## _080511 ## Source details **Canonical URL:** [_080511](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2011/_080511.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2011/_080511.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2011/_080511.pdf.json) --- ### Executive summary — modernization objective and headline proposals - Modernize the framework for fiscal policy and public debt sustainability analysis (DSA) in light of the recent crisis and rising sustainability concerns in some advanced economies. - Propose moving to a risk-based approach to DSAs for all market-access countries where: - The depth and extent of analysis are commensurate with concerns regarding sustainability. - A reasonable level of standardization is maintained. - Key areas of greater focus: - Realism of baseline assumptions: primary fiscal balance, interest rate, and growth rate assumptions subject to close scrutiny using country-specific information and cross-country experience. - Level of public debt as trigger: trend and level of the debt-to-GDP ratio are key indicators; use a reference point of 60 percent of GDP flexibly to trigger deeper analysis for market-access countries. - Analysis of fiscal risks: stress testing tailored to country-specific risks, including contingent liabilities from the financial sector; develop full-fledged alternative scenarios and more regular use of fan charts. - Debt profile vulnerabilities: integrate assessment of debt structure and liquidity issues; propose indicative benchmarks to facilitate staff analysis. - Coverage: broaden coverage to include entities presenting significant fiscal risks, including state owned enterprises, public–private partnerships, and pension and health care programs. - Implementation: - Directors’ views to inform specific guidance in coming months; public DSA template to be revised. ### I. Motivation and scope - Rising public debt in advanced economies (AEs) driven by: - sharp deterioration of fiscal balances during the crisis; - government intervention in the banking sector in some cases; - long-term spending pressures related to population aging. - Objective: highlight areas for improvement and make proposals, mobilizing new and existing tools to address interconnections across sectors. ### II. Conceptual framework and prioritized improvements - DSA essentials: - Debt-to-GDP trajectory under baseline and alternative scenarios. - Whether debt ratio stabilizes at a level consistent with low rollover risk and preserving growth. - Realism of underlying assumptions and debt composition. - Priorities: - Realism of baseline assumptions. - Emphasis on level and trend of debt-to-GDP; deeper analysis when debt exceeds reference point. - Tailored fiscal risk stress testing, including contingent liabilities. - Integrate refinancing and liquidity risks into DSA. - Broaden coverage to off-budget entities and long-term spending pressures. - Risk-based approach principles: - Tailor depth of analysis to: level of public debt; need/size of fiscal adjustment; extent of fiscal risks; liquidity issues; operational requirements (e.g., exceptional access cases). - Maintain minimum standardization for comparability. ### III. Realism of baseline projections — key findings and evidence - First step: derive projected debt-to-GDP path; realism of primary fiscal path, growth, and interest rate projections is crucial. - Example estimate: adjustment in primary balances required for AEs to bring debt ratios to or below 60 percent of GDP is estimated, on average, at about 8 percentage points of GDP between 2010 and 2020. - Historical evidence on large sustained primary surpluses (sample of 87 countries): - Over 40 percent had a maximum primary surplus exceeding 5 percent of GDP in at least one year. - Only 16 countries (less than 20 percent) sustained surpluses exceeding 5 percent of GDP for five years or longer. - Of the 16 sustained cases, five linked to exogenous factors (Botswana, Chile, Egypt, Uzbekistan, Lesotho). - Episodes of sustained large surpluses without such factors limited to 11 countries (13 percent of sample). - There were 30 instances where countries improved their five-year average primary balance by at least 5 percentage points of GDP relative to previous three-year average; in only 9 episodes did ensuing five-year primary balance equal or exceed 5 percent of GDP. - Growth and interest-rate evidence: - Interest rate-growth differential (IMF, 2011): G-20 AEs average 1 percent; EMs -4 percent; LICs -8 percent. - Forecast bias evidence: WEO real GDP growth forecasts tended to exceed outcomes (Timmermann, 2006); national plans in some AEs were more optimistic than WEO/Consensus (Bornhorst et al, 2010). - October 2010 WEO: a 1 percent of GDP fiscal consolidation typically reduces GDP growth by ½ percent within two years. - Recommendation: scrutinize growth and interest-rate assumptions, especially when planned adjustments are near cross-country tails (e.g., sustained surplus ≈ 5 percent of GDP). ### IV. Role of the debt level in the DSA - Both trend and level of debt-to-GDP matter; definitions of “high”/“low” are country-specific. - Institutional anchors: many countries use debt ceilings; commonly used ceiling is 60 percent of GDP. - Empirical ranges from cross-study medians: - Long-run debt level: median estimates range from 50 to 75 percent of GDP for AEs, and 25 percent of GDP for EM (one estimate). - Maximum sustainable debt: median estimates range from 80 to 192 percent of GDP for AEs, and 35 to 77 percent for EMs. - Staff re-estimation for EM sample (1993–2009): - Long-run debt level range: 49–58 percent. - Maximum sustainable debt level range: 63–78 percent. - Policy guidance: - Use 60 percent-of-GDP as a flexible reference to flag when deeper analysis is warranted. - 60 percent is not a definitive distress threshold; country-specific vulnerabilities may warrant attention below 60 percent. ### V. Improving the analysis of fiscal risks — methods and limitations - Current practice relies on standardized sensitivity analysis and stress tests; standardized shocks (e.g., 10 percent-of-GDP contingent liability shock) are uniform across countries. - Limitations: - Risks and realized impacts vary considerably across countries. - Standardized contingent liability shock often understates public debt impact of banking crises. - Proposed emphasis: - Place greater emphasis on contingent liabilities and country-specific fiscal risks. - Use cross-country and country-specific tools to identify risks and calibrate shocks. - Tools and approaches: - Contingent-liability analysis (financial and non-financial). - Use FSAP, FSSA, BSA, and CCA where data permit. - Tailored bound tests, alternative scenarios, and stochastic simulations (fan charts). - Incorporate historical co-movements across variables (growth, inflation, exchange rate) when calibrating shocks. ### Financial-sector contingent liabilities — empirical magnitudes - Systemic banking crises fiscal impact (Laeven and Valencia): - Median overall increase in public debt (direct + indirect effects) close to 20 percent of GDP. - Summary statistics (Medians, percent of GDP) across crisis samples (old: 1970-2006; new: 2007-2009; all): - Old crises (1970-2006) — Advanced economies: 3.7, 36.2, 32.9; Emerging markets: 11.5, 12.7, 29.4; All: 10.0, 16.3, 19.5. - New crises (2007-2009) — Advanced economies: 5.9, 25.1, 24.8; Other economies: 4.8, 23.9, 4.7; All: 4.9, 23.9, 24.5. - Current DSA contingent-liability shock (10 percent of GDP) often below median historical impacts; half of historical banking-crisis episodes had public-debt increases exceeding 18 percent of GDP. ### Non-financial contingent liabilities and natural disasters — examples - SOE/PPP exposures: - Chile: revenue guarantees to concessionaires estimated ~4 percent of GDP exposure. - Portugal: guarantees of SOE debt ~7 percent of GDP. - Natural disasters: - Hurricane Ivan (2004): damage ~ twice Grenada’s GDP. - New Zealand earthquakes: central government cost ~4 percent of GDP. ### Identifying country-specific shocks and indicators - Use CGER/other exchange rate assessments to inform exchange rate shock calibration; where overvaluation is found, expect exchange rate shock at least the maximum estimated magnitude. - Vulnerability exercises provide sectoral risk ratings (low/medium/high); medium/high warrants further investigation. - Private-credit indicator: - Domestic private sector credit-to-GDP ratio above 70 percent provides an early warning signal of sovereign debt distress (caveat: may reflect financial deepening). ### Using financial sector assessments in DSAs - FSAP/FSSA: - Use estimates of capital shortfalls to calibrate bound tests. - Use FSSA RAM: medium/high-impact banking risks → contingent liability shock equal to half the impact on capital base (further calibrated). - Joint stress tests of financial sector and public debt recommended where risks are high. - Practical limits: stress test reporting not standardized; confidentiality concerns may limit public DSA use. - BSA and CCA: - BSA identifies net positions and vulnerabilities across sectors; useful where external DSA points to high private external debt. - CCA can derive forward-looking indicators of potential fiscal costs where data exist. ### Assessing shock impacts — bound tests, scenarios, stochastic simulations - Limitations of current bound tests: - Mechanistic, often shocking one variable at a time; historical shocks are correlated. - Current macro stress test includes a 30 percent exchange rate depreciation (tail risk). - Empirical regularities to incorporate: - In half of episodes with negative real GDP growth over five years, public debt increased by more than 14 percent of GDP. - In half of episodes with terms-of-trade deterioration >25 percent in one year, public debt increased by more than 11 percent of GDP. - Nominal exchange rate depreciations >30 percent in one year tended to coincide with significant inflation increases and sharp real GDP declines. - Tail-risk guidance: - Use banking sector credit indicators and other financial indicators to decide on tail-risk scenarios. - Frame stress testing contingent liabilities as analytical exercises for policy and contingency planning. ### Stochastic simulations — illustration and guidance - Advantages: - Generate country-specific confidence intervals accounting for interactions among key variables. - Persistence of shocks determined by empirical model dynamics. - VAR-based example (1995–2010 estimation; projections 2011–2015) for a market-access country: - Baseline: public debt declines to 75 percent by 2015. - Most extreme current stress test: public debt increases to 92 percent by 2015 (17 percentage points above baseline). - Stochastic-simulation probabilities: 25 percent chance debt exceeds 80 percent of GDP by 2015 (75th percentile); 10 percent chance debt exceeds 85 percent by 2015 (90th percentile). - Implication: extreme stress-test growth shock not likely based on recent historical experience in this example. - Caveats: - Confidence intervals sensitive to model specification and sample period (AR vs VAR differences; sample 1999–2010 narrower than 1995–2010 in example). - Less useful when structural shifts occur. - Procedural proposals: - Develop guidelines for statistical foundations and comparability. - Consider centrally generating confidence intervals where data available; allow country teams to tailor models. - Where stochastic simulations infeasible, develop well-specified alternative scenarios. - Data scope: quarterly VARs available for ~40 countries (majority AEs); fan charts used for Greece, UK, Germany, US; DSAs with fan charts: Morocco, Mauritius, El Salvador, Indonesia, Israel, Costa Rica. ### Debt profile vulnerabilities and indicative benchmarks - Vulnerabilities arise from maturity, currency composition, and creditor base. - High short-term debt share increases rollover and interest-rate risk. - High foreign-currency debt share increases exchange-rate risk and reserve pressure. - Creditor-base composition affects rollover risk. - Proposal: add six indicators to public DSA with benchmark levels (flexible application for MACs where data available); benchmarks facilitate staff analysis without rating countries. - Indicative benchmarks (preserved entries from source): - EMBI Gl obal spre ads ( Basi s Poi nts)1,17526< - Five-Year CDS spreads (Basis Points)81235< - External Financing Requirement (Percent of GDP)3141< - Public Debt in Foreign Currency (Percent of total)6875< - Short-term Public Debt at Original Maturity (Percent of total)1476< - Average Maturity of Debt to Private Sector (Years)697> ### Coverage of fiscal balance and public debt — expansion and long-term pressures - Use GFSM 2001/general government concept where possible; include sub-national and public corporations where they impose large fiscal risks. - Reclassifications (SOEs/PPPs) can raise reported debt (example: Portugal reclassifications raised reported debt by 10 percentage points of GDP in 2010). - Long-term spending pressures from aging: - In AEs, combined annual spending on old-age pensions and health care expected to increase by an average of 4 percent of GDP by 2030. - In EMs, projected increase in pension and health care spending is 2 percent of GDP on average over next two decades, with additional pressures beyond 2030. - DSA implications: - Five-year projection horizon constrains inclusion of long-term impacts. - Options: expand projection horizon where pressures materialize quickly (mainly AEs) or add memo items (e.g., present value of pension and health care costs as percent of current GDP). ### Gross versus net debt — definitions and policy guidance - Gross debt: liabilities in form of debt instruments (principal and interest obligations). - Net debt variants: - Debt net of liquid assets: gross debt less highly liquid assets (e.g., government deposits). - Net financial liabilities: total financial liabilities minus total financial assets (equivalent to net financial worth with reversed signs per GFSM 2001). - Tradeoffs: - Gross debt: better for cross-country comparability and captures rollover risk. - Net debt: complements gross debt where reliable asset data available; limits comparability across countries due to asset reporting differences. - Recommendations: - Use net debt to complement gross debt where reliable asset data exist. - Encourage AEs to report net debt and net financial liabilities systematically. - For LICs/EMs, pursue narrower net debt definitions if data permit; if not, consider debt net of liquid assets. - When net debt used, strip yields on assets from primary balance definition. ### Risk-based approach to DSAs — implementation summary - Move to risk-based DSAs, maintaining limited standardization: - Provide a limited set of debt indicators; baseline scenario; historical and “no policy change” comparisons; a few bound tests. - Depth of DSA evaluated against 60 percent-of-GDP reference and country-specific factors (debt structure, fiscal risks). - Always assess fiscal risks (currency valuation, financial sector developments, off-balance-sheet liabilities). - Where debt is high or fiscal risks significant, expect customized bound tests and alternative scenarios; discuss fiscal risk mitigation strategies. - Encourage stochastic simulations and consider centralized production of confidence intervals. - Resource implications: - Risk-based approach reconciles need for deeper DSAs with staff resource constraints. - Customized shocks/alternative scenarios have initial setup costs but limited year-to-year change if circumstances stable. - More specific operational guidance to be developed. ### Illustrative DSA implementations (Country A and Country B) - Country A (detailed DSA): - Public debt rose from ~50 percent in 2000 to ~100 percent in 2010; projected to stabilize ~120 percent medium term (well above 60 percent). - Baseline assumes primary balance improves from deficit ~6 percent of GDP in 2010 to surplus ~2 percent of GDP medium term; realism questioned. - Fiscal risks: SOE guarantees ~20 percent of GDP (about half explicitly guaranteed), PPPs ~20 percent of GDP with bailout risk, banking sector vulnerabilities. - DSA could include contingent-liability shock of 15 percent of GDP and larger shocks based on stress tests; full alternative lower-growth scenario recommended. - Mitigation: SOE privatization, PPP re-assessment, financial-sector liquidity policies, structural reforms. - Country B (lighter DSA): - General government gross debt declined from peak 60 percent in 2006 to ~50 percent in 2010; sovereign wealth fund assets increased net asset position from ~140 percent of GDP to ~160 percent. - Gross debt close to 60 percent but vulnerabilities limited; net asset position expected to increase medium term. - Standard bound tests sufficient; customized scenario for oil-price decline possible; fan chart may not be necessary. ### Issues for discussion posed to Directors - Agreement on need to modernize DSA framework? - Agreement on broad areas for improvement (realism of baseline, fiscal risk analysis, integrating debt profile)? - Support for reference levels to trigger additional discussion and use of existing tools (vulnerability exercises, FSAP) to quantify risks? - Support to move to a risk-based approach for all market-access countries with commensurate depth and reasonable standardization? *Source: _080511 - Executive Summary* ### Executive Summary....................................................................................................... ### _080511 - Executive Summary ### Executive summary — modernization objective and headline proposals - Modernizing the framework for fiscal policy and public debt sustainability analysis (DSA) is necessary in light of the recent crisis and rising sustainability concerns in some advanced economies. - The paper highlights areas for improvement and makes general and specific proposals to achieve improvements while recognizing the inherent challenges of DSA. - Proposes moving to a risk-based approach to DSAs for all market-access countries, where: - The depth and extent of analysis would be commensurate with concerns regarding sustainability. - A reasonable level of standardization would be maintained. - Key areas of greater focus proposed: - Realism of baseline assumptions: close scrutiny of primary fiscal balance, interest rate, and growth rate assumptions using country-specific information and cross-country experience. - Level of public debt as a trigger for further analysis: the trend and the level of the debt-to-GDP ratio are key indicators; suggests using a reference point of 60 percent of GDP flexibly to trigger deeper analysis for market-access countries. - Analysis of fiscal risks: stress testing tailored to country-specific risks, including contingent liabilities from the financial sector; develop full-fledged alternative scenarios and more regular use of fan charts. - Vulnerabilities associated with the debt profile: integrate assessment of debt structure and liquidity issues into the DSA; propose indicative benchmarks to facilitate staff analysis. - Coverage of fiscal balance and public debt: broaden coverage, with attention to entities that present significant fiscal risks, including state owned enterprises, public-private partnerships, and pension and health care programs. - Implementation: - Based on Directors’ views, specific guidance would be developed in the coming months to render proposals operational, facilitate country team work, and ensure adequate implementation. - The public DSA template would be revised accordingly. ### I. Introduction — motivation and scope - Large increases in public debt in advanced economies (AEs) have made fiscal policy and public debt sustainability central to policy discussions. - Recent worsening of the debt outlook in AEs reflects factors including: - sharp deterioration of fiscal balances during the crisis; - in some cases, government intervention in the banking sector; - long-term spending pressures related to population aging. - Before the crisis, Fund analysis did not always pay sufficient attention to public debt sustainability in market access countries (MACs), particularly in AEs; DSAs had often become routine with mechanical implementation and limited linkage to macroeconomic and financial policy discussion. - The paper’s objective: highlight areas for improvement and make proposals — including new tools and existing tools mobilized to address interconnections across sectors. - Organization of the paper: conceptual framework (Section II); realism of projections (Section III); role of debt level (Section IV); fiscal risks (Section V); debt profile vulnerabilities (Section VI); coverage issues (Section VII); implementation (Section VIII); issues for discussion (Section IX). ### II. Conceptual framework and prioritized areas for improvement - Conceptual framing: - Fiscal policy is unsustainable if, absent adjustment, the government would not be able to service its debt; DSA starts with a baseline trajectory for public debt and tests baseline assumptions and the effects of risks. - A proper assessment requires: (i) debt-to-GDP trajectory under baseline and alternative scenarios; (ii) whether the debt ratio stabilizes at a level consistent with an acceptably low rollover risk and with preserving growth; (iii) realism of underlying assumptions; (iv) debt composition. - Identified priorities for improvement: - Realism of baseline assumptions: subject primary fiscal balance path and other assumptions to greater scrutiny, particularly where significant fiscal adjustment is projected. - Level of public debt: emphasize the level and trend of the debt-to-GDP ratio as a key indicator; recommend more stringent analysis of vulnerabilities when the debt ratio exceeds a certain level. - Analysis of fiscal risks: better tailor stress testing to country-specific circumstances; improve identification of relevant risks, including contingent liabilities from the financial sector; customize bounds tests and alternative scenarios. - Debt profile vulnerabilities: integrate analysis of refinancing and liquidity risks into the DSA rather than treating them outside the framework. - Coverage of fiscal balance and public debt: broaden coverage to include off-budget entities, public–private partnerships, and long-term spending pressures from population aging. - Risk-based approach recommendation: - Adopt a risk-based approach to DSA in an environment of tight resource constraints. - Maintain minimum standardization for discipline and comparability, but tailor depth of analysis to: - level of public debt; - need for and size of fiscal adjustment; - extent of fiscal risks; - liquidity issues; - operational requirements such as need for “a rigorous and systematic analysis” for cases of exceptional access to Fund resources. ### Box 1 (Conceptual framework) — key conceptual distinctions and implications - Two conceptually distinct but related cases: - Case 1: Current primary balance insufficient to stabilize debt-to-GDP ratio, but a realistic fiscal adjustment (economically and politically) could achieve the necessary primary balance — fiscal policy currently unsustainable but public debt can be regarded as sustainable. - Case 2: Primary balance needed to stabilize debt ratio is politically/economically infeasible — both fiscal policy and public debt are unsustainable (solvency problem) and debt restructuring would be necessary. - Reasons high debt levels increase risk and warrant attention: - Higher initial debt magnifies the impact of interest-rate increases or growth declines on the primary surplus needed to stabilize debt. - Rollover crisis risk depends on borrowing requirements (linked to the deficit and interest bill), debt composition (e.g., short maturities), and investor base (e.g., high externally-held debt). - Beyond certain levels, higher debt may be associated with lower long-term economic growth. - Necessity of realistic macroeconomic baseline assumptions: economic growth, interest rates on public debt, and likelihood of fiscal risks materializing are critical inputs; these assumptions must be stress tested. - Acknowledges that for some countries (e.g., those with substantial infrastructure needs), increases in debt that finance public investment could positively affect long-term growth. ### Implementation and next steps (high-level) - Directors’ views will inform specific guidance to: - render proposals fully operational; - facilitate country team work; - ensure adequate implementation. - Public DSA template to be revised to reflect the proposed framework. *Source: _080511 - Executive Summary* ### 8. While the main focus of this paper is on public DSA for MACs, many of the ### _080511 - 8. While the main focus of this paper is on public DSA for MACs, many of the ### III. REALISM OF BASELINE PROJECTIONS - The first step in the DSA is to derive the projected path of the debt-to-GDP ratio; realism of underlying assumptions on the primary fiscal path, economic growth, and interest rate projections is crucial, especially where sustainability depends on large fiscal adjustments. - Example estimate: the adjustment in primary balances required for AEs to bring debt ratios to or below 60 percent of GDP is estimated, on average, at about 8 percentage points of GDP between 2010 and 2020 (Abbas et al, 2010). - Fiscal consolidations of the magnitude above have been achieved in several past cases but pose major challenges; the potential contractionary impact of fiscal consolidation on growth needs to be accounted for (note: October 2010 WEO found that a 1 percent of GDP fiscal consolidation typically reduces GDP growth by ½ percent within two years). ### A. Realistic Primary Fiscal Balance Paths - Cross-country evidence on large and sustained primary surpluses: - Out of 87 countries sampled, over 40 percent had a maximum primary surplus exceeding 5 percent of GDP in at least one year; only 16 countries (less than 20 percent) sustained surpluses exceeding 5 percent of GDP for five years or longer. - Of the 16 countries with sustained surpluses, five had this performance linked to exogenous factors: large increase in revenue related to natural resources (Botswana, Chile, Egypt, and Uzbekistan) or transfers arising from customs union membership (Lesotho). - Episodes of sustained large surpluses without facilitating exogenous factors were limited to 11 countries (13 percent of the sample); a few ran large primary surpluses without a large debt burden (Denmark, New Zealand, Turkey), but the majority adjusted when facing debt levels above 60 percent of GDP (Belgium, Canada, Dominica, Israel, Jamaica, Panama, Seychelles, and Singapore). - There were 30 instances where countries improved their five-year average primary balance by at least 5 percentage points of GDP relative to the average of the previous three years; larger improvements correlated with weaker starting fiscal positions; in only 9 episodes did the ensuing five-year primary balance equal or exceed 5 percent of GDP. - Assessing realism of planned large fiscal adjustments: - If planned adjustments are near the right-hand tail of the cross-country distribution (e.g., sustained surplus around 5 percent of GDP or more), particular scrutiny is warranted. - Country-specific information to consider includes: past record of fiscal adjustment, extent of political commitment, implementation of supporting policy measures, design of fiscal adjustment plans, existing legal and institutional mechanisms, and experience regarding budget forecast errors. - Comparative scenario analysis: - Baseline vs. “no policy change” and historical scenarios can help judge whether policy actions are sufficient to credibly break from past/current trends. ### B. Realism of Economic Growth and Interest Rate Assumptions - The interest rate-growth differential is critical for DSA; it has been a key benign force in EMs and LICs when strongly negative, while a generally positive differential in AEs has been unfavorable requiring primary surpluses for debt stabilization. - IMF (2011) summary: interest rate-growth differential in G-20 AEs has been on average 1 percent, while this differential has been negative in EMs (-4 percent) and LICs (-8 percent). - Empirical findings on forecast biases and differences: - Timmermann (2006) found WEO real GDP growth forecasts tended to systematically exceed outcomes, especially in next-year forecasts and for countries with an IMF-supported program. - Bornhorst et al (2010): differences in growth forecasts between WEO and country authorities in post-crisis medium-term fiscal adjustment plans (2010); in AEs with large adjustment needs, national plans were somewhat more optimistic than WEO or Consensus Forecast projections; in EMs, growth assumptions were largely in line with WEO and Consensus Forecast projections. - IMF (2011): shocks—especially to economic growth—often derailed fiscal adjustment among 20 episodes in G-7 countries; some plans derailed almost immediately by unexpected downturns (e.g., Germany in the 1970s, Japan); some successes were facilitated by higher-than-expected growth and asset prices (e.g., United States in the 1990s). - Differences between interest rate projections in WEO and those by authorities tend to be higher in EMs; projections are more aligned for AEs. - Recommendation: scrutinize growth and interest rate assumptions, especially when substantial fiscal adjustment is considered; significant deviations from historical trends or market forecasts should be fully justified. ### IV. ROLE OF THE DEBT LEVEL IN THE DSA - Both the trend and the level of the debt-to-GDP ratio matter; an increasing debt ratio from a “low” initial level may entail less risk than a stable but “high” ratio. Definitions of “high” and “low” are country-specific. - Common institutional anchors and observed practice: - Many countries use debt ceilings in fiscal responsibility laws or regional agreements; a commonly used ceiling is 60 percent of GDP. - Table excerpt: Economic and Monetary Union of the EU member ceiling 60 actual 85; Eastern Caribbean Currency Union member ceiling 60 actual 103; individual countries examples: Pakistan 60/59, Panama 40/40, United Kingdom 60/77 (actual at end-2009 for currency unions is aggregated). - Challenges of high debt levels (Box 1 summary): - Large primary fiscal surpluses needed to service high debt may be difficult to sustain economically and politically. - High debt exacerbates vulnerability to interest rate and growth shocks. - High debt is associated with higher borrowing requirements and greater rollover risk. - High debt may be detrimental to economic growth, which in turn affects long-term debt dynamics. - Empirical difficulty and ranges for sustainable debt thresholds: - Two concepts studied: “long-run debt level” and “maximum sustainable debt level.” - Cross-study median estimates vary: - Long-run debt level: median estimates from previous studies range from 50 to 75 percent of GDP for AEs, and 25 percent of GDP for EM (one available estimate). - Maximum sustainable debt level: median estimates range from 80 to 192 percent of GDP for AEs, and 35 to 77 percent of GDP for EMs. - Staff re-estimation for EM sample (1993–2009): - Long-run debt level range: 49–58 percent. - Maximum sustainable debt level range: 63–78 percent (Annex III). - These estimates, particularly long-run levels, are higher than earlier ones reflecting improved fiscal performance of EMs; estimated long-run debt levels for EMs are now closer to recent estimates for AEs based on Ostry et al (2010). - Policy approach for DSA use of a reference threshold: - Empirical dispersion precludes formal sustainability thresholds in MAC DSAs, but a 60 percent-of-GDP reference can be used to flag when deeper analysis is warranted. - When public debt exceeds or is projected to exceed 60 percent of GDP for a substantial part of the projection horizon, a detailed discussion of potential risks to sustainability arising from high debt levels would normally be expected. - The 60 percent reference should not be construed as a definitive distress threshold; it is an indication that more analysis is needed and country-specific vulnerabilities may warrant detailed analysis even below 60 percent of GDP. ### V. IMPROVING THE ANALYSIS OF FISCAL RISKS - Current practice: - Public DSA framework assesses risks around the baseline mainly through standardized sensitivity analysis. - Stress tests apply standardized macroeconomic shocks to illustrate impacts on the debt path; macro shocks are generally reported in a standardized way across countries. - Limitations and examples of realized risks: - Risks vary considerably across countries; realized events that materially raised public debt include large exchange rate depreciations, systemic banking crises, and bailouts of off-budget entities. - Analysis of contingent liabilities in DSAs has been typically too succinct and uniform, contrasting with their actual impact on public debt. - The related stress test assumes the same 10-percent-of-GDP shock for all countries, irrespective of size and risk of contingent liabilities. - Proposed emphasis: - Place greater emphasis on contingent liabilities and improve the analysis of country-specific fiscal risks. - Use a range of tools to improve identification of such risks and enrich analysis of shock impacts on the debt outlook. ### A. Contingent Liabilities - The section begins to propose approaches to better identify and analyze contingent liabilities as a key area for improving DSA fiscal risk assessment. *Italic: Source: _080511 - 8. While the main focus of this paper is on public DSA for MACs, many of the (PDF chapter/section).* ### 24. Strengthening the analysis of contingent liabilities is critical given the scope and ### 24. Strengthening the analysis of contingent liabilities is critical given the scope and magnitude of off-budget risk materialization ### Financial-sector contingent liabilities and past fiscal costs - Assistance to the financial sector has been particularly costly; systemic banking crises have been frequent and have often carried heavy fiscal costs. - The median overall increase in public debt, which reflects both direct and indirect effects of banking crises, is close to 20 percent of GDP. - Summary statistics from Laeven and Valencia (1970–2009) (Medians, percent of GDP): - Old crises (1970-2006) - Advanced economies: 3.7, 36.2, 32.9 - Emerging markets: 11.5, 12.7, 29.4 - All: 10.0, 16.3, 19.5 - New crises (2007-2009) - Advanced economies: 5.9, 25.1, 24.8 - Other economies: 4.8, 23.9, 4.7 - All: 4.9, 23.9, 24.5 - Notes on definitions: - Direct fiscal costs: fiscal outlays committed to the financial sector from the start of the crisis (t) up to t+5 (or up to end-2009 for recent crises). - Increase in public debt: difference between pre and post-crisis debt projections measured over [T-1, T+3], where T is the starting year of the crisis. - Output losses: deviations of actual GDP from its trend over a period of three years from the start of the crisis. - A standardized contingent liability shock in the current DSA template often understates the public debt impact of banking crises; large cross-country variation suggests a standardized shock is unlikely to capture crisis impacts meaningfully. ### Non-financial contingent liabilities (SOEs, PPPs, natural disasters) - Guarantees for SOEs and PPPs can pose sizable risks: - Chile: revenue guarantees to airport and toll-road concessionaires estimated to create government exposure of about 4 percent of GDP. - Portugal: guarantees of SOEs debt amount to about 7 percent of GDP. - Natural disasters can impose substantial demands on governments: - Hurricane Ivan (2004) inflicted damage estimated at about twice the size of Grenada’s GDP. - Recent earthquakes in New Zealand are estimated to have cost the central government about 4 percent of GDP. - Two approaches to reflect non-debt obligations (e.g., PPPs) in DSAs: - Add the net present value of future payments (such as under PPP contracts) to public debt. - Count known and potential costs of non-debt obligations as primary spending (analytically equivalent). ### Identifying country-specific shocks and indicators - Identification of relevant country-specific shocks should use cross-country and country-specific experience and tools to detect risks from macroimbalances, private liabilities, off-balance-sheet public liabilities, and natural disasters. - Exchange rate risks: - Exchange rate assessments using CGER or other methodologies can inform the identification and calibration of exchange rate depreciation shocks. - Where the exchange rate is found to be overvalued (undervalued), an exchange rate shock of at least the maximum estimated magnitude of the overvaluation (undervaluation) would be expected in the DSA. - Use of vulnerability exercises and spillover reports: - Vulnerability exercises provide sectoral risk ratings (low, medium, and high risk). A medium or high risk rating warrants further investigation and may suggest relevant macro risks to include in the DSA. - Risk of transformation of private debt into public debt: - Rapid credit growth, asset price bubbles, and sustained surges in capital flows have been shown to precede banking crises. - Staff analysis shows that the ratio of domestic private sector credit to GDP is an efficient predictor of sovereign debt distress. - A domestic banking sector credit to the private sector-to-GDP ratio above 70 percent provides an early warning signal of sovereign debt distress. - Caveat: increases in this indicator may reflect financial deepening in countries undergoing structural reforms. ### Using financial sector assessments (FSAP, FSSA, BSA) in DSAs - FSAPs and FSSA: - FSAPs include estimates of capital shortfalls under stress tests; these estimates can inform whether to reflect financial sector risks in DSAs. - Possible ways to reflect FSAP results: - Calibrate bound test(s) using stress test estimates of capital shortfalls of systemic (or largest) banks. - Use FSSA’s risk assessment matrix (RAM): for medium/high impact risks on banking sector, include a contingent liability shock equivalent to half of the impact on the capital base (further calibrated by banking sector characteristics and market conditions). - Undertake joint stress tests of the financial sector and public debt where risks are high; start with simple feedbacks and align underlying assumptions so contingent liabilities feed into fiscal projections and vice versa. - Practical limitations: reporting of stress testing is not standardized; confidentiality and FSAP conduct issues may affect use of FSAP results in public DSAs. - Balance Sheet Approach (BSA): - The BSA examines assets and liabilities across sectors and key linkages; it helps identify vulnerabilities in non-financial corporate and household sectors by providing net financial position, net foreign currency position, and net short-term position. - BSA insights have informed Fund surveillance (Argentina 2001, Turkey 2001, Uruguay 2002, Iceland Selected Issues). - Consider encouraging BSA where external debt sustainability analysis points to high private sector external debt. - Existing methodologies such as the contingent claims approach (CCA) can be used where data is available to derive forward-looking indicators of potential fiscal costs. ### Assessing the impact of shocks: bound tests, scenarios, stochastic simulations - Methods: - Bound tests and alternative scenarios assess specific shocks. - Stochastic simulations capture uncertainty by drawing series of shocks from historical experience and can produce fan charts. - Limitations of current bound tests: - Current bound tests are relatively mechanistic, generally shocking one variable at a time, while historical shocks tend to be correlated. - The current framework includes one macroeconomic stress test—a 30 percent exchange rate depreciation—that is a tail risk compared to other, less extreme stress tests. - Empirical regularities to improve shock assessment: - Growth collapses and large terms-of-trade shocks have played a prominent role in episodes of large public debt increases but are not reflected in current stress tests. - In one half of episodes where real GDP growth was negative over a five-year period, public debt increased by more than 14 percent of GDP. - In one half of episodes where the terms of trade deteriorated by more than 25 percent in one year, public debt increased by more than 11 percent of GDP. - Linkages among macro variables after shocks: - Episodes with nominal exchange rate depreciations of more than 30 percent in one year against the U.S. dollar tended to have significant inflation increases and sharp real GDP growth declines relative to trend. - Banking crises, growth collapses, and large terms-of-trade shocks tend to coincide with large output losses, exchange rate depreciations, and higher inflation, all affecting the public debt outlook. - Better practice: make impacts more realistic by taking into account historical co-movements of output, inflation, and the exchange rate, or develop full country-specific scenarios capturing endogenous linkages (e.g., increases in cost of borrowing). - Tail-risk and scenario guidance: - Indicators of banking sector credit, combined with other financial indicators, can inform the decision to run a “tail risk” scenario or a bound test incorporating a financial crisis. - Stress testing contingent liabilities from the financial sector should be framed as an analytical exercise to inform policy and contingency planning, not as a prediction of crises or a presumption of bailouts. - Communication challenges from such stress testing should be handled under existing policies on transparency. *Source: IMF chapter text on strengthening contingent liabilities analysis* ### 40. Stochastic simulation methods could be applied to improve estimates of ### 40. Stochastic simulation methods could be applied to improve estimates of ### Stochastic simulations to gauge uncertainty around baseline debt projections - Current stress-test configuration: - Provides a rough estimate of uncertainty surrounding debt projections. - Persistence of shocks is calibrated in an ad-hoc manner. - Individual shocks do not allow for feedback between key macroeconomic and fiscal variables. - Advantages of stochastic simulations: - Can generate country-specific confidence intervals based on empirical models that take into account interaction between key variables. - Dynamics of underlying empirical models determine persistence of shocks, which can vary greatly across countries, adding realism. - Recommendation: - Encourage stochastic simulations for countries where data are available. ### Empirical illustration and key quantitative findings (Box 2) - VAR-based stochastic simulation example (market access country): - VAR estimated using annual data over the period 1995–2010 to generate a probability distribution for public debt-to-GDP over 2011–2015. - Baseline projection: public debt projected to decline, reaching 75 percent by 2015. - Under the most extreme shock in current public DSA framework: public debt increases to 92 percent of GDP by 2015 (17 percentage points above baseline). - Stochastic-simulation probabilities: - 25 percent probability that public debt would exceed 80 percent of GDP by 2015 (the 75th percentile). - 10 percent probability that public debt would exceed 85 percent of GDP by 2015 (the 90th percentile). - Implication: the growth shock in the extreme stress test is not a likely outcome based on recent historical experience in this example. - Sensitivity to model specification and sample period: - Confidence intervals from an autoregressive (AR) model are significantly wider than those from a VAR model — underlying dynamic specification and covariance of shocks matter. - Estimating the VAR over a shorter sample period (1999–2010 versus 1995–2010) resulted in narrower confidence intervals in this case. ### Implementation considerations and guidance - Caveats: - Confidence intervals are sensitive to model specification and sample period used for estimation. - May be less useful when structural shifts occur (e.g., changes in exchange rate regime, fiscal/monetary policy objectives/tools). - Procedural proposals: - Develop guidelines to ensure statistical foundations are sound and comparability across countries is maintained, recognizing DSA is best used for single-country analysis. - Consider centrally generating confidence intervals for countries where data are available to ensure comparability, while allowing country teams to tailor models to country-specific circumstances. - Where stochastic simulations are not feasible or useful, country teams should develop well-specified alternative scenarios to analyze risks. - Data coverage: - Most applications to date have used quarterly VARs available for around 40 countries, majority AEs. - Examples of fan charts cited: Greece, the UK, Germany, the US (November 2010 Fiscal Monitor); Greece, Ireland, Italy, Portugal, Spain (Fall 2010 Vulnerability Exercise); DSAs including fan charts: Morocco, Mauritius, El Salvador, Indonesia, Israel, Costa Rica. ### Methodologies for countries with limited data - Possible approach: - Exploit cross-country experience by estimating models using panel data for broadly similar countries. - Issues to address: - Classification of countries with similar attributes (reliance on concessional resources, exchange rate regime, dependence on commodity exports). - Trade-off: allow country-specific attributes (notably volatility) while constraining attributes harder to estimate robustly (e.g., correlation between shocks). - Safeguard against over-fitting: - Ensure confidence intervals are broadly consistent with historical forecast errors based on the WEO projection database to alleviate risk of data mining and understating risks. ### Vulnerabilities associated with the profile of public debt (Section VI) - Debt vulnerabilities arise from both level and profile (maturity, currency composition, creditor base). - Key risks: - High share of short-term debt at original maturity increases rollover and interest rate risk. - High share of foreign currency-denominated debt increases exchange rate risk and can pressure foreign exchange reserves. - Debt distress events typically preceded by increases in short-term debt and foreign-currency denominated debt. - Creditor-base composition (diversified, reliable, captive, domestic, foreign) affects rollover risk. - Additional useful indicators: - External financing needs and risk pricing (bond and CDS spreads) tend to rise before sovereign debt distress episodes, though spreads are noisy and only sustained increases may be relevant. ### Enhancing DSA with debt-structure and liquidity indicators - Proposal: - Add six indicators to the public DSA framework with associated benchmark levels based on their ability to signal debt distress events (flexible application for MACs where data are available). - Benchmarks intended to facilitate staff analysis, not to rate countries or provide firm early warning of market access loss. - Where creditor-base data are lacking, provide information where relevant (e.g., domestic vs foreign creditor breakdown) in DSAs. ### Indicative benchmarks for debt vulnerability indicators (Box 3) - Sample and methodology: - Examined 49 market access countries, mostly EMs; identified 38 debt distress events in 32 countries between 1993 and 2010. - Used the signal approach on annual information to derive benchmarks by minimizing noise-to-signal ratio; benchmarks are reference points, not precise indicators. - Table of variables and benchmark-related entries (preserved exactly as in source): - EMBI Gl obal spre ads ( Basi s Poi nts)1,17526< - Five-Year CDS spreads (Basis Points)81235< - External Financing Requirement (Percent of GDP)3141< - Public Debt in Foreign Currency (Percent of total)6875< - Short-term Public Debt at Original Maturity (Percent of total)1476< - Average Maturity of Debt to Private Sector (Years)697> ### Coverage of fiscal balance and public debt (Section VII) — key points - Importance of defining public debt: - Questions: appropriate government coverage (central vs general), how to include long-run spending pressures, whether to integrate assets. - Data availability is a critical constraint; a flexible approach to implementing good practices is required. - Expanding coverage of fiscal accounts (VII.A): - In line with GFSM 2001 and Manual on Fiscal Transparency, DSAs should use the concept of general government. - Decentralization can allow sub-national governments to create debt and offset fiscal adjustments; lack of transparency can lead to costly central government bailouts. - Country teams should seek to include all relevant general government entities in DSAs. - Wider aggregates should be reported where data are available, particularly public corporations that impose large fiscal risks (including sub-national). - Reclassifications of SOEs and PPPs into general government have increased reported debt (example: Portugal reclassifications increased reported debt by 10 percentage points of GDP in 2010). - Integrating long-run spending pressures into DSA (VII.B): - Age-related and health care spending can significantly impact primary balances and debt sustainability. - In AEs, combined annual spending on old-age pensions and health care is expected to increase by an average of 4 percent of GDP by 2030. - In EMs, projected increase in pension and health care spending is 2 percent of GDP on average over next two decades, with additional pressures beyond 2030. - Implications: - DSA five-year projection horizon constrains ability to include these long-term impacts. - Expand projection horizon in cases where pressures materialize quickly (mainly AEs); alternatively add memo item such as present value of pension and health care costs as a percent of current GDP. - For countries where these pressures are significant, assess impact on debt sustainability and potential reforms. - Assessing gross and net debts (Section VII.C begins after provided content). *Source: IMF staff excerpts from the provided PDF chapter/section.* ### 52. The assessment of the public debt may differ substantially depending on whether ### _080511 - 52. The assessment of the public debt may differ substantially depending on whether ### Gross versus net debt: key findings - Assessment can differ substantially depending on whether a gross or net debt measure is considered (Annex V). - Differences arise in both the level of debt and its evolution over time: - Level differences reflect large assets held by some governments (example countries cited: Norway, Finland, and Japan). - Evolution differences depend on whether increases in gross debt are accompanied by increases in assets. - Policy implication: the level of net debt should be discussed explicitly in the analysis, particularly for countries with high levels of gross debt (e.g., above the 60 percent mark discussed in Section III). - Tradeoffs: - Gross debt: permits cross-country comparisons with the greatest degree of reliability and coverage; better captures rollover risks and risks associated with the overall debt burden. - Net debt: limits scope for comparison because of differences in the types of assets reported by different countries, but complements gross debt when reliable government asset data are available. ### Data, measurement, and reporting recommendations - Whenever reliable data on government assets are available, net debt should be used to complement the assessment based on gross debt. - Advanced economies (AEs) should be encouraged to report net debt and net financial liabilities data on a more systematic basis. - For low-income countries (LICs) and emerging markets (EMs): - Data availability and reliability are important constraints, but reporting a narrower net debt definition should be pursued. - When financial assets and liabilities in the form of non-debt instruments are significant, it would be relevant for a country to use net financial liabilities. - If data are not reliable or are partial, a measure such as debt net of liquid assets would instead be considered. - Country-specific tailoring: - The relevant set of assets to be considered in the net debt definition may need to be tailored to each country, which creates complexity as the type of assets considered by each country might differ. - Note on methodology: - Importantly, when a net debt concept is explicitly introduced, yields on assets should be stripped away from the calculation of the primary balance definition. - A homogeneous basis for comparisons could be provided by using the concept of debt net of liquid assets where full asset coverage is not available. ### Risk-based approach to DSAs — implementation (major proposals) - Move towards a risk-based approach to debt sustainability analyses (DSAs) and away from standardization of most DSA elements. - Depth of the DSA would depend on extent of identified vulnerabilities, including debt level, fiscal risks, and debt profile. - A reasonable amount of standardization would be maintained, including: - Provision of a limited set of debt indicators. - A baseline scenario. - Comparisons with historical and “no policy change” scenarios. - A few bound tests. - Additional implementation elements: - Thorough assessment of the realism of baseline projections, particularly where a large fiscal adjustment or a large primary surplus is required; attention to long-term spending pressures due to population aging. - Depth of DSA would generally be evaluated against a reference point of 60 percent of GDP but would also consider country-specific factors (debt structure and other sources of fiscal risk) that might raise concerns at lower debt levels. - Fiscal risks should always be assessed, including risks from currency valuation (e.g., CGER measures), financial sector developments (e.g., evolution of banking sector credit, FSAP stress tests, liquidity and debt profile indicators, and balance sheet vulnerabilities where available), and off-balance-sheet public liabilities (e.g., size and experience with PPPs, and SOEs). - Where public debt is relatively high or fiscal risks are significant, customized bound tests and/or alternative scenarios would be expected, with design based on identification of the most important country-specific fiscal risks. - Analysis should include, where available, a brief discussion of fiscal risk mitigation and management strategies (e.g., restructuring and/or privatization of SOEs, strengthening of regulatory frameworks). - Use of stochastic simulation methods encouraged to improve understanding of uncertainty surrounding baseline projections; centralized generation of confidence intervals could facilitate this. ### Illustrative DSA implementation: Country A (detailed) and Country B (lighter) - Country A (detailed DSA) - Background: Public debt increased from around 50 percent of GDP in 2000 to close to 100 percent of GDP in 2010 due to sluggish growth and expansionary fiscal policies. - Vulnerabilities: Public debt projected to stabilize at roughly 120 percent of GDP in the medium term; high debt level (well above 60 percent of GDP) calls for detailed risk analysis. - Realism of baseline: Baseline assumes primary balance improves from a deficit of about 6 percent of GDP in 2010 to a surplus of 2 percent of GDP in the medium term; realism of this large adjustment should be discussed versus historical experiences and policy credibility. - Fiscal risks identified (examples): - Lower-than-projected growth would lead debt to increase and not stabilize. - About half of SOE debt (approximately 20 percent of GDP) is explicitly guaranteed by the government; SOE financial position is generally weak. - PPPs are very large (roughly 20 percent of GDP) and some PPPs need government bailout. - Banking sector has a very high loan-to-deposit ratio and credit to the corporate sector has been contracting. - Assessing shocks: DSA could include a contingent-liability shock of 15 percent of GDP given size of debt guarantees to SOEs and PPP risks; larger shocks could be envisaged to account for banking sector vulnerabilities based on stress tests; full-fledged alternative lower-growth scenario could be undertaken. - Debt profile: Four indicators associated with the debt profile show vulnerabilities; discussion of mitigating factors expected. - Coverage: Given risks from SOEs and PPPs, discussion of their coverage warranted; highlight large spending pressures from pension and health care costs and relatively low government financial assets. - Mitigation strategies: Authorities committed to actions such as SOE privatization, re-assessment of PPP contracts, financial sector policies to manage liquidity, and structural reforms to boost potential growth and sector efficiency. - Country B (lighter DSA) - Background: General government gross debt declined from a peak of 60 percent of GDP in 2006 to about 50 percent in 2010; sovereign wealth fund assets increased net asset position from roughly 140 percent of GDP to 160 percent over same period. - Vulnerabilities: Gross debt close to 60 percent but not likely to give rise to significant vulnerabilities; net asset position expected to increase medium term and unwind gradually over longer term as oil declines and aging pressures mount. - Realism of baseline: Assessment of realism of primary balance projection may not be necessary; sizeable general government overall surpluses expected even with modest oil price declines; adequate fiscal guidelines in place. - Fiscal risks: Risks relatively low and not likely to jeopardize sustainability; net asset position resilient; stress tests show banks would remain well capitalized under tail risk scenarios. - Assessing shocks: Standard bound tests expected; possible customized scenario to reflect decline in oil prices; a fan chart may not be necessary. - Debt profile and coverage: Debt profile analysis may not be necessary; public debt measured using general government definition; SOEs and PPPs do not pose serious vulnerabilities. - Mitigation strategies: Not necessary. ### Implementation and resource considerations - The risk-based approach helps reconcile need for deeper DSAs in some cases with severe constraints on staff resources. - In countries where fiscal policy and public debt sustainability are concerns, additional resources may be needed to analyze the issue. - Where it is not a concern, resources need not be diverted from more pressing matters. - Customization of DSA shocks and alternative scenario analysis has initial setup costs but is not expected to change considerably year-to-year if country circumstances remain broadly unchanged. - More specific guidance would be developed to implement the risk-based approach without overburdening staff resources and to clarify proposals to facilitate country team work and ensure adequate implementation. ### Issues for discussion (questions posed to Directors) - Do Directors agree there is a need to modernize the framework for fiscal policy and public debt sustainability analysis? - Do Directors agree with proposed broad areas for improvement (assessing realism of baseline projections, improving analysis of fiscal risks, integrating debt profile into the DSA framework)? Are there other areas needing greater attention? - Do Directors support use of reference levels for certain debt indicators (on level and profile) to trigger additional discussion of debt sustainability risk, as discussed in paragraphs 19 and 45? Do they agree existing tools and analyses (e.g., vulnerability exercises and FSAP) could be used as inputs to identify and quantify macroeconomic risks and contingent liabilities (paragraphs 28–35)? - Do Directors agree to move to a risk-based approach to DSA for all market-access countries, with depth and extent of analysis commensurate with sustainability concerns while maintaining reasonable standardization? *Source: IMF staff paper (content unit: _080511 - 52. The assessment of the public debt may differ substantially depending on whether).* ### Annex I: Debt Sustainability Analysis in Selected Countries ### Annex I: Debt Sustainability Analysis in Selected Countries ### Overview: pre-2009 Article IV coverage and outcomes - The countries experienced a rapid increase in public debt-to-GDP over 2007–2010, ranging from 15 percent (Italy) to 68 percent (Iceland). - A review of Article IV staff reports issued between 2006 and 2008 shows public and external DSAs were not systematically included; in most cases the magnitude of the subsequent increase in public debt was not anticipated, even under stress tests. - Table A1 (summarized in source) shows for selected countries: 2007 actual debt-to-GDP, 2010 actual debt-to-GDP, baseline public debt projections in the pre-2009 Article IV report, the projection year, year of Article IV, and whether public and external DSAs were included. Key exact values from Table A1: - Greece: 2007 = 105; 2010 = 142; Baseline = 72; Most Extreme Shock = 98; By year = 2013; Year of Art. IV = 2007; Public DSA included? Yes; External DSA included? No - Iceland: 2007 = 29; 2010 = 97; Baseline = 36; Most Extreme Shock = - ; By year = 2013; Year of Art. IV = 2008; Public DSA included? No; External DSA included? Yes - Ireland: 2007 = 12; 2010 = 57; Baseline = 6; Most Extreme Shock = 16; By year = 2012; Year of Art. IV = 2007; Public DSA included? Yes; External DSA included? No - Italy: 2007 = 104; 2010 = 119; Baseline = 111; Most Extreme Shock = 122; By year = 2013; Year of Art. IV = 2008; Public DSA included? Yes; External DSA included? No - Latvia: 2007 = 8; 2010 = 40; Baseline = - ; Most Extreme Shock = - ; By year = - ; Year of Art. IV = 2006; Public DSA included? No; External DSA included? Yes - Portugal: 2007 = 63; 2010 = 83; Baseline = 62; Most Extreme Shock = - ; By year = 2013; Year of Art. IV = 2008; Public DSA included? No; External DSA included? Yes - Ukraine: 2007 = 12; 2010 = 40; Baseline = 14; Most Extreme Shock = 39; By year = 2011; Year of Art. IV = 2006; Public DSA included? Yes; External DSA included? Yes - United Kingdom: 2007 = 44; 2010 = 77; Baseline = 43; Most Extreme Shock = 53; By year = 2012; Year of Art. IV = 2008; Public DSA included? Yes; External DSA included? No - United States: 2007 = 62; 2010 = 92; Baseline = 55; Most Extreme Shock = 67; By year = 2013; Year of Art. IV = 2008; Public DSA included? Yes; External DSA included? Yes ### Greece - 2007 Article IV staff report: - Recommended sustained fiscal consolidation given high public debt and projected increases in pension and health care costs. - Included a public DSA (charts with standard bound tests) and brief discussion. - Baseline projection: public debt-to-GDP projected to fall from 93 percent in 2007 to 72 percent in 2013. - Growth shock scenario: debt projected to rise to 98 percent of GDP by 2013. - Subsequent warnings: - Two years later staff warned public debt could rise to 115 percent of GDP by 2010 even after fiscal consolidation, and recommended further adjustment. ### Iceland - 2008 Article IV staff report: - Recommended a tighter fiscal stance given risk of króna depreciation and banking sector concerns. - Iceland’s gross external debt ~560 percent of GDP at end-2007, largely due to external borrowing by domestic banks. - 2007 and 2008 Article IV reports included a full external DSA (charts, table, written discussion); public DSA not included. - 2008 external DSA baseline: net external debt projected to decrease gradually to 229 percent of GDP in 2013 from 252 percent in 2007. - Bound tests: external debt most vulnerable to a real exchange rate depreciation shock. - Despite minimal public sector debt (<30 percent of GDP in 2007), sovereign risk premium was elevated. - Outcome: - After government intervention in the banking sector in 2008, public debt jumped to 72 percent of GDP by year-end. ### Ireland - 2007 Article IV staff report: - Included a public DSA showing government net debt (gross debt minus assets of National Pensions Reserve Fund and Social Insurance Fund) low and declining. - Baseline: net debt projected to fall from 12 percent of GDP in 2006 to 6 percent of GDP by 2012. - Worst outcome: growth shock raised net debt to 16 percent of GDP in 2012. - Staff identified age-related spending pressures as the most significant long-run threat. - Noted banks’ large property exposures but stress tests suggested cushions were adequate. - Outcome: - Net debt to GDP increased nearly fivefold from 2007 to 2010 due to sharp GDP contraction and large fiscal deficits tied mainly to bank recapitalization costs. ### Italy - 2008 Article IV staff report: - Advised balancing counter-cyclical recession measures with the need to maintain debt sustainability given high debt and widening sovereign spreads. - Included a public DSA (charts with bound tests plus table). - Baseline: public debt projected to rise to 111 percent of GDP in 2013 from 104 percent in 2007. - Most extreme shock (contingent liabilities): debt could rise to 122 percent by 2013. - Staff noted public debt likely to rise further under more realistic macro assumptions and possible bank support operations; discussed long-term sustainability concerns beyond five-year horizon. - Outcome: - Fiscal position deteriorated sharply in 2009; public debt = 119 percent of GDP at end-2010. ### Latvia - 2006 Article IV staff report: - Warned of overheating risks and recommended sizable front-loaded fiscal consolidation. - Included an external DSA (charts with bound tests plus table); no public DSA because public debt was low (10 percent of GDP). - External DSA baseline: external debt projected to rise to 118 percent of GDP in 2011 from 101 percent in 2005. - Most extreme shock (real depreciation): external debt projected to rise to 170 percent of GDP by 2011. - Staff cautioned that fiscal deterioration had been masked by EU grants and cyclical conditions, risking a hard landing. - Outcome: - Real GDP fell 4 percent in 2008 and 18 percent in 2009 (largest contraction in the world). - Public debt increased rapidly due to borrowing from official sources to address revenue collapse and crisis effects. ### Portugal - 2007 and 2008 Article IV staff reports: - Highlighted weak but improving fiscal position; private debt high and growing. - 2007 report: pension reforms substantially improved long-term fiscal sustainability. - Reports included a chart on fiscal sustainability risks showing alternative government debt-to-GDP paths. - Staff projected public debt would fall below 60 percent of GDP in 2011 and remain below 60 percent through 2040 assuming medium-term growth and primary surplus targets met. - 2008 report: acknowledged stronger fiscal footing from decisive actions but urged further consolidation given high public debt. - Both reports contained an external DSA table (no charts, bound tests, or discussion). - 2008 external DSA projection: external debt to rise to 220 percent of GDP in 2013 from 194 percent in 2007. ### Ukraine - 2006 Article IV staff report: - Noted public debt had plunged but contingent liabilities remained high at some 30 percent of GDP. - Included both external and public DSAs (charts with bound tests plus table) and brief discussion. - Staff emphasized risks to external debt sustainability; public debt viewed as more benign unless contingent liabilities materialized. - Public DSA included a customized shock (realization of contingent liabilities linked to “lost savings deposits”): public debt-to-GDP increased from 17 percent in 2006 to about 45 percent in 2007 under that scenario. - Outcome: - Public debt more than tripled in following years, reaching an estimated 40 percent in 2010 due to government support to national energy company, bank recapitalization, and VAT refund arrears. ### United Kingdom - 2008 Article IV staff report: - Staff agreed fiscal adjustment required and authorities should maintain net public debt below a ceiling of 40 percent of GDP. - Warned that using weaker growth and revenue projections, net public debt could breach 40 percent as early as 2009 even with planned adjustment. - Included a public DSA (charts with bound tests) but no written discussion. - Bound tests showed (gross) public debt could rise to 53 percent of GDP by 2012 in the most extreme scenario (contingent liabilities). - Outcome: - Recession produced unprecedented fiscal deterioration; public debt reached 77 percent of GDP in 2010. ### United States - 2007 and 2008 Article IV staff reports: - Discussed need for medium-term fiscal consolidation, but lacked detailed discussion of public debt. - Both reports contained a public DSA (charts with bound tests) with no written discussion. - Bound tests did not anticipate the surge in public debt from 62 percent of GDP in 2007 to 92 percent in 2010. - 2008 DSA baseline: public debt projected to rise to 55 percent of GDP by 2013. - 2008 most extreme shock (constant primary balance): debt rose to 67 percent of GDP by 2013. - Report characterized general government debt as “manageable” but cautioned authorities’ medium-term adjustment plans were premised on unrealistic assumptions. --- ### Annex II: Overview of Current Framework for Public DSA in Market-Access Countries (MACs) and LICs ### Implementation and template - Framework for MACs introduced in 2002 and refined in 2003 and 2005. - Framework for LICs developed jointly with World Bank in 2005. - Country teams implement the framework using a standardized DSA template that generates tables and charts with main elements: - Baseline scenario for debt-to-GDP over a five-year projection horizon (MACs). - Sensitivity analysis via two alternative scenarios and six bound tests. - Charts comparing baseline and alternative scenarios, impact of bound tests, and gross financing needs under the baseline (percent of GDP). - Customized scenarios are possible; they capture interaction between macro and fiscal variables but are not common. ### Alternative scenarios used in MAC public DSA - A1. Key variables (real GDP growth, real interest rate, and primary balance) set at historical averages (calculated over a 10-year period). - A2. No policy change – constant primary balance / GDP (fixed at level projected in first year). - Purpose: illustrate divergence from historical macro experience (A1) or current fiscal stance (A2) and act as a disciplining device. ### Bound tests in MAC public DSA - Permanent shocks to three key variables each set at one-half of a standard deviation over a ten-year historical period: - B1. real interest rate; - B2. real GDP growth rate; - B3. primary balance / GDP. - Combined permanent shock, each set at one-quarter of a standard deviation: - B4. real interest rate; real GDP growth rate; and primary balance / GDP. - Large exchange rate shock: - B5. 30 percent permanent exchange rate depreciation. - Contingent liability shock: - B6. 10 percent permanent increase in “other debt creating flows.” - Notes: - Exchange rate shock captures valuation effects on foreign-currency or exchange-indexed debt. - Contingent liability shock raises public debt by ten percent of GDP in all countries. ### Key distinctions for Public DSA in LICs - Debt measured in present value rather than nominal terms. - Projection horizon: 20 years (vs five years for MACs). - Bound tests do not include a real interest rate shock. - Shocks persist for two years but are larger in magnitude (one SD for individual shocks; one-half SD for combined shock). - Additional alternative scenario: permanently lower growth included. - Alternative scenarios and bound tests also shown for debt-to-revenue and debt service-to-revenue indicators. - Charts and tables report grant-equivalent financing and grant element of new borrowing. *Source: Annex I and Annex II, "Debt Sustainability Analysis in Selected Countries" (IMF Article IV staff reports and framework descriptions).* ### Annex III. Estimation of Indicative Public Debt Thresholds ### Annex III. Estimation of Indicative Public Debt Thresholds ### Theoretical underpinnings - Partial-equilibrium foundations: - Domar (1944): necessary condition for sustainability; interest rate and GDP growth exogenous. - Blanchard et al. (1990) (building on Buiter (1985)): two conditions for public debt sustainability: - (i) the ratio of debt to GDP should converge in the long run to its initial level; and - (ii) the present value of the ratio of the primary budget deficit to GDP should be equal to the negative of the current level of public debt to GDP. - General-equilibrium literature: - Diamond (1965) overlapping-generations model: government debt decreases utility when the economy is dynamically efficient; may increase utility if dynamically inefficient. - Rankin and Roffia (2003), Brauninger (2005), Yakita (2008): extensions in overlapping-generations and endogenous growth settings; not all explicitly model sustainable public debt. - Ghosh et al. (2011) framework: - Models default as inability-to-pay from inability to roll-over debt amid rising interest rates and stochastic shocks. - Distinguishes long-run public debt (d* in Figure A1) and maximum sustainable public debt (d̄ in Figure A1): - Long-run debt level: economy’s normal convergence point; shocks above this level should be offset by subsequent primary balances returning debt to long-run average. - Maximum sustainable debt level: point where interest rate approaches infinity (market access lost); beyond this no feasible sequence of primary balance adjustments can restore sustainability. - Markets tend to raise interest rates before the maximum sustainable debt level is reached. ### Empirical evidence - Partial-equilibrium / deterministic approaches: - 2003 WEO: thresholds estimated at 75 percent for AEs and 25 percent for EMs (long-run oriented). - Mendoza and Oviedo (2003): incorporating uncertainty and credible servicing needs suggests thresholds for representative AE within 100–150 percent of GDP and for representative EM within 35–75 percent of GDP (maximum-sustainable oriented). - Model-based / reaction-function approaches: - Bohn (1998) approach: primary surplus responds to debt to satisfy intertemporal budget constraint. - 2003 WEO estimates: 80 percent for AEs and 50 percent for EMs; Abiad and Ostry (2005) confirm EM result. - Mendoza and Ostry (2008): larger sample, control for autocorrelation; find marginal response of primary balance to debt weaker at high debt levels; combined-sample threshold of 48 percent. - Ostry et al. (2010) unifying framework: - Combines non-linear primary-balance response with uncertainty to quantify long-run and maximum sustainable public debt. - Implementation steps: - (i) estimate primary balance reaction function; - (ii) determine interest rate–growth differential; and - (iii) calculate country debt limits and fiscal space. - Results: - Long-run public debt median based on historical market interest rates: 50 percent. - Long-run public debt median based on projected interest rates: 63 percent. - Maximum sustainable public debt median based on historical market interest rates: 192 percent. - Maximum sustainable public debt median based on projected interest rates: 183 percent. - Signal approach (Kaminsky et al. (1998) lineage): - Defines thresholds as early-warning signals; optimizes noise-to-signal ratios. - Hemming et al. (2003) application to EMs suggests threshold at 77 percent (maximum-sustainable oriented). - Forecast performance: - Signal approach: statistically and economically significant predictors of crises; out-of-sample accuracy only slightly inferior to in-sample. - Parametric approaches: perform substantially worse out of sample than in sample. - Overall: signal approach outperforms parametric on out-of-sample criteria, but underperforms parametric on in-sample criteria. ### Estimating public debt thresholds for EMs (methodology and dataset) - Motivation: update EM estimates using data post early-2000s. - Hybrid methodology: - Long-run debt estimated using partial-equilibrium framework (following Ostry et al. (2010)). - Maximum sustainable debt estimated using: - parametric method (panel estimations of fiscal reaction functions, following Abiad and Ostry (2005)); and - non-parametric signal approach (Hemming et al. (2003)). - Dataset: - 50 EMs over the period 1993–2009: - Albania, Algeria, Argentina, Armenia, Bosnia and Herzegovina, Brazil, Bulgaria, Chile, China, Colombia, Costa Rica, Croatia, Dominican Republic, Ecuador, Egypt, El Salvador, Estonia, Georgia, Guatemala, Hungary, India, Indonesia, Israel, Jamaica, Jordan, Kazakhstan, Latvia, Lebanon, Lithuania, Macedonia, Malaysia, Mexico, Morocco, Pakistan, Panama, Peru, Philippines, Poland, Romania, Russia, Serbia, South Africa, Sri Lanka, Thailand, Tunisia, Turkey, Ukraine, Uruguay, Venezuela, and Vietnam. - Data sources: Vulnerability Exercise for EMs and the WEO. ### Long-run debt levels (results and sensitivity) - Operationalization: - Expected future primary balances approximated by historical sample average. - Future interest rates and growth assumed equal to historical averages. - Formula (constant path assumption): d* = p / (r - g) where d* is long-run debt, p is historical average primary balance, r is historical average interest rate, and g is historical average GDP growth rate. - Results: - Long-run debt level estimated on average at 58 percent of GDP (using historical averages). - Using average projected interest rate in the WEO as discount rate (following Ostry et al (2010)) yields long-run debt level of 49 percent of GDP, on average. - Results are sensitive to macroeconomic assumptions. ### Maximum sustainable debt levels: parametric method (fiscal reaction functions) - Empirical specification: - Primary balance reaction function estimated in panel with country-specific intercepts and macro controls: - p_it = α_i + β b_it-1 + Σ γ_j X_j,it + ε_it (formula as presented) - p_it: primary balance in country i at time t. - α_i: country-specific intercept. - b_it-1: public debt level at end of previous period. - X_j: vector of macroeconomic variables affecting primary balance unrelated to solvency requirement. - Key empirical findings: - Primary surpluses respond positively to increases in public debt, indicating intertemporal budget constraint satisfaction. - Reaction strengthens when public debt crosses a 78% threshold, suggesting fiscal tightening after crossing the threshold. - Other macro variables in the specification have expected signs and are statistically significant. - Summary statistics presented (Table A3 / Table context): - Spline at 78% coefficient positive and statistically significant. - Number of countries: 50. - Observations: between 607 and 696 across specifications. - Adj. R-squared values: 0.58, 0.64, 0.65, 0.68, 0.68 across specifications. ### Maximum sustainable debt levels: signal approach (non-parametric) - Methodology: - Signal approach applied to debt/GDP indicator to identify threshold that minimizes noise-to-signal ratio and maximizes predictive power for debt distress. - Result: - Maximum sustainable debt level estimated at 63 percent of GDP for EMs. - Noise-to-signal ratio well below one and other efficiency indicators point to relatively high predictive power for this indicator. ### Comparative results and cross-study ranges (selected findings) - Cross-study median and ranges as summarized in Table A4: - IMF, 2003 WEO (deterministic/intertemporal tests): - Long-run debt: AEs: 75 percent; EMs: 25 percent. - Model-based results for AEs: 80 percent; EMs: 50 percent. - Deterministic range for AEs: 100–150 percent; for EMs: 35–75 percent (Mendoza and Oviedo approach). - Hemming et al. (2003) (signal approach): - EMs: 77 percent (maximum sustainable oriented). - Ostry et al. (2010): - Long-run debt (LRD): median 50–63 percent (50 percent based on historical market interest rates; 63 percent based on projected interest rates). - Maximum sustainable debt (MSD): median 183–192 percent (192 percent historical; 183 percent projected). - Individual-country ranges reported: LRD: 50–111; MSD: 150–263. - This annex’s EM estimates (explicit results): - Long-run debt: average 58 percent of GDP (historical averages); 49 percent of GDP (using average projected interest rate). - Parametric reaction-function evidence: fiscal reaction strengthens after crossing 78 percent of debt/GDP. - Signal approach MSD estimate: 63 percent of GDP. ### Annex IV — Using historical episodes to calibrate tail risks (methodology and stylized outcomes) - Purpose: - Specify stress tests for tail risks based on historical episodes where public debt increased significantly. - Empirical sample and episodes: - Sample: 127 countries over 1970–2008. - Episodes: 166 episodes where public debt increased by more than 20 percent of GDP over a five-year period. - Key drivers frequently present in these episodes: currency crises, banking crises (contingent liabilities), large persistent declines in output growth (“growth collapses”), large deteriorations in terms of trade. - Calibration principles: - Linkages across macro variables (exchange rate, inflation, real GDP growth) accounted for to produce realistic joint shocks and resulting debt outcomes via debt accumulation identities. - Currency-depreciation episodes: - Analysis of 104 episodes with nominal exchange rate depreciation ≥ 30 percent in one year against the US dollar: - Median (50th percentile) outcome: 56 percent depreciation in one year. - Lower and upper quartiles (25th and 75th percentiles): 39 to 99 percent depreciation. - Current framework uses a permanent 30 percent exchange rate depreciation affecting debt via valuation of foreign-currency debt. - The framework’s 30 percent shock produces an initial public debt-to-GDP impact close to the 50th percentile across the 104 historical episodes and close to the 10th percentile by the end of a five-year projection. - The 10th percentile of exchange rate changes across episodes is 31 percent; combined with 10th percentile output and inflation changes it yields a similar medium-term debt accumulation but with more gradual response. - Contingent-liability (banking crisis) episodes: - Framework includes a permanent 10 percent increase in contingent liabilities (all else equal). - Analysis of 97 banking crises (Laeven and Valencia (2010)): - Such episodes tend to coincide with large output losses, exchange rate depreciations, and higher inflation. - Direct fiscal costs of banking crises: median outcome of 10 percent of GDP (Laeven and Valencia (2010), Table 4). - The framework’s 10 percent contingent-liability shock has an initial impact above the median (50th percentile) but stabilizes below the median over the remainder of the five-year projection — indicating that half of historical banking-crisis episodes had public-debt increases exceeding 18 percent of GDP, above the 10 percent framework shock. - Growth collapses: - Defined as episodes with negative real GDP growth over a five-year period. - 99 growth-collapse episodes analyzed for joint outcomes in real GDP growth, exchange rate changes, and inflation. - The framework’s growth shock outcome is close to the 25th percentile of historical outcomes, implying three-quarters of historical growth-collapse episodes had larger output losses than the calibrated growth shock. - Terms-of-trade episodes: - Not explicitly modeled in the core framework, despite empirical importance. - Applying the historical-episode methodology to terms-of-trade shocks produces a median public debt-to-GDP increase of almost 12 percent — below the 18 percent increase median for contingent-liability and depreciation shocks. - Stylized figure references (staff calculations): - Figure A2: Large exchange rate depreciation — 30% depreciation (DSF) compared with historical episodes’ 10th and 50th percentiles. - Figure A3: Contingent liability shock — 10% contingent liability shock (DSF) compared with historical episodes’ 25th and 50th percentiles. - Figure A4: Growth shock — Growth shock (DSF) compared with historical episodes’ 25th percentile. - Figure A5: Comparison of Terms of trade, Contingent liability and Large Depreciation shocks — historical episodes’ 50th percentiles; median public-debt increases: contingent liability and depreciation shocks ~18 percent of GDP, terms-of-trade median ~12 percent of GDP. *Source: Annex III and Annex IV, "Estimation of Indicative Public Debt Thresholds" (IMF staff material provided).* ### Annex V. Defining Gross and Net Debt ### Annex V. Defining Gross and Net Debt ### Definitions: Gross debt - Gross debt includes only those liabilities that are in the form of debt instruments. - Conceptually, gross debt refers exclusively to financial claims that exist and require the payment of principal and interest to creditors. ### Definitions: Net debt and variants - Net debt is derived by stripping away a number of specific financial assets from the gross debt definition. - Debt net of liquid assets: netting out highly liquid assets from gross debt gives rise to the definition of debt net of liquid assets, a useful concept when data are either scarce or not reliable. - Liquid assets are defined as those that can be liquidated at short notice without a significant loss in value, such as government’s deposits. - Net financial liabilities: when data availability is not a major constraint, the broader concept of net financial liabilities may be computed as the difference between total financial liabilities and assets. - This definition is essentially equivalent to that of net financial worth, as specified in the Government Finance Statistics Manual 2001(GFSM 2001), but with reversed signs. ### Total financial liabilities and assets - Total financial liabilities is a broader concept to assess government indebtedness that adds specific public sector obligations to the gross debt measure such as equity, investment fund shares and financial derivatives. - This definition may be better suited to understand the overall balance sheet exposure of the government when liabilities in the form of equity and derivatives are significant. - Total financial assets can be derived considering those assets that are equivalent in scope to the obligations included in the total financial liabilities definition. - Non-financial assets are generally excluded from the analysis of the government’s balance sheet position because they are difficult to value. ### Data availability and measurement choices - In some circumstances data availability may point toward the use of debt measures that are less stringent in terms of the information required on government’s assets. - A particular case is that of highly liquid assets, which in general are readily available to assess at least partially the balance sheet position of the government. - Difficulties in cross-country comparisons may arise because: - (i) the categories of reported assets differ substantially across countries and do not necessarily conform to international statistical definitions to compute net debt; - (ii) certain countries only report information on the most liquid assets (government’s deposits), thus narrowing the definition of net debt on an ad-hoc basis; - (iii) information on total financial assets and liabilities is only provided by a few countries, whose method of valuation may not be fully consistent across them. *Source: Annex V. Defining Gross and Net Debt (PDF).* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2011/_080511.pdf_