What happens when a country defaults — the economics and politics of sovereign insolvency

December 23, 2001. Adolfo Rodríguez Saá had been president of Argentina for four days when he stood before a joint session of Congress and announced the largest sovereign default in history — $95 billion in foreign debt, suspended immediately. The legislators gave him a standing ovation. They began chanting “Argentina, Argentina, Argentina.”

Not what you’d expect from a room watching economic extinction.

Argentina was about to enter one of the worst collapses of the modern era: currency devaluation, a frozen banking system, mass impoverishment. Within months a quarter of the workforce would be unemployed and more than half the population would fall below the poverty line. The default announcement was supposed to mark the catastrophe. So why were the elected representatives on their feet?

Because the alternative — three more years of austerity that was visibly not working, debt service on obligations that every economist in that chamber had known for months were unpayable, a political class maintaining a fiction at enormous human cost — was worse. The announcement wasn’t the disaster. The disaster had been ongoing. What Rodríguez Saá was ending was the pretense.

Hold that image.

What a default actually is

A sovereign default is a failure to make scheduled payments on government debt when they fall due. That sentence covers an enormous range — from a 30-day payment miss caused by administrative error to complete repudiation: a government declaring it will pay nothing to no one, ever. The same word applies to both, which is part of why the subject generates more heat than light.

In between are haircuts, where creditors accept less than face value; maturity extensions, where repayment moves further out; interest rate reductions; and combinations of all three. These are restructurings. They can precede a formal default — the government acts voluntarily before missing payments — or follow one, as part of the resolution process. They can achieve near-universal creditor participation or become contested, with holdout creditors litigating for years in whatever jurisdiction they choose.

Three distinctions govern how any specific crisis plays out.

Does the defaulting country have its own currency, or does it belong to a currency union? A country with monetary sovereignty can devalue. Imports cost more immediately, but export competitiveness eventually recovers and the real value of domestic-currency debt declines. A country in a currency union cannot devalue; all adjustment must come through wages, prices, and employment. This structural asymmetry — more than any inventory of fiscal sins — explains why Iceland recovered in years while Greece took a decade and a half.

Is the debt issued under domestic law or foreign law? Foreign-law bonds — issued under New York or English law — can be litigated in those jurisdictions by any creditor who refuses to restructure, including a hedge fund that bought the bonds for 27 cents on the dollar two years after the original default. Not a technical footnote — the legal infrastructure that makes holdout strategies viable and enormously profitable, as Argentina would discover over fifteen years of litigation in US courts.

How heavily are domestic banks loaded with sovereign debt? If they hold large volumes of government bonds, a sovereign default triggers a banking crisis simultaneously: the banks become insolvent, credit contracts, output falls. This feedback mechanism — the sovereign-bank doom loop — is the primary channel through which defaults transmit their most severe secondary damage to the real economy.

“Default” in newspaper headlines almost always implies the worst version: complete repudiation, permanent exclusion from capital markets, civilizational catastrophe. But the word describes a continuum. Where a crisis lands on it depends less on whether a country defaults than on when, under what monetary and institutional conditions, and with what framework — if any — governing the process.

Three crises, three fates

Start with Argentina, since that’s where we began.

The $95 billion default of December 2001 came at the end of three years of IMF-supported austerity that had failed to restore growth or stabilize debt dynamics. GDP had already contracted 15.7% from the recession’s start; the cumulative 1998–2002 decline was 19.9%. The peso had been fixed one-to-one against the dollar since 1991 — a commitment maintained through increasingly punishing fiscal contraction that was slowly destroying export competitiveness and draining reserves. By December 2001 the debt was unpayable and everyone involved knew it. The peg lasted another three weeks before it was abandoned; the peso fell to more than three to the dollar within months of the January 2002 devaluation. Unemployment hit 25%. Poverty exceeded 50%.

Capital markets closed. Argentina became the most notorious sovereign debtor on the planet.

And yet it continued to exist. Two debt exchanges — in 2005 and 2010 — produced settlements covering roughly 93% of eligible creditors. Market access returned. The story was messy: the restructuring was contested, litigated, and eventually produced a 2014 technical default when a US court blocked Argentina from paying restructured bondholders without simultaneously paying holdouts who had refused the deal. In 2020, Argentina defaulted again — its ninth — restructuring $65 billion in foreign bonds and obtaining $37.7 billion in debt relief over a decade. The pattern of serial mismanagement deserves no romanticization. But the permanent institutional collapse implied by the 2001 coverage never arrived. The ceiling on catastrophe proved lower than the catastrophism account suggested.

Iceland is the control case — and it’s here for a specific reason: what happened in Iceland was not a sovereign default. That distinction is the entire point.

In October 2008, Iceland’s three largest banks — Kaupthing, Landsbanki, Glitnir — collapsed in a single week. Their combined assets were approximately ten times Iceland’s GDP, the largest banking collapse relative to economy size in recorded history. What the Icelandic government did was the opposite of what every other developed country was doing that autumn: it let them fail. Each bank was split into a “new” domestic institution, which took deposits and domestic mortgages, and an “old” foreign creditor entity that inherited the cross-border liabilities and went into receivership. Foreign depositors and bondholders absorbed the losses. Iceland’s own sovereign obligations were untouched; the government had not guaranteed the banks’ foreign liabilities.

Compare this to Ireland, which on September 30, 2008 issued a blanket guarantee covering all liabilities of its six major banks — a commitment that eventually cost Irish taxpayers approximately €64 billion in gross injections (net cost approximately €46 billion, per the Comptroller and Auditor General’s 2022 report), converting a banking crisis into a sovereign debt crisis that would define Irish fiscal policy for the following decade. One country let its banks fail and kept its sovereign solvent. The other absorbed private banking losses into public debt and spent years paying for the decision.

Iceland accepted a $2.1 billion IMF standby arrangement in November 2008 — the first Western European IMF program since the UK in 1976 — and accepted a managed currency devaluation. GDP contracted roughly 10% peak to trough; unemployment peaked around 9%. Iceland had returned to growth by 2011 while Ireland remained mired in post-guarantee contraction. Iceland reached its pre-crisis per-capita peak by 2016.

The lesson is not that Iceland managed brilliantly. The lesson is structural: a government that refuses to absorb private banking losses into public debt is a government whose sovereign solvency stays intact. The severity of the crisis depends not just on what goes wrong in the financial sector but on who is made to bear the losses — and whether those losses are forced onto the population through public debt or absorbed by the private investors who took the original risk.

Then Greece — the hardest case, precisely because it never formally defaulted.

From 2010, Greece accepted troika conditions — European Commission, ECB, IMF — that made avoidance of formal default the explicit organizing principle of a decade of policy. GDP fell 26.3% peak to trough in per-capita terms, a decline comparable in scale to the United States during the Great Depression but shaped entirely differently. The US depression was V-shaped: a brutal fall, then recovery. Greece’s was U-shaped: six years to reach bottom, fifteen years to return to pre-crisis per-capita output. The comparison is not to the depth of the American experience — it’s to its duration and its character as a generational catastrophe.

A private-sector debt exchange did eventually happen in 2012, imposing a 53.5% haircut across roughly €200 billion in bonds — the largest sovereign debt restructuring in history. But it arrived three years into a program that the IMF’s own independent evaluators would later conclude had been built around implausible growth projections and rules that management had changed specifically to allow it to proceed. The restructuring that should have happened in 2010 happened in 2012, after two years of depression-grade contraction, and was presented as a success.

The comparison the three cases establish: formal default (Argentina) produced genuine catastrophe that was survivable. Non-default engineered to protect creditors (Greece) produced depression-scale damage lasting fifteen years. Not-a-default (Iceland) produced a sharp but short contraction followed by full recovery. The variation cannot be explained by which country was more reckless or more deserving. It can be explained by monetary sovereignty, institutional structure, and when the inevitable was permitted to happen.

The 2012 Greek PSI

The Private Sector Involvement of 2012 was the largest sovereign debt restructuring in history by nominal value — approximately €107 billion in debt reduction, with a 53.5% nominal haircut on privately held bonds. Greece retrofitted collective action clauses into its domestic-law bonds retroactively, achieving a participation rate of approximately 97% on the domestic-law tranche. But the exchange subordinated private creditors to official-sector debt — ECB holdings, bilateral EU loans, IMF facilities — which remained entirely untouched. By 2012, European banks had used the two preceding years to reduce their Greek sovereign debt exposure substantially. The private investors who took the haircut in 2012 were disproportionately not the banks whose balance sheet exposure had been the political obstacle to restructuring in 2010.

The three channels

Iceland recovered in roughly four years. Greece took fifteen. Argentina’s catastrophe was severe but survivable. Understanding why requires looking past the label — “default,” “bailout,” “program” — to the mechanisms through which sovereign distress transmits damage to the real economy. There are three. In each of the three cases above, they fired differently, and the differences explain the outcomes.

Start with the banking system. Most countries’ regulations designate domestic sovereign debt as a zero-risk asset — banks hold no capital against it, because the government is assumed to always pay. Banks respond rationally: they load up on government bonds, which yield more than cash and require no capital buffer. When a government defaults, banks holding large sovereign debt portfolios become insolvent simultaneously. Credit contracts; businesses can’t borrow to make payroll; households can’t finance spending; output falls. This is the doom loop. In Iceland, the banks held private-sector assets, not government bonds; the loop didn’t engage because the banking failure wasn’t triggered by sovereign insolvency. In Argentina, domestic banks held peso-denominated government bonds and froze when the default hit in December 2001. In Greece, the doom loop operated in slow motion before the 2012 exchange — Greek banks’ balance sheets were visibly impaired by mid-2011, and the bank run (gradual, then sudden) preceded rather than followed formal restructuring.

Then the exchange rate. Sovereign debt crises almost always trigger capital flight: investors pull money out before or during the crisis, reducing demand for the local currency, driving it down. Imports cost more; foreign-currency obligations become harder to service in local terms. Argentina’s January 2002 devaluation — the peso fell to less than a third of its dollar parity within weeks — made dollar-denominated private debt instantly unsustainable, producing a wave of corporate defaults alongside the sovereign one. For countries with monetary sovereignty, devaluation is painful but eventually self-correcting: exports become competitive, domestic currency debt shrinks in real terms. For Greece in the eurozone, no devaluation was possible. The entire adjustment burden fell on wages, pensions, and employment. This asymmetry is why Greece’s adjustment took so much longer and cost so much more than Iceland’s.

And then market access closes.

Most governments run structural deficits — they spend more than they raise in taxes and fund the difference by issuing new debt. When a country defaults, markets close. If it was running a primary deficit — spending more than it raised even before interest payments — it must achieve immediate primary balance: cutting spending or raising taxes sharply in a recession, exactly when the economy needs the opposite. This pro-cyclical tightening deepens the recession, shrinks the tax base, and makes the underlying fiscal adjustment harder over time. The speed of the effect depends on debt maturity structure — a country with long-dated bonds can hold on longer — but the direction is invariant.

When all three channels fire simultaneously — banks collapse, currency falls, new borrowing stops — the severity compounds. When a country retains monetary sovereignty and limits banking sector exposure, as Iceland did, the channels interact differently, and the damage is bounded. When a country surrenders monetary sovereignty and allows the banking sector to load up on sovereign debt, as Greece did within the eurozone framework, the channels become mutually reinforcing and the damage becomes very hard to control.

None of this is mysterious or unforeseeable. The mechanisms are understood by every finance ministry, every multilateral institution, every sophisticated creditor before the crisis begins. If it were ignorance producing these outcomes, the pattern would be random. It isn’t.

The doom loop by design

European banking regulation still classifies eurozone sovereign bonds as zero-risk assets under Basel rules; banks hold no capital against them. This was identified as a structural problem during the 2012 eurozone crisis — banks in fiscally stressed countries held disproportionately high sovereign debt precisely because the zero-capital treatment incentivized concentration. Reform proposals have been raised periodically; the Basel Committee's December 2017 review concluded without consensus — the Committee published a discussion paper but decided not to proceed to formal consultation — with resistance concentrated among jurisdictions carrying the largest bank-sovereign exposures. The consequence is direct: under current rules, any eurozone sovereign debt crisis is simultaneously a banking crisis by regulatory design. This is not an oversight that has persisted from inertia. It is a policy choice that every major actor has repeatedly chosen to preserve.

The political economy of delay

If the damage from sovereign debt crises is mechanically predictable — and it is — then the relevant question is why the actors involved consistently allow it to compound.

The political cost of admitting sovereign insolvency is enormous and immediate. The cost of delay is diffuse and slow. A government that announces it cannot pay its debts faces instant execution: it has conceded that its predecessors lied about the trajectory, that the debt is unsustainable, that creditors were misled. The ministers who must make this admission are the ones whose credibility depended on preventing it. They have every incentive to delay, and they delay.

Even when a restructuring finally happens, it rarely resolves cleanly: 86% of sovereigns that restructured their debt between 1980 and 2012 did so more than once. In the majority of cases, the first restructuring was insufficient and had to be repeated. This is not a series of unfortunate exceptions. It is the pattern.

The mechanism is called “extend and pretend.” Creditors keep rolling over loans to avoid recognizing losses on their balance sheets. Governments keep borrowing to postpone the political cost of admission. Multilateral institutions extend emergency programs to prevent immediate disorderly collapse — knowing the numbers don’t add up, because the alternative, an immediate disorderly default, is politically unacceptable in creditor countries. The hole grows. The eventual reckoning becomes more expensive. And the costs accumulate not on the balance sheets of the institutions choosing to extend, but on the population that had no say in the original borrowing.

Greece is the textbook case. The 2010 troika program was explicitly designed to avoid restructuring — not because restructuring was economically unnecessary (the IMF’s own staff had determined Greek debt was not sustainable with high probability before the program was designed) but because European banks held large amounts of Greek sovereign debt and a haircut in 2010 would have required recapitalizing those banks at European government expense. The IMF’s independent evaluation office, in its 2016 report on the crisis, found that management had amended the institution’s exceptional-access framework to allow the program to proceed without the restructuring that staff analysis said was needed. The evaluation describes the decision as “perceived as favouring the eurozone” — bureaucratic language that, read plainly, means the IMF changed its own rules to accommodate the preferences of the eurozone’s creditor governments.

By 2012, European banks had used the two intervening years to reduce their Greek sovereign debt exposure. They did reduce it. The losses that fell in 2012 landed on private bondholders and Greek domestic institutions. Greek unemployment exceeded 25% at the bottom. Per-capita GDP did not return to its 2008 level for fifteen years. The emigration of working-age Greeks to Germany, the UK, and elsewhere represented a permanent demographic transfer.

The actors who chose to delay were not the actors who bore the cost of delay. This asymmetry runs through every case in this article, and it is almost never named directly.

Creditors and the architecture of bad incentives

The asymmetry that governs creditor behavior in these crises is not complicated: the upside of lending to a high-yield sovereign accrues privately; a significant portion of the downside of disorderly default is socialized. When a major debtor country approaches crisis, multilateral institutions and creditor-country governments intervene — not out of generosity but because contagion is politically unacceptable at home. The private creditor who lent at high yield and now faces losses can expect some combination of official-sector lending that services existing debt (keeping the creditor whole through the crisis period) and structured bailout arrangements that prioritize private creditor recovery. The incentive is not subtle: lend, collect the high yield, expect intervention if things go wrong.

The holdout problem makes this worse. NML Capital, a subsidiary of Paul Singer’s Elliott Management, bought Argentine bonds at approximately 27 cents on the dollar after the 2001 default. So did Aurelius Capital Management. Both refused Argentina’s 2005 and 2010 debt exchanges — deals that 93% of eligible creditors accepted — and litigated instead in US federal court under the pari passu clause, arguing that Argentina could not pay restructured bondholders without simultaneously paying holdouts in the same proportion.

Judge Thomas Griesa of the Southern District of New York agreed, in February 2012. The Second Circuit upheld the ruling in October 2012. The consequence: Argentina could not service its restructured debt without simultaneously paying holdouts who had accepted no terms. It was pushed into a fresh technical default in 2014, despite having the money to pay exchange bondholders, because New York courts made doing so without also paying holdouts legally impermissible.

Elliott Management extracted approximately $2.4 billion from the eventual settlement — roughly 10 to 15 times what NML Capital paid for the distressed paper.

This is not anomalous behavior. It is a business model the framework enables, and rational actors will pursue it as long as the framework permits. A country whose bonds are issued under New York law, in a jurisdiction whose courts will uphold a plain-text reading of “equal treatment” clauses, cannot restructure its debt in any orderly way — because any deal that 93% of creditors accept can be blocked and litigated for a decade by the 7% who chose not to.

The collective-action clause (CAC) was the contractual response: after 2003, CACs became standard in sovereign bond contracts, allowing a qualified majority — typically 75% — to bind the holdout minority. But CACs apply within a single bond series and can be circumvented by accumulating positions across multiple series before a restructuring is proposed. They do nothing for the extend-and-pretend dynamic on the government side, the doom loop on the banking side, or the exchange-rate channel. A contractual patch on a structural problem doesn’t solve the structure.

The absent framework — and why it’s absent

Corporate insolvency has structured resolution frameworks because the alternative — creditors racing to seize assets, value destroyed in the scramble, contagion spreading through the financial system — was eventually judged intolerable. The US Bankruptcy Code’s Chapter 11 provides an automatic stay the moment a filing is made: all creditor action freezes while reorganization is negotiated, preventing the race to grab assets that makes disorderly failure so destructive. Priority rules establish the order in which different classes of creditors are paid. Cram-down provisions allow a reorganization plan to be imposed on holdout creditors if a qualified majority approves. The machinery works; corporate insolvency, even at scale, is manageable as a result.

For sovereigns: nothing.

The Sovereign Debt Restructuring Mechanism — the SDRM — was a serious attempt to build something equivalent. Anne Krueger, then IMF First Deputy Managing Director, proposed it publicly in November 2001, immediately after Argentina made the absurdity of the vacuum unmistakable. The SDRM would have created a quasi-judicial process within the IMF framework: an automatic stay on creditor litigation while negotiations proceeded, a supermajority voting mechanism to bind holdouts across all bond series, and structured negotiations with defined timelines and creditor hierarchies.

Dead by 2003.

The US government under George W. Bush opposed it, preferring contractual solutions — specifically, collective action clauses — that left discretion with market actors and kept US courts in their existing role as adjudicators of sovereign debt disputes. Large emerging-market debtor countries opposed it: formalizing IMF adjudication would entrench IMF political power over their economic policies in ways they had no interest in accepting. Private creditors opposed it because it would have eliminated holdout strategies. And the IMF itself, documented in the contemporary accounts, had limited enthusiasm for running the process. Krueger’s proposal died at the hands of nearly every constituency it would have needed to succeed. The institution that proposed the SDRM did not particularly want to run it.

What replaced it: CACs in bond contracts; the IMF’s amended exceptional-access policies, later bent again for Greece; and, since 2020, the G20 Common Framework for Debt Treatment, which covers only the world’s poorest countries. Zambia and Ghana — defaulted in 2020 and December 2022 respectively — took years to complete their restructurings under its provisions; Sri Lanka, which had its first-ever default in April 2022, was not eligible for the framework at all and navigated its restructuring outside any formal mechanism, completing it in 2025.

The Common Framework’s sluggishness is structural, not accidental. China — a major bilateral creditor in both Zambia and Ghana — prefers to negotiate separately, outside creditor committees, on undisclosed bilateral terms. Coordinating Paris Club creditors, China, and private bondholders with divergent incentives, across different legal jurisdictions, with no binding timeline and no cram-down mechanism, produces years of delay by design. A predictable outcome of a system shaped by actors who benefit from bilateral discretion and have chosen to preserve it.

The blueprint for something better is not secret. The SDRM specified it: automatic stay, supermajority holdout binding, structured timelines, defined creditor hierarchies. The mechanics are understood. So is what each category of actor would lose under any framework that matched those mechanics. US courts lose jurisdiction over sovereign debt disputes. Hedge funds lose the holdout strategy. European creditor governments lose the discretion to delay restructuring until their banks have reduced their exposure. The IMF loses the flexibility to amend its own access rules to accommodate major shareholders.

These are advantages being actively extracted right now from the absence of rules.

The G20 Common Framework

Launched in November 2020, the Common Framework applies only to low-income countries eligible for IMF-World Bank debt sustainability assessments. Zambia defaulted in 2020 and completed its Eurobond exchange by June 2024, with bilateral creditor agreements finalized through late 2024. Ghana defaulted in December 2022 and completed its Eurobond restructuring in October 2024. In each case, years passed between default and completed resolution while austerity conditions constrained investment in recovery. China — a major bilateral creditor in both cases — insists on negotiating outside the Paris Club on undisclosed bilateral terms, creating coordination problems that extend every timeline. Sri Lanka, which had its first-ever default in April 2022, was not eligible for the Common Framework — it is a lower-middle-income country — and completed its restructuring through a parallel multi-creditor process in 2025. The Common Framework has no equivalent for middle-income countries, which is exactly what Argentina and Greece were.

Back to that chamber in Buenos Aires in December 2001.

What the legislators understood, and what the coverage mostly missed, was that the catastrophe was not the announcement. The catastrophe had been the preceding three years of maintaining a fiction that Argentina could service debt everyone in that room knew was unpayable. The relief was at the end of pretending. And the pretending wasn’t irrational — it was the product of the exact incentive structure this article has spent several thousand words tracing. Ministers whose credibility depended on debt sustainability. IMF officials under pressure from creditor-country governments. Creditor banks with balance sheet exposure they needed time to reduce. All of them had reasons to extend the fiction. None of them bore the cost when it finally ended.

The mechanics are understood. The problem has never been technical.

What US courts lose under an automatic stay: jurisdiction. What European creditor governments lose under binding restructuring timelines: the discretion to delay until their banks have exited. What Elliott Management loses under enforceable cram-down: something close to $2.4 billion.

These are present extractions — advantages taken from the absence of rules right now, not hypothetical future costs.

The question the SDRM’s defeat answers is not whether a better system is architecturally possible — it demonstrably is. It’s whether the parties who would need to agree to build one have a reason to do so that outweighs their reason not to.

The chamber in Buenos Aires already knew how that calculation tends to come out.

Penafian Gen AI

Beberapa konten halaman ini dibuat dan/atau diedit dengan bantuan AI Generatif.

Media

Jorge Soto Farias – Pexels

Key Sources and References

Adolfo Rodríguez Saá, inaugural address to the Argentine Congress, December 23, 2001. Contemporaneous partial transcript at networkideas.org, archived December 29, 2001;

Miguel Kiguel, “Argentina’s 2001 Economic and Financial Crisis: Lessons for Europe,” Brookings Institution, November 2011.

Martin Guzman, “An Analysis of Argentina’s 2001 Default Resolution,” CIGI Paper No. 110, Centre for International Governance Innovation, October 2016.

International Monetary Fund Independent Evaluation Office, “The IMF and the Crises in Greece, Ireland, and Portugal,” IEO Report, 2016; IEO Background Paper BP/16/02/11.

Comptroller and Auditor General of Ireland, “Net Cost of Banking Stabilisation Measures,” September 2022.

Irish Times, “The bailout cost €64bn,” 2017. Gross injection figure cited.

IMF Press Release No. 08/296, “IMF Executive Board Approves US$2.1 Billion Stand-By Arrangement for Iceland,” November 19, 2008. https://www.imf.org/en/news/articles/2015/09/14/01/49/pr08296

IMF Country Report No. 08/362, “Iceland: Request for Stand-By Arrangement — Staff Report,” November 2008. https://www.imf.org/external/pubs/ft/scr/2008/cr08362.pdf

Desmond Lachman, “The Economic Divergence of Iceland vs Ireland,” American Enterprise Institute. https://www.aei.org/economics/international-economics/the-economic-divergence-of-iceland-vs-ireland/

MercoPress, “Paul Singer will have made a $2.4bn profit with the Argentine defaulted bonds,” March 4, 2016.

Udaibir Das, Michael Papaioannou, and Christoph Trebesch, “Sovereign Debt Restructurings 1950–2010: Literature Survey, Data, and Stylized Facts,” IMF Working Paper WP/12/203, August 2012. https://www.imf.org/en/Publications/WP/Issues/2016/12/31/Sovereign-Debt-Restructurings-1950-2010-Literature-Survey-Data-and-Stylized-Facts-26190

Leonardo Martinez, Francisco Roch, Francisco Roldán, and Jeromin Zettelmeyer, “Sovereign Debt,” IMF Working Paper WP/2022/122, June 2022. https://www.imf.org/en/publications/wp/issues/2022/06/17/sovereign-debt-519809

Pierre-Olivier Gourinchas, Thomas Philippon, and Dimitri Vayanos, “The Analytics of the Greek Crisis,” London School of Economics Hellenic Observatory, GreeSE Paper No. 100, July 2016.

Cyrille Lenoël, Corrado Macchiarelli, and Garry Young, “Greece 2010–18: What could we have done differently?” London School of Economics Hellenic Observatory, GreeSE Paper No. 172.

Anne Krueger, “A New Approach to Sovereign Debt Restructuring,” speech, IMF, November 26, 2001. https://www.imf.org/en/News/Articles/2015/09/28/04/53/sp112601

Nouriel Roubini and Brad Setser, “Improving the Sovereign Debt Restructuring Process: Problems in Restructuring, Proposed Solutions, and a Roadmap for Reform,” Peterson Institute for International Economics, March 2003. https://www.piie.com/commentary/speeches-papers/improving-sovereign-debt-restructuring-process-problems-restructuring

Basel Committee on Banking Supervision, press release, “Regulatory treatment of sovereign exposures,” December 7, 2017. https://www.bis.org/press/p171207a.htm