Debt Vulnerabilities in IDA Countries October 4, 2018 ACRONYMS AND ABBREVIATIONS Fiscal year (FY) = July 1 to June 30 AfDB African Development Bank IEG Independent Evaluation Group ADF African Development Fund IFC International Finance Corporation AFR Sub-Saharan Africa IMF International Monetary Fund ASA Advisory Services and Analytics IT Information Technology Cat DDO Catastrophe Deferred Draw- LCR Latin America and Caribbean Down Option LIC Low-Income Country CPF Country Partnership Framework LIC-DSA Debt Sustainability Analysis for CPIA Country Policy and Institutional Low-Income Countries Assessment LIDC Low-Income Developing Country DAC Development Assistance LMIC Lower Middle-Income Country Committee MAC-DSA Debt Sustainability Analysis for DeMPA Debt Management Performance Market-Access Countries Assessment MDB Multilateral Development Banks DLP Debt Limits Policy MDRI Multilateral Debt Relief Initiative DMF Debt Management Facility MNA Middle East and North Africa DMO Debt Management Office MTDS Medium-Term Debt Strategy DOD Debt Outstanding and Disbursed NCB Non-Concessional Borrowing DRC Democratic Republic of Congo NCBP Non-Concessional Borrowing DRF Debt Reduction Facility Policy DRM Domestic Revenue Mobilization ODA Official Development Assistance DRS Debt Reporting System OECD Organization for Economic DSA Debt Sustainability Analysis Cooperation and Development DSF Debt Sustainability Framework PBA Performance-Based Allocation EAP East Asia and Pacific PER Public Expenditure Review ECA Europe and Central Asia PFM Public Financial Management EIB European Investment Bank PFRAM Public-Private Partnership-Fiscal ESW Economic and Sector Work Risk Assessment Model FCS Fragile and Conflict-Affected PIM Public Investment Management States PNG Private Nonguaranteed FCV Fragility, Conflict and Violence PPG Public and Publicly Guaranteed FSAP Financial Sector Assessment PPP Public-Private Partnership Program SAR South Asia GAF Grant Allocation Framework SCD Systematic Country Diagnostic GDP Gross Domestic Product SDGs Sustainable Development Goals GNI Gross National Income SOE State-Owned Enterprise G-20 Group of Twenty SUF Scale-up Facility HIPC Heavily Indebted Poor Countries TA Technical Assistance IBRD International Bank for UMIC Upper Middle-Income Country Reconstruction and Development WB World Bank IDA International Development WBG World Bank Group Association WDI World Development Indicators IDS International Debt Statistics WEO World Economic Outlook TABLE OF CONTENTS EXECUTIVE SUMMARY ........................................................................................................................... i I. INTRODUCTION ................................................................................................................................ 1 II. DEBT VULNERABILITIES IN IDA COUNTRIES ........................................................................... 1 III. ACTIONS TO HELP ADDRESS DEBT-RELATED RISKS............................................................ 18 IV. IDA POLICY OPTIONS .................................................................................................................... 20 V. CONCLUSION AND ISSUES FOR DISCUSSION .......................................................................... 24 LIST OF ANNEXES Annex 1: List of IDA Countries by Country Groupings and Selected Indicators....................................... 25 Annex 2: Main Constraints for Expanding Debt Coverage ........................................................................ 27 Annex 3: IMF and World Bank Agenda Going Forward on Strengthening Debt Transparency ................ 28 LIST OF BOXES AND FIGURES Boxes Box 1. Evolution of Rwanda’s Debt Situation .............................................................................................. 5 Box 2. Drivers of the Public Debt Build Up ................................................................................................. 6 Box 3. Hidden Debt ...................................................................................................................................... 7 Box 4. The Changing Role of Official Development Assistance ............................................................... 10 Box 5. Taking Stock of Debt Management Capacity in IDA Countries: DeMPA Results ......................... 13 Box 6. The Joint World Bank IMF LIC Debt Sustainability Framework ................................................... 15 Figures Figure 1. Public Debt in IDA Countries, 2000-17 ........................................................................................ 2 Figure 2. Public Debt by IDA Groupings and Region .................................................................................. 3 Figure 3. Public Debt in HIPCs and non-HIPCs ........................................................................................... 3 Figure 4. Public Debt in Selected IDA Countries ......................................................................................... 4 Figure 5. External PPG Debt in IDA Countries ............................................................................................ 8 Figure 6. External PPG Debt by IDA Groupings and Region ...................................................................... 8 Figure 7. Domestic Currency Denominated Debt by IDA Groupings .......................................................... 9 Figure 8. IDA Countries: Change in External PPG Debt Creditor Composition.......................................... 9 Figure 9. IDA Countries: International Bonds Issuances.............................................................................. 9 Figure 10. External PPG Debt at Variable Interest Rate ............................................................................. 11 Figure 11. Change in the Composition of External PPG Debt ................................................................... 11 Figure 12. Data Coverage in DSAs under the DSF .................................................................................... 12 Figure 13. Improvements in DeMPA Scores in IDA Countries ................................................................. 14 Figure 14. Evolution of Debt Distress Risk Ratings for IDA Countries ..................................................... 16 Figure 15. Safety Margins in Low and Moderate Debt Distress Risk Countries ........................................ 17 EXECUTIVE SUMMARY i. This paper responds to the interest expressed at the IDA Technical Briefing during the 2018 Spring Meetings to be updated on debt vulnerabilities in IDA countries. As noted in the September 2018 Bali Development Committee Paper on Debt Vulnerabilities in Emerging and Low-Income Economies, over 40 percent of Low-Income Developing Countries are currently assessed at high risk of external debt distress or in debt distress, doubling the number of countries in such categories since 2013. Forty-nine percent of IDA-eligible countries covered under the joint World Bank-IMF Debt Sustainability Framework (33 countries) are currently in high risk of external debt distress or in distress. This share declines to 29 percent without small states. Safety margins have eroded in countries at low or moderate risk of debt distress which increases the likelihood of future risk rating downgrades should debt continue to build up or in the event of shocks. Access to new sources of financing, if managed well, can help countries achieve their development goals. However, debt accumulation beyond sustainable levels constitutes a threat to development outcomes. It can also have implications for IDA’s balance sheet, as higher volumes of grants and/or higher borrower credit risk will reduce IDA’s overall financing capacity unless donors increase their contributions. IDA countries at high risk or in debt distress received 27.5 percent of IDA commitments under the performance-based allocation and 26 percent of total IDA commitments in fiscal year 2018. ii. Rising debt levels and the shift in the composition of debt have increased debt vulnerabilities in IDA countries: • Public debt in IDA countries has increased significantly in recent years. Countries classified simultaneously as commodity dependent and affected by fragility, conflict and violence saw the largest increase in public debt levels. At the same time, some countries managed to contain the debt build up. It is worth noting that the median public debt level in heavily indebted poor countries (HIPCs) is similar to the median public debt level in non-HIPCs. However, there remains substantial variation across HIPCs. • Changes in the composition of public debt increased debt service costs and refinancing risks. Several IDA countries accessed international bond markets and/or significantly increased domestic debt. In addition, the share of non-traditional, non-Paris Club, external creditors increased. Access to new sources of financing – if managed well – can help countries achieve their development goals. However, it can lead to higher debt service costs and refinancing risks. It can also pose challenges for debt resolution. The above highlights the need for strengthening debt management. • Key drivers of the increase in public debt include weak macro-fiscal management, changing composition of debt towards more expensive and riskier sources of financing and adverse shocks. The analysis of the drivers of the increase in public debt across IDA countries reveal three main factors: weak macro-fiscal policy frameworks, greater reliance on new sources of financing (which compounded with weaknesses in debt management, led to increased debt service and refinancing risks), and exogenous shocks (i.e., commodity price volatility, natural disasters, security challenges). In several cases, “hidden” debt, fraud or corruption were also contributing factors. iii. Vulnerable IDA countries have the primary responsibility to act decisively to contain debt- related risks, with support from the international community. Countries need to have a sound macro- economic policy framework. Also, to ensure that risks are detected and addressed, countries need to strengthen debt transparency, including improved public debt recording, monitoring and reporting, enhance their debt management capacity, and better assess debt sustainability. Building capacity to identify and manage fiscal risks from contingent liabilities would be important. Lowering debt vulnerabilities will also hinge on enhancing domestic revenue mobilization, improving the efficiency of public spending and - ii - prudent debt management, while building external and fiscal buffers and promoting growth-enhancing reforms. Tailored policy reforms that reflect country-specific vulnerabilities will be important. iv. The International Monetary Fund and World Bank are implementing a multi-pronged approach to address debt vulnerabilities. The multi-pronged approach was called for in April 2018 by the Development Committee and the International Monetary and Financial Committee and is described in detail in the September 2018 Bali Development Committee paper. Building on the progress achieved and on lessons learned in recent years, the multi-pronged approach on debt comprises ongoing and planned actions: • Strengthening debt analytics and early warning systems to help countries better understand debt vulnerabilities; • Strengthening debt transparency to help countries have a more complete picture of their debt; • Strengthening capacity on debt/fiscal risk management to help countries deal with existing debt more effectively, including through operational support to strengthen macro-fiscal policy frameworks and manage fiscal risks; • Reviewing the IMF Debt Limits Policy (DLP) and the IDA Non-Concessional Borrowing Policy (NCBP). v. Since the successful implementation of the HIPC Initiative and MDRI by a broad global coalition, IDA has taken the initiative to develop a framework of sustainable lending and put it into practice. IDA’s framework centers around three pillars: • The traffic light system based on forward-looking debt analysis under the joint Bank/Fund debt sustainability framework for low-income countries; • The grant allocation framework (GAF) to prevent a worsening of the debt burden due to IDA credits in low-income countries that are already at risk; • The NCBP that seeks to preserve the gains to debt sustainability from debt relief under the HIPC Initiative and MDRI, and to ensure prudent management of scarce IDA concessional resources. vi. As part of the IMF-WB multi-pronged approach, the NCBP review will consider the extent to which the NCBP has been effective in meeting its objectives and explore options to improve its signaling effect on borrowers and creditors. The review will take stock of what remedies are appropriate under increasingly diverse country circumstances, as well as approaches to ensure consistency and simplify decision-making. The NCBP review will take account of incentive effects of the grant allocation framework on borrower decisions. The review will also consider options for enhancing disclosure of NCBP decisions, while observing confidentiality requirements. vii. The impact of a revised NCBP could be greatly enhanced by anchoring it in an inclusive, global platform focused on information exchange, sustainable lending practices and effective use of concessionality. The forthcoming review of the NCBP in parallel with the IMF review of the DLP provides an opportunity to strengthen the partnership with the IMF, other multilaterals, Paris Club and other external partners to promote prudent debt management and sustainable lending practices. Recognizing the primary role of country borrowers in ensuring debt sustainability, the World Bank will deepen outreach to all creditors including non-Paris Club and private creditors. It would build on the global coordination that characterized the HIPC Initiative. A key goal would be to promote debt transparency which is a shared responsibility of borrowers (in the leading role) and creditors. This would include disclosure of public debt information by creditors in a manner similar to the World Bank. I. INTRODUCTION 1. This paper responds to the interest expressed at the IDA Technical Briefing during the 2018 Spring Meetings to be updated on debt vulnerabilities in IDA countries. As noted in the September 2018 Bali Development Committee Paper on Debt Vulnerabilities in Emerging and Low-Income Economies1, over 40 percent of Low-Income Developing Countries are currently assessed at high risk of external debt distress or in debt distress, doubling the number of countries in such categories since 20132. Forty-nine percent of IDA-eligible countries3 assessed under the joint World Bank-IMF Debt Sustainability Framework (33 out of 68 countries) are currently in high risk of external debt distress or in distress. This share declines to 29 percent without small states. Increased debt vulnerabilities are the result of rising public debt levels and a changing composition of debt towards more expensive and riskier sources of financing. Access to new sources of financing – if managed well – can help countries achieve their development goals. However, when countries accumulate debt beyond sustainable levels, development outcomes are at stake, including reaching the World Bank’s twin goals and the Sustainable Development Goals (SDGs). Debt sustainability concerns can also have implications for IDA’s balance sheet, as higher volumes of grant- making and/or higher borrower credit risk will reduce IDA’s overall financing capacity unless donors increase their contributions. IDA Countries at high risk of external debt distress or in debt distress received 27.5 percent of IDA commitments under the performance-based allocation and 26 percent of total IDA commitments in fiscal year 2018. 2. The paper aims to promote informed, inclusive discussion of debt vulnerability issues in IDA countries and possible IDA policy options to be considered during the upcoming IDA18 Mid-Term Review. The analytical work and initial policy options build on actions taken since the HIPC Initiative and Multilateral Debt Relief Initiative (MDRI) were implemented, including introduction of the debt sustainability framework (DSF), the grant allocation framework and the NCBP. Based on the results of the consultative process, management will bring forward specific policy recommendations including in the context of the 19th IDA Replenishment (IDA19) discussions. 3. The paper comprises five sections. Following the introduction, the second section presents an overview of the growing indebtedness problem in IDA countries. The third section summarizes the key pillars of the joint Bank/Fund multi-pronged approach to help countries address debt vulnerabilities as presented in full in the World Bank/IMF Bali Paper to the Development Committee. Aligned with the multi-pronged approach, the fourth section outlines the sustainable lending framework IDA has been implementing to help its client countries better manage and mitigate their debt risks, and policy options to enhance the impact of this framework going forward. The fifth section concludes. II. DEBT VULNERABILITIES IN IDA COUNTRIES 4. This section elaborates on IDA debt vulnerabilities, particularly the rising public debt levels and the changing composition of debt towards costlier and riskier sources of financing. The section links them to the policy dimensions driving them. The section also takes stock of the quality of debt 1 See the Bali Development Committee paper “Debt Vulnerabilities in Emerging and Low-Income Economies,” September 2018. 2 Currently 59 countries are classified as LIDCs, as defined in Annex 1 of the Bali Development Committee paper, of which 58 are currently IDA-eligible. 3 As of September 2018, 75 countries are IDA-eligible, of which 68 countries are covered under the joint World Bank-IMF Debt Sustainability Framework (excluding Somalia for which no LIC-DSA is available). See Annex 1 for the list of countries and different country groupings included in the analysis. -2- monitoring and debt management in IDA countries. It concludes with an assessment of the debt outlook and debt vulnerabilities, drawing on the joint World Bank-IMF Low-Income Country Debt Sustainability Framework (DSF). Public Debt Dynamics4 5. Following a significant decline through 2013, public debt levels in IDA countries have increased substantially in recent years (figure 1). The median public debt-to-GDP ratio in IDA countries declined from a peak of 88 percent in 2000 to 35 percent in 2013.5 The decline largely reflects the substantial debt relief provided to 35 IDA countries in the context of the Heavily Indebted Poor Countries (HIPC) Initiative and the Multilateral Debt Relief Initiative (MDRI) in the early 2000s.6 From 2013 to 2017, the median public debt-to-GDP ratio in IDA countries increased by 14 percentage points, reaching 49 percent in 2017.7 Figure 1. Public Debt in IDA Countries, 2000-17 (Percent of GDP) 140 Average 25 percentile Median 75 percentile 120 100 80 60 40 20 0 Source: WEO April 2018. World Bank staff estimates. 6. The increase in public debt levels over the 2013-17 period was broad-based across IDA- eligible countries (figure 2).8 At the regional level, Sub-Saharan Africa (AFR) experienced the largest increase in the median public debt-to-GDP ratio (20 percentage points, from already high levels). Smaller increases were observed in East Asia and Pacific (EAP) and Europe and Central Asia (ECA), while Latin America and Caribbean (LCR) and South Asia (SAR) experienced declines. Regarding IDA lending categories9, the largest increase in the median public debt-to-GDP ratio was observed for Blend countries (23 percentage points to reach 67 percent in 2017), followed by IDA-only non-gap countries and gap 4 Figures and charts in this sub-section refer to general government gross debt as defined in the World Economic Outlook (WEO). 5 Figures throughout the document are subject to data availability in the different databases. 6 Excludes Bolivia, which benefited from debt relief under both initiatives but graduated from IDA at the end of the IDA17 replenishment. 7 Unless otherwise specified, aggregate figures are medians to better capture what a “typical” country would be (by co ntrolling for outliers or the impact of large countries). The analysis and conclusions remain broadly the same if other aggregation methods are used. 8 Unless otherwise specified, the term “IDA-only” in the document refers to IDA-only non-gap countries. 9 Aggregate references to IDA lending categories and figures by IDA lending categories exclude inactive countries. -3- countries (17 percentage points and four percentage points, respectively). The increase for countries affected by fragility, conflict and violence was relatively modest (nine percentage points, compared to 13 percentage points for non-FCS), while small states experienced a decline of four percentage points.10, 11 The debt increase in commodity-dependent countries was 14 percentage points.12, 13 Figure 2. Public Debt by IDA Groupings and Region a. IDA Groupings b. Region (Percent of GDP) (Percent of GDP) 2007 Average 2013 Average 2017 Average Median 2007 Average 2013 Average 2017 Average Median 80 67 70 70 58 55 53 53 60 51 51 60 51 51 50 49 47 48 50 50 44 44 42 42 40 38 36 35 35 33 34 32 40 31 33 40 26 28 29 30 30 20 20 10 10 53 36 51 46 44 48 48 52 64 51 53 58 68 38 56 62 43 62 35 30 36 32 26 38 52 60 59 43 48 51 0 0 IDA-only IDA Gap IBRD Blend Small states FCS AFR EAP ECA LAC SAR Source: WEO April 2018. World Bank staff estimates. Note: MNA, which has only two IDA-eligible countries, has been excluded from figure 2.b. 7. The median public debt level Figure 3. Public Debt in HIPCs and non-HIPCs in HIPCs (HIPCs are countries that benefitted from debt relief under the (Percent of GDP) HIPC Initiative and MDRI) is similar Median HIPC Median Non-HIPC 120 to the median public debt level in non- HIPCs (figure 3). The median public 100 debt-to-GDP ratio in HIPCs declined sharply from a peak of 106 percent in 80 2002 to 35 percent in 2007 (the year 60 immediately following the launch of MDRI). From 2008 to 2013 the ratio 40 declined slightly to 34 percent; but, like 20 most other country groupings, it experienced an increase in recent years 0 (13 percentage points between 2013 and 2000 2002 2004 2006 2008 2010 2012 2014 2016 2017) reaching 47 percent in 2017. Since Source: WEO April 2018. World Bank staff estimates. 10 In this document FCS refers to the IDA countries affected by fragility, conflict and violence (FCV). For the definition and full list of FCV countries, see http://www.worldbank.org/en/topic/fragilityconflictviolence/brief/harmonized-list-of-fragile- situations. 11 Small states are defined as countries with a population of 1.5 million or less. 12 Commodity-dependent countries refer to countries listed in the June 2018 World Bank Global Economic Prospects as having, on average in 2012-14, either (i) total commodities exports accounted for 30 percent or more of total goods exports or (ii) exports of any single commodity accounted for 20 percent or more of total goods exports. Economies for which these thresholds were met, re-exports were excluded. As per this definition, there are 44 commodity-dependent IDA countries. 13 The commodity price shock had a major impact on debt dynamics of commodity-dependent countries, negatively affecting both growth and debt stocks. Within this group, energy exporting countries have driven the increase of the debt levels (19 percentage points as a group). -4- 2014 (coinciding with the sharp commodity price decline) the gap between HIPC and non-HIPC median levels narrowed significantly. Currently the median public debt level in HIPCs is similar to the median public debt level in non-HIPCs. There is, however, significant heterogeneity across HIPCs. Nine countries had relatively modest increases (five percentage points or less) or a reduction of the public debt-to-GDP ratio. Eleven countries had public debt-to-GDP increases of 20 percentage points or more; and in four cases public debt has reached the levels observed in the early 2000s (all of them are commodity-dependent and three are FCS). 8. Countries classified simultaneously as commodity dependent and FCS had the highest increases in public debt levels over the 2013-17 period. Most of the public debt to GDP changes (50 percent) across IDA countries during that period ranged between 3 to 20 percentage points. There were, however, several outliers (minimum and maximum of -37 to 93 percentage points, respectively). Increases of 20 percentage points or more were observed in 18 countries. All but three of these countries were in AFR, 12 were HIPCs and 13 were commodity dependent (which included eight out of 11 FCS in the group).14 All but three of these countries were IDA-only. Eight countries experienced an increase beyond 30 percentage points (Republic of Congo, The Gambia, Mozambique, South Sudan, Sudan, Yemen, Zambia and Zimbabwe). Five of these countries are fragile and conflict-affected commodity-dependent countries (figure 4.b). 9. Countries with the highest public debt levels tended to accumulate debt faster in recent years (figure 4.a). Except for Grenada, Guyana and Samoa, no country with a debt level higher than the median for IDA countries reduced the public debt-to-GDP ratio in the last five years. In seven of the top 15 fast- borrowing countries15, the increase occurred between 2013 and 2017 and represents over 50 percent of their current public debt-to-GDP ratio (figure 4.b). Six of the eight countries with public debt levels over 100 percent of GDP in 2017 had increases in the public debt-to-GDP ratio between 2013 and 2017 well above the median for IDA countries. Five of these countries were either FCS or commodity-dependent countries, four were both. This group includes three IDA-only countries (Yemen, The Gambia and Mozambique). Figure 4. Public Debt in Selected IDA Countries (Percent of GDP unless otherwise specified) a. Highest Public Debt Levels b. Fast Borrowers 2017 Public debt 2013-17 Change (percentage points) 2013-17 Change (percentage points) 2017 Public debt 150 150 2017 Median public debt 2013-17 median change (percentage points) 100 100 50 50 0 0 Source: WEO April 2018. World Bank staff estimates. 14 Among the 13 commodity-dependent countries there were 3 energy exporters. 15 Fast borrowing countries defined as those with the largest increase in the public debt-to-GDP ratio over the 2013-17 period. -5- 10. Twenty-four countries managed to reduce or contain the increase of their public debt to GDP ratios. Eleven countries had a limited increase in their public debt-to-GDP ratio (10 percentage points or less) without a deterioration of their assessed risk of external debt distress.16 Thirteen additional countries managed to reduce their debt level. Half of these 24 countries are HIPCs. Only five of them are countries classified as both fragile and commodity-dependent countries; these are Democratic Republic of Congo, Côte d’Ivoire, Guinea-Bissau, Mali and Myanmar, none of which is an energy exporter. Box 1 discusses the case of Rwanda, a country that has improved its assessed risk of debt distress (from high risk of debt distress in 2006 to low risk of debt distress in 2017) while experiencing an increase in the public debt-to- GDP larger than 10 percentage points. Box 1. Evolution of Rwanda’s Debt Situation In 2006, Rwanda was assessed at high risk of external debt distress. By 2009, its rating had improved to moderate risk of debt distress. In 2014, at a point when debt vulnerabilities in IDA countries started to deteriorate, Rwanda was rated at low risk of debt distress and it has kept this status ever since. Rwanda’s positive story has been underpinned by a strong export performance over the past two decades (see figure), and a strong commitment to good governance and sound macroeconomic policies. Figure B1.1. Exports (Percent of GDP) Sound fiscal management has been key in ensuring that public resources are used effectively, with growth- enhancing infrastructure investment directed to carefully selected projects. Prudent fiscal and monetary policies have been instrumental in maintaining macroeconomic stability. Exchange rate flexibility has been the centerpiece of adjustment policies to address external imbalances. Rwanda also pursued structural reforms with notable success in improving the policy and regulatory environment for the private sector. Rwanda’s pace of structural transformation has also been faster than most peers in Sub-Saharan Africa, promoting a rapid move out of agriculture mainly into services. Reflecting the above, Rwanda’s Country Policy and Institutional Assessment (CPIA) score has been steadily improving, placing the country among IDA’s top performers (as measured by the CPIA), which resulted in upward shifts in the applicable World Bank IMF LIC DSF debt indicator thresholds. Going forward Rwanda needs to manage the risks to its public debt outlook. Its public debt burden has doubled between 2010 (20 percent of GDP) and 2017 (41 percent of GDP). The key risks to the debt sustainability outlook include the narrow export base (still heavily weighted towards commodities) and the refinancing of Eurobonds coming due in 2023. 16 The analysis excludes countries already assessed at high risk of debt distress or in debt distress. -6- 11. A decomposition of public debt dynamics across IDA countries points to weak fiscal management, weak macroeconomic policy frameworks to support growth, adverse shocks (commodity price volatility, security challenges, natural disasters) and greater reliance to costlier and riskier sources of finance (box 2). Several countries were affected by internal conflict (Yemen and Burundi – both of which experienced increases in the public debt-to-GDP ratio of 93 and 21 percentage points, respectively, between 2013 and 2017) or by shocks (Ebola epidemics in the case of Liberia and Sierra Leone – where the public debt-to-GDP ratio increased by about 20 percentage points between 2013 and 2017). In a limited number of cases, issues of hidden debt, fraud and corruption contributed to the increase in the public debt-to-GDP ratio. Examples include Mozambique (undisclosed debt, as detailed in box 3) and The Gambia (serious governance issues including embezzlement of funds), where the public debt-to-GDP ratio increased by 49 and 34 percentage points between 2013 and 2017, respectively, to levels above 100 percent. In some countries exchange rate depreciation also pushed up public debt levels. Box 2. Drivers of the Public Debt Build Up A combination of weak fiscal management, weak macro-fiscal management frameworks to support growth, greater reliance on costlier and riskier sources of finance, and adverse shocks largely explains the increase in public debt levels. Widening fiscal deficits played a particularly important role in commodity-dependent countries after the end of the commodity price boom. These countries ran sizeable fiscal deficits during the commodity price boom. Failure to build adequate buffers during that period left countries with the choice between higher borrowing and sharp fiscal adjustment when prices – and with them domestic revenues – dropped. Most chose the former. In some countries exchange rate depreciation also pushed up debt. For countries less reliant on commodities, primary deficits remained at the core of debt dynamics, but changes in debt ratios by 2017 came to be driven increasingly by currency depreciation and a higher contribution from interest spending as countries contracted less concessional sources of financing. Figure B2.1. Drivers of Total Public Debt (Percentage point of GDP) Commodity-Dependent Countries Non-Commodity-Dependent Countries Primary deficit Real GDP growth Primary deficit Real GDP growth Real exchange rate depreciation Real interest rate Real exchange rate depreciation Real interest rate Other Change in public sector debt Other Change in public sector debt 6 4 4 3 2 2 0 1 -2 0 -4 -1 -6 -2 -8 -3 2007-12 2013-17 2007-12 2013-17 Source: World Bank/IMF LIC-DSA database. Note: The category “Other” includes other debt-creating flows (e.g., recognition of contingent liabilities, privatization receipts, debt relief, bank recapitalization) and a residual stemming from data and measurement errors under the DSF. Empirical evidence for 34 Low-Income Developing Countries (LIDCs) suggests higher levels of public investment explain the build-up of debt only to a limited extent.1 Larger fiscal deficits are fully explained by the increase in public investment in about 30 percent of cases, and partially explained in one quarter of cases –with investment falling in the remainder. The potential for boosting future growth and development outcomes of debt-financed investments will ultimately depend on the quality of the investments. 1 LIDCs are countries with low per-capita gross national income and comparatively weak socioeconomic indicators. For a full definition and the list of countries see “Macroeconomic Developments and Prospects in Low-Income Developing Countries,” IMF, March 2018. -7- 12. A temporary increase in public debt levels – if well managed – may to some extent reflect a reasonable policy choice, but some of the steepest public debt increases reflect weak macroeconomic policy frameworks, hidden debt, fraud and corruption. A case can be made for smoothing consumption following a terms of trade shock, with the optimal speed of adjustment depending on a range of factors. Several countries also took advantage of the very low global interest rates and risk aversion in the wake of the global financial crisis to finance higher public investment to support growth. However, countries with weak macroeconomic policy frameworks, particularly fragile commodity-dependent countries, that failed to implement growth-enhancing reforms to diversify their economies and build buffers to mitigate the impact of external shocks experienced the steepest increases. Some countries also suffered from hidden State-Owned Enterprises (SOEs) debts (e.g., Mozambique and Republic of Congo – see box 3) as a result of significant deficiencies in the management and oversight of SOEs and deep-rooted governance challenges. Box 3. Hidden Debt There have been several recent cases of hidden debt among IDA countries: • In the Republic of Congo, in mid-2017, it was confirmed that pre-financing contracts with oil traders were contracted by the oil SOE on behalf of the government but remained undisclosed to the Debt Management Office (DMO), for over two years, as they were considered to be outside the DMO’s mandate. • In Togo, the government had pre-financed debt in 2015 – a form of de facto government debt which was not reflected in official government debt statistics, amounting to 7 percent of GDP at end-2016. • In Mozambique, two state-guarantees issued in 2013 and 2014 by the Minister of Finance to SOEs incorporated as private enterprises- amounting to US$1.15 billion (9 percent of GDP at end-2015)- were not disclosed to the debt management staff and the public, and broader governance issues appear to have been an issue. The case of Mozambique illustrates the consequences of the fallout that hidden debt can have. The authorities’ announcement to restructure the “tuna bonds” and the discovery of potential misuse of the borrowed funds led to the country being cut off from donor funding, and its subsequent default to some private creditors. It also set off a sharp widening of the bond yields since mid-2015. Just as the debt exchange for the “tuna bonds” was concluded in April 2016, two large previously undisclosed external loans were revealed, which saw bond yields spike again in October 2016.1 1 See box 2 of the document “G-20 Note: Improving Public Debt Recording, Monitoring and Reporting Capacity in Low and Lower Middle- Income Countries: Proposed Reforms.” External Public and Publicly Guaranteed (PPG) Debt Dynamics17 13. The evolution of the external PPG debt levels mirrors that of public debt, albeit with a much smaller increase during recent years (figure 5). From a peak of 72 percent in 2000-02, the median external PPG debt-to-GDP ratio decreased to 24 percent in 2013. From 2013, the ratio slowly increased to reach 28 percent in 2016. 18 17 Figures and charts in this sub-section refer to PPG debt as defined in the World Development Indicators (WDI). 18 2016 is the last year for which data is available in WDI. Given different time-period, increases in recent years of public and external debt to GDP ratios are not comparable. -8- Figure 5. External PPG Debt in IDA Countries (Percent of GDP) 120 Average 25 Percentile Median 75 Percentile 100 80 60 40 20 0 Source: WDI and World Bank staff estimates. 14. The increase in external PPG debt across IDA groupings was not as significant nor as wide- spread as the increase in public debt (figure 6). At the regional level, the median external PPG debt-to- GDP ratio increased between 2013 and 2016 in AFR and ECA (three and four percentage points, respectively) while it remained broadly stable in EAP and SAR and it experienced a reduction in LCR. Within IDA lending categories, the steepest increase was observed in Blend countries (five percentage points). FCS and commodity-dependent countries experienced relatively modest increases (two and three percentage points, respectively). At the country level, 26 countries saw their external PPG debt-to-GDP ratio decline or remain constant. Figure 6. External PPG Debt by IDA Groupings and Region a. IDA Groupings b. Region (Percent of GDP) (Percent of GDP) 2007 Average 2013 Average 2016 Average Median 50 2007 Average 2013 Average 2016 Average Median 70 58 38 60 40 44 48 32 31 33 50 35 38 29 30 29 34 27 27 26 40 27 24 27 25 23 28 29 30 24 25 24 26 22 30 20 20 20 20 10 43 29 33 30 31 35 40 33 39 59 25 30 47 47 50 0 10 45 29 35 33 28 29 33 22 26 36 40 38 32 30 31 0 AFR EAP ECA LAC SAR Source: WDI. World Bank staff estimates. Shifts in Debt Composition 15. Public debt in some IDA countries has shifted in recent years towards more domestic currency denominated debt. Foreign currency denominated debt still accounts for the larger part of total debt, but domestic currency denominated debt has increased to represent a median of about one-third of total public debt in IDA countries. For IDA countries as a group, the median foreign currency denominated -9- debt-to-GDP ratio stood at 30 percent of GDP in 2017. This compares to a median of 17 percent for domestic currency denominated debt. Between 2007 and 2017, some country categories posted a significant increase in domestic currency denominated debt Figure 7. Domestic Currency Denominated Debt (figure 7). For example, the median domestic by IDA Groupings currency denominated debt-to-GDP ratio (Percent of GDP) increased by nine percentage points for Blend countries and by five percentage points for IDA- 35 30 only countries. FCS posted an increase of eight 30 24 percentage points, while at the regional level 23 25 21 20 20 AFR posted an increase of six percentage points. 19 While domestic debt market development can 20 14 15 15 12 13 11 13 help countries deepen financial markets and 15 10 reduce exchange rate risks, a significant 10 expansion of domestic government borrowing 5 13 16 24 13 13 14 20 23 29 17 20 20 20 22 33 can crowd-out private investment and reduce 0 economic growth in the presence of shallow IDA-only IDA Gap IBRD Blend Small States FCSs 2007 Average 2013 Average 2017 Average Median financial markets, financial repression and poor debt management capacity. Source: WEO 2018. World Bank staff estimates. 16. The composition of external public debt in IDA countries has shifted towards riskier and more expensive sources of financing (figure 8). Multilateral creditors and Paris-Club creditors have declined in importance as a source of financing for IDA countries in recent years. Non-Paris Club creditors have emerged as an increasingly important source of financing for IDA countries (for some countries, mainly commodity-dependent ones, they account now for over half of the external debt stock).19 External borrowing from private creditors has also grown significantly in recent years, albeit from a low base (box 4).20 In recent years, IDA countries have actively issued international commercial bonds (figure 9). Commodity-dependent countries, including some HIPCs, constitute the bulk of issuers. Large associated redemptions in the next few years could create significant roll-over risks over the medium-term. Figure 8. IDA Countries: Change in External Figure 9. IDA Countries: International Bonds PPG Debt Creditor Composition Issuances (US$ billion) (Percent of GDP, 2007-16) 12 Non-Paris Club Bilateral 10 8 Private 6 Multilateral 4 2 Paris Club Bilateral 0 -6 -4 -2 0 2 4 2010 2011 2012 2013 2014 2015 2016 2017 2018 Source: WDI. World Bank staff estimates. Source: Bloomberg and World Bank staff estimates. Note: Chart reflects weighted average. Note: 2018 figure as of August 2018. 19 Debt owed by IDA countries to the top 5 non-Paris Club creditors account for 87 percent of the debt disbursed and outstanding to all non-Paris Club creditors as of end-2016. 20 External PPG debt owed by IDA countries to private creditors as a share of total PPG is estimated to have increased from 8 percent in 2007 to 15 percent in 2016. - 10 - Box 4. The Changing Role of Official Development Assistance (ODA) Net external financing to IDA countries increased until 2014 but has declined since then (figure B4.1). The increase and decline in net external financing to IDA countries has largely been driven by debt owed to private creditors, which includes international bond issuances and commercial bank loans. Since 2013, grant financing (net of debt forgiveness) is on a declining trend and its share of total net external financing has decreased from 70 percent on average over 2007-11 to 55 percent in the period 2012-16. The share of grants in total gross disbursement has declined across all IDA categories (figure B4.2). Commodity-dependent and HIPC countries suffered the steepest decline in the share of grants. Grants remain the key source of net disbursements for IDA-only countries, FCS and HIPCs, but contribute less than 50 percent of total net disbursements in small states and commodity-dependent countries. Figure B4.1. Net External Financing Figure B4.2. Average Public External Debt Gross (US$ million) Disbursements and Grants (Percent of total) 80,000 100% 70,000 60,000 80% 50,000 60% 40,000 30,000 40% 20,000 20% 10,000 - 0% 2007-11 2007-11 2012-16 2007-11 2012-16 2007-11 2012-16 2007-11 2012-16 2012-16 (10,000) 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Grants Bilateral non-concessional IDA-only FCS Small States Commodity HIPC Bilateral concessional Multilateral non-concessional exp. Multilateral concessional Private creditors ODA Grants Official concessional* Official non-concessional* IMF Private Source: OECD DAC and International Debt Statistics. Source: OECD DAC and International Debt Statistics. A significant share of ODA grants and official concessional financing goes to go Gap and Blend countries (figures B4.3 and B4.4). Over the period 2012-16, Gap and Blend countries benefited from 33 percent of ODA grants to IDA countries, this share remained unchanged relative to the 2007-11 period. Gap and Blend countries also benefited from close to 60 percent of the official concessional financing granted to all IDA countries during the 2012-16 period. This represents an increase of 7 percentage points relative to the 2007-11 period. Figure B4.3. Distribution of ODA Grants Figure B4.4. Distribution of Official Concessional by IDA Lending Category Financing by IDA Lending Category 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 2007-2011 2012-2016 2007-2011 2012-2016 IDA-only Gap Blend IDA-only Gap Blend Source: OECD DAC. Source: IDS. - 11 - 17. Reflecting this shift, external PPG debt carrying variable interest rates has risen markedly in recent years (figure 10). In IDA countries, the share of external PPG debt carrying variable interest rates debt has doubled since 2007 reaching an average Figure 10. External PPG Debt at Variable of 30 percent of total PPG debt in 2016. By 2016, Interest Rate this share reached an average of 46 percent in (Share of total external PPG) blend countries and lower levels in IDA-only countries (21 percent for IDA-only non-gap and 50% 25 percent for Gap countries, respectively). Debt service costs for countries with a high share of 40% variable rate debt coming to maturity over the 30% next years may rise as monetary policy in advanced economies normalizes, raising 20% concerns for financial stability and debt service 10% burden. The increase in external PPG debt in IDA countries has been accompanied by a 0% significant increase on private nonguaranteed IDA-only Gap Blend Small states FCS (PNG) external debt (164 percent in nominal 2007 2016 US$ during the 2010-16 period based on WDI data).21 An increase in interest rates and Source: International Debt Statistics (IDS). World Bank staff exchange rate fluctuations would also affect the estimates. cost of PNG debt financing, exacerbating financial stability concerns.22 18. The share of concessional external debt has declined (figure 11). Between 2007 and 2016, official debt (concessional and non-concessional) declined as a share of the total external PPG debt outstanding from 93 percent to 84 percent, Figure 11. Change in the Composition of respectively. At the same time, external PPG debt External PPG Debt owed to commercial creditors has increased, (2007-2016 – percentage point) reaching 16 percent. This shift is observed across 15% all country groupings except for small states. It is 10% more pronounced in IDA Gap and Blend 5% countries, as well as in commodity-dependent countries and HIPCs, where the share of external 0% PPG debt owed to commercial creditors has -5% increased by more than 10 percentage points. -10% Concessional financing as a share of total external PPG debt outstanding has decreased significantly -15% Blend Gap IDA-only FCS Small HIPC less in IDA-only and it has slightly increased for States FCS. For these countries, the increase in external Official concessional Official non-concessional Private PPG debt to commercial creditors is mostly associated to a reduction of non-concessional Source: International Debt Statistics. official external financing. 21 Only one third of the 59 International Development Association (IDA)-only countries report private non-guaranteed external debt. The WB’s Development Data Group (DECDG) compiles private non-guaranteed external debt for 70 percent of reporting countries. About 50 percent of these countries provide data. For the other countries, DECDG estimates aggregate external statistics from other data sources. 22 This would also affect the cost of PNG debt financing, raising financial stability concerns. See World Bank, June 2018 Global Economic Prospects. Special Topic: Corporate Debt: Financial Stability and Investment Implications. - 12 - Debt Monitoring and Debt Management in IDA Countries 19. Despite significant improvements in debt data, current public debt statistics suffer from limited debt data coverage and debt transparency (specially related to SOE debt, fiscal risks related to PPPs, and collateralized debt) – which leaves room for unpleasant debt surprises (figure 12). Data on domestic and external debt is often recorded in separate databases. Contingent liabilities, such as those arising from government guarantees, debt of SOEs, and public-private partnerships (PPP), are rarely collected in the central debt recording system. Several countries lack a systematic mechanism for collecting and recording information on collateralization. As a result, debt coverage is often incomplete as evidenced in DSAs covered under the World Bank IMF LIC DSF. These shortfalls as well as lack of sufficient information on the terms and conditions of public debt (including collateral requirements), impede reliable debt projections and, in some cases, have led to “unpleasant debt surprises” (box 3). Main constraints for expanding coverage are summarized in Annex 2. Figure 12. Data Coverage in DSAs under the DSF (Share of DSAs in LIDCs) Central Government (General 31% Government only) Central Government (General Government) + Government 61% Guarantees Central Government (General Government) + Non-Financial 8% Public Corporations 0% 20% 40% 60% 80% Source: IMF/World Bank LIC-DSAs. 20. There are several factors driving the weaknesses in debt recording, monitoring, and reporting. These include: weak incentives to produce reliable data (e.g., limited public scrutiny and lack of integration with other public financial management (PFM) systems); weak procedures; weaknesses in IT infrastructure for debt recording and outdated software; and insufficient human resources and lack of appropriate skills. Weaknesses are also driven by broader institutional and governance related issues, for example fragmented responsibilities and uncoordinated institutional arrangements; limited/inadequate mandate for performing debt management functions; and weak audit capacity. 21. Despite increases in public debt and a shift towards riskier debt portfolios, debt management capacity in IDA countries remains weak. While there have been improvements in debt management capacity and institutions, available evidence suggests that there are still significant gaps in IDA countries. The World Bank’s Country Policy and Institutional Assessment (CPIA) indicates that the quality of the debt management policy and institutions in about half of IDA countries falls short of the 3.5 rating for which countries are considered to have “adequate” capacity.23 Furthermore, results from Debt Management Performance Assessments (DeMPAs) suggest that IDA countries have had uneven improvements on debt 23 The CPIA is a diagnostic tool that captures the quality of a country’s policies and institutions, with ratings ranging from 1 to 6. Countries that have a rating below 3.5 are considered to have “weak” capacity. Among the sixteen criteria in the CPIA, one indicator assesses the central government’s capacity to record reliable and comprehensive public debt data and to produce public debt reports. - 13 - management functions over time (figure 13) and face significant debt management challenges.24 Areas of particular concern include (box 5): weaknesses in debt management governance (e.g., legal frameworks, insufficient audits, poor quality of debt management strategies and implementation capacity); weak coordination with fiscal policy; weaknesses in public financial management and in regulatory frameworks for domestic borrowing, loan guarantees, on-lending and derivatives; lack of operational risk management; poor quality of debt data; and insufficient staff capacity in debt management offices, to adequately assess fiscal and debt risks and deal with a diverse and fragmented landscape of investors and emerging creditors. Box 5. Taking Stock of Debt Management Capacity in IDA Countries: DeMPA Results An analysis based on DeMPAs performed between 2015−17 shows that the minimum requirements in debt recording and with regards to monitoring guarantees are met by only 41 percent and 33 percent of the sample countries, respectively. Given the problems with the accuracy, timeliness, coverage and completeness of debt records only 35 percent of the sample countries meet the minimum requirements for debt reporting and evaluation. 1 The DeMPA results also highlight broader debt management challenges: • Weak legal frameworks: Only about half of the sample countries have legal frameworks that clearly define the delegation of authority to borrow and undertake debt management activities including the issuance of guarantees. • Insufficient audits: Few countries undertake external financial audits on an annual basis or have conducted compliance audits in the past two years; hardly any country has had a debt management performance audit. Where audits are conduced, only about half of the countries address the outcomes of the audits. • Poor quality of debt management strategies and implementation: Although the number of countries producing medium-term debt management strategies has increased, the quality of such policies is lagging. In addition, few countries are aligning medium term debt and fiscal policy formulation processes. • Poor data administration and internal control: Even fewer countries evaluated complied with all the minimum requirements to have procedures (about 10 percent) and internal controls (about 20 percent) in place that ensure accurate, timely, and secure processing with minimal errors of public debt transactions. 18 percent of the countries had adequate data security with documented procedures for controlling access to the data recording system. • Low staff capacity: less than half of countries in the sample meet the minimum requirement for staff capacity and human resource management. Overall, countries in the sample score very poorly on operational risk management, business continuity, and disaster recovery planning. • Limited performance in cash flow management and forecasting as well as its coordination with debt management operations: Domestic treasury bills issuing programs are not linked to the government cash position and future needs due to absence or low quality of cash flow forecast. This may lead to the excessive cash position in government account and unnecessary carrying cost, or on the other hand insufficient cash resulting in cash rationing. • Operational risk management for government debt: Insufficient risk management can often mean significant exposure to the loss of data and other failures in implementation of critical debt management processes. 1 The sample included 9 LMICs and 8 LICs. Although this is a limited data set, evidence from DeMPAs in a broader set of countries and those conducted prior to 2015 suggests the findings here may be representative. 24 The DeMPA is a methodology for assessing performance covering the full range of government debt management operations. It is focused on central government debt and loan guarantees. See www.worldbank.org/debt for a description of the DeMPA. - 14 - Figure 13. Improvements in DeMPA Scores in IDA Countries (Percent of countries that meet the minimum requirement) Legal framework 80% Debt Records Managerial Structure Segregation of Duties, Staff 60% Debt Management Strategy Capacity and BCP 40% Debt Administration and Data 20% Evaluation and Reporting Security 0% Cash Flow Forecasting and Cash Audit Balance Management Loan Guarantees, On lending Coordination with Fiscal Policy Derivatives Coordination with Monetary External Borrowing Policy Domestic Borrowing 1st DeMPA Most recent DeMPA Source: WB staff calculations. Note: The sample includes 33 IDA countries for which two DeMPA exercises have been undertaken up to end-2017. Evolution of Risk of Debt Distress 22. Rising debt levels and shifts in the composition of debt have increased debt vulnerabilities in IDA countries as a group. As noted in the September 2018 Bali Development Committee Paper on Debt Vulnerabilities in Emerging and Low-Income Economies, over 40 percent of Low-Income Developing Countries are currently assessed at high risk of external debt distress or in debt distress, doubling the number of countries in such categories since 2013.25 Forty-nine percent of IDA-eligible countries covered under the joint World Bank-IMF Debt Sustainability Framework (33 countries) are currently in high risk of external debt distress or in distress.26 This share declines to 29 percent without small states. Twenty-four IDA countries are at moderate risk of external debt distress, and 11 are at low risk of external debt distress. 23. Changes in risk of debt distress ratings in IDA-eligible countries since 2013 are driven by the following dynamics (see box 6 and figure 14)27: 25 Currently 59 countries are classified as LIDCs as defined in Annex 1 of the Bali Development Committee paper “Debt Vulnerabilities in Emerging and Low-Income Economies,” of which 58 are currently IDA-eligible. 26 As of September 2018, 75 countries are IDA-eligible. They include 58 countries classified as LIDCs (32 of which are LICs and 26 LMICs) plus 17 countries classified as emerging economies (1 LIC, 6 LMICs and 10 UMICs) as defined in Annex 1 of the Bali Development Committee paper “Debt Vulnerabilities in Emerging and Low-Income Economies,” September 2018. Currently, 68 IDA-eligible countries are covered under the joint World Bank-IMF Debt Sustainability Framework (excluding Somalia, for which no LIC DSA is available) 27 Figures in this section draw on a sample of 64 IDA countries until 2013, 67 in 2014 and 68 from 2015 to 2018. Of the 68 IDA countries in the sample in 2018, 56 are LIDCs, see footnote 1 in Box 2 for a definition of LIDCs. AFR represents more than half of the countries in the sample. - 15 - • Twenty-one risk rating downgrades.28 This includes: (i) seven countries whose risk rating deteriorated from low to moderate; (ii) four countries that experienced a deterioration from low to high; and (iii) ten countries that shifted from moderate to high risk. Most of the downgrades correspond to countries in AFR. • Six risk rating upgrades. Four countries improved from moderate to low risk of debt distress and two from high to moderate risk of debt distress. • The addition to the sample of four additional IDA countries, all of which are currently at high risk of debt distress. Box 6. The Joint World Bank IMF LIC Debt Sustainability Framework The joint World Bank-IMF Debt Sustainability Framework for Low-Income Countries (DSF) is a methodology for conducting standardized debt sustainability analysis (DSA). The objective of the DSF is to support efforts by LICs to achieve their development goals while minimizing the risk that they experience debt distress. Reflecting the empirical finding that a country with better policies, institutions, assets, and macroeconomic prospects can sustain higher debt levels, the framework: • classifies countries based on their assessed debt-carrying capacity in three categories (strong, medium, and weak); • for each debt-carrying capacity category, it establishes thresholds for selected public and external debt burden indicators (in terms of GDP, exports and revenues). Thresholds for countries with strong debt carrying capacity are the highest; • evaluates baseline projections and stress-test scenarios relative to the relevant thresholds; and • combines mechanical rules and staff judgment to assign risk ratings for external and overall debt distress. 1 The framework has demonstrated its operational value since its development in 2005, but it has been periodically revised to enhance its performance in assessing risks in the context of a changing financial landscape. The last set of reforms took place in 2017 and was informed by both staff analysis and stakeholder feedback.2 While the core architecture of the LIC DSF remains the same, the revised framework introduces a number of important reforms, including: • Using a composite index with an expanded set of country-specific information to classify countries according to their debt carrying capacity. In addition to using the CPIA, the composite index includes country growth, international reserves coverage and remittance flows, and global growth. • Introduction of “realism” tools that aim at supporting stronger baseline debt forecasts and better implementation of the new classification methodology. • Streamlining the mechanical rules by reducing the number of debt indicators, thresholds and standardized stress tests, and the debt thresholds are also re-estimated. • Increasing the granularity within the risk categories, including a tool for characterizing countries in the moderate risk category in terms of its “space to absorb shocks”. • Introducing tailored stress tests on contingent liabilities, natural disasters, volatility of commodity export prices and market-financing shocks. Other risks stemming from high domestic debt levels and market- financing pressures are better assessed. Implementation of the revised DSF started in July 2018. 1 See “Guidance Note on the Bank-Fund Debt Sustainability Framework for Low-Income Countries,” December 2017, for a full discussion on how mechanical rules and judgment are used to inform final risk of debt distress ratings. 2 See “Review of the Debt Sustainability Framework for Low-Income Countries: Proposed Reforms,” August 2017, for a full discussion of the 2017 revisions to the DSF. 28 Risk rating upgrades/downgrades in this section cover the period from 2013. - 16 - Figure 14. Evolution of Debt Distress Risk Ratings for IDA Countries (Percent of IDA countries with LIC-DSAs) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Calendar Year Low Moderate High Source: World Bank/IMF LIC-DSA database. As of August 2018. Note: Figures are in calendar year. Sample includes 64 countries up to 2013, 67 in 2014 and 68 from 2015-2018. 24. IDA countries facing structural challenges account for a large share of the high or moderate risk of debt distress categories and/or have been affected by risk rating downgrades in recent years. Among countries in the sample: • Fragile and conflict-affected countries. Out of 29 FCS, only one is at low risk of debt distress, 10 are at moderate and 18 at high risk of debt distress. Seven out of 10 reclassifications in recent years were downgrades. • Commodity-dependent countries. The distribution of these countries is concentrated in the moderate risk of debt distress category. However, since 2013, 16 countries experienced risk rating downgrades compared to only three upgrades. • Countries that are commodity dependent and FCS. This group includes 16 countries. Their distribution is heavily concentrated among the high (7 countries) and moderate (8 countries) risk of debt distress categories. Furthermore, 8 of these countries saw their risk of debt distress modified between 2013 and 2017, all but two of which were downgrades. • Small states. All 19 countries in the group are at moderate or high risk of debt distress (6 and 13, respectively). Since 2013 there have been six risk rating downgrades and one upgrade. Small states are disproportionately exposed to economic shocks, natural disasters and the impact of climate change. They also suffer from many inherent vulnerabilities given their small population and economic size, including narrow potential for domestic revenue generation, lack of economies of scale in the provision of government services, and limited economic diversification. In combination, these factors have contributed to these countries’ vulnerable debt outlook.29 25. As of September 2018, ten countries are in debt distress, of which seven have publicly available DSAs. Two of these countries are inactive IDA borrowers (Sudan and Zimbabwe), while five 29 See technical note: “Small States: Vulnerability and Concessional Finance,” World Bank, July 2018. - 17 - (The Gambia, Grenada, Mozambique, São Tomé and Príncipe and South Sudan) have recently requested debt restructuring or accumulated large external arrears. Chad’s risk of debt distress has been recently upgraded to high thanks to a successful debt restructuring. Of these seven countries, 4 are commodity- dependent FCS. 26. Relative to non-HIPCs, HIPCs (defined as countries that have benefitted from debt relief under the HIPC Initiative and MDRI) are not disproportionately represented in higher risk categories or disproportionately affected by downgrades in recent years. Relative to non-HIPCs, HIPCs exhibit a higher concentration in the moderate risk of debt distress category and a significantly lower share of countries in both low and high risk of debt distress categories. However, relative to 2013, the number of HIPCs at high risk of debt distress increased from 9 to 16. This reflects nine downgrades and two upgrades. 27. Safety margins in many countries currently assessed at low or moderate risk of debt distress have eroded. Eighteen countries in the sample are currently at moderate or low risk of debt distress with ratings unchanged since 2013.30 The median public debt-to-GDP ratio for these countries increased 5 percentage points between 2013 and 2017 (5 percentage points for moderate and 8 percentage points for low risk of debt distress countries). For most countries, the distance between debt indicators and the relevant LIC DSF indicative threshold has narrowed since 2013 (figure 15). This signals an increased likelihood of a risk of debt distress downgrade should debt continue to build up or in the event of a shock. Figure 15. Safety Margins in Low and Moderate Debt Distress Risk Countries a. PV of Debt to Exports Ratio b. Debt Service to Revenue Ratio (Distance to threshold – percentage points) (Distance to threshold – percentage points) 200 25 Safety margin 20 Safety margin 150 expanded expanded 2017 DSAs 2017 DSAs 15 100 10 50 Safety margin Safety margin eroded 5 eroded 0 0 0 50 100 150 200 0 5 10 15 20 25 2013 DSAs 2013 DSAs Source: LIC-DSA Database. Notes: 1/ Distance to the threshold defined as the difference between the maximum value for a debt burden indicator (under the DSA baseline scenario) over the period 2017-26 and the relevant DSF indicative threshold. 2/ Yellow and green dots represent countries currently at moderate and low risk of debt distress whose risk rating did not change between 2013 and 2017. The sample excludes countries where the debt-carrying capacity changed over that period. 28. Debt related risks in some IDA countries not covered under the DSF are elevated, but there is variation among countries. Countries with large external financing needs, a reliance on volatile capital inflows and low reserve adequacy ratios are particularly vulnerable without adequate buffers such as sovereign wealth funds. Also, countries with a high stock of public debt that is subject to market risks 30 Excludes countries that experienced a reclassification of their debt-carrying capacity over the period. - 18 - (interest-rate refixing, refinancing and exchange rate risks) and a large fiscal sustainability gap31 are likely to be at heightened risk. III. ACTIONS TO HELP ADDRESS DEBT-RELATED RISKS 29. The primary responsibility for addressing debt vulnerabilities lies with borrowers. Vulnerable countries need to take determined action to contain debt-related risks. Countries need to have a sound macro-economic policy framework. Also, to ensure that risks are detected and addressed, countries would need to strengthen public debt recording, monitoring and reporting, and enhance their debt management capacity. Building capacity to identify and manage fiscal risks from contingent liabilities would be important as many countries are embarking on large public infrastructure investments through SOEs and are making increased use of PPPs. New borrowing should be considered only for investment projects with credibly high rates of return and using fiscal risk management tools. Lowering debt vulnerabilities will also hinge on enhancing domestic revenue mobilization, improving the efficiency of public spending and the implementation of prudent fiscal and debt management medium-term strategies, while building external and fiscal buffers and promoting growth-enhancing structural reforms. 30. Tailored policy reforms that reflect country-specific vulnerabilities will be important. Commodity-dependent countries would need to take active steps to better insulate themselves from volatile commodity prices, diversify their economies and improve external competitiveness. Elevated public debt levels in countries with fixed exchange rates may require a different set of risk management policies than in countries with flexible exchange rates (e.g., greater reserve buildup). Policy options may differ for countries facing solvency or liquidity constraints. Countries with high level of SOE debt and PPPs may benefit from strengthening corporate governance and improving fiscal risk management. 31. Creditors, under the leadership of borrowers, have also a role to play in containing debt vulnerabilities. The Addis Ababa Action Agenda calls for debtors and creditors to work together to prevent and resolve unsustainable debt situations.32 Against this background, creditors should aim for good practice in lending, drawing on principles for sustainable lending such as those being championed by the Group of Twenty (G-20).33 32. The World Bank is working with the IMF to implement a multi-pronged approach to strengthen the support they provide to help countries address debt vulnerabilities.34 As detailed below, the work program is organized around four areas. The work program under the multi-pronged approach will be complemented by additional support for the strengthening of fiscal frameworks (including domestic revenue mobilization efforts, improving the efficiency of public expenditure and strengthening public investment management). a. Area 1: Strengthening debt analytics and early warning systems to help countries better understand debt vulnerabilities. The IMF and World Bank are continuing to strengthen debt-related 31 Gap between primary deficit and debt-stabilizing primary deficit. 32 See “Addis Ababa Action Agenda of the Third International Conference on Financing for Development.” 33 G20 countries have endorsed the operational guidelines to enhance access to financing for development while ensuring debt sustainability through a collaborative approach among borrowers, creditors and the international financial institutions. See “G20 Operational Guidelines for Sustainable Financing,” March 2017. 34 The multi-pronged approach on debt has been presented in the Bali Development Committee paper “Debt Vulnerabilities in Emerging Markets and Low-Income Countries.” September 2018. - 19 - analytical work and increasing the focus on debt issues, public finance and fiscal risks in analytical products: • Debt sustainability analysis (DSA). o The IMF/World Bank Debt Sustainability Framework for Low-Income Countries (LIC-DSF) has been revised, and now places greater emphasis on debt data coverage, on accounting for contingent liabilities, on analyzing customized shock scenarios, and on flagging possible optimism bias in projections. Implementation began in July 2018, supported by a new staff guidance note35 and increased training of staff and country officials. o The IMF’s methodology for assessing debt sustainability in countries with significant access to external capital markets (MAC-DSA) is also being reviewed with a view to enhancing its coverage of debt including for contingent liabilities. • Increased focus on debt issues, public finance and fiscal risks in analytical products. The World Bank plans to deepen the analysis of debt vulnerabilities and fiscal space in core diagnostics, such as public expenditure reviews, systematic country diagnostics and in special-topic reports as relevant. This will complement the IMF’s work on fiscal space assessment pilots, which is being extended to a broad range of countries.36 • Fiscal risk assessments. The IMF and World Bank will continue to roll out analytical toolkits for fiscal risk assessments from contingent liabilities, including the PPP-Fiscal Risk Assessment Model (PFRAM). In parallel, assessments of macro-financial risks and their potential fiscal and economic implications will be strengthened. • Early warning systems. The World Bank is planning to strengthen its internal early warning system on debt vulnerabilities for emerging market economies. The IMF and World Bank staff will exchange assessments and policy views on highly vulnerable countries. b. Area 2: Strengthening debt transparency to help countries have a more complete picture of their debt. The full work program in the area of debt transparency is presented in Annex 3 and in the IMF- World Bank G-20 notes on debt transparency. 37 It includes: • Raising awareness at the highest political level. The IMF and World Bank staff will use policy and technical assistance (TA) engagement to raise awareness of debt sustainability and of reform options to reduce debt vulnerabilities with governments and parliaments or equivalent legislative bodies (in coordination with the executive branch – as relevant). • Provision of TA and development of tools to build borrower capacity to record, monitor, and report debt. This will be done by tailoring TA to country specific needs in this area, while ensuring adequate TA funding (including for the Debt Management Facility, DMF) and enhancing information sharing among TA providers.38 Another initiative is the development of a tool to better monitor a country’s capacity and performance on debt recording, monitoring and reporting, building on the World Bank’s DeMPA. • Efforts to provide greater clarity about the requirements and accessibility of debt data collected and disseminated by the IMF and World Bank. There is scope to improve the 35 See IMF and World Bank, “Review of the Debt Sustainability Frame work for Low-Income Countries: Proposed Reforms,” August 2017, and “Guidance Note on the Bank-Fund Debt Sustainability Framework for Low Income Countries,” December 2017. 36 See https://www.imf.org/en/News/Articles/2018/06/26/pr18260imf-board-takes-stock-of-work-on-fiscal-space. 37 See “Highlights and Work Agenda,” June 2018, “G-20 Note: Improving Public Debt Recording, Monitoring, and Reporting Capacity in Low and Lower Middle-Income Countries: Proposed Reforms,” June 2018, and “G-20 Note: Strengthening Public Debt Transparency: The Role of the IMF and the World Bank,” June 2018. 38 The DMF is a multi-donor trust fund supporting debt management capacity building. - 20 - accessibility and user friendliness of this data and encourage countries to improve reporting and compliance with established statistical standards. • Efforts to enhance creditor outreach. The IMF and World Bank are exploring how to build on existing creditor platforms (such as the DMF, the DMF Stakeholder Forum and Multilateral Development Banks meetings) to strengthen engagement with non-Paris Club bilateral and plurilateral creditors. By expanding the scope of the DMF, the facility could be leveraged into a multilateral platform for dialogue on debt issues. The IMF and World Bank are also planning workshops for non-Paris Club bilateral creditors on debt sustainability analysis and lending frameworks. As per the recent IMF-World Bank G-20 notes on debt transparency, both institutions will also support the implementation of the G-20 Principles and Operational Guidelines for Sustainable Financing and private sector lending disclosure initiatives. c. Area 3: Strengthen capacity on debt/fiscal risk management to help countries deal with existing debt more effectively. The IMF and World Bank are scaling up efforts to address demand for more focused and expanded capacity building on debt and fiscal risk management. Activities include: • Debt management. The IMF and World Bank plan to scale up effective debt management TA, including through the DMF, to provide support for the development of national and sub-national debt management reform plans, medium-term debt strategy formulation, targeted TA support to reform implementation, and domestic debt market development. • Operational support to strengthen debt and fiscal policy frameworks and manage fiscal risks. The IMF and World Bank will conduct joint missions to help countries improve their ability to better monitor and manage fiscal risks stemming from contingent liabilities (including from SOEs and PPPs). They are also strengthening the link between debt management TA, fiscal risk management and related policy reforms anchored in IMF-supported programs and World Bank development policy operations. • Debt reduction. The World Bank will extend the mandate of the Debt Reduction Facility (DRF) for IDA-only countries and may adapt the facility’s scope to address identified implementation challenges. d. Area 4: Reviews of the IMF Debt Limits Policy and the IDA Non-Concessional Borrowing Policy. The reviews are to start in the second half of 2018. They will be informed by implementation experience and will include extensive consultations with stakeholders. The World Bank is also considering the review of operational guidelines on debt-related disclosure requirements. IV. IDA POLICY OPTIONS 33. Aligned with the multi-pronged approach, this section outlines the sustainable lending framework IDA has been implementing to help its client countries better manage and mitigate their debt risks, and policy options to enhance the impact of this framework going forward. Since the successful implementation of the HIPC Initiative and MDRI by a broad global coalition, IDA has taken the initiative to develop a framework of sustainable lending and put it into practice. IDA’s framework centers around three pillars: • The traffic light system based on forward-looking debt analysis under the joint Bank/Fund debt sustainability framework for low-income countries; • The grant allocation framework (GAF) to prevent a worsening of the debt burden due to IDA credits in low-income countries that are already at risk; - 21 - • The NCBP that seeks to preserve the gains to debt sustainability from debt relief under the HIPC Initiative and MDRI, and to ensure prudent management of scarce IDA concessional resources. 34. The focus is on encouraging and supporting early action by country authorities to identify and address the underlying determinants of elevated debt risk that may threaten progress on their overall development agenda. Debt per se is not the problem; used prudently, debt – including non- concessional debt – is a powerful tool for development. However, the returns from debt must exceed the costs: first and foremost, in terms of development outcomes, but also in terms of strict economic rates of return. Otherwise, countries can become increasingly at risk of debt distress. Therefore, the challenge for the international community as a whole is two-fold. First, to assist IDA client countries to ensure that the benefits exceed the costs of servicing their debt; and to flag situations where the costs may be exceeding the benefits early enough in order that corrective action may be taken by country authorities with support from IDA and other partners. Second, to respect the principles of debt transparency and sustainable lending that include provision of concessional financing where and when needed. IDA’s Policy Toolkit 35. Over the past decade, IDA has adapted its toolkit of financial terms and conditions, and introduced new financial instruments, to reflect the core principles of sustainable lending. The performance-based allocation (PBA) system remains the basis for allocating core IDA resources to support national development priorities as per IDA’s country-based model. IDA also provides non-core financing through a series of tailored financial instruments to respond to key specific needs of its clients. The grant allocation framework (GAF) was introduced under IDA14. It covers ‘grant-eligible’ IDA countries, i.e., active clients with non-gap IDA-only status.39 Under the GAF, the DSF debt risk rating determines the terms of concessional IDA financing: low-risk clients receive credit terms, high-risk clients receive grants, and moderate-risk clients receive a 50/50 mix of credits and grants. The terms for concessional financing for a specific, grant-eligible or MDRI recipient country can be modified by the non-concessional borrowing policy introduced in 2006. 36. As part of the IMF/WB multi-pronged approach, IDA intends to undertake a further review of its Non-Concessional Borrowing Policy in coordination with the IMF which is undertaking a parallel review of its Debt Limit Policy. The NCBP seeks to help preserve the gains to debt sustainability from debt relief under the HIPC Initiative and MDRI, and to limit moral hazard when providing concessional finance. The policy operates on the premise that concessional financing remains the most appropriate financing for LICs, while recognizing that NCB can be an important source of financing for development needs. The NCBP also seeks to ensure prudent management of scarce IDA concessional resources. By exacerbating the risk of debt distress, excessive NCB can undermine development objectives supported by IDA concessional resources and create the need for increased IDA grants in the future. In cases of exceptions under the policy, IDA can either issue waivers or take remedies by: (i) reducing allocated IDA volumes, (ii) hardening terms, or (iii) a combination of both. 37. The NCBP is reviewed periodically along with updates on NCBP actions in reports to the Board of Executive Directors. The most recent review (2015) highlighted enhancements to the NCBP’s implementation arrangements including: (i) streamlined assessment of debt management capacity, (ii) the introduction of debt ceilings for countries at low or moderate risk of debt distress, and (iii) enhanced transparency and reporting including publicly available information on the IDA website.40 These reports are publicly disclosed. 39 Countries covered under the GAF include certain countries eligible for the small islands exception that are not low-income. 40 IDA’s Non-Concessional Borrowing Policy: Review and Update, DFIRM, October 2015. - 22 - 38. Building on the lessons learned during the first decade of implementation of the NCBP, the upcoming review will be organized around a set of guiding principles. These principles could include: (i) clarity of objectives – what can be realistically achieved under the policy going forward; (ii) simplicity of procedure – in particular, what approaches to reviewing country-level NCB and possible remedies make sense under the different circumstances prevailing in IDA countries; (iii) consistency of implementation – notably ensuring fair and equitable treatment of country clients; and (iv) transparency of decision-making – including expanding public access when appropriate. 39. The review will consider the extent to which the NCBP has been effective in meeting its objectives. Perhaps the NCBP’s biggest benefit comes from its signaling effect on other creditors. A pro- active outreach effort enhances this effect by promoting the use of concessional lending to IDA countries as defined by the widely recognized 35 percent grant element benchmark. The efforts taken to improve implementation arrangements have probably strengthened the NCBP’s effectiveness in practice, as has enhanced coordination with creditors including the European Investment Bank (EIB), the African Development Bank (AfDB) which has adopted a policy on non-concessional borrowing similar to the NCBP, other MDBs, and export credit agencies of the Organization for Economic Cooperation and Development (OECD). However, the NCBP faces important limitations as well. The 2015 and earlier reviews highlighted several recurring themes, including: • Limited ability to affect borrowing decisions particularly when IDA allocations are small and other sources of concessional financing are lacking; • Limited public access to information on NCBP decisions weakens the signaling effect. 40. The review will take stock of what remedies are appropriate under different country circumstances, as well as approaches to ensure consistency and effectiveness. The 2006 Board Paper introducing the NCBP41, while noting that no one size fits all, indicated that volume cuts (that preserved the existing grant element) were more suitable when debt sustainability is fragile; while hardening terms was more suitable when debt risk is low. It also proposed providing hardened, but still concessional, terms that ‘follow the market’ for green-light countries with high market access. Finally, the 2006 paper envisaged combining volume cuts and hardened terms for yellow and red-light countries with substantial market access including those relying on collateralization of natural resources. The NCBP Review will consider whether these guidelines should be revised to take account of the substantial changes in the financing landscape for IDA countries over the past decade as summarized in Sections II above. The NCBP review will take account of incentive effects of the grant allocation framework on borrower decisions. 41. The review will examine the pros and cons of setting the default IDA terms to credits for all countries. This would send a strong signal that grants are the exception, even as the GAF remained fully in place. Grant-eligible countries at moderate or high risk of debt distress could submit a request to IDA for grant terms under the GAF. This could be accompanied by disclosure of all NCB contracted over the past year and any intention to incur new NCB over the coming year. This option would establish an operational link between grant eligibility and NCBP debt reporting obligations as set by IDA’s general conditions. 42. As part of the NCBP review, IDA will consider options for enhancing disclosure of NCBP decisions. Presently, IDA terms for each recipient country are published on the web42, as are DSAs, so informed observers can determine when NCBP remedies have been applied. Enhanced public disclosure of 41 IDA Countries and Non-concessional debt: Dealing with the ‘Free Rider’ Problem in IDA14 Gran t-recipient and post-MDRI Countries, FRM, June 2006. 42 When remedies applied under the NCBP result in changes to a country’s IDA terms, this is stated in a footnote to Annex 1 of the Bank Directive on Financial Terms and Conditions of Bank Financing. See: OPS 5.09, June 2018. - 23 - NCBP decisions would amplify the signaling effect to all creditors, thereby potentially making the policy more effective. The specific terms and conditions of disclosure would need to be carefully designed to protect applicable confidentiality requirements. Debt Transparency and Coordination 43. The impact of a revised NCBP could be greatly enhanced by anchoring it in an inclusive, global platform focused on information exchange, sustainable lending practices and effective use of concessionality. The forthcoming review of the NCBP in parallel with the IMF review of the DLP provides an opportunity to strengthen the partnership with the IMF, other multilaterals, Paris Club and other external partners to promote prudent debt management and sustainable lending practices. Recognizing the primary role of country borrowers in ensuring debt sustainability, the World Bank will deepen outreach to all creditors including non-Paris Club and private creditors. It would build on the global coordination that characterized the HIPC Initiative. A key goal would be to promote debt transparency which is a shared responsibility of borrowers (in the leading role) and creditors. This would include disclosure of public debt information by creditors in a manner similar to the World Bank. 44. The DSF, as well as the IDA policy toolkit, can only be effective in helping countries manage their debt prudently to the extent that all public debt – including SOEs, publicly guaranteed debt and other contingent liabilities – is transparently reported. IDA’s general conditions43 oblige all client countries to report long-term external debt through the Debt Reporting System (DRS) as is the case for IBRD clients as well.44 Aggregate DRS data is made public through the IDS portal and publications. For IDA countries covered by the NCBP, the general conditions also require ex-ante reporting of any intention to incur NCB as defined by the Association. While reporting to the DRS by IDA clients has generally improved, some gaps persist. Reporting under the NCBP, by contrast, is often late and sometimes ex-post. As described above in box 3, debt surprises as have occurred in several IDA countries can damage credibility and confidence. A first option would involve comprehensive outreach to all clients to ensure awareness and understanding of IDA’s debt reporting requirements. 45. IDA is considering options to strengthen client countries’ understanding and adherence to their external debt reporting obligations to IDA. To enhance client awareness, the relevant clauses on debt reporting in the general conditions could be reflected in IDA credit and grant agreements or the minutes of negotiation. To improve accountability and transparency, the annual DRS status report on compliance could be made public. Similarly, consideration could be given to preparing an annual report on the status of reporting under the NCBP. 46. IDA is stepping up efforts to promote sustainable lending practices. The IMF and World Bank actively reach out to creditors through implementation of the DLP and NCBP. The forthcoming reviews of the DLP and NCBP provide an opportunity to strengthen this outreach including promoting links between debt transparency and sustainability. Efforts are needed to expand and improve channels for sharing information among creditors including IDA’s Lending to LICs system. 47. IDA is engaged with its counterparts in developing options to deepen dialogue and coordination among MDBs. The Multilateral Development Bank Forum facilitates information sharing and coordination among multilateral creditors including on issues related to the indebtedness of low-income countries. In April 2018, for example, staff of the African Development Fund (ADF) Policy Innovation Lab presented a proposal on moderately concessional lending to ADF countries, and issues regarding the revised 43 General Conditions for IDA Financing, IDA, July 2017. 44 Client countries report quarterly to the DRS on new external debt commitments and annually on disbursements and other flows. Some countries also voluntarily report on external debt quarterly to the Bank’s QEDS database. - 24 - LIC DSF led the agenda at the Forum’s annual meeting in May. Looking ahead, members of the MDB meetings on debt are considering increasing the frequency of their formal meetings. 48. The World Bank and IMF maintain a long-standing and productive collaboration with the Paris Club, as well as dialogue with export credit agencies and the OECD. The Paris Club currently publishes annual reports on total claims by its members, organized by borrowing country and with a breakdown of ODA and non-ODA. The Paris Club could consider the benefits of including in the annual report a breakdown by member country and providing data on commitments as well as DOD. Dialogue with the OECD has helped guide the on-going effort by the DAC to operationalize its 2014 requirement that “Loans whose terms are not consistent with the IMF Debt Limits Policy and/or the World Bank’s Non- Concessional Borrowing Policy will not be reportable as ODA”.45 49. Further outreach to non-Paris Club and private creditors aims to enhance the combined impact of actions to help IDA clients manage debt risks. Non-Paris Club and private creditors are playing a growing role in providing financing to IDA countries and may benefit from increased awareness of IDA’s programs and toolkit. The Bank, in concert with the Fund, can step up engagement with these creditors in areas such as debt sustainability analysis, debt management and sustainable lending. All creditors can benefit when lending decisions are based on established principles for sustainable lending and information sharing. For official creditors, expanding membership in the Paris Club would be an important step. V. CONCLUSION AND ISSUES FOR DISCUSSION 50. This paper presents recent debt trends in IDA countries and emerging challenges. The analysis shows that following a significant decline through 2013, public debt levels in IDA countries have increased substantially in recent years. Combined with changes in the composition of debt, this has heightened debt vulnerabilities in many countries. 51. Aligned with the IMF and World Bank multi-pronged approach, IDA is considering policy options to enhance the impact of its existing framework of sustainable lending practices to help its client countries better manage and mitigate their debt risks. In coordination with the IMF which is undertaking a parallel review of the Debt Limits Policy, IDA intends to undertake further review of its Non- Concessional Borrowing Policy, in particular to enhance its signaling effect to both borrowers and creditors. The impact of a revised NCBP could be greatly enhanced by anchoring it in an inclusive, global platform focused on information exchange, sustainable lending practices and effective use of concessionality. IDA’s convening role could support this global platform. 52. The analytical work and IDA policy options will be further developed in the context of the IDA19 discussions. Based on the results of this consultative process, management will bring forward specific policy recommendations including for the IDA19 replenishment. 45 Operationalization of the IMF Debt Limit Policy and the [IDA] Non-Concessional Borrowing Policy in the ODA context, note for the DAC Working Party on Development Finance Statistics, 13-14 June, Consultation Draft, April 2018. - 25 - Annex 1: List of IDA Countries by Country Groupings and Selected Indicators Country grouping External DSA (DSF) Debt indicators Macroeconomic indicators Risk of Total public External PPG debt 4/ (2014-17 average) IDA debt 2017 Debt Out- Debt Average Country Commodity Small State Debt carrying external Current lending FCS status DSA date 2/ (% GDP) 3/ standing Multilateral Bilateral Commercial service grant element GDP Primary Dependent 1/ status capacity debt Account category 2016 (% (% of total) (% of total) (% of total) 2016 (% of new growth Balance distress Balance GDP) exports) commt. (%) Afghanistan IDA-only Yes No No Nov-17 Weak High 7.3 11.3 57.7 42.3 0.0 3.2 2.2 -0.8 -21.2 Bangladesh Gap 5/ No No No May-18 Medium Low 32.4 13.6 74.2 25.8 0.0 3.2 63.4 6.8 -1.5 -0.1 Benin IDA-only No Yes No Jun-18 Medium Moderate 54.6 26.4 81.5 16.3 2.2 3.7 56.1 4.5 -4.3 -9.2 Bhutan Gap No No Yes Jun-16 Medium Moderate 102.4 100.8 21.0 77.1 1.9 11.5 70.3 6.8 1.6 -25.6 Burkina Faso IDA-only No Yes No Mar-18 Medium Moderate 38.3 24.7 88.1 11.9 0.0 3.7 70.4 5.2 -3.2 -7.8 Burundi IDA-only Yes Yes No Apr-15 Weak High 56.7 20.0 83.7 15.6 0.7 17.2 66.3 0.2 -4.4 -12.2 Cabo Verde Blend 6/ No No Yes Mar-18 Strong High 126.0 100.0 50.9 32.9 16.2 7.0 80.7 2.3 -2.7 -5.5 Cambodia IDA-only No No No Oct-17 Medium Low 35.1 29.8 34.4 65.6 0.0 1.5 70.3 7.0 -1.6 -9.4 Cameroon Blend No Yes (*) No Jul-18 Weak High 33.8 18.0 28.6 56.2 15.1 7.2 47.7 4.8 -4.2 -3.7 Central African Republic IDA-only Yes n.a. No Jul-18 Weak High 53.4 34.1 45.6 47.1 7.3 11.2 3.7 1.2 .. Chad IDA-only Yes Yes (*) No Jul-18 Weak High 52.5 17.5 46.0 42.2 11.8 1.5 76.0 0.1 -1.4 .. Comoros IDA-only Yes No Yes Jun-18 Weak Moderate 28.4 25.7 50.8 49.2 0.0 6.4 0.0 1.9 -1.5 .. Democratic Republic of the Congo IDA-only Yes Yes No Oct-15 Weak Moderate 15.7 13.8 56.1 43.8 0.0 4.5 75.0 5.6 -0.6 -4.2 Republic of Congo Blend Yes Yes (*) No Jul-15 Weak Moderate 119.1 46.4 9.9 60.9 29.2 3.3 40.2 0.5 -16.9 -24.7 Côte d'Ivoire Gap Yes Yes No Jun-18 Weak Moderate 46.4 26.5 26.1 31.7 42.3 10.2 53.8 8.4 -1.8 -0.1 Djibouti Gap Yes No Yes Apr-17 Weak High 30.6 93.7 32.4 67.3 0.3 9.9 47.3 7.9 -12.0 -14.6 Dominica Blend No No Yes Dec-17 Strong High 87.6 49.6 52.5 31.9 15.6 9.5 65.7 0.1 3.5 -2.8 Eritrea Inactive Yes No No 131.2 n/a 75.2 20.9 3.9 0.0 .. -11.6 .. Ethiopia IDA-only No Yes No Jan-18 Medium High 56.2 30.3 38.4 35.6 26.0 21.1 69.2 9.6 -2.1 -11.1 The Gambia IDA-only Yes Yes No Jun-18 Weak In distress 123.2 50.6 76.6 22.5 0.9 16.8 72.5 2.7 -0.6 -10.3 Ghana Gap No Yes (*) No May-18 Medium High 71.8 42.9 33.2 22.2 44.6 10.3 28.0 5.0 -0.9 -7.0 Grenada Blend No No Yes Jul-18 Medium In distress 71.4 58.1 48.3 12.7 39.1 5.3 68.3 5.3 3.0 -3.8 Guinea IDA-only No Yes No Jul-18 Weak Moderate 39.7 13.9 77.0 20.0 3.0 2.3 76.6 6.6 -1.6 -17.4 Guinea-Bissau IDA-only Yes Yes No Jun-18 Weak Moderate 42.0 22.3 61.8 38.2 0.0 1.7 39.1 4.8 -2.3 1.2 Guyana Gap No Yes Yes Jul-18 Medium Moderate 50.7 33.3 69.5 28.7 1.8 3.3 53.8 3.3 -2.8 -3.4 Haiti IDA-only Yes No No Feb-17 Weak High 31.1 27.0 14.2 85.8 0.0 5.3 58.7 1.7 -2.1 -4.2 Honduras Gap No Yes No Jul-18 Medium Moderate 43.9 28.6 65.3 14.6 20.1 4.3 54.6 3.9 -0.4 -4.0 Kenya Blend No Yes No Feb-17 Strong Low 55.6 27.4 41.7 35.7 22.7 8.4 46.5 5.5 -5.1 -6.7 Kiribati IDA-only Yes No Yes Dec-17 Weak High 26.3 3.9 21.9 40.1 Kosovo Gap Yes Yes No 20.9 7.4 88.6 6.9 4.5 5.5 46.7 3.5 0.0 -7.4 Kyrgyz Republic IDA-only No Yes No Feb-18 Medium Moderate 59.1 56.7 42.4 57.6 0.0 6.4 68.8 4.2 -1.1 -12.3 Lao P.D.R. Gap No Yes No Mar-18 Medium High 62.8 46.6 20.8 62.8 16.4 7.1 44.3 7.2 -3.0 -10.8 Lesotho Gap No No No Feb-18 Medium Low 34.7 38.5 85.7 13.9 0.4 5.9 58.4 3.4 -2.7 -5.7 Liberia IDA-only Yes Yes No Jun-18 Weak Moderate 34.4 42.2 85.9 14.1 0.0 1.1 69.8 0.4 -3.6 -61.2 Madagascar IDA-only No Yes No Jul-18 Weak Moderate 37.3 27.2 83.3 13.4 3.3 3.3 70.0 3.7 -1.8 -1.7 Malawi IDA-only No Yes No May-18 Weak Moderate 59.3 33.2 76.7 23.3 0.0 4.0 71.4 3.7 -1.9 -16.6 Maldives IDA-only No No Yes Dec-17 Weak High 68.1 21.5 29.0 66.0 5.0 2.7 34.5 6.1 -5.7 -13.6 Mali IDA-only Yes Yes No May-18 Medium Moderate 35.6 26.4 75.3 24.7 0.0 3.8 52.2 6.0 -2.2 -4.7 Marshall Islands IDA-only Yes No Yes Sep-18 Weak High 28.0 0.8 3.5 -5.0 Mauritania IDA-only No Yes No Dec-17 Medium High 91.1 76.9 60.5 39.5 0.0 12.8 55.8 3.1 -1.0 -19.0 Micronesia IDA-only Yes No Yes Sep-17 Weak High 24.5 1.2 10.2 7.0 Moldova Blend No No No Dec-17 Strong Low 37.7 26.3 85.9 12.8 1.3 4.3 56.9 3.3 -0.8 -6.5 Mongolia Blend No Yes No 40.9 27.5 23.5 49.0 4.0 23.2 4.3 -6.3 -10.1 Mozambique IDA-only Yes Yes No Mar-18 Medium In distress 102.2 86.6 42.4 38.5 19.1 11.8 76.2 5.4 -5.2 -32.5 Myanmar Gap Yes Yes (*) No Mar-18 Weak Low 34.7 8.9 30.8 59.9 9.2 0.7 72.6 6.8 -1.6 -4.1 - 26 - - 27 - Annex 2: Main Constraints for Expanding Debt Coverage INSTITUTIONAL OPERATIONAL CAPACITY Fragmented responsibilities and Weak procedures and Insufficient and/or not qualified uncoordinated institutional processes on data staff arrangement administration Narrow mandate for the DMOs Weak IT infrastructure for Weak incentives for the DMOs to collect debt data for the whole debt recording and outdated to produce reliable data (limited public sector. As a result, software public scrutiny, lack of narrow coverage in debt integration in the PFM) recording/reporting Limited audits Multiple data requests from different internal/external stakeholders - 28 - Annex 3: IMF and World Bank Agenda Going Forward on Strengthening Debt Transparency Areas Main work areas Institutions Borrower • Develop tailored and targeted action plans with support of diagnostic IMF/WB capacity TA building in debt recording, • Provide TA to support implementation of needed reforms where IMF/WB monitoring and diagnostic TA has taken place and weaknesses have been identified reporting • Provide support to broaden debt coverage and strengthen legal and IMF/WB governance frameworks, and to support the implementation of appropriate tools for debt recording, monitoring and reporting, with contributions from specialized providers • Deliver adequate and effective TA by ensuring sufficient TA funding, IMF/WB including to the DMFIII facility, enhancing information sharing among TA providers, and regularly reporting on progress in the debt- related work program • Strengthen country efforts to build debt management capacity by IMF/WB identifying measures to support in IMF-supported programs and World Bank financing operations • Simplify debt recording guidance and standardize data reporting IMF/WB templates to help alleviate capacity constraints Collection and • Implement IMF’s “Overarching Strategy on Data Statistics in the IMF dissemination of Digital Age” to strengthen the collection of the broader institutional debt data and instrument coverage of existing debt databases by integrating IMF-wide work streams • Implement and scale up D4D and Financial Sector Stability Fund IMF (statistics module) to close data gaps and strengthen capacity • Implement World Bank initiatives to improve private external debt WB and public domestic debt statistics • Improve accessibility of various debt databases by providing on an IMF/WB IMF/WB website a summary of information by country; supplemented by links to published implementation status and assessment reports Public debt • Support implementation of the new LIC DSF including with IMF/WB analysis (DSA supplementary guidance on expanding debt coverage and assessing and MTDS) fiscal risks • Define options for stronger debt coverage and disclosures in the IMF review of the MAC DSA • Facilitate access to published DSA information through: an extended IMF/WB LIC DSA summary table with key debt information (beyond just rating); clearer guidance on sharing of DSA files with country authorities; a webpage listing published MAC DSAs; and a platform for voluntary sharing of DSA files by country authorities • Strongly encourage country authorities to publish MTDS to increase IMF/WB transparency Sustainable • Review of DLP/NCBP with a view to strengthening data provisioning IMF/WB lending (creditor requirement and simplify conditionality framework - 29 - outreach and • Consideration of enhanced commentary on member countries’ debt IMF debt limits) issues in context of IMF surveillance • More structured outreach to non-Paris Club and plurilateral creditors IMF/WB • Prepare and provide workshops for emerging creditors on: DSA IMF/WB analysis, lending frameworks, internal coordination of lending agencies, and external coordination in debt resolution situations • Enhanced information sharing with multilateral and plurilateral IMF/WB creditors • Clarification of perimeter of official and commercial debt, and IMF multilateral and plurilateral debt (for IMF policy purposes) Support to • Support G20 self-assessment of sustainable financing principles IMF/WB creditor initiatives • Support private sector lending disclosure initiative IMF/WB