## _100608

## Source details

**Canonical URL:** [_100608](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2008/_100608.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2008/_100608.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2008/_100608.pdf.json)

---

### I. Introduction and Country Coverage — Key Points
- Objective:
  - Support LICs in achieving development goals without creating future debt problems; maintain sustainability for countries receiving HIPC and MDRI relief.
- Joint preparation and frequency:
  - Country DSAs are prepared jointly by World Bank and IMF staffs with close collaboration on macroeconomic baseline, alternative scenarios, debt distress rating, and write-up.
  - Joint Bank-Fund LIC DSAs are generally expected to be prepared once a year for PRGF-eligible, IDA-only countries.
  - For PRGF-eligible countries that are not IDA-only, Fund staff expected to produce a LIC DSA once a year unless the country has significant access to market financing.
- Role and communication:
  - DSF intended as an upstream device to inform country teams’ dialogue with authorities rather than an ex-post consistency check.
  - Country teams should communicate frequently with MDBs, present DSA results to authorities, share final DSA files with authorities, and encourage authorities to consent to publication.

### Box 1 — Key Enhancements to the DSF (selected)
- Historical scenarios must be used actively; large differences between baseline and historical scenarios must be justified.
- Past projections should be scrutinized against outcomes; post-mortems are expected in write-ups.
- High projected growth dividends associated with large upfront borrowing (defined as an annual increase in the PV of public external or total public debt of 5 percent of GDP or more) trigger:
  - Inclusion of an alternative “high-investment, low-growth” scenario.
  - Detailed and explicit justification of projected growth dividends.
- Financing assumptions that materially affect debt evolution require explicit justification.
- Public DSAs must be included in all DSAs; write-ups must flag when inclusion of domestic debt changes interpretation versus external-only analysis.
- Additional analysis required where increased private external capital into sovereign debt instruments may create new vulnerabilities.
- Use three-year moving average CPIA score to reduce volatility of thresholds.
- DSAs to be published as supplements to Fund staff reports, self-contained in Bank and Fund documents, and made easily accessible.
- Preliminary DSA should be included in IMF briefing papers for early feedback; Bank-Fund collaboration prior to briefing papers emphasized.

### LIC DSA Framework — Pillars and Structure
- Three framework pillars:
  1. Standardized forward-looking analysis of debt and debt-service dynamics under baseline, alternative scenarios, and standardized stress tests (bound tests).
  2. Debt sustainability assessment based on indicative country-specific debt-burden thresholds that depend on policy and institutional quality.
  3. Recommendations on borrowing/lending strategy to limit debt-distress risk while maximizing resources for MDGs.
- Templates and projection horizon:
  - One mandatory pre-set template for PPG and private external debt in external DSA and for total public sector debt including domestic debt in public sector DSA.
  - Template designed for a twenty-year projection period.
  - Uniform discount rate used to calculate PV of future external debt-service obligations; adjusted when the six-month average U.S. dollar CIRR deviates from the template rate by more than 100 basis points for six months or more.
  - Template automatically produces output tables for baseline dynamics, standardized alternative scenarios, and stress tests.

### A. Debt and Debt-Service Projections and Indicators
- Coverage and definitions:
  - Public and publicly-guaranteed (PPG) debt includes debt owed by the central bank to the IMF, if any.
  - Private external debt is included in external DSA though not used for IDA grant allocations; inclusion can change overall sustainability assessment.
  - External debt defined on a residency basis and may include domestic-currency denominated debt.
  - Analysis covers medium- and long-term debt; short-term debt affects financing need within the year.
- Scenarios and stress tests:
  - Bound tests are partial and assume passive fiscal policy.
  - Fiscal reaction functions may be included in alternative scenarios if important; need separate modeling.

### B. Country-Specific Debt-Burden Thresholds (CPIA-based)
- CPIA-based performance categories:
  - Strong performance: CPIA rating at or above 3.75.
  - Medium performance: CPIA rating between 3.25 and 3.75.
  - Poor performance: CPIA rating at or below 3.25.
- Use three-year moving average CPIA score to determine policy performance under the DSF.
- Indicators interpretation:
  - PV of debt captures concessionality and future obligations (solvency).
  - Debt-service indicators measure immediate burden and potential liquidity problems.
  - Repayment capacity measured by GDP, exports of goods and services, or government revenues depending on constraints.

### C. Debt Distress Risk — Classification Guidance
- Four classifications required in every joint Fund-Bank and Fund-only DSA:
  - Low risk:
    - All debt indicators well below country-specific debt-burden thresholds.
    - Stress testing and alternatives do not result in significant breaches.
    - If only one indicator is above its benchmark, judgment required to determine cause.
  - Moderate risk:
    - Baseline does not indicate breach, but alternative scenarios or stress tests cause significant rise in debt-service indicators (nearing thresholds) or breaches.
  - High risk:
    - Baseline indicates protracted breach of thresholds but country not currently facing payment difficulties; alternative scenarios/stress tests exacerbate condition.
  - In debt distress:
    - Current debt and debt-service ratios in significant or sustained breach of thresholds; existence of arrears generally suggests debt distress unless other reasons explain nonpayment.

### Judgment versus Mechanistic Classification (paragraph 11)
- Mechanistic classification must be balanced with judgmental adjustments in specific cases, e.g.:
  - Marginal and temporary breaches.
  - Ability to pay evidenced by foreign exchange reserves not captured in template.
  - Lack of available CPIA scores.
- When judgment applied, it should be explained in the DSA write-up.
- CPIA scores are averages of 16 indicators; CPIA thresholds should not be used mechanically.
- Where three-year averaging causes undue volatility in ratings and IDA grant share, country teams may apply discretion on timing upon confirmation with reviewing departments (PRMED and SPR).

### Role and Design of DSAs (upstream device)
- DSAs should:
  - Function upstream of Bank and Fund program design.
  - Inform dialogue with authorities and early detection of debt-related vulnerabilities.
  - Serve as a cornerstone for medium-term debt strategies, fiscal frameworks, and public expenditure planning.
- Promote realism through:
  - Scrutinizing baseline projections via reality checks.
  - Using precautionary features of the DSF.

### Standard Reality Checks and Scaling-Up Scenarios
- Historical scenario comparisons:
  - Baseline debt ratios significantly lower than historical 10-year averages raise concerns; require explicit justification.
- Post-mortems:
  - DSAs must explain differences between prior assumptions and outcomes; revise assumptions where previous DSAs were overly optimistic.
- Financing and fiscal shifts:
  - Notable improvement in financing terms or sharp fiscal/structural shifts that materially affect sustainability require explicit justification.
- Scaling-up guidance:
  - Justify assumptions on impact of additional public investment on GDP, exports, revenues using country history, analytic considerations, and empirical work.
  - Empirical literature general conclusions: prolonged growth accelerations are rare; macro returns often lower than micro returns; policy and institutional quality strongly influence returns; volatility argues for caution.
  - Mandatory inclusion of an alternative “high-investment, low-growth” scenario if baseline assumes growth rates at least one standard deviation above historical average or if reversion to historical growth implies significantly worse outlook.
  - Detailed justification required when large upfront borrowing defined as PV increase of 5 percent of GDP or more.

### Box 2 — Indicators Linking Debt-Financed Investment and Growth (selected)
- Rates of Return: microeconomic project returns; implementation lags; public capital stocks and shortfalls; composition of expenditures.
- Structural Constraints: CPIA, governance, Doing Business, PEFA, public investment completion rates, skill shortages.
- Macroeconomic Constraints: cost of capital, private investment growth, banking lending capacity, real exchange rate measures.
- Aggregate Trends: per capita GDP growth, TFP growth, binding constraints analyses.

### Treatment of Domestic Debt
- Requirement:
  - All LIC DSAs must include a public DSA regardless of domestic debt size.
- Rationale:
  - Public domestic debt typically higher cost and shorter maturities; rising domestic debt increases likelihood of external debt distress.
- Coverage considerations:
  - Public domestic debt defined on a residency basis and may include foreign currency-denominated obligations.
  - Data should aim to cover broader public sector liabilities; data limitations may limit initial coverage.
  - Contingent liabilities (including PPPs and financial sector weaknesses) should be taken into account when possible and flagged.
- Staff assessment focus:
  - Domestic debt risks where domestic debt stocks significant (i.e., above 15-20 percent of GDP); rapid recent build-up warrants explanation.
  - Assess creditor base, likely duration, financing burden, medium-term implications, and primary balance consistency.
  - Inclusion of domestic debt may change fiscal sustainability assessment but does not affect external debt distress classification or IDA grant allocation.

### Treatment of Debt Held by Private External Creditors
- New vulnerabilities identified:
  - Abrupt reversals in market sentiment causing sudden capital outflows.
  - Non-standard financing terms (e.g., collateralization with future export receipts) weakening medium-term sustainability.
  - Secondary balance sheet effects on domestic financial system creating contingent public liabilities.
- Complementary analyses recommended:
  - Short-term debt-related vulnerabilities and financial sector soundness analyses.
  - Additional indicators to highlight sovereign liquidity risks, external liquidity and rollover risks, reserve cover adequacy, and financial sector weaknesses.
- Policy measures to discuss with authorities (illustrative):
  - Assign legal borrowing authority and permissible instruments.
  - Strengthen debt recording, liquidity forecasting, indicators to monitor private borrowing risks.
  - Provide technical assistance where needed.
  - Re-assess reserve adequacy.
  - Sequence banking supervision and prudential reforms before capital account liberalization.

### Suggested Indicators for Vulnerability Analysis (selected)
- Public sector solvency: PV public sector debt-to-GDP (public sector revenue); PV external public sector debt-to-GDP (exports); PV foreign-currency denominated public sector debt-to-GDP; PV contingent liabilities.
- External solvency: PV external debt-to-GDP (exports); external debt-to-GDP; non-interest external current account deficit that stabilizes external debt-to-GDP.
- Public flow indicators: public sector debt service-to-revenue; external public debt service-to-exports; public sector gross financing need (in percent of GDP); short-term public debt-to-total debt.
- External flow indicators: external debt service-to-exports (revenue); external gross financing need (billions of U.S. dollars); gross official reserves-to-short-term external debt.
- Financial soundness: regulatory capital-to-risk-weighted assets; nonperforming loans-to-total loans; claims on the Government and Central Bank-to-total banking sector claims; private sector credit growth; foreign currency loans-to-total loans.

### Role of LIC DSAs in IDA Allocations and Fund Program Design (paragraph 26)
- LIC DSAs inform grant/loan mix in IDA allocations and other multilaterals.
- Financing rules by risk classification:
  - High risk or in debt distress: 100 percent grant financing from IDA.
  - Moderate risk: 50/50 blend of grants and traditional credits.
  - Low risk: 100 percent credit financing on standard IDA terms.
- DSA risk ratings determine IDA grant allocations; assume regular IDA credit terms in projection years where grant finance not already committed.
- For the Fund, programs where debt sustainability is a concern could:
  - Supplement existing debt limits with conditionality related to PV of external debt.
  - Make more systematic use of limits on overall fiscal deficit relative to limits on domestic financing.
- Conditionality on PV of external debt expected to be indicative targets due to complexity.
- Recommended grant element for concessional loans presumed to increase with debt distress risk.
- DSA should assess impact of alternative financing strategies and recommend minimum concessionality for new lending.
- When risk classification shifts upward, staff should conduct comprehensive reassessment of recommended debt accumulation strategy.

### IV. Modalities for Preparing DSAs — Frequency, Presentation, and Process
- Frequency and scheduling:
  - Joint LIC DSA expected once a year for each IDA-only, PRGF-eligible country.
  - Fund typically produces a DSA for an Article IV consultation or program requests/reviews; Bank requires DSA for Country Assistance Strategies and IDA allocations.
  - Teams should agree schedule each calendar year and seek agreement on broad DSA parameters at least 90 business days ahead of prospective Board meeting.
  - Preliminary joint DSA should be included in the Fund’s brief; preliminary DSA in briefs contains broad medium-to-long-term macro parameters and is subject to change.
- Presentation and content:
  - LIC DSAs should be self-contained documents including clear description of macroeconomic assumptions without referring to Fund staff report.
  - DSAs should be concise; suggested limit of 2000 words (excluding tables or appendices).
  - A new DSA must be available before end of second quarter of calendar year to be reflected in IDA allocation for coming fiscal year.
- Coordination and updates:
  - Each institution can update DSA for its own purposes if changes in assumptions are relatively minor; other institution must be notified and given minimum three business days to comment.
  - Major changes (e.g., non-concessional borrowing episodes) require consultation with the other institution.
- Review and clearance process:
  - Bank country teams transmit preliminary draft to PRMED for initial guidance; completed DSAs reviewed by PRMED and regional PREM director with three days for review.
  - Final DSA files should be submitted to SPR review box in Fund and to PRMED in Bank when DSA sent to Executive Boards.
  - If comments not received within agreed timeframe despite efforts, originating front office should contact commenting institution; failure to comment by agreed timeline implicitly waives right to comment and should be documented.

### Division of Responsibilities Between Fund and Bank Staff
- Fund leads on medium-term macroeconomic projections (three to five years) developed with the member country — starting point for baseline consultation.
- Bank leads on long-term growth prospects.
- External sector DSA responsibilities:
  - Fund: debt-service projections for bilateral and commercial creditors; assumptions on new borrowing from these creditors.
  - Bank: debt-service projections for multilateral creditors and assumptions on new multilateral borrowing based on current allocations.
- Teams should agree baseline assumptions, collaborate on alternative scenarios and stress tests, and share electronic data files confidentially.
- External DSA output together with Fund fiscal projections provide basis for public sector DSA; reconcile any differences in sustainability assessment.

### Dispute Resolution
- Seek common Bank-Fund assessment; resolve disagreements at working level first.
- If material differences remain:
  - Mission chief and Bank regional PREM director attempt agreement.
  - If unresolved, consult review departments (SPR in Fund, PRMED in Bank) and seek resolution within five working days.
  - If still unresolved, escalate to area department director at Fund and vice president at Bank with another five working days.
  - Failures to resolve brought to managements; managements can resolve within five working days or decide to present different staff views to Boards.

### V. Arrangements for HIPCs — Differences and Treatment
- HIPC DSA vs LIC DSA:
  - HIPC DSA: tool to calculate debt relief; uniform thresholds for PV debt-to-exports and PV debt-to-revenue; denominators derived from three-year backward-looking averages; predetermined currency-specific discount rates.
  - LIC DSA: forward-looking; uses single-year denominators; incorporates exchange rate projections and a uniform 5 percent discount rate; applies policy-dependent indicative thresholds.
- Application and consistency:
  - DSF to be applied to both HIPCs and non-HIPC LICs.
  - For HIPCs that have started process but not reached completion point, LIC DSA updated annually; selected HIPC indicators included as memorandum items.
  - When HIPC DSA and LIC DSA in same document, they must be based on consistent baseline macro assumptions and debt data.
  - Debt ratios and debt-service projections will differ due to different exchange rates and discount rates; write-up should decompose differences into (i) discount and exchange rates and (ii) exports (three-year averages versus current levels).
- Discount rate:
  - Discount rate currently set at 5 percent.
  - Adjusted by a full percentage point whenever the U.S. dollar CIRR (six-months average) deviates from prevailing discount rate by at least this amount for a consecutive period of six months.
- Treatment of HIPC Initiative and MDRI relief:
  - Accounted for in baseline or alternative scenarios depending on HIPC status with specified guidance for post-completion, interim period, and pre-decision-point countries.

### VI. Use, Dissemination, and Coordination (effectiveness of DSF)
- DSF as coordination and communication tool among borrowers and creditors, including emerging creditors.
- Country teams to involve relevant MDBs early with authorities’ consent and thoroughly discuss DSA assumptions and results with authorities.
- Final DSA templates should be provided to authorities; staff should encourage early consent to publication.
- Both Bank and Fund maintain dedicated websites on published DSAs and background material.
- Dissemination may require consent of third parties if DSA includes nonpublic third-party information.

### Template Design, Inputs, and Outputs — Practical Modalities
- Template overview:
  - Set up for a 20-year projection period.
  - Covers: (i) debt incurred externally by domestic residents (public and private); (ii) debt incurred by public sector (gross or net), including domestic public debt.
  - For each debt type includes baseline, sensitivity tests, output tables, and charts.
- Template components:
  - Language sheet, template navigator, two input sheets, four output tables, two output figures, multiple worksheets, two customized-scenario worksheets, and a summary of instructions.
- Navigator and inputs:
  - Navigator sections: INPUT, SCENARIOS, OUTPUT; “Return to Navigator” link present.
  - Two input sheets: “Data_input” and “Inp_Outp_debt”; required inputs shaded in yellow.
  - Required data: total stock of existing debt and new borrowing terms by main creditors; macroeconomic series for baseline; select template scale in cell E17.
  - Coverage decisions for public DSA must be documented in write-up; contingent liabilities reported separately when available.
- “Inp_Outp_debt” specifics:
  - Focuses on medium and long-term PPG debt; divided into terms of new borrowing, debt service on existing PPG external debt, amounts of new external PPG borrowing, and output stock/debt service/PV.
  - Template accommodates creditor disaggregation and different lending terms; terms of marginal public borrowing split by foreign-currency, domestic medium- and long-term, and domestic short-term.
- PV and discounting:
  - PV of public debt = PV of external public debt + nominal value of public domestic debt (assumes nominal interest rate equals discount rate for domestic debt).
  - PV measure emphasized given reliance on concessional financing.
  - Discount rate set at a uniform 5 percent; adjusted when U.S. dollar CIRR (6-month average) deviates by at least 100 basis points for 6 consecutive months.
  - Use WEO exchange rate assumptions for medium-term conversion.

### Debt-Service and PV Calculations; Outputs and Stress Tests
- Inputs required:
  - Debt-service projections on existing outstanding external PPG debt by main creditor groups over entire maturity period and projected disbursements by creditor over projection period.
- Outputs:
  - “Table Baseline External” reports nominal external debt-to-GDP evolution, decomposes changes into non-interest current account deficit, non-debt creating capital inflows (net FDI), endogenous debt dynamics, and a residual “unexplained change”.
  - Five external debt-burden indicators shown: PV of PPG external debt-to-GDP, PV-to-exports, PV-to-revenues, debt service-to-exports, debt service-to-revenues.
  - “Table Baseline Fiscal” decomposes public debt-to-GDP changes into primary balance, endogenous debt dynamics, and other debt-creating/reducing flows; reports PV debt-to-GDP, PV contingent liabilities, gross financing need, PV debt-to-revenue, debt-service-to-revenue, and primary deficit that stabilizes debt-to-GDP.
  - Large residuals flagged as potential data errors.
- Stress tests and sensitivity:
  - Template runs standardized sensitivity tests and produces panel charts showing baseline, historical scenario, and most extreme stress test.
  - Distinguishes two “alternative scenarios” and six “bound tests” for external debt; three “alternative scenarios” and five “bound tests” for public debt.
  - Alternative scenarios are mechanical, calibrated using averages and standard deviations over last 10 years; do not reflect full macro adjustment—users should develop consistent alternative projections if needed.
  - Bound tests provide upper-bound confidence-interval-like shocks (confidence interval corresponds on average to a 25 percent probability over a ten-year period).
- External debt alternative scenarios:
  - Historical Average Scenario (“A1_ Historical”): assumes key variables at historical averages throughout projection period.
  - Level of concessionality (“A2_ Financing”): assumes interest rate on new borrowing is 200 basis points higher than baseline.
- Public debt alternative scenarios:
  - Historical Average (“A1_ historical”).
  - Primary Balance Unchanged (“A2_ PB unchanged”): primary balance unchanged from last actual observation.
  - Lower long-run GDP growth (“A3_LR growth”): real GDP growth lower than baseline by one standard deviation divided by square root of projection period.
- External bound tests (main assumptions):
  - B1_GDP: real GDP growth one standard deviation below historical average in each of first two years.
  - B2_Exports: exports growth one standard deviation below historical average in each of first two years.
  - B3_Deflator: U.S. dollar GDP deflator one standard deviation below historical average in each of first two years.
  - B4_non-debt flows: net non-debt flows one standard deviation below historical average in first two years.
  - B5_Combo: combines all four variables at one-half standard deviation below historical average in first two years.
  - One-time depreciation test: one-time 30 percent depreciation of domestic currency in first year.
- Public bound tests (main assumptions):
  - B1_GDP: real GDP growth one standard deviation below historical average in first two years.
  - B2_PB: primary balance one standard deviation below historical average in first two years.
  - B3_combo: GDP growth and primary balance one-half standard deviation below historical average.
  - B4_depreciation: one-time 30 percent depreciation in first year.
  - B5_other flows: debt increases by 10 percent of GDP in first year due to other debt-creating flows.
- Notes on GDP shocks and fiscal assumptions:
  - GDP shocks assume revenues constant as share of GDP and expenditures constant in nominal terms; users may alter and must justify changes.
- Customized scenarios:
  - One customized scenario for external DSA and one for public DSA; status set to “ON” to include in output tables and charts.
  - External customizable variables include exports, imports, official/private transfers, net FDI, real GDP growth, GDP deflator; default shocks and shock profiles adjustable.
  - Public customizable variables include revenues and grants, primary expenditure, real GDP growth, inflation, nominal depreciation, public sector assets for net debt, terms and composition of new borrowing.

### Stress Test Outputs and Panel Chart (paragraph 58)
- “Stress Test External” shows sensitivity of the five key external debt-burden indicators to standardized shocks and alternative assumptions.
- “Stress Test Fiscal” shows PV debt-to-GDP, PV debt-to-revenue, and debt-service-to-revenue evolution similarly.
- Panel chart displays external and public debt evolution in baseline, historical scenario, and the most extreme stress test (most extreme defined as highest indicator after 10 years).
- For external debt, chart shows rate of debt accumulation and average grant element under baseline; customized scenarios also displayed when active.

*Source: IMF–World Bank joint guidance note on the debt sustainability framework for low-income countries (excerpts and annexes as provided).*

### References..............................................................................................................

### References

### I. Introduction and Country Coverage — Key Points
- Objective: Support LICs in achieving development goals without creating future debt problems; maintain sustainability for countries receiving HIPC and MDRI relief.
- Joint preparation: Country DSAs are prepared jointly by World Bank and IMF staffs with close collaboration on macroeconomic baseline, alternative scenarios, debt distress rating, and write-up.
- Frequency:
  - Joint Bank-Fund LIC DSAs are generally expected to be prepared once a year for PRGF-eligible, IDA-only countries (Table 1).
  - For PRGF-eligible countries that are not IDA-only, Fund staff is expected to produce a LIC DSA once a year, unless the country has significant access to market financing (in which case a middle-income template could be used).
- Role of DSF: Intended as an upstream device to inform country teams’ dialogue with authorities rather than an ex-post consistency check.
- Communication and publication:
  - Country teams should communicate frequently with MDBs, present DSA results to authorities, and share final DSA files with authorities.
  - Staff should encourage authorities to consent to publication of the DSA; with consent, present DSA findings to donors and other interested parties.

### Box 1 — Enhancements to the DSF (Main Changes from Previous Practice)
- Historical scenarios should be used actively; large differences between baseline and historical scenarios must be justified.
- Scrutinize past projections against outcomes to improve future projections.
- High projected growth dividends associated with large upfront borrowing (5 percent of GDP or more in PV terms) trigger:
  - Inclusion of an alternative “high-investment, low-growth” scenario.
  - Detailed and explicit justification of projected growth dividends.
- Financing assumptions require explicit justification when assuming a significant improvement in terms that materially affect debt evolution.
- Public DSAs should be included in all DSAs; write-ups must flag when inclusion of domestic debt changes interpretation of sustainability relative to external-only analysis.
- Additional analysis required where increased private external capital flows into sovereign debt instruments may create new vulnerabilities.
- Use a three-year moving average CPIA score to reduce volatility of thresholds and potential unwarranted fluctuations in IDA grant share.
- Build capacity and ownership: actively discuss DSA assumptions and outcomes with authorities and encourage use of the instrument.
- DSAs should be:
  - Published as supplements to Fund staff reports.
  - Self-contained in both Bank and Fund documents.
  - Easily accessible to enhance DSF effectiveness.
- Review process:
  - Preliminary DSA needs inclusion in IMF briefing papers to provide early feedback.
  - Bank-Fund collaboration should occur prior to preparation of briefing papers to ensure joint DSAs.
  - Similar timelines should be adhered to when DSAs are needed for Bank documents.

### II. LIC DSA Framework — Structure and Analytical Aspects
- Framework pillars:
  1. Standardized forward-looking analysis of debt and debt-service dynamics under baseline, alternative scenarios, and standardized stress tests (bound tests).
  2. Debt sustainability assessment based on indicative country-specific debt-burden thresholds that depend on policy and institutional quality.
  3. Recommendations on borrowing/lending strategy to limit debt-distress risk while maximizing resources for MDGs.

A. Debt and Debt-Service Projections and Indicators
- Templates and scope:
  - One mandatory pre-set template used for PPG and private external debt in the external DSA and for total public sector debt including domestic debt in the public sector DSA.
  - Template designed for a twenty-year projection period.
  - Uses a uniform discount rate to calculate PV of future external debt-service obligations.
  - Discount rate adjusted when the six-month average U.S. dollar CIRR deviates from the template rate by more than 100 basis points for six months or more.
  - Template automatically produces output tables for baseline dynamics and standardized alternative scenarios and stress tests.
- Coverage and definitions:
  - Public and publicly-guaranteed (PPG) debt includes debt owed by the central bank to the IMF, if any.
  - Private external debt is included in external DSA though not used for IDA grant allocations; inclusion can change overall sustainability assessment.
  - External debt is defined on a residency basis and may include domestic-currency denominated debt.
  - Analysis covers medium- and long-term debt; short-term debt affects financing need within the year.
- Scenarios and stress tests:
  - Bound tests are partial and assume passive fiscal policy.
  - Fiscal reaction functions may be included in alternative scenarios if deemed important; need separate modeling.

B. Country-Specific Debt-Burden Thresholds
- Policy dependence:
  - Debt-burden thresholds depend on Country Policy and Institutional Assessment (CPIA) index.
  - Countries divided into three performance categories: strong, medium, poor.
  - CPIA thresholds:
    - A rating at or above 3.75 corresponds to strong performance.
    - A rating between 3.25 and 3.75 reflects medium performance.
    - A rating at or below 3.25 corresponds to poor policy performance.
  - Use three-year moving average CPIA score to determine policy performance under the DSF to reduce undesirable uncertainty.
- Interpretation of indicators:
  - Debt stock indicators: PV of debt captures concessionality and summarizes future obligations (implications for solvency).
  - Debt-service indicators: measure immediate burden and potential liquidity problems.
  - Repayment capacity measured by GDP, exports of goods and services, or government revenues; relevance depends on country constraints.

- Table 2 (Debt Burden Thresholds under the DSF): (note: thresholds apply to external public debt; table headings indicate PV of debt in percent of Exports/GDP/Revenue and Debt service in percent of Exports/GDP/Revenue across policy categories — preserved structure from source)

C. Debt Distress Risk — Classification Guidance
- Every joint Fund-Bank and Fund-only DSA must include an explicit assessment of debt distress risk with four classifications:
  - Low risk:
    - All debt indicators well below country-specific debt-burden thresholds.
    - Stress testing and alternative scenarios do not result in significant breaches.
    - If only one indicator is above its benchmark, judgment required to determine cause (e.g., data problem).
  - Moderate risk:
    - Baseline does not indicate breach of thresholds, but alternative scenarios or stress tests cause significant rise in debt-service indicators (nearing thresholds) or breach thresholds.
  - High risk:
    - Baseline indicates a protracted breach of debt or debt-service thresholds but country currently does not face payment difficulties.
    - Alternative scenarios or stress tests exacerbate condition.
  - In debt distress:
    - Current debt and debt-service ratios are in significant or sustained breach of thresholds.
    - Existence of arrears generally suggests debt distress unless other reasons explain nonpayment.

### Operational and Implementation Notes
- Country coverage: Table 1 lists PRGF-eligible countries according to IDA status; last update: July 1, 2008.
- New financial environment challenges:
  - Post-debt-relief perception of large borrowing space in some LICs.
  - Emergence of new creditors and rising importance of domestic debt expanding volume and sources of funds.
  - Staff expected to:
    - Strengthen DSF application using precautionary aspects.
    - Design realistic baseline macroeconomic and growth scenarios.
    - Integrate domestic debt systematically into debt sustainability assessment.
    - Introduce additional vulnerability indicators where debt to private external creditors is significant.
- Process guidance:
  - Section III: DSA design and operational implications.
  - Section IV: Technical modalities including timing, information sharing, review, and clearance.
  - Section V: Arrangements for HIPCs.
  - Section VI: Communications strategy.
  - Annex I: User’s guide to templates.
  - Annex II: DSA outline template.

*Source: IMF–World Bank joint guidance note on the debt sustainability framework for low-income countries (excerpts: References, Annex, Boxes, Tables, and selected sections I–II).*

### 11.      The assessment of the risk of debt distress needs to strike a balance between a

### 11.      The assessment of the risk of debt distress needs to strike a balance between a

### Judgment versus mechanistic classification
- Mechanistic classification of debt distress risk must be balanced with judgmental adjustments in specific cases.
- Examples where judgment is warranted:
  - A marginal and temporary breach of thresholds.
  - An ability to pay evidenced by the level of foreign exchange reserves not captured in the template.
  - Lack of available CPIA scores (e.g., countries not active in IDA or newly re-engaging).
- When judgment is applied, it should be explained in the DSA write-up.
- CPIA-related guidance:
  - CPIA scores are averages of 16 indicators of policy and institutional quality; CPIA thresholds should not be used mechanically.
  - When moving to a three-year average causes undue volatility in performance ratings and potential loan-grant mix fluctuations, country teams may apply discretion on timing of the change, upon confirmation with reviewing departments (PRMED and SPR).

### Role and design of DSAs (DSA as upstream device)
- DSAs should:
  - Function upstream of Bank and Fund program design.
  - Inform broader dialogue with authorities.
  - Serve early detection of debt-related vulnerabilities.
  - Be a cornerstone for medium-term debt strategies, fiscal frameworks, and public expenditure planning in support of sustainable development goals (the third DSF pillar).
- DSAs need realistic macroeconomic scenarios; realism is promoted by:
  - Scrutinizing baseline projections via reality checks.
  - Using precautionary features of the DSF.

### Standard reality checks (to promote realism)
- Historical scenario comparisons:
  - Baseline debt ratios significantly lower than ratios under the historical scenario (historical 10-year averages) raise concerns of excessive optimism and require explicit justification.
  - Plausible reasons for deviations: recent performance improvements not in historical averages or structural breaks (e.g., end of civil conflict).
- Post-mortems on past projections:
  - “Post-mortems” explaining differences between assumptions and outcomes from previous DSAs are expected in the write-up.
  - If previous DSAs were overly optimistic, assumptions should be scrutinized and revised if previous forecast errors are not adequately accounted for.
- Financing assumptions:
  - Any notable improvement in financing terms that materially affects debt indicators requires explicit justification (e.g., already contracted concessional loans or firm donor commitments).
  - Review of the accuracy of past financing assumptions is recommended.
- Fiscal and structural shifts:
  - Explicit justification required if sustainability hinges on sharp shifts in fiscal policy, investment rate, financing mix, or productivity growth.

### Reality checks and precautionary features in scaling-up scenarios
- Scaling-up analytical challenge:
  - Projecting impact of additional public investment on GDP growth, exports, and public revenues is complex and requires careful justification.
- Guidance for teams:
  - Justify underlying assumptions carefully; use country history, analytical considerations, and empirical cross-country work (see referenced Box 2 and Appendix 3).
- Empirical literature general conclusions:
  - Prolonged growth accelerations are rare.
  - Macro returns on individual high-return projects tend to be considerably lower than microeconomic returns.
  - Quality of policies and institutions strongly influences macroeconomic return of public investment.
  - Economic volatility, including aid volatility and shocks, argues for caution in average growth/export projections.
- Special scrutiny where ambitious borrowing implies high projected growth dividends:
  - Mandatory inclusion of an alternative “high-investment, low-growth” scenario if the baseline assumes growth rates of at least one standard deviation above the historical average or if changing growth alone to historical levels implies a significantly worse debt outlook.
  - Baseline must be supported by compelling evidence that assumed growth dividends are very likely to materialize; absent such evidence, baseline should be revised.
  - Detailed justification required if baseline includes very large upfront borrowing. Large upfront borrowing defined as an annual increase in the PV of public external or total public debt of 5 percent of GDP or more.

### Box 2 — Indicators for analysis of the link between debt-financed investment and growth (selected indicators)
- Rates of Return:
  - Microeconomic studies on rates of return of projects.
  - Implementation lags/gaps for investment and recurrent budgets.
  - Estimates of stocks and shortfalls in public capital.
  - Composition of public expenditures in terms of growth impact.
- Structural Constraints:
  - Policy and institutional constraints as indicated by the CPIA, public governance indicators, Doing Business surveys, PEFA, other public expenditure management analyses.
  - Level and growth rates of public investment.
  - Completion or implementation rate of public investment projects.
  - Skill shortages that can only be alleviated in the long run.
- Macroeconomic Constraints:
  - Cost of capital (firm-level surveys, real interest rates).
  - Rate (or rate of growth) of private investment.
  - Excess reserves/lending capacity in banking system.
  - Various real exchange rate measures (unit labor costs, export market share).
- Aggregate Trends:
  - Growth rate of per capita GDP.
  - Growth rate of TFP.
  - Results of “binding constraints to growth” analyses.

### Treatment of domestic debt
- Requirement:
  - Regardless of size, all LIC DSAs must include a public DSA.
- Rationale:
  - Public domestic debt typically involves higher costs and shorter maturities; rising domestic debt increases likelihood of external debt distress.
- Coverage and definition considerations:
  - Public domestic debt defined on a residency basis and may include foreign currency-denominated obligations.
  - Data should seek to cover liabilities of the broader public sector (central government, local governments, government-owned enterprises, central bank).
  - Data limitations may restrict coverage to central or general government initially.
  - Public sector contingent liabilities (including PPPs and financial sector weaknesses) should be taken into account when possible; staff should flag coverage problems and steps to improve coverage in the DSA write-up.
- Staff assessment focus (guided by stress tests and scenarios):
  - Domestic debt risks: thorough review where domestic debt stocks are significant (i.e., above 15-20 percent of GDP). Any rapid recent build-up warrants explanation. Assessment should cover creditor base, likely duration, financing burden, and medium-term implications.
  - Primary balance: public DSA should assess whether projected primary fiscal balance evolution is consistent with debt sustainability and assess risks posed by baseline primary fiscal deficit path.
  - Debt distress classification: inclusion of domestic debt and debt service may change overall fiscal sustainability assessment and should be discussed in expanded commentary, but does not affect the country’s external debt distress classification or IDA grant allocation. The risk of debt distress rating is guided only by results of the external DSA relative to the thresholds.

### Treatment of debt held by private external creditors
- New vulnerabilities from increased private sector capital flows:
  - Abrupt reversals in market sentiment causing sudden capital outflows.
  - Non-standard financing terms (e.g., collateralization with future export receipts) weakening medium-term sustainability.
  - Secondary balance sheet effects on the domestic financial system potentially creating contingent public liabilities.
- Complementary analyses for significant private borrowing:
  - DSF should be complemented with analyses of short-term debt-related vulnerabilities and financial sector soundness.
  - Additional indicators (subject to data availability) can highlight:
    - Risks to sovereign liquidity from debt composition and maturity structure.
    - External liquidity and rollover risks and adequacy of reserve cover (especially for short-term debt and risk of market reversals).
    - Weaknesses in the financial sector that could create contingent public liabilities.
  - Where significant, these factors should be explained and taken into account in sustainability assessment.
- Policy discussion with authorities:
  - Country teams should discuss policies to alleviate these risks. Illustrative measures include:
    - Assign legal authority to borrow and identify permissible instruments and accountability mechanisms.
    - Facilitate portfolio management through effective recording of debt stock, liquidity forecasting framework, and availability of indicators to monitor benefits, costs, and risks from private borrowing.
    - Potential need for technical assistance.
    - Re-assessment of reserve adequacy.
    - Sequencing reforms to strengthen banking supervision and prudential regulation before liberalizing the capital account.

### Suggested indicators for vulnerability analysis (high-level)
- Public sector solvency indicators:
  - PV of public sector debt-to-GDP (public sector revenue).
  - PV of external public sector debt-to-GDP (exports).
  - PV of foreign-currency denominated public sector debt-to-GDP.
  - PV of contingent liabilities (not included in public sector debt).
  - Public sector debt-to-GDP and its breakdowns (External; Foreign currency denominated; Foreign currency linked; Indexed to the CPI).
  - Primary deficit that stabilizes public sector debt-to-GDP.
- External sector solvency indicators:
  - PV of external debt-to-GDP (exports).
  - External debt-to-GDP.
  - Non-interest external current account deficit that stabilizes external debt-to-GDP.
- Public sector flow (liquidity/rollover) indicators:
  - Public sector debt service-to-revenue.
  - External public debt service-to-exports.
  - Public sector gross financing need (in percent of GDP).
  - Short-term public debt-to-total debt (at remaining maturity).
  - Domestically-issued public debt held by nonresidents-to-GDP.
- External sector flow (liquidity/rollover) indicators:
  - External debt service-to-exports (revenue).
  - External gross financing need (billions of U.S. dollars).
  - Gross official reserves-to-short-term external debt (at remaining maturity).
  - Extended reserve cover.
  - Gross official reserves-to-broad money (M2).
  - Foreign currency deposits-to-foreign assets of the banking system.
- Financial system soundness indicators:
  - Regulatory capital-to-risk-weighted assets.
  - Nonperforming loans-to-total loans (gross and net of provisions).
  - Claims on the Government and Central Bank-to-total banking sector claims.
  - Private sector credit growth.
  - Foreign currency loans-to-total loans.
  - Foreign currency deposits-to-total banking sector deposits.
  - Share of foreign currency deposits held by nonresidents.

*Source: _100608 - 11.      The assessment of the risk of debt distress needs to strike a balance between a*

### 26.      LIC DSAs play a pivotal role in determining terms of IDA assistance and

### 26.      LIC DSAs play a pivotal role in determining terms of IDA assistance and

### Role of LIC DSAs in IDA allocations and Fund program design
- LIC DSAs form the basis for determining the grant/loan mix in IDA allocations and those of some other multilaterals, including the African Development Fund.
- IDA-only countries judged to be at high risk of debt distress risk or in debt distress receive 100 percent grant financing from IDA.
- Countries at moderate risk receive a 50/50 blend of grants and traditional credits.
- Countries at low risk continue to receive 100 percent credit financing on standard IDA terms.
- Because DSA risk ratings determine IDA grant allocations, regular IDA credit terms on all IDA lending should be assumed for all years in the projection period for which grant finance has not already been committed to by IDA.
- The same assumption on credit terms applies to other major MDBs who link terms of assistance to the DSF risk rating.
- For the Fund, programs for LICs in which debt sustainability is a concern could:
  - supplement existing debt limits with conditionality related to the PV of external debt,
  - make more systematic use of limits on the overall fiscal deficit as compared to existing limits on domestic financing.
- Conditionality on the PV of external debt would be derived from the DSA on a fiscal year basis and expected to be in the form of indicative targets due to complexity and monitoring difficulties.
- It is presumed that the recommended grant element for concessional loans, including in Fund-supported programs, would increase with the risk of debt distress.
- The DSA should be the primary means of assessing the impact of alternative financing strategies and recommending the minimum concessionality for new lending.
- When the debt distress risk classification shifts to a higher level, staff should conduct a comprehensive reassessment of the recommended debt accumulation strategy.
- DSAs can be used to analyze the potential impact of unexpected borrowing, especially in the context of IDA’s nonconcessional borrowing policy.

### IV. MODALITIES FOR PREPARING DSAs — A. Frequency and Presentation
- A joint LIC DSA is expected to be prepared once a year for each IDA-only, PRGF-eligible country.
- Each calendar year, Fund and Bank country teams need to agree on a schedule for the preparation of DSAs for individual countries.
- For the Fund, a DSA will normally be produced for an Article IV consultation, and otherwise in the context of program requests or reviews.
- For the Bank, the DSA will be required for Country Assistance Strategies and for IDA allocation purposes.
- In cases where a joint DSA is needed but the Fund’s Board is not expected to consider a review or conclude an Article IV consultation within two months, the joint DSA would be sent to the Fund’s Board for information at the same time the DSA is sent to the Bank’s Board; the corresponding procedure applies when Fund requirements drive timing.
- Each institution can update the DSA for its own purposes if changes in assumptions are relatively minor; the other institution must be notified and given at minimum three business days to comment.
- Major changes (e.g., due to non-concessional borrowing episodes) require consultation with the other institution.
- LIC DSAs should be prepared as self-contained documents, including a clear description of macroeconomic assumptions without referring to the Fund staff report.
- DSAs should be concise; a limit of 2000 words is suggested (excluding tables or appendices).
- The Bank publishes LIC DSAs on a stand-alone basis and sends all LIC DSAs to its Board for information in this format.
- A new DSA would need to be available before the end of the second quarter of the calendar year in order to be reflected in the IDA allocation for the coming fiscal year.
- As of FY08 in the Bank, for all IDA-only countries the LIC DSA Board document will become a formal deliverable with associated code and budget norm range.

### DSA process (summary of timeline and coordination)
- Bank and Fund country teams should agree on schedule each calendar year and seek agreement on broad DSA parameters at least 90 business days ahead of the prospective Board meeting.
- A preliminary joint DSA should be included in the Fund’s brief, incorporating available new information and revised assumptions consistent with the Fund’s latest macroeconomic framework.
- The preliminary DSA in briefs contains the broad parameters of a medium-to-long-term macro framework and is subject to change depending on mission findings.
- No Bank clearance of the preliminary DSA in briefs is required but significant differences should be reported.

### B. Division of Responsibilities between Fund and Bank Staff
- Fund leads on medium-term macroeconomic projections (three to five years) developed with the member country; this is the starting point for consultation with the Bank on the baseline scenario for the DSA.
- Bank leads on long-term growth prospects.
- For the external sector DSA:
  - Fund staff responsible for debt-service projections for bilateral and commercial creditors and assumptions on new borrowing from these creditors.
  - Bank staff provides debt-service projections for multilateral creditors together with assumptions on new multilateral borrowing based on current allocations.
- Fund and Bank country teams should agree on baseline assumptions, collaborate on alternative scenarios and stress tests, and consider additional country-specific factors.
- All relevant data files should be shared across Bank and Fund teams but treated confidentially.
- The external sector DSA output, together with Fund fiscal projections, provide the basis for the public sector DSA; teams should discuss and reconcile any differences in sustainability assessment between external and public sector DSAs.

### C. Dispute Resolution
- Common Bank-Fund assessment should be sought in most cases; where disagreements occur, teams should first seek resolution at working level.
- Working-level country economists should discuss differences and determine whether viewpoints lead to a material difference in risk classification; if not material, accommodate differences.
- If material, the Fund mission chief and the Bank’s regional PREM director should attempt to reach agreement.
- The mission chief and regional director should, after consultation with review departments (SPR in Fund, PRMED in Bank), seek resolution within five working days; if unsuccessful, escalate to area department director at the Fund and vice president at the Bank with another five working days to resolve.
- Failures to resolve at that level will be brought to managements of the two institutions.
- Managements can, within five working days, resolve the dispute or decide the DSA document will present different staff views to the Boards; in the latter case each institution presents its views in its own words.

### D. Review Process
- LIC DSA documents prepared by Fund and Bank staff are subject to the regular review process; timing details are given in Box 3.
- Any substantive changes by Fund management will be communicated to Bank staff.
- Bank country teams should transmit a preliminary draft to PRMED for initial guidance; completed DSAs reviewed on a stand-alone basis by PRMED and the regional PREM director, who will have three days for review.
- Once clearance comments are incorporated, PRMED sends the DSA as a stand-alone document via PRMVP to SECBO for transmission to the Board for information.
- Each institution is expected to abide by the agreed timeline; any major disagreement should be communicated immediately.
- If comments are not received within the agreed timeframe despite efforts, the originating institution’s front office should contact the commenting institution; failure to comment by the agreed timeline implicitly waives the right to comment and should be documented and brought to management attention.
- Final versions of the DSA files (external and fiscal templates) should be submitted to the SPR review box in the Fund and to PRMED in the Bank when the DSA (and staff report) is sent to the Fund’s Executive Board or respective Executive Boards.
- To avoid discrepancies between published tables and electronic files, all electronic links to external files should be broken. The files should also be sent to the authorities.

### V. ARRANGEMENTS FOR HIPCS
- There are important conceptual and methodological differences between HIPC Initiative DSAs and LIC DSAs:
  - HIPC DSA: tool to calculate debt relief under HIPC Initiative; thresholds for PV of debt-to-exports and PV of debt-to-revenue ratios are uniform; denominators derived from three-year backward-looking averages; predetermined currency-specific discount rates used to calculate PVs.
  - LIC DSA: forward-looking; uses single-year denominators; incorporates exchange rate projections and a uniform 5 percent discount rate; applies policy-dependent indicative thresholds.
- The DSF should be applied to both HIPCs and non-HIPC low-income countries.
- For HIPCs that have started the HIPC process but not reached completion point, arrangements include:
  - DSF remains main tool and LIC DSA should be updated annually.
  - Selected debt indicators drawn from the HIPC DSA should be included in LIC DSA tables as memorandum items (debt-service and debt-stock ratios).
  - Decision and completion point documents should contain a LIC DSA as a supplement for forward-looking analysis.
- When HIPC DSA and LIC DSA are in the same document, they must be based on consistent baseline macroeconomic assumptions and debt data; baseline scenarios including assumptions on new borrowing should generally be identical.
- Debt ratios and debt-service projections will differ between HIPC and LIC DSAs because of different exchange rates and discount rates; write-up should decompose differences into components attributable to: (i) different discount and exchange rates, and (ii) different exports (three-year averages versus current levels).
- The discount rate is currently set at 5 percent. It will be adjusted by a full percentage point, whenever the U.S. dollar CIRR (six-months average) deviates from the prevailing discount rate by at least this amount for a consecutive period of six months.
- HIPC Initiative and MDRI debt relief should be accounted for in the baseline or alternative scenario, depending on HIPC status.
- LIC DSA baseline and alternative scenario treatment:
  - Post-completion point countries: LIC DSA should incorporate HIPC Initiative and MDRI relief in the baseline scenario; debt-service projections should take into account mechanisms under which relief is delivered (e.g., debt forgiveness or rescheduling).
  - Countries in the interim period: baseline should assume HIPC interim relief (risk rating should not be predicated on reaching completion point). Alternative scenario should assume irrevocable HIPC and MDRI relief beyond expected completion point date; PV indicators affected by relief only beyond expected completion point date; in years preceding expected completion, PV based on debt-service projections before completion point debt relief.
  - Countries not yet at decision point but with reviewed HIPC preliminary document: baseline should incorporate only traditional debt relief; alternative should incorporate HIPC relief assumed delivered through debt rescheduling beyond assumed decision point date, noting estimates of value of debt relief will only be approximate because actual HIPC debt reduction factor depends on the decision point date.

*Source: IMF/World Bank joint guidance on LIC DSAs (text as provided).*

### 40.      The effectiveness of the DSF ultimately depends on its broader use by

### The effectiveness of the DSF ultimately depends on its broader use by borrowers and creditors.

### Use, dissemination, and coordination
- The DSF should be seen as a tool for better communication and coordination between creditors and borrowers, and among creditors, including emerging creditors with limited tradition of regular coordination and information sharing.
- Country teams are to involve relevant MDBs, as appropriate and with the consent of the authorities where needed, in the early phases of the DSA process.
- DSA assumptions and results should be thoroughly discussed with the authorities (preferably early in the Fund mission in cases where the DSA is being produced for an IMF Board document).
- The final DSA templates should be provided to the authorities.
- The DSF, combined with technical assistance, should help build capacity in public debt management; over time borrowers can develop their own medium-term debt strategy to support development objectives while containing risks of debt distress and macroeconomic vulnerability.
- Following the Board meeting, country teams, in consultation with the authorities, are encouraged to disseminate DSA results, if and when the authorities consent.
- Where possible, staff should encourage the authorities to consent as early as possible to the publication of the DSA (for the Fund, the staff report to which the DSA is supplemented).
- Both the Bank and the Fund maintain a dedicated website on published DSAs and relevant background material (http://www.imf.org/dsa and http://worldbank.org/debt).
- Consent considerations:
  - Dissemination may require consent of third parties to the extent that the DSA contains nonpublic third party information; in practice this may arise only rarely because third party information is generally processed and aggregated.

### DSF role with MDBs and donors
- MDBs that would typically be included: AfDB, AsDB, IDB, and EBRD.
- Information sharing is subject to the Fund’s policy on sharing confidential information and may require consent of the authorities or third parties.
- Dissemination formats suggested include presentations in the context of regular contacts with the donor/creditor community (e.g., consultative group meetings), including emerging creditors.

### LIC DSA: purpose and scope
- Assessing fiscal and external sustainability underpins Article IV surveillance and the use of Fund resources, and informs IDA’s allocation of grants and the Bank’s dialogue with governments on economic management.
- The LIC DSA has been used since its inception in April 2005.
- The template provides practical guidance to country teams and country authorities on conducting external and fiscal sustainability analyses in low-income countries.

### Template design and projection horizon
- The template is set up for a 20-year projection period in light of the long maturity of concessional debt.
- The template covers two types of debt: (i) debt incurred externally by domestic residents (both public and private sectors); and (ii) debt incurred by the public sector (either in gross or net terms), including domestic public debt.
- For each type of debt (external or public) the template includes:
  - a baseline scenario,
  - a set of sensitivity tests,
  - output tables,
  - and a set of charts summarizing DSA results.

### Template components (high level)
- The template consists of:
  - (i) a language sheet,
  - (ii) a template navigator,
  - (iii) two input sheets,
  - (iv) four output tables—two for each type of debt,
  - (v) two output figures,
  - (vi) a range of worksheets that transform input data into output tables,
  - (vii) two worksheets that allow for customized scenarios for each type of debt,
  - (viii) a summary of the instructions laid out in more detail in this Annex.

### Navigator functionality
- Main sections in the navigator:
  - INPUT: Data Input, Input-Output Debt and Approach to Public Debt with corresponding drop-down menus.
  - SCENARIOS: boxes to open and navigate through scenario worksheets for each type of debt; users click the blue arrow next to “Open fiscal all” or “Open external all” to display boxes; close unused approach by clicking the red arrow next to “Close fiscal all” or “Close external all”.
  - OUTPUT: drop-down menu to navigate output tables and charts and a link to the Worksheet “Output database”.
- To return to the navigator from any worksheet, click the link “Return to Navigator” in the top left corner.

### Input worksheets: data requirements and conventions
- Two input sheets: “Data_input” and “Inp_Outp_debt”.
- Required inputs are the cells shaded in yellow; non-shaded cells are formulas calculated automatically.
- Data required include:
  - total stock of existing debt and new borrowing terms by main creditors (loan-by-loan data not required);
  - macroeconomic series for baseline: stock of total external and public debt, associated debt service (including on new borrowing), exports, imports, current transfers, revenue, expenditure, grants, nominal GDP, GDP deflator, etc.
- Before working on the data table, the scale for the template must be selected in cell E17.
- Historical data requirements: public and publicly guaranteed external debt (stock and debt service) and concessional loans.
- In formulating the baseline scenario for public DSA:
  - determine coverage of the public sector (e.g., central government, general government, nonfinancial public sector),
  - choose debt concept (net or gross),
  - note coverage and type of debt in the DSA write-up,
  - ensure consistency of coverage across fiscal series.
- Contingent liabilities should be reported as a separate item (if available) rather than as part of the debt stock.
- If public debt is on net terms, report public sector assets as a separate line.
- To ensure consistency with IMF BOP files, series from BOP files should be entered only on the External Debt disaggregation and are automatically calculated on the Public Debt disaggregation.

### “Inp_Outp_debt” sheet specifics
- Focuses only on medium and long-term PPG debt; coverage differs from the “Data-input” debt-service concept (which includes external debt service on private sector debt, public debt service on domestic debt, and short-term debt).
- Divided into four parts:
  - (i) terms of new external and public borrowing;
  - (ii) debt service on existing PPG external debt;
  - (iii) amounts of new external PPG borrowing;
  - (iv) output – stock, debt service and PV of external PPG debt.
- For external debt:
  - template accommodates disaggregation of creditors and allows different terms;
  - users enter creditor names in the “Descriptor” cell and lending terms in adjacent cells; these are used to calculate PV of new disbursements.
- For marginal public borrowing (stress tests):
  - allow division between foreign-currency borrowing, domestic medium- and long-term borrowing, and domestic short-term borrowing;
  - specify interest rate and maturity structure for each type of marginal borrowing;
  - interest rates specified in nominal terms for foreign-currency borrowing (assumed to be in U.S. dollars) and in real terms for domestic borrowing.
- PV of public debt is calculated as:
  - PV of external public debt plus the nominal value of public domestic debt (assumption: nominal interest rate equals discount rate for domestic debt).
- Given reliance on concessional financing in low-income countries, the PV of debt is a more informative measure of effective debt burden.

### Debt-service and PV calculations
- Users must enter debt-service projections on existing outstanding external PPG debt by main creditor groups over the entire maturity period (until existing claims are paid off) and projected disbursements by creditor over the projection period.
- For debt-service ratios, only debt service on existing and new (calculated in template) PPG external debt is needed.
- PV of PPG external debt is calculated based on projected debt service and new disbursements using terms of new borrowing.
- Terms of additional external financing from stress tests are set to coincide with average terms assumed in the baseline, but may be adjusted if warranted.

### Discount rate and exchange-rate conversion
- The discount rate has been set at a uniform 5 percent, consistent with the proposal in IMF and IDA, 2004, and should not be altered.
- The discount rate will be adjusted whenever it deviates from the U.S. dollar CIRR (6-month average) by at least 100 basis points for a consecutive period of 6 months.
- Any changes to the discount rate will be reflected in the most recent version of the template.
- Use of one discount rate for all external loans implies need for explicit exchange-rate projections to convert debt service on existing debt into U.S. dollars; medium-term conversion should use WEO exchange rate assumptions.

### Outputs: stress tests, tables, and indicators
- Once inputs are populated, the template automatically runs stress tests and produces output tables and panel charts.
- “Table Baseline External” reports evolution of nominal external debt-to-GDP ratio in the baseline and decomposes changes into:
  - (i) non-interest current account deficit and its basic breakdown,
  - (ii) non-debt creating capital inflows (net FDI),
  - (iii) endogenous debt dynamics (calculated using GDP growth, interest rates, and price and exchange rate movements; these movements are not shown for the projection period in line with Fund convention).
- The unexplained change in debt is included in a residual.
- “Table Baseline External” also presents five external debt-burden indicators:
  - PV of PPG external debt relative to GDP,
  - PV of PPG external debt relative to exports,
  - PV of PPG external debt relative to revenues,
  - debt service on PPG external debt relative to exports,
  - debt service on PPG external debt relative to revenues.
- The table shows key macroeconomic assumptions underlying the baseline scenario, how they compare with historical averages, and the rate of debt accumulation—which would trigger more thorough analysis if in excess of 5 percent.
- “Table Baseline Fiscal” decomposes changes in public debt-to-GDP ratio into:
  - (i) the primary balance and its basic breakdown,
  - (ii) endogenous debt dynamics (using GDP growth, interest rates, and currency appreciation/depreciation; currency movements not shown for projection period in line with Fund convention),
  - (iii) other debt-creating/reducing flows including privatization receipts, asset purchases (when debt is gross), debt relief, or recognition of contingent liabilities.
- The unexplained change in debt is reported as the residual; a very large residual may indicate data errors and should be further analyzed.
- “Table Baseline Fiscal” presents additional debt burden indicators:
  - PV of debt-to-GDP ratio,
  - PV of contingent liabilities,
  - gross financing need,
  - PV of debt-to-revenue ratio,
  - debt-service-to-revenue ratio,
  - primary deficit that stabilizes the debt-to-GDP ratio (defined as the difference between the actual primary deficit and the actual change in the public debt-to-GDP ratio).
  - Notes on primary deficit indicator:
    - a negative number corresponds to a primary surplus;
    - the estimated primary deficit stabilizes the PV ratio in the year in question assuming previous years followed the baseline path.

*Source: Excerpt from IMF user guide and annex on the Debt Sustainability Framework and LIC DSA templates.*

### 58.      The table “Stress Test External” shows the sensitivity of the five key external debt-

### _100608 - 58.      The table “Stress Test External” shows the sensitivity of the five key external debt-

### Overview
- The table “Stress Test External” shows the sensitivity of the five key external debt-burden indicators to standardized shocks and alternative assumptions.
- The table “Stress Test Fiscal” shows the evolution of the PV of debt-to-GDP, PV of debt-to-revenue and debt-service-to-revenue ratios in the same fashion.
- The template produces a panel chart (Worksheet “Panel Chart”) showing external and public debt evolution in the baseline, the historical scenario, and the most extreme stress test.37

### Panel chart details
- For external debt, the rate of debt accumulation and the average grant element under the baseline is shown.
- If the user designs a customized scenario, the resulting debt dynamics will also be shown in the charts.
- Footnote: The most extreme stress test is defined as the test that results in the highest indicator after 10 years of projections.

### Sensitivity analyses (general)
- The template includes standardized sensitivity tests to assess robustness of sustainability indicators to changes in key assumptions and parameters.
- Distinguishes between:
  - Two “alternative scenarios” and six “bound tests” for external debt.
  - Three “alternative scenarios” and five “bound tests” for public debt.
- Debt dynamics under alternative scenarios and bound tests are derived in separate worksheets and summarized in output tables and the panel chart.
- Customized scenarios can be constructed for each approach to debt sustainability.

### Alternative scenarios — calibration and purpose
- Standard stress tests are calibrated using the average and standard deviations over the last 10 years of history.
- Alternative scenarios present mechanical responses in debt-burden indicators due to changes in critical variables; they:
  - Do not reflect a comprehensive and consistent alternative macroeconomic framework.
  - Ignore joint second round dynamic responses of macroeconomic variables and potential policy responses.
- To reflect adjustment paths or policy responses, users should develop an alternative baseline projection consistent with the scenario and rerun the template, or use the customized scenario.

### Alternative scenarios — external debt
- Historical Average Scenario (Worksheet “A1_ Historical”):
  - Assumes key variables are at their respective historical averages throughout the projection period.
  - Provides indications about the extent of optimism in the baseline relative to the country’s historical performance.
- Level of concessionality (Worksheet “A2_ Financing”):
  - Assumes that the interest rate on new borrowing is 200 basis points higher than in the baseline scenario.

### Alternative scenarios — public debt
- Historical Average Scenario (Worksheet “A1_ historical”):
  - Assumes key variables are at historical averages throughout the projection period; gauges baseline optimism relative to history.
- Primary Balance Unchanged (Worksheet “A2_ PB unchanged”):
  - Assumes the primary balance is unchanged from the last actual observation to replicate a “status quo”.
- Lower long-run GDP growth (Worksheet “A3_LR growth”):
  - Assumes real GDP growth in all future years is lower than under the baseline by one standard deviation divided by the square root of the projection period.
  - Intended to illustrate effects of persistently lower-than-projected growth.

### Bound tests — purpose and adjustability
- Bound tests are standardized tests akin to providing the upper bound of a confidence interval to the baseline projections.
- Sensitivity-test parameters can be adjusted for country-specific circumstances (e.g., choice of calibration period, exclusion of non-representative events).
- Changes to parameters should be made in worksheets “Baseline” and “Baseline-fiscal” and justified in the write-up.
- Note: The confidence interval corresponds, on average, to a 25 percent probability over a ten-year period.

### Bound tests — external debt (main assumptions)
- B1_GDP: real GDP growth is one standard deviation below its historical average in each of the first two years.
- B2_ Exports: exports growth is one standard deviation below its historical average in each of the first two years.
- B3_Deflator: inflation, measured by the increase in the U.S. dollar GDP deflator, is one standard deviation below its historical average in each of the first two years.
- B4_non- debt flows: net non-debt flows, including both FDI and current transfers, are one standard deviation below their historical average in each of the first two years.
- B5_Combo: combines all four variables and assumes they are half a standard deviation below their historical average in each of the first two years.
- One-time depreciation test: assumes a one-time 30 percent depreciation of the domestic currency in the first year of the projection period.

### Bound tests — public debt (main assumptions)
- B1_GDP: real GDP growth is one standard deviation below the historical average in the first two years of projection.
- B2_ PB: the primary balance is one standard deviation below the historical average in the first two years of projection.
- B3_combo: combines shocks to GDP growth and the primary balance, assuming they are one-half standard deviation below the historical average.
- B4_depreciation: assumes a one-time 30 percent depreciation of the domestic currency (relative to the baseline) in the first year of projection.
- B5_other flows: assumes that debt increases by 10 percent of GDP in the first year of projection due to other debt-creating flows (e.g., bank recapitalization or recognition of contingent liabilities).

### Notes on GDP shocks and fiscal assumptions
- Shocks to GDP growth assume revenues stay constant as a share of GDP while expenditures stay constant in nominal terms.
- Consequently, GDP shocks increase (decrease) the primary deficit (surplus), which can produce large effects.
- Users may alter these assumptions if unreasonable for the country, but must note and justify such changes.

### Customized scenarios — general
- Facilitates design of country-specific analysis (commodity prices, terms of new borrowing, path of primary balance, etc.).
- One customized scenario exists for external DSA and one for public DSA.
- Top of each “Customized Scenario” worksheet contains user instructions.
- To display results in output tables and charts, set the Customized Scenario’s status to “ON” by clicking the blue box at the top right; switch “OFF” by clicking the red box.

### Customized scenario — external debt
- Modifiable variables displayed in rows 8 to 19:
  - Exports, imports, official and private current transfers, net FDI, real GDP growth, GDP deflator.
- Standardized table on debt dynamics reads from rows 8 to 19.
- A check line for the current account deficit is included in line 12 to ensure overall consistency.
- Default shock to exports or imports accounts for commodity price changes for net exporter or net importer countries; shock intensity and persistence can be customized via the shock profile line (see worksheet instructions).
- Terms of new borrowing can be adjusted by modifying average interest rate, maturity, and grace period in cells C23:C26.

### Customized scenario — public debt
- Modifiable variables shown in rows 8 to 15:
  - Revenues and grants, primary expenditure, real GDP growth, inflation, nominal depreciation of the exchange rate.
- Standardized table on debt dynamics is fed directly from these rows.
- A check line for the primary balance is included in line 12.
- For cases where net public debt is more appropriate, public sector assets size and path can be modified in line 11.
- Terms of new borrowing can be adjusted in cells A25:A28 and A33:A35 (average interest rate, maturity, grace period).
- Composition of new public borrowing (external, MLT domestic, short term domestic) can be modified in cells A20:A22.

*Document: _100608 - 58. The table “Stress Test External” shows the sensitivity of the five key external debt-*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2008/_100608.pdf_