## _052214a ## Source details **Canonical URL:** [_052214a](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_052214a.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_052214a.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_052214a.pdf.json) --- ### Overview - Describes evolution of the Fund’s approach to lending in sovereign debt crises since the early 1980s, with emphasis on origins and evolution of the current framework for exceptional access. - Identifies three broad phases: - Emerging markets debt crises of the 1980s: policy adjustment with rescheduling and new money initially, moving to debt reduction when adjustment proved insufficient. - Capital account crises during 1995–2002: non-sovereign in nature; rescue programs relied on exceptional access to catalyze private flows; private flows sometimes failed to materialize. - Post-1990s reforms and the 2008–10 period: experience (including Argentina and Russia defaults) led to the 2002 Exceptional Access Policy and a more rigorous DSA framework; 2010 amendment created a “systemic exemption” to the “high probability” public debt sustainability requirement (invoked in programs for Greece, Portugal, and Ireland). ### The Emerging Markets Debt Crises of the 1980s — key findings and chronology - Origins and context: - Debt crisis of 1982–83 resulted from massive external shocks and policy errors that reduced developing countries’ capacity to service external debts. - Turnaround in borrowing conditions and U.S. policy changes led to Mexico’s moratorium in 1982 and a regional crisis. - Fund policy and tools: - Supplementary Financing Facility (SFF) introduced in 1979 to on-lend borrowed resources and permit access above normal limits “in special circumstances.” - “Exceptional circumstances” clause established in the 1983 review of the Enlarged Access Policy (EAP); the EAP replaced the SFF in 1981. - Strategy evolution: - Early phase through 1986: “concerted lending” required creditor bank commitments for rescheduling and new lending as preconditions for Fund financing, typically within normal access limits. - From early 1987: shift to market-based debt operations to reduce present value of debt, preceding the Brady Plan (early 1989). - Role in debt reduction: - Staff advocacy for debt relief by 1987; early operations included Bolivia’s buybacks, Chile’s debt-equity swaps, and Mexico’s exchange of bank debt for lower face value collateralized bonds. - Brady Plan (March 1989) plus Fund support delivered debt sustainability. ### Fund Support for Debt and Debt Service Reduction (DDSR) Operations — structure and outcomes - Policy evolution and financing mechanics: - 1989: Fund allowed members to use Fund resources, with other official support, to finance upfront costs of DDSR operations (buybacks, escrow collateral). - Financing through set-asides (25 percent of access under Fund arrangements) and augmentations (up to 30 percent of quota). - Set-asides and augmentations released only after commercial creditor agreement and Executive Board assurance of effective use; must leave adequate margins for general balance of payments needs. - Implementation examples: - Early 1990s beneficiaries included Venezuela, Costa Rica, Mexico, and the Philippines. - Policy terminated in 2000 as countries shifted to capital markets. - Poland case (1994 SBA): - Regularized USD 14.4 billion of commercial debt at an upfront cost of USD 1.9 billion (buybacks and purchase of collateral). - Operation judged consistent with guidelines: voluntary participation; market-based terms better than secondary market; significant authorities’ contribution; crucial for external viability. - Technical adjustments: - Segmentation between set-asides and augmentations modified in 1992 and eliminated in 1994 for flexibility. ### Capital account crises (1995–2002) — nature, Fund response, and lessons - Nature: - First wave began with Mexico “tequila crisis” in 1995; Asian crisis in 1997; driven by short-term funding dependence, currency and maturity mismatches, and weaknesses in financial/corporate sectors. - After 1998: Russia and Argentina sovereign defaults reflected fiscal imbalances and structural rigidities. - Fund financing strategy: - Rescue programs relied on exceptionally large Fund access to catalyze private returns; access heavily frontloaded. - Average length of arrangements was 2½ years with purchases of more than 70 percent of total access during the first year available in all cases. - “Exceptional circumstances” clause invoked 14 times between 1995 and 2002. - 1997: Supplemental Reserve Facility (SRF) created (not subject to access limits); first use for Korea in December 1997; later used by Argentina, Brazil, Russia, Turkey and Uruguay. - Private sector involvement: - Pure catalytic approach often insufficient as creditor base became more diffuse and bonded debt more prevalent. - Examples of limited creditor engagement: Argentina (1995), Mexico (1994), Turkey (1994), Philippines (1997). - Examples of creditor pressure/engagement: Thailand (1997); Brazil (1999); Korea and Indonesia (1998). - Concerns and reform impetus: - Large-scale Fund financing raised moral hazard and systemic risk concerns, especially where debt sustainability was doubtful; Russia and Argentina defaults amplified concerns. - Led to Prague Framework (2000) and later the 2002 Exceptional Access Policy (EAP), more rigorous DSA, advocacy for CACs, and consideration of an SDRM. ### Development and evolution of the Exceptional Access Policy (EAP) - 2002 EAP (operational Feb 2003) substantive criteria for exceptional access in capital account crises: 1) Member experiencing exceptional balance of payments pressures on the capital account requiring Fund financing beyond normal limits. 2) Rigorous and systematic analysis indicates a high probability that debt will remain sustainable. 3) Member has good prospects of regaining private market access within Fund resources outstanding. 4) Policy program provides a reasonably strong prospect of success, including institutional and political capacity. - Emphasis on a modified DSA for market access countries (May 2002). - Board view: access where debt restructuring was needed would normally be within normal limits; rare exceptions possible. - 2003–2004 debates: - SDRM proposed but lacked international support; focus shifted to CAC enhancement. - Argentina and Brazil granted exceptional access despite not meeting all criteria (Argentina: access of 424 percent of quota; Brazil sought precautionary support). - Early 2004: Board declined staff amendments to broaden exceptions; agreed procedures would apply “in light of the four substantive criteria” without conditioning approval on meeting all criteria in noncapital account cases. ### 2009 reform and post-2008 experience - Rationale: - Pre-2009 asymmetry between capital and noncapital account crises generated perceptions of inconsistency; framework precluded exceptional access where debt unsustainable at time of request even if sustainability could be restored. - 2009 revised criteria (all must be satisfied for GRA access beyond normal limits): 1) Member is experiencing or has the potential to experience exceptional balance of payments pressures on the current or capital account requiring Fund financing beyond normal limits. 2) A rigorous and systematic analysis indicates a high probability that the member’s public debt is sustainable in the medium term, evaluated on a forward-looking basis and may take into account intended restructuring to restore sustainability; applies only to public (domestic and external) debt but incorporates potential contingent liabilities. 3) Member has prospects of gaining or regaining access to private capital markets within the timeframe Fund resources are outstanding. 4) Policy program provides a reasonably strong prospect of success, including institutional and political capacity. - Complementary changes: - Doubled access limits: annual limit of 200 percent of quota and cumulative limit of 600 percent of quota. - Added new “precautionary” GRA facilities. - Eliminated the SRF. - Post-2008 demand and systemic exemption: - After 2008, commitment of more than US$250 billion in loans to members amid global financial crisis. - May 2010 EAP revision created an exemption to the “high probability” debt sustainability requirement where there was a “high risk of international systemic spillovers”; invoked 34 times by end-May, 2014 in the three EA programs for Greece, Portugal, and Ireland. ### Greece case highlights (timing and magnitudes) - By 2011 a debt restructuring was needed amid deep recession, massive financing needs, and an undercapitalized banking system. - Mid-2011 staff acknowledged PSI necessary to close financing gaps. - March/April 2012 debt exchange reduced more than 50 percent of eligible debt (about €200 billion) with participation close to 97 percent. - Late 2012 second PSI round (buyback) retired sovereign bonds issued in the debt exchange. - Private creditors paid out a total of almost €50 billion drawing on Fund and other official resources prior to 2012 restructuring. ### Annex II — retrospective review of sovereign debt restructurings (1980–2012) — core findings - Sample and scope: - 107 countries (34 advanced, 73 emerging) with public debt data for at least 10 consecutive years over 1980–2012. - Total of 229 debt restructuring episodes: 15 domestic, 99 private foreign, 115 official. - Key findings: - Many countries avoided restructuring, but a significant number required restructurings. - Lower initial debt and stronger fundamentals increased likelihood of exiting distress without restructuring. - Restructurings were often delayed and insufficient in size; “repeat” reprofilings should be avoided. - Two-thirds of restructurings with private foreign creditors in a representative sample resulted in a “non-cured” outcome requiring further restructuring. - Where initial debt was high, significant NPV reduction was typically needed to improve success probability. ### Definitions and empirical thresholds used - Large debt reduction (LDR) requires: - (i) initial debt > 60 percent of GDP; - (ii) debt fell for at least 5 years subsequently; - (iii) total reduction ≥ 10 percentage points of GDP. - Durable recovery in sovereign debt ratings requires: - (i) ratings reach a trough and recover to at least an average B- rating; - (ii) remain above the trough 5 years later; - (iii) allow negative blips of at most one notch during the 5 years following the trough. - Signal approach indicative benchmarks: 60 percent for emerging markets and 70 percent for advanced economies as predictive of debt distress events. ### Evidence on restructuring effectiveness — “too little too late” - Delays and repeat restructurings: - From 1980 to 2012, 44 countries restructured; “86 percent had more than one restructuring.” - On average each country had “over 5 restructurings,” about half with private foreign creditors. - Outcomes with private foreign creditors: - Of 99 restructurings with private foreign creditors, “only about a third met the criteria for a ‘cured’ outcome”; two-thirds did not establish sustainability. - Determinants of success (econometric findings, Box AII2): - Probit model using 96 cases. - Key drivers improving cure probability: larger face value reductions, larger NPV haircuts, presence of an IMF program, better policy/institutional framework, lower initial debt. - Interaction between initial debt and debt reduction significant and positive: larger debt relief is more effective when initial debt is high. - Adverse factors: unfavorable global conditions, natural disasters, high initial debt, exogenous shocks. ### Quantitative illustrations and case studies - LDR episodes (summary characteristics): - LDR with restructuring (32 cases) vs without restructuring (21 cases) — selected averages preserved exactly: - Public debt at t = 0: 100.4 (with restructuring) and 79.8 (without restructuring). - Public debt build-up (t = -3 to t = 0): 43.8 (with) and 7.1 (without). - External public debt at t = 0: 66.0 (with) and 47.2 (without). - External public debt build-up (t = -3 to t = 0): 24.1 (with) and 10.7 (without). - Primary balance average, t = -3 and t = 0: 0.8 (with) and 1.7 (without). - Current account deficit average, t = -3 and t = 0: 3.0 (with) and 2.0 (without). - GDP growth average, t = -3 and t = 0, in percent: 1.1 (with) and 1.8 (without). - Bond spreads average, t = -3 and t = 0, in basis points: 1108 (with) and 462 (without). - Depreciation of the REER average, t = -3 and t = 0, in percent: 4.6 (with) and 1.9 (without). - Case studies: - Brazil (2002 program): access of SDR 22.8 billion (US$30 billion, 752 percent of quota); regained market access in April 2003. - Turkey (1999–2001): SBA initial access SDR 2.9 billion, augmented to SDR 15 billion (about US$19 billion, 1560 percent of quota); maintained market access and organized voluntary domestic debt swap. ### Ex-post case evidence on reprofiling and restructuring - Reprofiling outcomes: - Reprofiling sufficient in some liquidity-driven or moderate cases: Dominican Republic, Pakistan, Uruguay. - Reprofiling failed where debt too high or banking fragility present: Grenada, Jamaica 2010. - Uruguay (2003 reprofiling) — selected exact figures: - Public debt rose from 40 percent of GDP in 2001 to about 100 percent in 2003. - Average maturity extension: international bonds extended by 6.4 years; domestic bonds and treasury bills extended by 8.6 years. - NPV reduction: about 13.3 percent for international bonds and 23.3 percent for domestic bonds. - Participation: average 89 percent for international bonds and 99 percent for domestic obligations. - Market access regained in June 2003 with a US$53 million bond with 7.7-year maturity and 10.6 percent yield to maturity; subsequent issuance in October 2003. ### Annex III — comparative costs/benefits of reprofiling, bailout, and restructuring (gray zone) - Market steady-state implications of a credible reprofiling regime: - Investors price higher probability of credit events; spreads may increase, especially for higher-risk countries. - Shorter-term bonds more likely to be reprofiled → demand declines for short tenors → maturity shifts to longer tenors → higher average cost, reduced rollover risk. - Countries with limited domestic markets may issue more international bonds → higher currency risk. - Sovereign borrowing cost increases likely transmit to corporate sector where pass-through strong. - Empirical sample-based main finding: - On a sample of 12 episodes (half reprofilings, half restructurings with face-value cuts), reprofilings on average appeared less costly than face-value cuts across spreads, rating downgrades, time to market re-access, negotiation length, participation rates, and litigation. - Comparative statistics (selected exact figures preserved): - Spreads at announcement (median): - Restructuring cases: 2,119 bps. - Reprofiling cases: 1,053 bps. - Change vs median prior 12 months: - Restructurings: +727 bps. - Reprofilings: +216 bps. - Time to re-access international capital markets (median, global bond issuance): - Restructurings: 57 months. - Reprofilings: 35 months. - Time to re-access by spread normalization (median): - Restructurings: 35 months. - Reprofilings: 8 months. - Credit rating recovery within 12 months after exchange (average notches): - Reprofilings: 3 notches. - Restructurings: 2 notches. - Caveats: - Small sample size; low statistical power for reprofiling cases. - Difficulty disentangling debt-exchange effects from subsequent policies and international environment. - Reprofiling not viable where restructuring is necessary to restore sustainability. ### Sovereign contagion, measurement, and findings - Vulnerability Index (VI) framework: - VI measures probability of distress of a country conditional on other countries’ distress; ranges zero to one; estimated using EMBI for EMs and sovereign CDS for AEs. - Contribution to contagion computed as percentage contribution of each crisis country to related countries’ VI. - CIMDO methodology used to estimate joint and conditional probabilities; non-parametric, minimum cross-entropy approach. - Evolution and findings: - Contagion rose sharply in run-up to crises and subsided once credible policy responses (reprofiling or restructuring) were undertaken. - Contagion was empirically larger in restructuring cases (e.g., Russia, Argentina, Greece) than in reprofiling cases in the sample. - International environment (global risk aversion, liquidity) significantly affects speed of market re-access and spreads normalization. - Determinants of sovereign spreads (empirical model highlights): - Global risk aversion, fiscal fundamentals (overall balance, debt-to-GDP), and contagion (VI) significantly affect spreads. - Effects varied across crisis phases: build-up, systemic outbreak, systemic response, sovereign risk phase; contagion became more significant as the crisis progressed into sovereign stress phase. ### Reprofiling and domestic banking system — empirical review (Annex IV and AIV boxes) - Past reviewed reprofiling cases: Cyprus, Jamaica (2010, 2013), Pakistan (1999), Uruguay (2003). - Main findings: - Reprofilings in reviewed Fund-supported programs did not have a disruptive impact on domestic banking systems. - Containing factors included: design accounting for financial stability, banks holding sovereigns as HTM, regulatory incentives for participation, presence of capital and liquidity backstops, limited bank holdings of sovereign bonds, and selective forbearance. - Measured SD durations in past cases (staff findings, exact figures): - Cyprus (July 2013): 0.2 months. - Uruguay (May 2003): 0.5 months. - Jamaica (2010, 2013): 0.7–1.3 months. - Pakistan: 11 months (outlier). - Greece (March 2012): 2¼ months. - Central bank liquidity and accounting: - Debt reprofiling typically leads to a Selected Default (SD) rating, affecting collateral eligibility; ELA often used as temporary liquidity support. - Impact depends on banks’ accounting portfolios: Trading, AFS, HTM — reprofiling not per se requiring reclassification; impairment assessments and prudential filters matter. ### Risks, safeguards, and alternative of bailout - Bailing out creditors does not guarantee financial stability or remove program safeguards; past programs show Fund resources did not always avoid banking crises (Greece example). - Forbearance as a regulatory tool: - Permitting temporary relaxation of provisioning/capital rules can be last-resort but risks undermining market confidence. - Definition used in annex: bailout = creditors paid out before an eventual restructuring. - Measures to protect banking system and encourage participation in reprofiling: - Backstops (Financial Sector Stability Fund, central bank liquidity, FX swap lines), regulatory incentives (risk weights, temporary forbearance), capital support and bank resolution tools (bail-in). ### Model of comparative gains from slower consolidation enabled by reprofiling (Annex AV summary) - Framing and intuition: - Slower consolidation can exploit higher fiscal multipliers during recessions and hysteresis, preserving GDP and potential output. - Calibration overview and exact figures preserved: - Sample of ten exceptional access cases between 2009 and 2013: Armenia, Belarus, Greece, Ireland, Jordan, Portugal, Romania, Sri Lanka, St. Kitts and Nevis, and Ukraine. - Starting at t5, primary structural balance stabilizes at 1.5 percent of potential GDP. - Potential growth from t5 onwards set to 2 percent; real interest rate from t5 onwards set to 2.5 percent. - Table AV2 calibrated parameters (as reported): - Persistence of the output gap: 0.5 - Growth of potential GDP from t5 onwards: 1.02 - Hysteresis in the output gap: 0.1 - Real interest rate from t5 onwards: 1.025 - Elasticity of primary balance with respect to output gap: 0.48 - Intertemporal discount factor: 1/R - Fiscal multiplier range: between 0.25 and 1.5; larger in recessions. - Simulation results (selected exact outcomes): - Both frontloaded and gradual consolidation paths bring debt down to 75 percent of GDP in fifteen years. - Gradual consolidation involves a temporary increase in debt-to-GDP, limited to less than 10 percentage points of GDP and possibly persisting beyond 2–3 years under reprofiling. - Delaying consolidation yields a temporary GDP increase of more than 0.5 percent during first 5 years. - Over first 15 years, cumulative present discounted value effect is +0.15 percent of GDP. - After 15 years, potential GDP is 0.1 percent higher than under the baseline (permanent gain via reduced hysteresis). - Robustness and caveats: - Case for delayed adjustment strongest when multipliers are larger during crises and hysteresis present. - Delaying consolidation is counterproductive when overheating or if commitment to future medium-term adjustment is not credible. - Short reprofiling (1–3 years) may or may not provide sufficient space depending on liability maturity structure. ### Main policy implications drawn across annexes - Exceptional access should be constrained by rigorous DSA and clear criteria to limit moral hazard; forward-looking assessment of sustainability can incorporate intended restructuring. - Where initial debt is high, significant NPV reductions are generally required for a higher probability of a “cured” restructuring; light reprofilings should not be repeated if ineffective. - Reprofiling can be an effective, lower-cost tool in “gray zone” and liquidity-driven cases—if designed to protect financial stability and accompanied by credible macro policies and program conditionality. - Protecting domestic banking systems during reprofiling requires careful design: regulatory incentives, liquidity backstops, clear accounting guidance, and, where necessary, capital support or resolution measures. - Credible, timely policy action—whether reprofiling or restructuring—reduces contagion and shortens time to market re-access; international environment and global risk appetite materially affect outcomes. *Source: THE FUND’S LENDING FRAMEWORK AND SOVEREIGN DEBT—ANNEXES (excerpts provided in content unit _052214a).* ### Annex V was prepared by Damiano Sandri (RES) and Suchanan ### Annex V was prepared by Damiano Sandri (RES) and Suchanan Tambunlertchai (SPR). ### Overview - The annex describes how the Fund's approach to lending in sovereign debt crises evolved since the early 1980s, with emphasis on the origins and evolution of the current framework for exceptional access. - Three broad phases of Fund crisis involvement are identified: - Emerging markets debt crises of the 1980s: strategy relied initially on policy adjustment with rescheduling and new money as an interim source of finance, eventually resorting to debt reduction when adjustment proved insufficient. - Capital account crises during 1995–2002: non-sovereign in nature, rescue programs relied on exceptional access of Fund resources to catalyze private flows; in some cases private flows failed to materialize. - Post-1990s reforms and the 2008–10 period: experience (including Argentina and Russia defaults) led to the 2002 Exceptional Access Policy and a more rigorous DSA framework; the 2010 amendment created an exemption to the “high probability” public debt sustainability requirement for cases with high risk of international systemic spillovers (invoked in programs for Greece, Portugal, and Ireland). ### The Emerging Markets Debt Crises of the 1980s — key findings and chronology - Origins and context: - The debt crisis of 1982–83 resulted from massive external shocks and policy errors that reduced developing countries’ capacity to service external debts. - The turnaround in borrowing conditions from the 1970s, driven by U.S. policy changes, led to Mexico’s moratorium in 1982 and a regional crisis. - Fund policy tools and institutional developments: - The Supplementary Financing Facility (SFF) was introduced in 1979 to on-lend borrowed resources and to permit access above normal limits “in special circumstances.” - The “exceptional circumstances” clause was established in the 1983 review of the Enlarged Access Policy (EAP); the EAP replaced the SFF in 1981. - Strategy evolution: - Early phase through 1986: a “concerted lending” approach required firm creditor bank commitments for debt rescheduling and new lending as preconditions for Fund financing, typically within normal access limits. - Second phase from early 1987: recognition of the need for market-based debt operations to reduce the present value of debt, preceding the Brady Plan of early 1989. - Rationale and limits: - Initial responses treated the problems as liquidity-driven, expecting policy adjustment plus financing and growth to restore debt-servicing capacity. - By 1987, creditor banks had reduced exposure, growth remained elusive, and debt overhangs impeded growth; commodity price declines and delayed structural reforms contributed to faltering recovery. - Role in eventual debt reduction: - In 1987 staff advocated debt relief for heavily indebted countries; early debt operations included Bolivia’s buybacks, Chile’s debt-equity swaps, and Mexico’s exchange of bank debt with lower face value collateralized bonds. - The Brady Plan (introduced March 1989) provided debt relief that, together with Fund support, delivered debt sustainability. ### Fund Support for Debt and Debt Service Reduction (DDSR) Operations — structure and outcomes - Policy evolution: - In 1989 the Fund adopted a debt strategy allowing countries to use Fund resources, in conjunction with other official support, to finance upfront costs of DDSR operations with commercial banks (e.g., buybacks, escrow collateral). - Prior to 1989 the Fund's role had been mainly as facilitator for LICs. - Financing mechanics: - Fund support for DDSR operations was financed through set-asides (25 percent of access under Fund arrangements) and additional resources from augmentations (up to 30 percent of quota). - Set-asides and augmentations were released only after a member reached agreement with commercial bank creditors and the Executive Board agreed the resources would be used effectively. - Set-asides had to leave adequate margins to meet the member’s general balance of payments need. - Implementation and examples: - During the early 1990s countries such as Venezuela, Costa Rica, Mexico, and the Philippines benefited from Fund support for commercial creditor agreements. - The policy was pivotal in resolving the long debt crisis that started in the early 1980s but had limited use by the late 1990s as countries shifted to capital markets; the policy was terminated in 2000. - Poland case: - Under a 1994 stand-by arrangement, Poland received set-asides and augmented access to finance a commercial bank debt operation that regularized USD 14.4 billion of commercial debt at an upfront cost of USD 1.9 billion (divided among debt buybacks and purchase of collateral for principal). - The Fund judged Poland’s DDSR operation consistent with guidelines: voluntary commercial bank participation; market-based terms more favorable than secondary market conditions; significant authorities’ contribution; and a crucial step toward external viability. - Technical adjustments: - Initially DDSR support was segmented between set-asides and augmentations; in 1992 and later in 1994 the segmentation was modified and then eliminated to allow more flexible use of both resources for debt reduction and collateralization. ### Transitional lessons toward the 1995–2002 capital account crises (summary points appearing in text) - Capital account crises differed from 1980s sovereign debt crises: they were often triggered by broader balance sheet concerns and sudden stops rather than by excessive public debt per se. - Rescue programs in 1995–2002 relied on exceptional access to catalyze private flows; these catalytic roles sometimes failed, exposing members to excessive adjustment and prompting policy reforms. - Experience in the 1990s (including Argentina and Russia defaults) prompted the 2002 Exceptional Access Policy, adoption of a more rigorous DSA framework, consideration of a Sovereign Debt Restructuring Mechanism (SDRM), and advocacy for collective action clauses in sovereign bonds. - The global financial crisis increased demand for Fund resources and led to the 2010 amendment creating an exemption from the “high probability” debt sustainability requirement where there was a high risk of international systemic spillovers (invoked for Greece, Portugal, and Ireland). *Annex prepared within the IMF annexes document (May 22, 2014) — material excerpted from the supplied content.* ### 10. Capital account crises took different forms in the 1990s, reflecting distinctive domestic ### 10. Capital account crises took different forms in the 1990s, reflecting distinctive domestic ### Nature of 1990s capital account crises - First wave (beginning with Mexico’s “tequila crisis” in 1995 and followed by the Asian crisis of 1997): sudden reversals of private capital flows were driven by investor concerns about excessive dependence on short-term funding, currency and maturity mismatches, and perceived weaknesses in the financial and corporate sectors. - Exchange rate pressures and banking sector problems transformed private external financing gaps into sovereign funding stress. - After 1998: sovereign defaults of Russia and Argentina reflected sovereign debt crises engendered by significant fiscal imbalances and structural rigidities. ### Fund response and financing strategy in the 1990s - Rescue programs relied on massive Fund and other official financing aimed at catalyzing a return of private capital flows. - Strategy combined exceptionally large Fund access (and sometimes other official financing) with economic adjustment and reform by the member to restore confidence, halt capital outflows, and support transition back to private capital markets. - Access was heavily frontloaded given large up-front needs and the assumption of a short-lived crisis of confidence: - The Fund expected a quick turnaround of capital outflows; average length of arrangements was 2½ years and purchases of more than 70 percent of total access during the first year of the arrangement were available in all cases. (footnote text preserved) - Exceptional provision above normal access limits was enabled by a highly discretionary access policy: - The “exceptional circumstances” clause was invoked 14 times between 1995 and 2002 to provide access above normal limits in the credit tranches or the EFF. - In 1997 the Fund created the Supplemental Reserve Facility (SRF), not subject to access limits; first use agreed for Korea in December 1997 and later used by Argentina, Brazil, Russia, Turkey and Uruguay. ### Private sector involvement and creditor treatment - Pure catalytic approach relied primarily on Fund financing to attract private capital; limited and cautious forms of private sector involvement were attempted. - By the late 1990s, bonded debt prevalence and a more diffuse creditor base made coordination more difficult than in the 1980s. - Cases where private sector involvement was not needed or not attempted: Argentina (1995), Mexico (1994), Turkey (1994), Philippines (1997). - Where catalytic effect was weaker and financing shortfalls large, some form of private sector involvement was pursued; approaches varied by country circumstances. - Examples of creditor engagement measures: - Informal pressure on international banks to maintain exposure (Thailand, 1997; Brazil, 1999). - Direct pressure to transform interbank credit lines into new short- to medium-term instruments, with rescheduled bank debt benefiting from explicit government guarantees (Korea and Indonesia, 1998). ### Concerns about large-scale Fund financing and moral hazard (early 2000s) - Growing concern that large-scale Fund packages created adverse systemic risks and moral hazard when used to repay maturing debt obligations where debt sustainability was in doubt. - Russia and Argentina defaulted shortly after approval of augmentations to their Fund arrangements, heightening concerns. - Degree of discretion and flexibility in the framework was seen as making the Fund vulnerable to pressure to provide exceptional access when prospects for success were poor and sovereign debt burdens likely unsustainable. - Market participants argued lack of clarity in the framework created uncertainty in financial markets when a country approached the Fund. - Provision of exceptional access in these circumstances raised questions about adequacy of safeguards for Fund resources. ### Prague Framework, IMF/IMFC guidance, and move toward a new Exceptional Access Policy - “Prague Framework for Private Sector Involvement” (endorsed by the IMFC in September 2000) principles: - Extraordinary access to Fund resources should be exceptional. - A spectrum of actions by private creditors, including comprehensive debt restructuring, may be warranted to provide for an adequately financed program and a viable medium-term payments profile. - Operational framework for private sector involvement must be based on the Fund’s assessment of a country’s underlying payment capacity and prospects of regaining market access, and should rely on market-oriented solutions and voluntary approaches. - These principles set the stage for the development of a new Exceptional Access Policy in 2002. ### Box AI2 — Creditor Bailout in Fund-Supported Programs (summary findings) - Analysis of external debt restructuring episodes between 1995 and 2012 during Fund-supported programs assessed whether Fund (and other official) resources were used to pay out private creditors prior to restructuring or default. - Defined “bailout” as net positive financing from the official sector, including the IMF, that offsets reduced financing from the private sector (where data available, compared financing received against the debt service schedule). - Findings: - Private sector bailouts tended to occur in larger (more “systemic”) countries, with varying degrees. - Argentina: in months leading up to default in December 2001, received large net financing from the Fund which helped meet external private debt service. - Russia: bailouts occurred over a longer period; IMF disbursements since 1995 made up for shortfalls in 1995–1996; assessment of immediate pre-August 1998 bailout is more difficult due to net positive financing from both official and private sectors. - Pakistan: official financing (Fund’s share relatively small) began replacing private flows as the crisis intensified in 1998. - Greece: IMF and EU financing of around €48 billion, or 23 percent of Greece’s GDP, was used in part to finance the exit of private creditors in the two years leading up to the 2012 restructuring. - Turkey: Fund financing contributed to large sovereign debt repayments; extent of bailout limited since Turkey negotiated a voluntary PSI that maintained private sector exposure. - In less systemic economies, Fund support was conditional on, or came after, debt restructuring (Dominican Republic, Ecuador, St. Kitts & Nevis, Seychelles, Uruguay); restructuring was assumed in program baselines. - Ukraine: IMF financing and program NIR targets were coordinated so that authorities had no choice but to reschedule debt falling due; program designed to avoid private sector bailout. - Table summary of country cases (selected entries preserved in text): - Argentina: IMF program Mar-00 to Apr-05; External debt restructuring Yes; Private creditor bailout Yes; IMF disbursements were stepped up as crisis intensified; IMF provided net positive financing prior to default in December 2001. - Dominican Republic: IMF program Jan-05 to May-05; External debt restructuring Yes; Private creditor bailout No; Restructuring announced prior to IMF program; net IMF financing during program was small. - Ecuador: IMF program Jan-00 to Jul-00; External debt restructuring Yes; Private creditor bailout No; Restructuring announced prior to IMF program; roughly zero net IMF and official sector funding during the program. - Greece: IMF program May-10 to Mar-12; External debt restructuring Yes; Private creditor bailout Yes; IMF and EU financing of around €48 billion, or 23 percent of Greece’s GDP, used to pay out private creditors prior to restructuring in 2012. - Pakistan: IMF program Oct-97 to Nov-99; External debt restructuring Yes, partially; Private creditor bailout Yes, partially; IMF disbursements made until debt restructuring; net IMF financing small. - Russia: IMF program Jul-98 to Jul-98; External debt restructuring Yes, partially; Private creditor bailout Yes, partially; IMF disbursements stepped up as crisis intensified; IMF and private sector provided net positive financing, some used to repay domestic debt falling due. - Turkey: IMF program Dec-00 to n/a; External debt restructuring Yes; Private creditor bailout Yes; Turkey serviced debt until debt exchange in June 2001; significant IMF disbursements in interim. - Ukraine: IMF program Sep-98 to Sep-98 to Aug-99; External debt restructuring No; Private creditor bailout No; Program designed to avoid private sector bailout; net purchases small, tight NIR targets forced rescheduling. - Dominican Republic, Ecuador, Seychelles, St. Kitts & Nevis, Uruguay: in these less systemic cases, restructuring either preceded or was assumed in the program baseline and Fund support did not constitute a bailout. ### Development of the Exceptional Access Policy (EAP) since 2003 - New EAP adopted in 2002 and fully operational in February 2003 to constrain discretion and make exceptional access systematic. - Board agreed the following substantive criteria to justify exceptional access for members facing a capital account crisis: 1) 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. 2) A rigorous and systematic analysis indicates that there is a high probability that debt will remain sustainable. 3) The member has good prospects of regaining access to private markets within the time Fund resources would be outstanding. 4) 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. - Criterion 2 emphasized a “rigorous and systematic analysis of debt sustainability,” giving more prominence to the recently improved DSA: - A modified debt sustainability framework for market access countries, including standardized sensitivity analysis, was developed in May 2002. - Board agreed that access where debt restructuring was needed would normally be expected to be within normal access limits, though rare exceptions could warrant exceptional access (e.g., where country could not repay the Fund or additional financing was needed to limit disruption associated with debt reduction). - Staff cautioned that Fund financing should not unduly increase rigidity of the member’s debt stock (e.g., leaving countries heavily dependent on official financing). ### Debates and subsequent adjustments (2003–2004) - Prague Framework for PSI did not provide a clear mechanism to prevent holdouts; staff proposed a statutory approach (SDRM) to provide incentives for cooperative restructuring and to prevent holdouts via binding majority rules and an arbitration body (Debt Resolution Forum). - SDRM lacked sufficient international support and was not feasible by 2003 Spring Meetings; the IMFC welcomed enhancement of the contractual approach, notably inclusion of collective action clauses (CACs) in international sovereign bonds. - Board discussed whether regional and systemic contagion could justify exceptional access: - Staff argued cross-border contagion should not justify exceptional access if other criteria were not met, to preserve uniformity of treatment and avoid reviving moral hazard concerns. - Board concurred that Fund should provide exceptional access for regional/systemic contagion only if the other criteria are met. - Late 2003: Argentina and Brazil were granted exceptional access though neither satisfied all EAP criteria: - Argentina’s exceptional access was justified on the basis of sizable external financing gaps projected over the program period (424 percent of quota, with more than half available during the first year of the program). - Brazil sought precautionary support, which fell outside Criterion 1 (capital account crisis). - Early 2004: staff proposed revisions to address anomalies (allow lending when there was a potential rather than existing balance of payments need or when debt sustainability could be achieved on a “forward looking” basis). Board declined to approve these amendments at that time, agreeing instead that in noncapital account cases the procedures would apply and requests judged "in light of the four substantive criteria" without necessarily conditioning approval on meeting all criteria. *Source: THE FUND’S LENDING FRAMEWORK AND SOVEREIGN DEBT—ANNEXES (excerpts provided).* ### 22. It was not until 2009 that the Board adopted reforms aimed at resolving the perceived ### _052214a - 22. It was not until 2009 that the Board adopted reforms aimed at resolving the perceived ### Background and rationale for reform - Prior to 2009 the Exceptional Access Policy (EAP) exhibited an asymmetry of treatment between capital and noncapital account crises, which generated perceptions of inconsistency and ad hoc decision making. - In capital account crisis cases the framework effectively precluded the use of exceptional access when the debt position at the time of the member’s request was judged to be unsustainable even if sustainability could have been restored through policy adjustment and/or debt restructuring. - By contrast, in situations where financing needs did not stem from a capital account shock, the policy was far more permissive of access requests above normal limits. - The pre-2009 framework was also incompatible with provision of exceptional access under potential rather than existing balance of payments needs. - The staff proposal that led to the 2009 reform intended to streamline and clarify EAP criteria by adopting the general principle that lending above access limits was permitted as long as there was a credible strategy to address the debt situation and restore debt sustainability. ### 2009 reform: revised criteria for exceptional access - Under the 2009 reform all revised criteria for exceptional access would henceforth need to be satisfied for the approval of any request involving GRA access beyond the normal limits. - Criterion 1 to 3 were revised (revisions underlined) as follows: 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. 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. 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. 3) The member has prospects of gaining or regaining access to private capital markets within the timeframe when Fund resources are outstanding. 4) 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. ### Complementary toolkit and other 2009 changes - The 2009 reform doubled access limits to an annual limit of 200 percent of quota and cumulative limit of 600 percent of quota. - The 2009 reform added new “precautionary” facilities to the GRA lending toolkit to help contain contagion risks for countries with broadly sound fundamentals and policies. - The 2009 reform eliminated the SRF given its relative inflexibility and limited use. ### Recent experience under the EAP framework (post-2008) - The global financial crisis brought an unprecedented demand for Fund-supported programs after 2008. - Between 2004 and 2007 only two exceptional access stand-by arrangements were approved (Turkey and Uruguay in 2005); after 2008 the global financial crisis led to a spate of high-access programs and the commitment of more than US$250 billion in loans to member countries. - The global financial crisis exposed deep-rooted external imbalances in advanced economies; except for Cyprus (extended arrangement under normal access limits), other Euro Area (EA) program countries required historically unprecedented levels of Fund and official support. - EA program financing relied almost entirely on the official sector because EA program countries were taken out of the market for extended periods. - In May 2010 the EAP was revised in the context of the Greek program approval to create an exemption to the requirement for achieving debt sustainability with a high probability when there was a “high risk of international systemic spillovers.” - Since the systemic exemption was created, it has been invoked 34 times by end-May, 2014 in the three EA programs for Greece, Portugal, and Ireland. ### Greece: timing, restructuring, and financial magnitudes - By 2011, in the context of a deep and prolonged recession, massive financing needs and an undercapitalized banking system, it became clear that a debt restructuring would be needed for Greece. - At the time of the 4th Greece review, in mid-2011, the staff report acknowledged that PSI would be necessary to close financing gaps and address debt sustainability concerns. - The debt exchange in March/April 2012 resulted in a reduction of more than 50 percent of eligible debt (about €200 billion) after attracting a participation rate of close to 97 percent. - In late 2012, a second round of PSI was implemented through a buyback operation that retired sovereign bonds issued in the context of the debt exchange. - By the time of these debt restructurings private creditors had been paid out a total of almost €50 billion, drawing on resources provided by the Fund and other official creditors to support Greece’s program. ### Annex II: retrospective review of sovereign debt restructurings since the 1980s — core questions and findings - Purpose: address whether the 2013 Board paper’s assessment (that unsustainable debt situations often fester and restructurings do not always restore sustainability) holds over a broader period and sample. - Key questions examined: - When countries experience debt distress, how often and under what circumstances are they able to exit stress without needing a restructuring? - When restructurings have been used, have they generally been timely and adequately-sized? - In what circumstances have restructurings been effective in establishing sustainability? - Main findings (summary): - In many cases countries were able to avoid a debt restructuring; nevertheless, in a significant number of cases an exit from distress entailed a restructuring. - Lower initial debt levels and stronger fundamentals made it more likely that a country could exit debt distress without a restructuring. - Countries often delayed needed debt reduction and debt overhangs tended to persist for several years before a restructuring was undertaken. - Restructurings often appear to have been insufficient to bring debt down to lower and safer levels; reprofilings can be appropriate at moderate debt levels but should be followed by definitive debt reduction if they fail to pave the way to a sustainable outcome—“repeat” reprofilings should be avoided. - On average, two-thirds of a representative sample of restructurings with private foreign creditors in market access countries resulted in a “non-cured” outcome, requiring further restructuring. - Where initial debt was high a significant NPV reduction was needed to help increase the likelihood of a successful restructuring; where initial debt was at more moderate levels, a relatively small NPV reduction was enough to raise the likelihood of success. ### Data and sample used in Annex II analysis - Sample: 107 countries (34 advanced and 73 emerging economies) for which public debt data are available for at least 10 consecutive years over the period 1980–2012. - Total of 229 debt restructuring episodes over 1980–2012 considered in the sample: - 15 associated with domestic debt, - 99 with private foreign creditors, - 115 with official creditors. - Sources for data on debt restructurings include Reinhart and Rogoff (2009), Cruces and Trebesch (2013), Das and others (2012), and IMF staff reports. Main sources for debt and other macroeconomic data include Abbas and others (2010), Mauro and others (2013), the World Bank’s World Development Indicators and WEO. Data on credit ratings come from the three rating agencies (S&P, Moody’s and Fitch) and go through 2013. ### Definitions and thresholds used in the empirical work - Large debt reduction (LDR) defined as cases where: - (i) initial debt at the beginning of the episode exceeded 60 percent of GDP; - (ii) debt fell for at least 5 years subsequently; and - (iii) the total reduction was at least 10 percentage points of GDP. - Durable recovery in sovereign debt ratings defined as cases where: - (i) ratings reached a trough and recovered to at least an average B- rating; - (ii) they remained above the trough 5 years later; and - (iii) negative blips of at most one notch were allowed during the 5 years following the trough. - The selected sample of rating-based episodes includes countries that experienced a significant ratings downgrade (the average rating downgrade was about 4 notches) and excludes ongoing distress episodes (e.g., Greece, Jamaica, Belize) and cases of ratings downgrades leading to a final rating above A-/A3. *Source: IMF Staff Reports.* ### 32. While many countries were able to exit from debt distress without a restructuring, a ### _052214a - 32. While many countries were able to exit from debt distress without a restructuring, a ### Prevalence of restructurings in large debt reductions - Out of 68 cases of large debt reductions, about one half (32, all in emerging economies) entailed a restructuring of debt owed to private foreign, official, or domestic creditors. - The 32 debt reduction episodes correspond to 93 debt restructurings during the episode.* - Most restructurings occurred when debt peaked or during the 5 years thereafter. - When measured on the basis of recovery in credit ratings, one fifth of all cases (7 out of 34) entailed a debt restructuring. - Some of the 32 cases were preceded by other debt restructurings; only those restructurings that strictly coincided with the debt reduction episodes were considered in the analysis. - The remaining 36 cases (15 in advanced and 21 in emerging economies) entailed only two restructurings in the three years preceding the reduction episodes and none during or thereafter. *Footnote text retained from source: “The 32 debt reduction episodes correspond to 93 debt restructurings during the episode. The latter is larger because several restructurings may occur during a debt reduction episode and because restructurings of different types may occur in a single year (these were not aggregated in a single event in this exercise).” ### Characteristics of episodes that avoided restructuring versus those that required restructuring - Number of LDR episodes: 32 (associated with debt restructuring) and 21 (without debt restructuring). - Number of debt restructurings in sample: 93 (associated with debt restructuring) and 0–2 (without restructurings; two restructurings occurred before the debt peak in the sample). - Public debt level at t = 0: 100.4 (LDR associated with debt restructuring) and 79.8 (LDR without debt restructuring). - Public debt build-up (increase from t = -3 to t = 0): 43.8 (with restructuring) and 7.1 (without restructuring). - External public debt level at t = 0: 66.0 (with restructuring) and 47.2 (without restructuring). - External public debt build-up (increase from t = -3 to t = 0): 24.1 (with restructuring) and 10.7 (without restructuring). - Primary balance average, t = -3 and t = 0: 0.8 (with restructuring) and 1.7 (without restructuring). - Current account deficit average, t = -3 and t = 0: 3.0 (with restructuring) and 2.0 (without restructuring). - GDP growth average, t = -3 and t = 0, in percent: 1.1 (with restructuring) and 1.8 (without restructuring). - Bond spreads average, t = -3 and t = 0, in basis points: 1108 (with restructuring) and 462 (without restructuring). - Depreciation of the REER average, t = -3 and t = 0, in percent: 4.6 (with restructuring) and 1.9 (without restructuring). - Qualitative conclusion: countries that exited debt distress without the need for a restructuring generally faced milder initial conditions—lower initial levels of total public and public external debt, smaller rise in debt preceding the peak, higher primary and current account balances, stronger growth, lower spreads, and smaller exchange rate depreciation. ### Drivers of large debt reductions (with and without restructurings) - Debt decomposition components used: real interest rate (RIR), GDP growth, real exchange rate (RER), primary balance (PB), nominal reduction (Haircut), and a residual component. - Findings: - Real exchange rate changes played an important role during both the debt buildup (t = -3 to t = 0) and the debt reduction (t = 0 to t = 5). - Debt reduction was also driven by recovery of growth, some fiscal adjustment, debt treatment (nominal haircut), and lower real interest rate payments. - In large debt reduction episodes without the need for restructuring, the bulk of the debt reduction was achieved through strong growth and fiscal discipline. - The contribution of the nominal reduction is probably understated because of data availability. - Real interest rate payments are lower in LDR episodes with restructurings in part because they probably reflect flow treatment but also because of higher average inflation. - Note on data and methodology: the debt decomposition is based on a standard equation breaking down debt changes into the listed contributions; the residual reflects both data limitations (e.g., incomplete reconciliation between stocks and flows) and economic factors (e.g., privatization receipts, recognition of contingent liabilities, bank recapitalization, etc). ### Case studies: Brazil and Turkey - Context: - Both Brazil and Turkey experienced crises in the early 2000s but exited debt distress without deep debt restructuring. - The main finding: initial conditions mattered in both cases, but neither would have been reprofiled under the proposed new framework—Brazil because of confidence in fiscal fundamentals despite loss of market access; Turkey because it did not lose market access. - Brazil: - Crisis drivers: continued depreciation of the currency and large share of foreign-currency denominated debt leading to a large increase of public debt; political uncertainty around October 2002 presidential election triggered sharp increase in spreads, downgrades, and loss of market access. - Fund response: three Stand-by Arrangements prior to largest program; largest program in September 2002 with access of SDR 22.8 billion (US$30 billion, 752 percent of quota). - Assessment: staff expressed confidence in fiscal fundamentals and did not express concerns about debt sustainability; exceptional access aimed to coordinate political endorsement of the Fund-supported adjustment program; market access was regained in April 2003. - Turkey: - Crisis drivers: widening current account, fragile banking system reliant on short-term funding, years of weak economic and policy performance with continued high inflation; market pressure returned in early 2001 requiring augmentation of the Fund-supported program in mid-2001. - Fund response: SBA in December 1999 with access of SDR 2.9 billion, augmented to SDR 15 billion (about US$19 billion, 1560 percent of quota) by May 2001. - Assessment: despite weaker fundamentals, Turkey maintained market access (albeit at high interest rates) and was not downgraded to “selective default”; Turkey organized a voluntary and market-priced domestic debt swap and negotiated restored exposure from foreign creditors; staff noted reliance on voluntary private sector involvement and domestic debt swap to reduce borrowing needs and address roll-over problems. - Comparative indicators on eve of crises (averages between t-1 and t-3 where t = 2001 for Turkey and 2002 for Brazil): - Real GDP growth (annual percent change): Brazil 2.0, Turkey 1.4. - Overall fiscal balance (percent of GDP): Brazil -3.8, Turkey -14.6. - Primary balance (percent of GDP): Brazil 3.3, Turkey 1.9. - Net public debt (percent of GDP): Brazil 50.7, Turkey 55.7. - Net public debt (projected for 1st program year): Brazil 58.5, Turkey 79.7. - Current account balance (percent of GDP): Brazil -4.1, Turkey -1.1. ### Observations on measurement and thresholds - Large debt reduction (LDR) episode definition: - (i) initial debt was larger than 60 percent of GDP; - (ii) debt fell for at least 5 years subsequently; - (iii) the total reduction was at least 10 p.p. of GDP. - Durable rating recoveries definition: - (i) ratings downgrades and recovery to at least B- rating; - (ii) ratings remained above the trough 5 years later; - (iii) ratings experienced blips of at most one notch 5 years after the trough. - Additional notes from source text: - No country has been able to regain market access with a rating below B-. - Such cases (high-grade downgrades within AA/A3 segments) were excluded as credit agencies consider payment capacity of sovereigns with ratings above A-/A3 to be strong. - The share of restructurings measured by rating recovery is lower partly because sovereign ratings can fall below B- even if debt is not too high when facing temporary liquidity problems. - The signal approach used to estimate debt-to-GDP benchmarks indicates levels of 60 percent for emerging markets and 70 percent for advanced economies as predictive of debt distress events; alternative indicators from MAC DSA not used due to data limitations. ### Transition to assessment of restructuring effectiveness - The source concludes this section by posing the evaluative question: "When restructurings have been used, have they been timely and adequately sized?" *Italic source attribution retained from the supplied content.* ### 36. A broader sample corroborates the message of the 2013 paper that restructurings ### 36. A broader sample corroborates the message of the 2013 paper that restructurings have tended to be “too little too late.” ### Delayed and insufficient restructurings - Restructurings have often been unduly delayed; "debt overhang had been lingering for a number of years before a restructuring." - Reasons for delaying restructurings include weighing costs of delay against risks to financial stability, contagion, and perceived risk of protracted loss of market access. - Example cases and context: - Jamaica (2010): "65 percent of the debt was held by domestic financial institutions," raising concerns about financial sector distress. - Greece (2010): concerns about spillovers on other euro area countries. - Staff reports prior to restructurings were often not explicit about whether the debt problem was liquidity or solvency, even in some high-debt cases. - Short-term forecasts of GDP growth and inflation in IMF-supported programs "tended to be optimistic in high-profile cases characterized by exceptional access to IMF resources," according to the IEO report (2014). ### Repeat restructurings and scope - From 1980 to 2012: - Sample: 44 countries restructured their debt. - "86 percent had more than one restructuring." - On average each country had "over 5 restructurings," of which about half were with private foreign creditors (see Table AII2). - Repeat restructurings suggest one-time restructuring was often insufficient to solve the debt problem. - Example: Poland (1981–1990) underwent "six restructurings with private creditors and four with the Paris Club," mostly reschedulings; full resolution required Paris Club "50 percent debt forgiveness in 1991" and private banks' "45 percent debt reduction in 1994." ### Outcomes of restructurings with private foreign creditors - Of the 99 restructurings with private foreign creditors in the sample: - "Only about a third met the criteria for a ‘cured’ outcome" (i.e., a long-lasting solution). - Conversely, "two-thirds of all restructurings with private foreign creditors did not successfully establish sustainability and led to repeat restructurings." - Definition of a "cured" restructuring (paragraph 40): - (i) Not followed by another restructuring with private foreign creditors in the next four years; - (ii) The country does not remain in a state of default; - (iii) The country regains market access sometime following the restructuring and before another restructuring with private foreign creditors occurs. - Market re-access measured as: the first international bond/loan issuance after the restructuring (if it increased public and publicly guaranteed external debt with private creditors) and/or by positive net transfers from private creditors to the public sector. ### Determinants of successful restructurings (empirical findings) - Successful restructurings starting from high initial debt typically required larger principal reduction and also benefited from: - A good policy framework. - Benign external conditions. - Comparative evidence: - Larger debt reduction was needed for success "particularly when initial debt was high." - Low to moderate debt reduction can suffice when initial debt is relatively low. - Illustration from estimated probabilities (Figure AII7): - Two restructuring types considered: NPV reduction of "40 percent" (high) and "20 percent" (low). - Two initial debt levels considered: "50 percent of GDP" and "70 percent of GDP." - For a country with initial debt of "50 percent of GDP," there is a "50 percent chance it would reestablish sustainability with a small NPV reduction." - At an initial debt level of "70 percent of GDP," a small NPV reduction is more likely to perpetuate the problem; a larger debt reduction yields a larger increase in the probability of success. - Presence of an IMF program at the time of restructuring "also improves the likelihood of success." - Unfavorable global conditions, high initial debt level, and exogenous shocks (e.g., a natural disaster) reduce the probability of a cured restructuring. - Interpretation: "A reprofiling (even small) is more beneficial at more moderate levels of initial debt and should generally be avoided at high initial debt levels." ### Quantitative summary from Table AII2 (Restructuring Episodes by Country and Type, 1980–2012) - Total restructurings in table: - Private Foreign: 99 - Official: 115 - Domestic: 42 - Total: 228 ### Econometric approach (Box AII2) - Outcome variable: - Binary indicator for a "cured" restructuring (value 1 if cured, 0 otherwise), cross-sectional unit = restructuring episode with private foreign creditors. - Model: - Probit regression: P(cured)i = Φ(Xi'β), where Φ is the normal CDF and Xi is the vector of control variables. - Sample: - "There are in total 96 such cases" used in regressions after excluding some early and late-sample cases lacking information. - Control variable groups included: 1. Terms/characteristics of the restructuring: - Size of face value reduction (percent of debt treated) — expected sign: positive. - Size of haircut (NPV reduction, percent of PV of old debt) — expected sign: positive. - Instrument (bond exchange dummy) — expected sign: uncertain. - Debt treated previously (dummy) — expected sign: positive. 2. External shocks: - Natural disasters: average cost (fraction of GDP) during the four years following restructuring — expected sign: negative. - Global liquidity: average spread between 30-year U.S. bond yields and 30-year Baa corporate bond yields (four years after restructuring) — expected sign: negative (low spread = higher liquidity = more demand for sovereign debt). 3. Policy/institutions: - IMF program dummy (first year of a program starting in year of restructuring or previous two years) — expected sign: positive. - Political institutions: new government dummy (first two years of a new government over the four-year pre-restructuring period) — expected sign: positive. 4. Initial conditions/macro fundamentals: - Debt level in the year before restructuring — expected sign: negative. - Interaction between initial debt and debt reduction (nominal or NPV) — expected sign: positive. - Level of development: log real GDP per capita — expected sign: positive. - Openness: average (exports + imports)/GDP in four years before restructuring — expected sign: positive. - Macro volatility: standard deviation of GDP growth in a 10-year window before restructuring — expected sign: negative. - Coverage ratio: average reserves-to-imports ratio (months) in the four years prior — expected sign: positive. - Results: - Four baseline models estimated (two "long" models with full controls; two "short" models excluding political institutions, income volatility, openness, and coverage ratio). - Most estimated coefficients were statistically significant and had the expected sign (GDP per capita being the notable exception). - Key findings: high initial debt reduces odds of success; an appropriately large haircut improves odds of success; interaction between initial debt and debt reduction is significant and positive. - Estimated probabilities from the “long model” (NPV reduction case) are shown in Figure AII7. - Robustness: - Robustness checks with alternative controls did not overturn baseline results but generally reduced regression fit. *Source: THE FUND’S LENDING FRAMEWORK AND SOVEREIGN DEBT—ANNEXES (extracted content unit: 36–Box AII2).* ### 42. Evidence on the effectiveness of restructurings can also be drawn from case studies. ### 42. Evidence on the effectiveness of restructurings can also be drawn from case studies. ### Ex-ante assessment - Staff was not always confident ex ante that planned debt restructurings would re-establish debt sustainability in cases including: Ecuador 2000, Belize 2007, Jamaica 2010 and 2013, Moldova, Greece 2012. - Expected outcomes were often conditioned on continued ambitious adjustment. Example: January 29, 2010 request for an SBA for Jamaica stated the debt exchange was “expected to result in a significant extension of maturities and interest savings”, but also that “Given Jamaica’s debt overhang problem, public sustainability risks remain high.” ### Ex-post assessment - In many cases it is unclear whether staff assessed debt sustainability to have been achieved ex post. - Re-establishing sustainability often still hinged on further adjustments or overall positive macroeconomic developments. Cases cited include Argentina, Greece, Jamaica 2010, Grenada. - Example: Grenada (2006)—staff noted program adjustment would bring public debt on a sustainable path only if fiscal and growth objectives were accomplished, warning “Failure to adhere to program targets would rapidly result in an unsustainable debt trajectory.” ### Final outcomes from case studies - Of eight cases that were assessed to have solvency issues before restructuring, only five included a cut in face value. - Two of the three cases that only underwent a reprofiling continue to have debt problems: Grenada and Jamaica. ### Examples and effectiveness of reprofiling - Reprofiling was sufficient to restore market access and debt sustainability in some cases when sustainability was uncertain or the problem was liquidity-related, provided program implementation was strong: Dominican Republic, Pakistan, Uruguay. - Reprofiling was not successful where debt was too high or financial stability concerns required addressing banking-sector fragility: Grenada, Jamaica 2010. - Reprofiling was mostly used in normal access cases; Uruguay (2003) is a notable exception in the context of a Fund-supported program with exceptional access. ### Box AII3 — Reprofiling in Uruguay (summary of case study) - Trigger: - Macroeconomic deterioration after Brazilian and Argentine crises; high dollarization, dependence on nonresident deposits, large share of public debt in foreign currency. - 2001 Argentine default → run on deposits → depletion of reserves, rising borrowing costs, ratings downgrade, 50 percent exchange rate devaluation, severe recession in 2002. - Public debt rose from 40 percent of GDP in 2001 to about 100 percent in 2003. - Fund-supported program approved March 2002, augmented June 2002 and August 2002. Authorities sought an extension of maturities in May 2003. - Debt exchange design: - Intended to achieve debt service relief in a creditor-friendly way: low NPV haircut and no nominal reduction; “preemptive”, comprehensive (targeted about half of total debt), distinct instruments for domestic and external creditors. - Bondholders offered a choice between two approximately par-for-par options: a maturity extension option and a benchmark bond option. - Nearly two-thirds of bondholders chose the maturity extension option. - Average maturity extension: international bonds under foreign law extended by an average of 6.4 years; domestically issued bonds and treasury bills extended by 8.6 years on average. - NPV reduction: about 13.3 percent for international bonds and 23.3 percent for domestic bonds (resulting from implementing reprofiling at original issuance interest rates rather than prevailing market rates). - Inter-creditor equity measures: (i) asymmetric exchange ratio (new instruments’ face value varied by maturity while aggregate face value equaled the old), (ii) asymmetric cash payments (some bondholders received small upfront cash payments to compensate accrued interest). - Timing: launched April 10 and closed May 29, 2003. - Participation: average participation was 89 percent for international bonds and 99 percent for domestic obligations. - Risks flagged by Fund staff: - Potential problems in implementation, difficult economic environment, impact on banking system. - Post-exchange uncertainty persisted: “vulnerability has decreased but risks to the program remain significant...The main risks are in the still-fragile banking system and the continuing challenges to meeting the primary surplus targets required for debt sustainability” (IMF 2003). - Outcome: - Macro fundamentals: public debt declined from crisis highs driven by robust growth recovery, sizeable fiscal balances and some reversal of the exchange rate devaluation. - Policies: tight control over current expenditure; adjustment of public tariffs in line with costs; implementation of a bank restructuring program; maintenance of a floating exchange rate; prudent debt management with longer maturities and better terms. - Market access: regained market access in June 2003 issuing a US$53 million dollar-denominated bond with 7.7-year maturity and 10.6 percent yield to maturity; subsequent issuance in October 2003 of an international bond denominated in domestic currency with 3-year maturity. Spreads returned to pre-crisis levels and credit ratings recovered over the following months. Selective default downgrade lasted less than a month; investment grade was regained only in 2012. - Domestic financial stability: maturity extension had a small impact on domestic banks (low exposure to government bonds); bank supervisory authorities provided regulatory incentives for bank participation (e.g., central bank only allowed new bonds as eligible collateral). ### Conclusions drawn in the retrospective - Three main lessons: - (i) A debt restructuring has often been part of the solution to addressing debt overhang. - (ii) Restructurings have often been delayed and inadequately sized. - (iii) ‘Curing’ debt distress often requires not only resilience to shocks and prudent macroeconomic policy but also adequate debt reduction when initial debt is high. - Policy implication on scope of restructurings: - A light debt restructuring is sufficient to address debt distress when initial debt is low to moderate. - Light restructurings generally should not be used for high debt cases; if tried and they fail, they should not be repeated but followed by a deeper and more definitive debt reduction. - Cost-benefit insight: - Considering the costs associated with deep restructurings, light restructurings have reasonable prospects for restoring sustainability at moderate levels of public debt. *Source: _052214a - 42. Evidence on the effectiveness of restructurings can also be drawn from case studies.* ### ANNEX III. COMPARATIVE COSTS AND BENEFITS OF ### ANNEX III. COMPARATIVE COSTS AND BENEFITS OF REPROFILING, BAILOUT, AND RESTRUCTURING IN GRAY ZONE CASES—MARKET DYNAMICS AND INTERNATIONAL SPILLOVERS ### A. Implications of a Reformed Fund Policy Framework on Spreads and Maturity Structures of Government Debt - Context and assumptions - Analysis assumes adoption of a new Fund lending policy that leads markets to perceive a lower probability of being bailed out and a higher probability of being reprofiled. - Assumes reprofiling option would only be applied to government bonds coming due within the duration of the Fund program. - Steady state refers to tranquil periods when a country is not in debt distress. - Steady-state market implications - Investors would factor in a greater probability of a credit event in the pricing of risks; spreads may increase, especially for countries perceived as more at risk of debt distress. - Markets would better differentiate sovereigns based on macroeconomic fundamentals and fiscal discipline; investors demand higher interest rates from governments with higher debt levels, fiscal deficits, and external financing requirements. - Countries with better fundamentals could see increasing demand for their government bonds and possibly lower spreads. - For riskier countries, credit risk premia would tend to rise across the yield curve and more so at the shorter end: - Shorter-term bonds are more likely to be reprofiled than longer-term bonds; investors would be more hesitant to buy or roll over shorter-duration bonds in the steady state. - Demand for shorter-term bonds would decrease; the maturity structure would shift toward longer tenors. - A shift toward longer tenors (assuming a positively-sloping yield curve) implies: - Higher average cost of the debt portfolio. - Reduced rollover risk. - Impact depends on country cost-risk preferences; long-term investors less likely to be reprofiled could benefit and, if reprofiling averts deeper restructuring, would face a lower haircut due to a bigger creditor base. - Governments with less developed domestic debt markets may rely increasingly on international bond issuance: - Placement of longer-term bonds is limited by domestic investor absorption; riskier countries that rely on shorter-term instruments will be most affected. - Issuing more international bonds would heighten currency risks. - Sovereign borrowing cost increases would likely raise corporate sector borrowing costs (sovereign yields often benchmark corporate debt), with stronger transmission where sovereign-to-private cost pass-through is higher. - Box AIII1: The Long-Term Effects of Reprofiling on EM Borrowing Costs (key points) - Net effect on yields depends on whether the new policy is viewed as a tightening relative to a bailout or a more flexible approach for gray zone countries. - Evidence from three episodes (shifts in policy/regime or increased risk aversion) suggests effects for EM sovereigns with strong fundamentals would be small or possibly beneficial. - Three specific empirical observations: - EM yields following the introduction of the 2002 Exceptional Access framework: For strongly rated countries, the change resulted in falling borrowing costs; middle-rated countries saw little change; poorly rated sovereigns saw yields increase. Date of policy changes: 3/2003. Ratings for countries displayed on the chart refer to Mar/2014. - Borrowing costs and collective action clauses (CACs): In theory ambiguous; empirical evidence (Mexico, Turkey) shows a non-discernible difference in maturity-adjusted yields between bonds with and without CACs; evidence consistent with literature that CAC impact depends on sovereign credit rating. - EM yields during the global financial crisis (post-Lehman): Borrowing costs spiked briefly for all countries but soon decoupled according to macro fundamentals; countries grouped according to IMF Vulnerability Exercise as of 09/2008; series indexed to value as of 7/1/2008. - Periods of heightened stress - Reprofiling option may calm market concerns at times of stress, though it could advance the onset of distress if seen as triggered earlier. - Reprofiling may reduce spreads volatility in times of stress: average steady-state borrowing costs would be higher, but spreads may widen less relative to bailouts or deep NPV-haircut restructurings. - By preventing payments to external creditors, reprofiling could leave more resources domestically and permit countercyclical fiscal policy and reforms; however, if a large share of bonds is held domestically, stopping payments may hurt banks’ profitability and slow the real economy. - Contagion may spread to more countries (more sovereigns considered “at risk” for reprofiling) but the magnitude of the shock is likely reduced—spreads rise in more countries but at a lesser magnitude than under the old policy. - Reprofiling is growth-enhancing and volatility-reducing only if it successfully avoids a future restructuring; reprofilings used when debt is clearly unsustainable will eventually be followed by deeper restructurings. ### B. Do Reprofilings and Restructurings Have Significantly Different Impacts on Spreads, Credit Ratings, and Market Access? - Main empirical finding (sample-based) - Based on a sample of 12 restructuring country episodes, equally divided between reprofiling and debt restructuring with face value cut, reprofilings on average appeared to be less costly than face value cut restructurings. - Comparative outcomes (sample evidence) - Sovereign spreads were generally higher in restructurings compared to reprofilings, both at announcement and at the time of the debt exchange. - Credit downgrades were more severe in restructurings with nominal reductions than in reprofilings. - Sovereign credit ratings recovered faster in reprofilings than in restructurings within 12 months after the exchange (it took twice as long for restructurings to recover to precrisis credit ratings). - Time to re-access markets was longer for restructuring countries than for reprofiling ones (both time to a new global bond issuance and normalization of spreads to precrisis levels). - Negotiations in restructurings took longer than in reprofilings. - Participation rates were higher and litigations were fewer in reprofilings than in restructurings with nominal debt reductions. - Caveats and interpretation limits - Results should be interpreted with caution due to: - Lack of sufficient reprofiling cases for a rigorous empirical study (low statistical power). - Difficulty disentangling effects of the debt exchange itself from policies implemented after the debt operation. - Uncertainty from introduction of a new framework on market dynamics and domestic and cross-border financial stability. - When restructuring is necessary to resolve sovereign distress, reprofiling may not be a viable alternative. - Sample composition (as specified in source) - Reprofiling cases in the sample: Pakistan (1999), Moldova (2002), Uruguay (2003), Grenada (2004), the Dominican Republic (2005), Belize (2006). - Restructurings with nominal reductions in the sample: Ecuador (1999), Russia (1999), Dominica (2004), Argentina (2005), Seychelles (2008), St. Kitts and Nevis (2011), Greece (2012), Belize (2012). - For the purposes of the annex, reprofiling is defined as a moderate, face-value preserving maturity extension and NPV reduction. *Source: ANNEX III. COMPARATIVE COSTS AND BENEFITS OF REPROFILING, BAILOUT, AND RESTRUCTURING IN GRAY ZONE CASES—MARKET DYNAMICS AND INTERNATIONAL SPILLOVERS (IMF).* ### 50. Restructurings have been associated with higher spreads at the time of the ### _052214a - 50. Restructurings have been associated with higher spreads at the time of the ### Spreads at the Time of Announcement - Sovereign spreads at the time of the announcement of the credit event: - Restructuring cases: median of 2,119 bps. - Reprofiling cases: median of 1,053 bps. - Change relative to the median spread of the previous 12 months: - Restructurings: spreads increased by 727 bps over the median of the previous 12 months. - Reprofilings: spreads increased by 216 bps over the median of the previous 12 months. - Note: Pakistan, with the spread of almost 20,000 bps at the time of the reprofiling, was a clear outlier and is not reported in Figure AIII2. ### Spreads at the Time of the Debt Exchange - Sovereign spreads at the time of the debt exchange: - Spreads were higher in restructuring cases than in reprofiling cases (with the exception of Pakistan). - The difference between restructurings and reprofilings is typically smaller at the time of the debt exchange and statistically not significant. - Interpretation: uncertainty may have been resolved by the time the country concluded the debt exchange. ### Effect on Credit Ratings - Default/selective default assignment: - For rating agencies that assign a default rating, both reprofilings and restructurings resulted in a selective default rating (technical, not judgmental). - Credit downgrade severity: - Median downgrade in restructuring cases was almost twice as large (in number of notches) as in reprofiling cases (Figure AIII4). - Credit rating recovery within 12 months after the exchange: - Reprofiling cases: credit ratings recovered by an average of 3 notches within 12 months. - Restructuring cases: credit ratings recovered by an average of 2 notches within 12 months. - Median recovery (alternative presentation in Figures AIII5): - Reprofiling cases: median = 2.9 (st. dev. = 2.6) and median = 3 (st. dev. = 0.3) in two panels. - Restructuring cases: median = 5.3 (st. dev. = 3.4) and median = 1.7 (st. dev. = 1.4) in two panels. - Time to recover to pre-crisis ratings: - It took on average four times as long for restructuring cases to recover to precrisis credit ratings compared with reprofilings. - Heterogeneous outcomes: some rose to investment grade (Russia and Uruguay); some languished in the B category (Grenada, Ecuador, Belize, Jamaica, Pakistan, Greece, and Cyprus); some were withdrawn at the government's request (Seychelles). ### Effect on Market Access - Time to re-access international capital markets (defined as issuance of another international security): - Median time for restructurings to re-access and issue global bonds: 57 months. - Median time for reprofilings to re-access and issue global bonds: 35 months. - Caveats: Argentina has not re-accessed international markets since the exchange in April 2005. Moldova has not yet regained market access following its 2002 reprofiling, partly due to Fund program conditionalities. Belize (2013) and Cyprus (2013) were not included in the sample. - Time to re-access when defined as return of spreads to precrisis levels (normalization): - Median time for restructuring cases: 35 months. - Median time for reprofiling cases: 8 months. - Note: normalization is defined as the time it took for spreads to drop to the 12-month average prior to the distress event (announcement of default, restructuring, Fund program, suspension of debt payments). Should be interpreted with caution for countries that took a long time to finalize the exchange (Argentina). - Examples and medians reported: - Reprofiling cases (spreads normalization): median = 8 (st. dev. = 9). - Restructuring cases (spreads normalization): median = 35 (st. dev. = 24). - Time to market access (global bond issuance) panels: reprofiling median = 35 (st. dev. = 52); restructuring median = 57 (st. dev. = 42). - Time to market access (alternative panels): reprofiling median = 10 (st. dev. = 36); restructuring median = 45 (st. dev. = 65). ### Factors Affecting Speed of Market Normalization After Restructuring/Reprofiling - Restructuring characteristics: - Size of NPV loss/haircut: - Governments imposing smaller haircuts tended to normalize market conditions faster. - Empirical evidence (Cruces and Trebesch, 2013): size of NPV loss correlated with length of time until market re-access. - Examples: Uruguay and the Dominican Republic (small NPV losses) re-accessed markets relatively quickly; Argentina and Russia (large nominal reductions and NPV losses) experienced much longer periods before market re-access. - Government indebtedness: - Market conditions took longer to normalize in countries with high debt-to-GDP ratios at the time of the restructuring. - Governments with lower debt-to-GDP ratios at the time of restructuring achieved normalized spreads faster than governments with higher debt-to-GDP ratios. - Caution: time to re-access counted from the time of the credit event; for countries like Argentina with long delays between announcement and completion, results require caution. - Credibility of fiscal adjustment and policies: - Credible fiscal adjustment was associated with faster normalization and quicker market re-access. - Credibility often embodied in conditionality of Fund-supported programs that accompany debt restructurings and assure medium- to long-term debt sustainability. - Governments that went off track (fiscal slippage) and did not achieve projected fiscal consolidation took longer to re-access the market (evidence using global bond issuance measure). - Fiscal slippage definition: cumulative deviation (over two years) of primary balance-to-GDP ratios from the primary balance envisioned at the time of the program request. *THE FUND’S LENDING FRAMEWORK AND SOVEREIGN DEBT—ANNEXES; INTERNATIONAL MONETARY FUND* ### 62. The international environment also played a significant role for the time taken for ### 62. The international environment also played a significant role for the time taken for restructuring country to re-access markets ### International environment and market re-access - Prevailing international investors’ risk appetite, global financial liquidity conditions, issuance and volatility in mature bond markets, and the general economic and financial situation in major countries affect the length of time taken to re-access markets. - A favorable external environment cannot substitute permanently for required domestic policy changes, which are a prerequisite for a restructuring country seeking market re-access. - Empirical relationship: there was a positive relationship between global risk aversion and volatility (proxied by VIX index) prevailing at the time of restructuring/reprofiling and the time it takes for market conditions to normalize (Figure AIII12). - Time to market reaccess measured in two ways: - Time to reaccess based on global bond issuance after the last debt exchange. 3/ - Time to reaccess based on spreads normalization following the last debt exchange. Normalization is defined as the time it took for spreads to drop to the 12-month average prior to the distress event (announcement of default, restructuring, Fund program, or suspension of debt payments). 4/ - Notes: - Reprofilings are proxied by cases involving moderate (face-value preserving) maturity extensions and NPV reductions. - For countries that have not yet regained market access (as defined in the source), the duration of market access loss is based on April 2014 cut-off date. - Higher risk aversion is reflected by increases in the VIX index. ### Sovereign contagion: measurement and methodology - Sovereign contagion assessed using the vulnerability index (VI): - VI characterizes the probability of distress of a country conditional on other countries becoming distressed; the index ranges from zero to one. 65 - VI embeds distress dependence across sovereigns and their changes throughout the economic cycle, capturing direct linkages (trade, capital flows, financial sector) and indirect linkages (common risk factors and contingent liabilities). - VI estimated using EMBI spreads for emerging economies and sovereign CDS spreads for advanced economies. - When data availability allowed, data spanned from at least 12 months before the beginning of each crisis through at least 12 months after spreads appeared to go back to normal levels. - Assessment approach: - Evaluated how vulnerable crisis countries were to distress in other countries and how relevant crisis countries were as a source of contagion to related countries. - For each crisis, quantified the percentage contribution of crisis countries to the VI of related countries. - Related country set included: Argentina, Brazil, Chile, Colombia, Indonesia, Malaysia, Mexico, Philippines, Poland, Russia, Thailand, Turkey, the United States of America, Japan, the United Kingdom, France, Germany, Italy, Spain, the Netherlands, Belgium, Austria, Greece, Ireland, Portugal, and Sweden. 66 ### Evolution of contagion in crisis episodes (summary findings) - Sovereign contagion rose significantly during crises until policy actions to resolve sovereign distress were taken, irrespective of whether the policy decision was restructuring or reprofiling. - Once policy responses to address sovereign distress were embarked upon, sovereign contagion subsided irrespective of the debt exchange policy decision taken. - Contribution of crisis countries to international contagion increased in periods of heightened uncertainty and declined when uncertainty diminished. - Complementary empirical analysis using a different approach (Box AIII4 in source) finds similar results. ### Case studies — Restructuring cases (selected evidence) - Russia (1998): - Increased uncertainty: weakening fiscal position in first half of 1998, exogenous shocks to oil prices and the Asian crisis, market access loss in second half of 1998; sovereign default and currency devaluation in August 1998; sovereign debt restructuring announced in September 1998 leading to exploding spreads and peak contagion (Figures AIII13 and AIII15). - Diminished uncertainty: by May 1999, start of debt negotiations and improvements in economic prospects (coincident with commodity price improvement) diminished uncertainty and contagion subsided. - Argentina (2002): - Increased uncertainty: economic deterioration from end-2000, loss of market access, failed liquidity measures in 2001 (the “mega-swap” in June, voluntary debt exchange in November), sovereign default in December 2001; 2002 devaluation and GDP collapse; contagion soared in last quarter of 2002 as situation in Uruguay and other Latin American countries worsened (Figure AIII13). - Diminished uncertainty: in 2003, positive expectations of a new government and announcement of intention to implement a debt exchange diminished contagion. - Greece (2010): - Increased uncertainty: October 2009 revelation of understated debt and deficit figures, rating downgrades in December 2009, IMF program announced March 2010, spreads over German bonds soared in second quarter 2010 excluding Greece from bond markets; April 2011 European Commission deficit figures worse than expected led to a spike in contagion. Greece became highly vulnerable and a major source of contagion within the euro area (Figure AIII13). - Diminished uncertainty: various policies in Greece and other euro area members culminating with announcement of the OMT appear to have diminished uncertainty and reduced contagion. ### Case studies — Reprofiling cases (selected evidence) - Uruguay (2002): - Increased uncertainty: close links to Argentina caused spillovers after Argentina’s collapse; deposit runs as 40 percent of deposits in Uruguay were from Argentina; banking system and reserves rapidly depleted leading to floating currency; sovereign risk soared by mid-2002 (Figure AIII14). Uruguay’s contribution to contagion was not the highest in the region. - Decreased uncertainty: March 2003 announcement of debt restructuring through maturity extension and improved communication about economic prospects diminished uncertainty and contagion. - Dominican Republic (2005): - Increased uncertainty: slowdown in 2003 due to decreased tourism, a banking-sector fraud scandal and bank closure, resulting fiscal costs and liquidity injections that fueled currency depreciation and high inflation; debt–to-GDP ratio doubled from 26 percent in 2002 to 54 percent at end-2003; fiscal slippages and high inflation in early 2004 led to loss of market access and significant contagion (Figure AIII14). - Decreased uncertainty: second half of 2004 and first half of 2005 saw successful restructuring of commercial banks and external bonds, reducing uncertainty and contagion. ### Comparative magnitude of contagion: restructuring vs reprofiling - Empirical results indicate contagion appears higher in restructuring than in reprofiling cases for the sample analyzed. - Contributions to international contagion from Russia, Argentina, and Greece were significant and in each case the highest among their related countries during periods of increased uncertainty. - In contrast, in periods of increased uncertainty the contributions to international contagion from Uruguay and the Dominican Republic were not the highest among their related countries (Figure AIII15). *Source: THE FUND’S LENDING FRAMEWORK AND SOVEREIGN DEBT—ANNEXES (excerpt).* ### 66. Nevertheless, empirical results are subject to important caveats. Data availability and ### _052214a - 66. Nevertheless, empirical results are subject to important caveats. Data availability and ### Caveats and limitations of the empirical results - Data availability and quality restricted the sample size, making it difficult to draw robust inference. - Crisis countries in the sample that went through reprofiling are not major economies and do not host major financial centers; contagion from these countries could have been limited irrespective of debt exchange decisions. - It is difficult to separate effects of debt exchange decisions (policies that affect uncertainty about the resolution of sovereign distress) from other factors affecting uncertainty. - Example: contagion in the Greek restructuring dwarfs that observed in all other cases under analysis, but contributions from other euro area countries (Portugal, Ireland, Italy, and Spain) to international contagion during the crisis were highly significant even though these countries did not go through debt exchanges (Figure AIII15). This suggests magnitude of contagion may reflect a composite of factors. ### Vulnerability Index (VI) — purpose and formulation - Purpose: - The VI quantifies sovereign contagion by assessing how vulnerable a country is to distress in other countries. - The VI characterizes the probability of distress of a country conditional on other countries becoming distressed, and allows assessment of how relevant crisis countries are as sources of contagion to related countries. - Formula: - SC(Ai) = ∑ P(Ai/Aj) · PoD(Aj) for all j ≠ i - SC(Ai) is the vulnerability index for country Ai. - PoD(Aj) denotes the probability of distress of country Aj. - Contribution to contagion: - Contribution to Contagion (Ai) = ∑[ [P(Ai/Aj) · PoD(Aj)]/SC(Ai)] for all j ≠ i - This is the percentage contribution of each country to the vulnerability index of related countries. - Construction of P(Ai/Aj): - (i) Marginal probabilities PoD(Ai) and PoD(Aj) are extracted from EMBI or CDS spreads. - (ii) Joint probability P(Ai, Aj) is obtained using the CIMDO methodology (Segoviano (2006)) to estimate the multivariate empirical distribution (CIMDO-distribution) characterizing probabilities of distress and distress dependence across sovereigns. - (iii) Conditional probability obtained via Bayes’ law: P(Ai/Aj) = P(Ai,Aj) / P(Aj). ### CIMDO methodology and key technical features - CIMDO is a non-parametric methodology based on the Kullback (1959) cross-entropy (minimum cross-entropy) approach that: - Does not impose parametric pre-determined distributional forms. - Is constrained to match empirical probabilities of distress observed for each sovereign (from EMBI or CDS spreads). - Estimates joint probabilities for the group and pairwise combinations (P(Ai, Aj)) from the CIMDO-distribution. - Advantages over traditional risk models: - Characterizes entire distributional links (linear and nonlinear distress dependence) and their evolution across the economic cycle. - Reflects that dependence increases in periods of distress (time-varying dependence captured by a CIMDO-copula). - Technical outline (selected elements preserved as in source): - Objective function: C[p,q]=∫ ∫ p(x,y) ln( p(x,y) / q(x,y) ) dx dy, where q(x,y) is the prior distribution and p(x,y) the posterior distribution. - Moment-consistency constraint equations incorporate empirically observed probabilities of distress (PoDs) and indicator functions defined with distress thresholds. - The CIMDO density is recovered by minimizing a functional that includes cross-entropy terms and Lagrange multipliers ensuring moment-consistency and probability additivity constraints. - The optimization via calculus of variations yields the posterior multivariate density. ### Determinants of sovereign risk — empirical model specification - Model framework: - A modified GARCH(1,1) model with explanatory variables in the mean equation was used to analyze determinants of sovereign swap spreads at daily frequency (Box AIII3). - Swap spreads measured as spreads of sovereign bond yields to a common numeraire (for euro area: 10-year euro swap rate). - The matrix of explanatory variables includes: - Index of Global Risk Aversion (IGRA): market price of risk (Espinoza and Segoviano (2011)). - Contagion: captured by the vulnerability index (VI). - Country-specific fundamentals: overall balance as percent of GDP and debt-to-GDP ratio. - The model includes a lagged dependent variable to capture high persistence in daily swap spread data. - Complementary approach: - A multivariate GARCH framework (Dynamic Conditional Correlation, DCC, Engle (2002)) was employed to jointly model multiple daily bond spread series and extract heteroskedasticity and time-varying correlations (co-movements) in conditional variances (Box AIII4). - Interpretation: low estimated co-movements between a crisis country and related economies indicate relatively low spillovers; high co-movements indicate higher outward financial spillovers. ### Empirical findings on determinants and timing (Euro area during the GFC) - Overall: - Global risk aversion, fiscal fundamentals, and contagion have a significant impact on sovereign spreads. - The effect of these factors changed across different stages of the crisis (Figure AIII16 and Figure AIII17). - Stages and observed effects: - Financial crisis build-up (July 2007–September 2008): - Contagion not a particularly significant factor. - Germany and other core sovereigns benefited from global risk aversion (flight-to-quality). - Peripheral countries saw rising spreads as global risk aversion weighed adversely on lower-rated issuers. - Fundamentals were supportive (deficits and debt levels improving), so contagion was not a primary driver. - Systemic outbreak (October 2008–March 2009): - Contagion gained significance as banking sector problems spilled over to sovereign balance sheets and fundamentals deteriorated. - Spreads rose due to contagion from countries directly involved in the financial crisis (Austria, the Netherlands, Ireland) and from deteriorating fundamentals. - Global risk aversion no longer favored sovereigns as crisis interventions and fiscal stimulus diluted perception of sovereigns as riskless. - Systemic response phase (April 2009–September 2009): - Effect of contagion on spreads decreased as lower probability of distress in some countries favorably affected others. - Spreads converged; larger corrections among underperformers (Italy, Greece) although spreads remained wider than before. - Sovereign risk phase (October 2009–August 2012): - Swap spreads increasingly driven by country-specific developments (short-term refinancing risk and long-term fiscal sustainability). - From March 2010 a clear decoupling occurred for countries with weaker fundamentals; contagion remained marginal for stronger-fundamentals countries but increased substantially for weak-fundamentals countries (Greece, Portugal, Ireland, and also Spain and Italy). - Contagion began decreasing from end-2011 after pan-European policies (EFSF effective October 2011; ESM treaty signed February 2012). - From August 2012, sovereign spreads and hence contagion decreased significantly across European countries after the OMT announcement; spreads remained differentiated according to fundamentals and some countries showed improvement by mid-2012 due to fiscal consolidation. ### Interaction of factors and implications - Factors interact and can exert a compounded effect on sovereign spreads irrespective of debt exchange policy decisions. - The magnitude of contagion appears to depend on a compound of factors, especially when: - A country transmits contingent liabilities through interconnected financial systems and/or economies. - A country belongs to a monetary union without fiscal transfers, as markets may assign a high probability that a distressed country will be supported by related countries. - Conclusion: evidence indicates contagion magnitude is influenced by multiple interacting channels (global risk aversion, contagion via VI, and fiscal fundamentals), and is not solely determined by debt exchange policy decisions. *Source: IMF staff calculations and analysis as presented in the supplied content.* ### 1. In the run-up to most crisis cases, contagion increased significantly, as evidenced by high bond spread ### 1. In the run-up to most crisis cases, contagion increased significantly, as evidenced by high bond spread correlations. ### Contagion: core findings - In the run-up to most crisis cases, contagion increased significantly, as evidenced by high bond spread correlations. - Credible policy decisions on sovereign debt resolution reduce contagion: in Greece, contagion declined once credible policy decisions on the sovereign debt restructuring were made. - In Argentina, sovereign bond spread spillovers with other EMs (Brazil, Mexico, and Russia) only started to decisively subside after the eventual default in early 2002. - For Uruguay, the magnitude of sovereign debt contagion appears much smaller in comparison to Argentina. - Overall conclusion: credibility in policy decisions is key to the solution of contagion, whether it is a bailout or debt restructuring. ### Greece: illustrative timeline and correlations - Estimated GARCH Bond Spread Correlation of Greece with Euro Area Countries — sources: Bloomberg and Fund staff calculations. - Notable dated events marked on the Greece correlation series: - Greece -IMF Program, May 9, 2010 - Deauville Declaration, October 18, 2010 - Greece -Announces terms of Debt Exchange, February 24, 2012 - Greece: IIF proposal for restructuring, July 2011 - Greece -IMF Approves Second Program, March 15, 2012 - Draghi London Speech, July 26, 2012 - OMT, September 6, 2012 - Announcement IMF Program Ireland - Announcement IMF Program Portugal - Two correlation series shown: Average of Gr with Ire, It, Por and Sp; Average of Gr with Aus, Neth and Fra. ### Argentina: illustrative timeline and correlations - Estimated GARCH Bond Spread Correlations of Argentina with Other EMs — sources: Bloomberg; and Fund staff calculations. - Three series shown: Argentina-Brazil; Argentina-Mexico; Argentina-Russia. - Notable dated events marked on the Argentina correlation series: - IMF Program, March 10,2000 - Final Exchange offer, November2001 - Default Date, January 2002 - (Also listed) Announcement of Restructuring, October 2011 ### Methodology pointers - Multivariate GARCH (generalization of the Constant Conditional Correlation (CCC) model by Bollerslev (1990)). - Followed Frank and Hesse (2009) for related approaches. - Sources cited for methodology and related literature include Bollerslev (1990), Engle (2002), Frank and Hesse (2009), Caceres, Guzzo and Segoviano (2010), Espinoza and Segoviano (2011), Cruces and Trebesch (2013), and others. ### Reprofiling and domestic financial stability: summary of annex findings - Annex IV analyzes past reprofiling cases (Cyprus, Jamaica, Pakistan, and Uruguay) to assess destabilizing effects on domestic banking systems. - Main findings: - In the reviewed Fund-supported programs entailing reprofiling, reprofilings did not have a disruptive impact on the domestic banking system. - Factors that contained effects on the banking system included: - Financial stability concerns were taken into account in the design of reprofiling and program strategy. - Banks mainly held sovereign assets as held-to-maturity (HTM); reprofilings were not assessed to be impairment events requiring write-downs (e.g., Cyprus, Jamaica). - Regulatory incentives for banks to participate in bond exchanges were provided (e.g., Jamaica, Uruguay). - Capital and liquidity support mechanisms were established or present (e.g., Jamaica, Cyprus). - Bank holdings of sovereign bonds were not very large in most cases. - Some forbearance was used. - The Jamaican case: restructuring intentionally designed to be light to minimize impact on the financial system. - Cyprus and Uruguay: reprofilings worked smoothly and in parallel with programs to address banking vulnerabilities; Cyprus included bail-in of bank creditors to close capital gaps. - Caveats: - Past reprofilings were not done under the regime being considered and evidence is not conclusive. - Temporary forbearance can contain impacts but may adversely affect credit growth. - Benign outcomes may partly reflect relatively better initial macroeconomic conditions in past reprofilings; however the text asserts initial macroeconomic conditions were very similar in past cases of reprofilings and deeper restructurings. ### Potential disruptive effects of reprofiling (listed in annex) - Banks can suffer mark-to-market losses on government bond holdings leading to undercapitalization. - Deposit runs could be triggered by concerns about domestic banks’ health. - Banks’ short-term external wholesale funding could become more expensive or evaporate. - Larger haircuts on sovereign exposure used as collateral for interbank or central bank repo operations. - Exchange rate depreciation could increase bank FX funding costs and expose unhedged FX borrowers. ### Potential beneficial effects of reprofiling - If reprofiling improves debt sustainability and helps avoid deeper restructuring, it can speed recovery, reduce debt overhang, and benefit the financial sector. - Reprofiling can remove uncertainty and lead to relief rallies and market-to-market gains on bank balance sheets, depending on bond portfolio structures. ### Measures to protect the banking system and encourage bank participation in reprofiling - Creation of backstops if stress tests indicate possible stability concerns (examples: Jamaica’s Financial Sector Stability Fund (FSSF) in 2010 and 2013; Fund for Fortifying the System of Banks in Uruguay). Fund program resources could potentially contribute to the backstop. - Increased central bank liquidity provision prior to and during debt reprofiling episodes, including: - Temporary expansion of eligible collateral if banks face large haircuts. - FX swap lines with other central banks for temporary FX funding buffers. - Regulatory measures to facilitate participation and limit impact, such as: - Applying different risk weights on old versus new bonds. - Providing liquidity support subject to eligible collateral for viable institutions. - Granting temporary forbearance (e.g., on provisioning requirements). - Capital and deposit controls in extreme cases (examples: Cyprus and Iceland), with careful implementation, clear communication, and credible exit strategies. - Bank resolution tools such as bail-in to require bank shareholders and possibly other bank creditors to contribute if banks are undercapitalized (example: Cyprus bail-in closing imminent capital hole). ### Examples and specific case notes - Cyprus: reprofiling was not assessed to be an impairment event; bonds were close to maturity and principal and coupon were preserved; bail-in of bank creditors addressed capital gaps in some banks. - Jamaica: restructuring designed to be light; regulatory incentives and a Financial Sector Stability Fund were used; capital and liquidity support mechanisms were established. - Uruguay: Fund for Fortifying the System of Banks present; reprofiling worked smoothly in parallel with program measures addressing banking vulnerabilities. - Pakistan: included in the set of reviewed reprofiling cases (Annex IV analyses past maturity extensions across Cyprus, Jamaica, Pakistan, and Uruguay). ### Indicators of fundamentals and policy track record (as presented) - The text table presents indicators with the following strings exactly as shown: - MeanMedianMeanMedian - Public Debt (percent of GDP, year before restructuring) 96.893.0100.2101.6 - Current Account Balance (percent of GDP, average over three years before restructuring) -6.4-7.7-5.7-5.5 - Overall Fiscal Balance (percent of GDP, average over three years before restructuring) -5.0-4.6-3.4-2.9 - Primary Balance (percent of GDP, average over three years before restructuring) 0.50.40.70.9 - Reprofilings listed: Belize (2007), Cyprus, Dominican Republic, Grenada, Jamaica (2010), Jamaica (2013), Pakistan, Paraguay, Uruguay - Debt reductions listed: Argentina, Belize (2013), Ecuador (2000), Ecuador (2009), Greece, Russia, Seychelles, St. Kitts & Nevis, Ukraine _The Fund’s Lending Framework and Sovereign Debt—Annexes (excerpt)._ ### 75. The alternative to reprofiling—bailing out holders of sovereign debt—would not ### The alternative to reprofiling—bailing out holders of sovereign debt—would not ### Risks to financial stability and program safeguards - Bailing out holders of sovereign debt would not necessarily eliminate risks of financial instability or the need to build in appropriate program safeguards. - Past programs show access to Fund resources did not always avoid banking crises, particularly where a restructuring was eventually needed (case example: Greece). - In Greece (first Fund-supported program started in 2010): - The banking sector was perceived as relatively sound in 2010. - With the severity of the recession, ongoing sovereign debt problems and fears about a Greece euro exit, the domestic banking system increasingly faced existential liquidity, asset quality and capital problems even before a restructuring was envisaged. - Strong program measures were taken, including on bank recapitalization and restructuring and substantial ECB liquidity support. ### Forbearance considerations for regulators - Regulators must strike the right balance in the use of forbearance. - Forbearance would typically involve allowing affected financial institutions more grace time to book any bond losses from the bond exchange than is stipulated under IFRS accounting standards. - Forbearance is in general not recommended because it could undermine market confidence if provisioning and capital holes are suspected as a result of adverse bond yield developments around and after the bond exchange. - If financial stability concerns are a serious issue following debt reprofiling, temporary relaxation of some regulations could be justified as a last resort to avoid “cliff effects” on affected banks. ### Definition and context: bailout - For the purposes of this annex, a bailout is defined as cases where creditors are paid out before an eventual restructuring. ### Findings from past maturity extensions and banking-sector impacts (Box AIV1) - Criterion used: a bond exchange had a material impact on the banking sector if, as a direct result of the bond exchange, any bank needed additional provisioning or recapitalization. - Cases reviewed: Cyprus (2013), Jamaica (2010, 2013), Pakistan (1999), Uruguay (2003). - Cyprus (2013): - Bonds exchanged weeks ahead of original due dates; maturities extended through 2019-2023. - Banks mostly held affected sovereign bonds as Hold-to-Maturity (HTM) and, with regulator tacit consent, assessed no impairment event. - Temporary SD assessment did not bind banks to book losses because sovereign did not default on payments and issued new bonds with the same face value and other terms. - Prior bail-in of bank creditors had closed the imminent capital hole of the main Cypriot banks. - Jamaica (2010, 2013): - Restructuring designed to be light to avoid destabilizing the financial system. - Most affected domestic debt held by banks as HTM; no immediate additional provisioning or capital hit. - Rating agencies ruled domestic government debt as SD then upgraded the sovereign after successful completion. - A Financial Sector Stability Fund (FSSF) was set up but not tapped. - Pakistan (1999): - Restructuring of external sovereign debt in November 1999 had limited impact on domestic banking sector. - About 30 percent of restructured bonds were held by domestic investors. - One participating domestic bank received a capital injection 6 months after the exchange following a bank audit; undercapitalization did not seem related to the bond exchange. - Uruguay (2003): - Immediate direct impact on domestic banking system was small. - More than 50 percent of bonds were held by domestic creditors, mostly retail investors. - Domestic bank holdings of government bonds were relatively low, at less than 5 percent of total bank assets, and mostly held as HTM. - Bank supervisor provided strong regulatory incentives for banks to participate in the exchange. ### Central bank liquidity provision during reprofiling (Box AIV2) - General principles: - Central banks provide liquidity to banks subject to eligible collateral and appropriate haircuts; banks need to be solvent to access central bank liquidity. - If banks lose eligible collateral due to reprofiling, they can often gain access to Emergency Liquidity Assistance (ELA) with a wider pool of collateral but more expensive and conditional. - Debt reprofiling typically leads to a Selected Default (SD) rating; affected sovereign securities cease to be eligible for regular central bank liquidity, so banks need ELA until SD rating is lifted. - Duration of SD in past successful reprofiling cases (staff findings): - 0.2 months in Cyprus (July 2013) - 0.5 months in Uruguay (May 2003) - 0.7–1.3 months in Jamaica (2010, 2013) - Pakistan as outlier at 11 months - Greece in March 2012: SD rating lasted 2¼ months - Country-specific arrangements: - Cyprus (2013): Affected banks already had only ELA from Central Bank of Cyprus, not ECB; SD duration 0.2 months. - Jamaica (2010, 2013): FSSF set up to provide liquidity and solvency support; financial institutions qualified if they exchanged at least 90 percent of old bonds; banks had access to Bank of Jamaica temporary discount window subject to liquid collateral; maximum liquidity maturity 6 months or increased regulatory intervention; no bank drew on FSSF. - Pakistan (1999): State Bank of Pakistan provided liquidity through its lender-of-last-resort discount window and regular open market operations; no take-up in immediate months after the exchange. - Uruguay (2003): Central bank allowed only new bonds as eligible collateral, encouraging participation; old bonds announced as no longer eligible for liquidity assistance. ### Accounting treatment of bank holdings of government bonds (Box AIV3) - Impact depends on accounting portfolio: - Trading portfolio: mark-to-market (MTM) affects P&L and regulatory capital; following reprofiling, banks with trading holdings could make MTM gains as yields usually fell after past maturity extensions. - Available-for-Sale (AFS) portfolio: unrealized gains and losses taken to OCI, excluded from capital calculation via a prudential filter in Basel II; impairment tests can recycle OCI to P&L and affect capital. - In practice, reprofiling has not been treated as an impairment event in cases reviewed (Cyprus, Jamaica). - Prudential filter for AFS is to be phased out under Basel III, but European CRDIV/CRR allows an exemption (CRR article 467 allows the AFS prudential filter). Some EU countries (Ireland, Italy) already allowed their banks to keep the AFS prudential filter under Basel III. - Held-to-Maturity (HTM) portfolio: not marked-to-market; subject to impairment test; impairment losses affect P&L and capital. - Reclassification and forbearance: - Reclassifications from AFS into HTM were allowed in certain crisis-era amendments (e.g., October 2008 IAS 39/IFRS 7 amendments) and used by some supervisors (Turkey, Philippines). - IAS39 paragraph 9 restricts classification as HTM after sizable sales or reclassifications in current or prior two financial years, but exceptions exist (e.g., sale/reclassification close to maturity such as three months, collection of significant principal, or an “isolated event that is beyond the entity’s control, is nonrecurring and could not have been reasonably anticipated”). - In practice there appears to be accounting scope during a reprofiling for banks not to move HTM securities to AFS and re-measure at fair value, depending on timing and assessments. - Even if HTM classification is retained, an impairment test based on objective evidence could indicate provisioning is required if bond prices are well below par. - Conclusion: A reprofiling will not per se require a reclassification. ### Greece banking-sector developments during the crisis (Box AIV4) - Key developments: - Bank capital ratio was 11.7 percent when the Stand-by Arrangement program began in May 2010 (aided by a recapitalization in 2009). - Balance sheets came under pressure from higher nonperforming loans (NPLs) as the economy weakened. - Liquidity conditions tightened in 2009 due to banks losing wholesale market access and some deposit outflows. - By 2011, deleveraging and restructuring of state-owned banks was perceived as necessary; ATE (largest state-owned bank) failed Europe-wide stress tests in mid-2010 and had to be recapitalized. - Sizable deposit outflows began in the second half of 2011, fanned by fears of a Greek euro exit. - ECB support: - From May 2010, the ECB suspended the link between sovereign credit ratings and eligibility of collateral for refinancing operations and intervened directly in government bond markets under the Securities’ Market Program (SMP). - ECB began accepting uncovered bank bonds guaranteed by the government as collateral eligible for refinancing. - Impact of the debt restructuring (PSI) and remedial measures: - Greek banks were heavily exposed to the sovereign, holding government bonds with a book value of about €40 billion (after some initial June 2011 impairments). - Core capital was €22 billion, about the same magnitude as the capital needs arising from the PSI. - Only €1.5 billion was drawn from the Hellenic Financial Stabilization Fund (HFSF) during the SBA-supported program, but banks’ subsequent capital needs dwarfed the HFSF provision. - As of the fourth review, the purpose of the HFSF changed to providing a substantial injection of public funds for banks severely affected by the PSI and the deep recession: the amount needed for the HFSF in the context of the EFF was estimated at €50 billion. *Source: Excerpt from “The Fund’s Lending Framework and Sovereign Debt—Annexes” (selected boxes and paragraphs).* ### References ### _052214a - References ### Annex purpose and framing - Examines potential gains from slower fiscal consolidation in debt-distress scenarios, focusing on debt reprofiling as an alternative to bailing out creditors and its role in allowing a more gradual pace of fiscal consolidation. - Key conceptual points: - Fiscal multipliers are likely to be higher during recessions and periods of negative output gaps. - Slower consolidation can postpone adjustment to a period when multipliers are lower, potentially increasing cumulative GDP. - Hysteresis effects (temporary underutilization of resources causing permanent reductions in potential output) magnify the gains from smoothing consolidation. - Non-output costs of sharp consolidation (political/social unrest, program ownership issues, disruptions in public services) are acknowledged though not fully captured in the model. ### Model (Section A) — structure and equations (qualitative description) - Output dynamics: - Actual output deviates from potential GDP (P_t) through an output gap Γ_t. - Persistence of the output gap is governed by ρ. - Fiscal multiplier μ_t varies over time and depends on the output gap. - Structural primary balance is S_t; demand shocks are ε_t and used to match the baseline. - Potential GDP dynamics: - Potential GDP evolves with gross growth G_{t+1} net of hysteresis h (h captures strength of hysteresis). - Fiscal and debt dynamics: - Primary fiscal balance responds to the output gap with elasticity τ. - Debt stock evolves with real interest rate R. ### Calibration (Section B) — baseline and parameter values - Calibration approach: - Based on six years of actual and projected data for ten exceptional access cases between 2009 and 2013, starting with the onset of distress, the program request, and five years after the outbreak of the crisis. - The ten countries are: Armenia, Belarus, Greece, Ireland, Jordan, Portugal, Romania, Sri Lanka, St. Kitts and Nevis, and Ukraine. - Time t0 is the year of program request; the last year of actual data is 2013. - After t4, model dynamics apply; from t5 onwards the structural assumptions include stabilization and long-run settings. - Specific calibrated assumptions stated in text: - Starting at t5, the primary structural balance stabilizes at 1.5 percent of potential GDP. - Potential growth and the real interest rate are, respectively, set to 2 percent and 2.5 percent (text). - Table AV2 calibrated parameters (as reported): - Persistence of the output gap: 0.5 - Growth of potential GDP from t5 onwards: 1.02 - Hysteresis in the output gap: 0.1 - Real interest rate from t5 onwards: 1.025 - Elasticity of primary balance with respect to output gap: 0.48 - Intertemporal discount factor: 1/R - Fiscal multiplier specification: - Fiscal multiplier varies nonlinearly with the output gap, ranging between 0.25 and 1.5. - Assumption: multipliers larger in recessions (negative output gaps) than in expansions. ### Results (Section C) — comparative simulations and quantitative findings - Comparative setup: - Baseline: frontloaded fiscal consolidation (sharp initial tightening of structural primary balance). - Alternative: more gradual fiscal adjustment (moderate early tightening, compensated by additional medium-term consolidation) such that debt returns to baseline path. - Debt outcomes: - Both consolidation paths bring debt down to 75 percent of GDP in fifteen years. - A more gradual pace of consolidation involves a temporary increase in the debt-to-GDP ratio; this increase can persist beyond the 2–3 year horizon under reprofiling and remains limited, less than 10 percentage points of GDP. - GDP and output-gap outcomes: - Delaying consolidation leads to a temporary increase in GDP of more than 0.5 percent during the first 5 years. - Over the first 15 years, the cumulative effect remains positive at 0.15 percent (present discounted value of GDP relative to baseline). - Gradual consolidation preserves higher GDP growth and produces a smaller negative output gap in the early years, reducing contemporaneous fiscal multipliers and supporting GDP in a virtuous circle; later required additional consolidation can trigger larger negative output gaps and moderately higher multipliers (a countervailing vicious cycle). - Potential GDP and hysteresis: - After 15 years, potential GDP is 0.1 percent higher than under the baseline; this is characterized as a permanent gain driven by reduced negative output gaps and associated hysteresis effects. ### Robustness, caveats, and conditionalities - Results are consistent with existing literature that emphasizes time-varying fiscal multipliers and hysteresis (cited studies and working papers noted in the text). - The case for delayed adjustment is strongest when: - Fiscal multipliers are larger during crises, and - Hysteresis effects are present. - Important caveats: - If a crisis occurs during a period of overheating (positive output gaps), delaying consolidation may be counterproductive. - Successful delayed consolidation requires the country’s ability to commit to larger medium-term adjustment; feasibility depends on country-specific characteristics. - The analysis focuses on the benefits of reprofiling insofar as it allows slower consolidation; implications for financial markets, contagion, and other creditor effects are analyzed in other annexes. - Whether short reprofiling (pushing maturities out by only 1 to 3 years) generates sufficient financing space for delayed consolidation depends on the maturity structure of liabilities. ### Key references cited within the annex (selected) - Auerbach, Alan, and Yuriy Gorodnichenko, 2012, “Fiscal Multipliers in Recession and Expansion,” in Fiscal Policy after the Financial Crisis. - Batini, Nicoletta, Giovanni Callegari, and Giovanni Melina, 2012, “Successful Austerity in the United States, Europe, and Japan,” IMF Working Paper No. 12/190. - Bi, Ran, Haonan Qu, and James Roaf, 2013, “Assessing the Impact and Phasing of Multi-year Fiscal Adjustment: A General Framework,” IMF Working Paper No. 12/182. - Blanchard, Olivier, and Daniel Leigh, 2013, “Growth Forecast Errors and Fiscal Multipliers,” IMF Working Paper No. 13/1. - Christiano, Lawrence, Martin Eichenbaum, and Sergio Rebelo, 2011, “When is the Government Spending Multiplier Large?” Journal of Political Economy 119(1), pp. 78–121. - Delong, Bradford, and Lawrence Summers, 2012, “Fiscal Policy in a Depressed Economy,” Brookings Papers on Economic Activity, Spring 2012: pp. 233–97. - Woodford, Michael, 2011, “Simple Analytics of the Government Expenditure Multiplier,” American Economic Journal: Macroeconomics 3(1), pp. 1–35. *Source: THE FUND’S LENDING FRAMEWORK AND SOVEREIGN DEBT—ANNEXES (extracted content from the provided PDF).* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_052214a.pdf_