Debt Transparency in Developing Economies Debt Transparency in Developing Economies © 2021 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This study is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 4.0 International License (CC BY 4.0) https://creativecommons.org/licenses/by/4.0/ Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution: Please cite the work as follows: Rivetti D. 2021. Debt Transparency in Developing Economies. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 4.0. Translations: If you create a translation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content: The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Graphic Designer: Maria Lopez / lopez.ten@gmail.com Contents Foreword Acknowledgments i iii Acronyms iv Executive Summary 1 01 Chapter 1: Debt Transparency: A Framework 9 1.1 Motivation 10 1.2 What is Debt Transparency? 14 1.3 Debt Transparency: Incentives and Disincentives 16 1.4 Current Efforts to Improve Debt Transparency 19 1.5 Conclusions 21 References 22 02 Chapter 2: Transparency in Debt Reporting 24 2.1 Debt Statistics: Direct Data Disclosure 26 2.2 Debt Statistics: Indirect Debt Data Disclosure 34 2.3 Financial Reporting 42 2.4 Conclusions 46 References 50 03 Chapter 3: Transparency of Debt Operations 53 3.1 Domestic Debt 55 3.2 External Debt to Private Creditors 61 3.3 Resource-Backed Loans 66 3.4 Central Bank Repos and Swap Operations 73 References 80 04 Chapter 4: The Role of Public Debt Management Legal Frameworks in Transparency 83 4.1 The Authority to Borrow 85 4.2 Debt Management Responsibilities 88 4.3 Public Debt Definitions and Limits 90 4.4 Disclosure Requirements 92 4.5 Audit Requirements 95 4.6 Stipulation of the Consequences of Noncompliance with the PDMLF 96 4.7 Public Availability of the PDMLF 97 4.8 Extending the PDMLF’s Scope to the Entire Public Sector Borrowing 99 4.9 Conclusions 100 References 102 Annex I: Drivers of Public Debt Disclosure: Estimation Strategy and Results 103 Annex II: WB and IMF Main Public Debt Databases 107 Annex III: Debt Restructuring and Transparency: Estimation Strategy and Results 110 Annex IV: Systems, Institutional Arrangements, and Resources for Accrual Accounting 115 Annex V: Proposal for an International Loan Repository 119 Boxes Box 2.1 Debt Restructuring: A Missed Opportunity for Transparency? Box 2.2 Impact of Including SOEs Debt in Public Debt Coverage 32 38 Box 2.3 LIC-DSA Ex-Post Debt Revisions 40 Box 2.4 Status of Sovereign Financial Reporting 43 Box 2.5 Total Government Liabilities: IPSAS vs. Direct Reporting 45 Box 3.1 Benefits of Loan Contract Standardization 62 Box 3.2 Reducing the Information Asymmetry: The Role of Credit Rating Agencies 63 Box 3.3 RBLs, Collateral, Collateral-Like Features, and Contractual Covenants 71 Box 3.4 Over-Collateralized REPO in Egypt (2016) and Ecuador (2018) 75 Box 3.5 Current Use of FXSLs 77 Box 4.1 Public Debt Definitions in Selected Countries 91 Box 4.2 Legal Consequences of Noncompliance in Selected Countries 97 Debt Transparency in Developing Economies Greater debt transparency is a vital step in the development process. It facilitates new, high quality investment, reduced corruption and provides accountability. The public debt of many low-income economies remains difficult to pin down. Foreword Debt is sometimes incompletely reported in official statistics or hidden through confidentiality clauses. Our new Debt Transparency in Developing Economies report marks the first comprehensive assessment of debt transparency in low-income countries. It presents a sobering picture. Global debt surveillance today depends on a patchwork of databases with different standards and definitions. The analysis shows that publicly available tallies of debt stocks in low-income countries can vary by as much as 30 percent of a country’s GDP because of divergent definitions and standards in local and international databases. The report also reveals that 40 percent of low- income countries have not published any data about their sovereign debt for more than two years. The report outlines a new way forward for national policymakers, debt-management offices, and international financial institutions. The need for reform is urgent in the wake of COVID-19. Declining revenues and wider public-sector deficits have increased the risk that unreported liabilities will emerge and make it difficult for these countries to service or restructure their debt. Today, 44 percent of low-income countries face a high risk of debt distress and 12 percent are already experiencing it. Countries that cannot access international bond markets increasingly rely on off- budget transactions, opaque collateralized debt instruments, and non-market-based domestic issuances—or they run arrears that go unreported. The report documents that between 2004 and 2018, resource-backed loans—which use future revenue streams as collateral—represented nearly 10 percent of total new borrowing in Sub-Saharan Africa. Among developing economies, more than 15 countries have outstanding collateralized debt, yet none has disclosed the details of collateral used to secure the borrowing. Moreover, low-income countries are starting to use central-bank repurchases and foreign-currency swaps to support external borrowing rather than as tools of monetary policy. These operations do not show up in government debt statistics, and the databases of international financial institutions do not capture them either. There is also a proliferation of debt being taken on by state-owned enterprises, SPVs, joint ventures, and other private sector entities that often carry various forms of government guarantee, including through future revenue streams, further complicating the understanding of the actual size of public debt and other fiscal space implications. i Debt Transparency in Developing Economies Even more uncertainty stems from the opacity of domestic debt markets in low- income countries. The report finds that just 41 percent of these governments use market-based auctions as the main channel to issue domestic debt. Moreover, even those that do use auctions, divulge at best spotty information to investors. The good news is that these problems can be fixed. The study lays out a detailed list of recommendations, ranked in order of urgency. Prominent among them: publishing public and publicly guaranteed debt statistics annually, encouraging coordinated data collection and reporting, and instituting integrated debt recording and management systems that align with international standards. The report also calls for creating sound national legal frameworks for public debt management to specify debt-management objectives, clarify who has authority to borrow, and spell out the full cycle of authorization. In addition, the new rules would require public debt reporting to adhere to international standards and delineate a list of permitted instruments, transactions, or sources of funding. The solutions detailed in the report flesh out the five key recommendations that I outlined in the early days of the COVID-19 crisis. First, publish loan terms and payment schedules. Second, fully disclose public and publicly traded debt, the liabilities of state-owned enterprises, and other debt-like instruments. Third, liberate borrowers from excessive confidentiality clauses so they can carry out more transparent debt reporting. Fourth, promote more prudent use of collateral and liens in sovereign borrowing. Fifth, insist that borrowers and lenders do not violate the legal requirements of other creditors, such as negative pledge clauses. These are not easy fixes, but the experiences described in the report show that they are possible. It is also clear, however, that individual countries acting alone cannot bring about changes of this scale. Progress—and the ability to prevent debt crises—depends on international cooperation to promote more effective global surveillance systems. International financial institutions, including the World Bank, have a crucial role to play—along with private creditors, credit-rating agencies, and other stakeholders. Debt Transparency in Developing Economies should end any complacency about addressing debt transparency challenges in low-income countries. The time to act is now. David Malpass President World Bank Group ii Debt Transparency in Developing Economies Acknowledgments The report was led by Diego Rivetti (Senior Debt Specialist, Macroeconomics, Trade & Investments - MTI) under the guidance of Doerte Doemeland (Practice Manager, MTI) and Marcello Estevão (Global Director, MTI), in collaboration with staffs from the World Bank’s Governance Global Practice (GOV), Finance, Competitiveness & Innovation (FCI), and the Legal Vice-Presidency (LEG). The authorship of the chapters is as follows: Chapter 1 was written by Sebastian Essl (Economist, MTI) and Diego Rivetti. Chapter 2 was written by Diego Rivetti with inputs from James A. Brumby (Senior Advisor, GOV), Sebastian Michael Essl and Vasileios Tsiropoulos (Economists, MTI), Juan Carlos Serrano-Machorro and Bonnie Ann Sirois (Senior Financial Sector Specialists, GOV), Paul Mason and Mike Seiferling (consultants). Chapter 3 was written by: • Section 3.1: Zsolt Bango (Senior Financial Sector Specialist, FCI) • Section 3.2: Andre Proite, Leandro Secunho and Diego Rivetti (Senior Debt Specialists, MTI) • Section 3.3: James Cust (Senior Economist, Africa Region), Susan Maslen (Senior Counsel, LEG), Diego Rivetti, Jyhjong Hwang and David Mihalyi (consultants) • Section 3.4: Steen Byskov (Senior Financial Officer, Treasury) and Diego Rivetti. Chapter 4 was written by Cigdem Aslan (Lead Debt Specialist, EFI), Susan Maslen, Ximena Talero (Lead Counsel, LEG) and Bilge Yalcin Bastimur (consultant). Many colleagues, inside and outside the World Bank Group, provided useful suggestions at various stages: Luca Bandiera, Rodrigo Cabral, Marie-Helene Cadoret, Anna Corcuera, Asli Demirguc-Kunt, Sebastian Eckardt, J. Clifford Frazier, Veronica Frisancho, Anna Gelpern, Susan Maslen, Paulo Moreira Marques, Ugo Panizza, Balint Parragi, Gabor Piski, Rita Ramalho, Joe Rebello, Mellany Pintado Vasquez, Evis Rucaj, Karlis Smits, Daniel Villar, and Albert Zeufack. All finalized papers, data, and tools have been posted on the World Bank website and are available at https://www.worldbank.org/en/topic/debt iii Debt Transparency in Developing Economies ABP Annual Borrowing Plan BOP Balance of Payments Acronyms CF Common Framework CG Central Government CRA Credit Rating Agency CSD Central Security Depository DeMPA Debt Management Performance Assessment DLP Debt Limits Policy DM Debt Management DMO Debt Management Office DMRS Debt Management and Recording System DPF Development Policy Financing DPO Development Policy Operation DRF Debt Reduction Facility DSA Debt Sustainability Analysis DSF Debt Sustainability Framework FXSL Foreign Currency Swap Line GDP Gross Domestic Product GFS Government Finance Statistics DSSI Debt Service Suspension Initiative HIPC Heavily Indebted Poor Country ICMA International Capital Market Association ICSD International Central Security Depository IDA International Development Association IDS International Debt Statistics IFI International Financial Institution IFMIS Integrated Financial Management System IMF International Monetary Fund IFRS International Financial Reporting Standards IIP International Investment Position INTOSAI International Organization of Supreme Audit Institutions IPSAS International Public Sector Accounting Standards FED US Federal Reserve GRID Green, Resilient and Inclusive Development iv Debt Transparency in Developing Economies LIC Low-Income Country LICDs Low Income Developing Countries LIC DSA Joint Bank-Fund Debt Sustainability Framework for LICs LMA Loan Market Association LSTA Loan Syndications and Trading Association MDB Multilateral Development Bank MTDS Medium-Term Debt Strategy NFPS Non-financial Public Sector OECD Organization for Economic Co-operation and Development NPC Negative Pledge Clause NRGI National Resource Governance Institute PBOC People’s Bank of China PDMLF Public Debt Management Legal Framework PEFA Public Expenditure and Financial Accountability PFM Public Financial Management PFRAM Public-Private Partnerships Fiscal Risk Assessment Model PPG Public and Publicly Guaranteed PPP Public-Private Partnership PSDS Public Sector Debt Statistics RBL Resource-Backed Loan SDFP Sustainable Development Finance Policy SDG Sustainable Development Goals SDR Special Drawing Right SFA Stock-Flow Adjustment SOE State-Owned Enterprise SPV Special Purpose Vehicle TA Technical Assistance UCOA Unified Charts of Accounts WBG World Bank Group v Debt Transparency in Developing Economies Analyzing public debt in low-income developing countries is like solving a puzzle with many missing pieces. Forty percent of LIDCs have not published any Executive sovereign debt data in the last two years. Public debt data disclosed in different publications show discrepancies of up to 30 percent of GDP across sources, and Summary relative to the records of relevant authorities. Over 15 LIDCs have outstanding collateralized debt but no details of the collateralization are provided in official statistics.1 Restructuring of bilateral and commercial debt is often handled privately. All these problems have different origins and implications. Yet, they all amount to a lack of transparency. The international community has become acutely aware of the importance of debt transparency after recent cases of “hidden debt.” The “Tuna Bond” case in Mozambique highlighted the dangers of inadequate debt transparency. In 2016 two large previously unreported loans totaling US$1.15 billion—equal to about 9 percent of the country’s GDP—were revealed.2 As a result, donor support was frozen, the economy plunged, and the government was forced to make deep cuts in public spending. The biggest losers were poor Mozambiquans. Nontransparent public debt can quickly alter the lives of millions of ordinary citizens. This report is the first comprehensive assessment of debt transparency in LIDCs. It presents a complete picture of the current challenges and the pending policy agenda for all stakeholders. It draws upon new databases and surveys to take stock of key gaps in debt reporting, borrowing practices and legal frameworks, offering a detailed and timely view on the current state of debt transparency in LIDCs. It also synthesizes recent studies and policy discussions on debt transparency and offers practical policy recommendations required to further improve debt transparency in LIDCs. The COVID-19 pandemic has highlighted the central role of debt transparency in better assessing debt sustainability, addressing vulnerabilities, and facilitating debt restructuring. The crisis has increased the balance sheet of the public sector and exacerbated the likelihood of contingent liabilities materializing. According to the WB/IMF Low Income Country Debt Sustainability Assessment (LIC-DSA), 44 percent of LIDCs are at high risk of external debt distress and 12 percent of them are already in debt distress. Inadequate transparency could delay debt restructurings and curb the ability of low-income countries to overcome the pandemic and generate a green, resilient, and inclusive recovery. 1. In this report, LIDCs are broadly defined as countries eligible for support from the International Development Association (IDA): http://ida.worldbank.org/about/borrowing-countries. 2. The loans were part of a larger financing operation organized by two state-owned enterprises (SOEs) under guarantee of the central government. The revelation led to large upward revisions to Mozambique’s official debt figures. Subsequent investigation showed that the guarantees were not subject to scrutiny by the Ministry of Finance before being approved, nor were they subject to oversight by the Parliament. 1 Debt Transparency in Developing Economies Greater public debt transparency is essential for macroeconomic stability and sustainable development. To meet the Sustainable Development Goals (SDGs) by 2030, LIDCs will need to invest at least 4.5 percent of national GDP each year on infrastructure alone (WB, 2019). With growing current account and budget deficits following the global economic slowdown, initial WB estimates indicate that external financing needs in LIDCs will reach USD$429 billion between 2023 and 2025. Most of these financing needs will have to be met through new borrowing. To ensure that this financing contributes effectively to development outcomes and does not undermine long-term debt sustainability, debt transparency must be improved. This could contribute to mitigating the severity of “boom-bust” cycles and help avoid setbacks in poverty reduction and other development objectives (Reinhart, Pazarbasioglu, 2021). This report aims to inform the full range of stakeholders in sound and transparent debt management. This includes not only public debt-management practitioners but also national policymakers, private sector creditors, and the public at large. It provides countries with tools to better identify their main challenges and offers concrete policy advice that can help them improve their debt management practices. It also recommends ways for international organizations to better calibrate their policy actions and contribute to the standardization of debt disclosure practices. Who Debt transparency benefits three categories of stakeholders: Benefits? 1. Authorities in borrowing countries: It enables them to make informed decisions about future borrowing with an accurate understanding of the cost and risk of the existing debt portfolio. 2. Creditors: It helps them fully understand borrowers’ debt-sustainability challenges, accurately price debt instruments and estimate comparability of treatment in the event of debt restructuring. 3. Citizens and civil society: Debt transparency assists them in holding governments accountable for the debt they take on, thereby facilitating better governance, increasing accountability, and helping to counter corruption. All play an important role in promoting transparency. Debt transparency is mainly the responsibility of authorities in borrowing countries. However, creditors can promote transparent financing practices and provide detailed information about their lending portfolio to their own constituencies, possibly filling in gaps in borrower’s statistics. IFIs can promote reforms in national legal and operational frameworks and encourage the harmonization of debt reporting and operations standards. Finally, citizens can actively engage with their national institutions and political. 2 Debt Transparency in Developing Economies Although the report focuses on LIDCs, some of its key findings can be applied to all sovereign borrowers. LIDCs face the greatest transparency challenges, given the (i) more limited availability of debt data beyond direct central government debt, (ii) higher share of non-marketable debt in their portfolios, (iii) weaker capacity and institutions, and (iv) higher political instability that creates additional opportunities for opaque borrowing. KEY FINDINGS Despite progress in recent years, public disclosure of debt data by LIDCs’ authorities (“direct reporting”) is still limited, particularly in fragile countries. Forty percent of LIDCs have never published any debt data or have not updated their data in the last two years. When available, debt statistics tend to cover only central government loans and securities, omitting other public sector components and debt instruments. Existing statistical standards for public debt are proving to be very ambitious for LIDCs. Debt management offices (DMOs) or offices in charge of producing debt statistics in LIDCs typically do not have the legal mandate, the incentives, or the capacity to collect data or report them beyond the central government level and standard debt instruments (e.g., loans and securities). As a result, instrument and sectoral coverage in public debt statistics fall short in adhering to international standards and differ significantly by country. According to our estimates, adding the state-owned enterprises’ (SOEs’) debt portfolio alone would increase the median of public debt by more than seven percent of national GDP. Any gaps in borrower statistics can be only partially covered by creditor data or by IFIs’ “indirect reporting”. Information disclosed by creditors is scant and depends on the disclosure policies of each lending institution. IFI databases extend the coverage, but only to the extent that data are collected and shared by the borrowing country. Even when comprehensive data exist, they may not be comparable because of different debt standards and definitions. In fact, the two internationally recognized standards for statistical and accounting reporting (GFS and IPSAS) are adopted on a voluntary basis, and the level of compliance with them depends on national preference, capacity, and legislation. In addition, deviations from international standards are not disclosed. Debt restructuring offers a unique opportunity to reconcile, update, and improve debtor and creditor records—but it has been a missed opportunity so far. The existing reporting ecosystem, depicted in Figure 2.6, is not entirely conducive to transparency. Multiple direct and indirect sources of public debt data co-exist. Uncoordinated data requests by IFIs and other external agents overburden often short- staffed DMOs. Differences in debt definitions and recording errors lead to discrepancies of up to 30 percent of national GDP across sources with the same expected coverage. Contingent liabilities are sizeable in LIDCs, but data thereon are not easily accessible. In 30 percent of LIDCs, statistics on guarantees are not disclosed. Among LIDCs with available data, the median stock of guarantees is 3.1 percent of GDP. Similarly, estimates of contingent liabilities from public private partnerships (PPP) are available in official debt/fiscal documents in less than 10 percent of the cases. PPPs have the potential to increase debt by a median of 2 percent of GDP. Finally, expenditure arrears, typically converted to debt through securitization, are hard to quantify in the absence of well performing accounting systems. Recent evidence suggests that they 3 Debt Transparency in Developing Economies accounted for an average of 3 percent of GDP in Sub-Saharan Africa in 2018 (IMF, 2018), and the COVID-19 crisis has further increased their stock by an additional 2 percent (WB, 2020). Central bank repos and foreign-currency swaps are increasingly being used to facilitate external borrowing, rather than to implement monetary policy or manage liquidity. This has the potential to generate “debt surprises” because central bank accounts are not consolidated with government accounts. The presence of repos and swaps is not clearly identifiable in central bank balance sheets, and it is not captured in current IFI databases. Domestic debt markets in LIDCs tend to be opaque. Only 41 percent of LIDCs use market-based auctions as the main issuance mechanism for domestic debt. Among those that do issue via auction, only half properly communicate with investors ex-ante by publishing a borrowing plan, and ex-post by disclosing the auction results on the same day. Non-tradable external debt is a source of possibly opaque operations. Bilateral and syndicated loans are more prone to misreporting or non-disclosure than are Eurobonds because loans are not traded in official markets and are more likely to include confidentiality clauses. In addition, information about their restructuring or re-profiling does not enter the public domain in a timely manner. Resource-backed loans (RBLs) pose distinct transparency challenges. Between 2004 and 2018, RBLs made up at least 8 percent of total new borrowing in Sub-Saharan Africa, between 10 percent and 30 percent of the median country’s total external public debt stock in the year following their signature. No country using these instruments reports collateralization details. The existence of these loans is often omitted in debt statistics because: (i) they are not systematically recognized and classified as debt by the borrower; (ii) they are often contracted by SOEs or special purpose vehicles (SPVs) which are off-budget and/or beyond the data collection mandate of the DMOs; or (iii) international databases do not require countries to report collateralization features. POLICY A sound public debt management legal framework (PDMLF) is needed for RECOMMENDATIONS promoting greater debt transparency. This report identifies eight key properties of a national PDMLF that make it conducive to transparency. A sound PDMLF (i) clarifies the borrowing authority, the delegation of power, and the debt authorization cycle; (ii) outlines debt management roles and responsibilities; (iii) defines public debt according to international standards, sets debt management objectives, and provides a list of permitted instruments, transactions, or sources of funding; (iv) regulates debt data disclosure statistics to ensure comprehensiveness, timeliness and full accessibility; (v) includes public debt audit requirements; (vi) stipulates the consequence of non-compliance; (vii) is publicly available: and (viii) extends its scope to the entire public sector. 4 Debt Transparency in Developing Economies The other key enabler of transparency is the presence of a sound institutional and operational framework. Many debt offices in LIDCs are now structured according to the international sound practices of back, middle, and front office. However, results from the World Bank’s Debt Management Performance Assessment (DeMPA) suggest that less than 50 percent of LIDCs meet the minimum requirements in terms of staff capacity (IMF-WB, 2020d). Improvements in IT systems for debt recording and management could also help enhance transparency. However, debt transparency cannot be addressed by individual countries by themselves. It requires broad international consensus and better global debt surveillance systems. IFIs, creditors, and other stakeholders, like credit rating agencies, have a key role to play in fostering debt transparency. Because of the current risks to low-income countries, improvements in debt transparency have to be accelerated. The debt transparency agenda has become urgent given the larger number of countries facing a high risk of debt distress or going through debt restructuring processes. This report makes several key recommendations and organizes them by their overall level of priority (high, medium, low). For borrowing countries, the relevance of each recommendation and the timing of their implementation will depend on their starting conditions. Topic Policy Recommendations Agent Priority Publish core public and publicly guaranteed (PPG) debt statistics at general government High level on an annual basis, including information on individual debt instruments contracted. Expand the coverage and improve timeliness of PPG debt reports in the categories identified in Medium the World Bank’s reporting heatmap. Publish regular estimation of the exposure to Transparency in Debt contingent liabilities (including those stemming Borrowers Medium Reporting (Chapter 2) from PPPs) and domestic arrears, in debt statistical reports and fiscal risk statements. Accelerate transition to IPSAS as the normative accrual accounting framework for financial Medium reporting. Publish information on final bondholders by category of investor, based on data provided Low by local and international central security depositories (CSDs). 5 Debt Transparency in Developing Economies Topic Policy Recommendations Agent Priority Limit and define the scope of confidentiality clauses and refrain from those that require High secrecy. Borrowers, creditors Fully disclose debt data reconciled in the context Medium of debt restructuring. Publish granular information on the lending portfolio (at loan level, possibly including terms) Creditors Medium on a single website, with annual updates. In indirect reporting databases, specify the country-specific instrument and sectoral High coverage (as opposed to their expected one) and explain deviations from direct statistics. Foster coordinated data collection processes and explore potential for streamlining and High consolidating existing IFI debt databases. Support the development and implementation of modern and integrated debt recording and management systems (DRMS), with definitions High and calculation methods aligned with international standards, and capable of easily recording all debt instruments. IFIs, other external Explore the feasibility of developing an agents Medium international loan repository system. Provide a regular assessment of borrowing countries’ adherence to international statistical Medium and accounting standards. In comprehensive debt restructurings, such as the G20 Common Framework, disclose key methodological information (e.g., on Medium comparability of treatment) ex ante and publish detailed reporting about their enforcement ex post. Promote alignment of debt statistical and Low financial reporting standards. 6 Debt Transparency in Developing Economies Topic Policy Recommendations Agent Priority Explicitly disclose the source of debt data used for credit ratings and the rationale for downward Low adjustment as a result of data transparency. Adopt market-based issuing mechanisms for domestic debt and develop clear, transparent High rules in the categories identified in the WB’s domestic debt securities heatmap. Borrowers Develop and adopt strict analytical and monitoring processes for approval and implementation of High resource-backed loans. Transparency in Debt Operations (Chapter 3) Minimize deviations from standard commercial loan templates and include provisions to require Borrowers, Medium any secondary market transaction of the loan to creditors be communicated to the borrower. Strengthen disclosure requirements on Central Bank repos/swaps. Collect information on their Borrowers, IFIs Medium use in international databases. Define clear rules, procedures, and processes on the authority to contract debt or issue guarantees in the PDMLF. Introduce “enhanced High authorization” for heavily structured transactions (e.g., collateralized debt), and new debt instruments. Outline the roles and responsibilities of the unit in charge of executing debt operations. High Public Debt Management Provide a definition of public debt in line with Legal Frameworks Borrowers international standards; announce the country’s (Chapter 4) debt-management objectives and provide a list High of permitted debt instruments, transactions or sources of funding. Require the publication of comprehensive and timely debt statistics and disclosure of a core set High of transaction-level debt information. Require regular audits (external and internal) of Medium DM activities and publish audit reports. 7 Debt Transparency in Developing Economies Topic Policy Recommendations Agent Priority Stipulate the consequences of not complying Medium with the PDMLF rules under domestic law. Publish the PDMLF rules and regulations in a single location, ideally in a unified document to Medium increase their coherency and accessibility. Extend the PDMLF’s scope to the entire public Medium sector borrowing. The rest of this report is organized as follows: Chapter 1 introduces a framework that defines debt transparency and lays out its main determinants. It also discusses the current efforts of the international community to promote the transparency agenda. Chapter 2 describes the debt reporting ecosystem in which two recording frameworks (financial and statistical) and multiple data sources coexist. The chapter also offers a number of recommendations to make the reporting process more effective, comprehensive, and coherent. Chapter 3 focuses on specific debt operations and instruments that may give rise to transparency concerns and discusses policy reforms that can serve to mitigate those concerns. Chapter 4 presents the key elements of a PDMLF that are conducive to debt transparency and offers clear guidance to authorities on how to strengthen their PDMLF. 8 Debt Transparency: A Framework Debt Transparency in Developing Economies 1.1 MOTIVATION Public debt levels in LIDCs have deteriorated in the last decade. Between 2010 and 2020, median public debt in IDA countries increased from 35 to 50 percent of GDP (see Figure 1.1a). In half of the countries that obtained debt relief from the Heavily Indebted Poor Countries (HIPC) initiative, the interest-to-revenue ratio on external debt currently exceeds pre-HIPC levels, particularly in Sub-Saharan Africa (WB-IMF, 2020a). In addition to the fall in commodity prices over 2011-14 causing volatile primary deficits, a key driver has been the rise of real interest-growth differentials; this is in direct contrast to advanced economies where interest-growth differentials have been steadily declining as a result of accommodative monetary policy (WB-IMF, 2020a). This means that the share of IDA countries classified by the IMF/World Bank Debt Sustainability Analysis for Low-Income Countries (LIC-DSA) as under or at high risk of debt distress more than doubled between 2013 and 2021, from 24 to 56 countries. The COVID-19 pandemic further deteriorated the debt situation in most LIDCs. Declining revenues and the rapid expansion of government balance sheets have challenged debt-service capacity, particularly in the world’s poorest and most vulnerable countries, which have limited fiscal and monetary policy instruments. Overall external public and publicly guaranteed (PPG) debt-service-to-revenue ratios for IDA countries increased from 8.2 percent to an estimated 11.8 percent between 2017 and 2019 (WB- IMF, 2021). The most recent LIC-DSA data from June 2021 show that 44 percent of LIDCs are at high risk of external debt distress and 12 percent of them are already in debt distress (see Figure 1.1b). Furthermore, safety margins have eroded in many countries at moderate risk of external debt distress. Figure 1.1 Figure 1.1a: Public Debt in IDA Countries Public Debt Vulnerabilities in LIDCs (Percent of GDP) 80 70 60 50 40 30 20 10 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020e Median Mean 25th percentile 75th percentile Source: World Economic Outlook October 2020. 10 Debt Transparency in Developing Economies Figure 1.1b: Risk of External Debt Distress (Percent of countries with LIC DSA) 100 24 80 51 54 56 60 45 40 31 32 32 20 31 18 13 12 0 2013 2019 2020 2021 Low Moderate High Source: Joint Bank-Fund LIC DSF database, June 2021. The creditor landscape has changed significantly over the last 20 years. Until the late 1990s, LIDCs borrowed primarily from official Paris Club creditors and IFIs on concessional standardized terms. The relative share of total debt of these creditors has decreased over time, giving up space to other bilateral creditors (mostly China) and commercial debt. Non-Paris Club creditors now account for more than twice the outstanding external debt as Paris Club creditors3. Over the last decade, the share of external debt to private creditors more than tripled in LIDCs, increasingly taking the form of Eurobonds as opposed to syndicated loans (see Figure 1.2). As a result, commercial banks are no longer the only holders of commercial debt; the expanding universe of creditors now includes, hedge/credit funds, investment management firms and commodity trading companies. Domestic debt has also been on the rise in LIDCs. Between 2011 and 2019, the median domestic debt-to-GDP ratio in IDA countries almost doubled from 7 percent to 13 percent of GDP (WB-IMF, 2021). The structure of debt is expected to continue to shift, leading to further increases in interest costs and to higher risk. In the absence of significant increases in ODA and based on current trends,4 the share of concessional debt to total public external debt is expected to fall from the current level of 39 percent to 30 percent by 2030. (WB-IMF, 2020a). 3. Source: WB’s IDS. 4 . Assuming constant share of ODA (in percent of donor countries’ GDP), and the current DSA assumption of average annual growth of external debt of 8 percent. 11 Debt Transparency in Developing Economies Figure 1.2 (share of total, 2010-19) Composition of PPG External 100% Debt in LIDCs 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Multilateral Bilateral Non PC Bilateral PC Bonds Commercial Source: International Debt Statistics. The range of lending instruments and borrowers within the public sector has also expanded. LIDCs are increasingly relying on off-budget borrowing using state-owned enterprises (SOEs) and special purpose vehicles (SPVs). The range of instruments includes resource-backed loans as well as other forms of collateralized debt and PPPs whose future obligations governments often find hard to anticipate. The higher diversification of creditor and borrower base and the use of innovative and complex debt instruments create important challenges for transparency. Legal frameworks, and debt reporting and management practices in LIDCs are still based on a model of sovereign lending centered around standard concessional borrowing from the central government. As discussed in the report, the increased sophistication of public debt portfolios has not been accompanied in all LIDCs by a corresponding upgrade in capacity, institutions, or legal and operational frameworks for debt management. Increasing levels of debt as a result of the pandemic have further highlighted the urgency for transparency. In past crises, it became apparent that borrowed funds had been diverted to purposes that did not raise export proceeds, productivity, or potential output (Kose et al., 2020). Any deterioration in debt-service capacities tends to increase incentives for governments to move debt off-budget or to take advantage of the lack of internationally enforceable standards for accounting and statistics to disguise debt and avoid possible repercussions from higher disclosure (i.e., higher cost of borrowing, possible downgrades). The last column in Table 1.1 shows the very limited availability of debt data in LIDCs beyond central government level. 12 Debt Transparency in Developing Economies Table 1.1 Data Availability: General Government and Non-Financial Public Corporation Debt (end-2019) All AEs EMs LIDCs Number of countries General Government 107 35 45 27 Non-Financial Public Corporations 81 29 34 18 Percent of membership General Government 56.6 100 52.9 39.1 Non-Financial Public Corporations 42.9 82.9 40 26.1 Source: WB-IMF, 2020a. Collateralization by public sector borrowers also tends to be higher in times of stress (WB, IMF, 2020c). Collateralization takes different forms and has appeared in bilateral official lending as well as commercial lending (WB, IMF, 2020a) as means to maintain access to commercial funding or to reduce the cost of borrowing. Collateralization may be beneficial from a financial standpoint, as it may provide access to finance not otherwise available or allow the borrower to obtain more favorable pricing. However, experience from countries that entered into such contracts following the commodity price shock in 2014-15 (e.g., Chad and Republic of Congo) shows that those collateralized debt contracts were non-transparent and likely exacerbated debt vulnerabilities. Finally, debt transparency is crucial for ongoing and future debt restructuring workouts. Three countries (Chad, Ethiopia, Zambia) have already requested the G20 Common Framework for debt restructuring and others may receive some form of bilateral or comprehensive debt restructuring in coming months. Debt restructuring negotiations can be protracted when creditors raise doubts about the size and composition of the debtor’s debt portfolio or find it difficult to assess the level of debt relief needed to restore debt sustainability. Creditors also seek to avoid subsidizing other creditors whose debt has not been fully disclosed.5 This problem is even more acute when restructuring is carried out in private negotiations with single creditors rather than in a coordinated manner.6 Moreover, in the absence of accurate and comprehensive data, debt restructuring is delayed by the need of undertaking time-consuming reconciliations between debtor and creditor records. 5. In Zambia, a group of bondholders representing around 40% of total outstanding Eurobonds held out from accepting a request from the government for a suspension of coupon payments on the grounds that the debt owed to some bilateral lenders was not fully disclosed. Zambia defaulted in November 2020. 6. In the of Republic of Congo, for instance, the restructuring that began in early 2018 remains incomplete. After a year-long negotiation, the authorities restructured the debt owed to a non-Paris Club creditor, but they have continued to be in discussions with three commercial commodity traders to restructure their debt. An agreement with one of the traders was reached in 2020, but no official report detailing its terms has been published. 13 Debt Transparency in Developing Economies Countries will need debt financing for a recovery based on a green, resilient, and inclusive development (GRID). The COVID-19 pandemic has eroded fiscal space and the macro-economic environment remains highly uncertain; recovery from the pandemic will require significant investment. In this context, improvements to debt transparency are needed to help ensure that borrowing and lending decisions are not subject to asymmetric information, and that debt is contracted at the lowest possible costs—subject to an acceptable level of risk. 1.2 WHAT IS DEBT Debt transparency covers the availability of debt data and borrowing processes TRANSPARENCY? that are legitimate, rule-based, and traceable. Borrowers and creditors need detailed information on the outstanding stock of public debt, including terms and conditions, to make informed borrowing and lending decisions; citizens also need this information to hold their governments accountable. However, this is only part of debt transparency as reporting needs to be complemented by borrowing processes and practices that ensure that new debt is contracted responsibly and in line with sound legal and operational frameworks to minimize enforcement uncertainty. As depicted in Figure 1.3, debt transparency in this report is therefore defined by the following two interrelated dimensions: • Transparent debt reporting: Debt reports should comprise comprehensive, timely, and consistent debt data at public sector level. To facilitate cross-country comparability and comprehensive debt analyses, public sector debt statistics (PSDS) should be compiled and reported based on internationally accepted statistical definitions and concepts (WB-IMF, 2020d) • Transparent borrowing operations: Transparency around borrowing practices is needed to ensure that debt is contracted legitimately, shielded from undue political interference, and grounded on a sound analysis of the legal implications and financial cost and risks of the different borrowing alternatives. A country’s performance in these two dimensions depends on two enabling factors: • A sound public debt management legal framework (PDMLF) and adherence to it. Public debt reporting and borrowing operations must be based on a comprehensive set of legal requirements. As detailed in Chapter 4, a PDMLF promotes debt transparency when it: (i) clearly specifies the authority to borrow and the debt authorization cycle; (ii) clarifies the institutional arrangements of debt management; (iii) discloses national debt policies; (iv) adopts reporting standards in support of debt transparency; (v) introduces audit requirements; (vi) regulates consequences of non-compliant debt; (vii) is publicly accessible; and (viii) extends its scope to the entire public sector. 14 Debt Transparency in Developing Economies • An effective DM institutional and operational framework. Transparency of debt data and borrowing operations also depends on organizational structures that ensure segregation of duties, avoid conflicts of interest and are well-equipped with skilled staff and robust and integrated IT systems for debt recording and management.7 External stakeholders can also play a key role in fostering debt transparency in LIDCs. • IFIs promote reforms in national legal and operational frameworks and encourage the harmonization of debt reporting and operations standards. IFIs have developed standards and guidelines to promote accurate debt reporting (e.g., Public Sector Debt Statistics: Guide for Compilers and Users)8 and sound practice with respect to debt management (e.g., WB’s DeMPA) that promote transparency. However, international standards are currently not enforceable and therefore implemented unevenly. • Creditors’ data disclosure practices and lending policies are also key drivers of debt transparency. In addition to providing information about their lending portfolio to their own constituencies, creditors can fill the gaps in borrowers’ statistics, and facilitate transparent and legally sound debt operations. Acknowledging this role, the G20 Operational Guidelines for Sustainable Financing encourages sovereign creditors to publish loan-by-loan information on new debt, including all terms of the new debt (on a single website with regular updates); to refrain from confidentiality clauses; and to use publicly available legal documentation templates. Figure 1.3 International Standards and Guidelines Debt Transparency Framework National PDM National PDM Institutional Legal Framework and Operational Framework Creditors’ Reporting and Lending Practices Debt Reporting: Debt Operations: Comprehensive, Accurate, Legitimate, Rule-Based and Timely Debt Data and Traceable Debt Process Debt Transparency Source: Authors’ elaboration. 7. The existence of an effective institutional and operation framework constitutes a pre-condition of any sound debt management activity. The World Bank’s Debt Management Performance Assessment (DeMPA) and the performance indicator 13 of the Public Expenditure and Financial Accountability (PEFA) methodology capture the elements considered essential for achieving this goal. The performance of LIDCs against these international benchmarks is regularly assessed. 8. The Public Sector Debt Statistics: Guide for Compilers and Users, 2013 (PSDS Guide) provides the international statistical standard for compiling and reporting PSDS. The PSDS Guide is fully harmonized with the standards set out in the System of National Accounts of 2008 (2008SNA) and the IMF’s Government Finance Statistics Manual of 2014 (GFS, 2014). 15 Debt Transparency in Developing Economies The definition of debt transparency used in this report does not capture how the proceeds of debt financing are used. The sound selection and implementation of debt-financed projects and the coherent and effective use of proceeds will help create successful development outcomes, but the related analysis is outside the scope of the current report. The analysis of the use of proceeds is more closely linked to the transparency of the budget process, which has become increasingly important as LIDCs shift their borrowing focus to commercial debt for general budget purposes.9 1.3 DEBT TRANSPARENCY: Governments, acting in the interest of the public, should treat debt transparency INCENTIVES AND as an objective in itself, irrespective of financial implications. Similar to budget DISINCENTIVES transparency, public debt is a matter of public interest and citizens should be able to obtain information on public borrowing and hold policymakers accountable for their decisions. Enhancing debt transparency also reduces the uncertainty premium on public borrowing. Empirical studies found that increased accountability in debt-management practices results in higher credit ratings (Arbatli and Escolano, 2012) and stronger investor appetite (Gelos and Wei, 2005), which may ultimately reduce the cost of external borrowing (Choi and Hashimoto, 2017 and Kubota and Zeufack, 2020). A recent WB study on “hidden debt” in South Asia concluded that transparency reduces the likelihood of shocks from contingent liabilities, by aligning policy maker incentives with fiscal responsibility (Melecky, 2021). These findings are consistent with other empirical work that finds a negative correlation between transparency and the “stock-flow adjustments” in a cross-country analysis (Weber, 2012).10 However, without a credible debt transparency commitment mechanism, markets cannot fully identify countries with enhanced transparency, which may reduce the financial benefits for transparent borrowers. IFIs and credit rating agencies have a role to play in this (see Box 3.2). A lack of debt transparency can lead to debt mis-pricing and subsequent sudden price corrections. Investigating a liquid bond market like the EU’s, Bernoth and Wolff (2008) show that creative accounting increases risk premia.11 In Mozambique, the disclosure of previously unknown external loans led to a sharp widening of bond spreads (see Figure 1.4a). In Zambia, reporting lags and uncertainty around public debt coverage led to speculation about the true level of indebtedness and a sharp increase in bond yields (see Figure 1.4b). 9. A number of international initiatives are in place to foster more open and inclusive budgeting processes (see, for instance: www.internationalbudget.org). 10. The Stock-flow adjustment (SFA) is the year-on-year change in debt after accounting for the fiscal deficit. It may represent the largest driver of public debt spikes (Campos, Jaimovich, & Panizza, 2006). SFA can stem from below-the-line acquisition of liabilities (and assets); changes to the valuation of the existing debt stock; or statistical discrepancies. SFAs can be considered blind spots in public debt dynamics as they cannot be properly modelled or forecasted (Jaramillo, Mulas-Granados, & Kimani, 2016, and Jaramillo, Mulas-Granados, & Jalles, 2017). 11. “Creative accounting” is defined by the authors as fiscal and debt window-dressing or shifting expenditures off the budget. 16 Debt Transparency in Developing Economies Figure 1.4 Figure 1.4a: Mozambique EMBI Spread Lack of Transparency: Market Reactions (basis points, 2015-16) 2,400 2,100 - Mozambique authorities announced “tuna bond” 1,800 restructuring; investors lost confidence over its “misuse” 1,500 1,200 900 600 - Just as “tuna bond” - Mozambique exchange drew to an authorities published 300 end, two large “hidden DSA showing public loans” were revealed debt in distress 0 01/2015 02/2015 03/2015 04/2015 05/2015 06/2015 07/2015 08/2015 09/2015 10/2015 11/2015 12/2015 01/2016 02/2016 03/2016 04/2016 05/2016 06/2016 07/2016 08/2016 09/2016 10/2016 11/2016 12/2016 Source: WB-IMF, 2018. Figure 1.4b: Zambia EMBI Spread (basis points, 2018) 550 - The Finance Minister was demoted - IMF issued statement 500 to call off program discussions (due to debt- related concerns) 450 400 - Africa Confidential published article raising concerns that Zambia is - Mozambique published its annual 350 economic report pulling a Mozambique - BAML published investor report clarifying issue on “hidden debt” 300 1/01/2018 1/08/2018 1/15/2018 1/22/2018 1/29/2018 2/05/2018 2/12/2018 2/19/2018 2/26/2018 3/05/2018 3/12/2018 3/19/2018 3/26/2018 4/02/2018 4/09/2018 4/16/2018 Source: WB-IMF, 2018. 17 Debt Transparency in Developing Economies Despite these advantages, borrowers and creditors may have incentives to keep the existence, or terms, of some transactions confidential. While there may be also occasional legitimate reasons for keeping part of the agreements confidential (see Chapter 4.4), Table 1.2 outlines other kind of incentives for lenders and borrowers to avoid higher levels of transparency or even to actively hide debt. Table 1.2 Lack of Debt Transparency: Possible Incentives for Borrowers and Lenders Borrower Lender To avoid a higher cost of debt in the short term by To acquire the favor of the borrower to facilitate hiding (i) the true extent of indebtedness or (ii) the the lending operation or promote broader business/ collateral granted to selected creditors politically strategic objectives To circumvent policies that may impact borrowing cost To avoid disclosing financial or legal terms or availability (e.g., fiscal rules, borrowing limits, or to competitors negative pledge clauses) To avoid public scrutiny or to maintain confidentiality around politically sensitive issues, such as security investments or access to natural resources To obtain personal gains, particularly in environments with limited checks and balances Source: Authors’ elaboration. However, lack of transparency is not necessarily the consequence of deliberate actions by borrowers and lenders. Recent episodes of “hidden debt” have linked debt transparency with intentional efforts to conceal debt exerted by the creditor, the borrower, or both. While this is undoubtedly part of the problem, our analysis shows that debt transparency is also a story of “missing debt portfolios”. In several cases, entire sectors or instruments are mis- or under-reported as a result of weak domestic legal and operational frameworks. Inadequate debt reporting may result from a narrow public debt definition at national level, in the absence of enforceable international reporting standards; or it may also come about as a result of the lack of capacity, legal mandate, or organizational procedures to collect, process, and disclose data related to certain debt operations. It is challenging to empirically disentangle unintentional from intentional lack of transparency. A desire to maintain discretion in borrowing decisions may affect the level of investment in debt management institutions and legislation. For example, authorities may prefer to limit the mandate of the DMO, thus hampering its ability to become a key debt management decision-maker. In this case, weaknesses in capacity area direct consequence of a deliberate lack of commitment to debt transparency. This causality loop limits the extent to which either intention or capacity can be isolated as the cause of non-transparency. 18 Debt Transparency in Developing Economies Restructuring episodes offer unique opportunities to foster debt transparency. Restructuring episodes temporarily relax some of the main constraints of debt transparency, i.e., capacity and willingness to report. On the one hand, the ad-hoc assistance provided to borrowing countries by IFIs, creditors and financial/legal advisors temporarily solves capacity limitations. On the other hand, the high stakes attached to the restructuring process increase the borrower’s reputational risk in the event that hidden debt is discovered, increasing the likelihood of truthful reporting. However, as discussed in Chapter 2, the opportunity for more complete disclosure and information sharing has not been yet fully leveraged. 1.4 CURRENT EFFORTS Debt transparency has moved to the forefront of International Financial Institutions’ TO IMPROVE DEBT (IFI) agenda, particularly the World Bank. Under the Addis Ababa Action Agenda TRANSPARENCY for Financing for Development to meet the Sustainable Development Goals, debt transparency is a key commitment of the international community.12 Acknowledging its importance for sustainable development, the World Bank Group has been working on five key principles for achieving full debt transparency, as communicated by its President in June 2020.13 The five principles are the following: (i) spell out loan contract terms and payment schedules; (ii) full disclosure of the stock of public and publicly guaranteed debt, SOE liabilities, and debt-like instruments; (iii) enable borrowers to seek relief from excessive confidentiality clauses so they can proceed with more transparent data reporting; (iv) promote effective and prudent use of collateral and liens in sovereign borrowing; and (v) insist that borrowers and lenders avoid violations of legal requirements of other creditors, such as negative pledge clauses. Debt transparency has become a cornerstone of the World Bank’s public debt program and is a cross-cutting theme within the institution. It has been integrated into World Bank policies and operations and is supported by scaled-up technical assistance. Enhancing debt transparency is a key pillar of the Multi-Pronged Approach (MPA) to reducing debt vulnerabilities, approved by the WB board in 2018 (updated in 2020);14 the new Sustainable Development Finance Policy (SDFP) also provides new incentives for IDA countries to improve debt transparency.15 The revised Low-Income Country Debt Sustainability Framework (LIC DSF) has been successfully rolled out, contributing to increased debt coverage in more than 15 LIDCs. Regular publications support monitoring public debt vulnerabilities in developing countries;16 tools to assess the transparency of public debt reporting and domestic borrowing in LIDCs are being developed and published.17 12. “We recall the need to strengthen information sharing and transparency to make sure that debt sustainability assessments are based on comprehensive, objective and reliable data. We will work towards a global consensus on guidelines for debtor and creditor responsibilities in borrowing by and lending to sovereigns, building on existing initiatives”. 13. https://blogs.worldbank.org/voices/june-19-2020-debt-and-investment-transparency-better-outcomes. 14. https://documents.worldbank.org/en/publication/documents-reports/documentdetail/351311607717096586/update-on-the-joint-world-bank-imf-multipronged-approach- to-address-debt-vulnerabilities. 15. https://ida.worldbank.org/debt/sustainable-development-finance-policy. 16. https://documents.worldbank.org/en/publication/documents-reports/documentdetail/695971579921244762/the-evolution-of-public-debt-vulnerabilities-in-lower-income- economies. 17. https://www.worldbank.org/en/topic/debt/brief/debt-transparency-report; https://blogs.worldbank.org/voices/making-domestic-borrowing-transparent. 19 Debt Transparency in Developing Economies The World Bank also supports creditor initiatives to enhance sustainable financing practices. In 2017, the G20 endorsed the Operational Guidelines for Sustainable Financing, which aim to enhance sustainable financing by official bilateral lenders, ensuring debt sustainability and improved creditor coordination. In 2019, the guidelines were translated into a set of financing principles structured around five main objectives, including improved information sharing and debt transparency. To this end, the World Bank, with the IMF, has published a diagnostic tool allowing creditors to self-assess their financing practices.18 World Bank support of debt transparency is aligned with ongoing initiatives led by other IFIs. The OECD Recommendations on Sustainable Lending Practices and Officially Supported Export Credits were adopted by the Council at Ministerial level in 2018.19 Directed at bilateral providers of export credits, the recommendations recognize that export credits should be allocated sustainably and should create a positive net economic return for the borrower. A specific recommendation is that the decision to provide export credit should follow the most recently available LIC DSF, be in line with IMF and World Bank debt limit policies and enhance debt transparency through annual data sharing with the IMF and World Bank, via the OECD Secretariat, on all transactions to lower-income countries. In 2019, the Institute of International Finance (IIF) published the Voluntary Principles for Debt Sustainability for private sector lenders.20 These principles aim to facilitate the disclosure of private sector foreign- currency lending to public sector entities and recommend disclosure of the amount and terms of private lending within 120 days of the first financial flow, including reference to any collateralization. This initiative was followed-up by the OECD through the development of a commercial debt data repository and reporting platform.21 Debt transparency is also at the core of the G20 Debt Service Suspension Initiative (DSSI) and Common Framework (CF). For eligible countries that requested it, the DSSI is enacting a temporary suspension of debt service to bilateral official creditors; the CF facilitates comprehensive debt restructuring including burden sharing across creditors. Both initiatives rely on participating countries’ commitment to full disclosure of all public sector financial commitments. The World Bank has been publishing updated PPG external data from the International Debt Statistics database (IDS) online, including a previously unavailable breakdown by creditor.22 Data on the potential debt- service-suspension amounts from the DSSI are also published.23 18. https://www.worldbank.org/en/programs/debt-toolkit/resources#5. 19. http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=tad/ecg(2018)4&doclanguage=en. 20. https://www.iif.com/Publications/ID/3387/PageID/3387/Voluntary-Principles-For-Debt-Transparency. 21. https://www.oecd.org/finance/OECD-Debt-Transparency-Initiative.htm#:~:text=29%2F03%2F2021%20%2D%20The,poorest%20and%20most%20vulnerable%20 countries. 22. https://datatopics.worldbank.org/debt/ids/TDSS. 23. https://www.worldbank.org/en/topic/debt/brief/covid-19-debt-service-suspension-initiative. 20 Debt Transparency in Developing Economies 1.5 CONCLUSIONS In the last decade, the deterioration of public debt levels in LIDCs has been paired with a changing creditor landscape and a greater range of lending instruments that place transparency at the forefront of the policy agenda. The COVID-19 pandemic has eroded the fiscal space while rendering the macro-economic environment highly uncertain. The road to recovery demands significant investments in the face of global financial distress. Fostering debt transparency can help countries undertake better borrowing choices and make efficient, effective use of the resources. Debt transparency entails availability of debt data and legitimate, rule-based and traceable debt processes. A country’s performance in these two dimensions depends on two factors: a sound public debt management legal framework (and adherence to it) and an effective DM institutional and operational framework. External stakeholders like IFIs and creditors can also play a role in fostering debt transparency in LIDCs. Enhancing debt transparency creates long-term material benefits. Yet, governments may face operational challenges to transparency or have incentives to keep transactions confidential. Enhancing debt transparency has therefore become a cornerstone of the World Bank’s public debt program and has been established as a cross-cutting theme within the institution. The World Bank has developed a comprehensive debt transparency agenda to deal with the multilayered nature of the issue while remaining aligned with ongoing initiatives led by other IFIs. 21 Debt Transparency in Developing Economies REFERENCES Asonuma, T., & C. Trebesch. 2016. Sovereign Debt Restructurings: Preemptive or Post- Default. Journal of the European Economic Association 14, 175–214. Association for Financial Markets in Europe (2019/2020), 2020, “European Primary Dealers Handbook” Beers, D., Jones, E. and Walsh, J. (2020) BoC-BoE Sovereign Default Database: Methodology, Assumptions and Sources 2020. Bank of England Research Paper, June 2020. Bank of England, London. Baduel, Benedicte & R. T Price. 2012. “Evolution of Debt Sustainability Analysis in Low-Income Countries; Some Aggregate Evidence,” IMF Working Papers 2012/167, International Monetary Fund, Washington, D.C. Beers, D., E. Jones and J. Walsh. 2020. BoC-BoE Sovereign Default Database: What’s new in 2020? Bank of England Research Paper, June 2020. Bank of England, London. Beers, D., E. Jones and J. Walsh. 2020. Special Topic: When sovereigns default, how big a share of government debt is involved? Bank of England Research Paper, June 2020. Bernoth, K., and G. B. Wolff. 2008. “Fool the Markets? Creative Accounting, Fiscal Transparency and Sovereign Risk Premia.” Scottish Journal of Political Economy 55 (4): 465–87. Campos, C. F.S., D. Jaimovich & U. Panizza. 2006. “The unexplained part of public debt,” Emerging Markets Review, Elsevier, vol. 7(3), pages 228-243. Choi, S., Hashimoto, Y. 2017. “The Effects of Data Transparency Policy Reforms on Emerging Market Sovereign Bond Spreads” Working Paper No. 17/74. International Monetary Fund, Washington, D.C. Escolano J., Arbatli, E., 2012. “Fiscal Transparency, Fiscal Performance and Credit Ratings”. Working Paper No. 12/156. International Monetary Fund, Washington, D.C. Gelos, G., Wei, S. 2005. “Transparency and International Portfolio Holdings”. J. Finance, 2005, vol. 60/6. Horn, Sebastian & C. M. Reinhart & C. Trebesch. 2019. “China’s Overseas Lending,” NBER Working Papers 26050, National Bureau of Economic Research, Inc. Jaramillo, Laura & Carlos Mulas-Granados & Elijah Kimani. 2016. “The Blind Side of Public Debt Spikes,” IMF Working Papers 2016/202, International Monetary Fund. Washington, D.C. Jaramillo, Laura & C. Mulas-Granados & J. Tovar Jalles. 2017. “Debt spikes, blind spots, and financial stress,” International Journal of Finance & Economics, Vol. 22/4. 255-437. 22 Debt Transparency in Developing Economies Kubota, M., and M. Zeufack. 2020. “Assessing the Returns on Investments in Data Openness and Transparency”. Policy Research Working Paper No. 18; World Bank. Washington, D.C. Kose, M., P. Nagle, F. Ohnsorge, N. Sugawara. 2021. Global Waves of Debt: Causes and Consequences. World Bank. Washington, D.C. Melecky, Martin. 2021. “Hidden Debt: Solutions to Avert the Next Financial Crisis in South Asia”. South Asia Development Matters. World Bank. Washington, D.C. United Nations. 2015. “Addis Ababa Action Agenda of the Third International Conference on Financing for Development, outcome document (General Assembly resolution 69/313, annex)”. August 17, 2015. https://www.un.org/en/development/desa/ population/migration/generalassembly/docs/globalcompact/A_RES_69_313.pdf. Weber, A. 2012. “Stock-Flow Adjustments and Fiscal Transparency: A Cross- Country Comparison.” IMF Working Paper 12/39, International Monetary Fund, Washington, D.C. World Bank (2009). Debt Management Performance Assessment. World Bank. Washington, D.C. World Bank and International Monetary Fund. 2021. “World Bank Group and International Monetary Fund Support for Debt Relief Under the Common Framework and Beyond,” Washington, DC. World Bank and International Monetary Fund. 2020a. “The Evolution of Public Debt Vulnerabilities in Lower Income Countries”. Washington, DC. World Bank and International Monetary Fund. 2020b. “Update on the Joint IMF-WB Multipronged approach to address debt vulnerabilities “, Washington, DC. World Bank and International Monetary Fund. 2020c. “Collateralized Transactions: Key Considerations for Public Lenders and Borrowers”. Washington, D.C. World Bank and International Monetary Fund. 2020d. “Public Sector Debt Definitions and Reporting in Low Income Countries”. Washington, D.C. World Bank and International Monetary Fund. 2018. “G20 Notes on Strengthening Public Debt Transparency”. Washington, D.C. 23 Transparency in Debt Reporting Debt Transparency in Developing Economies The publication of accurate, comprehensive and timely debt data is the basis of the concept of public debt transparency. Public debt constitutes the largest financial portfolio in most LIDCs, and governments owe it to their citizens and creditors to provide regular current data. Lack of full disclosure may have important financial consequences and may diminish trust in governments, as discussed in Section 1.3. In light of this, one might expect to find essential information about government debt operations publicly accessible to citizens and other stakeholders in standard form. Yet, public debt information in LIDCs is often absent or partial. Close to 40 percent of LIDCs have never published debt data on their websites or have not updated their data in the last two years. When available, statistics tend to cover only central government loans and securities, often omitting other public sector entities, as well as non-standard debt instrument and contingent liabilities. As a result of the different instrument and sectoral coverage levels, debt data are not easy to compare across time and countries. In addition, lack of visibility around exact public debt stock may affect the quality of debt sustainability analysis and limit the implementation of swift and fairly designed debt restructuring. Finally, government capacity to make sound borrowing decisions is limited when comprehensive debt data are not available. At its core, the production of debt data is governed by two frameworks: financial and statistical reporting. Although these frameworks share the same underlying data and overlap in terms of coverage, they serve different purposes. Public sector government finance statistics are used to analyze fiscal policy options, determine their impact on the economy, and compare outcomes nationally and internationally. Financial statements are used to evaluate financial performance and position, hold management accountable, and inform decision-making (GFS, 2014). Both are meant to inform external stakeholders, but the two frameworks present differences in key methodological principles. Public debt statistics are disseminated in two forms: direct or indirect reporting. Direct reporting takes place when debt data are directly made public by national authorities in dedicated bulletins or monetary/macroeconomic publications. National authorities also make data available to external agents like the World Bank, the IMF, rating agencies, etc., for their review, compilation, and disclosure in their own publications or websites. This is indirect reporting (Figure 2.1). 25 Debt Transparency in Developing Economies Figure 2.1: Debt Reporting Frameworks and Dissemination Debt Data Channels Statistical Reporting Financial Reporting Direct Indirect Source: Authors’ elaboration. The existence of multiple data disclosure frameworks is justified by the respective objectives and audiences but complicates the interpretation of debt figures. Different definitions and standards across different frameworks and dissemination channels result in diverging debt records across reporting sources. Dipplesman (2015) shows that the public debt-to-GDP ratios of a country at any given time could range from 40 to over 100 percent depending on the definition of “public debt”. It is therefore critical to understand these methodological differences to identify the impacts of sub- reporting or misreporting in debt data. This chapter measures the extent of transparency in statistical and accounting reporting and studies its determinants. It is structured as follows: Section 2.1 portrays current debt statistics practices in LIDCs and describes factors that may foster or impede these practices. Section 2.2 focuses on indirect reporting and presents possible sources of discrepancies with direct sources. Section 2.3 identifies the key differences between statistical and financial debt reporting. Section 2.4 concludes and provides policy recommendations on how to make the reporting ecosystem more comprehensive, consistent and reliable. 2.1 DEBT STATISTICS: Public debt data are produced and disseminated by borrowing countries with DIRECT DATA DISCLOSURE diverse objectives and final users in mind. Format, timing, and content of the information depend on the final audience (policymakers, investors, IFIs, general public) as well as on the final objectives (marketing, information, internal governance, and compliance with national or international rules). Debt reports are expected to cover the entire public sector to meet the highest standard of transparency (Figure 2.2).24 According to international standards, public debt reports should provide a comprehensive and detailed overview of the stock of public sector debt, including data on guarantees and other explicit contingent liabilities, and (when relevant) collateral provided. In terms of debt instruments, the following six instruments should be included: special drawing rights (SDRs);25 currency and deposits; debt securities; loans; insurance, pension, and standardized guarantee schemes (IPSGS); and other accounts payable (IMF. WB, 2020). 24. As outlined in the Public Sector Debt Statistics: Guide for Compilers and Users, 2013 (PSDS Guide) which provides the international statistical standard for compiling and reporting public sector debt. The PSDS Guide is fully harmonized with the standards set out in the System of National Accounts of 2008 (2008SNA) and the Government Finance Statistics Manual of 2001 and 2014 (GFS) http://tffs.org/pdf/method/2013/psds2013.pdf. 25. SDR refer to SDR allocations, not WB/IMF loans denominated in SDR, which should be classified as loans. 26 Debt Transparency in Developing Economies Figure 2.2 Public Sector Coverage Public Sector General Public Government Corporations Central Government Public Nonfinancial Public Financial Corporations Corporations State Government Central Bank Local Government Public Deposit-Taking Corporations Except the Central Bank Other Public Financial Corporations Source: PSDS Guide 2013. The statistical targets for both instrument and sectoral coverage are proving to be very ambitious for LIDCs. The DMOs or the office in charge of producing debt statistics in LIDCs typically do not have the legal mandate, the incentives, or the capacity to collect data or report them beyond a certain level (e.g., central government) or standard debt instruments (e.g., loans and securities). As a result, instrument and sectoral coverage in public debt statistics differ significantly by country, as documented in the literature (Dippelsman et al, 2012; Gelpern, 2018 and Seiferling, 2020). In addition to completeness, accessibility and timeliness are key dimensions of transparent debt statistics. Debt data should be easy to access and should be published with a limited time lag. To benchmark and track a country’s level of direct disclosure of debt statistics along these three dimensions (completeness, accessibility and timeliness), the World Bank recently launched a tool (the debt reporting “heatmap”) based on analysis of information available on national authorities’ websites.26 Additional heatmap indicators cover the publication of key debt management documents and the reporting on risks from contingent liabilities.27 Each indicator is evaluated using a scale divided into four categories, which ranks reporting standards from low (red) to high (green), according to the criteria presented in Figure 2.3.28 26. The heatmap and the methodology can be found at: https://www.worldbank.org/en/topic/debt/brief/debt-transparency-report. 27. This dimension assesses the disclosure of debt statistics “memorandum items”, including account payables, central governments guarantees and other contingent liabilities. 28. Selection of indicators and relative thresholds is based on guidelines contained in recognized debt management “best practices” documents, such as the WB’s DeMPA. 27 Debt Transparency in Developing Economies Figure 2.3 Methodology Underpinning the Debt Reporting Heatmap 1. Public Debt Reporting Data Accessibility Completeness Timeliness Information on Instrument Coverage Sectoral Coverage Periodicity Time Lag New Loans No publicly available N.A. or N.A. or N.A. N.A. N.A. reporting or incomplete/ incomplete coverage incomplete budgetary outdated debt reports central (CG) Multiple websites Limited coverage: Limited coverage: Limited Information: > 1 year > = 6 months external or domestic complete CG lender’s name (e.g. every debt only or purposes of two years) borrowing, amount Single website & Partial coverage: Partial coverage: Partial information: Annual basis > 3 months multiple docs external and domestic complete general lender’s name, (if applicable) government (CG) amount, purpose but or complete public no financial terms corporations Single document Full coverage: Full coverage Full information: < 1 year < = 3 months external and domestic (CG and public lender’s name, (e.g. quarterly and guarantees corporations) amount, purpose update) (if applicable) and financial terms 3. Other Debt Statistics/ 2. Public Debt Management Contingent Liabilities (CLs) Debt Management Strategy (DMS) Annual Borrowing Plan (ABP) N.A./No Information No DMS published No ABP published or a partial plan No reporting or insufficient reporting with a delay > 3 months of existing CLs Yes, but no targets Partial: only for domestic or external Limited reporting: guaranteed debt debt) with a delay < 3 months or full by beneficiary (if applicable) coverage with delay > 3 months Yes, with target for total debt Full coverage within 3 months Partial reporting: audited/recognized fiscal arrears and collateralized debt (if applicable) Yes, with target for domestic Full coverage before the fiscal/ Comprehensive reporting: & external debt calendar year starts publication of a framework covering existing CLs Source: Author’s elaboration. 28 Debt Transparency in Developing Economies The heatmap reveals great variability in public debt data disclosure among the 76 LIDCs included in the analysis. The key take-aways of the analysis of the heatmap results are: • Public debt data are not systematically published or updated, particularly in Sub-Saharan Africa and small states. Both the 2019 and 2020 assessments show that close to 40 percent of LIDCs did not reach minimum data disclosure standards as defined by the methodology, either because they have never published any debt data, or because their debt data are more than two years old. • In available publications, standard debt instruments (loans and securities) are typically covered. Of the 46 LIDCs that publish regular debt data, 60 percent cover loans, guarantees and securities (which can be regarded as an adequate level of instrumental coverage for LIDCs)29. In 40 percent of LIDCs, at least one of these instruments is missing (Figure 2.4). • Extending sectoral coverage remains a challenge. Only 18 percent of the countries publish debt statistics where coverage aligns with what is expected based on their legal framework and borrowing practices (Figure 2.4). In most cases, statistics focus on central government direct debt only, and omit subnational debt (required to reach general government level) and/or SOE debt. • Total stock of central government guarantees is usually reported, but beneficiaries are rarely mentioned. In 70 percent of the countries with guarantees, the total stock of the guaranteed portfolio is reported. The median guaranteed debt in LIDCs is estimated at 5.1 percent of national GDP,30 with a significant dispersion of up to 23 percent. When total stock is reported, the beneficiary of the guarantee is mentioned in less than 20 percent of the cases. • Public private partnerships (PPPs) represent the other main category of explicit contingent liability that is under-reported.31 PPPs are used by over two thirds of LIDCs. However, less than 10 percent of LIDCs quantify the total direct and contingent exposure for the government in public debt or fiscal risk documents as published by Ministries of Finance (MOF) or Central Banks (CBs).32 The exposure to PPP-risk in LIDCs is estimated at an average of 2 percent of national 2020 GDP,33 but in 15 percent (8 LIDCs) this would exceed 5 percent of GDP. 29. Based on DeMPA methodology (2021). 30. Based on the latest LIC-DSA available. 31. Explicit liabilities are specific government obligations defined by law or contract. Implicit liabilities represent an expected burden for the government not in the legal sense but based on public expectations or political pressures. 32. The Public-Private Partnerships Fiscal Risk Assessment Model (PFRAM), jointly developed by the IMF and the World Bank Group (WBG), can guide countries in assessing fiscal implications of PPPs. 33. As per LIC DSF Guidance Note, the contingent liability component of a PPP is estimated as 35 percent of the country’s PPP capital stock (proxying for the present value of direct and potential future fiscal costs from PPP distress and/or cancellations). The capital stock is drawn from the World Bank Database on PPPs: https://ppi.worldbank.org/. 29 Debt Transparency in Developing Economies • Expenditure arrears are typically excluded from domestic debt statistics. With very few exceptions (e.g., Nigeria federal government), expenditure arrears are not included in the debt stocks.34 This reflects one of the key differences between the cash-based frameworks largely applied in LIDCs and accrual-based ones. A 2019 IMF report showed that at least 70 percent of countries in Sub-Saharan Africa had domestic expenditure arrears in 2018 amounting to an average of about 3 percent of GDP, with arrears to private suppliers accounting for the largest share (IMF, 2019). A subsequent report by the World Bank showed that arrears are expected to increase by more than 2 percent on average as a consequence of the COVID-19 pandemic (Utz et al, 2020). • There is ample room for improvement when publishing key debt-management documents. Medium-term Debt Strategies (MTDS) and Annual Borrowing Plans (ABP) should guide future borrowing and provide key references to investors and stakeholders. Forty five percent of IDA countries (34) publish a debt-management strategy; however, the MTDS is translated into a comprehensive ABP only in 9 countries (Figure 2.5). Figure 2.4 (% over total) Coverage of Public Debt Reports 100% in LIDCs 80% 60% 40% 20% 0% Instrument Coverage Sectoral Coverage No data Limited Adequate Source: Author’s elaboration 34. Expenditure arrears are financial obligations incurred by any level of the public sector for which payments have not been made by the due date. 30 Debt Transparency in Developing Economies Figure 2.5 (% over total) Publication of Debt 100% Management Documents in LIDCs 80% 60% 40% 20% 0% Strategy ABP No Partial (only external or domestic debt) Yes Source: Author’s elaboration. An analysis of the heatmap data shows that debt data disclosure is lower in fragile countries, due to the lack of solid legal and operational frameworks (Annex 1). The existence of specific legal requirements for disclosure (see section 4.4), increased capacity at DMOs and the usage of standard debt recording, and management systems (DRMS) significantly contribute to improving the level of debt disclosure in direct reporting. Our analysis also shows that availability of ratings and Eurobond issuances are highly correlated to levels of transparency in LIDCs, as a result of their own data requirements and the investor relations practices that they encourage. Introducing integrated and universal DRMS has a huge potential for increasing transparency. Our analysis shows that LIDCs benefit from standardized DRMS, despite the limitations of the existing options (e.g., inability to register complex debt instruments,35 limited integration with other PFM systems).36 Very few LIDCs have in- house expertise to develop and maintain a DRMS and this solution may not be cost effective. Instead, well designed, standardized and integrated DRMS would limit operational risk by simplifying debt recording, which is still a mostly manual task. In fact, the use of different references for key parameters (e.g., exchange rates, dates of loan repayment or disbursements, etc.), inconsistent computational formula and inaccurate data entries currently result in the need for frequent reconciliation between creditors and debtors, through letters, emails or even country visits. This practice, common to all LIDCs, is very costly and time-consuming and it can significantly delay and complicate restructuring negotiations (Box 2.1). 35. “It is a challenge for system providers to stay abreast of financial innovation and keep products and services relevant against increasingly sophisticated debt portfolios and transactions (i.e., securities, liability management transactions, etc.)” (WB, IMF, 2018). 36. Current DRMSs are mostly stand-alone software, which may limit the data quality control, as incorrect data entry has no impact in the public financial operations. On the contrary, if data registered in the DRMSs were to be used at different levels in integrated systems (for instance, to mobilize budget lines or generate payments), DMOs would have a significant incentive to maintain accurate and comprehensive data. 31 Debt Transparency in Developing Economies Box 2.1 Debt Restructuring: A Missed Opportunity for Transparency? Successful debt restructuring strongly depends on full disclosure of public debt records. Without accurate and comprehensive information, restructuring agreements may fail to put debt-distressed countries on a sustainable macroeconomic trajectory. Full transparency in debt restructuring is also a key condition for ensuring a common understanding of the debt relief requested from the different creditors, thus facilitating effective negotiations and earlier resolutions. Despite this, reporting inadequacies continue to be identified in the context of debt restructuring (Bon and Cheng, 2020; Minsat and Dossia 2020; Horn, Reinhart, and Trebesch 2020; Cheng et al 2016; Cruces et al 2013). These inadequacies have led to high-level calls for improved debt disclosure in the context of G20 Common Framework. Debt restructuring offers an opportunity to reconcile, update, and improve debtor and creditor records. Ideally, the first step in a restructuring process involves a data-reconciliation exercise between the debtor and its different creditors. Given the heterogeneity of reporting practices and the increasingly diverse pool of creditors, data-reconciliation processes have been lengthy and resource-intensive. This complexity means that financial advisors and IMF/World Bank staff are often involved in coordinating the data exchange and reconciliation.37 Under these conditions, debt transparency is expected to improve after a restructuring episode. Borrowers may have a positive incentive to fully disclose their debt portfolio to make the case for higher debt relief. In addition, the reconciliation process and the conditionality that is often attached to restructuring should produce more comprehensive and accurate debt records. Our analysis, however, (see Annex 3) indicates that restructuring episodes have limited impact on debt transparency. Figure B2.1.1 shows that countries that undergo debt restructuring do not improve their level of reporting to international debt databases. The effect of restructuring episodes on indirect debt disclosure is statistically insignificant in both Paris Club- and non-Paris Club-led restructuring. Figure B2.1.1: Debt Restructuring and Transparency Estimated Marginal Effects BoE - BoC Bon and Cheng Paris Club 1 Paris Club 1 Paris Club 2+ Paris Club 2+ Non Paris Club* 1 Non Paris Club* 1 Non Paris Club* 2 Non Paris Club* 2 DeMPA 1 DeMPA 1 DeMPA 2 DeMPA 2 IMF Program IMF Program Current Account (%GDP) Current Account (%GDP) GDP per Capita (In) GDP per Capita (In) Inflation (In) Inflation (In) State Fragility Index State Fragility Index -3.5 -3 -2.5 -2 -1.5 -1 -.5 0 .5 1 1.5 2 2.5 3 3.5 4 4.5 -3.5 -3 -2.5 -2 -1.5 -1 -.5 0 .5 1 1.5 2 2.5 3 3.5 4 4.5 sample size = 1,656; countries=97; R2 (within) 0.32 sample size = 1,498; countries=88; R2 (within) 0.30 Source: Author’s elaboration. 37. Under the HIPC Initiative, staffs of the IMF and the World Bank conduct a loan-by-loan debt reconciliation exercise jointly with the national DMOs. In all 37 HIPC-eligible countries that have reached the Decision Point under the HIPC Initiative, at least 80 percent of the debt stock has been reconciled. The reconciliation of the debt database is also conducted prior to the Completion Point, when countries receive full debt relief. The debt records resulting from the reconciliation process are used in the Paris Club framework agreement and bilateral agreements. 32 Debt Transparency in Developing Economies These findings may be explained by the following reasons: (i) Restructuring deals seldom require accurate and comprehensive recording of new terms, nor do they request the expansion of the instrument/sectoral coverage. The Paris Club website recognizes that “each organization compiling and publishing debt figures may have slightly different ways to categorize and to measure debt” and discussions about the comprehensiveness of debt data are not featured in the more detailed Paris Club agreements.38 In the case of non-Paris club creditors, little is known about their restructuring process including the extent to which comprehensive disclosure is emphasized in restructuring agreements. (ii) Recording of post-restructuring debt transactions can be complex. As discussed, capacity constraints create significant barriers to expanding the coverage of debt statistics. Restructuring episodes create additional burdens for DMOs as they need to analyze and negotiate proposals and reflect any changes in debt contracts as a result of the restructuring agreement.39 Improvements to reporting standards are therefore expected to take some time to materialize. (iii) Incentives to extend the debt coverage could diminish in the immediate aftermath of a restructuring episode. Unless improvements in data coverage is established ex-ante as eligibility criteria for creditors to engage in restructuring, recipient governments have no immediate benefit for improving it ex-post. In fact, this will likely result in the reporting of larger debt burdens which may be interpreted as a breach of the debt reduction plan negotiated by creditors and stakeholders. Any gaps in borrowers’ statistics can only be partially filled by creditors’ statistics. The IMF and the World Bank publish comprehensive and timely data on their financing.40 41 The Asian Development Bank, the African Development Bank, and the Inter-American Development Bank also have solid reporting standards as confirmed by their rankings in the Aid Transparency Index.42 The level of transparency of the other multilateral creditors is mixed and largely depends on the degree of accountability imposed by their internal policies and procedures.43 As for bilateral creditors, only some Paris-Club members disclose information on their lending operations via their website or publications.44 Their records therefore tend to be used for validation purposes only, particularly in the case of debt restructuring. The availability of private creditor data tends to depend on whether or not debt is tradable. With few exceptions, key information about tradable debt is available on main trading platforms (see Chapter 4.5 for details). Information disclosed by private investors on non-tradable debt is scant and ultimately depends on internal corporate policies as well as applicable laws and regulations. 38. See: https://clubdeparis.org/en/communications/page/classification. 39. In principle, there is not much difference between recording changes to existing debt records that are a result of restructuring agreements and recording regular debt transactions. However, recording complexities specific to the restructuring process include: (i) translating debt reduction extended on present value terms into nominal values for debt stock and service; (ii) capturing debt service deferral resulting from activation of deceleration clauses; (iii) accounting for complex debt-grant offset mechanisms (e.g., France C2D or IMF Catastrophic Containment Trust); and (iv) keeping track of the multiple ways that individual creditors restructure their claims (e.g. in the case of Paris Club, the time between conclusion of an Agreed Minute and completion of all bilateral agreements may take at least 12 months, and in the meantime debtors undertake parallel accounting to reflect the PC’s decision). Important differences may also arise in debtor/creditor records, notably when debt is forgiven; debtors will write-off the whole amount when the agreement is concluded, whereas creditors may record it sequentially as the forgiven debt service payments fall due. 40. www.imf.org/en/Data/IMF-Finances. 41. The World Bank Policy on Access to Information is available at: https://www.worldbank.org/en/access-to-information. 42. Source: https://www.publishwhatyoufund.org/the-index/2020/. Assessments under this index also cover the publication of loan terms. 43. In a 2018 survey of 25 MDBs, 16 had policies on public communication or disclosure (Engen and Prizzon, 2018). 44. According to the IMF/WB G20 Operational Guidelines for Sustainable Financing – Survey Results and Policy Recommendation, “Most countries provide information on their lending on the web, but it is not being done in a way that consolidates lending by all agencies, and for some, there is room to improve on the comprehensiveness of the data being reported”. In particular, only one third of the G20 countries that responded to the survey report lending terms in line with OECD requirements. Progress was achieved by the Finance Ministers and Central Bank Governors of the G7 in July 2021 by committing their respective countries to “publish our own creditor portfolios on a loan-by-loan basis for future direct lending by the end of 2021”. 33 Debt Transparency in Developing Economies 2.2 DEBT STATISTICS: Direct reporting aims to provide citizens and stakeholders with comprehensive INDIRECT DEBT DATA and timely data. In principle, indirect reporting shares the same goal but, over time, DISCLOSURE this channel has also dealt with concerns about misreporting and comparability of data published by national sources (Monnet, Truong-Loi, 2020). IFIs and rating agencies have thus become key in the process of integrating and standardizing records, filling the gaps in direct reporting channels and even replacing them in some cases. LIDCs may report debt data to four main statistical databases hosted by the IMF and the World Bank, which are closely aligned with international definitions: Quarterly Public Sector Debt Statistics (QPSDS), Quarterly External Debt Statistics (QEDS), Government Finance Statistics (GFS, annual), and the Debtor Reporting System (DRS). Each of these databases was created for a different purpose, and so their debt coverage and definitions differ (World Bank, IMF, 2020c). The DRS provides the most granular data breakdown, and the broadest coverage for external debt. It is also the only compulsory data disclosure exercise; the others are voluntary. In fact, countries that have borrowed at least once from the World Bank are required to meet certain reporting obligations, including the completion of DRS standardized templates on their external public and publicly guaranteed (PPG) debt on a quarterly and annual basis. These data are then reviewed by World Bank staff45 and published in the International Debt Statistics (IDS) reports.46 Other IFIs or regional institutions maintain databases with narrower geographical or instrument coverage (e.g., the African Development Bank’s Bond Market Database or the forthcoming OECD database on external commercial debt). Additional databases that rely on one, or multiple, indirect sources add to the complexity of this environment.47 The list of the main indirect reporting databases with their respective features is presented in Annex II. In parallel to this structured data collection, indirect reporting may also occur on an ad hoc basis. On top of the direct and indirect standardized channels, additional requests for updated debt data typically come from IFIs (e.g., for the purpose of IMF Article IV missions or IMF/World Bank lending operations)48 or other key stakeholders (e.g., rating agencies). Official lenders (bilateral and multilateral) also feed into international databases, such as the OECD’s Creditor Reporting System. The resulting reporting eco- system is portrayed in Figure 2.6. 45. In the case of the World Bank International Debt Statistics, “data accuracy and comprehensiveness is ensured by validation with other sources such as market data, creditor data, other external statistics such as BOP/IIP, QEDS —including data used in debt analytical exercises led by the World Bank and IMF, such as the medium-term debt strategy (MTDS) or DSA—and rigorous follow-up with government authorities” (IMF/WB, 2020a). 46. In June 2020, the WB published detailed external debt stock and service data for 68 IDA eligible countries in the IDS database, including a previously unavailable breakdown by creditor. http://datatopics.worldbank.org/debt/ids/. 47. For example, WB World Development Indicators (WDI) access its debt data from the IMF Government Finance Statistics Yearbook (GFSY). 48. The IMF Articles of Agreement establish an obligation for every IMF member to provide a minimum set of data to the Fund for its activities, as set forth in Article VIII, Section 5 (supplemented in 2004 by a decision by the IMF Executive Board). This is a narrow set of debt data: stocks of gross national external debt, and central government-guaranteed debt including currency and maturity composition and, if possible, the extent to which debt is held by residents or nonresidents. In addition, and while not as a legal obligation under the IMF Articles of Agreement but as a condition for the provision of financing under IMF-supported programs, the IMF can request debt information beyond the above-mentioned minimum set of data requirements if it determines that such information is of critical importance for achieving the goals, or monitoring the implementation, of the member’s program (and can decide not to lend if such critical information is not forthcoming). 34 Debt Transparency in Developing Economies Figure 2.6 Direct and Indirect Data Reporting of Government Debt Regional/ WB International Organizations Indirect (e.g., OECD) Reporting Private data Following IMF platforms beneficiaries’ (e.g., Bloomberg definitions Multi & Bilateral and standards Debtor Creditors Other IFIs, Credit Rating Agencies Direct Reporting Creditors’ Debtors’ Following Official Statistics Official Statistics compliers’ definitions and standards Data disclosure Additional data Data reconciliation Source: Rivetti 2021. The main sources of direct and indirect reporting in LIDCs are debt records from the national DMOs; however, final debt data differ significantly. Discrepancies across public debt data, available via direct and indirect debt reporting can be significant. Figure 2.7 portrays the differences in 2019 total debt stocks between the WB/IMF’s LIC- DSA and authorities’ websites, as a percentage of each country’s GDP. The LIC-DSA increases the debt stock that is reported in official statistics by an average of 5 percent of GDP. However, the degree of dispersion fluctuates between -15 and 30 percent. Figure 2.7 30.0 Indirect-Direct Reporting Debt 25.0 Stock Gap (in percent of 2019 20.0 national GDP) 15.0 10.0 5.0 - (5.0) (10.0) (15.0) Source: Rivetti 2021. 35 Debt Transparency in Developing Economies This degree of mismatch can be explained by the following: • Sectoral and instrument coverage: Indirect reporting contributes to the expansion of coverage beyond what is officially published. In particular, external agents may include debt instruments or sectors that do not fall under the national definition of public debt (see Chapter 4) or are not covered in direct sources, thus increasing a country’s debt coverage. These factors highlight the potential for higher direct disclosure levels by tapping into existing data sources within the country, and a more effective standardization of reporting requirements by external agents. In Box 2.2 we study the hypothetical impact of the inclusion of SOEs debt. • Use of different debt definitions and valuation methods: In the absence of unified and universal guidelines for reporting debt statistics, several divergences emerge across countries and reporting sources. For example, the Public Sector Debt Statistics: Guide for Compilers and Users (PSDS, 2011) requests countries to report all debt at nominal value,49 while securities should be reported at market value. Ten years after the PSDS publication, however, all LIDCs still report their debt at face value only.50 Similarly, while external debt should be defined by the creditor’s residency - according to PSDS and External Debt Statistics guide (EDS, 2013), as well as Balance of Payments (BoP) / International Investment Position (IIP)–direct reports rely on the more accessible currency definition. This practice also reflects debt managers’ preference for a criterion that facilitates the analysis of the debt and budget exposure to foreign exchange volatility. Indirect agents are able to comply with the residency definition only in specific cases when information is available, as acknowledged in the LIC DSA Guidance Note.51 In fact, debt denominated in foreign currency is still used as a proxy for external debt in 43 percent of the 2019 DSAs. Finally, ambiguity over the classification of certain debt may have severe implications in the case of debt restructuring. As shown by recent discussions around the DSSI and G20 Common Framework, the classification of certain claims on SOEs, debt held by state-owned banks or development banks has been interpreted differently by some creditors and debtors. In addition to leading to inconsistent treatment in statistics, the lack of adherence to international definitions for official/commercial debt by some creditors can create uncertainty around the perimeter of debt to be covered in any restructuring, it can also create uncertainly around the terms of treatment, affecting prospects for reaching swift agreement (IMF, 2020). 49. Nominal and face value definitions tend to be used interchangeably. However, face value is “the amount to be repaid at maturity”, while nominal value is the “amount that debtor owes to the creditor at any moment in time. Conceptually, the nominal value of a debt instrument can be calculated by discounting future interest and principal payments at the existing contractual interest rate on the instrument” (PSDS, 2011). 50. LIDCs may not have liquid issuances to price their debt, particularly domestic debt, and most LIDCs use debt recording systems that define debt at face value in their software and do not allow computation of market value” (WB/IMF, 2020). The computation therefore has to be done manually for each loan, which is a time-consuming exercise that requires sound financial skills. Moreover, DMOs may have clear incentives to produce debt statistics at face value, since most domestic law and fiscal rules, and the LIC-DSA, require this evaluation method. 51. “Because of difficulties in record-keeping (e.g., as a result of secondary market trading and data limitations in LICs), and where non-resident participation in domestic debt market is not significant, debt denominated in foreign currency can be used as a proxy for external debt”. (WB/IMF, 2018). 36 Debt Transparency in Developing Economies • Recording errors: Data collected in over 20 WB-led missions on Medium-Term Debt Strategy (MTDS)52 in LIDCs between 2010 and 2020 reveal severe recording mistakes in debt databases in more than 30 percent of cases. The assessment of the quality of DRS data conducted by the WB on an annual basis confirms these findings: of 74 LIDCs, 15 (20%) do not submit comprehensive reporting templates for DRS or, if submitted, the quality is not considered satisfactory. Among those who provide a complete template, 28% have at least one type of recording mistake (Table 2.1). Table 2.1 Assessment of the 2020 Reports by IDA Countries to DRS No/Partial/ Unsatisfactory Complete Reporting Reporting Number of 20% 80% countries (%) of which 28% affected by one or more of the errors below stocks/flows missing information transactions not Error types inconsistencies on new commitments reported Frequency (%) 15% 17% 17% Source: WB Development Data Group. Note: Author’s elaboration. • “Hidden debt”: external agents may be able to identify debt instruments not disclosed in official statistics (“hidden debt”). However, this rarely happens as non-disclosed debt instruments can be identified only when granular information is provided by the creditor, which is seldom the case. Otherwise, given the existing voluntary-based data disclosure framework, any uncovering of “hidden debt” can only be triggered by stronger parallel engagements at governance level, through audits or an analysis of project funding that may allow some form of cross-validation. 52. Prior to any MTDS mission, a data validation exercise is conducted to verify that (i) financial terms for each debt instrument are present and (ii) debt stock-flows are consistent. Inconsistencies are regarded as “severe” if the error is higher than 1% of the total debt outstanding. 37 Debt Transparency in Developing Economies Box 2.2 Impact of Including SOEs Debt in Public Debt Coverage As discussed in Chapter 1.1, sectoral coverage of public debt statistics includes, at best, general government, both in LIDCs and advanced economies.53 The use of this debt measure derives from the principle of “the predominantly tax-financed nature of government, compared to the market revenues of government-owned corporations (Abbas et al, 2020). Although not included in debt figures, SOE debt data make up part of the reporting on SOEs made available by the majority of advanced economies (OECD, 2018).54 By contrast, data on SOE debt are seldom centralized or disclosed in LIDCs. As described in Chapter 2, in most LIDCs, SOEs are subject to weak reporting requirements both in terms of direct and indirect reporting. Many LIDC financial statements are not systematically available or audited, furthermore, their debt position is rarely collected and centralized by the MOF; the information may also not be disclosed to the unit in charge of producing debt statistics, unless the debt is on-lent to the SOEs or guaranteed by the central government. These factors represent a challenge for indirect reporting that are expected to cover the entire public sector. The solution found in the LIC-DSA, is to introduce a contingent liability stress test based on a standard default value of 2 percent of GDP in countries where SOE debt is not fully captured under the debt coverage.55 Given the rapid evolution of SOE debt portfolios, the World Bank launched a data collection initiative to better estimate the size and composition of SOEs’ debt in LIDCs and to strengthen the debt coverage in LIC DSAs.56 Preliminary results on a first sample of 15 countries57 for which data are deemed comprehensive (i.e., including on-lent, guaranteed and non- guaranteed) show the following: • The median SOE debt levels (domestic and external) amounts to 7.3 percent of GDP, with a broad distribution across countries (ranging from 0.4 percent to 18.1 percent). Nearly 90 percent of total outstanding SOE debt is external. On-lent activity is the dominant financing source (49 percent of the total). • More than 80 percent of the debt was contracted by SOEs operating in the transportation, and the energy and extractives sectors (41 percent and 40 percent of total debt, respectively). SOEs in the transportation and energy & extractives sector receive mostly on-lent and non-guaranteed funds (90 percent and 75 percent of the debt to that sector, respectively) while the industry, trade & service sector tends to be financed mostly via guaranteed debt by the central government (80 percent of the debt). • The creditor groups that provide the largest financing support to SOEs are financial institutions and export credit agencies (about 33 percent, and 30 percent of total debt respectively). In the current sample, 94 percent of export agency financing is provided by Chinese institutions. International organizations provide 18 percent of total financing to SOEs, and this is via on-lending in 88 percent of the cases. • In our sample, China is the largest country-counterpart, representing 41 percent of total SOE financing, distributed across export agencies, official creditor roles, corporations and private banks, and financial institutions. 53. See the Maastricht definition of debt, which is the general consolidated gross debt in line with the 2010 European Systems of Account. 54. The SOE survey conducted by OECD includes the 35 OECD member countries as well as selected emerging economies. Around half of the 57 countries surveyed make available online some form of aggregate reporting on state-owned enterprises. In cases where no report is produced, 7 countries in the sample provide online inventories of SOE’s, in most cases with links either to entities consolidated financial statements, or to entities’ websites where annual reports are generally available. 55. According to an IMF survey (2016), the median SOE external liability is equal to 2 percent of the GDP. 56. Significant efforts are on-going to ensure a larger sectoral coverage in LIC-DSA Since the roll-out of the revised LIC-DSA (2019), sectoral debt data coverage has been broadened in more than 15 countries 57. The data submission covers 165 SOEs and 481 individual loans in the following countries: Benin, Bhutan, Cameroon, Central African Republic, Comoros, Côte d’Ivoire, Ethiopia, The Gambia, Honduras, Kenya, Madagascar, Mozambique, Senegal, Togo, Tuvalu. A methodological caveat is that single data points reflect different base years (ranging from 2018 to 2021), as SOEs fiscal statements are not systematically available on annual basis. 38 Debt Transparency in Developing Economies • On-lent loans offer the most advantageous financial conditions. 100 percent of the loans bear a maturity greater than 10 years and 48 percent have an interest rate not exceeding 2 percent. • Non-guaranteed debt represents 34 percent of total SOE debt. The majority of which is provided by private financial institutions, export agencies, and private corporations (40 percent, 30 percent and 25 percent of total non-guaranteed debt, respectively). China remains a key stakeholder providing 35 percent of total non-guaranteed debt. Although financial conditions are less advantageous than those of on-lending, average terms remain concessional with 23 percent of the debt having a maturity greater than 10 years and 21 percent bearing an interest rate not exceeding 2 percent. Figure B2.2.1: SOEs’ Debt by Category 50% 40% 30% 49% 20% 34% 10% 17% 0% On-Lent Non Guaranteed Guaranteed Source: Authors’ elaboration. Figure B2.2.2: Sectoral Distribution of SOEs’ Debt Energy and Extractives Information and Communications Technologies Water, Sanitation & Waste Management Industry, Trade & Services Transportation Agriculture, Fishing, & Forestry Other Source: Authors’ elaboration. 39 Debt Transparency in Developing Economies The same factors above discussed (different debt coverage, definitions, recording errors and “hidden debt”) explain existing differences in the debt figures across indirect sources, as well as those across different vintages of the same database. Box 2.3 shows how they have been triggering significant revisions of debt data in LIC DSAs. Box 2.3 LIC-DSA Ex-Post Debt Revisions Non-comprehensive and inconsistent public debt data manifest across different databases, and also across different vintages of the same data source. Comparing different vintages of nominal public and publicly guaranteed (PPG) external or total public debt series for countries applying the LIC DSF reveals substantial differences in historical actuals for several of them. Forecast errors are well documented, including in LIC DSF data, but differences across vintages are less extensively discussed and documented. Revisions of historical debt data may occur for several reasons. Nominal debt data revisions, particularly for the most recent year, may arise because of statistical errors and revisions or updated exchange rates. Revisions can also be the result of increased coverage of government sectors or instruments (see examples below for Madagascar and Senegal). Figure B2.3.1: Madagascar’s Public Debt across Different DSA Vintages (billions of MGA) 25,000 20,000 15.000 10,000 5,000 CB and SOE liabilities included in debt coverage 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2020 DSA 2018 DSA Actuals 2018 DSA Projected 40 Debt Transparency in Developing Economies Figure B2.3.2: Senegal’s Public Debt across Different DSA Vintages (billions of CFA) 12,000 10,000 SOE debt and extrabudgetary funds includedin debt coverage 8,000 6,000 4,000 2,000 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2020 DSA 2017 DSA Actuals 2017 DSA Projected Discrepancies in historical debt data (between the most recent and older DSA vintages) that are linked to an expansion of their debt coverage are identifiable under the LIC DSF. In Madagascar and Senegal, the inclusion of additional components of the public sector in the debt coverage resulted in significant upward revisions of PPG external and total public debt. Senegal’s 2020 DSA expanded debt coverage to include para-public entities and SOEs, compared with the debt coverage of the 2017 DSA (and previous vintages) that was limited to the central government. As a result, the end-2017 debt stock increased by 28.5 percent or 14.2 percentage points of GDP, with PPG external debt increasing by 3.0 percent of 2020 GDP.58 For Mozambique, the 2020 DSA expanded the debt coverage to include SOEs’ domestic debt and central bank external liabilities. This coincided with a yearly average increase of USD 460 million over the period 2007 to 2017, or 2.2 percent of 2020 GDP. Large debt revisions can also result from “hidden debt”, as in the case of Mozambique. In 2016, previously undisclosed external loans in the amount of more than USD 1.15 billion, or roughly 7 percent of 2020 GDP were discovered. These loans were contracted by SOEs during 2013/2014. They were included in LIC DSAs after 2016, explaining the gap in historical actuals of PPG external debt. 41 Debt Transparency in Developing Economies Figure B2.3.3: Mozambique’s Public Debt across Different DSA Vintages (billions of USD) 16 14 12 10 8 6 4 Additional loans of three SOEs included in debt coverage 2 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2020 DSA 2015 DSA Actuals 2015 DSA Projected 2.3 FINANCIAL REPORTING Accurate financial reporting can strengthen debt transparency efforts. Including the output of financial reporting can augment the picture of public debt and broader public finances than just public debt statistics taken on their own. Financial reporting standards generally refer to the standards captured by the International Public Sector Accounting Standards (IPSAS). Many IPSAS are based on International Financial Reporting Standards (IFRS) and aim to improve the quality of general-purpose financial reporting by public sector entities,59 leading to better informed assessments of resource allocation decisions made by governments. Despite not being mandated by an international regulator,60 IPSAS are increasingly applied by national and subnational governments, as well as intergovernmental organizations and institutions. However, the standards used at country level are often home-grown, or only influenced by IPSAS (Box 2.4) and their use in LIDCs is still very limited. By 2019, only 3 LIDCs (Côte d’Ivoire, Nigeria, and Tanzania) had directly adopted accrual basis IPSAS and an additional 23 IDA-eligible countries had adopted cash-basis IPSAS or are transitioning to accrual basis with varying degrees of implementation. 58. The increase in percentage points of GDP is estimated using GDP values of 2020 DSA to correct for the GDP rebasing that occurred between 2017 and 2020. 59. General purpose financial statements, prepared on an accrual basis, can provide a primary source of information to assess whether government has met its commitments to its multitude of stakeholders. While some specific aspects of their design and implementation are left to local preferences, the minimum requirements should include the following three core statements: (i) a statement of financial position, recognizing assets and liabilities and disclosing matters of material impact; (ii) a statement of operating performance showing how the financial position changed through the year in terms of revenues / expenses and gains / losses; and (iii) a statement of cash flows, showing inflows and outflows and changes in cash stocks organized by cashflows from operating performance, from investing, and from financing transactions. 60. IPSAS are issued by the IPSAS Board, an independent standard-setting body created under the auspices of IFAC (International Federation of Accountants). 42 Debt Transparency in Developing Economies Box 2.4 Status of Sovereign Financial Reporting About two-thirds of all countries intend to publish their accounts on an accrual basis by the end of 2023.61 As of 2019, 52 percent of jurisdictions have adopted some or all IPSAS, including 15 percent of LIDCs.62 Many jurisdictions have adopted a gradual approach to accrual-based IPSAS due to national political and economic realities, as well as capacity constraints of skills and resources. Other countries do not adopt IPSAS directly, and instead adopt national standards based on IPSAS to varying degrees. Country practice shows three different strategies for adopting IPSAS in public sector generally accepted accounting principles (PS GAAP), as follows: 1. Direct adoption: Countries move to replace their own national public sector accounting standards with IPSAS through legislative reference. 2. Modified adoption: Countries adopt (all or some) IPSAS standards through replication or reference, adjusting for any specific jurisdictional features. This could be achieved by revising legislation or issuing national public sector accounting standards which are equivalent to IPSAS. 3. Local standards informed by IPSAS: Countries modify their own PS GAAP using IPSAS for guidance, resulting in national PS GAAP that are consistent with only selected parts of selected IPSAS. This could be achieved by revising legislation or by issuing national public sector accounting standards derived from IPSAS. Many countries are developing the underlying systems and institutional arrangements necessary to support the transition to accrual accounting. Accrual accounting records the economic substance of transactions and events when they occur rather than when cash is exchanged, and therefore recognizes liabilities based on the nature of the obligation. As they prepare for the transition to accrual basis IPSAS, LIDCs are transitioning to Cash Basis IPSAS as a first step. Efforts are being made to modernize charts of accounts, which are often rudimentary and do not contain coding structures to capture funding source, projects, activities, and programs. Investments are being made in technology to help develop the current fragmented, suboptimal systems to provide reliable data for compiling government financial statistics (GFS) and financial reporting. Accrual information and disclosures are being progressively included in government financial statements to supplement cash basis information. For countries currently struggling with timely and accurate cash basis accounting, the transition to accrual accounting will take considerable time and effort. Differences between IPSAS and debt statistical principles and rules can have a significant impact on the level of debt reported under the two frameworks. As in the case of the deviations among debt statistics shown in section 2.2, the differences between financial reporting and the statistics approach on public debt depends on key methodological criteria such as: (i) sectoral coverage; (ii) instrument coverage, (iii) valuation; (iv) asset/liability approach. 61. https://www.ifac.org/system/files/uploads/IFAC/IFAC-CIPFA-Public-Sector-Index-2018-Status.pdf. 62. https://www.ifac.org/system/files/publications/files/IFAC-International-standards-2019-global-status-report.pdf. 43 Debt Transparency in Developing Economies (i) Sectoral coverage and consolidation Under IPSAS, consolidated financial statements are prepared for “whole of government” based on the concept of control. Control is defined as an entity’s ability to influence the nature and number of benefits through its power over another entity, and the ability to exert influence over the returns through power over the investee. The concept of control brings more arrangements directly onto the balance sheet for recording and reporting purposes. For instance, while independence of technical judgment is accorded to many state bodies under law, this independence does not typically apply to the way resources are used, and the activities in which they can engage. If these bodies were to cease functioning, the state would have a residual claim on the assets devoted to their purpose and any assets may incur the residual liabilities on closure. As a result, the consolidation criteria may not align with the general government sector, which is the basis of GFS, and often results in a broader scope of consolidation. For example, GFS consolidation does not include entities engaged in market activities such as SOEs. Conversely, local governments may not be consolidated under IPSAS if they are not controlled by central government. (ii) Instrument coverage Rather than defining specific categories of debt, IPSAS provide a definition of liabilities, based on principles and standards determining the conditions under which liabilities arise. These standards apply to all financial liabilities, including any assets that have been pledged as collateral; coverage is also extended to debt-creating arrangements such as those that are the result of PPPs, pension obligations, central bank swap lines, deposits, and long-term trade credits. As highlighted in Chapter 3, lack of information on some of these of transactions is a significant hindrance to debt transparency. Expenditure arrears are also covered, contrary to debt statistics that typically only cover arrears on debt payments. The rationale is that an overdue bill owed by government to a private sector supplier or an overdue salary to a public official are de facto forms of enforced borrowing by government from that supplier or worker. As a result, measuring and capturing fiscal arrears in debt reports is key to transparency, especially in LIDCs where these can be substantial, as shown in Section 1.3. Contingent liabilities are disclosed in the notes of the balance sheet under IPSAS. Professional judgement and estimates are required to determine the likelihood that contingent liabilities may materialize, as well as to determine amounts. Unquantifiable amounts are disclosed unless likelihood is deemed to be remote.63 (iii) Valuation method IPSAS recognizes debt based on fair value. While IPSAS recognizes debt based on fair value, as shown in Section 2.2, statistical reporting frameworks recognize debt at nominal value (market value is only on securities). This difference is particularly critical for concessional loans: under IPSAS loans are recognized at fair value (e.g., present value of future cash flows discounted at market rates), while the difference between the face value and fair value is accounted for as a gain or a loss; the loan is subsequently measured at amortized cost using the effective interest method, which 63. GFS capture financial guarantees as explicit contingent liabilities and disclose these as a memorandum account at face value (GFS Manual 2014, para 7.255). 44 Debt Transparency in Developing Economies calculates the effective interest for each period based on contractual cash flows.64 This approach leads to adjustments to the liability/asset and the balance sheet of the debtor/ creditor,65 depending on the difference between contractual interest rates and market interest rates. In the case of a restructuring, the debt would be treated as repaid and a new loan recognized if the changes significant.66 If the changes are not significant, the carrying amount would be adjusted to reflect the new terms. Box 2.5 Total Government Liabilities: IPSAS vs. Direct Reporting As a result of significant methodological differences, total liabilities reported under IPSAS are not directly comparable with GFS or other statistical debt figures. The discrepancy–driven by account payable, provisions and other liabilities not recognized as debt under GFS is sizeable as shown by the following examples: United Kingdom: according to IPSAS, as of 2019 there are GBP 2.5 trillion net liabilities per Whole of Government Accounts (WGA), while according to GFS public sector net debt amounts to GBP 1.8 trillion. Ecuador.67 As of 2019, IPSAS reflects liabilities for an amount of US$73.3 billion while debt statistics figures show US$51.7 billion. Philippines.68 As of 2018 IPSAS reflects PHP 8 billion total liabilities while GFS shows PHP 6.8 billion of general government gross debt. Tanzania: As of 2016, according to IPSAS there are TSh. 61 trillion total liabilities, including 41 trillion of short and long-term borrowings, however, GFS presents TSh. 37.6 trillion of total national debt. (iv) Asset/liability approach Under accrual-basis IPSAS, all assets and liabilities of consolidated entities are recognized in the government balance sheet, and transactions between members of the same economic entity are eliminated. Ideally, a government balance sheet will provide a comprehensive view of all accumulated assets and liabilities that the government controls. Active balance sheet management enables countries to better plan revenues, consider risks, and improve the information base for fiscal policymaking. 64. The carrying amount of the loan is increased as interest is charged and decreased as payments are made. Interest is calculated based on the carrying amount of the loan and the effective interest rate. The effective interest rate is the rate that exactly discounts estimated future cash payments through the expected life of the loan to the amortized cost of the loan. 65. In the wake of its sovereign debt crisis, for instance, the face value of Greece’s debt was reported at 180% of GDP under PSDS. However, due to the concessionary terms of the bailout (low interest rates and long-term maturities) the debt was estimated to be 68% of GDP under IPSAS. On the counterparty’s side, the receivable on the books of Germany was also recorded at nominal value and was not discounted to reflect the concessional arrangements. 66. Terms are substantially different if the discounted present value of the cash flows under the new terms discounted using the original effective interest rate, is at least 10 percent different from the discounted present value of the remaining cash flows of the original financial liability. 67. Ecuador is gradually transitioning to IPSAS. 68. The Philippines reports its annual financial statements largely consistent with accrual-based IPSAS. Its 2018 Annual Financial Report reported general government gross debt of 38.9% of GDP. 45 Debt Transparency in Developing Economies 2.4 CONCLUSIONS GFS and IPSAS have generally come to be recognized as normative architecture for financial and statistical reporting, yet compliance is uneven. In line with the debt transparency framework described in Chapter 1, the production of accurate, timely, and comprehensive debt statistics and the adoption of sound accrual standards are subject to the same challenges: • Lack of adoption of non-enforceable international regulations. GFS and IPSAS are voluntary, and countries adopt them to varying degrees based on national preference, capacity, or national legislation. In particular, very few LIDCs adopt international debt reporting standards and definitions or IPSAS rules. In addition, the specific deviations from international standards and their effects are seldom disclosed, which ultimately affects transparency and limits cross-country comparability. • Lack of operational pre-requisites. The production of debt statistics and implementation of IPSAS requires several prerequisites, including supportive institutional and legal frameworks, adequate technological capacity and information systems, and sufficient staff capacity (see Annex 3 for a detailed discussion of the enabling factors for accrual accounting). These preconditions are not easily met in LIDCs and the situation is aggravated by the low priority given to reporting in public financial management (PFM) reforms in resource-constrained environments. Against this backdrop, this report calls for the following policy actions: (I) Align international reporting standards and foster their adoption Efforts to expand sound debt reporting through direct and indirect statistical reporting and through financial reporting should be accelerated to strengthen debt transparency. As shown in this chapter, idiosyncratic differences across the different standards are difficult to overcome. The long-term goal should be a gradual alignment of the debt statistical and financial standards along common international standards, monitored and enforced by an international organization. In the meantime, there are intermediate and easily achievable steps that can be pursued: • Strengthen countries’ direct reporting. Public and direct access to debt statistics is a key accountability mechanism that is still absent in over 30 percent of LIDCs. Comprehensive and accessible direct databases would also improve efficiency in data exchanges with external agents, reducing the burden for DMOs to meet uncoordinated data requests from external agents. • Improve clarity and consistency in indirect reporting standards. External agents responsible for indirect reporting should clarify the methodological criteria supporting any deviation from direct sources and explicitly mention the level of sectoral and instrument coverage of their analysis. • Make progress towards compliance with international financial reporting standards. With the support of development partners, LIDCs are encouraged to implement accrual financial reporting based on IPSAS, which have evolved as the normative international framework. 46 Debt Transparency in Developing Economies • Document progress toward compliance with financial reporting standards. This can be achieved by disclosing all material deviations from IPSAS and the impact of the deviation in the financial statements, to be attested by external auditors, or by relying on internationally recognized formal assessment on a periodic basis.69 Establishing strong governance arrangements over the assessment framework and engaging international stakeholders, such as the IPSASB, would ensure robustness of the framework and consistency in its application. More ambitious is the creation of an agency within an existing IFI tasked with producing “soft law” (i.e., non-binding guidelines and opinions) on fiscal and debt statistics, as well as liaising with national authorities to periodically monitor and report on countries’ adherence to international standards. A reference, in that sense, is Eurostat that has played a key role in supervising the quality and harmonizing statistics at EU level.70 (ii) Promote the design and implementation of modern and integrated IT systems allowing systematic data cross-validation. As shown in 2.3, the debt recording process is still a largely manual exercise in which weak understanding of financial and legal terms and human errors may trigger significant operational risk. This typically becomes evident during debt restructuring when data reconciliation between debtor and creditor is needed. An integral solution to these issues could be the creation of a repository for public debt data. Annex 4 proposes a concrete strategy to develop and implement an international loan repository (ILR), the goal of which is to provide a platform to reconcile debt records between creditors and debtors, thus improving the accuracy of debt records and limiting operational risk. The ILR would also enable the creation of a central database, which can be used for public reporting purposes.71 The ILR can also pave the way to the creation of a unique DMRS thereby ensuring that all participating countries benefit from the latest technological advances and standardizing debt reporting. One of the key features of a new generation of DMRS should be the capacity to allow for a smooth transition to accrual accounting to ensure a reciprocal information flow between the DMO and accrual accounting systems. This would ensure full consistency between debt data used for statistical purposes and budget preparation. 69. An IPSAS gap analysis is typically the initial step in the IPSAS adoption process. Several developments partners, including the World Bank, and consulting firms can conduct this analysis. 70. Eurostat is a Directorate-General of the European Commission. 71. A similar approach has been proposed by civil society (Eurodad and others, April 2019) and formalized by a G20 working note (Ayadi, Avgouleas, 2020). However, the proposal in Annex V differs from the approach mentioned in several ways: (i) participation in the ILR is entirely voluntary, although its use can be incentivized by official lenders; (ii) the scope of the ILR is limited to loans, as bonds tend to be more transparent (e.g., bond terms are disclosed through the prospectus and commercial platforms); (iii) the ILR is not replacing existing well-established platforms for issuing or trading debt instruments; (iv) ILR is not tracking the use of funds; (v) ILR is fed by the respective DRMS of creditors and debtors to facilitate reconciliation between the two sources. 47 Debt Transparency in Developing Economies (iii) Develop capacity at the debtor level Improving debt reporting is a long-term commitment that typically progresses through a phased approach over several years. Technical capacity for debt recording and reporting, as well as accrual accounting, should be built through appropriate prioritization of PFM reforms. Countries should target recruitment and retention of staff with the relevant financial and accounting skills. (iv) Promote data disclosure at the creditor level The G20’s Operational Guidelines on Sustainable Financing emphasize the need for official bilateral creditors to share debt information (guideline 2). The IMF-World Bank Diagnostic Tool on the implementation of the G20 guidelines identifies as a strong practice for transparency to publish loan-by-loan information by debtors, including terms, on a single website, with regular updates. In addition, strong practice requires creditors to use publicly available templates for their financing and to refrain from confidentiality clauses. (v) Make debt restructuring processes more transparent and conducive to debt data disclosure Transparency is one of the tenets of the IIF’s Principles for Stable Capital Flows and Fair Debt Restructuring72, which lay out a voluntary code of conduct for borrowers and lenders in sovereign debt restructuring. In its principles, IIF defines transparency as a disclosure obligation of the borrower towards its relevant creditors.73 The G20’s Operational Guidelines on Sustainable Financing (guideline 2.3) expand this definition, requiring creditors to engage in debt data reconciliation and to disclose information about their participation in debt restructurings in a timely manner. This report shows that restructuring episodes, as conducted until now, do not lead to significant improvements in the availability of debt data. However, they have the potential to facilitate greater levels of transparency if these processes incorporate the enhancement of debt data disclosure as an explicit objective. As discussed in Chapter 1, restructuring episodes temporarily relax both capacity and willingness to report, both of which are considered to be the main constraints of debt transparency. Based on the above, this report recommends the following: • Creditors and IFIs should consider making the reporting of comprehensive and complete public and publicly guaranteed (PPG) debt-stock of the public sector a prerequisite for granting debt relief. Extending the instrument and sector coverage of debt reporting is seldom achievable after restructuring has taken place. Conversely, countries would have strong incentives to produce timely, comprehensive, and accurate data if the debt relief process explicitly depended on it. 72. https://www.iif.com/Advocacy_old/Policy-Issues/Principles-for-Stable-Capital-Flows-and-Fair-Debt-Restructuring 73. “In the context of a restructuring, the debtor should disclose to all affected creditors maturity and interest rate structures of all external financial sovereign obligations, including the proposed treatment of such obligations; and the central aspects, including assumptions, of its economic policies and programs”. 48 Debt Transparency in Developing Economies • Clear pre-restructuring debt reconciliation guidelines and post-restructuring reporting rules should be applied. Irrespective of the restructuring channel, a set of procedures guiding debt reconciliation would be extremely useful especially for countries with low capacity. Similarly, clarity on the exact reporting treatment of the restructured instruments is needed. This could be facilitated by a recording framework involving creditor/debtor continuous validation, such as the one suggested in the international loan repository described in Annex 2. • Greater incentives should be established to encourage creditors and debtors to publish information on their participation in debt restructuring, as well as details, including amounts and changes in terms.74 This is not currently standard practice, especially for bilateral restructuring episodes. • Transparency around the restructuring process itself should also promoted. This goal needs to be balanced against legitimate needs for confidentiality during the negotiation phase, where disclosure of the terms under consideration may compromise the constructive participation of creditors.75 Comprehensive debt restructuring initiatives like the G20 Common Framework could benefit from disclosing key methodologies ex ante (such as on comparability of treatment)76 and publishing reasoned decisions ex post. This would marginally diminish the forum’s capacity to deal with diverse debtor and creditor country circumstances in a flexible way. However, it would assure constituents that the outcomes of a workout are not arbitrary and facilitate outside monitoring (UNCTAD, 2015). 74. In the context of the HIPC initiative, well established debt reconciliation processes and the disclosure of HIPC documents with detailed public debt information helped enhance debt transparency, providing important lessons for the implementation of the G20 Common Framework. 75. For this purpose, the borrower and lenders typically agree on a non-disclosure agreement (NDAs) and ensure confidentiality of the material non-public information. 76. Comparability of treatment (CoT) is critical for fair burden sharing in the context of the Paris Club-led restructuring, including the Common Framework ones. In Paris Club restructurings, CoT is evaluated ex-post on the basis of the assessment of one or more of the following three parameters: (i) changes in nominal debt service; (ii) debt reduction in net present value terms; and (iii) extension of the duration of the treated claims. The lack of a methodology to determine CoT implies some degree of judgement and may result in uneven treatment of some creditors or borrowers. 49 Debt Transparency in Developing Economies REFERENCES Ayadi, R., E. Avgouleas. 2020. “Time to Implement a Tech-Driven Sovereign Debt Transparency Initiative: Concept, Design and Policy Actions”; T20 Policy Brief. Bon, G. and G. Cheng. 2020. China’s Debt Relief Actions Overseas and Macroeconomic Implications. EconomiX Working Papers 2020. Brautigam, D., J. Hwang, J. Link. and K. Acker. 201) Chinese Loans to Africa Database, Washington, D.C.: China Africa Research Initiative, Johns Hopkins University School of Advanced International Studies. Cheng, G., J. Díaz-Cassou, and A. Erce. 2018. Official Debt Restructurings and Development. World Development 111, pp: 181–195. Cheng, G., J. Díaz-Cassou, and A. Erce. 2016. From Debt Collection to Relief Provision: 60 Years of Official Debt Restructurings through the Paris Club. European Stability Mechanism, 2016 Cheng, G., J. Díaz-Cassou, and A. Erce. 2018. Official Debt Restructurings and Development. World Development vol. 111(C), pp: 181-195. Cheng, G., J. Diaz-Cassou, and A. Erce. 2019. The Macroeconomic Effects of Official Debt Restructuring: Evidence from the Paris Club, Oxford Economic Papers, vol. 71(2), pp: 344-363. COMCEC. 2017. Improving Public Debt Management in the OIC Member Countries. Standing Committee for Economic and Commercial Cooperation of the Organization of Islamic Cooperation. Cruces, J.J., C. Trebesch. 2013. Sovereign Defaults: The Price of Haircuts. American Economic Journal: Macroeconomics, 5 (3): 85-117. Das, U.; Papaioannou, M. and Trebesch, C. 2012. Sovereign Debt Restructurings 1950– 2010: Literature Survey, Data, and Stylized Facts. IMF Working Paper WP/12/203. Dippelsman, R., Dziobek C., Gutiérrez Mangas C. 2014. “What Lies Beneath: The Statistical Definition of Public Sector Debt”. International Monetary Fund, Washington, DC. Dreher, A., Fuchs, A., Parks, B., Strange, A., Tierney, M.J. 2017. Aid, China, and growth: Evidence from a new global development finance dataset. Aid Data Working Paper No. 46. William & Mary’s Global Research Institute. Williamsburg, VA. Driscoll, J.C., Kraay, A.C. (1998). Consistent Covariance Matrix Estimation with Spatially Dependent Panel Data. Review of Economics and Statistics 80, 549–560. Bradbury, S., Brumby, J., and Skilling, D. 1999. “Sovereign Net Worth; An Analytical Framework”, NZ Treasury WP 99/3. Wellington. 50 Debt Transparency in Developing Economies Engen, L., A. Prizzon, A. 2018. “A Guide to Multilateral Development Banks”. ODI, London. Gelpern, A., 2018. “About Government Debt… Who Knows?” Capital Markets Law Journal. Vol. 13; No. 3. International Monetary Fund, 2019 “Regional Economic Outlook: Sub-Saharan Africa – Navigating Uncertainty. Background Paper: Online Annexes and Statistical Data arrears”. IMF. Washington, D.C. International Monetary Fund, 2014 “Government finance statistics manual 2014”. IMF. Washington, D.C. Kratz, A., Feng, A., Wright, L. 2019. New Data on the ‘Debt Trap’ Question. Rhodium Group Note New York, Rhodium Group. Murara, L. K., Irwin, T. C. and Wang, R. F. 2015. Trends in Fiscal Transparency: Evidence from a New Database of the Coverage of Fiscal Reporting. IMF Working Paper WP/15/188. Washington, D.C. Reinhart, C. 2019 The Curious Case of the Missing Defaults. Harvard Kennedy School Faculty Research Working Paper Series RWP19-004 Rivetti, D. 2021. “Debt Reporting in Developing Countries”, World Bank Paper, forthcoming. Seiferling M. 2020. “Apples, Oranges and Lemons: Public Sector Debt Statistics in the 21st Century”. Financial Innovation; London. Tudela, M., E. Duggar, A. Metz and B. Oosterveld. 2011. Sovereign Default and Recovery Rates, 1983–2010. Moody’s Investors Service. United Nations Conference on Trade and Development (UNCTAD), 2015. Sovereign Debt Workouts: Going Forward Roadmap and Guide Utz, R.; Mastruzzi, M.; Vazquez A.; Tawfik, E., 2020, “An Overview of the Potential Impact of the COVID-19 Crisis on the Accumulation of Government Expenditure Arrears”, World Bank, Washington, DC World Bank and International Monetary Fund. 2020. “G20 Note on Public Sector Debt Definitions and Reporting in Low-Income Developing Countries.” Washington, DC. World Bank and International Monetary Fund. 2018a. “G-20 Note: Strengthening Public Debt Transparency –The Role of the IMF and the World Bank”. Washington, DC. World Bank and International Monetary Fund. 2018b. “G-20 Note: Improving Public Debt Recording, Monitoring, and Reporting Capacity in Low and Lower-Middle Income Countries” Washington, DC. 51 Debt Transparency in Developing Economies World Bank and International Monetary Fund. 2013 “External Debt Statistics: Guide for Compilers and Users” (EDS). Washington, D.C. World Bank and International Monetary Fund. 2011. “Public Sector Debt Statistics: Guide for Compilers and Users” (PSDS). Washington, D.C. 52 Transparency of Debt Operations Debt Transparency in Developing Economies Debt reporting is the end-result of a multi-step operational process, and these steps tend to differ across debt instruments. While some forms of borrowing (such as funds from multilateral institutions) meet the highest standards in terms of procedures and overall reporting requirements, others are more prone to opaque borrowing processes. Moreover, given the same debt instrument, different procedures to acquire debt can lead to different transparency outcomes. For instance, when reporting domestic debt in national statistics it is impossible to know whether it comes from an open, market-based auction, or if it is the result of an agreement with a commercial bank against the payment of fees that are not disclosed in the cost of the instrument. These idiosyncratic differences add an additional level of complexity to the reporting challenges described in the previous chapter and deserve additional scrutiny. This is particularly important given the evolving borrowing landscape outlined in Chapter 1, as LIDCs begin to move away from traditional concessional financing sources. This chapter sheds light on the transparency challenges related to selected borrowing operations. Section 3.1 explores transparency in domestic debt issuances; Section 3.2 focuses on external commercial instruments (tradable and non-tradable) and discusses the role played by international rating agencies in promoting transparency; Section 3.3 investigates the characteristics of resource-backed lending; Section 3.4 examines public debt-related Central Bank operations. As in the previous chapters, the focus of the analysis is on LIDCs. 54 Debt Transparency in Developing Economies 3.1. DOMESTIC DEBT77 Deep and efficient domestic government debt markets help provide resilience to shocks in times of financial turbulence and convey multiple economic benefits. Recent financial crises, including the financial markets turmoil caused by the COVID-19 pandemic, have shown that the efficient development of domestic bond markets can increase financial resilience by mitigating rollover and currency risks, which are often a source of financial distress (World Bank-IMF, 2021). Transparency of domestic borrowing supports efficient market functioning. Transparent domestic borrowing increases investor confidence, thus reducing market uncertainty and creating positive expectations about the consistency of future policy decisions. (Wheeler, 2004). This can eventually lead to the reduction of the cost of funding over time. Moreover, transparent primary and secondary markets facilitate better price discovery, provide a level playing field for market participants, and play a critical role for market development. Yet, despite the increasing importance of domestic debt78 and the clear benefits of transparency, important transparency gaps remain in LIDCs. Transparency gaps include lack of market-based issuance procedures, irregular publication of issuance plans or auction calendars, and frequent cancellation of auctions. These practices may discourage investors from entering the local market. In addition, a lack of transparency around the issuance processes (auctions, private placements, etc.) reduces confidence and negatively impacts market development. The World Bank has recently launched a tool to track the transparency of domestic government securities issuances in LIDCs.79 80 The methodology is consistent with the relevant indicators of the WB/IMF Guidance Note for Developing Government Local Currency Bond Markets.81 The heatmap tracks five indicators presented in the WB/IMF note that are the most critical to ensure transparent domestic borrowing, namely: (i) use of market-based mechanisms to borrow from the domestic market; (ii) predictability of the government securities issuances; (iii) adherence to the issuance calendar; (iv) publication of the results of the borrowing transactions; and (iv) disclosure of secondary market operations. A country’s performance in each indicator is evaluated under a four- category scale. It ranks standards from low (red) to high (green), according to the criteria presented in Figure 3.1.82 77. Given the focus on operational and legal features of the related debt operations, in this chapter, domestic debt instruments are defined as those issued/contracted domestically, under the jurisdiction of a national court. 78. Between 2011 and 2019, the median domestic debt-to-GDP ratio in IDA countries almost doubled from 7 percent to 13 percent of GDP (WB-IMF, 2021). 79. Non-marketable securities issued for dedicated retail programs are also excluded from the assessment. 80. It should be noted that, in line with the findings of Chapter 1, the assessment of domestic debt practices is complicated by suboptimal reporting standards. Annual reports and statistical bulletins should be reliable sources for assessing the composition of the debt as well as implementation of the annual borrowing plan and the domestic issuance calendar. However, annual reports sometimes include only aggregate data for the domestic debt without clear separation of the marketable and non-marketable instruments. Moreover, it is often difficult to find information on the issuance mechanism of specific instruments. 81. https://documents.worldbank.org/en/publication/documentseports/documentdetail/790921615526044752/guidance-note-for-developing-government-local-currency-bond- markets. 82. An additional category is also introduced for countries that do not issue government securities in their domestic market. Indicators for these countries are not assessed and presented in grey. 55 Debt Transparency in Developing Economies Figure 3.1 Methodology Underpinning the Domestic Debt Securities Heatmap Domestic Market Use of market based Secondary market Dissemination of issuance calendars and results issuance mechanisms information Share of domestic debt Publication of the Implementation of the Publication of the Post-trade transparency issued through auction issuance calendar issuance calendar auction results No issuance of No issuance of No issuance of No issuance of No issuance of government securities government securities government securities government securities government securities in the domestic market in the domestic market in the domestic market in the domestic market in the domestic market N.A. (no auctions or N.A. (no publicly N.A. (no information or N.A. (no auctions or no N.A. (no post-trade no information available) available information no auctions or more than auction result published) information available) or issuance calenda 20% of the announced at least for the next month auctions cancelled) is not published) At least 20 percent of A monthly issuance Less than 20% but 10% Auction result including the Daily information is domestic debt is placed calendar (with auction or more of the auctions key parameters is published not published, however through auctions dates and instrument type) cancelled but not on the same day less frequent (e.g. weekly is published before the or monthly) publications beginning of the respective are available month and offering amounts for each type of instruments are published at least one day prior to the auction date. At least 50 percent of A quarterly issuance Less than 50% but 5% Auction result including the Information published domestic debt is placed calendar (with auction or more of the auctions key parameters is published on the next day through auctions dates and instrument type) cancelled on the same day but more is published before the than one hour after the beginning of the respective auction cut-off time quarter and offering amounts for each type of instruments are published at least one day prior to the auction date At least 80 percent of Yellow score is fulfilled and Less than 5% of the Auction result including the Information published domestic debt is placed a yearly issuance calendar auctions cancelled key parameters is published at the end of the day through auctions (with auction dates and within one hour after the indicative total issuance auction cut-off time amounts) is also published Source: Authors’ elaboration. 56 Debt Transparency in Developing Economies Figure 3.2 Domestic Debt Securities Heatmap: Summary from the 2020 Assessment (percent of LIDCs) Domestic Market Use of market based Secondary market Dissemination of issuance calendars and results issuance mechanisms information Share of domestic debt Publication of the Implementation of the Publication of the Post-trade transparency issued through auction issuance calendar issuance calendar auction results 20% 20% 20% 20% 20% 31% 41% 33% 20% 58% 8% 5% 3% 24% 8% 9% 16% 5% 36% 4% 32% 18% 39% 0% 10% Source: Authors’ elaboration. The key take-aways of the analysis of the heatmap results are the following: • The use of market-based issuance mechanisms is not a common practice in LIDCs. Fifteen of the 76 LIDCs have not developed a framework for local currency government securities and only 41 percent of LIDCs issue at least 50 percent of their domestic securities through auctions, which is the issuance mechanism that ensures the highest level of transparency and price discovery.83 On the other hand, twenty- four countries (32 percent) met the highest standard and used the auction method for at least 80 percent of their domestic borrowing. • The publication of the auction calendar is a weak point in LIDCs’ domestic borrowing. Primary markets would, in normal times, publish and adhere to at least a monthly issuance calendar, including details on tenors and indicative and/ or aggregate issuance volumes. Contrary to those “good practices” 32 countries (52 percent of the countries that issue securities) do not publish an issuance calendar, or the published calendar does not meet the minimum standards.84 Only 14 countries (22 percent) meet the highest standard by publishing a quarterly calendar and disclosing the auction dates with the indicative gross issuance for the entire year. Underperformance under this indicator is problematic, as predictability in the timing of issuance in primary markets reduces uncertainty for investors and allows them to plan their investments over a longer time horizon. 83. Issuance by syndication are also considered market-based issuance mechanisms and are typically used for new maturities or a new type of instruments. However, in LIDCs syndications are often accompanied by a lower level of price disclosure and transparency of regulations. Furthermore, domestic syndications have limitations and drawbacks, which need to be considered carefully. (See more in World Bank Domestic Syndication Background Note). To reflect the possible role of syndications in the domestic borrowing plan, the highest threshold for the first indicator (i.e., the number of securities to be issued with auctions) has been reduced to 80 percent. The other domestic debt sources are banking loans, issuances on a tap basis or private placements. Contrary to auctions, they do not facilitate price discovery. 84. Several countries publish an auction calendar for the entire year, typically including the auction date and type of instrument (bond or treasury bill). Some countries even specify the instruments by ISIN code or tenor. However, in the absence of indicative aggregate issuance amounts, all these countries fail to meet the minimum standards. An indicative gross issuance for the entire year is a key piece of information for investors to be able to assess the expected supply and adjust their demand and investment strategy accordingly. 57 Debt Transparency in Developing Economies • Once the issuance calendar is published, countries must adhere to it. Unjustified auction cancellations or rejections of bids are detrimental to the transparent primary market operation and may adversely impact investor participation. In an encouraging sign, 30 countries (49 percent) canceled less than 5 percent of the auctions in the first half of 2020, and most did not cancel a single auction. However, one-third of countries did not manage to meet the minimum standards. • No LIDC publishes the auction result within one hour after the cut-off time, with a minimum level of information.85 More than one-third of the countries manage to put out the publication at least on the same day; although they often do not meet the minimum standard given the lack of granular information. Some countries follow the practice of first informing auction bidders or primary dealers and make dissemination publicly available later. These countries do not meet the minimum standard, as the publication must ensure equal access to information to all investors. • The transparency of secondary markets also has significant room for improvement. Availability of pre-and post-trade information86 is still insufficient. Less than a quarter of LIDCs managed to demonstrate a minimum level of transparency in the secondary market by publishing average price and/or yield and aggregated volume.87 This outcome should be viewed in context as secondary markets tend to be very illiquid in LIDCs and countries may not have strong incentives to disclose details of single operations especially when their pricing is not aligned with the primary market pricing. • Regional centralized frameworks tend to improve the transparency of domestic securities issuance. The Eastern Caribbean Securities Exchange (RGSM), the Bank of Central African States (CEMAC), and UMOA-Titres (WAEMU) run very informative and transparent websites. Issuance calendars and auction results are published for each member country in a timely manner. Different statistics, public debt bulletins, and countries’ annual reports are published. These practices facilitate assessment of the implementation of the issuance calendar and overall help advance the level of transparency. 85. A minimum level of information includes data on total bid and accepted volume; cut-off price; and minimum, average, and maximum prices (yields) of accepted bids. 86. Pre-trade transparency refers to information about prices and volumes quoted by market participants (typically by primary dealers and/ or market makers), related to trades can be executed in case of firm prices. Post-trade transparency refers to information about the prices and volumes of trades that have already taken place. 87. Four of the eight countries with a green score belong to the Eastern Caribbean Currency Union (Regional Government Securities Market; RGSM), where the secondary market trading of the government securities takes place on the Eastern Caribbean Securities Exchange (ECSE), and individual trades are published on the same day; consequently, this is assigned a green score. However, secondary market trades are still very infrequent, and there are often weeks without a single trade in government securities. 58 Debt Transparency in Developing Economies RECOMMENDATIONS Transparency in domestic borrowing enhances domestic market development and AND CONCLUSIONS supports smooth implementation of the debt management strategy. Transparency and predictability of the borrowing requirements and the issuance are crucial preconditions for the institutional investors to engage in the government bond market and for attracting non-resident investors (Velandia and Secunho, 2021). The debt-management strategy, the annual borrowing plan, and the auction calendar, combined with timely and detailed announcements of individual transactions (auctions), are standard tools to foster transparency (DeMPA,2021). Based on the findings of this report, the following policy actions are recommended to improve transparency in domestic debt issuances: • Market-based issuing mechanisms should be regarded as the default option for a transparent primary market. To ensure that instruments are priced in line with the fundamentals and the market conditions, the government issuer must be a price taker on the primary market rather than a price maker. Market-based pricing through competitive auctions is the basis for extending maturities of debt, establishing a reliable yield curve and developing the secondary market. If non-market based domestic instruments are contracted, a full disclosure of their all-in cost should be ensured. • Auction rules should be transparent, clear, and consistently applied. It is vital that the rules and procedures for tenders are available to all eligible investors. The eligibility criteria to participate directly or indirectly in the auctions, the bidding rules, the applied auction method, the availability of non-competitive bidding, the allocation method for competitive and non-competitive bids, the treatment of the possible outliers (off-market bids), and whether and under what conditions, and how, the auction size can be changed are all necessary pieces of information for investors to be able to formulate the best investment and auction strategy. • Publication of the annual borrowing plan with volumes indicated for the domestic borrowing through marketable instruments is good practice and promotes transparency. The annual borrowing plan should include and disclose the gross and net annual borrowing need for the government, broken down between local and foreign currency instruments. Within the local currency instruments, an indication of the volume of gross and net domestic marketable borrowing can help investors to plan their investments for the fiscal-or calendar year. • The announcement of the auction dates and instruments at the beginning of the year, and the rule-based sequence of auctions enhance the transparency of the issuances. Many countries publish the auction dates for treasury bills and government bonds one year in advance, prior to the beginning of the calendar year. Additional considerations can further strengthen the transparent domestic market operation, such as: i) assigning a specific auction day to the treasury bill and government bond auctions each week; and/or ii) designating a specific week in the month for the auction of particular instruments. 59 Debt Transparency in Developing Economies • An issuance plan that is made publicly available in advance, at least on a monthly basis, further strengthens transparency. The ultimate goal is to publish full details, including the offered amount prior to the auction, so that investors can formulate their bidding strategy accordingly. The plan can be refined to include at least: (i) the auction dates; (ii) the type of instruments (fixed, floating or index-linked, and tenor); and (iii) the indicative gross issuance aggregated at the level of treasury bills and government bonds, to be fine-tuned as the date of the auctions get close. • Predictable implementation of the issuance calendar is crucial for strengthening investor confidence. Frequent, unpredictable, and/or unreasonable cancellations may deter investors in the long run since it can demonstrate that the issuer does not accept the market price. Frequent and unjustified partial rejections of bids should also be discouraged as they undermine price discovery. In the event of canceled auctions or reduced issuance, clear communication is key: DMOs should clearly explain the motivation/reasons behind the cancellation or partial rejections of the otherwise competitive bids to dispel any doubts that may arise in such circumstances. • Full and equal access to the auction results is another critical element of efficient primary market operation. The issuer should disclose, as soon as possible after the cut-off time,88 a sufficiently wide range of information on the auction that enables investors to assess the real demand and efficiency of the price discovery on the auction. The minimum content should include the aggregate volume of the bids, the accepted amount, the cut-off, and the average price and yield. The number of bids submitted and accepted can reveal additional information on the concentration of the auction. Moreover, the issuer should establish policies that ensure that the auction results are disclosed at the same time on each publication platform (for example, primary market auction system, vendor platforms, issuer and/or central bank website, etc.). The practice of informing primary dealers or auction participants first and the rest of the market later does not support the desired creation of a level playing field for investors. • In the absence of well-established reporting regimes,89 particularly in less developed markets, national authorities should consider taking the leading role in collecting, aggregating, and publishing post-trade data. An electronic registry and central security depository (CSD) can easily extract information on transaction activity. This information can be published daily either by the CSD or the authorities (Central Bank or DMO). Regular publication of trading volumes, values, and rates helps investors to assess the constantly changing underlying demand for government securities and also helps to understand yield curve response to market conditions. 88. For example, according to the AFME European Primary Dealers Handbook (2019/2020), most European Union DMOs publish the auction results within a few minutes of the cut-off time, but not later than 30 minutes. 89. Post-trade transparency regimes have developed significantly across the world since the global financial crisis. Strict regulations have been developed across jurisdictions to promote the transparency of the secondary market. The International Capital Market Association (ICMA) has been collecting and publishing pre- and post-trade reporting obligations across multiple jurisdictions from Europe, the Americas and Asia-Pacific. The ICMA Bond Market Transparency Directory provides a consolidated view to compare the regulatory rules and sound practice guidance on bond trade reporting transparency regimes. 60 Debt Transparency in Developing Economies 3.2 EXTERNAL DEBT External bonds and loans differ in terms of data disclosure requirements. As TO PRIVATE CREDITORS90 discussed in Chapter 1, Eurobonds have surpassed loans in the composition of LIDCs’ external portfolios (Figure 1.2). Bonds tend to be more transparent than loans because of stricter disclosure requirements from regulatory bodies, and continuous trading on international platforms. However, there are several factors that affect the level of disclosure of bonds, including the regulation governing the registration and distribution of the security;91 and the relevant law determining the type of information to be provided in the prospectus.92 Common challenges to transparency in connection with bonds include the non-availability of the legal documentation that underlies their issuance (e.g. trust deeds/ indentures, agency agreements) and the limited availability of prospectuses except via subscription to proprietary/commercial databases. Commercial loans contracted with private creditors, typically financial institutions, are more prone to misreporting or nondisclosure. Most loans are typically based on standard-form documents prepared by the Loan Market Association (LMA) and the Loan Syndications and Trading Association (LSTA), frequently following London/ New York jurisdictions/ governing laws. The debtor nondisclosure obligations observed in these model forms are narrowly drawn, usually limited to price-sensitive information. The LMA and LSTA templates impose more robust non-disclosure obligations on lenders, as they may obtain confidential business information in the course of their credit assessment of the borrower before issuing a loan. (Gelpern et al., 2021). Yet, ad-hoc confidentiality clauses may prevent disclosure of a loan (in direct and/or indirect reporting sources) or the publication of its terms. As a result, the risk of partial reporting or even non-disclosure is higher in external loans than Eurobonds, whose main financial features are confirmed in real time by the main trading platforms. A higher standardization of loan contracts can help reduce this gap (see box 3.1). 90. Given the focus on operational and legal features of the related debt operations, in this chapter, external debt is defined as debt issued/contracted in foreign countries, under the jurisdiction of a foreign court. ‘Eurobond’ is applied to all international commercial financing through bond issuances. 91. Among Eurobonds, those issued outside the US market (regulation S) do not require the full US-SEC disclosure criteria; securities under Rule 144A follow a stricter set of disclosure criteria but trading in these instruments is limited to qualified US institutional investors. 92. For instance, Eurobonds sold to investors in Europe are subject to the EU Prospectus Directive. The Prospectus Directive requires the inclusion of (i) a “Summary” section; (ii) risk factors; and (iii) other standard information about the issuer, including information about their history and current affairs, government and politics, economy (including, GDP, balance of payments, external sector, monetary policy, fiscal position and debt). In addition to format, the EU Directive regulates the dissemination of such prospectuses (Van der Wansem et al, 2019). 61 Debt Transparency in Developing Economies Box 3.1 Benefits of Loan Contract Standardization Governments may benefit from standardizing critically important provisions or key elements of their contracts. Greater standardization facilitates the signature of contracts that are legally sound and consistent with the PDMLF. Contract standardization is particularly important for countries with low institutional capacity, as it helps reduce the legal and operational risks stemming from possible contradictions or inconsistencies. Moreover, if not carefully prepared, negotiated, and implemented, bespoke terms and conditions in any transaction may prove overly burdensome for the borrower to actually implement, and may also cause inadvertent violation of the borrower’s obligations to other lenders as specified in other/existing debt contracts. Although not currently common practice among LIDCs, requirements about standardization of terms in PDMLF may help promote transparency. For example, Ghana’s PFM Law of 2016 (Article 56) and Nigeria’s Debt Management Office Act (Article 27) call for approval by Parliament of such standard terms and conditions. Standardization may include adaptations for contractual provisions related to standard covenants, such as: confidentiality, events of defaults, dispute resolution mechanisms, waivers of sovereign immunity, choice of law, governing law, and provisions related to debt standstills and moratoria or the restructuring of contract terms (Gelpern et al., 2017). Standardization is also beneficial during the loan negotiation phase. Provided that the proposed standardization is not rigid or unrealistic (in light of the relevant debt instrument, jurisdiction, commercial context or market), it can help finalize negotiations in a timely manner. In some countries, the use of a standard loan template could facilitate the approval process at the national level. For instance, in Nigeria and Ghana, an external loan or guarantee transaction is exempt from further approval by the National Assembly if the debt agreement conforms with the terms and conditions approved by the National Assembly. Guidance for staff that help negotiate for borrowers is also useful as it can elaborate on important elements of specific standard clauses for the borrower and thereby help inform negotiation strategies. The different level of disclosure between Eurobonds and loans has a significant impact in the case of debt restructuring. Unlike Eurobonds, information about the restructuring of bilateral and syndicated loans are not necessarily disclosed and may only enter the public domain months later when the new terms are reflected in direct/ indirect reports. This underscores the importance of involving private creditors in the recently launched OECD data collection initiative or in the project for an international loan repository described in Annex 2. While these initiatives may facilitate the availability of financial terms, the identification of final creditors remains a challenge. In the case of bonds, DMOs often are not in possession of this information, particulary on external issuances,93 as they may not have access to central securities depositories’ (CSDs’) data on bondholder identities. Therefore, DMOs tend to rely on aggregated information provided by the banks that arrange the deals. Bloomberg has some information on “current” bondholders based on surveys with large investment banks and global custodians, but the coverage for LIDCs is limited. The lack of visibility on the final creditor is not a prerogative of bonds, since 93. The identification of final creditors is less of problem for domestically-issued securities, as domestic markets tend to be more “captive”, national banks and pension funds typically adopt a “buy and hold approach”, and with limited secondary markets transactions, that can be tracked by Central Banks or local CSDs. 62 Debt Transparency in Developing Economies there is also a secondary market for loans, which is less regulated and transparent than the market for securities. In fact, depending on the terms and governing law, a loan creditor may decide to transfer its exposure to other investors without being under any obligation to amend the original agreement or inform the borrower about the transaction. Having detailed and timely understanding of the final creditors is important for implementing debt strategies and critical for debt restructuring. The lack of visibility on the secondary market with respect to creditor base composition may limit the analysis of refinancing risk as well as the opportunity for liability management transactions. The absence of these data is far more detrimental in the case of restructuring, as countries need to launch a call for data to reach out to final investors, which delays the negotiation process. However, the publication of detailed statistics on the composition of the investor base for tradable debt is not common practice in LIDCs. A general breakdown by investor category is presented in some more mature markets. This lack of granular information may respond to legal, operational or strategic reasons. Firstly, this information can be protected by local privacy regulations, legislation or by an arrangement with the CSD that would keep real-time data of the current holder, but would only share data by investor type. Secondly, investors may need to be pulled together to facilitate cross-border transactions, possibly reducing borrowing cost. This is the case for the International Central Security Depository (ICSD) (e.g. Euroclear and Clearstream), that serves as an important avenue for sovereign issuers to attract non-resident investors in LIDCs (Velandia and Secunho, 2021). The bridge between local and international CSDs that enables non-resident investors to trade local bonds without opening a domestic account is underpinned by the recognition of so-called “omnibus” accounts. Using “ominibus accounts”, foreign investors hold their securities in the name of the ICSD and neither the issuer nor the the local CSD have access to information related to the final holder. Box 3.2 Reducing the Information Asymmetry: The Role of Credit Rating Agencies A prerequisite of international debt issuance or loan is typically a credit rating from a credit-rating agency (CRAs). The CRAs’ goal is to produce an independent and objective assessment of the creditworthiness of a debt issuer using methodologies to ensure cross-country comparability. For that, they need to collect and process a considerable amount of macro and fiscal data, including on debt stock, composition, and debt service. Debt statistics are requested from the borrower in specific templates, typically at the general government level. Gaps or missing information are usually filled in by referring to indirect reporting from the IMF/World Bank or other international source. Indirect data sources play a central role, especially if the rating is “unsolicited” (i.e., assigned without being requested by the issuer). The general public may perceive this data analysis by CRAs as involving a certain degree of due diligence on the country’s data quality and comprehensiveness. CRAs need a sufficiently updated and comprehensive set of data to provide an informed assessment and may withdraw the rating in the absence of data.94 However, assessing the completeness or quality 94. For instance, Fitch withdrew its rating to Benin in 2012 and to Suriname in 2020 as insufficient public information was available to assign “unsolicited” ratings. 63 Debt Transparency in Developing Economies of the data directly received (or obtained through indirect sources) is beyond their mandate. As a result, rating agencies use relevant data collected, without providing details on the statistical collection and compilation methods whose heterogeneity, as described in Chapter 2 of the report, may affect their quality and ultimately compromise country cross-comparability. Transparency - defined as the availability of data - is factored in by CRAs. The methodology applied by the three main CRAs (S&P, Fitch, and Moody’s) contains clear criteria regarding the impact of sovereign data transparency on the rating assessment. S&P’s sovereign rating methodology, for instance, uses five assessments to determine its indicative rating.95 Of these, the level of sovereign data transparency directly impacts the institutional assessment, which evaluates the quality and consistency of all data used by the agency. S&P may also apply adjustment factors related to the transparency of statistics, these factors are important for the economic96 and external assessment.97 (Figure B3.2.1). Figure B3.2.1 Key Areas Affected by Transparency in S&P`s Sovereign Rating Methodology Institutional Economic External Fiscal Monetary Assessment Assessment Assessment Assessment Assessment Primary Primary Primary Primary Primary Factors Factors Factors Factors Factors + + + + + Adjustment Adjustment Adjustment Adjustment Adjustment Factors Factors Factors Factors Factors Assessments with factors directly related to transparency Source: S&P Sovereign Rating Methodology. It is possible to estimate the potential impact on the indicative rating from material data revisions and the suppression of data and information flows. For instance, ceteris paribus, a country rated BBB- (lowest level of investment grade level) would be downgraded to BB+ in the case of “material gaps in sovereign data or reporting delays”98 affecting its institutional assessment. Similarly, Fitch’s and Moody’s methodologies also take the level of transparency into account, with explicit reference to debt data: 94. For instance, Fitch withdrew its rating to Benin in 2012 and to Suriname in 2020 as insufficient public information was available to assign “unsolicited” ratings. 95. The indicative rating comes from the combination of five key assessments: institutional, economic, external, monetary, and fiscal. After this indicative rating, the agency may also apply supplemental adjustment factors which could impact the final rating. 96. The primary indicator for economic assessment is GDP per capita in USD (for further details, see S&P`s Sovereign Rating Methodology). 97. Primary indicators for external assessment comprise the status of the country’s currency in international transactions, and its external liquidity and position, including external debt. 98. In this hypothetical case, the country does not score more than 5 on each key assessment (in S&P`s methodology each assessment has a six-point numerical scale from 1 (strongest) to 6 (weakest)) and does not score less than 2 on each key assessment. 64 Debt Transparency in Developing Economies “Additionally, in the event of data limitations that are potentially material to the rating outcome, Fitch will consider making an adjustment in the qualitative overlay (QO) within the relevant analytical pillar; e.g., if there is a lack of information on external assets and liabilities, a negative notching adjustment could be made to the External Finances section of the QO. If Fitch believes that this lack of information is so significant as to render any analysis insufficiently robust to support a rating or rating action, Fitch will not assign a rating, or will withdraw an existing rating.”99 “While accounting standards can be complex and evolve over time, leading to ex-post revisions of fiscal performance and debt levels, a track record of frequent and large revisions in past budget accounts would typically weigh negatively on our assessment of a sovereign’s fiscal effectiveness.”100 Rating agencies have made progress in the level of transparency of their methodology.101 Since 2019 S&P has published more details on adjustment factor assessments (usually more qualitative) in the sovereign reports, showing the main drivers behind those adjustments; Fitch has made an interactive sovereign rating model available (in an Excel file) in which it is possible to observe the exact weights and inputs considered for each country;102 Moody’s published a new version of its sovereign rating methodology increasing the transparency of its adjustment factors. RECOMMENDATIONS The use of a standardized key clauses in loan agreements should be promoted. Using AND CONCLUSIONS standardized provisions helps ensure adherence to the PDMLF and reduce the borrower’s exposure to legal and operational risks. Moreover, it can promote transparency to the extent that it excludes confidentiality clauses that require secrecy and unreasonable confidentiality clauses, as described in section 4.4. Provisions should be included in loan agreements to require secondary market transactions (e.g., participations, assignments, transfers) of loans or interests in loans to be communicated to the borrower. Since loan trading does not take place in organized markets, the borrowers should be informed about the transactions, as their size and pricing may affect the primary market and should be taken into account in strategy formulation. Updated, reliable information about government debt holders should be available to the DMO. Local and international CSDs can collect and disclose updated information about final debt holders based on their operational and fiduciary role. However, full disclosure at the final holders’ level may entail operational and legal constraints for the borrower, thus making disclosure by category (e.g., banks, mutual funds, pension funds, retail investors, non-resident) a more achievable and balanced objective. 99. Fitch’s Sovereign Rating Methodology. 100 Moody’s’ Sovereign Rating Methodology. 101. Addis Ababa Action Agenda of the Third International Conference on Financing for Development, outcome document (General Assembly resolution 69/313, annex). 102. The interactive sovereign rating model for each country is available on the Fitch Ratings-Issuer Portal. 65 Debt Transparency in Developing Economies In international commercial debt, CRAs play a key role in reducing the information asymmetry between sovereign issuers and their investors; however, they could contribute more to transparency than they currently do. There is room for further improvement in the granularity of information on the rationale for CRA’s downward adjustment as a result of data transparency. A revision of CRAs’ methodology ensuring a more detailed and systematic disclosure of the specific events that trigger adjustments would add clarity to the process and allow final users to appreciate possible “red flags” in their own approach to debt data transparency. CRA assessments could also benefit from an explicit disclosure of the source of debt data (i.e., direct/indirect). This would demonstrate full alignment with the debt definitions and coverage used in their methodologies and would indirectly promote stronger disclosure practice at the country level. 3.3 RESOURCE-BACKED Developing countries often face difficulties accessing large-scale financing to meet LOANS their development needs. Major constraints include the lack of a sufficiently deep domestic market and limited or costly access to international capital markets. In response to this challenge, a resource-linked financing model, referred to herein as resource- backed loans (RBLs),103 has become fairly popular. RBLs provide new opportunities for countries to access finance in exchange for, or collateralized by,104 future streams of income from the underlying commodities. RBLs can be beneficial to LIDCs under certain circumstances, including if there is full transparency of contractual terms (WB-IMF, 2020). Collateralized borrowing practices along the lines of RBLs go back at least a century,105 and became widely used across resource-rich developing countries during the commodity boom (NRGI, 2020). The COVID-19 pandemic has generated considerable focus on these types of transactions because they pose unique challenges for debt sustainability. First, they are tied to volatile commodity prices. Second, they tend to be embedded in legally and financially complex deals. Third, they are often omitted in existing direct/indirect reporting sources, thus becoming part of the “hidden debt”. The collateral arrangements in RBLs can be grouped into four categories: (i) Resource sales receivables are the most common collateral arrangements. These arrangements require relevant parties to agree upon specific amounts of natural resources (barrels per day, tons, etc.) to be sold to a designated buyer(s). The amount that the designated buyer is obligated to pay for the resources is paid to the benefit of the lender, either directly or into a deposit account owned by the borrower but over which the lender has rights to deduct funds in the event of a loan repayment failure. 103. As used in this report, RBLs refer to loans provided to a government or state-owned enterprise (SOE) where: (i) the repayment is either made directly in natural resources (i.e., in kind) or from a natural-resource-related future income stream; (ii) the repayment is guaranteed by a natural-resource-related income stream; or (iii) a natural resource asset serves as collateral. 104. As used in this section of the report, a debt instrument is “collateralized” when the creditor has rights over an asset or revenue stream whereby if the borrower defaults on its payment obligations the creditor can rely on the asset or revenue stream for debt repayment. In a legal sense, collateralization entails a borrower granting liens over specific existing assets or future receivables to a lender as security against repayment of the loan. 105. E.g., Peru’s guano-backed borrowing in mid-19th century (Vizcarra, 2009). 66 Debt Transparency in Developing Economies (ii) Resource sales pre-payments (SPP) are advances made in respect of purchases of resources, most commonly used by commodities traders who want to purchase natural resources. (iii) Resource development access ties the expected returns from granting mineral development rights to an investor to the repayment of a loan. They are sometimes labeled “resource-for- infrastructure” deals or “barter deals.” (iv) Direct resource collateral involves collateralizing undeveloped resources still in the ground, usually mineral or oil deposits. It is not easy to find information on RBLs. Based on publicly available information and focusing on Sub-Saharan Africa, Cust, Hwang, Mihalyi and Rivetti (2021) identified 30 RBLs signed between 2004 and 2018 in 11 African countries, mostly commodity- exporters.106 The analysis considers only loans for which the proceeds are used for purposes external to the revenue stream that is used for repayment or as collateral.107 To compile the database, the following sources have been used: the Johns Hopkins SAIS China-Africa Research Initiative’s (CARI) dataset on Chinese lending to Africa;108 government sources,109 including non-debt-related official documents and SOEs’ financial statements.110 The identified loans account for USD 46.5 billion committed, including loans only partially disbursed. The primary sources of RBLs are foreign state policy banks/development banks and private trading companies. RBLs are heavily concentrated in a few countries where they represent a substantial share of the relevant countries’ borrowings. The total new external debt contracted by Sub-Saharan African countries over the 2004-2018 period is USD 600 billion (source: WB’s IDS) of which USD 450 billion was disbursed. As a result, based on the information presented in the sample, RBLs represent approximately 8 percent of total new borrowing in Sub-Saharan Africa and between 10 percent and 30 percent of the median country’s total external public debt stock in the year following the RBLs’ signature. Figure 3.3 shows the size of RBL-committed amounts relative to the total external PPG debt disbursements, further confirming that when countries decide to take on RBLs, they typically do it in large volumes. 106. Based on WB-IM definitions, 3 fuel exporters: Angola, Chad, the Republic of Congo; 6 non-fuel exporters: DRC, Guinea, Niger, South Sudan, Sudan, and Zimbabwe; and 2 diversified exporters: Ghana and São Tomé and Príncipe. 107. In the typology of IMF-WB (2020a), these are labelled as “unrelated collateral” transactions. These types of loans may be considered substitutes to regular government borrowing. 108. http://www.sais-cari.org/data. 109. For instance, DRC debt reports contain details of the collateralized loans. South Sudan and Republic of Congo report intermittently on their debt outstanding with commodities traders. 110. While significant effort has been spent on ensuring the accuracy of the data, an important caveat to this analysis is that it relies on mixed direct/indirect sources that may not follow the same definitions and valuation methods as discussed in Chapter 1). 67 Debt Transparency in Developing Economies Figure 3.3 Angola Chad New Disbursements of 15,000M$ 1,500M$ External PPG Debt vs. RBL Committed Amounts 10,000M$ 1,000M$ 5,000M$ 500M$ 0M$ 0M$ 2000 2005 2010 2015 2020 2000 2005 2010 2015 2020 Congo, Democratic Republic Congo, Republic 3,000M$ 2,500M$ 2,000M$ 2,000M$ 1,500M$ 1,000M$ 1,000M$ 500M$ 0M$ 0M$ 2000 2005 2010 2015 2020 2000 2005 2010 2015 2020 Ghana Guinea 3,000M$ 750M$ 2,000M$ 500M$ 1,000M$ 250M$ 0M$ 0M$ 2000 2005 2010 2015 2020 2000 2005 2010 2015 2020 Niger Sao Tome & Principe 1,000M$ 40M$ 750M$ 30M$ 500M$ 20M$ 250M$ 10M$ 0M$ 0M$ 2000 2005 2010 2015 2020 2000 2005 2010 2015 2020 Public debt disbursement RBL agreement 68 Debt Transparency in Developing Economies Sudan Zimbabwe 3,000M$ 600M$ 2,000M$ 400M$ 1,000M$ 200M$ 0M$ 0M$ 2000 2005 2010 2015 2020 2000 2005 2010 2015 2020 Public debt disbursement RBL agreement Source: Cust, Hwang, Mihalyi and Rivetti 2021. Given the magnitudes observed, having access to the terms of RBLs is critical to monitoring debt sustainability. This is evident in the recent restructuring process in Angola, Chad, and the Republic of Congo, where the renegotiation of RBLs has been a key stumbling block. RBLs are distinctively opaque for five main reasons: • They tend to include stringent confidentiality clauses, which may be due to the political sensitivity of pledging natural resources or permitting certain ownership transfers; or a desire of the parties to conceal or play down the nature of collateralization. • RBLS are not systematically recognized and classified as debt by the debtor country (instrument coverage, see Chapter 2). In many cases, RBLs are treated more as advance payments to a supplier. • Countries relying on such borrowing methods tend to have weaker debt reporting practices. Seven out of ten RBL borrowing countries perform below the median of IDA countries on the WB Debt Reporting Heat Map (Cust, Hwang, Mihalyi and Rivetti, 2021). • RBLs are often contracted by SOE or SPV (sectoral coverage) which may not publish audited financial statements and do not necessarily provide data to the DMOs, which are the main sources of debt statistics in LIDCs (see Chapter 3). As a result, RBLs are often omitted in debt statistics.111 • Finally, existing public debt data collection by IMF/WB does not currently request countries to report on collateralization features. 111. Only 15 of the 30 loans analyzed by Cust, Hwang, Mihalyi and Rivetti (2021) could be identified on the World Bank’s Debtor Reporting System (DRS). 69 Debt Transparency in Developing Economies The lack of transparency on RBLs has a number of implications • First, these deals are often highly complex, and their approval requires specialized knowledge to estimate the all-in-cost and legal risks that they entail. In fact, collateralized deals often consist of different related agreements, and may involve additional costs such as export credit premia, remuneration of financial intermediaries, legal fees, and non-monetary “costs”, such as lender step-in rights, and other lender controls over the management and disposal of the secured asset (WB-IMF, 2020). Even without considering these additional fees, the analysis of RBLs shows that collateralization does not always bring down the cost compared to uncollateralized lending. Although a few RBLs extended by bilateral lenders have relatively low rates, most RBLs reviewed in the sample come with rates above the median of other financing sources with comparable features (e.g., maturity, lender category, currency). This may be because RBLs are often contracted in scenarios where the borrower has limited market access or limited funding sources. That makes transparency around RBLs even more critical for debt sustainability. From a legal perspective, borrowers need to consider whether the structure or terms of an RBL would cause non-compliance with contractual undertakings already given in favor of other lenders by the sovereign or SOE borrower (Box 3.3), as well as domestic legal constraints on resource ownership. • Second, given the complexity of these transactions, there is a significant risk of poor negotiation outcomes, given the asymmetry in capacity between borrowers and lenders, their different bargaining power as well as the limited competition among providers of such loans (few lenders offer them). This risk is higher when the outcome is subject to limited public awareness and there is therefore little scrutiny and accountability by the borrowing government. • Third, RBLs may use structures that are not fully reflected in standard national debt and confer undisclosed seniority or payment advantages to some lenders. This may undermine the soundness of lending allocations based on debt sustainability analysis, can cause mistrust among creditors and complicate the finalization of arrangements in sovereign debt restructurings. 70 Debt Transparency in Developing Economies Box 3.3 RBLs, Collateral, Collateral-Like Features, and Contractual Covenants The key aim of a negative pledge provision is to prevent subordination of a lender by precluding or limiting the ability of the borrower to grant collateral in favor of other lenders—unless the borrower secures the lender in question, provides equivalent security or otherwise obtains its consent. Negative-pledge undertakings are used in loans made by a wide range of lenders, including multilateral development banks, private or commercial lenders, and state-owned policy banks. They are also usually included in the terms and conditions of sovereign bonds. Why is a negative-pledge clause important to a lender? When specific borrower assets are secured in favor of specific lenders, it leaves fewer borrower assets from which unsecured lenders can be repaid in the event of a default. Some negative- pledge clauses have an expanded scope in that they apply not only to transactions involving the creation of legal security or a lien which is legally enforceable, but also to transactions with collateral arrangements that offer lenders commercial security even if they do not amount to true, legally enforceable security in an insolvency or default scenario. Such transactions may be considered to have an equivalent (economic) effect to legal security, albeit not the same legal effect or remedies. Under these arrangements, contractual rights and transaction structuring techniques are used to give the relevant lenders advantages vis-à- vis other lenders, despite technically falling short of legal security. These contractual features are often referred to as “quasi- collateral”, “quasi security,” or “collateral-like features”. Given the possible variety in scope of negative-pledge undertakings,112 the question of whether a particular transaction violates a negative-pledge undertaking will depend on the language used in the negative pledge clause as well as the structure of the transaction and the language of the legal documentation. Prior to entering into an RBL, a borrower needs to consider whether the proposed transaction would put it in breach of any negative-pledge undertakings (or other contractual covenants) given to other lenders. In addition, even where an RBL structure does not violate any of the borrower’s existing contractual undertakings, sovereign and SOE borrowers in RBLs should consider the extent to which a collateralized transaction structure effectively ties up the borrower’s assets - for example if resource sale proceeds are directed and held in specific accounts where the borrower’s ability to withdraw funds is subject to contractual limitations tied to the loan. These limitations may significantly counterbalance the attraction of any lower interest rate offered in connection with a collateralized loan. In some cases (such as borrowings by SOEs), loan contracts also include “permitted lien” provisions through which the lender explicitly permits the borrower to enter into certain collateralized transactions. Transaction types are usually listed and specifically described with clear parameters. A lender who permits certain types of liens or secured transactions upfront, may stipulate maximum monetary values of a permitted loan; may limit permitted liens to loans with a specific purpose; or specify other limitations that require the borrower to obtain the lender’s consent or to equally and ratably secure such a lender. The lender’s motivation for permitting, upfront, the borrower to give such liens or security in favor of other lenders over the life of the loan will vary, but may include: (i) recognizing that such liens are usual in the borrower’s business context or sector; (ii) recognizing that such liens are likely to be given in connection with specific investment types that enhance the borrower’s overall business and value; and (iii) a desire by the parties to limit the amount of time and transaction costs expended over the life of a loan in obtaining the lender’s consent. In some instances, the number and scope of permitted liens or transaction types is so extensive that it could be argued that the extensive number of carve-outs impairs the purpose of the permitted lien (and negative pledge) language in the first place. 112. The breadth of a negative-pledge and undertaking initiatives more than just the definitions of concepts such as “lien” or “security”. Exceptions are often contemplated within the negative-pledge clause itself. In the case of the WB’s NPCs, the following exceptions apply: (i) the giving of liens or security by the borrower in connection with debt that has a maturity of 12 months or less; and (ii) transactions that entail a lien or security created on property at the time of purchase of such property, where the lien is created solely to secure payment of the purchase price or to secure the repayment of debt incurred for the purpose of purchasing the property. In addition, depending on the context of the debt in question, a lender may limit the scope of the negative pledge and undertaking to external debt incurred by the borrower. 71 Debt Transparency in Developing Economies Compliance with a negative-pledge clause and permitted lien provisions is the borrower’s responsibility. The limitations of these clauses, although often viewed as a burden, can potentially function as an ally to assist sovereign and SOE borrowers by giving them a focus in seeking to minimize any legal security or quasi-security proposed by lenders. It is not the responsibility of lenders to monitor or check compliance by a borrower with the lender’s negative pledge and permitted lien provisions. However, if information about a sovereign’s or SOE’s borrowing activities is available, a lender may decide to conduct their own review - although possibly facing the same disclosure or confidentiality challenges described in Chapter 2. RECOMMENDATIONS Public debt operations that leverage natural resources as credit enhancement may AND CONCLUSIONS represent an opportunity to gain better financial terms and pricing, or to gain access to finance that would not otherwise be available. However, the net benefit of such operations depends on a country’s debt sustainability, the use of proceeds, the soundness of their legal and financial terms, and compliance with applicable negative pledge clauses (NPCs). On the downside, these contractual arrangements inhibit the ability of a sovereign or SOE borrower to freely direct their own cashflows, potentially impairing the borrower’s ability to respond to economic shocks and commodity price volatility. The COVID-19 pandemic has exposed some Sub-Saharan African countries to these risks, as they face narrow fiscal space and reduced control over scarce resources tied up in highly structured loan arrangements. Full transparency is key for beneficial outcome from RBLs. Collateralized transactions need to be properly disclosed to ensure a timely and accurate assessment of their impact on debt sustainability, thus allowing borrowers and creditors to correctly assess risks and borrow (lend) sustainably. A full disclosure of RBLs’ terms would also limit the prospect of corruption or poorly negotiated deals (e.g., the effective “privatization” of assets on highly advantageous terms to the lender). Strengthening disclosure requirements is important since RBLs tend to be large, highly complex and contracted in scenarios where the borrower has limited funding sources. The borrower needs to provide a minimum level of information such as: memorandum items in national debt statistics, including asset, amount, and type of security. In parallel, IFIs can capture collateralization features in indirect reporting databases (see section 2.2) and can promote specific financial and legal technical assistance to better equip and empower borrowers in negotiations with lenders. Relative to standard loans, RBLs also require stricter analytical processes in the approval phase to promote better development outcomes. This includes the following steps: (i) a careful assessment of how sustainability might be impacted; (ii) a check that the proposed terms and conditions account fairly for the value of the security given; (iii) a check that the legal and technical dimensions of the proposed structure are fully taken into account; and (iv) careful assessment of how granting collateral might impact other financing, in the context of the country’s debt management strategy (WB-IMF, 2020). 72 Debt Transparency in Developing Economies 3.4 CENTRAL BANK REPOS In most countries, the central bank is not consolidated with government accounts AND SWAP OPERATIONS for public sector reporting. As described in Chapter 2, the CB is beyond the sectoral coverage of LIDC’s public debt statistics, its asset and liabilities are typically only reported in its balance sheet without consolidation with other sectors. In national debt definitions, the exclusion of CB debt also reflects the principle of separating debt instruments contracted for fiscal use from those contracted for monetary purposes. For this reason, securities issued by the CB in its role of fiscal agent of the government are included in the debt stock, while those issued solely for monetary or liquidity purposes are excluded. This is also the criterion followed by the LIC-DSA.113 However, establishing the exact purpose of a CB instrument (fiscal vs. monetary), is not always straightforward. This difference is evident when countries use different policy instruments for fiscal and monetary purposes or, in the case that the same instruments are used (e.g., T-bills), the respective amounts are clearly separated in the reporting and accounting (DeMPA, 2021). Some key monetary policy instruments - such as Repurchase Agreement (“repos”), Foreign Currency Deposits or Swap Lines (FXSLs) - have the potential to generate external borrowing and have been used to this end by some CBs in emerging markets and LIDCs. As such, they represent material risk for the government and warrant a risk-based inclusion in DSA. The non-conventional use of CB instruments raises transparency concerns. Repos, FX Deposits and FXSLs are key features of international financial architecture. They are important tools for CBs to ensure financial stability by controlling market liquidity and bolstering reserves. Their use has grown significantly following the 2008 global financial crisis, the COVID-19 crisis has further underscored their importance. However, their presence is not always clearly identified in the CB’s balance sheet, but rather lumped in the broad category of international reserve and FX liability. They are also not captured by existing debt databases, as they are considered to be short-term instruments114 or assumed to be for monetary/liquidity purposes. As a result, repos and FXSLs may generate “debt surprises”115 and deserve further analysis. 113. “Central bank debt issuance or foreign exchange swaps for the purposes of monetary policy or reserves management are excluded from external public debt”. (WB, 2018). 114. In the case of WB DRS, central bank deposits and FXSLs are understood to be short-term and, therefore outside the reporting parameters. 115. In 2020, for instance, the Reserve Bank of Malawi, could not reverse a FXSL open position with a local bank and had to convert it into a medium-term forex facility of US$450 million (6 percent of the GDP). 73 Debt Transparency in Developing Economies Central Bank Repos for External Borrowing A repurchase agreement (“repo”) is the sale of securities to a counterpart in exchange for cash, under the agreement to repurchase the same or similar securities at an agreed price in the future. Repos are widely used in money markets including: (i) interbank lending, (ii) central bank open market operations with domestic commercial banks, and (iii) securities dealers to finance their inventories. Starting from the mid-90s, repos have been increasingly used in LIDCs as they are a safer, more flexible, and often cheaper source of funding than unsecured borrowing for market makers (WB-IMF, 2021). The “price differential” between the price at the start of the transaction and the price at the end of the transaction reflects the interest rate, whilst the “haircut” is the difference between the market value of the securities and the amount of cash lent against their transfer. The haircut depends on a number of variables such as maturity, quality, scarcity value, and price volatility of the underlying collateral; terms of the repo; and creditworthiness of the counterpart (Figure 3.4). Figure 3.4 Trade Initiation Example of a Standard Central Bank Repo Transaction Securities (101) Central Commercial Bank Bank Cash (100) Trade Termination Securities (101) Central Commercial Bank Bank Cash (100) + interest Source: Authors’ elaboration. 74 Debt Transparency in Developing Economies The economic nature of a repo is that of a collateralized loan. The market arrangements for repos, including the payments of margin, the ability to substitute securities, and the retention of market risk by the security provider, support the view that repos should be classified as loans, with the security remaining on the balance sheet of the security provider. However, from a legal perspective, a repo is a true sale/purchase of assets for a purchase price, with an agreement to re-purchase at a price differential; this is different to a loan that bears interest. As a result of this, the “legal owner” of the securities (i.e., the security receiver) in repos may differ from the “economic owner” for statistical purposes (i.e., the security provider). In the absence of adequate disclosure of the collateralization details, this difference may generate severe information asymmetries, particularly when the repo is overcollateralized and the securities used are not marketable. Overcollateralization of repos that utilize the seller’s own securities can lower the cost of borrowing by providing credit protection in case of default. Box 3.4 shows examples of repos that, in contrast to normal repos (which use third party typically high-grade securities), utilized the countries’ own sovereign bonds as collateral for their borrowing with large haircuts. These repos—signed when the borrowers were experiencing difficult financial conditions—gave creditors the right to claim, in the case of default, a larger amount than was lent (i.e., paid as the initial purchase price), thus de facto diluting the rights of other creditors. Such transactions would only be cost-effective if the potential impact of the collateral in the case of a default (e.g., external securities increasing in value) is not observed by other investors. Otherwise, theoretically, their inclusion in the reported debt portfolio should trigger an increase in the cost of future un-collateralized debt. Box 3.4 Over-Collateralized REPOs in Egypt (2016) and Ecuador (2018) In 2016, Egypt’s central bank entered into a one-year repo to increase its foreign reserves. As Egypt did not hold the Eurobonds to secure the transactions, the Egyptian Ministry of Finance issued non-marketable bonds to collateralize the repo. USD 4 billion bonds were transferred directly to the CB and therefore not included in the external debt stock. These bonds were used to back a USD 2 billion loan from HSBC (haircut: 50 percent). The over-collateralization and cash-margining managed to reduce the cost of borrowing, relative to Egypt’s likely cost of borrowing attainable through a Eurobond issuance. The operation was replicated in 2018, when the CB of Egypt agreed to enter into a new repo with a consortium of international banks for a total amount of USD 3.8 billion and a final maturity of 4.5 years and an average life of 3 years. The repo was extended in November 2020 by 1.5 years. A similar operation was conducted by the Government of Ecuador in 2018. Ecuador also over-collateralized a series of four-year repos with Goldman Sachs and Credit Suisse by using ad hoc- issued bonds to cover funding gaps during difficult financial conditions. The amount of bonds pledged was USD 2.4 billion against USD 1 billion received (haircut: 58.3 percent). Although the repo interest rates were 30 percent below Eurobond market rates at the time, Ecuador saw the true costs of repo increasing due to margin calls linked to the mark-to-market collateral. In May 2020 Ecuador agreed to repay the full repo, which was no longer included in the subsequent restructuring. 75 Debt Transparency in Developing Economies Foreign Currency Swap Lines (FXSLs) FXSLs provide foreign exchange (FX) liquidity to the receiving CBs. CB Foreign Swap Lines (FXSLs) are bilateral arrangements between two CBs to exchange a certain amount of their currencies at the spot exchange rate, with the commitment to unwind the operation at a future date, at an agreed exchange rate (normally, the spot rate of the date of the original transaction).116 They are typically set up to enable a CB to supply its domestic banking sector with liquidity in currencies other than the domestic one (i.e., the CB would lend the foreign currency on to institutions in its jurisdictions, on its own terms and at its own risk) or as a way to temporarily increase international reserves. Figure 3.5 summarizes how FXSLs typically work. Figure 3.5 Credit limit, often time bound, and How CB FXSLs Work amount based on credit risk Central Bank A Central Bank B Drawing on the FXSL CB A currency provided to CB B Central Bank A Central Bank B CB B simultaneously deposits an equivalent amount in its own currency in favor of CB A Repayment CB A currency repaid to CB A Central Bank A Central Bank B Deposits are liquidated. The exchange rate is the same as when the funds were drawn. CB B also pays interest. Source: Authors’ elaboration 116. Depending on the contract, interest may be paid by one CB to the other. Also, margin calls may be required to account for exchange rate fluctuations, amounts in respect of which are to be credited to the partner’s deposit account if certain thresholds are crossed or at certain frequency. 76 Debt Transparency in Developing Economies The receiving CBs are typically selected based on high credit quality and close trade relationships. The providing CB is exposed to credit risk should the receiving CB not honor its agreement to sell the FX at the agreed exchange rate. Therefore, the CB offering the FXSL will consider both the credit quality of the receiving CB and the potential spill-over to the domestic economic activity of not offering the FXSL. However, as in the case of bilateral lending, political reasons may be more pertinent (Box 3.5) Box 3.5 Current Use of FXSLs The use of FXSL has grown to be an important part of the global financial safety net in the past two decades. Perks et al. (2021) document three bilateral FXSLs in 2000 growing to 25 in 2010 after the global financial crisis and to 91 in 2020 after the onset of COVID-19 pandemic. The swap line limit amounts in the same period grew from US$6 billion in 2000 to US$500 billion in 2010 and US$1.9 trillion in 2020.117 The role of FXSLs is key when funding markets in one currency deteriorate and it becomes difficult or expensive for banks outside that currency area to fund their FX assets. During the 2008 financial crisis, for example, funding markets dried up because of an extreme aversion to risk. Under these circumstances, it became difficult for euro area banks to obtain USD to fund their USD-denominated assets. To prevent disruptions, such as banks having to sell assets abruptly and thus provoking extreme price movements, the European Central Bank (ECB) and the United States Federal Reserve (Fed) set up a FXSL, allowing the ECB to provide USD to banks located in the euro area. Similarly, in the immediate aftermath of the COVID-19 crisis, an international dollar shortage led to a deviation from covered interest rate parity among major currencies, which materialized as a cross-currency basis spread.118 The Fed provided USD funding to other major CBs to help relieve the USD funding pressure.119 Six major CBs have mutual standing FXSL arrangements although only few FXSLs have been activated, they are usually only valid for a limited amount of time.120 The US Federal Reserve (US Fed), the European CB (ECB), the Bank of Japan, the Bank of Canada, the Bank of England, and the Swiss National Bank each have FXSL arrangements with each other. The currencies of these CBs account for the large majority of international finance and trade settlement. Given the role of the USD in international trade and finance, the Fed has been the most important provider of FXSLs. USD liquidity swaps from the Federal Reserve Bank have maturities ranging from overnight to three months. The US Fed publishes daily outstanding volumes of outstanding amounts by counterparty thus providing transparency on its transactions. In response to the COVID-19 pandemic, additional lines were established, and the outstanding amount reached almost USD450 billion, this has since declined to just over half a billion. The other main player in FXSL - both in terms of notional limits and number of beneficiaries – is the People’s Bank of China (PBoC). PBoC’s FXSL involve several emerging markets121 and tend to have much longer maturities. 117. Some FXSLs do not have limits. In those cases, the amounts are estimated based on past usage – see Perks et al (2021). 118. The cross-currency basis is the interest rate difference between borrowing one currency and borrowing in another currency and swapping it to the first currency. If the covered interest rate parity holds, this spread is 0. 119. See for example Bahaj and Reis (2018). 120. For instance, in March 2019 amid Brexit uncertainty, the BoE activated its Euro swap line with the ECB. 121. E.g., Argentina, Belarus, Mongolia, Nigeria, Pakistan, Sri Lanka, Suriname, and Ukraine 77 Debt Transparency in Developing Economies FXSL withdrawals add to gross international reserves and external debt. The establishment of the FXSL itself does not impact the balance sheet of the receiving CB until funds are withdrawn. Once funds are drawn under the FXSL, the FX leg adds to international reserves. However, because CBs are obligated to return the FX upon maturity, it has an equivalent reserve-related liability and thus net international reserves are unaffected. The deposit in local currency is an external liability of the CB and adds to its external debt. Undrawn amounts should be reported under contingent foreign exchange resources. A recent use of some FXSL appears to be motivated by covering a country’s funding needs rather than for liquidity reasons.122 When FXSL are systematically rolled over, the difference with long-term debt becomes less straightforward. Therefore, given the size and the potential impact of these operations on a country’s debt sustainability, full transparency around their use and conditions is key. Yet, accurate data on FXSLs’ LIDCs are not readily available. The borrower’s use of FXSLs is reflected in the CB’s balance sheet at face value, but its market value, which will fluctuate depending on the spot exchange rate, interest rates, and time to maturity at the moment of valuation, is not reported.123 Similarly, the lender’s leg of FXSLs is not usually reported or marked to market (PBoC, for instance, announces the FXSLs and their established limits but does not provide information on usage).124 Moreover, the external debt leg of the FXSLs is not systematically captured through indirect debt reporting (e.g., main IMF/WB debt databases). The FXSLs’ accounting treatment may create incentives for creditors to favor these instruments as opposed to standard bilateral loans. While keeping the same political and financial return of a loan, CB’s liabilities under FXSL are typically excluded from the standard restructuring perimeter in the case of comprehensive restructuring.125 This feature, together with the collateralization via local-currency CB deposit, may represent significant incentives for bilateral lenders to prefer this form of official assistance over more regular forms of lending. 122. Pakistan drew about USD 600 equivalent in May 2013 after a dip in reserves and pressure on its currency. Argentina drew about USD 2.7 billion equivalent in 2015 after losing access to international capital markets. Mongolia drew about USD 1.7 billion equivalent in 2015 while facing balance of payments pressures. Russia reportedly drew on its line in 2015 and 2016, and Ukraine did so in 2016. 123. Two options are possible: (i) Since the liability deposit account in domestic currency is fully indexed to a foreign currency, a valuation adjustment account linked to the former should be created and its carrying balance periodically adjusted to reflect the total amount of domestic currency needed to buy the foreign currency to be delivered, including any interest payment, or (ii) creation of a financial derivative, namely a forward contract (IMF, 2017). 124. McDowell (2019) used press releases and media reports to estimate the total size of the PBOC’s FXSL limits at almost USD500 million equivalent. 125. Various considerations may prompt a member to exclude the FXSL from restructuring, such as: a central government may not be able to seek a restructuring of CB liabilities under domestic law, including CB independence provisions; Including FXSLs in debt restructuring could impact CB’s credibility, potentially impairing its capacity to implement monetary and prudential policies; FXSLs help address financing gaps access and support macroeconomic stability in the face of shocks. 78 Debt Transparency in Developing Economies RECOMMENDATIONS Stronger focus should be placed in the statistical treatment of repos and swaps. AND CONCLUSIONS Collateralization features and financial derivatives are excluded from the standard debt presentation tables in LIDCs, despite the significant risks they pose. Given the level of discretion in how they are accounted for and reported in LIDCs, specific guidance is needed to ensure that they are fully reflected in debt statistics so that their use does affect other stakeholders. In parallel, data should be collected though indirect reporting database to facilitate DSAs and other analytical exercises. The non-standard use of CB repos and swaps demands heightened governance and transparency. The more complex the operation, the wider the borrower/lender capacity gap becomes. Therefore, specific TA and adequate risk management systems are needed to ensure that the borrowing countries appreciate the implications of these operations. As discussed in Chapter 2, this would also require an upgrade of the existing systems that should facilitate the approval and monitoring of these instruments by allowing a real-time pricing of financial collateral or derivatives. 79 Debt Transparency in Developing Economies REFERENCES Aldasoro, Iñaki, C. Cabanilla, P. Disyatat, T. Ehlers, P. McGuire and P. von Goetz. 2020. “CB swap lines and cross-border bank flows”, Bank for International Settlement Bulletin No. 34, 14 December 2020. BIS. Basel. Aldasoro, Iñaki, T. Ehlers, P. McGuire and P. von Goetz. 2020. “Global banks’ dollar funding needs and CB swap lines”, Bank for International Settlement Bulletin No. 27, 16 July 2020. BIS. Basel. Association for Financial Markets in Europe (2019/2020). 2020. “European Primary Dealers Handbook” Bahaj, Saleem and Reis, Ricardo (2018), “CB Swap Lines”, June 2018. BIS/ECB/IMF 2015. Handbook on Securities Statistics. International Monetary Fund, Washington, DC, Cust, J., Hwang, Mihalyi D., and Rivetti, D. 2021. Resource Backed Loans in SSA. 2021. The World Bank, Washington D.C. (forthcoming) Essl, S., K. Celik, S., P. Kirby, and A. Proite. 2019. Debt in Low-Income Countries: Evolution, Implications, and Remedies. World Bank Group Policy Research Working Paper 8794. World Bank. Washington, D.C. Fitch Ratings. 2021. “Sovereign Rating Criteria.” April 26, 2021. https://www. fitchratings.com/research/sovereigns/sovereign-rating-criteria-26-04-2021. Gelpern, A. 2017. About Government Debt… Who Knows. Capital Markets Law Journal, 13(3). Gelpern, A., S. Horn, S. Morris, B. Parks, and C. Christoph Trebesch. 2021. How China Lends: A Rare Look into 100 Debt Contracts with Foreign Governments. Peterson Institute for International Economics, Kiel Institute for the World Economy, Center for Global Development, and AidData at William & Mary. Williamsburg, VA. Wheeler, G. 2004. “Sound Practice in Government Debt Management” The World Bank. Washington D.C. Gruić, B. and P. Wooldridge. 2012. Enhancements to the BIS debt securities statistics. BIS Quarterly Review, December 2012. Heller, P. and Mihalyi, D. 2019. “Massive and Misunderstood: Data-Driven Insights into National Oil Companies.” Natural Resource Governance Institute. 2019. Horn, S., Reinhart, C. and Trebesch, C. 2019 China’s Overseas Lending. NBER Working Paper Series Working Paper 26050. Horn, S., Reinhart, C. and Trebesch, C. 2020. Coping with Disasters: Two Centuries of International Official Lending. Kiel Institute for the World Economy Working Paper 2157. 80 Debt Transparency in Developing Economies IMF. 2011. Public Sector Debt Statistics: Guide for Compilers and Users. International Monetary Fund, Washington, D.C. IMF. 2020. Review of the Debt Sustainability Framework for Market Access Countries. IMF Policy Paper, November 2020. IMF. Washington, D.C. IMF. 2020. The International Architecture for Resolving Sovereign Debt Involving Private-Sector Creditors —Recent Developments, Challenges, And Reform Options. IMF. Washington, D.C. IMF. 2003. “Assessing Public Sector Borrowing Collateralized on Future Flow Receivables.” IMF. Washington, D.C. IMF/WB. 2020. “Collateralized Transactions: Key Considerations for Public Lenders and Borrowers.” IMF/WB. 2018. G20 Note: Improving Public Debt Recording, Monitoring, and Reporting Capacity in Low- and Middle-Income Countries. June 14, 2018.The World Bank. Washington, D.C. IMF/WB. (2019). Debt Vulnerabilities in Emerging and Low-Income Economies. IMF, World Bank Development Committee Meeting, Washington, DC. International Capital Market Association. 2020. (ICMA “ICMA Bond Market Transparency Directory”) (January 2020) International Monetary Fund. 2017. “Recording of CB Swap Arrangements in Macroeconomic Statistics”, Statistics Department. IMF. Washington, D.C. McDowell, Daniel. 2019, “The (Ineffective) Statecraft of China’s Bilateral Swap Agreements”, Development and Change 50 (1): 122-143, 2019. Mihalyi, David, Aisha Adam and Jyhjong Hwang. 2020. “Resource-Backed Loans: Pitfalls and Potential”. Natural Resource Governance Institute. New York. Moody’s Investor Services. 2019. “Sovereign Ratings Methodology.” November 25, 2019. https://www.moodys.com/researchdocumentcontentpage.aspx?docid=PBC_1158631. Panetta, Schnabel. 2021. “The provision of euro liquidity through the ECB’s swap and repo operations”, ECB Blog 8/19/2021. Perks, M. and Yudong Rao, Jongsoon Shin, and Kiichi Tokouka. 2021. “Evolution of Bilateral Swap Lines“, IMF Working Paper WP/21/210, Washington D.C. Reinhart, C. and Trebesch, C. 2016. Sovereign Debt Relief and Its Aftermath. Journal of the European Economic Association. 14(1) p.215-251. 81 Debt Transparency in Developing Economies S&P Global Ratings. 2017. “Sovereign Ratings Methodology.” December 18, 2017. https://www.spratings.com/documents/20184/4432051/Sovereign+Rating+Methodolo gy/5f8c852c-108d-46d2-add1-4c20c3304725. The World Bank & International Monetary Fund. 2020. “G20 staff note on Local Currency Bond Market Development IMF, World Bank, Washington D.C. The World Bank & International Monetary Fund. 2021. “Guidance Note for Developing Government Local Currency Bond Markets” IMF, World Bank, Washington D.C. The World Bank. 2015. “Domestic Syndications” Background Note (May 2015), World Bank, Washington, D.C. Van Der Wansem, P. B. G., L. Jessen, & D. Rivetti. 2019. Issuing International Bonds. World Bank. Washington, D.C. “Velandia, Antonio; L. Secunho. 2021. How to Attract Non-Resident Investors to Local Currency Bonds: The Cases of Ukraine, Panama, Colombia, and Brazil. Equitable Growth, Finance and Institutions Insight. World Bank, Washington, DC Vizcarra, Catalina. 2009. “Guano, Credible Commitments and State Finances in Nineteenth Century Peru”. The Journal of Economic History. Vol. 69, No. 2. 2009. Wang, Hongying (2014). China and Sovereign Debt Restructuring. CIGI Paper no. 45 — September 2014 Zhitao, L., Z. Wenjie, and Y-W Cheung. 2016. “China’s Bilateral Swap Lines”, CESIfo Working Paper No. 5736 82 The Role of Public Debt Management Legal Frameworks in Transparency Debt Transparency in Developing Economies As discussed in Chapter 1, a key enabler of debt transparency is the presence of a conducive public debt management legal framework (PDMLF). A PDMLF will encompass a range of legislation and regulations126 to govern a country’s public debt management. A PDMLF can directly enhance transparency by establishing clear debt authorization provisions and requiring the disclosure of public debt information, regulating its content and frequency, and making it easily accessible to all stakeholders. An easily accessible and coherent PDMLF will enable public scrutiny of DM operations and enhance accountability. Publicly available DM laws and regulations allow interested parties to understand the decision-making process, the institutions and the rules that govern the debt-management operations of a country. A coherent framework of legislation and guidance documents will help eliminate gaps that could lead to legal uncertainty. Transparency and scrutiny of compliance with debt management operations with applicable laws can increase accountability of government officials and reduce the possible differences between de jure and de facto implementation. LIDCs have heterogeneous PDMLFs. Our analysis shows that LIDCs tend to have fairly clear and traceable provisions that identify who is authorized to borrow and manage the public debt portfolio. The legal frameworks, however, vary significantly in areas such as the debt disclosure and audit requirements, approval of non-standard instruments (e.g., involving collateral or collateral-like structures) or the consequences of noncompliance. 126. The term legislation will be used for primary legislation, the term regulation for the secondary legislation, and guidelines for lower-level operational guidelines. Primary legislation refers to the constitution, laws and decrees enacted by a legislature such as Parliament or Congress (for example, public debt management law, public financial management law, budget law). Secondary legislation refers to regulations referenced in the primary legislation that provide details for the implementation of a procedure and to regulations that are published by the government, a minister, or the DMO (for example, bylaws, communiques, directives, resolutions). Guidelines refer to documents prepared by a unit of the Ministry of Finance to further describe the internal steps for the implementation of the legislation and regulation. 84 Debt Transparency in Developing Economies This chapter highlights the essential elements of a transparent PDMLF and provides an overview of PDMLFs in LIDCs. Given the difficulties of collecting country-specific data on legislation and regulations pertaining to public sector entities, most findings and recommendations refer to central-government borrowing. However, the evolving public borrowing landscape described in Chapter 1 means that a robust PDMLF would need to contain rules and procedures to regulate the monitoring and reporting of borrowing beyond central government. Lessons are drawn for the debt-related operations of the entire public sector (such as subnational governments or state-owned enterprises). The findings are mostly supported by the results of the 2020 World Bank PDMLF Survey (the “survey”) conducted with IDA countries.127 A literature review and analysis of other select countries’ practices were also carried out. The chapter is organized into sections around the key features that promote debt transparency in a PDMLF: (a) authority128 to borrow and the debt authorization cycle; (b) debt management institutional arrangements; (c) debt management limits; (d) disclosure requirements; (e) audit requirements; (f) the consequences of noncompliance with the PDMLF; (g) public availability of the PDMLF; and (h) relevant rules and regulations for overall public sector monitoring and oversight. 4.1 THE AUTHORITY Clarity in determining and delegating the borrowing authority is a key enabler TO BORROW of debt transparency. A PDMLF lists all the rules and procedures related to incurring debt and issuing guarantees in a clear manner, free of any ambiguity, thus reducing the likelihood for the debt to become invalid or legally contestable. This is important for the following reasons: (i) at the borrower level, the PDMLF sets out a check list of rules and procedures government officials should follow when deciding, preparing, and signing debt contracts; (ii) at the lender level, it provides information for appropriate risk assessment through proper due diligence, and monitors compliance of government borrowing operations with applicable laws; and (iii) for the general public and civil society, as well as CG supervisory agencies, it enables oversight of public accounts and debt, and strengthens accountability of government officials. Clear debt authorization provisions in the PDMLF help base transactions on legally valid and enforceable contracts. If the legality of a public debt transaction is contested based on lack of authority, the courts decide which party should carry the burden of such a noncompliant transaction. This generally involves determining whether the lender was aware of the lack of authority for the transaction and whether it “acted in good faith”. Therefore, if the PDMLF provides clarity and certainty for lenders, it will be easier for them to verify compliance with relevant legislation, thus lowering the risk of debt becoming contestable due to lack of authority. 127. The survey was conducted by the World Bank from March to December 2020. Representatives of debt management offices in countries classified as borrowers from the International Development Association (IDA) were invited to voluntarily complete an online survey (Annex I). Thirty-nine countries replied. 128. In theory, the authorization cycle refers to all stages of a debt or guarantee operation as well as the debt management strategy that is underpinning those debt and guarantee transactions. However, this section will focus on the former, which encompasses the delegation of power by the legislature and the execution of debt and guarantee transactions. The stages of a debt operation include planning, origination, negotiation, signing, closing, recording, settlement, and servicing of the debts. The words operation and transaction are used interchangeably to refer to individual debt or guarantee contracts. 85 Debt Transparency in Developing Economies Our analysis shows that countries usually define roles and responsibilities of the legislative and executive body according to the type of debt instrument and the desired level of scrutiny. For a riskier debt instrument, or for increased scrutiny, the national legislation usually keeps the authority to borrow at the legislative level; it may also require additional authorization for certain instruments (see 4.1 for country examples), like external commercial debt and guarantees. Some countries keep the borrowing authority at the legislative level within specific debt ceilings.129 The legislation also usually grants the authority to borrow directly to the relevant executive official (e.g., the council of ministers or the minister of finance) for frequently issued instruments such as market-based domestic bonds that have standard terms and conditions (see Figure 4.1).130 Figure 4.1 Examples of Common Authorization Cycles (*) Market-Based Borrowings Authorization Cycle Law delegates authority to issue securities and describes general Delegate the authority to terms and conditions issue individual securities Council of Ministers/ Debt Management Parliament/Congress/ Prime Minister/ Office/Treasury/ National Assembly Minister of Finance Ministry of Finance Approval of issuance External Loan Authorization Cycle Law delegates authority Delegate authority to borrow to borrow and enter into and enter into individual loan loan agreements agreements Council of Ministers/ Debt Management Parliament/Congress/ Prime Minister/ Office/Treasury/ National Assembly Minister of Finance Ministry of Finance Submission for Submission for signing or approval/ratification approval Source: Authors’ elaboration. (*): Market-based funding corresponds to issuing of domestic and international securities. 129. The Argentinian law on sustainable public debt, approved in 2021, introduces a limit for foreign-currency–denominated public securities and securities that are governed by foreign law or subject to foreign jurisdiction. The limit is determined for each fiscal year under the Budget Law. For any debt that exceeds the limit, Argentina’s Congress must grant authorization to issue. 130. See dimension PI-13.2: Approval of Debt and Guarantees, World Bank’s Public Expenditure and Financial Accountability Framework for Assessing Public Financial Management (PEFA 2019). DPI 1 Legal Framework, Debt Management Performance Assessment (DeMPA) tool, 2021, https://www.worldbank.org/en/programs/debt- toolkit/DeMPA. 86 Source: International Debt Statistics Debt Transparency in Developing Economies Table 4.1 Enhanced Reviews Required by Law for Different Transactions in Selected Countries Transaction/ Derivatives/Market Type of Enhanced External Loan Guarantee Transactions/Liability Review Management Operations Approval of Brazil (Federal Senate Brazil (Federal Senate Ghana (derivatives, approval legislature approval) a approval) a of Parliament) i Kosovo (ratification by Jamaica (prior approval of National Assembly per loan) b House of Representatives) e Bosnia and Herzegovina North Macedonia (law to be (a Parliamentary decision enacted for each guarantee in per loan) c favor of an external lender) f Nigeria (signed by Minister Honduras (approval) g as mandated by Parliamentary resolution) d Nigeria (external loans require prior approval of terms and conditions by National Assembly) d Approval of higher Kosovo (prior mandate before North Macedonia Honduras (conversion, executive organ each loan) b (government decision for consolidation or renegotiation, (Government/ guarantees in favor of prior recommendation of Council of Bosnia and Herzegovina domestic lenders) f Public Credit Commission) g Ministers/President) (approval by Council of Ministers for each loan, Turkey (President approval The Gambia (derivative and based on the specific decision for build-operate-transfer swap transactions are approved of the Parliament) c [BOT] projects) h by the Cabinet) j Honduras (certain external Honduras (prior approval by loans are approved by the President) g President) g Source: World Bank staff. a. Article 32 of the Brazilian Fiscal Responsibility Law, 2001. b. Article 11 of Kosovo’s Law on Public Debt 2010, Law Nr. 03/L-175. c. Article 27 of Law on Debt, Borrowing, and Guarantees in the Federation of Bosnia and Herzegovina. d. Article 19 of Nigeria’s Debt Management Office (Establishment Act). e. Article 17 of Jamaica’s Public Debt Management Act of 2012, No. 339. f. Article 22 of North Macedonia’s Public Debt Law. g. for external loans Article 73, for guarantees Article 78, for conversion, consolidation or renegotiation Article 67 of Honduras’s Organic Law of the Budget of 2004. h. Article 7 of Turkey’s Law on Regulating Public Finance and Debt Management and Law, No. 4749. i. Article 64 of Ghana’s Public Finance Management Act of 2016. j. Article 43 of The Gambia’s Public Finance Act of 2014. Parliamentary approval, ratification, or any other enhanced authorization can aid transparency by promoting checks and balances within government. This will require greater consultation among relevant government agencies and a review of the whole transaction to ensure that it is PDMLF-compliant. It also gives the legislature an opportunity to scrutinize the operation to ensure that any terms and conditions of the transactions correspond to the originally approved parameters. 87 Debt Transparency in Developing Economies Enhanced authorization is particularly useful for non-standard or novel debt instruments. This applies to instruments involving collateral or collateral-like structures (see Chapter 3.4). Enhanced authorization for such transactions helps deter unintended or surprise allocations of national assets in favor of specific lenders and also supports any commitments made under existing debt to treat creditors equally. At a minimum, the legislative branch should be informed and should clearly identify the body with the authority to provide collateral and approve the transaction.131 While enhanced authorization processes provide added scrutiny, flexibility is important for avoiding delays and limiting time-sensitive debt management operations. Debt managers need to be able to respond to changes in market conditions and a good balance between accountability and flexibility within the authorization cycle is important (Awadzi, 2015). Legislation that clearly spells out executive authority, such as allowing certain types of transactions up to a certain ceiling, or with predefined terms and conditions, might provide the executive with discretion to to proceed without unnecessary delays. Ghana and Nigeria provide such flexibility in their legislation by not seeking further approval from Parliament for contracts that are compliant with preapproved terms and conditions. In Kenya, the terms and conditions of derivatives are pre-approved by the National Assembly up to a certain ceiling determined in the budget policy statement132 so that the Cabinet Secretary does not need to request additional approval. In Honduras, public debt may be renegotiated, converted, or consolidated with the prior recommendation of the “Public Credit Commission”, and the National Congress must be informed about the operation. 4.2 DEBT MANAGEMENT Transparency of debt operations is promoted by clear legislation and regulations RESPONSIBILITIES that define the roles and responsibilities of the unit in charge of executing debt operations. Typically, the highest level executive official with the delegated authority to sign debt contracts is tasked, through the PDMLF, with assigning the management of public debt to the debt management office (DMO) for operational efficiency and technical proficiency. Ideally, a single entity should be assigned sovereign debt-management functions (DPI 1, DeMPA 2021); when multiple institutions, or entities within the same institution are in charge of debt-management activities, proper coordination mechanisms are essential and should be clearly set out in the legal framework (IMF/ World Bank, 2014). To promote transparency, countries should enable verification of actual authority in individual transactions. Debt-management legislation usually grants the authority to borrow to an office or position (such as Minister of Finance, Cabinet Secretary, or Permanent Secretary). For operational simplicity, the office-bearer with such authority may have to assign additional relevant officials the actual authority to sign and process 131. In Rwanda, the Minister of Finance has the authority to provide collateral on behalf of the government, and local administrations with financial and legal autonomy are also authorized to pledge collateral for their debt (Organic Law on State Finances and Property of Rwanda No. 12/2013, Article 50). Bosnia and Herzegovina legislation lists the types of collateral that may be provided to lenders (Law on Debt, Borrowing, and Guarantees in the Federation of Bosnia and Herzegovina, Article 6). In Kosovo, the law explicitly permits the granting of collateral for some state debt but prohibits the pledge of assets that are essential for delivery of public service (Kosovo’s Law on Public Debt 2010, Law Nr. 03/L-175, Article 10). 132. Article 56 of Kenya’s Public Finance Management Act, 2012. 88 Debt Transparency in Developing Economies debt operations through a power of attorney or other form of delegation. Since no staff is permanently assigned to the same unit or position, the lenders should verify the actual authority of the signatory to conduct a specific debt management activity at the time of transaction. To promote transparency in borrowing operations, guidelines for issuing and verifying the power of attorney should be published. Furthermore, the institution with the authority to borrow should make the most recent power of attorney, or other legal document such as the conditions precedent evidencing the delegation, available upon lenders’ request during the negotiations or before closing a transaction. Publishing standardized templates of the legal documentation to be used to verify the delegation of authority would facilitate transactions both for debt management officials and lenders. Some LIDCs describe the processes for the delegation of authority in their regulations and operating guidelines to avoid transactions being considered unauthorized and becoming contestable in the courts. In the Gambia,133 the legislation mandates that a power of attorney by the minister, or similar legal documentation, is required for the delegation of authority from the Minister of Finance to the Permanent Secretary in charge of transaction execution. Furthermore, the Permanent Secretary may delegate its functions to other DMO officials by way of a similar documentation. Similarly, in Kosovo, government consent is required before the Minister of Finance can delegate authority to a government official to sign an external debt contract. However, in Ghana, the Public Financial Management (PFM) law expressly prohibits the Minister of Finance from delegating their authority to borrow. Nigeria and Rwanda have detailed lower-level guidelines on borrowing procedures that describe all steps for signing a debt contract. In Nigeria134 and Cabo Verde,135 the regulations also describe the procedures required to establish negotiation teams and the role of legal departments. Within the DM roles and responsibilities, lawyers tasked with providing a legal opinion play a critical role in transparency. The legal opinion constitutes an early- detection system vis-a-vis a borrowing transaction, as it helps verify the overall legality,136 validity, and enforceability of the transaction under local and other applicable laws. When entering into a transaction, lenders typically request such legal opinions as part of the conditions precedent (CP) documents.137 LIDCs very often involve lawyers when issuing external bonds. According to the survey, 82 percent of LIDCs are required to provide legal opinions to their lenders in new borrowing or guarantee transactions. All of the respondent LIDCs confirmed that they had received legal assistance during loan contracting and bond issuances (either in-house legal counsel or external legal advisers or law firms) (See 4.2). 133. Article 60 of The Gambia’s Public Finance Act of 2014. 134. Article 33 of Nigeria’s Debt Management Office (Establishment Act). 135. Article 12 of Cabo Verde’s Lei Dívida Pública No. 43/IX/2018. 136. In Cabo Verde, the Attorney General has to issue a legal opinion to certify the legality of public debt issuance (“Lei Dívida Pública No. 43/IX/2018”, Article 11). In Nigeria, the Attorney General has a chair in the supervisory board of the debt management office (Nigeria’s Debt Management Office (Establishment Act, Article 5). In Zimbabwe, the Attorney General’s written prior opinion on the legal aspects of a debt contract is required (Article 13, Public Debt Management Act of 4/2015). According to the survey, in the Republic of Djibouti, the Supreme Court is the body providing the legal opinion. In other countries, the Ministry of Justice (Cambodia, The Gambia, and Madagascar) or the Treasury Counsel (Rwanda) has this role. 137. The CP documents include the legal evidence for all necessary conditions for creating the debt or guarantee, such as the internal and official government documents evidencing the authority to borrow, the actual authority of the government officials to negotiate and finalize the transaction, the power of attorney, and the cabinet or parliamentary approval or ratification. 89 Debt Transparency in Developing Economies Figure 4.2 Bond issuance 44.8% 20.7% 27.6% 6.9% Involvement of Lawyers in Bond Issuance, Loan Contracting, and Compliance Loan negotiation 57.0% 13.0% 30.0% Loan contract advice and drafting 53.1% 15.6% 6.3% 25.0% Compliance and 41.2% 23.5% 5.9% 29.4% regulatory advice In-house lawyers (lawyers within DMO and lawyers within institution) In-house lawyers and external counsels External counsels Lawyers from other institutions Source: PDMLF Survey. 4.3 PUBLIC DEBT DM practices can be limited in the PDMF to ensure transparent and accountable DEFINITIONS AND LIMITS decision making. It is sound practice for national legislations to: (a) provide a definition of public debt in line with international standards (see chapter 2); (b) announce the country’s debt-management objectives; and (c) provide a list of permitted debt instruments, transactions or sources of funding (e.g., domestic and external markets, type of lender).138 National or regional definitions of public debt in LIDCs often deviate from international standards. As shown in box 4.1, LIDC public debt definitions are very heterogeneous, both in terms of instruments and borrowing sectors. As discussed in Section 3.4, RBLs are the typical example of instruments that fall through the cracks in national debt definitions. A narrow definition of public debt in national legislation contributes to “debt surprises” and ultimately underestimates the government’s debt burden in national statistics. It may also undermine efforts to standardize coverage of public debt statistics in indirect sources (see Chapter 2.2), as debt statistics compilers may not have the mandate to collect debt-related information beyond what constitutes debt according to national legislation. 138 For example, Ghana, The Gambia, Kosovo, Kyrgyz Republic, Vietnam, and Zimbabwe all include these factors in their primary legislation. 90 Debt Transparency in Developing Economies Box 4.1 Public Debt Definitions in Selected Countries Countries make a strong commitment to enhance transparency and accountability when their legal framework specifies what constitutes public debt. Among survey respondents, 69 percent of countries define debt in their legal framework. The following examples show that countries often define debt as “all financial liabilities created as a result of borrowing.” They also often list the types of instruments (for example, loans, bonds, and guarantees) and specify if the sectoral coverage is limited to central government or if it extends to local authorities or public enterprises. Ghana, PDM Act of 2016: “Government debt” means a financial claim on the government that requires payment by the government of the principal sum or the principal and interest to a creditor. Sierra Leone, PDM Act 2010: “Public debt” includes all financial liabilities created as a result of borrowing by the government, local councils, and public enterprises, and includes outstanding liabilities that have been securitized by issuing government securities. Kosovo, Law on Public Debt 2010: “Debt: any financial obligation to repay or otherwise pay money created by a Financing Contract or by notes of treasury, a bond, overdraft or other security issued as consideration for the disbursement of funds, as well as the obligation to repay principal, interest, discount, and any fees, commissions or penalties of any nature.”(continued): “State Debt: Debt incurred on behalf of the Central Governmental Institutions that the Republic of Kosovo is obligated to pay, but shall not include any obligation of certain other governmental entities, including but not limited to Municipalities, public enterprises, or the Central Bank of Kosovo.” Rwanda, Organic Law on State Finances and Property 2013: “Public debt: any State monetary liability or treasury bill issued by Central Government or decentralized entity or any other debt the State may take on.” Kyrgyz Republic, Law on the state and non-state debt of the Kyrgyz Republic, 2001: “Public debt: the total amount of disbursed and outstanding internal and external public debt of the Kyrgyz Republic on a specific date under a loan agreement or other debt obligations of the state.” (continued): “Debt obligation: all types of obligations under securities, loan agreements, agreements and other obligations in paper form and/or electronic records establishing the fact of borrowing certain funds or purchasing goods and/or services with an obligation to reimburse the corresponding amount or the contractually agreed value of goods and/or services in whole or in part.” Vietnam, Law on PDM, 2009: “Debt” means a loan to be repaid, including the principal, interests, charges, and other related expenses at a point of time, which arises from the borrowing by a borrower that is permitted to take loans under the law of Vietnam” (continued): “Government debt” means a debt arising from a domestic or foreign loan which is signed or issued in the name of the state or the government or a loan signed or issued by or under the authorization of the Ministry of Finance under law. Government debts do not include debts issued by the State Bank of Vietnam to implement monetary policies in each period.” Zimbabwe, PDM Act, 2015: “Public debt” comprises domestic and external (a) government debt, lending, and guarantees; (b) local authority debt, lending, and guarantees; and (c) public entity debt, lending, and guarantees; and it includes the debt of any other entity as the Minister may specify by notice in the Gazette. Jamaica, PDM Law, 2013: “Public debt” means all financial liabilities created as a result of borrowing or guarantees by government and includes government securities. 91 Debt Transparency in Developing Economies The most important consequence of the differences in the way countries define debt would be in relation to reporting. The detailed definitions, which address sectoral coverage and list types of eligible debt instruments, eliminate ambiguities in identifying the authority to borrow and the scope of debt to be reported. By setting overall debt management objectives and providing a list of permitted instruments, transactions or funding sources, the PDMLF adds an accountability dimension to DM operation assessments. Most PDMLFs in LIDCs include at least one of these (see Figure 4.3), and almost two-thirds include debt management objectives in a PDMLF, for example: (i) to meet the government’s borrowing requirements; (ii) to minimize the medium- to long-term expected cost, while keeping risks in the debt portfolio at acceptable levels; and (iii) to promote the development of the domestic debt market (WB, 2013). Many more LIDCs spell out permitted purposes of borrowing in their legislation (for example, to finance the annual budget deficit, to use for investment projects, or to build international reserves). It is sound practice to clarify that operations not specifically listed in the law should be separately authorized by the legislature. Figure 4.3 Sources of borrowing 79.5% 20.5% National Debt and Guarantee- and debt instruments Related Policies Included in the Legal Framework Clear debt management 64.1% 2.6% 33.3% objectives Purposes for which the government can 56.4% 5.1% 38.5% grant a loan guarantee or on-lend a loan Purposes for which the 76.9% 7.7% 15.4% government can borrow Yes No N.A. Source: PDMLF Survey. 4.4 DISCLOSURE Legal requirements to publicly disclose debt-related information in a comprehensive REQUIREMENTS and timely manner directly enhance transparency. As discussed in Chapter 2, debt data are disclosed in different formats depending on the final audience (e.g., authorities, investors, general public) and objectives, and it is good practice to publicly disclose debt statistics at least annually (preferably quarterly or semiannually). At a minimum, public debt statistics must include information on public debt stocks (by creditor, instrument, currency, interest rate); debt flows (by maturity); cost and risk measures of the debt portfolio and stock of guaranteed debt (WB, 2021). Additionally, legislation that further promotes transparency and accountability requires publication of debt management strategies and annual borrowing plans (ABP), as well as reporting of debt management activities to the legislature or within the executive branch. 92 Debt Transparency in Developing Economies Most LIDCs have reporting requirements in their PDMLF. According to the survey, 79 percent of LIDCs have legal requirements to publish “debt statistics” and 77 percent are required to publish “debt management strategies.” Publication of an ABP, is required in only 46 percent countries, and there are requirements to publish a “debt sustainability analysis” in 56 percent of countries (see figure 4.4). Most countries (72 percent) require reports on debt management activities to the legislature. For many countries it is mandatory to report the performance of debt management operations on a yearly basis,139 while reporting on contingent liabilities is required in only 36 percent of the countries. Figure 4.4 Reporting to Parliament on debt management 71.8% 5.1% 23.1% Disclosure Requirements activities in the Legal Framework Reporting on contingent liabilities/ 35.9% 10.3% 53.8% fiscal risks Publication of 79.5% 5.1% 15.4% debt statistics Publication of debt 76.9% 5.1% 17.9% management strategy Publication of annual 46.2% 5.1% 48.7% borrowing plan Publication of debt 56.4% 5.1% 38.5% sustainability analysis Yes No N.A. Source: PDMLF Survey. In addition to aggregated statistics, a PDMLF requiring public disclosure of granular information related to individual debt transactions signals a country’s commitment to debt transparency. Public availability of information related to individual debt transactions (e.g., size, financial terms, etc.) encourages public scrutiny of the borrower’s performance in promoting and protecting the country’s financial interests, it also facilitates more in-depth debt analysis by existing and potential lenders and other stakeholders. 139. E.g., Ghana, India, Kosovo, Lao People’s Democratic Republic, Madagascar, Mauritania, Mozambique, Nepal, and Zimbabwe. 93 Debt Transparency in Developing Economies Secrecy and unreasonable confidentiality clauses limit the ability to disclose transaction-level information to other stakeholders. The use of confidentiality clauses that require parties to maintain secrecy in debt transaction details is problematic as they deprive the public of the opportunity to scrutinize debt terms, and they leave other lenders and stakeholders in the dark with respect to the true nature and scope of the borrower’s debt portfolio (Gelpern et. al. 2021). In contrast to secrecy, confidentiality per se is not necessarily unreasonable.140 A PDMLF that requires adequate public disclosure and accessibility of transaction- level information can influence confidentiality practices in individual debt transactions. Legislation in the borrowing country that requires public disclosure of an adequate amount of debt transaction information helps the borrower avoid entering into secrecy provisions and reduces the chance of entering into unreasonable confidentiality requirements. 33 percent of countries in the survey acknowledged the existence of confidentiality clauses in more that 10 percent of their contracts. Survey responses revealed that in most cases, confidentiality clauses refer to keeping financial terms of the operation confidential, but they may also preclude disclosure of the total amount of the loan, the lender’s name, or the purpose of borrowing. Some countries have adopted a requirement to disclose entire debt contracts. For example, the assembly in Barbados, Kosovo, Kyrgyz Republic, Philippines and Sierra Leone publish the ratification law and entire external debt contracts in the official gazette as required by their respective primary legislation. In those cases, the relevant rules of the PDMLF need to include a way to address and comply with confidentiality obligations, and the treatment of commercially or market-sensitive information (for example, through redaction or delayed publication). In addition, the applicable legal framework would need to account for scenarios where the statutory rules of a relevant jurisdiction limit public disclosure of certain information or specify modalities for compliant disclosure (e.g., the lender’s country or of the country whose laws govern the debt contract). Recognizing and catering to these kinds of considerations may increase the workload of government officials and expose the process to operational risks. Adequate public disclosure ultimately aims at a degree of disclosure of public debt transaction information that is meaningful, and sufficiently granular to facilitate stakeholder awareness, scrutiny of government actions and public accountability. What might this look like in practice? The following information about public debt transactions should be made publicly available and could be mandated at the appropriate legislation level, regulations or procedural rules: (a) core terms and conditions of the relevant transaction (including financial and legal conditions); (b) identification of main legal documents that make up the transaction; (c) a summary of the transaction structure/ design; (d) the nature of any collateral or quasi-collateral granted in connection with the transaction; and (e) the nature of any amendments, supplemental agreements or 140. Possible drivers for keeping certain information confidential include where the information is proprietary or technical (e.g., financial calculations or formulae), price- sensitive or commercially sensitive (e.g., fee information); where non-disclosure (or actual disclosure) is required or where non-disclosure is time-limited by securities laws, market abuse or insider dealing rules (e.g., interest rate-related information), or privacy and freedom of information-related rules. Such rules may be binding for either or both the lender or the borrower, depending on the context, and rules may vary depending on whether the debt instrument is a loan or a bond etc. and on which relevant governing law and jurisdiction is involved. Other possible drivers related to relevant information pertain to matters of national security or national strategic interests (e.g., a debt transaction for the sale of military or intelligence equipment). 94 Debt Transparency in Developing Economies side letters entered into in connection with the transaction throughout its life. At the transaction level, the relevant parties need to agree to such public disclosure (within the debt contract itself). The precise nature of what constitutes “adequate public disclosure” would vary by transaction and by country, depending on the relevant legal frameworks. Statutory/ legislative requirements, which may be outside the PDMLF itself, may influence the ability of parties to publicly disclose and share transaction-level information; and may specify non-disclosure for a certain period of time. For example, relevant limitations or modalities for a sovereign borrower’s public disclosure of debt transaction-related information may be affected by domestic laws related to freedom of information, national security, or privacy rules. There may also be public policy considerations where the interests of disclosure are weighed against strategic or other national interests. In addition, where a debt transaction is governed by the laws of a jurisdiction other than the borrower’s, there may be governing jurisdiction laws that limit public disclosure or regulate the timing of public disclosure. Examples include antitrust laws, insider dealing and market abuse rules. 4.5 AUDIT REQUIREMENTS Public debt accountability and transparency is strengthened by introducing regular external and internal audits of government DM activities. A PDMLF should promote frequent and comprehensive financial audits, compliance audits, and performance audits (i.e., of the effectiveness and efficiency of government debt-management operations, including the internal control system). It should also promote publication of the external audit reports within six months of completion (DeMPA, 2021).141 Among the countries surveyed, more than half require external auditing,142 while the rest impose internal audits. Internal auditing is generally the responsibility of internal audit units, such as the general inspectorate of finance or a department of internal audit within the Ministry of Finance. In most countries, external audits are undertaken by the Court of Accounts or national audit authorities. In some countries (Afghanistan and Bhutan), this task is also the responsibility of executive branches, such as the Treasury or the Ministry of Finance. According to the survey, the legal framework of 39 percent of LIDCs requires external and internal audits. 141. External audit practice should be consistent with international standards, such as those set by the International Organization of Supreme Audit Institutions (INTOSAI). 142. “External audit” refers to auditing conducted by auditors outside the organization of the DMO/Treasury/Ministry of Finance. Such an external audit could be conducted by the central government’s audit department or Court of Accounts, National Audit Office, or Chamber of Accounts. 95 Debt Transparency in Developing Economies Figure 4.5 Audit Requirements in the Legal Framework 43.6% 35.9% 2.6% 2.6% 53.8% 61.5% Internal Audit External Audit Yes No N.A. Source: PDMLF Survey. 4.6. STIPULATION OF The PDMLF can increase its enforceability by stipulating the consequences of THE CONSEQUENCES OF not complying with its rules. Robust legislation that promotes understanding of the NONCOMPLIANCE WITH compliance mechanisms functions as an incentive to the parties to lessen the likelihood THE PDMLF of engaging in collusive or fraudulent transactions. In that respect, the PDMLF would also need to specify how non-compliant debt will be treated and indicate any legal or administrative consequences for government officials who generate non-compliant debt. This would enhance debt transparency by reducing the scope for hidden debt, or otherwise contestable debt. As in other PDMLF dimensions assessed in this chapter, LIDCs show considerable heterogeneity in rules on this topic. Box 4.2 presents a few country examples. 96 Debt Transparency in Developing Economies Box 4.2 Legal Consequences of Noncompliance in Selected Countries Many LIDCs specify the consequences of conducting non PDMLF-compliant transactions in their legislation. The most common consequences consider any debts contracted in violation of law as invalid or void. Among countries surveyed, 44 percent say that the contracted debt would be considered void if there is a violation of law. Most PDMLFs impose civil and/or criminal sanctions on government officials who take part in invalid debt transactions (49 percent of the countries in the survey). In Brazil (Brazil’s Fiscal Responsibility Act, Article 33), Côte d’Ivoire (Organic Law 2014-337, Article 76), Honduras (Organic Law of Budget, Article 70), Kosovo (Public Financial Management and Accountability Law, Article 49.4), and South Africa (Public Finance Management Act 1999, section 68), debt laws specify that operations in breach of legislation would be considered null and void. In Brazil any noncompliant operation would need to be canceled, and disbursed amounts refunded, excluding interest and other financial charges As per Zimbabwe’s Public Debt Law (Article 24), the government is not bound by a debt contract or the guarantee, indemnity, security, or other transaction that is not in compliance with the law. However, if the lender obtained a written legal opinion from the attorney general before the financial close of the transaction, this legal opinion would prevail for determination of compliance. Suriname’s law (Law on the Government Debt, Articles 4 and 12) invalidates any debt that is in excess of the debt ceiling and prohibits any payment of debt or guaranteed obligation exceeding the limit. In addition, lenders and the government, as the parties to a transaction, are required to provide each other with all data, statements, or documents to conclude a contract. The law determines that providing incomplete or incorrect data, statements, or documents nullifies the government’s obligations under the agreement in question. In The Gambia, the Public Finance Act (Article 45/2) dictates that if lenders are involved in corruption practices with government officials, the government is not obliged to service the debt. 4.7. PUBLIC AVAILABILITY A single piece of legislation entirely devoted to public debt management (or a specific OF THE PDMLF section within broader public finance legislation) contributes to transparency by providing easy access to the rules governing borrowing operations. Fragmented legal frameworks can make it difficult to identify applicable rules and procedures for a debt transaction (i.e., when two ministries have the authority to borrow under separate laws). Several countries have dedicated public debt management laws, while others have a dedicated chapter in their PFM laws (see Table 4.3 for examples). Even though it might be hard to achieve such consolidated legislation under some legal systems, countries should seize the opportunity to reform their PDMLF to reduce such fragmentation. 97 Debt Transparency in Developing Economies Table 4.2 Examples of PDMLFs Accessible on the Web Central Finance Institution/DMO Website Third-Party Website (*) Kenya a Jamaical l Honduras b Indonesia m Rwanda c Peru n Kosovo d Lao PDR o The Gambia e Vietnam p Chad f Nepal g Nigeria h Bhutan i Maldives j South Africa k Source: Authors’ elaboration. * National digital government service platforms, legislation, or regulation portals: a https://www.treasury.go.ke/publications/bills-acts-agreements.html b. https://www.sefin.gob.hn/normativa/ c. http://www.minecofin.gov.rw/index.php?id=135 d. https://mf.rks-gov.net/page.aspx?id=2,43 e. https://www.mofea.gm/constitution f. https://finances.gouv.td/index.php/publications/lois-des-finances?start=20 g. https://mof.gov.np/en/document/?c=83&t=&y= h. https://www.dmo.gov.ng/publications/other-publications i. https://www.mof.gov.bt/publications/acts-policy/ j. https://www.finance.gov.mv/public-finance/legislation k. http://www.treasury.gov.za/legislation/default.aspx l. Jamaica’s debt management legislation is found both in the Ministry of Finance website (partially) and also the Ministry of Justice’s website. https://moj.gov.jm/search/site/debtpercent20management; https://mof.gov.jm/documents/documents-publications/legislations.html m. https://www.kemenkeu.go.id/en/publications/regulation-updates/ n. https://www.gob.pe/ o. laoservicesportal.gov.la p. http://vbpl.vn/TW/Pages/vbpqen-toanvan.aspx?ItemID=10471&Keyword=publicpercent20debt A single piece of legislation entirely devoted to public debt management, (or a specific section within broader public finance legislation) contribute to transparency by providing access to the relevant DM documents. Fragmented legal frameworks, however, make it difficult to identify applicable rules and procedures for a specific transaction. Several countries have dedicated public debt management laws, while others have a dedicated chapter in their PFM laws (see Table 4.3 for examples). 98 Debt Transparency in Developing Economies Table 4.3 Examples of PDMLF Architecture Dedicated PDM Law a Chapter in PFM Law b Jamaica Afghanistan Kosovo Bhutan Mauritius The Gambia Sierra Leone Ghana Vietnam Guyana Zimbabwe Honduras Kenya c Source: Authors’ elaboration. a. Jamaica’s PDM Act of 2012, Kosovo Law on Public Debt, Mauritius’s Public Debt Management Act of 2009 (amended through 2012), 2010, Sierra Leone’s PDM Act of 2011, Vietnam Law on PDM of 2009, Zimbabwe’s PDM Act of 2015. b Afghanistan’s Public Finance & Expenditure Management Law of 1983, Bhutan’s Public Finance Act of 2007, The Gambia’s Public Finance Act of 2014, Ghana’s Public Financial Management Act of 2016, Guyana’s Fiscal Management and Accountability Act of 2003, Honduras’s Organic Law of the Budget, 2004, Kenya’s Public Finance Management Act of 2012. c. Kenya Public Finance Management Act of 2012; at the time of this report Kenya had an outstanding Public Debt Authority Management Bill of 2020, expected to become a dedicated PDM law once enacted. 4.8. EXTENDING THE A comprehensive PDMLF helps promote debt transparency beyond the central- PDMLF’S SCOPE TO government level, to include other public sector debt. Countries that have policies THE ENTIRE PUBLIC to monitor and oversee public sector debt-related obligations (i.e., including SOEs and SECTOR BORROWING sub-nationals) in their legislation are contributing strongly to enhanced transparency. This can be done by: (a) requesting central government authorization at the onset of a debt transaction; (b) requesting central government approval for the relevant entity’s borrowing limit, debt management strategy, or annual borrowing plan; and (c) providing detailed reporting to the central government on its debt portfolio. Many LIDCs have provisions in their legal framework regarding non-financial Public Sector (NFPS) borrowing. More than half the countries in the survey require central government authorization for external and domestic borrowing operations by public sector entities. A few countries (Chad, Republic of Congo, Côte D’Ivoire, The Gambia, Mauritania, Niger, and Yemen) impose central government approval for the debt management strategy and public entity debt limits (see Table 4.4). 99 Debt Transparency in Developing Economies Table 4.4 Rules and Regulations of NFPS Borrowing and Reporting by Surveyed Countries Local Authorities Yes% No% N.A.% Approval of domestic loans / bonds by the CG 54 15 31 Approval of external loans / bonds by the CG 49 21 30 Reporting requirements to central government 59 10 31 Debt limit 36 33 31 Public Entities Yes% No% N.A.% Approval of domestic loans / bonds by the CG 54 10 36 Approval of external loans / bonds by the CG 56 8 36 Reporting requirements to CG 56 8 36 Debt limit 26 38 36 Source: PDMLF Survey. The central government’s responsibilities regarding monitoring and oversight of all public sector borrowing operations should be described in the PDMLF. For example, the legislation can specify if and when the central government can regulate the authority of the NFPS to borrow (for example, the central government may regulate an NFPS entity’s ability to borrow above a certain limit or only for external debt). It can also introduce special oversight arrangements for the financial transactions of the NFPS entities that include collateral or collateral-like characteristics; or structures that generate finance for the borrower but that would not, in a technical legal sense, necessarily be characterized as debt. There could also be provisions about the reporting obligations of the NFPS entities to the central government as well as applicable public disclosure and information-sharing requirements. 4.9. CONCLUSIONS Given different government types, organizational structures, and legal systems, there is not a one-size-fits-all approach for developing and implementing a PDMLF conducive to debt transparency. Table 4.5 summarizes the key recommendations that can guide PDMLF reform in LIDCs.143 143. These recommendations indicate key components that can be included in the legal framework of a country either in primary legislation or secondary regulations. 100 Debt Transparency in Developing Economies Table 4.5 PDMLF: Summary of Recommendations I. Clearly Identify the Authority to Borrow • Describe: i) the authority to borrow and issue guarantees and its limits; ii) the authorization cycle. • Design enhanced authorization and scrutiny mechanisms for non-standard or new debt instruments II. Clarify the DM Responsibilities • Identify institutions, and their roles and responsibilities in public debt-management operations. • Enable verification of actual authority in individual transactions. • Ensure that rules and procedures include lawyers’ legal opinions in debt-management operations. III. Define Public Debt and Set the DM Limits • Disclose a country’s debt policies, including: definition of public debt, debt management objectives and strategies, purposes of borrowing, financial instruments and sources of funding that can be used; also include whether collateral and/or collateral-like features are permitted. IV. Adopt Public Disclosure and Reporting Requirements that Promote Transparency • Provide clear rules for comprehensive and timely debt data and information disclosure and reporting standards, bearing in mind that these rules may shape types of confidentiality arrangements that the government enters into. • Provide adequate and timely public disclosure of information about individual debt transactions. V. Introduce Auditing Requirements • Introduce requirements for performing and publishing financial, compliance and performance audits. VI. Regulate Consequences of Noncompliant Debt • Regulate legal consequences for debt that does not comply with national laws. VII. Provide Accessibility to PDMLF Itself • Ensure that the debt management rules and procedures can be clearly identified and traced. • Ensure accessibility of the PDMLF documents on the relevant government websites. VIII. Extend the Scope of the PDMLF to the Entire Public Sector • Regulate the authority of the non-financial public sector to borrow. • Regulate the monitoring and oversight of the nonfinancial public sector borrowing. 101 Debt Transparency in Developing Economies REFERENCES Awadzi, Elsie Addo. 2015. “Designing Legal Frameworks for Public DM.” IMF Working Paper, WP/15/147. Buchheit, Lee C., Gulati, Mitu. 2010. “Responsible Sovereign Lending and Borrowing.” UNCTAD Discussion Papers 198. UNCTAD/OSG/DP/2010/2. Gelpern, Anna, Sebastian Horn, Scott Morris, Brad Parks, and Christoph Trebesch. 2021. How China Lends: A Rare Look into 100 Debt Contracts with Foreign Governments. Peterson Institute for International Economics, Kiel Institute for the World Economy, Center for Global Development, and AidData at William & Mary. Gelpern, Anna, Gulati Mitu, and J. Zettelmeyer. 2017. “If Boilerplate Could Talk: The Work of Standard Terms in Sovereign Bond Contracts.” SRRN Duke Law School Public Law & Legal Theory Series 2017-45. IMF. 2014. “Revised Guidelines for Public DM.” IMF Policy Paper, IMF, Washington, DC. https://www.imf.org/en/Publications/Policy-Papers/Issues/2016/12/31/Revised- Guidelines-for-Public-Debt-Management-PP4855 IMF. 2018. “G-20 Note: Strengthening Public Debt Transparency –The Role of the IMF and the World Bank” Washington, DC. IMF. 2020. “Public Sector Debt Definitions and Reporting in Low-Income Developing Countries”, IMF. 2003. Assessing Public Sector Borrowing Collateralized on Future Flow Receivables. IMF. Washington, D.C. IMF. 2002. Development of Standards for Security Interests, available at https://www. imf.org/external/np/leg/sem/2002/cdmfl/eng/pdb.pdf. Organization for Economic Cooperation and Development, OECD (2016), Sovereign Borrowing Outlook. https://www.oecd-ilibrary.org/governance/oecd- sovereign-borrowing-outlook-2016_sov_b_outlk-2016-en PEFA (Public Expenditure and Financial Accountability). 2019. Framework for Assessing Public Financial Management. https://www.pefa.org/sites/pefa/files/news/ files/PEFA-Framework_English.pdf Proite, Andre. 2020. Recording, Monitoring and Reporting Public Debt- Organizing a Back Office, A Guidance Note, World Bank Discussion Papers, No: 18. World Bank. Washington, D.C. UNCTAD (United Nations Conference on Trade and Development). 2012. Principles on Promoting Responsible Sovereign Lending and Borrowing. UNCTAD. Geneva. World Bank Group. 2015. “Debt Management Performance Assessment (DeMPA) Methodology.” World Bank, Washington, D.C. 102 Annex I Drivers of Public Debt Disclosure: Estimation Strategy and Results Debt Transparency in Developing Economies Drawing on the data of the WB debt reporting heatmap, Rivetti (2021) explores the correlation between the degree of public debt disclosure and its key driving factors. A data disclosure index is obtained as a simple average of the heatmap’s sub-components that measure instrument and sectorial coverage of debt statistics, availability of financial terms on new loans, and publication of DMS and ABP. The index is standardized to have mean zero and variance 1 in the sample of IDA countries to facilitate the interpretation of the results. Relying on cross sectional data of 74 IDA countries, the role that different factors play on debt transparency is estimated by the following equation: Yij = α + βXij + γZij + δj + ϵij where Yij denotes the transparency level for country i in sub-region j.144 The matrix Xij includes the set of key determinants such as: • Type of debt recording and management system (DRMS). Debt Management Offices (DMOs) in LIDCs uses either off-the shelf recording systems, developed by the Commonwealth Secretariat (CS-DRMS/Meridian) or UNCTAD (DMFAS), or idiosyncratic systems (mostly Excel-based). • Portfolio Composition: presence of collateralized debt or Eurobonds in the public debt portfolio. • External scrutiny: availability of a rating from one of three major rating agencies (Fitch, Moody’s, Standard & Poor’s). • Debt management capacity, proxied by the share of college graduates among DMO staff (source: WB 2020 survey). • DM Legal framework: presence of legal requirements to produce debt statistics, strategy and/or annual borrowing plan (source: WB 2020 survey). The matrix adds a set of controls. They include categorical variables that capture the country’s income level, debt sustainability risk as rated by WB/IMF’s LIC-DSA (low, moderate, high, or in debt distress), status of fragile and conflict-afflicted country as defined by the WB,145 and participation in the Highly Indebted Poor Countries (HIPC) initiative. Even though country fixed effects to address endogeneity could not be added, sub regional fixed effects () have been included to partially deal with this issue.146 144. Regional patterns are relevant since debt disclosure requirements could be enforced by regional institutions (e.g., Central Bank of West African States or Eastern Caribbean Central Bank). 145. https://www.worldbank.org/en/topic/fragilityconflictviolence/brief/harmonized-list-of-fragile-situations 146. Subregions are defined as follows: Central America and Caribbean (8), Eastern Europe and Central Asia (4), East Asia and Pacific (18), South Asia (6); Sub-Saharan Africa non-CEMAC (23); Sub-Saharan Africa CEMAC (4), West Africa non-WAEMU (9); West-Africa WAEMU (7). 104 Debt Transparency in Developing Economies The results from estimating the linear model in equation are the following: I II Debt Recording Systems COMSEC 0.694** 0.590** (0.272) (0.270) DMFAS 0.475* 0.389 (0.266) (0.261) Portfolio Composition Has Eurobond 0.491** 0.337 (0.245) (0.243) Has collateralized debt -0.164 -0.006 (0.281) (0.285) External Scrutiny Has rating 0.510** 0.504** (0.229) (0.227) Capacity Percentage of staff with college degree 0.012** 0.009* (0.006) (0.005) Legal Requirements Index (statistical bulletin, ABP, and Strategy) 0.747** (0.336) Debt Recording Systems Lower middle income -0.220 -0.182 (0.253) (0.249) Upper middle income 0.332 0.390 (0.435) (0.428) High DSA risk -0.751*** -0.758*** (0.227) (0.226) Fragile state -0.522** -0.567*** (0.211) (0.207) HIPC 0.180 0.425 (0.294) (0.305) N. of Obs. 74 74 R-squared 0.63 0.65 Note: Syria and Yemen are excluded from the analysis since they are currently facing armed conflicts. Standard errors in parentheses, * p<0.10, ** p<0.05, *** p<0.01. Column 2 in Table 1 adds an index that captures the presence of laws requiring the publication of debt statistics and key DM documents 105 Debt Transparency in Developing Economies The model shows that the use of standard DRMS, the availability of ratings and qualified DMO’s staff significantly contribute to improve debt transparency. Relative to no standardized systems, off-the-shelf DRMS (CS-DRMS/Meridian and DMFAS) facilitate production of debt statistics, with a greater contribution registered for the former. The results also show that the issuance of Eurobonds and the availability of ratings increase transparency, while countries with collateralized debt tend to be less transparent. Skills of DMO staff are key to improving transparency: a 10 percentage- point increase in the share of college graduates in the DMO staff increases transparency by 0.12SD. The effects of other controls are shown at the bottom. On the one hand, higher income levels seem to be positively correlated with transparency, but these effects are not statistically significant. On the other hand, debt transparency is a challenge primarily for fragile and highly indebted countries. Finally, past multi-country debt relief initiatives (e.g. HIPC) may have played a supporting role in debt transparency, but the estimated coefficient is quite noisy to be statistically significant. This may be due to the practice of privileging “indirect” reporting (e.g., to Paris Club, IMF/WB) over direct disclosure. 106 Annex II WB and IMF Main Public Debt Databases Debt Transparency in Developing Economies The results from estimating the linear model in equation are the following: Database Coverage ISD JEDH QEDS QSPD GFS IFS GDD Administered by World Bank World Bank World Bank World Bank IMF IMF IMF Number ~123 ~200 ~123 ~83 ~80 ~40 ~190 Country reporting coverage Country Low & middle All countries All countries All countries All countries All countries All countries groups income Country Country Country Country Source 1/ 2/ 3/ authorities authorities authorities authorities Monthly/ Frequency Annual Quarterly Quarterly Quarterly Annual Annual Quarterly Year earliest 1951 1990 1995 1950s data available Validation Medium Medium Medium High Low High process Budgetary Central Government Central General Government Nonfinancial Institutional Government sector coverage Financial Public Corporations Public Sector EBG, SSF, Other PPG PPG PPG SG, LG Dom/Ext Ext Ext Ext Dom/Ext Dom/Ext Dom/Ext Dom/Ext Securities Instrument coverage Loans SDR, C&D, SDR, C&D, Other IPSGS, OAP IPSGS, OAP Original maturity Analytical Remaining coverage maturity Currency of denomination 108 Debt Transparency in Developing Economies Database Coverage ISD JEDH QEDS QSPD GFS IFS GDD Residency of creditor Counterparty sector Source: WB-IMF. 2018, EBG = Extrabudgetary units, SSF = social security funds, SG = Senate Governments, LG = Local Governments, PPG = Public and Publicly Guaranteed, SDR = Special Drawing Rights, C&D = Cash & Deposits, IPSGS = Insurance, pension, and standardized guarantee, schemes, OAP = Other Accounts Payable. 1/ Country aurhorities (long term debt), BIS and country authorities (Short term debt), IMF Treasurers Department, staff estimates. 2/ QEDS, BIS, IMF, OECD, and World Bank. 3/ Country authorities, international institutions, and academic researchers. 109 Annex III Debt Restructuring and Transparency: Estimation Strategy and Results Debt Transparency in Developing Economies This section seeks to assess whether debt transparency across different types of creditors improved during debt restructuring episodes. Given (i) the need for accurate debt reconciliation, (ii) the incentives to fully disclose the debt portfolio to make the case for higher debt relief and (iii) debt management conditionality often attached to restructuring, debt transparency is expected to improve after a restructuring episode. The analysis aims to test this hypothesis using an index that captures levels of reporting for the 2005-2018 period. This index measures transparency as the reporting of data to high profile databases whose quality has been validated in accordance with the IMF Public Sector Debt Statistics Guide (PSDSG) definitions. As a result, two voluntary and detailed public databases are considered: the IMF Government Finance Statistics Yearbook (GFSY) and the IMF/World Bank Quarterly External Debt Statistics database (QEDS). Both of these databases are built on internationally agreed methodological foundations (PSDSG), are consistent with the reporting requirements set out in the General Data Dissemination System (GDDS) and Special Data Dissemination Standard (SDDS) frameworks and provide detailed breakdowns of debt portfolios. No official database currently exists on debt restructuring, but several academic and institutional efforts have been undertaken to build comprehensive statistics on them. Restructuring episodes have been measured in a variety of ways, but the two most comprehensive databases are the joint Bank of Canada -Bank of England (2020)147 on debt default and the dataset of Bon and Cheng (2020).148 For the purpose of the analysis below, the focus is on i) debt restructuring episodes and ii) cumulative restructuring volume (in billions of USD). 147. The Bank of England/Bank of Canada Sovereign Default Database contains the most detailed breakdown of sovereign defaults covering a long time period (1960-2020) and a wide range of countries. The methodology and data builds on past research from Beers and Chambers 2006, Suter 1992, Cruces and Trebesch 2011, Tudela et al 2011, Papaioannou and Trebesch 2012, Tweedie et al 2012, Paris Club, IMF, IDA, the World Bank, national sources and other private creditors. The database considers a default to occur when debt service is not paid on the due date or within a specified grace period, when payments are not made within the time frame specified under a guarantee or, absent an outright payment default, in circumstances where creditors incur material economic losses on the sovereign debt they hold. 148. The Bon and Cheng database is a cumulative exercise covering a total of 140 restructuring episodes across 64 countries over the 2000-2019 period. The database builds on past work from, Development Reimagined, Dreher et al (2017), Bluhm et al (2018), Hurley et al, (2018), Kratz et al, (2019), Debtwire , Factiva , the John Hopkins China- Africa Loan Database , Paris Club, Cheng et al (2018), Cruces and Trebesch (2013) and Asonuma and Trebesch (2016). The definition of a restructuring episode is more narrowly focused on the traditional Paris Club definition of official restructuring which is generally consistent with IMF PSDSG. This leads to significantly fewer cases than what was recorded in the BOE-BOC database. 111 Debt Transparency in Developing Economies Consistent with past research, this analysis examines the effects of restructuring episodes with different creditors, distinguishing those that coordinate their efforts through the Paris Club from the others. For both databases (Bank of England/Bank of Canada and Bon/Cheng), two categories of restructuring episodes are considered (Paris Club, non-Paris Club), while controlling for other factors that may limit a debtor government’s ability to make progress on achieving higher levels of debt transparency. For example, a country that is experiencing a large economic downturn, hyperinflation, balance of payment problems, or conflict149 would not be likely to prioritize recommendations regarding the comprehensive disclosure of its debt obligations or improvements in debt governance. Finally, controls for external interventions that may influence government’s macroeconomic performance and accountability (Cheng et al 2018) are also added. Given the multiple restructuring episodes (in number and volume) over a relatively short period (2005-2018), a straightforward fixed effects specification is adopted.150 There is an advantage of our simple fixed effects approach in that it captures within-country changes in debt transparency and allows for a direct comparison of our dependent variables before and after restructuring episodes (controlling for other factors). As in Bon and Cheng (2020), robust standard errors are used to correct for heteroskedasticity, autocorrelation, and error correlation across panels (Driscoll and Kraay 1998) to estimate the following equation: 3 cn cn 3 pc pc 3 priv priv 2 dm dtit = αi + λt + ∑(q=1) βqrfqit + ∑(p=1) βprfpit + ∑(v=1) βv rfit + ∑(d=1) βd DeMPAdit + ωIMFit + τgdpcapit + ρcabit+ γinflit + δsfiit Where: dtit = debt transparency index in country i at time t x rfqit = debt restructuring by episode and volume (cumulative) for creditor x in country i at time t x [Note: rfqit = 1 for all years after a qth restructuring episode with creditor x and 0 otherwise] gdpcapit = GDP per capita in country i at time t inflit = inflation rate (CPI) in country i at time t sfiit = state fragility index score in country i at time t IMFit = 1 when country i is involved in an IMF program for year t DeMPAdit = 1 for all years following an assessment in country i And, βqcn, βp pc , βvpriv , ω, τ, ρ, γ, δ are unknown parameters to be estimated.151 , βddm 149. Conflict data was sourced from the Center for Systemic Peace State Fragility Index. The index is built on eight underlying factors that contribute to a country’s fragility (see Marshall and Elzinga-Marshall 2017). 150. Past statistical specifications for debt restructuring episodes have had sufficiently long time series to take advantage of treatment events (for example, Brady bonds) under a difference in difference approach (Reinhart and Trebesch 2016). Using a shorter time series, Cheng et al (2018) and Bon and Cheng (2020) use a local projection method to provide joint inference for impulse response coefficients. 151. As was the case with previous statistical specifications of debt restructuring episodes, there is a potential endogeneity, or reverse causality, problem (Bon and Cheng 2020; Horn, Reinhart and Trebesch 2020). While it is possible that the discovery of new debt may push a debtor country into restructuring, it is unlikely, especially in the context of multiple restructuring episodes. As argued in Reinhart and Trebesch (2016), Cheng et al (2016) and Bon and Gong (2020) reverse causality is even less likely in the context of group restructuring which qualify recipient countries based on being members of a specific group. 112 Debt Transparency in Developing Economies Building recent contributions to measuring fiscal and debt transparency (Murara et al 2015; IMF 2020a), an index of external debt transparency is constructed based on eight categories that are important for comprehensive debt reporting. The scoring scheme for each of these is depicted below in Table A1. This article uses the IMF classification of instrument coverage which range from narrow instrument coverage (D1) to broad coverage (D4).152 Table A3.1 External Debt Transparency Scoring Scheme Integrated Financing Stock of Coverage* Instrument Currency Counterparty Maturity Indicator Stocks and Terms Financial (cv) (inst) (cur) (cpt) (mat) Flows (sf) (ft) Assets (fa) No data= 0 No data = 0 No breakdown No breakdown No breakdown Stocks only = 0 Principal and No data = 0 BG = 1 D1 = 1 =0 =0 =0 interest = 0 CG = 2 D2 = 2 Stocks, Total stock= 1 Scoring GG = 3 D3 = 3 Domestic/ External by ST/LT transactions and Principal PS = 4 D4 = 4 foreign = 1 sector = 1 breakdown = 1 OEF = 1 and interest disaggregated =1 * BG = budgetary central government, CG = central government, GG = general government, PS = public sector Using this framework, a feasible measure of debt transparency can be computed as an additive index of the product of coverage and each of the remaining six components of the scoring scheme:153 dtit = cvit (instit) + cvit (curit) + cvit (matit) + cvit (sfit) + cvit (ftit) + cvit (fait) + cvit (cptit) Sectoral coverage is used multiplicatively because, i) coverage can vary by component, and, ii) higher degrees of coverage will increase transparency across the entire realm of debt statistics. 152. D1 = debt securities + loans; D2 = D1 + currency and deposits + SDRs; D3 = D2 + other accounts payable; D4 = D3 + insurance, pension and standardized guarantee schemes. 153. This indicator leaves out two dimensions discussed in the report: valuation method and contingent liabilities. Valuation methods tend to vary across and within databases, often with no clear distinction other than metadata. Contingent liabilities are not included as they are not reported in any of the existing databases in a systematic way. In some cases, countries report information on explicit guarantees, but these fall into a variety of definitions and degrees of coverage, which make them systematically not comparable across countries. 113 Debt Transparency in Developing Economies Table A3.2 Debt Transparency and Restructuring (Episodes and Volume) Episodes Volume BoC/BoE Bon/Gong BoC/BoE Bon/Gong PC first restructure -0.01 0.47 (0.61) (0.55) PC second restructure -0.64 -0.36 (0.40) (0.56) PC second restructure -0.46 -0.25 (0.32) (0.40) CN second restructure -0.59 -0.57 (0.67) (0.58) PC restructure volume (US Bn) -0.05*** -0.09 (0.01) (0.06) CN restructure volume (US Bn) 0.03 -0.21*** (0.04) (0.07) IMF Program 0.03 -0.13 -0.09 (0.34) (0.32) (0.32) DeMPA1 1.61*** 1.13** 1.37*** 1.06** (0.45) (0.52) (0.44) (0.48) DeMPA2 0.51 0.29 0.40 0.25 (0.59) (0.63) (0.55) (0.58) GDP pc (ln) 3.70 3.66*** 3.69*** 3.69 (0.32) (0.39) (0.27) (-.34) Current account -0.70 -1.33 -0.37 -1.38 (1.36) (1.26) (1.51) (1.33) Inflation (ln) 0.15 0.10 0.14 0.07 (0.11) (0.13) (0.13) (0.11) SFI -0.18** -0.17* -0.16** -0.16* (0.08) (0.08) (0.08) (0.08) Country fixed effects Y Y Y Y Countries 97 88 97 88 Obs 1656 1498 1656 1498 R2 0.32 0.30 0.33 0.31 *** - p<0.01; ** - p<0.05; * - p<.01 The model indicates that restructuring episodes do not have a statistically significant effect on indirect debt disclosure. Countries that undertake debt restructuring do not improve their level of reporting to international debt databases. There does not appear to be any impact from IMF programs. In turn, one of the strongest predictors of debt transparency is GDP per capita, which is assumed to be a proxy for the level of capacity after controlling for country fixed effects. 114 Annex IV Systems, Institutional Arrangements and Resources for Accrual Accounting Debt Transparency in Developing Economies Several enabling systems and institutional arrangements should be in place to reap the benefits of accrual accounting in debt reporting. These include: • Legal framework: In the absence of internationally mandated standards, the legal framework should clearly identify whether IPSAS is adopted as a basis for national standards, whether directly, indirectly or otherwise, in order to promote transparency, accountability and comparability of financial information. Any national standard setting arrangements should support objectivity, independence, and integrity in government financial reporting. Having an interested party involved in standard setting may cause conflicts of interest. • Unified Chart of Accounts (UCOA): A UCOA allows for seamless preparation of both statistical reports and whole of government financial statements. Many of these accounts are the same in both frameworks. Differences arise, however, in response to the reporting differences between these two frameworks described earlier. These differences should be properly captured in the UCOA and countries should be pragmatic about how differences between the frameworks are met.154 • Adequate technological capacity and information systems: An Integrated Financial Management Information System (IFMIS) supports the execution and management of public sector financial operations by enabling the electronic capture, recording, categorization, consolidation and reporting of transactional level financial data based on the UCOA and pre-defined standardized reporting formats. Access, process-related and administrative controls are typically programmed into the IFMIS to provide a level of security, data integrity and processing uniformity, enhancing credibility of government financial information, preventing data redundancy and thereby providing reliable accrual basis accounting data. Figure A.1 below denotes typical IFMIS architecture. 154. Optimizing the Unified Chart of Accounts Design, PEMPAL Treasury COP Public Sector Accounting Working Group, October 2020 116 Debt Transparency in Developing Economies Figure A4.1 Integrated IFMIS Architecture: An Example Core Modules Non-Core Modules IFMIS Modules Cash AR Reporting Analytics Asset PIM Budgeting Management Dash Fleet Employee GL AP Commitment Payroll Boards Management Loans Middleware/Big Data/Data Warehouse Intergration/Other E-GP Debt Tax Customs Systems HRMIS Others AMIS Non-Tax Bank PFM Systems Landscape Databases Legend: - AR: Accounts receivable - PIM: Public Investment Management - AP: Accounts Payable - AMIS: Audit Management System - GL: General Ledger - E-GP: Electronic Government Procurements Source: Authors’ elaboration. • Reciprocal information flow between the DMO and the accrual accounting system: Processes should exist for sharing information captured through the accrual accounting system and debt information captured by the DMO. For example, the DRMS should be integrated within the broader IFMIS. Additionally, the DMO maintains critical information on the inventory of all financial instruments, as well as details of these arrangements; this is needed to inform accrual accounting frameworks and related financial statement disclosures. Conversely, the accounting system captures information on payments arrears, lease arrangements, cash overdrafts and other debt-like instruments which may not be captured through the DRMS. 117 Source: International Debt Statistics Debt Transparency in Developing Economies • Inventory and control of financial instruments: A complete and reliable inventory and valuation of all financial instruments, as well as details of these arrangements, should be maintained to ensure that all related commitments, contingencies, and other relevant information are properly recorded in the financial statements and qualitative information is properly disclosed. In many countries, central coordination of the inventory for debt instruments is led by the Debt Management Office (DMO). • Internal controls and procedures to assist identification and management of ‘other contingent liabilities’: Processes for identifying and reporting ‘other contingent liabilities’ should be in place. “Other contingent liabilities” may include potential legal claims against government, guarantees provided, or any other form of implicit or explicit potential claim which are not automatically captured by government accounting systems. The information should be aggregated by nature and include information on estimated amount, probability and any possible reimbursement. • Internal controls to match spending with resources available: A standard form of internal control is a commitment control system, which limits spending commitments to budget authority either through computerized or, in some countries, through manual systems. The use of such a system can mitigate some of the pressures that may otherwise lead to payment arrears. • Clearly defined accounting policies, instructions (procedures) and competencies: Policies should be clearly defined at the central level to promote consistency of application among individual reporting units included in the consolidated financial statements. Likewise, clear and detailed accounting Procedures for both statistical and accounting frameworks can provide step by step guidance to enable operational level finance and accounting staff throughout government to process transactions in a consistent manner based on the accounting standards and policies in effect. • Adequate human resources and technical knowledge: Expertise is required in order to lead the reform process and oversee implementation. This core expertise should include strong technical knowledge of IPSAS, an understanding of the interrelationship between accounting and financial reporting with other PFM processes, and knowledge of the IFMIS or other information technology systems supporting the accounting processes. Capacity for establishing the market value of financial instruments may also need to be developed, as well as capacity for developing actuarial estimates of the expected cost of providing post-employment and other long-term benefits. • Adequate external audit capacity: Government financial statements are typically subject to external audit in accordance with national laws. External auditors, generally the Supreme Audit Institution (SAI) of the country, should possess the requisite expertise to complete an audit of IPSAS based financial statements, including the underlying systems, procedures, and controls governing the accounting process. 118 Annex V Proposal for an International Loan Repository Debt Transparency in Developing Economies An International Loan Repository (ILR) could strengthen and coordinate existing data collection initiatives. The goal of the ILR is to provide a modern platform to reconcile debt records between creditors and debtors, thus improving the accuracy of debt records and limiting operational risk. The ILR would also enable the creation of a central database, which can be used for public reporting purposes. Participation to the ILR would be voluntary. However, it is expected that the data reconciliation services, and reporting facilities provided by the ILR will act as a strong incentive for DMOs to participate. Furthermore, incentives could be introduced in MDB’s operations or creditor/debtor country legislation to spearhead its use and increase its relevance. While targeting official debt in a first step, the ILR can be broadened to include private external debt. Terms of existing and new public loan agreements would be registered by creditors or debtors participating in the initiative. Any subsequent loan transaction would be linked to an existing agreement and, once cleared by the debtor and creditor, stored in the ILR.155 The data will be encrypted to ensure data ownership and confidentiality. Disclosure clauses, and other relevant legal arrangements, would allow for debt reporting (e.g., WBs DRS) using the centralized ILR database (Figure A.2). 155. The use of block chain technology can be envisaged to include the full history of previous transactions and loan data. 120 Debt Transparency in Developing Economies Figure A5.1 ILR’s Automatic Reconciliation and Centralization of Loan Data Loan agreement Payments Clearing Payments 1. Loan agreement data Debtor 2. Loan disbursement transaction data Creditor Country or Government Multilateral Institution Acknowledges receipt of 3. Interest payments data Transmits data on 1. Loan agreement data 4. Principal payments data 1. Loan agreement data 2. Disbursements data 2. Disbursements data Transmits data on Acknowledges receipt of 3. Interest payments 1. Interest payments 4. Principal payments DRS reporting Statistics on demand 2. Principal payments International Debt Statistics publication Reconcilliation with QEDS and QPSDS data World Bank Source: Authors’ elaboration. 121 Debt Transparency in Developing Economies The proposed ILR’s main benefits are the following: • Provide a secure platform for data exchanges on loan transactions between creditors and debtors, replacing existing risk-prone and ineffective channels. • Ensure validation between creditor and debtor of all loan transactions. • Automate data recording in DRMS and thereby reduce data inconsistencies. • Ensure transparent data on loan financing conditions, with the possibility to introduce variables currently not covered by existing reporting templates (e.g., collateralization, quasi-collateralization/collateral-like features, jurisdiction, legislation). • Allow for real-time dissemination of statistics, and complement existing indirect reporting channels (e.g., WB’s DRS). • Promote the use of standardized debt definitions for statistics and debt reporting. • Assist creditors and debtors to move to digital end-to-end processing of loan transactions. 122