76258 WORLD BANK INTERNATIONAL MONETARY FUND Helping Developing Countries Address Public Debt Management Challenges A World Bank-IMF Capacity Building Partnership Prepared by Staffs of the World Bank and the International Monetary Fund Approved by Otaviano Canuto and Jose Viñals March 4, 2013 CONTENTS Executive Summary ................................................................................................................... i I. Context ..............................................................................................................................1 II. Delivery of the Bank-Fund TA Program ..........................................................................5 III. Status Of Performance and Areas of Progress ..................................................................9 A. Debt Management Strategy Development .................................................................11 B. Institutional and Organizational Areas.......................................................................13 IV. Links With Broader Bank and Fund Policy Agenda ......................................................14 V. Lessons, Gaps and Challenges Going Forward ..............................................................17 BOXES 1.The MTDS program—Partnership in action...........................................................................7 2. Examples of Country Program Support for Enhanced Debt Management ..........................15 3. Market-Based Sources of Financing: Challenges ................................................................19 FIGURES 1. Recent Developments in Developing Countries: Key Indicators ..........................................2 2. Financing Costs ......................................................................................................................4 3. Bank-Fund Debt Management TA Products and Services ....................................................6 4. Scaling Up of Debt Management TA Activities ....................................................................8 5. Cross-Country Summary Assessment of Debt Management Performance .........................10 6. Selected African countries―Bond Yields ...........................................................................25 TABLES 1. Portfolio Risk in Selected Countries: Key Indicators ............................................................3 2. Progress in Strengthening PDM Capacity: A Summary ......................................................11 3. Recent debut international issuance by emerging and developing countries ......................24 APPENDIXES I. List of Developing Countries ...............................................................................................22 II. Trends in Accessing International Capital ..........................................................................23 Glossary CEMLA Center for Latin American Monetary Studies CPIA Country Policy and Institutional Assessment DeMPA Debt Management Performance Assessment DMF Debt Management Facility DMFAS Debt Management and Financial Analysis System DSF Debt Sustainability Framework ECF Extended Credit Facility FSAP Financial Sector Assessment Program HIPC Heavily-indebted poor country IDA International Development Association IMF International Monetary Fund IP Implementing partners LCBM Local currency bond market LIC Low income country MEFMI Macroeconomic and Financial Management Institute of Eastern and Southern Africa MTDS Medium-term Debt Management Strategy MDRI Multilateral debt relief initiative ODA Official Development Assistance PDM Public debt management PEFA Public Expenditure and Financial Accountability PFM Public financial management PRGT Poverty Reduction and Growth Trust TA Technical assistance TFFS Task Force on Finance Statistics UNCTAD United Nations Conference on Trade and Development WAIFEM West African Institute for Financial and Economic Management i EXECUTIVE SUMMARY In 2009, the Boards of the World Bank and the IMF jointly endorsed a capacity building program to help developing countries strengthen their public debt management frameworks. A key aspect of the program was to help developing countries implement the framework developed by staffs to formulate an effective medium-term debt management strategy (MTDS). The Boards also supported the continued use of the complementary framework—the Debt Management Performance Assessment (DeMPA)—developed in 2007, to assess the effectiveness of the broader institutional arrangements for public debt management. This paper provides an update on the implementation of the program since its endorsement in 2009. The lessons of the global financial crisis reinforce the importance of the debt management agenda. Although developing countries have been relatively insulated from the direct effects, the crisis highlighted the importance of managing sovereign debt well. In addition, the scale of investment needs, coupled with potentially reduced access to concessional financing, means that many developing countries will become increasingly reliant on non-concessional and market- based financing. This will potentially change the risk profile of their debt portfolios significantly. Consequently, it will be imperative that countries have sufficient analytical capacity to design robust debt management strategies, leverage these new financing sources effectively, and manage the resultant portfolio risks. Since 2009, 59 countries have received technical assistance under this program. Individual country missions have been supported by extensive knowledge transfer through training and workshops. In some instances, the main pillars of the program—the MTDS and DeMPA—have been complemented by support to develop debt management reform plans or supplemented by other targeted technical assistance, including efforts to develop domestic government securities markets. A range of technical assistance providers have partnered in these efforts. Feedback from recipient governments has been very positive, and demand is growing. Country authorities have indicated that the MTDS framework is useful and sufficiently flexible to accommodate country- specific issues; some have already published a formal debt management strategy. Experience to date indicates that Bank and Fund efforts are having a positive impact, though much work remains to be done. This is also recognized in the preliminary findings of the external evaluation of the Bank’s Debt Management Facility that is being finalized. Furthermore, evidence from follow-up DeMPAs show clear gains in organizational structures, legal frameworks and borrowing procedures. There has been significant progress in the integration of debt management responsibilities and data reporting and recording, and improved procedures to mitigate operational risks. Nevertheless, despite progress in these areas, the overall DeMPA results across the full sample highlight ongoing weaknesses in critical areas. ii The implementation experience suggests a number of lessons, with strong country ownership critical for success: • Debt management reform and capacity building is complex and takes time, requiring the sustained commitment of senior policy makers. Building the capacity of debt management offices often entails significant legal, institutional and operational changes, especially given the persistence of fragmentation challenges. • Building analytical capacity remains a fundamental challenge and is especially important as debt portfolios become more complex and the need for the authorities’ to develop and implement effective debt management strategies becomes more acute. • Progress has been more sustained where it has been complemented by related development solutions in a Bank loan or design elements of a Fund program arrangement, and the corresponding engagement of Bank and Fund country teams. • In a number of cases, exogenous factors (such as natural disasters and conflicts) have impeded efforts to improve the institutional and operational capacity for debt management. The agenda will remain challenging. The importance of this work program will increase as countries expand their access to non-concessional and market-based sources in both domestic and international capital markets. Both the Bank and Fund remain committed to supporting these efforts. I. CONTEXT 1. The importance of strengthening public debt management (PDM) frameworks in developing countries has been explicitly recognized by the World Bank and IMF since 2007. 1 At that time, country experiences were discussed and Directors of both institutions expressed concern that developing countries faced significant challenges, not least because of institutional weaknesses and scarcity of skills combined with significant underlying macroeconomic vulnerabilities. These concerns were seen as particularly relevant given the borrowing space created by the heavily-indebted poor country (HIPC) and multilateral debt relief initiatives (MDRI), and the financing needs associated with countries’ infrastructure investment and broader development goals. 2 The Boards called for staff to make assistance for developing countries in this area a priority. 2. In 2009, staffs reported to the Boards on the progress in supporting efforts to strengthen PDM frameworks. A formal framework—the Medium-term Debt Management Strategy (MTDS) framework—to guide country authorities in their design of debt management strategies was put forward for endorsement. In addition, staff updated the Boards on countries’ observed strengths and weaknesses based on the findings of the Debt Management Performance Assessment (DeMPA) program. 3. Bank and Fund Boards endorsed the staffs’ 2009 strategy to strengthen PDM frameworks and capacity in developing countries. The strategy comprised (i) helping countries adopt and implement the MTDS framework; (ii) undertaking debt management performance assessments; and (iii) continuing the provision of other debt management and domestic debt market development technical assistance (TA) and advisory services to low- and middle-income countries. 4. Developments since 2009 have brought to the fore the critical interconnections between sound PDM and economic and financial stability, and heightened the importance of this agenda. While the direct effects of the global financial crisis were mostly felt by those developing countries contemplating accessing international capital markets, the broader lessons are relevant for all. In particular, the crisis has highlighted the importance of creating resilient and reliable sources of funding for mitigating financing risks and facilitating countercyclical fiscal policies as a crisis response. The crisis has also emphasized the risks associated with 1 This paper focuses on countries that are eligible to borrow from either the International Development Association (IDA) or the Poverty Reduction and Growth Trust (PRGT). It is the third in a series of joint Bank-Fund papers that focus on progress and challenges in strengthening public debt management in developing countries and responds to the request from the IMF Board for a subsequent update on developments. See “Strengthening Debt Management Practices—Lessons from Country Experiences and Issues Going Forward”, World Bank and IMF (2007), and “Managing Public Debt: Formulating Strategies and Strengthening Institutional Capacity”, World Bank and IMF (2009). 2 These debt relief initiatives created the perception of space for new borrowing, making the need to strengthen PDM frameworks particularly urgent to ensure countries did not rebuild the risk of future debt distress. 2 overreliance on specific creditors or investors, and the importance of robust debt management strategies in mitigating such crises. 3 5. Despite the virulence and persistence of the global financial crisis, recent economic performance in developing countries has been broadly positive.4 For instance, real GDP growth is estimated at around 5 percent for the median low-income country in 2011. The robustness of the real sector outlook reflects the positive impact of countercyclical measures that many developing countries were able to adopt because of the pre-crisis build-up of buffers and the strengthening of domestic economic and market capacity. The level of public debt in developing countries has remained broadly stable (Figure 1). Overall, the interaction of past debt relief and positive economic growth has helped reduce the number of developing countries in, or at high risk of, debt distress. 5 Nevertheless, despite improvements, significant financial resources are still being used in servicing debt (on average around 12 percent of revenues), indicating the debt burden remains a critical policy constraint. Figure 1. Recent Developments in Developing Countries: Key Indicators 6 Real GDP growth Debt-to-GDP ratio in percentage in percentage 12.0 70.0 10.0 60.0 8.0 50.0 6.0 40.0 4.0 Oil producers Other commodity producers 30.0 2.0 Non-commodity producers All countries 0.0 20.0 2007 2008 2009 2010 2011 2007 2008 2009 2010 2011 Interest-to-revenues ratio Risk of Debt Distress 2/ in percentage 45% 18.0 39% 40% 37% 2007 2011 15.0 35% 32% 30% 28% 12.0 24% 25% 9.0 20% 17% 16% 15% 6.0 10% 7% 3.0 5% 0% 0.0 In debt distress High risk Moderate risk Low risk 2007 2008 2009 2010 2011 Source: IMF WEO database, IMF, staff estimates. Note data excludes Bosnia and Herzegovina, India, and Sri Lanka. 1/ The sharp decline in debt in other commodity producers reflects debt relief for both the Democratic Republic of Congo and Liberia. 2/ The percentage of countries in the sample of 50 DSAs in 2007 and 41 to October 7, 2011. 3 See discussions in “Managing Public Debt and Debt Markets through a Crisis”, IMF (2011). 4 See IMF (2012), “Managing Global Growth Risks and Commodity Price Shocks—Vulnerabilities and Policy Challenges for Low-Income Countries”. 5 See World Bank and IMF (2012), “Revisiting the Debt Sustainability Framework for Low-Income Countries”. Debt distress signals the point at which countries face repayment difficulties. 6 See Appendix I for the list of countries included in the sample. 3 6. Overall, although portfolio structures are not uniform, developing countries’ debt portfolios continue to be dominated by official sector external debt. 7 This is reflected in a relatively higher level of exchange rate risk (Table 1). However, the generally long tenors and amortizing structures of multilateral and bilateral loans mitigate this risk, while their maturity and interest rate structure also helps contain refinancing and re-fixing risks. 8 These positive elements help offset vulnerabilities in domestic debt portfolios, where the still relatively short maturity of available debt instruments results in higher refinancing and re-fixing risks. Finally, the low cost of concessional borrowing and of some semi-concessional borrowing 9 also helps offset the generally higher cost of commercial and market-based borrowing (Figure 2). A key goal of an effective debt management strategy is to help manage portfolio costs and risks by identifying the mix of debt instruments (maturities, currencies and interest rate structures) that is consistent with the authorities’ medium-term debt management objectives. / Table 1. Portfolio Risk in Selected Countries: Key Indicators1 Average Max Min Level of debt Outstanding debt/GDP (%) 39% 73% 13% Exchange rate risk Foreign exchange debt/Total debt (%) 71% 94% 50% Refinancing risk Average time to maturity (years) 11.0 17.5 6.0 Average time to maturity (domestic debt) 3.6 9.1 0.5 Average time to maturity (external debt) 13.5 20.0 6.6 Debt maturing in next 12 months (% of total debt) 15% 34% 2% Interest rate risk Average time to re-fixing total debt (years) 10.3 15.3 4.0 Debt re-fixing in next 12 months (% of total debt) 23% 56% 3% 1/ Covers central government debt in countries that received joint Bank-Fund MTDS TA missions, including Armenia, Cape Verde, Côte d'Ivoire, Ethiopia, the Gambia, Ghana, Guatemala, Kenya, Kyrgyz Republic, Malawi, Moldova, Mongolia, Mozambique, Paraguay, Rwanda, Senegal and Tanzania. Data refer to the respective date of the last MTDS mission. Source: IMF-WB MTDS TA reports. 7 World Bank and IMF (2009), “Managing Public Debt: Formulating Strategies and Strengthening Institutional Capacity.” 8 The risk that interest rates move against the sovereign when it comes time to re-fix an interest rate; this differs from refinancing risk to the extent that the portfolio contains variable (or floating) rate debt. 9 Semi-concessional or quasi-concessional borrowings are loans from bilateral or multilateral creditors that are contracted at rates less than those the country could secure in the market, but which do not meet the minimum grant element of 35 percent. 4 Figure 2. Financing Costs Average effective interest rate (%) 1/ % International Market Conditions (%) 2/ % 8 12 6 EMBI EMBI Africa 7 10 Underlying U.S. Tsy Benchmark (RHS) 5 6 8 4 5 4 6 3 3 4 2 2 Oil producers 2 1 Other commodity producers 1 Non-commodity producers All countries 0 0 0 2007 2008 2009 2010 2011 Source: Bloomberg, IMF WEO database, staff estimates. 1/ Calculated as the ratio of interest payments relative to debt stock in previous year. 2/ EMBI spreads are expressed as their percentage equivalent. 7. The importance of effective capacity to manage debt is reinforced by recent changes to Bank and Fund debt-related policies. The review of the Bank-Fund Debt Sustainability Framework (DSF) for Low Income Countries (LICs) and greater flexibility in limits on non- concessional borrowing in Bank and Fund programs reflect the recognition that countries may need to borrow on non-concessional terms to finance key infrastructure investments. The shift towards non-concessional sources of financing is further aggravated by the potential decline in official development assistance (ODA). For example, net ODA by members of the OECD Development Assistance Committee fell by 2.7 percent in 2011, with country programmable aid anticipated to stagnate between 2013 and 2015. 10 8. This paper reviews progress in strengthening debt management practices in developing countries since the program was endorsed in March 2009. It responds to the request from the IMF Board for a progress report. It focuses on the delivery and results of the capacity building program and on challenges identified through its implementation. 11 It also discusses how changes in the Bank and Fund policy frameworks support these capacity building efforts. The following sections cover: delivery of the Bank-Fund TA program, status of performance and areas 10 See the UN Secretary General Report on the International Financial System and Development (2012). 11 Initially the program was targeted at low-income countries; however, it was subsequently expanded to include some lower middle-income countries that are IDA-eligible. See Appendix I for a full list of countries included in the sample. 5 of progress, links with the broader Bank-Fund policy agenda and lessons and challenges going forward. II. DELIVERY OF THE BANK-FUND TA PROGRAM 9. Since 2009, Bank and Fund staffs have provided debt management support through a range of TA channels (Figure 3). An integrated architecture of instruments has been developed that has proven to be flexible enough to meet the large variety of individual country situations, including taking a programmatic approach when needed. At the core of the engagement is an understanding of countries’ needs, which can be achieved through (i) a formal diagnostic, such as the DeMPA; (ii) a focused TA mission, including, for example, a joint Bank-Fund MTDS mission; (iii) a more general TA mission; or (iv) country self-assessment. Once needs are identified, a capacity building action plan can be developed, for example, through a Bank-assisted Debt Management Reform Plan mission. Tailored TA on key analytical and operational issues can then be delivered to help implement specific elements of the action plan. For instance, the Bank and Fund have provided tailored TA on the legal and institutional aspects of debt management that are critical for enabling the design and implementation of an effective debt management strategy, while other TA partners have helped address deficiencies in debt recording, another critical enabling element. 12 10. The key components of the work program are: • The Debt Management Performance Assessment (DeMPA), which assesses the strengths and weaknesses of a country’s institutional arrangements for PDM. • The MTDS program, which provides technical and operational guidance on how to determine a medium-term debt management strategy that targets a composition of the debt portfolio consistent with the government’s preferences with regard to the cost-risk trade-off and long- term debt sustainability. • The Debt Management Reform Plan which sets out a detailed and sequenced country-owned action plan that aims to alleviate the weaknesses identified in PDM institutions and operations. It details expected outputs and outcomes, actions, sequencing and milestones. 13 • Targeted capacity building and TA, that can be provided on request in specific areas of PDM (e.g., drafting a legal framework to underpin borrowings and guarantees). These activities are customized and sequenced according to country specific needs, priorities, and absorption capacity. 12 See “Developing a Medium-Term Debt Management Strategy (MTDS)—Guidance Note for Country Authorities”, World Bank and IMF (2009) for a discussion of the enabling framework. 13 In the case of Bank-assisted Debt Management Reform Plans, it will also provide an estimate of budget and resources required to implement the plan. 6 11. These activities complement one another in supporting overall PDM reform. They also complement other related capacity building and TA efforts, including financial sector TA related to the Financial Sector Assessment Program (FSAP), other domestic debt market development initiatives, such as the Global Emerging Markets Local Currency Bond Program, or targeted Fund TA and other macro-fiscal TA related to strengthening public financial management (PFM). Overall, these activities are supported by broad knowledge sharing activities, including an extensive program of training and outreach, and the dissemination of related policy, research and operational papers. 14 Figure 3. Bank-Fund Debt Management TA Products and Services DeMPA MTDS Reform Plans Other Fund- Bank TA 12. Under this overarching program, 59 developing countries have received some assistance from the Bank and/or Fund to support capacity building efforts in the area of debt management. 15 Depending on countries’ needs and preferences, support has either involved a discrete engagement or more comprehensive and sustained follow-up programs (see Figure 4 and Background Paper, Table 1). The pace of delivery of the DeMPA and MTDS programs has been broadly in line with that originally envisaged. 16 Collaboration and coordination between the Bank and the Fund staff in the area of PDM has been reinforced by the joint delivery of the MTDS TA 14 This includes activities supported by the Bank’s Debt Management Facility (such as the Debt Managers’ Practitioners’ Program (DMPP), the Debt Managers’ Peer Network, and the Annual Stakeholders’ Forum), along with the joint regional workshops on the MTDS framework. 15 See chapter 1 of the accompanying background paper for a detailed analysis of the delivery of the TA program. 16 The initial program envisaged up to 20 DeMPAs per year and applying the MTDS framework in 4-6 countries (implying up to 12 missions taking into account the anticipated need for a follow-up mission) per year; this level of activity was costed in the 2007 paper. On average, 13 DeMPAs have been undertaken and 7 MTDS missions delivered per year. During the same period an average of 6 Debt Management Reform Plan missions following up on the DeMPAs – that were not costed in the 2007 paper - were delivered, reflecting a shift in country demand from diagnostics to support for developing action plans. Note these figures also do not include DeMPAs or MTDS missions delivered beyond the set of developing countries identified in Appendix I. 7 program. The delivery and positive impact of this overall TA program illustrates the benefits of a broad-based partnership with country authorities and other TA providers (Box 1). 17 The positive impact and partnership benefits have also been recognized in the early findings of the forthcoming external evaluation of the Bank’s Debt Management Facility (DMF). Box 1.The MTDS program—Partnership in action Collaboration and coordination between the Bank and the Fund in the area of PDM has been reinforced by the joint delivery of the MTDS related TA program. This builds on the long history of collaboration on these issues, such as the drafting of the joint “Guidelines for Public Debt Management” and follow-up publications, as well as related TA activities. Bank and Fund staffs continue to work together to enhance the quality of the program. In particular, the Bank and Fund MTDS Practitioners Group meets regularly to coordinate mission delivery, brainstorm on technical and analytical issues and form a consensus on key issues. These activities complement other coordination efforts covering broader TA requests in the area of debt management and debt market development, which have been recognized as one area of progress in SM/10/54, “Implementation of the Joint Management Action Plan on Bank-Fund Collaboration”. Partnership with country authorities has proved critical in the effectiveness of MTDS missions. Bank and Fund staffs have built on their shared experience to enhance the effectiveness of country- level TA interventions. In particular, mission teams have adapted their approach to more actively engage TA-recipients in the hands-on design of the analyses and reports. For instance, in the initial stages of the program, the Bank-Fund team would present the outcome of the strategy analysis at the conclusion of the mission; now, technical counterparts are encouraged to take responsibility for presenting the analysis to senior policymakers, with the support of the Bank-Fund team. These changes have contributed to a ‘learning by doing’ culture and enhanced country ownership, fostered the active use of MTDS analysis by country authorities, and further facilitated the production of a formal debt management strategy. In addition to Bank-Fund collaboration, partnership with other providers of TA has proved beneficial. The MTDS program was designed to include the participation by other international and regional TA providers in debt management, such as the Center for Latin American Monetary Studies (CEMLA), the Debt Management Section of the Commonwealth Secretariat, the Debt Management Program of the United Nations Conference on Trade and Development (DMFAS-UNCTAD), Debt Relief International, the Macroeconomic and Financial Management Institute of Eastern and Southern Africa (MEFMI), and the West African Institute for Financial and Economic Management (WAIFEM). Their participation is supported through the DMF, which has the coordination in service delivery with these implementing partners (IPs) as a key objective.1/ This collaboration and continuous engagement helps build the capacity of IP staff and consultants, and enhances the quality and consistency of advice to client and developing countries while also reducing duplication. 1/ Launched in November 2008, the DMF for Low-Income Countries (LICs) is a grant facility financed through a multi-donor trust fund to support the scaling up and accelerated implementation of the Bank's debt management work program in DMF eligible countries. To date the DMF has supported work in over 50 developing countries See http://go.worldbank.org/BO5WOYO460. 17 Of those 23 countries that have received an MTDS TA mission, several have published a formal debt management strategy (including Ghana, The Gambia, Kenya and Tanzania) while others have presented a formal MTDS to ministers, cabinet or parliament as relevant (including Cape Verde, Kyrgyz Republic, Malawi, Moldova, Mozambique, and Rwanda). 8 Figure 4. Scaling Up of Debt Management TA Activities 60 Type of Activity 60 Regional Coverage Africa DeMPA MTDS Asia Reform Plans Training Europe, Central Asia and Middle East 50 50 Other Fund DM TA Latin and Central America, and Caribbean HQ- and WB-hub based training 40 40 30 30 20 20 10 10 0 0 2007 2008 2009 2010 2011 2012 2007 2008 2009 2010 2011 2012 Source: Bank and Fund staf f . 13. Building on this experience, the Bank has expanded its debt management TA program to strengthen capacity at the sub-national level. This responds to increased demand from sub- national entities that are borrowing either in the markets or from bi- and multi-lateral sources. To reflect the critical differences between sovereign and sub-national debt management, staff developed a sub-national DeMPA tool 18 that takes into account the specific inter-governmental framework, and the fiscal and procedural rules of the country on incurring debt. 19 14. The involvement of other TA partners has supported the delivery of the program. In addition to the Bank’s IPs, a variety of short-term external experts also helped resource the delivery of the TA work program. Their participation on missions has had the added benefit of ensuring the wider population of debt management experts and consultants share a common understanding of the key issues. This means that follow-up TA, or a more prolonged TA engagement, can be effectively delivered by other experts, allowing Bank-Fund staff to focus on priority cases. 18 The tool has been subject to external review and has been piloted in Lagos and Ondo states in Nigeria, Rio de Janeiro State in Brazil, DKI Jakarta in Indonesia and the metropolitan municipality of Lima, Peru. The subnational DeMPA tool and Guide are available at http://go.worldbank.org/5AHEF2KF70. 19 For instance, in most countries sub-national debt management activities may only marginally affect monetary policies. As such, these aspects are not given prominence in the sub-national tool. 9 III. STATUS OF PERFORMANCE AND AREAS OF PROGRESS 15. The application of the DeMPA tool over the past five years provides a rich information set on the status of debt management practices in developing countries. 20 Since 2007, a total of 54 developing countries have been assessed using the DeMPA tool, about sixty percent of which have received subsequent TA under the program. The results across all countries assessed post- 2009 highlight the key analytical, institutional and operational challenges countries continue to face (Figure 5). 16. In general, performance in the areas of legal framework, debt recording, and coordination with monetary policy has been relatively strong. At the same time, deficiencies in key analytical functions, including development of a robust debt management strategy, assessment of cost effective and beneficial terms for external borrowing, and cash flow forecasting and cash management, have been revealed. These weaknesses have the potential to become more acute, and will often impact more adversely, as the characteristics of the debt portfolio change and become more complex. 17. Poor performance in the area of debt management strategy was generally related to the quality of the exisiting strategy and weaknesses in the associated governance process. While many countries do have some form of strategy, it is often unpublished, does not have the approval of the highest policy makers and is not supported by an institutional decision making process that ensures its regular production, implementation, and updating. Moreover, most strategies are not underpinned by robust cost-risk analysis of the debt portfolio. 21 18. Weaknesses associated with policies and procedures related to external borrowing are of particular concern given the likely shift in countries’ borrowing choices. Less than one-third of countries in the sample analyzed met the minimum requirements in this area. Scores on this indicator show that there is: (i) a low degree of assessment of the most beneficial/cost-effective borrowing terms and conditions; and (ii) a general absence of documented procedures for borrowing in foreign markets. 19. Weaknesses in operational and institutional factors further aggravate deficiencies in countries’ analytical capacity. Deficiencies in operational controls, business continuity planning, and staff responsibilities significantly increase overall risk as the range of borrowing instruments increases. The absence of effective and independent auditing of debt management policies, 20 Using a set of standardized criteria, DeMPA measures performance across the following functions: (1) governance and strategy development; (2) coordination with macroeconomic policies; (3) borrowing and related financing activities; (4) cash flow forecasting and cash balance management; (5) operational risk management; and (6) debt records and reporting. 21 Note that these results may precede country efforts to strengthen specific areas. In particular, several countries have received a joint Bank-Fund MTDS mission subsequent to their initial DeMPA and have made progress in the area of strategy development. 10 functions and performance undermines accountability and becomes an increasing problem as decisions about borrowing choices become more complicated. Figure 5. Cross-Country Summary Assessment of Debt Management Performance Source: World Bank, Results aggregated from 28 countries that conducted DeMPAs and finalized reports during 2010-12. Note: Results in each category are aggregated to show the share of the sample meeting the minimum requirements as defined in the DeMPA tool. 20. While the overall results show weaknesses in critical areas, progress is evident in those countries that received subsequent TA support. In particular, where Bank-Fund TA engagement—through MTDS missions, debt management reform plan missions or other TA—has been sustained, countries appear to show most progress. Results of DeMPA follow-up missions show improvements in many aspects of debt management, including organizational structures, the legal framework, debt management strategy development and operational risk management (Table 2). At the same time, there were hardly any improvements in the areas of audit and evaluation of debt management operations, and cash management. This highlights the importance of ensuring that senior policy makers remain committed and focused on strengthening capacity. Progress has also been more sustainable where it has been complemented by the inclusion of debt management reform elements in the design of a Bank loan or Fund program arrangement. 21. Experience also illustrates the importance of training in sustaining country authorities’ capacity to understand and utilize the debt management TA received. Since 2009, 51 training events covering a range of debt management related issues have been delivered by the Bank and Fund, training over 300 country participants during the period. In addition, hands-on training is a key element of Bank-Fund MTDS missions and plays a key role in ensuring country authorities can apply the framework and analytical tool in their specific country circumstance. 11 Training events have also promoted the development of peer networks where learning and experiences can be shared among practitioners. 22. Key areas of progress and challenges that countries continue to face are detailed below. Table 2. Progress in Strengthening PDM Capacity: A Summary1/ DeMPA Indicator Steps taken Country Strengthened legal Burundi, Haiti, Malawi, Sierra Leone, Togo, Mongolia, Sao Legal framework framework Tome Principe, Sierra Leone Upgraded or modernized Burundi, The Gambia, Ghana, Kenya, Maldives, Malawi, Managerial structure organizational structure Tanzania, Senegal Enhanced institutional framework governing Malawi, Senegal, Sierra Leone loan guarantees Debt management Strengthened middle Bangladesh, Papua New Guinea, The Gambia, Liberia, strategy office Malawi, Nicaragua, Sierra Leone. Moldova Debt management Bangladesh,Burkina Faso, Ethiopia, Maldives, Cape Verde, strategy development The Gambia, Ghana, Malawi, Mali, Moldova, based on cost-risk Mozambique, Nigeria, Nicaragua, Rwanda, Tanzania analysis Strengthened audit Audit Malawi, Mongolia function Deepening domestic Domestic Borrowing Bangladesh, Kenya market development Cash Flow Forecasting & Cash Strengthened cash Burkina Faso, Malawi, Maldives Balance management Management Segregation of Duties, Staff Reduced operational risk Bangladesh, Bhutan, Maldives, Rwanda, Togo Capacity & BCP Debt Records and Improved debt recording Bhutan, Burundi, Liberia, Burkina Faso, Malawi, Mali, Debt Reporting and reporting Mozambique, Nigeria, Togo 1/ This reflects areas where countries have taken explicit steps to enhance capacity; it does not measure the effectivensess or completeness of the steps taken. This is based on information at the time of a mission visit. Source: DeMPA and Reform Plan updates, mission staff, and country authorities. A. Debt Management Strategy Development 23. Analytical capacity remains constrained and initiatives are needed to augment incentives and retain skilled debt management staff. The results of initial DeMPAs show that some countries did have a formal debt management strategy in place, but there were deficiencies with respect to analysis of cost and risk. This is of particular concern where countries contract (or could contract) significant amounts of non-concessional or market-based debt. 24. To help address this deficiency, capacity building efforts have centered on the MTDS TA program. The objective of the MTDS TA program is to help country authorities assimilate the 12 MTDS framework for developing a debt management strategy, as well as build capacity in the use of the associated quantitative tool (MTDS Analytical Tool). 22 TA missions are complemented with workshops that provide training for country officials on the conceptual and operational aspects of debt management. Sound understanding of the framework, along with focused training, should, over time, enhance the quality of debt management strategies and strengthen portfolio risk analysis. 25. Several countries have made progress in developing a formal and explicit debt management strategy. Ghana and Kenya have published and regularly updated their debt management strategy. Others (e.g., Cape Verde, Malawi, Moldova, Mozambique, and Tanzania) recently presented their first formal debt management strategy to ministers, cabinet or parliament as relevant. Other countries (e.g., Bangladesh) have yet to publish a formal strategy, but are using the MTDS analysis internally to inform policy making. Reflecting these developments, the results of follow-up DeMPAs show progress in almost half the assessed countries (e.g., The Gambia, Malawi, and Mongolia). 26. Success in developing a debt management strategy is reflected in the improved coordination between debt management and fiscal and monetary policies. Given the explicit recognition of key policy inter-linkages in the MTDS framework, success in understanding and using the toolkit should reinforce the importance of sound medium-term fiscal, monetary and financial sector policies. In addition, by providing insight into the country’s debt dynamics, strengthened capacity to formulate debt management strategy will enhance countries’ public debt sustainability analysis. 27. The flexibility of the MTDS toolkit has helped countries better evaluate changes in the structure of borrowing. The quantitative analysis has proved especially relevant to countries that have or anticipate graduating from IDA-only access. For instance, in Malawi the quantitative tool helped quantify the potential cost-risk mix of increasing access to international capital markets versus further reliance on domestic debt markets. The quantitative tool has also allowed borrowing constraints defined in the context of an economic program (e.g., limits on net domestic financing) to be examined. 28. Country authorities have proved capable of using the MTDS quantitative tool to evaluate the impact of alternative macroeconomic scenarios on their preferred borrowing compositions. For instance, country authorities in Cape Verde, Ghana, Kenya, Nigeria, and Tanzania used the MTDS quantitative tool to examine the potential implications of contingent liabilities.23 This analysis has helped country authorities better understand debt dynamics by showing that the ranking of various strategies in terms of their cost and risk does not necessarily 22 The tool helps quantify the costs and risks of alternative borrowing strategies. In response to user feedback—both Bank and Fund staff, and country authorities—the tool has been revised to make it more user friendly and operationally more robust. The new version has been extensively tested in the field and reviewed by external peers, and has been released on the Bank and Fund websites. 23 The nature of contingent liabilities may differ—including guarantees, on-lending to state owned enterprises (SOEs) and to parastatals, and support to the banking sector. 13 change with the size of the financing requirement. 24 However, it does challenge country authorities to consider whether their preferred strategy remains feasible when the financing requirement rises. For instance, while the preferred strategy might entail meeting 50 percent of financing needs through concessional financing, limits on the availability of that financing might render that strategy infeasible when the gross financing requirement increases. B. Institutional and organizational areas 29. Work to strengthen institutional and organizational arrangements has been delivered through Bank-assisted debt management reform plan missions and other Bank and Fund tailored TA. The focus of these TA activities included helping to strengthen the legal framework (e.g., Malawi, Sao, Tome & Principe, and Sierra Leone), address technical deficiencies in key back-office functions, and develop the evaluation and audit capacities of countries’ debt operations (e.g., Bhutan and Mongolia). 30. There has been significant progress in the integration of debt management responsibilities. A number of countries have taken steps to improve the coordination of debt management responsibilities to help enhance efficiency and reduce impediments to the development of an effective debt management strategy. While similar in purpose, the approaches have taken different forms. Some (e.g., Bangladesh, Bhutan, and Togo,) set up institutional coordinating committees, while Ghana upgraded the existing debt office. Some countries (e.g., Moldova, Maldives, Malawi, Tanzania, and Vietnam) set up a front, middle and back-office structure, while others (e.g., Bangladesh and Papua New Guinea), with a more complex division of responsibilities, have tried to strengthen the middle office capacity to analyze the total debt portfolio and formulate a debt management strategy. However, fragmentation of operations remains a challenge in a number of countries (e.g., Vietnam, Pakistan, and Zambia). In addition, when the central bank is (de facto) involved in the exercise as the principal (e.g., Zambia, Bhutan, and Tanzania), lack of clarity in its role in managing domestic or external public debt may contribute to potential policy conflicts. 31. Improvements have been seen in data reporting and recording, and in procedures to mitigate operational risks. The availability of timely and good quality debt data can mitigate some of the impediments to debt management strategy formulation that arise from organizational fragmentation, and support evidence-based policy-making and transparency. Some countries (e.g., Bhutan, Burkina Faso, Mali, Mozambique, and Nigeria) improved the quality of debt data reporting and /or prepared a debt statistical bulletin presenting the key risks of their existing debt portfolios. Guinea Bissau and Tanzania strengthened the debt data recording and management software systems; while others (e.g., Maldives, Bhutan, and Togo) improved operational risk management through enhanced back-up procedures. More than half of the countries in the sample now meet DeMPA requirements for timely and accurate debt data records for both external and domestic 24 Provided the change in the level of borrowing does not change the credit risk premium. 14 debt; this has facilitated enhanced debt monitoring across developing countries. 25 However, obtaining timely data on commitments, disbursements, and loan guarantees remains a challenge for several countries (e.g., Bangladesh). And, weaknesses in debt recording have resulted in some countries accumulating debt arrears, although these weaknesses are now being addressed. In addition, while the framework for debt monitoring focuses on public and publicly guaranteed debt, country authorities should also consider the potential risks associated with private sector debt, especially external debt, and potential strategies to monitor that. 32. Progress in supporting the development of domestic government securities markets has been facilitated by improvements in cash management practices. 26 Improvement in countries such as the Maldives, Malawi, and Bangladesh centered on the publication of Treasury bills/bonds issuance calendars and borrowing procedures, and, in some instances, establishing a single treasury account or setting up a cash management committee to improve cash forecasting and enhance coordination. These developments are expected to help improve adherence to the pre- established auctions calendar and further support market development. IV. LINKS WITH BROADER BANK AND FUND POLICY AGENDA 33. The strengthening of debt management practices is often explicitly recognized as a critical component of a country’s economic program. A number of Fund programs incorporate structural conditionality aimed at enhancing the country’s debt management practices and, in particular, its debt management strategy. Improvements in countries’ institutional arrangements and operational approaches have been targeted to help sustain the benefits of policy adjustment and improved macro-economic performance (Box 2). 34. Bank-Fund policy frameworks for non-concessional borrowings allow more flexibility to countries with enhanced debt management capacity. 27 Where the latest Country Policy and Institutional Assessment (CPIA) and Public Expenditure and Financial Accountability (PEFA) assessment measures of debt management capacity conflict, staff consider both the availability and quality of the debt management strategy, plus a country’s track record of accessing market financing and managing non-concessional borrowing, coupled with the availability of timely debt data, as necessary conditions for effective debt management. This highlights the complementarity between the staffs’ TA-program and the broader policy agenda. 25 A joint IMF/World Bank public sector debt database covering 40 developing countries can be found at: http://www.worldbank.org/qpsd.html 26 These include better cash forecasting. 27 See “Debt Limits in Fund-Supported Programs—Proposed New Guidelines”, IMF (2009) and “IDA’s Non- Concessional Borrowing Policy: Progress Update”, World Bank (2010). 15 Box 2. Examples of Country Program Support for Enhanced Debt Management • In Ghana, efforts to incorporate and embed the MTDS framework have been supported by two successive program structural benchmarks related to the publication of a medium term debt management strategy in both 2010 and 2011. That has helped ensure that the analytical capacity developed with the support of Bank and Fund advisory and technical assistance missions has been fully incorporated into the operational routines of the Aid and Debt Management Department. • In Kenya, the debt management operations were fragmented and disorganized before 2003. In 2004, the Bank recommended a sequenced capacity building plan based on the previous needs assessment. The Bank has led the effort in collaboration with other implementation partners. By 2009, all but one of the proposed reform actions had been completed. Subsequently, the 2011 Article IV report praised the authorities’ prudent medium term debt strategy, despite challenges driven by the increase in domestic financing costs and the depreciation of the currency. The MEFP for the 3-year ECF arrangement, approved on January 31, 2011, incorporates the authorities’ commitment to reflect exposures to contingent liabilities, including those that arise through projects funded through public private partnerships, into the medium term debt management strategy. • In Cape Verde, given the impact of the worsening of the global outlook, Fund staff and the authorities recognized the need to enhance debt management and continue to strengthen oversight of contingent liabilities to safeguard debt sustainability. Consequently, the authorities committed to present to the council of ministers a medium-term debt strategy report discussing debt levels, composition and desirable trends as a structural benchmark under the policy support instrument (PSI). This commitment was met, with the presentation of the country authorities’ first medium term debt management strategy in November 2011. • In Mauritania, the IMF ECF (2010) included the requirement to design an overall external debt management strategy. Given problems signaled in institutional arrangements and capacity (identified by the DeMPA 2010), the Bank, at the authorities request, sent a reform plan mission (March 2012) with a joint Bank-Fund MTDS mission following in June 2012. • In Burundi, the IMF ECF (2008) included the requirement to undergo a DeMPA and develop an action plan to increase staff capacity. Accordingly, to ensure sustained engagement, the DeMPA mission in April 2012 was followed by a debt management reform plan mission in July 2012. • In Comoros, the Bank-Fund HIPC Decision Point document (2010) included three triggers on debt management: installation of debt management software, publication of reports, and alignment of the debt management office mission with staff capacity. To support this agenda, a DeMPA followed in 2011, with a debt management reform plan mission in 2012. • Other IMF programs also incorporate measures focused on the development and publication of a strategy as part of the core structural conditionality. Examples include Kyrgyz Republic (2011) and Burkina Faso and Côte d’ Ivoire (both 2012). 35. The recent enhancement of the Bank-Fund DSF for LICs has strengthened the links with the MTDS work. The MTDS program, with a main focus on debt composition complements 16 the LIC DSF which focuses on the sustainability of debt and evaluates the risk of debt distress. The outcome of the MTDS analysis supplements a country’s debt sustainability analysis (DSA) by providing a more granular assessment of how the characteristics of different financing instruments and the debt redemption profile can transmit shocks into the stock of debt and be reflected in interest payments in the medium term. 28 36. Improvements in debt management play an important role in enhancing overall public financial management (PFM) at the country level. The design of an effective debt management strategy requires consistency with the medium-term fiscal framework. Furthermore, debt management is strongly linked to core aspects of PFM as it requires predictability in the budget process and of fiscal outturns, as well as sound cash management practices. Recognition of these complementarities and of the relevance of integrating debt management institutional reform into PFM capacity building programs has helped improve the downstream implementation of debt management TA. For example, in Bangladesh a project on implementing downstream PFM reforms included debt and cash management as core components. Similar projects to ensure downstream implementation and strengthening of debt management functions are being developed in Vietnam and Tanzania, among others. Likewise, linking with the PEFA yielded positive externalities in the case of Nigeria, where the DeMPA findings informed the PEFA exercise. 37. Work on debt management practices is complemented by a number of other initiatives led by the international community. In 1998, the Inter-Agency Task Force on Finance Statistics (TFFS) reconvened to enhance the quality of external debt and international reserves data. To improve debt data reporting, the TFFS prepared the External Debt Statistics Guide (2003); this is currently being updated, with the pre-publication version expected to be released in 2013. And to complement that, the TFFS published the Public Sector Debt Statistics Guide in December 2011. 38. Enhanced debt management also plays a critical role in supporting broader initiatives to develop domestic debt capital markets. An effective and published debt management strategy should provide a medium-term perspective on the volume and type of new domestic debt; this, coupled with clear operational procedures, provides a degree of transparency that can support more active markets. In turn, deeper and more liquid domestic debt markets increase the range of feasible debt management strategies, enabling country authorities to achieve different cost-risk trade-offs. Recognizing the importance of developing domestic debt markets for enhancing financial stability and economic growth, the G20 heads of state endorsed an action plan to support the development of local currency bond markets (LCBMs) in November 2011. Both the Bank and Fund are actively supporting this initiative, which will complement their respective debt management work. 28 This complements other efforts to enrich the DSA, such as recent work to explore the impact of public investment surges in LICs. See, Buffie, E., et al, (2012) “Public Investment, Growth, and Debt Sustainability: Putting Together the Pieces”, IMF Working Paper 12/144. 17 V. LESSONS, GAPS AND CHALLENGES GOING FORWARD 39. The implementation of the program suggests a number of relevant lessons. Debt management reform and capacity building is complex and takes time. Country authorities’ capacity to develop and implement effective debt management strategies is affected by both their analytical capacity and the enabling institutional structure. Lessons from the past three years of program implementation include: • Effective debt management cannot be seen in isolation. Building robust capacity requires efforts to strengthen practices across a number of parallel work streams, and often involves organizational and legal changes that require sustained commitment of senior policy makers. • Capacity building requires on-going support. To encourage country ownership and policy makers’ ongoing commitment, efforts are made to engage and coordinate with Bank and Fund country teams on the delivery of the TA work program. Close involvement of the country team ensures not only a continued connection with the authorities, but helps place the various debt management reforms into the broader context of the country program to ensure their further implementation. This also helps ensure debt management reform plans and specific TA are tailored to country-specific requirements and priorities. • The role of effective debt management within broader Bank and Fund policies has supported the absorption of TA efforts. Progress has been more sustained where it has been complemented by related design elements in a Bank loan or Fund program arrangement, or where TA engagement has been more extensive. This is reflected in their follow-up DeMPAs. For example, progress is evidenced in the legal framework (e.g., Haiti, Sierra Leone and Sao Tome & Principe), debt management strategy development (e.g., Cape Verde, Ghana, Kenya, Malawi and Nigeria), and operational risk management (e.g., Bangladesh and Bhutan). • Strong country ownership of the reform effort is critical for success. This has both increased demand for, but also helped secure commitment to, capacity building. It is critical that support is prioritized to countries where senior policymakers are committed to follow-through. • Demand for debt management TA is growing significantly. This partly reflects an increase in countries’ awareness of the MTDS and DeMPA frameworks, and their links to the debt limits and non-concessional borrowing policies, but also the broader recognition of the significance of debt management challenges. 40. Capacity constraints and institutional fragility remain acute in many developing countries. Exogenous factors and prioritization conflicts have also impeded efforts to improve the institutional and operational capacity for debt management. This argues for continued support from the Bank and Fund if PDM policies and practices are to become fully resilient. Such progress also requires significant time and sustained commitment on the part of country authorities. 18 41. The systematic management of operational risk must be improved in most countries. Most countries do not have adequate, explicit, and well understood procedures governing their debt management operations. For example, debt transactions that are entered into and verified by the same individual increase the risk of mistakes and fraud, and put at risk the integrity of the data. Codes of conduct guiding debt management operations are often missing or rarely implemented; these are especially important where regular interface with the market is involved. These risks are amplified by staff turnover and dependence on very limited resources, with many debt offices depending on only one or two key qualified staff. Consequently, the preparation of formal procedures manuals and implementation of systematic business process has been factored in as a key element in the staff-delivered TA. 42. Overall, the current work program has made significant progress in responding to demand by authorities for improving their debt management capacity. The program has earned a reputation for facilitating the delivery of quality debt management services and products. Country authorities reported their capacity has increased as a consequence of receiving support under the program. Preliminary findings from the forthcoming external evaluation of the DMF also affirm the program has contributed to positive achievements on the ground. 43. Looking forward, the economic landscape will remain challenging. The near-term economic outlook remains broadly favorable, with positive growth rates supported by domestic demand, including from investment. The main risks to the outlook include the risk of an intensification of financial stress in the euro zone and a sharp fiscal adjustment in the U.S. 29 Vulnerabilities are accentuated by the fact that policymakers have not had time to restore policy buffers to pre-crisis levels. This would limit the scope for any counter-cyclical measures were these risks to materialize. 44. In parallel, the nature of financing challenges facing developing countries is changing. Economic growth has led to higher income levels but it also implies that some countries will graduate from International Development Assistance (IDA) eligibility. This will change the nature of available official sector financing and reduce a country’s access to concessional financing. While this will likely imply higher cost, the financial choices available will increase. Furthermore, the depth and length of the current economic crisis is increasing fiscal pressure on official donors, which might also reduce the medium- and longer-term availability of aid. 45. Substituting market-based sources of financing may be viable for some. In the current low interest environment, foreign investors are aggressively searching for yield, with strong demand for bonds from new issuers (Appendix II). 30 However, accessing the international capital markets brings its own challenges. While the direct portfolio risks—refinancing and exchange rate 29 See World Economic Outlook (Fall 2012), Sub-Saharan Regional Economic Outlook (Fall 2012) and Global Economic Prospects for Development (January 2012). 30 A record $3.75 billion has been raised by debut issuers in international capital markets in 2012, with Bolivia’s public issue being three times over-subscribed and Zambia’s offering also being about 16 times over-subscribed, reflecting high investor demand. 19 risks—are generally recognized, the indirect challenges of absorption and cost of carry are often not adequately assessed (Box 3). 31 And, in most cases, domestic markets remain very constrained in both size and tenors. This points to a need for continued progress on both the macroeconomic front (to generate greater capacity for savings) and broader financial sector development (to provide a robust regulatory environment or enhance market intermediation) if domestic debt is to become a stable source of financing. Box 3. Market-Based Sources of Financing: Challenges Access to international capital markets is often seen as a viable alternative source of financing (e.g., Kenya, Tanzania and Zambia). While market conditions—though volatile—remain relatively favorable for a number of developing countries (Figure 2 above), refinancing risks linked to “sudden stops” cannot be overstated, especially in the current turbulent global outlook. Country authorities need to carefully consider the trade-offs involved in determining the size of any issue. Access to foreign markets is often contemplated for large amounts that cannot be readily absorbed. This increases risks (including of inefficient uses of borrowed resources) and the cost of carry (i.e., the costs incurred while the funds are under-utilized).1/ However, a larger size is more likely to be liquid and will allow inclusion in key indices; these factors can broaden the investor base and reduce the direct cost of borrowing. In contrast, a smaller issue size would be easier to absorb, mitigating the cost of carry, but typically carries an illiquidity premium. Anecdotal evidence suggests that a country could pay up to 200-250 basis points on an illiquid issue. Given the scale of the potential costs and risks involved, the trade-offs need to be well understood so that country authorities can make informed choices and not simply respond to market demand. The syndicated loan market may prove a suitable funding channel in testing access to international markets.2/ The lower volumes and more tailored structures that can be accessed can alleviate some of the problems associated with the cost of carry and absorption capacity. Also, this instrument may complement eventual international bond issuance (e.g., Kenya), and help mitigate potential rollover risks. However, costs in this market can be high and maturities may not be as long as in the bond market so, again, the trade-offs need careful consideration. Countries’ ability to rely on domestic market resources tends to be constrained by the relatively low level of market development. Countries with an already-established and relatively robust local institutional investor base may be able to access long-term funding for public investment purposes (e.g., Kenya, Nigeria) in their domestic market. However, the strategy needs to be carefully designed to mitigate the risks of “crowding out” and avoid creating unnecessary obstacles to private sector activity. However, in countries where banks are the only viable investor, domestic markets remain small, with short maturities; in this case, long-term financing is likely to be limited and very costly. More broadly, enhancing domestic debt market access will require a sustained policy commitment to financial sector deepening supported by a sound macroeconomic environment. Consequently, it will take considerable time to expand capacity in a meaningful way. 1/ For instance, assume a country borrows $750 million for five years to finance a three-year investment at an all-in-cost of 5 percent. Then, given returns on safe assets are effectively zero, the carry cost could amount to $25 million in the first year, with a further carry cost of $12.5 million in year two. This means the return on the associated investment has to be equivalent to at least 6 percent per annum over the five years of the bond just to help cover this cost. The broader challenges facing countries in accessing international capital markets are discussed in Appendix II. 2/ For instance, Kenya recently accessed this market with a $500 million two-year floating rate loan. Given the then volatile 31 In addition, an over-reliance on international capital markets could undermine efforts to develop the domestic debt market. 20 macroeconomic conditions, there was a risk that a debut bond would not have been well received by the market. The loan allowed them to secure the needed financing in a low key manner that helped mitigate reputation and execution risk. It has also provided scope to resolve some key macroeconomic challenges (e.g., inflation has declined) before going to the wider market. 46. As countries move towards more commercial borrowing, a robust debt management strategy, coupled with sufficient capacity to implement it, becomes critical. To the extent available, maximizing concessional financing tends to be a preferred debt management strategy for developing countries. However, this strategy becomes less feasible as income rises. As countries substitute non-concessional and market-based financing sources, higher financing costs could contribute to a deterioration in debt dynamics. This shift in debt composition will also aggravate portfolio risks. In particular, the typical bullet-payment structure of market-based debt increases refinancing risk and, to the extent countries need to continue their reliance on external debt, exchange rate risk. 32 In this context, the analytical capacity to measure and monitor portfolio risks and evaluate effectively alternative borrowing choices becomes critical. 47. Reflecting these factors, demand for TA is likely to increase further. Limited TA resources―both skills and financial, imply competing demands will need to be prioritized so that TA resources can be used most effectively. In particular, given the criticality of country ownership for success, TA delivery will need to prioritize countries that show strong commitment to the reform process. For instance, a two-track approach could be adopted with priority given to those countries with explicit commitments set out in a Bank loan or Fund program, while also maintaining a broader pipeline to meet the needs of other countries that have set out a clear commitment to capacity building. Similarly, requests for follow-up TA could be explicitly dependent on the implementation of earlier recommendations, or other evidence of progress. In addition, a country’s financing situation could also be taken into account, especially with respect to demands for Bank-Fund MTDS TA. Here, priority could be given to countries facing a near-term phasing out of IDA-financing and / or with specific plans to borrow from international capital markets or increase significantly domestic debt. For other countries, more emphasis could be given to strengthening key elements of the enabling framework, such as debt recording or institutional arrangements, than delivery of a Bank-Fund MTDS mission. 48. The importance of the Bank-Fund TA work program will increase as countries expand their access to non-concessional and market-based sources. To ensure such developments do not aggravate underlying vulnerabilities, analytical and risk management capacity will need further development, coupled with a strengthening of institutional and governance arrangements in many countries. 49. Bank and Fund staffs are committed to providing continued support in this area. However, this will be subject to the resource constraints that both the Bank and Fund face. Work is 32 Exchange rate risk increases because of its higher concentration on one point in time and dependence on the exchange rate value at the date rather than a smoothing effect coming from the currency fluctuations over time. 21 ongoing on addressing these constraints to ensure this work program can be sustained into the medium-term. 22 Appendix I. List of Developing Countries Countries covered by the TA program described in this paper consist of all IDA-eligible countries. Oil & Gas producers Others Angola Afghanistan Madagascar Bolivia Armenia Malawi Cameroon Bangladesh Maldives Chad Benin Mali Bhutan Congo, Republic of Bosnia-Herzegovina Marshall Islands Mauritania Burkina Faso Micronesia, Fed. Sts. Nigeria Burundi Moldova Papua New Guinea Cambodia Mozambique São Tomé and Príncipe Cape Verde Myanmar Sudan Central African Rep. Nepal Timor-Leste, Dem. Rep. of Comoros Nicaragua Yemen Côte d'Ivoire Niger Djibouti Rwanda Other commodity producers Dominica Samoa Congo, Dem. Rep. of Eritrea Senegal Ghana Ethiopia Solomon Islands Guinea Gambia, The Somalia Sri Lanka Kyrgyz Republic Georgia St. Lucia Liberia Grenada St. Vincent & Grenadines Mongolia Guinea-Bissau Tajikistan Sierra Leone Guyana Tanzania Uzbekistan Haiti Togo Honduras Tonga India Kenya Uganda Kiribati Tuvalu Kosovo Vanuatu Lao People's Dem. Rep. Vietnam Lesotho Zambia Zimbabwe 23 Appendix II. Trends in Accessing International Capital Markets 50. Several developing economies have started to explore new funding alternatives to expand their financing sources. In particular, countries have accessed, or are considering accessing, international capital markets for the first time. This may be a reflection of limitations in domestic market financing in these countries. 51. The prospect of accessing international capital markets gained considerable traction prior to the crisis. Global liquidity was abundant and international investors were actively searching for diversification and yield. Improvements in underlying macroeconomic conditions in issuing countries also supported this trend. However, the crisis resulted in effectively a “sudden stop” in international capital market access in late-2008, This brought to the fore one of the key risks involved in reliance on this market—sudden stops can be unpredictable and of uncertain length. 52. Market access appears to have been reestablished. Although established emerging market issuers began to regain access early in 2009, the first low-income country to test the waters post-crisis was Senegal with its debut bond in December 2009 (Table 3). Since then, seven different sovereigns have tapped the international capital markets for the first time, issuing a total of USS 5.6 billion. Three of these issuers were African—Angola, Namibia, and Zambia. Activity in these frontier markets has been further supplemented with a new issue and debt exchange by Senegal, and a further novel issuance by Namibia (South African rand-denominated bond). 53. Despite the historical low levels of interest rates (Figure 6), these new funding sources are not without challenges. 33 • The broad macroeconomic environment needs to be supportive: Typically, a track record of good economic performance and a stable macroeconomic outlook are necessary preconditions for an issuance to be successful. • The external environment also needs to be supportive, with significant investor interest and risk appetite in evidence. • The issuer requires sufficient technical capacity. These operations tend to be highly specialized and require specific technical knowledge. Although legal and financial advisers can provide some of this expertise, the issuer needs sufficient in-house capacity in order to make the right decisions. • Transaction costs can be significant. Issuance in international capital markets involves an array of other transaction costs. These range from fees for advisory services, to engaging trustees and paying agents, to fees for any related transactions (e.g., currency swaps). 33 See Das, et al, “Strategic Considerations for First-Time Sovereign Bond Issuers,” IMF Working Paper 08/261; November 1, 2008. 24 54. In addition, there are some key risk exposures that need careful management. • Dealing with large issuance proceeds. In many circumstances, the issuance size is larger than the amount of resources needed for the country at that immediate point in time. This means that there is an associated carry cost until resources can be invested; it also increases the potential risk that resources are allocated inefficiently or wasted. • Substantial refinancing risk. The typical bullet structure of these bonds concentrates refinancing exposure. In particular, issuers are exposed to the risk of a “sudden stop” in market access when the bond is due. The closure of the market for several months post-Lehman is one example of this. Similarly, concentrating large repayments of debt at one point in time also increases the impact of any shock to exchange rates. Table 3. Recent debut international issuance by emerging and developing countries Issuance Maturity Yield at Volume Country Tenor Currency Date Date issue (%) ($ mn.) Ghana 27-Sep-07 4-Oct-17 8.500 10 750 USD Sri Lanka 17-Oct-07 24-Oct-12 8.250 5 500 USD Gabon 6-Dec-07 12-Dec-17 8.250 10 1,000 USD Georgia 7-Apr-08 7-Apr-13 7.500 5 500 USD Senegal 15-Dec-09 22-Dec-14 9.473 5 200 USD Belarus 26-Jul-10 3-Aug-15 9.207 5 600 USD Montenegro 7-Sep-10 14-Sep-15 8.000 5 254 EUR Albania 28-Oct-10 4-Nov-15 7.625 5 407 EUR Jordan 8-Nov-10 12-Nov-15 4.125 5 750 USD Nigeria 21-Jan-11 28-Jan-21 7.126 10 500 USD Namibia 27-Oct-11 3-Nov-21 5.835 10 500 USD Angola 14-Aug-12 16-Aug-19 7.186 7 1,000 USD Zambia 13-Sep-12 20-Sep-22 5.625 10 750 USD Bolivia 22-Oct-12 29-Oct-22 4.875 10 500 USD Mongolia 5-Dec-12 5-Jan-18 4.125 5 500 USD Mongolia 5-Dec-12 5-Dec-22 5.125 10 1,000 USD Average 6.930 7 607 25 Figure 6. Selected African countries―Bond Yields Ghana Gabon 10 Nigeria Namibia Zambia Senegal 9 8 7 6 5 4 3 Source: Bloomberg WORLD BANK INTERNATIONAL MONETARY FUND Helping Developing Countries Address Public Debt Management Challenges A World Bank-IMF Capacity Building Partnership Background Paper Prepared By Staffs of the World Bank and the International Monetary Fund Approved by Otaviano Canuto and Jose Viñals March 4, 2013 CONTENTS I. Delivery of the Work Program .................................................................................................1 A. DeMPA in IDA-eligible Countries .............................................................................4 B. MTDS in IDA-eligible Countries ................................................................................4 C. Bank-assisted Debt Management Reform Plans .........................................................5 D. Other Fund TA in Debt Management .........................................................................6 E. Training .......................................................................................................................7 F. Debt Management Capacity Building in Middle Income Countries ...........................8 II. Overview of the MTDS Framework .......................................................................................8 A. What is a Debt Management Strategy? .......................................................................8 B. Benefits of the MTDS Framework ............................................................................ 10 III. Overview of DEMPA Framework ....................................................................................... 11 A. Results ....................................................................................................................... 13 IV. Overview of Bank-Assisted Debt Management Reform Plans ........................................... 17 A. Introduction ............................................................................................................... 17 B. What Is A Debt Management Reform Plan? ............................................................. 17 C. Design Factors for Success in Implementation of Reform Plans .............................. 17 BOXES Box1. Delivery of Debt Management Capacity Building to Middle-Income Countries ...........9 FIGURES Figure 1. DeMPA Missions in IDA-eligible Countries by Region, Income Group and HIPC..3 Figure 2. MTDS missions ..........................................................................................................5 Figure 3. Number of DeMPA and Reform Plan Missions Implemented ...................................6 Figure 4. DeMPA results over 2007-09 to 2010-121 ..............................................................14 Figure 5. DeMPA results since 2010 .......................................................................................16 TABLES Table 1. Detailed Country Coverage of the Work Program ......................................................2 Glossary BIS Bank for International Settlements CEMLA Center for Latin American Monetary Studies CP Completion Point DeMPA Debt Management Performance Assessment DMF Debt Management Facility DMFAS Debt Management and Financial Analysis System DP Decision Point DPI Debt Management Performance Indicator DRI Debt Relief International DSA Debt Sustainability Analysis EAP East Asia and Pacific ECA Europe and Central Asia FAD Fiscal Affairs Department FSAP Financial Sector Assessment Program HIPC Heavily-indebted poor country IDA International Development Association IMF International Monetary Fund LAC Latin America and Caribbean LCBM Local currency bond market LIC Low income country MCM Monetary and Capital Market Development Department MENA Middle East and North Africa MEFMI Macroeconomic and Financial Management Institute of Eastern and Southern Africa MIC Middle income countries MTDS Medium-term Debt Management Strategy OECD Organization for Economic Co-operation and Development PDM Public debt management PEFA Public Expenditure and Financial Accountability SAS South Asia SSA Sub-Saharan Africa TA Technical assistance TFFS Task Force on Finance Statistics UNCTAD United Nations Conference on Trade and Development WAIFEM West African Institute for Financial and Economic Management I. DELIVERY OF THE WORK PROGRAM 1. The initial work program aimed to complete Debt Management Performance Assessments (DeMPAs) for up to 20 countries and deliver technical assistance (TA) on the Medium-Term Debt Management Strategy (MTDS) framework to 4-6 countries per annum. The work program was initially targeted at countries included in the heavily indebted poor countries (HIPC) initiative. 2. Both the nature of activities and the country coverage of the program expanded to reflect the nature of demand and significance of needs. A range of additional complementary activities, such as Bank-assisted Debt Management Reform Plans, and other Bank and Fund TA and training, are now provided, and country beneficiaries have been expanded to cover all IDA- eligible countries. The nature of regional support has also changed over time, with Sub-Saharan Africa being the major beneficiary in the early phases of the program, and countries in South Asia and the Middle East having seen support increased significantly in recent years. Importantly, the actual delivery of the work program is demand-driven, based on formal requests from country authorities. Table 1 shows the cumulative stock of these activities. 3. Up to end-December, 2012, fifty-nine IDA-eligible countries have received support under the program (Table 1 1). Reflecting the programmatic approach adopted by the Bank— with countries encouraged to undertake a DeMPA, and receive joint Bank-Fund MTDS and Debt Management Reform Plan missions, many countries have received assistance through multiple channels. Many countries have also chosen to complement Bank support with targeted TA from the Fund. Being demand driven, the delivery of the work program reflects the priorities of country authorities, coupled with an assessment of country absorptive capacity. 4. The delivery of the work program is discussed in more detail in the following sections. 1 Table one covers the delivery of the full work program since 2007. Sixty-two countries have received debt management TA from the Bank and/or Fund. The Bank’s contribution to the work program has received considerable support from the Debt Management Facility (DMF) since its inception in November 2008. The DMF was created to allow donors to scale up the Bank’s work in its objective to strengthen debt management capacity in IDA-eligible countries. Fund support has been met from its general TA resource envelope. Table 1. Detailed Country Coverage of the Work Program Debt Management Debt Management Country HIPC Status Country HIPC Status Activities Activities Africa (35 national governments and 2 sub-national) East Asia and Pacific (7) Angola NA F Cambodia NA D Benin Post-CP D Mongolia NA D, D(F), F, M Burkina Faso Post-CP D, D(F), M Papua New Guinea NA D Burundi Post-CP D, D(F), RP, F Samoa NA D Cameroon Post-CP D, M, M(F), RP, RP(F) Solomon Islands NA D, RP Cape Verde NA D, M, M(F) Tonga NA RP Central African Republic Post-CP D, D(F), RP Vietnam NA D, RP Comoros Post-CP D, RP Congo, Dem. Rep. Post-CP D, RP Europe and Central Asia (5) Congo, Rep. of Post-CP D, RP, RP(F) Armenia NA M Cote d'Ivoire Interim Period D, M, F Kosovo NA D Ethiopia Post-CP M Kyrgyz Republic NA M Gambia, The Post-CP D, D(F), M, RP Moldova NA D, M, M(F), RP Ghana Post-CP D, D(F), M, M(F), RP Tajikistan NA D, M, RP Guinea Interim Period D N Guinea-Bissau Post-CP D Latin America and the Caribbean (7) Kenya NA M, M(F) Bolivia Post-CP D, F Lagos, Nigeria NA D(SN) Grenada NA D, F Lesotho NA D Guyana Post-CP D Liberia Post-CP D, RP Haiti Post-CP F Malawi Post-CP D, D(F), M, RP, RP(F) Honduras Post-CP D, F 2 Mali Post-CP D, D(F) Nicaragua Post-CP D, D(F), M, M(F) Mauritania Post-CP D, M, RP St. Lucia NA F Mozambique Post-CP D, M, M(F) Niger Post-CP D Middle East and North Africa (2) Nigeria NA D, D(F), M, M(F) Djibouti NA D Ondo, Nigera NA D(SN) Yemen NA D Rwanda Post-CP D, M Sao Tome and Principe Post-CP D, D(F), RP South Asia (6 national governments and one sub-national) Senegal Post-CP D, M Afghanistan Post-CP D Sierra Leone Post-CP D, RP Andhra Pradesh, India NA RP, RP(F) Sudan Pre-DP D Bangladesh NA D, F, M, M(F), RP Tanzania Post-CP D, F, M, M(F), RP Bhutan NA D, RP Togo Post-CP D, D(F) Maldives NA D, RP Uganda Post-CP D Nepal NA D Zambia Post-CP D, D(F), M, RP Pakistan NA D Zimbabwe NA D, RP Source: World Bank WDI Database, PRMED and Fund database. Notes: D - DeMPA, D(F) - DeMPA Follow-up, D(SN) – Subnational DeMPA, M - MTDS Baseline, M(F) - MTDS Follow-up, RP - Reform Plan, F – other Fund TA. 3 Figure 1. DeMPA Missions in IDA-eligible Countries by Region, Income Group and HIPC Status EAP ECA 5% 5% ECA EAP 4% 12% LAC LAC 6% 15% MNA MNA 0% 4% SAS 5% SSA 70% 64% SAS 10% By Region: 2007-08 By Region: 2009-12 Upper- middle- Upper- income middle- 0% income 4% Low- Lower- Low- income middle- income 45% income 50% Lower- 50% middle- income 51% By Income: 2009-12 By Income: 2007-08 Non- HIPC 20% Non- HIPC 43% HIPC HIPC 57% 80% By HIPC Status: 2007-08 By HIPC Status: 2009-12 Source: Staff calculations 4 A. DeMPA in IDA-eligible Countries 5. The coverage of the DeMPA work program has expanded significantly since the first report to the joint Boards in 2007.2 A total of 83 DeMPA missions have taken place since the methodology was piloted in 2007 to end-December 2012. Out of the total, 69 were undertaken in IDA-eligible countries, including two sub-national DeMPAs in Nigeria. The pace of demand for DeMPA missions has increased over time, with the bulk (49) taking place after end-2008. Also, the DeMPA program has become more diversified in terms of its regional coverage, with an expansion tilted towards the Middle East (Djibouti and Yemen) and South Asia (Afghanistan, Bangladesh, Bhutan, Maldives, Nepal, and Pakistan). Coverage of Latin America remains somewhat limited. Furthermore, 33 out of 39 HIPC-eligible countries have received a DeMPA mission. Of these 33 countries, 29 had reached completion point, three were post-decision-point, and one was a pre- decision-point country. 6. The DeMPA work program has also evolved in two additional directions—follow-up (second) DeMPAs at the country level and DeMPAs at the sub-national level. 13 countries 3 have requested a second DeMPA to assess progress made since the first assessment. In parallel, recognizing the increased decentralization in many countries and the particular risks in sub-national debt management, staff developed the Sub-national DeMPA tool and Guide by adapting the methodology of the sovereign tool to the sub-national context. Among decentralized IDA-eligible countries, pilot missions were undertaken in the State of Lagos and Ondo of Nigeria. 4 B. MTDS in IDA-eligible Countries 7. The MTDS program has gathered significant momentum since 2009. Twenty-eight joint Bank-Fund MTDS missions have been delivered across 23 IDA-eligible countries, 5 of which 13 were post-HIPC completion point. 6 Of these, 10 were follow-up missions. 7 In terms of regional coverage, delivery of the MTDS framework has been concentrated in Sub-Saharan Africa, with 21 missions overall, and Europe and Central Asia, with four missions. Recently, demand from Latin American economies has picked up. 2 This paper is third in a series of joint Bank-Fund papers that focus on progress and challenges in strengthening public debt management in developing countries. See “Strengthening Debt Management Practices—Lessons from Country Experiences and Issues Going Forward”, World Bank and IMF (2007), and “Managing Public Debt: Formulating Strategies and Strengthening Institutional Capacity”, World Bank and IMF (2009). 3 These comprise Burkina Faso, Burundi, Central African Republic, The Gambia, Ghana, Malawi, Mali, Mongolia, Nicaragua, Nigeria, Sao Tome and Principe, Togo, and Zambia. 4 Three other pilots were conducted in Rio de Janeiro State of Brazil, DKI Jakarta in Indonesia and the Metropolitan Municipality of Lima, Peru. 5 In total, 33 MTDS missions have been delivered since 2008. 6 These include Burkina Faso, Cameroon, Ethiopia, The Gambia, Ghana, Malawi, Mauritania, Mozambique, Nicaragua, Rwanda, Senegal, Tanzania and Zambia. 7 These were Bangladesh, Cameroon, Cape Verde, Ghana, Kenya, Moldova, Mozambique, Nicaragua, Nigeria, and Tanzania. 5 Figure 2. MTDS missions By HIPC Status Baseline vs. Follow-up Missions 12 12 10 10 8 8 6 6 4 4 2 2 0 0 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 Non-HIPC Follow-up Baseline Upper middle ECA, 5 EAP, 1 income, Lower 0 LAC, 2 middle income, Low SAR, 2 15 income, SSA, 23 18 By Region By Income Group Source: Bank-Fund Database. C. Bank-assisted Debt Management Reform Plans 8. Demand for this product has shown strong growth. In 23 IDA-eligible countries, a Bank-assisted Debt Management Reform Plan mission has followed a DeMPA and / or MTDS mission, and helped design a plan to address the weaknesses identified. 8 Fifteen of these countries were in Africa. The missions focused on helping countries develop a reform plan outlining actions and their sequencing, expected outputs and outcomes, and time-bound milestones. These plans also provided an estimate of the budget and resources required to implement the plan. 9 Experience 8 In total 29 Reform Plan missions were delivered across 24 countries since 2007, including with a subnational authority. Cameroon, Republic of Congo, Malawi and Andhra Pradesh, India had two missions. 9 Guidance for conducting Debt Management Reform Plans is available at http://siteresources.worldbank.org/INTDEBTDEPT/Resources/468980-1238442914363/5969985- 1329841432655/ReformPlanGuidanceNote.pdf. 6 suggests that reform plans are best sequenced after a diagnostic assessment and hence these missions have grown in line with the DeMPA coverage (see Figure 3.) Figure 3. Number of DeMPA and Reform Plan Missions Implemented 16 15 14 14 12 12 11 9 8 7 6 5 4 3 1 1 0 2007 2008 2009 2010 2011 2012 DeMPA Reform Plan Source: Staff calculations. D. Other Fund TA in Debt Management 9. In addition to the MTDS, the Fund provides a range of other targeted TA assistance on analytical, operational and institutional aspects related to debt management in developing countries. Both the Monetary and Capital Markets Department (MCM) and the Fiscal Affairs Department (FAD) are active in this area, though each Department provides different elements of support. In addition, the Legal Department and the Statistics Department are called on in specific cases. Since 2009, in addition to the joint Bank-Fund MTDS missions, the Fund has delivered 15 missions on debt management to 11 countries, of which half were to Africa. On occasion, these activities have been delivered as part of a broader mandate to cover complementary debt market development or public financial management issues. For instance, in addition to advice on portfolio risk analysis and strategy development, the Fund has provided tailored advice on the legal framework for debt management, the organization of the debt management office, staffing requirements and profiles, policy coordination challenges (cash and debt, monetary and debt, fiscal and debt), and debt recording and statistics. 10. MCM has also delivered complementary targeted support on debt market development. Over the period in question, eight developing countries received nine focused debt market development missions; about half of these were in Africa and the rest in Latin America or the Caribbean. FAD has also provided targeted advice on how to identify fiscal risks associated with contingent liabilities in one additional instance in Africa. 7 11. These efforts have also been complemented with other TA support provided by several resident regional advisers on debt management. These advisers have been posted at the Central and Western Africa, and Middle East Technical Assistance Centers, plus in Central Asia and Azerbaijan. In addition, public financial management advisers at the Caribbean, Central American, and East African Technical Assistance Centers, plus a resident adviser at the Center for Excellence in Finance (Slovenia), have also provided support on specific debt management issues. To ensure they can provide sound advice and effectively support the overall goals of this capacity building program, advisers are actively backstopped by staff at headquarters. In a few instances, advisers have participated in joint Bank-Fund MTDS missions. Overall, the support provided by these advisers helps countries further assimilate efforts to strengthen their institutional, operational and analytical capacity in debt management. E. Training 12. Training is an important component of debt management capacity building activities. The objectives of the training are to raise awareness of debt management issues, including the need to assess funding risks, and to build in-house technical capacity, which is critical as country access to concessional funding declines. 13. Training has been provided at two levels—training of trainers and direct training of debt management practitioners. The training of trainers has targeted Bank and Fund staff, other TA providers and implementing partners. Training of practitioners has targeted country officials dealing with the country’s debt management. These trainings not only boosted their knowledge and ability to understand and utilize the debt management tools, but have also promoted the development of peer networks where learning and experience can be shared across practitioners. 14. Since 2009, 51 training events covering a range of debt management related issues have been delivered by the Bank and Fund. These have covered the DeMPA framework, joint Bank-Fund MTDS, and other debt management concepts and issues. Sub-national debt management and fiscal reform were also covered in a few instances. These events have typically been open to country authorities from both developing and emerging market economies, although some events, especially joint-Bank-Fund training on the MTDS framework, have been more focused on developing economies. 15. Twenty-three training workshops have been delivered on the MTDS framework. Most have had a specific regional focus and have often been delivered in partnership with other technical assistance providers—two with MEFMI (East and Southern Africa), two with WAIFEM (West Africa), two for Francophone Africa (of which one was with DRI), two with CEMLA (Central and Latin America), one with Asia-Pacific Finance Development Centre (Asia) and one with the East Caribbean Central Bank (Caribbean) In addition, five have been hosted at the Joint Vienna Institute with a broad country coverage. 16. These have been complemented by twenty-eight additional training events. For instance, since 2009, the Bank has provided semi-annual training courses on debt management strategy design and debt management strategy implementation at the Joint Vienna Institute and in 8 Tokyo, Japan. Over this period, more than 300 participants from 67 countries attended these courses. Similarly, the Fund has provided five additional workshops on debt management issues, including in collaboration with the Joint Partnership for Africa (1), and the Central American (3) and Middle East (1) Technical Assistance Centers. 17. The new Debt Management Practitioners Program (DMPP), supported by the DMF, has also been launched. Under the DMPP, government officials from debt management offices in DMF-eligible countries are invited to join the World Bank’s Economic Policy and Debt Department (PRMED) for three-month assignments. Selected candidates participate directly in the DMF work program, including participation in missions and training. The program provides hands- on and in-depth training in debt management issues and exposure to country work. To date, nine practitioners from Bhutan, Cambodia, Lao P.D.R., Liberia, Malawi, Nicaragua, Burkina Faso, Senegal, and Sierra Leone have participated in the program 10. Of these four debt managers from Bhutan, Lao P.D.R, Vietnam, and Uzbekistan were hosted from the World Bank’s Voice Secondment Program together with the DMPP participants. F. Debt Management Capacity Building in Middle Income Countries 18. In addition to the work undertaken in developing countries, the Bank and Fund have continued to be active in delivering TA to middle income countries (MICs) (Box 1). In general, while developing countries typically have relatively limited financing choices and constrained institutional capacity, the situation of MICs is more varied. In addition, they often have a broader set of financing options, including options for active risk management. Consequently, while the foundations of effective debt management remain the same, their TA needs are more specialized. Since March 2009, the Bank and Fund have delivered capacity building through a variety of modalities and funded with different arrangements such as trust funds and fee-based services, as summarized below. II. OVERVIEW OF THE MTDS FRAMEWORK A. What is a debt management strategy? 19. The MTDS framework, and related analytical tool, provides a systematic and analytical approach for developing an effective debt management strategy. An effective debt management strategy is a plan that the government intends to implement over the medium-term in order to achieve a desired composition of the government debt portfolio, which reflects the government’s preferences on the cost-risk trade-off. It should operationalize country authorities’ debt management objectives—e.g., ensuring the government’s financing needs and payment obligations are met at the lowest possible cost consistent with a prudent degree of risk. A strategy 10 One each from Bolivia and Myanmar are currently being hosted at PRMED under the DMPP. 9 should have a strong focus on managing the risk exposures embedded in the debt portfolio—which could translate into a non-negligible change in the cost of debt servicing, with related budgetary impact. Box1. Delivery of Debt Management Capacity Building to Middle-Income Countries Since March 2009, both the Bank and Fund have delivered capacity building to a broad range of middle-income countries (MICs), through a variety of modalities, as summarized below. World Bank Country Work1/ The approach to TA is often based on a comprehensive needs assessment (e.g., Barbados, Belarus, Jamaica, and Trinidad & Tobago) and tailored reform plans. DeMPA assessments and/or reform plans were undertaken in Belarus, Gabon, Namibia, and Swaziland. Supported by a multi-donor trust fund—the Government Debt and Risk Management Trust Fund, with initial funding provided by SECO—work has been initiated or will be undertaken in Azerbaijan, Colombia, Egypt, Ghana, Indonesia, Peru, Serbia, South Africa, Tunisia, and Vietnam. The scope of work differs across countries but typically includes: governance, institutional capacity and management of internal operations, debt management strategy and risk management (e.g., contingent liabilities). A significant part of the country work (e.g., El Salvador, Guatemala, Kazakhstan, Panama, Romania, and Ukraine) consisted of TA and/or training in the design and review of quantitative models, and the design of the debt management strategy. In some cases TA was provided as a component of broader programs, and often included use of the MTDS tool (e.g., Armenia, Dominican Republic, and Paraguay). A number of countries requested assistance in institutional capacity building and strengthening of internal operations (e.g., India, Croatia, Cyprus, and United Arab Emirates). Support for strengthening debt management in Lebanon was provided under a broader public finance reform program. Other areas of support included assessing debt management information systems (e.g. Bosnia and Herzegovina), the design of new debt management laws, (e.g., Albania and Georgia) and issues related to cash management. A few MICs (e.g., Indonesia, Chile, and Colombia), sought TA for addressing broader balance sheet risk. Technical support to sub-national governments was provided in Andhra Pradesh (India), City of Warsaw (Poland), and Rio Grande do Sul (Brazil). IMF Country Work Since March 2009, the Fund has delivered about 27 TA missions in debt management or debt market development from headquarters. This includes 5 missions delivered in partnership with the Arab Monetary Fund (AMF) in the context of the Arab Debt Market Development Initiative. In addition, there has been a regional focus on debt management in the Caribbean with TA delivered in Antigua, the Bahamas, Dominica, Jamaica, Trinidad and Tobago, and St. Kitts and Nevis, complementing the TA to Grenada and St. Lucia reported in the main paper. Debt management issues have also been central to a number of UFR-related surveillance missions, such as St. Kitts and Nevis, while broader debt management and debt market development issues have been covered in detail in the context of Article IV surveillance in 9 countries, including Belize, Bulgaria, Dubai, Jordan, and United Arab Emirates. Other TA missions to help authorities deal with specific debt market development issues included Bolivia, Mauritius, and Serbia, while debt and cash management issues have been explored in Kosovo, Malaysia, Paraguay, and Vietnam. Joint Country Work Following the model of joint Bank-Fund MTDS missions to developing countries, and reflecting the general applicability of the framework, joint TA missions on developing a debt management strategy have been delivered in Armenia, Belarus, Guatemala, and Paraguay. In addition, the Bank and Fund have coordinated some of their TA delivery in Jamaica. 1/ Note Bosnia and Herzegovina and India are also IDA-eligible. 10 B. Benefits of the MTDS framework 20. The eight-step MTDS process provides a clear framework for making informed choices on how the government’s financing requirement should be met given the authorities’ objectives. Such a systematic approach to decision-making can help strengthen the debt management function, enhance analytical capacity and help reduce operational risk even where capacity is constrained. 21. Using the MTDS framework to develop an explicit and formal debt management strategy enables the authorities to: • Evaluate the cost-risk trade-offs: The MTDS framework allows informed decisions to be made, ensuring that the costs and risks associated with alternative debt strategies are clearly identified and taken into account. Setting clear medium-term strategic goals will help debt managers avoid poor decisions made solely on the basis of cost or to satisfy immediate pressures. • Identify and manage risk: Even where financing choices are limited, the MTDS helps identify and monitor key financial risks (refinancing, interest and foreign exchange risks), and establish strategies that ensure countries are well placed to take advantage of new borrowing opportunities, in a consistent and risk conscious way. The MTDS is also a risk management tool that helps to identify, analyze and manage portfolio risks. 22. In addition, adopting the MTDS framework provides benefits with respect to: • Coordination: The structured process of developing a strategy encourages coordination with fiscal and monetary management, helping to reconcile different objectives and constraints, including on domestic debt market development, cash management and balance of payments issues. Along with enhancing coordination, it clarifies each function’s core objectives. This helps to more clearly define the responsibilities and accountability for debt management decisions separately from fiscal and monetary policies. • Identification of constraints: The MTDS analysis helps identify the constraints that limit the debt manager’s choices, allowing, where possible, the adoption of measures to ease those constraints. 23. The steps involved in designing an MTDS are set out below. Note that although these steps are presented in a specific sequence, this is only indicative. In practice, the distinction between steps may not be so clear, and several steps may be undertaken simultaneously, and/or in a different order: • Identify the authorities’ objectives for debt management and the scope of the analysis. 11 • Examine the characteristics of the current debt management strategy and analyze the cost and risk properties of the existing debt portfolio. • Identify and analyze potential funding sources, including their cost and risk characteristics. • Identify baseline projections and risks in key policy areas—fiscal, monetary, external, and market. • Review key longer-term structural factors that could affect the design of the strategy. • Assess and rank alternative debt strategies on the basis of the cost-risk trade-off. • Review implications of candidate debt management strategies with fiscal and monetary policy authorities, and their implications for the market. • Submit and secure relevant policy makers’ agreement on the strategy. 24. Once the strategy has been agreed, it should be disseminated through a published strategy document. Publication provides benefits with respect to: • Cost: A debt management strategy can potentially lower the cost of debt servicing, as an effective and transparent strategy will support domestic debt market development, facilitate the relationship with investors, creditors and rating agencies, and support efficient cash management practices. • Transparency: A formal and explicit strategy helps build broad-based support for responsible financial stewardship, enhancing governance and accountability. 25. The debt manager should develop an annual borrowing plan that is consistent with the approved strategy and reflects any assumptions regarding enhanced market access. As the borrowing plan is implemented, its impact and alignment with the pre-defined strategy should be regularly monitored and evaluated. In addition, the strategy should be reviewed on a regular basis, at least annually, or more often if macro or market conditions change significantly. This monitoring and review process is an important element of effective risk management. III. OVERVIEW OF DEMPA FRAMEWORK 26. The DeMPA tool was developed in 2007 as a comprehensive methodology for assessing public debt management (PDM) performance. 11 It provides a standard to measure performance 11 The framework was developed under the oversight of the Bank-wide Debt Management Technical Working Group, and through a broad collaborative effort with country officials, TA providers, international standard setting agencies and related stakeholders. The current product also benefited from experiences and feedback gained through field testing in Albania, Guyana, The Gambia, Malawi, and Nicaragua. Inputs provided by the IMF; Debt Relief International (continued) 12 by assessing the strengths and weaknesses in country PDM operations. This assessment can form the basis for the design of an actionable reform program, thereby helping harmonize donor support in this area. It also permits country authorities, international donors and creditors to monitor progress in strengthening PDM operations in a country over time. 27. The DeMPA diagnoses six core functions of PDM: These functions are (1) governance and strategy development; (2) coordination with macroeconomic policies; (3) borrowing and related financing activities; (4) cash flow forecasting and cash balance management; (5) operational risk management; and (6) debt records and reporting. 12 Its scope is central government PDM and closely related functions such as issuance of loan guarantees, on-lending, cash flow forecasting, and cash balance management. However, in line with international standards on debt reporting, the debt reporting indicator requires that the central government report on both central government and total nonfinancial public sector debt, as well as loan guarantees. The scoring is modeled after the Public Expenditure and Financial Accountability (PEFA) indicators. 28. The DeMPA attaches a score to each aspect of PDM. Each dimension of the DeMPA is measured across a set of benchmarks and a rating allotted, adopting a scoring methodology similar to PEFA. 13 This allows country authorities to monitor improvements over time and benchmark their performance relative to international sound practice. Emphasis is placed on meeting the minimum requirement considered to be a necessary condition for effective performance, i.e., achieving a C score for a specific dimension. Failure to meet that minimum requirement is interpreted as signaling a serious deficiency in performance and a priority area for reform. The dimensions of each indicator provide a level of detail that can form the basis for the design of an actionable reform plan. 29. A key value of the assessment is the qualitative description that justifies each score. This is captured in the assessment report. By highlighting, in detail, the specific strengths and weaknesses, the report enables the development of a prioritized PDM reform plan. However, the report stops short of making explicit recommendations for reforms or setting out an action plan as these activities require careful follow-up. (DRI); the DMFAS Programme of the United Nations Conference on Trade and Development (UNCTAD); the Debt Management Division of the Commonwealth Secretariat; the United States Department of Treasury - Office of Technical Assistance; and international standard setting bodies (e.g., OECD, PEFA and the Task Force on Finance Statistics) enhanced the applicability of the tool. The tool was subsequently revised in December 2009, based on inputs received from its operational application and suggestions received at training/outreach events from inter alia, Crown Agents, MEFMI, WAIFEM, Pole Dette, CEMLA, Asian Development Bank, and African Development Bank. (http://go.worldbank.org/5AHEF2KF70). 12 The tool is available at the web-link below and is complemented by a Guide that provides supplemental rationale and information on each indicator to assist users (http://go.worldbank.org/5AHEF2KF70). 13 An important difference relative to PEFA is the emphasis placed on meeting a minimum requirement. 13 30. The assessment is discussed with the country authorities, but it is not a negotiated document. Where the authorities’ views differ from that of the assessment, the divergence of views is reflected in the report. It is a document owned by the authorities, to be used to help improve PDM operations as they determine, and it is at the country’s discretion whether or not the assessment report is published. 31. The DeMPA missions are demand driven and not linked to any conditionality. The mission process is initiated upon receipt of a request from the country authorities. The mission findings and the report are shared with the authorities and only released at their discretion. The DeMPA scores are not related to any lending or fiduciary conditions of the World Bank as a lending institution. A. Results 32. Since 2009, staff has undertaken 49 assessments in 46 IDA-eligible countries and with two sub-national authorities. 14 As the product is demand driven, there is no pre-defined target list of countries. However, assessments have been completed in a diverse set of countries, including lower MICs, some upper MICs and those at differing stages of the HIPC process. Over three- fourths of the assessments were carried out in partnership with regional TA providers. This partnership has helped raise awareness of regional/local issues, and also called for consistency with downstream implementing activities and for enhanced capacity among TA providers. 33. Methodological changes make the early DeMPA results (2007-09) not strictly comparable to subsequent assessments. This distinction corresponds to the revision of the methodology in December 2009 to incorporate suggestions based on its operational application across a diverse set of countries. The revision resulted in the assessment criteria being clarified and/or made stricter for some indicators, more flexible for others, while a few dimensions were dropped. 15 The decline in the ratings since 2010 for the “coordination with fiscal policy” indicator is partly explained by the amendments to this indicator (Figure 4). Staff strengthened the minimum requirement from “merely having access to key fiscal variables” to “having undertaken an in-house DSA within the last three years”; thereby assessing the capacity to undertake a DSA independently. Similarly, reduced scores for the indicators on ‘debt reporting’ reflect a stricter assessment, benchmarked against the requirements for public sector debt reporting as defined by the Task Force on Finance Statistics (TFFS). 16 14 Since 2007, 69 assessments have been undertaken in 54 IDA-eligible countries. 15 The indicators that were made stricter were those relating to coordination with fiscal policy and on the reporting of total public sector debt. In the respective dimensions, the requirements for assessing fiscal variables and the debt statistical bulletin were clarified. In addition, the dimension that assessed how the debt management entity operated the government accounts was deleted (in the cash management indicator). 16 The Task Force on Finance Statistics is a joint WB-IMF-OECD-BIS standard setting body that has developed benchmarks for public debt statistics with which the DeMPA indicator is benchmarked. 14 34. Differences in country coverage also likely explain changes in scores (in 2010-12) in a number of areas. Regarding “coordination with monetary policies”, a number of countries in the pre-2010 period were part of either the Economic and Monetary Community of Central Africa or the West African Economic and Monetary Union. As such the governments were bound by strict and prudent legislation that limited the access to direct funding from the regional central banks. The expansion in coverage since 2010 has resulted in a smaller share of countries in the sample that are part of currency unions with such legislation. Similarly, improvements in the dimensions assessing the policies and procedures for domestic borrowing may in part be attributable to the higher coverage of IDA-eligible MICs in the period since 2010 (Figure 4). Countries with higher shares of domestic debt also tend to have clearly defined and market-consistent debt issuance procedures and auction mechanisms that help maintain credibility with investors. A similar explanation applies for the better performance in the indicators assessing the accuracy and completeness of debt records and evaluation of debt management operations. Also, since 2010, the sample included a larger number of non-HIPCs or HIPCs that graduated and had strengthened their debt offices. Finally, twenty-eight countries in the sample had a follow-up MTDS and/or a Reform plan mission suggesting that the programmatic and sustained engagement could also be responsible for the identified improvements. Figure 4. DeMPA results over 2007-09 to 2010-121 2007-2009 2010-2012 DPI-1 DPI-1 DPI- 100% DPI-2 DPI-15 100% DPI-2 15 DPI- 80% 80% DPI-3 DPI-14 DPI-3 14 60% 60% DPI- 40% 40% DPI-4 DPI-13 DPI-4 13 20% 20% DPI- 0% 0% DPI-5 DPI-12 DPI-5 12 DPI- DPI-11 DPI-6 DPI-6 11 DPI- DPI-10 DPI-7 DPI-7 10 DPI-9 DPI-8 DPI-9 DPI-8 Source: World Bank staff calculations Note: DPI-1: Legal Framework; DPI-2: Managerial Structure; DPI-3: Debt Management Strategy; DPI-4: Evaluation of Debt Management Operations; DPI-5: Audit; DPI-6: Coordination with Fiscal Policy; DPI-7: Coordination with Monetary Policy; DPI-8: Domestic Borrowing; DPI-9: External Borrowing; DPI-10: Loan Guarantees, On-lending and Derivatives; DPI-11: Cash Flow Forecasting and Cash Balance Management; DPI-12: Debt Administration and Data Security; DPI-13: Segregation of Duties, Staff Capacity and Business Continuity; DPI-14: Debt Records; and DPI-15: Debt Reporting. 1/ Number of countries (including follow-ups) in period 2007-09 is 35, and 28 in 2010-12. Results from finalized reports are taken while those from the sub-national DeMPAs are not included. The results are normalized to reflect uniform percentages. 35. Methodological differences apart, the results across the two periods provide key pointers and help identify common priority areas for debt management reform across 15 countries. In both periods, there have been major deficiencies in operational risk management and cash flow forecasting and management (Figure 4). Less than half of the sample met the minimum requirements for sound governance, performance of public debt management audits and the development of a robust debt management strategy. Most were not able to assess the cost-risk trade-offs of the existing debt portfolio. 36. A closer look to the 15 individual indicators and their associated dimensions permits identifying key areas of concern (Figure 5). For example, the weak operational risk management results can be explained by the absence in most of the countries of (i) business continuity planning; (ii) strong operational controls; and (iii) well-articulated responsibilities for staff. The poor performance under sound governance and debt management strategy captures the fact that very few countries in the sample had a formal debt management strategy, or effective accountability frameworks with regular performance audits, in place. These weaknesses outweighed the positive factor that the majority of countries had effective legal frameworks underpinning borrowing. In the area of cash flow forecasting and cash balance management, performance was impeded by weak forecasting of the aggregate cash balances in government bank accounts. 37. Weaknesses observed in the area of external borrowing are of particular concern. Less than one-third of the countries in the sample met minimum requirements in this area. Scores on this indicator pointed at: (i) weak assessments of the most beneficial/cost-effective terms and conditions of available borrowing options; and (ii) a generalized absence of documented procedures for borrowing in foreign markets. These findings are of particular concern given that a number of the countries in the sample have expressed interest in issuing in international capital markets in coming years. 38. The results also show that very few developing countries have sound debt management strategies. While many countries in the sample had some form of a debt management strategy, it did not have approval of the relevant policy makers and was not supported by a decision making process that ensured its implementation, regular updating, and publication. Moreover, most strategies were not underpinned by any sound cost-risk analysis of the debt portfolio. Number of countries 0 5 10 15 20 25 30 DPI-1-1 DPI-2-1 DPI-2-2 DPI-3-1 DPI-3-2 DPI-4-1 DPI-5-1 Source: World Bank staff estimates. DPI-5-2 DPI-6-1 DPI-6-2 DPI-7-1 DPI-7-2 DPI-7-3 DPI-8-1 DPI-8-2 DPI-9-1 DPI-9-2 DPI-9-3 DPI-10-1 16 DPI-10-2 DPI-10-3 DPI-11-1 DPI-11-2 DPI-12-1 DPI-12-2 Figure 5. DeMPA results since 2010 DPI-12-3 DPI-12-4 DPI-13-1 DPI-13-2 DPI-13-3 DPI-14-1 DPI-14-2 DPI-15-1 DPI-15-2 DPI-15-3 Note: Data in the charts relates to 28 IDA-eligible countries for which the reports have been finalized since 2010. B C A D N/R 17 IV. OVERVIEW OF BANK-ASSISTED DEBT MANAGEMENT REFORM PLANS A. Introduction 39. As part of its programmatic approach to strengthening debt management in developing countries, the Bank provides support in developing a debt management reform plan. This section summarizes the lessons learned from cross country work in developing debt management reform plans. These plans tend to be country-specific and vary considerably depending on the existing circumstances of the country. B. What Is A Debt Management Reform Plan? 40. The debt management reform plan is a time-bound agenda designed to put in place an effective debt management framework. The timeframe of the plan should ideally not exceed 3-5 years (medium-term) to maintain momentum and focus. The key aspects of a debt management reform plan are: • A diagnosis of the key weaknesses in government debt management institutions, functions and operations, informed by a recent and comprehensive assessment, for example DeMPA. • The scope of the reform plan can cover all government debt management activities, loan guarantees, lending, and cash balance management, or can be limited to focus on country specific priorities as discussed and agreed with the authorities. Nevertheless, if the scope of the plan agreed is seen as narrow, the plan should at least be consistent with the broader institutional setting, in order to avoid a piecemeal approach. • A prioritized and sequenced action plan to address areas requiring improvement. The plan needs to be project-related and contain details on the expected outputs and outcomes, actions, sequencing and milestones. It also provides an estimate of the resources, budget and time required to implement the plan. • A country-specific approach to realistically set the speed at which institutional and structural reforms can be implemented, and reflect the willingness of the authorities to undertake them. C. Design Factors for Success in Implementation of Reform Plans 41. Experience suggests there are six key factors that are critical for successful implementation: • Political commitment is critical, especially for institutional reform. To ensure this support, it is essential that the reform plan is developed in close coordination with country authorities. • Realism in resourcing and timeframes. Although programs aimed at developing a sound government debt management capability can take many years, there should be clear timelines and enough funding for financing the main stages of the implementation process. 18 • Country specificity and appropriateness. Reform plans are best characterized as “good fit” rather than “best practice.” These must reflect country specific priorities, prevailing political climate, technical difficulty, and capacity constraints. These plans need to be practical enough to be implementable. • Reform champion. In line with the above, it is essential to have a committed project manager in the country, a local champion who can drive the process forward. In a debt management project this support must come from the minister or the deputy minister of finance. • Integration with broader reform efforts. Debt management reforms can be more effectively sustained by integrating them into broader programs, such as public sector or public financial management reforms. Such integration helps ensure project sustainability and continuity through financing, support by experts, and project supervision. • Proper sequencing. Sequencing of the project components will differ for each country depending on the initial state of the government’s performance. Nevertheless, experience does provide some specific insight on sequencing. In particular, the 12-country pilot program concluded that: “The basic building blocks that must necessarily come first are building capacity in the back office and establishing reliable debt recording systems. These are required to ensure timely servicing of the debt, without having to rely on creditors’ notifications, and to produce accurate and frequent reporting. Beyond these steps, sequencing has been varied.” Detailed guidelines are available at: http://go.worldbank.org/CFPMWGNZ70.