily generate significant adverse effects, which would lower the credibility of these events as alternative strategies to official bailouts.

It is thus evident that any substantive progress on this front would also require the sovereign-bank linkages to be severed, so as to limit risks of cross-sector and cross-border contagion. This is a challenging task, given that there are limits to what can realistically be achieved in practice. It will be virtually impossible to protect banks fully from a severely distressed sovereign, and other measures are needed to minimize the spilling over of banks’ difficulties back into sovereign risks. Sound supervisory and macroprudential policies are key to this end, while appropriate bank resolution regimes contribute to reduce investors’ expectations of government support for distressed banks (CGFS, 2011).

All in all, ensuring adequate capital (and liquidity) buffers for the financial systems is an integral part of the effort to make sovereign defaults more credible and, ultimately, to boost crisis prevention.

Collective Solutions to Global Challenges

Efforts to improve the IMF’s surveillance and lending frameworks need to be framed within the broader reform agenda that the international community has been pursuing since the start of the global financial crisis, under the aegis of the G-20 and with the involvement of other international bodies and organizations. The systemic nature of recent sovereign debt crises requires specific collective solutions, as systemic threats do not lend themselves to being tackled effectively with the instruments traditionally employed to address debt difficulties confined to individual countries.

These solutions pertain to all stages of sovereign debt difficulties (crisis prevention, crisis management, and crisis resolution). They require active contributions from a wide array of actors at various levels (national, regional, and global). They need to be part of a broader framework for a collective governance of systemic risks of various nature, but most notably fiscal and financial. Many notable progresses have been achieved in the aftermath of the recent crises.

While a “this time is different” attitude led to excessive complacency in assessing many pre-crisis dynamics that ultimately turned out to be contributors to the crises, nobody doubts that a “this crisis is different” argument is now fully warranted in the face of the unprecedented systemic threats posed by the events of the last few years: a meltdowns of the U.S. financial markets and the collapse of the euro area. We hope that a sort of Great Scare will act as a most powerful incentive for a further decisive strengthening of crisis prevention capabilities at all levels.

You Never Give Me Your Money? Sovereign Debt Crises, Collective Action Problems, and IMF Lending
Author: Mr. Marco Committeri and Francesco Spadafora
  • View in gallery

    IMF Lending Arrangements (*), Credit Outstanding, and Undrawn Balances Under GRA

    (SDR millions)