LE A R N ING PR O DUC T Findings from Evaluations of Policy-Based Guarantees Findings from Evaluations of Policy- Based Guarantees An IEG Learning Product © 2016 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work 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. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e- mail: pubrights@worldbank.org. Contents ACKNOWLEDGMENTS ........................................................................................................................ III ABBREVIATIONS .................................................................................................................................. V OVERVIEW........................................................................................................................................... VII 1. INTRODUCTION........................................................................................................................ 1 2. THE EVOLUTION OF PBGS ..................................................................................................... 3 3. THE QUALITY OF THE MACRO-FISCAL FRAMEWORK OF PBGS....................................... 7 4. DID PBGS IMPROVE ACCESS TO PRIVATE FINANCE AND CREDIT TERMS?................. 11 5. REFORM AGENDA UNDER PBGS ........................................................................................ 16 6. LESSONS ................................................................................................................................ 19 REFERENCES...................................................................................................................................... 23 Boxes Box 2.1. PBGs in the Context of Evolving Bank Policy ........................................................................... 4 Box 2.2. Outcome Ratings of Completed and Evaluated PBGs .............................................................. 6 Box 4.1. An OPCS Metric of Improvements in Financial Terms - 2012 ................................................. 14 Box 5.1. Summary of Reform Contents under Selected PBGs ............................................................. 16 Tables Table 2.1. World Bank Policy-Based Guarantees, 1975-2016 ................................................................ 5 Table 3.1. Financing Mobilized through World Bank Policy-Based Guarantees...................................... 7 Table 3.2. Macro-Framework for selected PBGs reviewed and evaluated by IEG .................................. 8 Table 4.1. Summary Terms of Commercial Loans Supported by PBGs................................................ 11 i CONTENTS Appendixes APPENDIX A: ASSESSING THE IMPLIED INTEREST RATE OF NON-GUARANTEED INSTRUMENT AND THE VALUE OF GUARANTEE: A METHODOLOGICAL APPROACH .............. 21 Evaluation Managers  Caroline Heider Director-General, Evaluation  Nicholas David York Director, Human Development and Economic Management  Anjali Kumar Acting Manager  Aghassi Mkrtchyan Task Manager ii Acknowledgments The task manager of the report was Aghassi Mkrtchyan under the supervision of Mark Sundberg andAnjali Kumar, and the general direction of Nicholas York and Caroline Heider. The report was prepared by Aghassi Mkrtchyan with substantial inputs from Mansoor Dailami and Chad Leechor. Team assistance by Yezena Zemene Yimer is gratefully acknowledged. The task team is grateful to Mark Sundberg, Zeljko Bogetic, and Anjali Kumar for substantive guidance and advice in all stages of the study, and to Lourdes Pagaran and Moritz Piatti for their advice and comments on the draft. The task team is also grateful to the World Bank colleagues in the Macro and Fiscal Management Global Practice, in the Treasury Department, and in Operational Policy and Country Services who substantially discussed early versions of this learning product. Peer reviewers were R. Sudharshan Canagarajah and Shahrokh Fardoust. iii Abbreviations DPF Development Policy Financing DPL Development Policy Lending DPOs Development Policy Operations FSAL Financial Sector Adjustment Loan IBRD International Bank for Reconstruction and Development IDA International Development Agency IEG Independent Evaluation Group IMF International Monetary Fund PBGs policy-based guarantees PCGs Partial Credit Guarantees PPAR Project Performance Assessment Report PRSCs Poverty Reduction Strategy Credit SSAL Special Structural Adjustment Loan v Overview In 1999, the World Bank introduced undertaken by IEG, in Serbia, policy-based guarantees (PBGs) to cover Macedonia, and Montenegro. It also private lenders against the risk of draws upon IEG’s evaluations of default by sovereign borrowers. As a Ghana’s Poverty Reduction Support new instrument, the PBG expanded the Credits, which provide background menu of financing options through material on the 2015 Ghana PBG. A policy lending. PBGs were built on the recent World Bank review of the World Bank’s policy-based operations Albania PBG was also taken into and existing guarantees, with the goal of consideration. The report draws on leveraging the World Bank’s resources IEG’s technical analyses, coordinated to attract more private financing to with relevant units of the World Bank enhance poverty reduction and growth. and, finally, interviews with a cross- section of World Bank staff. It therefore The World Bank’s early experience with represents an initial step in building PBGs was colored by its involvement greater awareness of a relatively new with the first guarantee in Argentina, and important tool in development which was called in 2002. However, finance. Based upon these experiences, interest in using the instrument grew the learning product offers the after the global financial crisis. Since following lessons: 2011, the World Bank has issued eight PBGs with seven countries. In 2013, the PBGs can represent an important operational framework was streamlined instrument of development policy to make PBGs more accessible to clients financing that can be used effectively to by bringing the instrument fully under help members overcome difficult Operations Policy 8.60 that covers all financing and reform challenges. The development policy financing . Interest PBG approach brings a judicious in PBGs now appears to be on the rise. combination of client benefits, private Member countries are particularly sector participation, and financial attracted to the scale of financing and leveraging of World Bank resources. the market access under financial stress From the point of view of the Bank, that PBGs have made possible. through a PBG, World Bank capital is tied up for a much shorter period than This paper is a survey of practices and for a standard policy loan. Additionally, results that have been observed in the the fact that only 25 percent of an issued implementation of PBGs, drawing guarantee counts toward the World largely upon four in-depth Project Bank’s country exposure limit creates an Performance Assessment Reports incentive for the use of the instrument, vii OVERVIEW especially for clients with large mitigating actions in the policy matrix is financing needs. And the PBG provides important if there are substantial an opportunity for the World Bank to macroeconomic risks. The World Bank’s leverage its own resources for the engagement through PBGs would also country’s benefit. In parallel, benefits to benefit from extended macroeconomic World Bank clients include improved monitoring that extends beyond the market access, potential diversification programs’ closing dates. This could be of the creditor base, longer maturity, managed in the context of the World and lower interest rates. IEG’s review of Bank’s country partnership framework. PBGs found that, with World Bank support, borrowers were able to meet The impact of PBGs on borrowers’ their financing needs during difficult credit terms varied from one program to market conditions. another. In all of the PBGs reviewed by IEG, the aggregate interest rates were A robust macroeconomic and fiscal lower than they would have been policy framework is essential for without guarantees. However, this fact sustaining benefits from improved alone may not be sufficient for assessing access to private finance for deficit the effectiveness of PBGs in improving financing. As a development policy borrowers’ credit terms. Estimating the financing instrument, PBGs can implied interest rate on the non- facilitate a range of important policy guaranteed portion of PBG-supported reforms in client countries. Some of the loans and the extent of possible PBG operations evaluated by IEG, “erosion” in the value of a guarantee is however, would have benefited from a important for more comprehensive stronger focus on mitigating fiscal and assessment of PBGs. In this respect, the macroeconomic risks. There are review finds that the financial benefits suggestions that in some cases, from improved credit terms varied relatively large loans mobilized through across the five PBGs that supported PBGs might have created incentives for commercial loans. The review therefore sub-optimal fiscal choices, particularly highlights the importance of in the absence of an International maximizing the financial benefits for the Monetary Fund (IMF) program. The borrowers from PBGs. Assessing the full report highlights the importance of extent of financial benefits using this rigorous application of the World metric in World Bank self-evaluation Bank’s operational policy on the could further enhance understanding of macroeconomic framework of the overall effectiveness of PBGs. development policy financing with a special focus on mitigating sovereign Greater attention to the modality for default risks through debt sustainability raising private finance is needed. The analysis. Incorporation of necessary choice between bank loans and bond issuance in capital markets involves viii OVERVIEW several trade-offs that need to be interest in PBGs, there is need for more considered in light of a particular comprehensive Bank group wide borrower’s circumstances and lessons corporate guidance on borrowing learned from previous operations. Most modalities and on the roles and recent PBG transactions have been responsibilities of relevant departments, through negotiated deals in the that could perhaps be undertaken in the international bank loan markets. For review of WBG guarantee instruments bank loan markets to ensure pricing launched in July 2016. A supportive efficiency, it is necessary for the World governance framework for operations Bank and the borrower to ensure that make use of PBGs is important for competitive bidding, transparency, and enhancing the overall effectiveness of due diligence. The review also this important instrument. emphasizes that, in light of growing ix 1. Introduction In 1999, the World Bank introduced policy-based guarantees (PBGs) as a new financing instrument to expand the menu of financing options in support of clients’ reform programs. PBGs cover private lenders against the risk of debt service default by the sovereign borrower. They are structurally the same as the existing “partial credit guarantees” of the World Bank, but they provide budget and balance of payments support and are not tied to any specific projects. Building on the World Bank’s policy- based operations and existing guarantees, the World Bank created PBGs with the ambitious goal of leveraging its resources in policy lending operations to attract more private financing so as to enhance poverty reduction and growth, and to reduce its own capital charges and exposure limits. With global financial markets regularly going through cycles of boom and bust, and given limited access to emerging markets and marginal borrowers, the World Bank saw PBGs as an instrument to enhance sovereign access to commercial financing, particularly during periods of market downturns. PBGs were expected to combine the financial role of credit enhancement with the developmental function of supporting reforms. They targeted a government’s overall performance in macroeconomic management and structural reforms, rather than specific projects or specific risks, as was the case under partial credit guarantees (PCGs) and partial risk guarantees (PRGs). The World Bank’s early experience with PBGs was colored by Argentina’s default on its guarantee in 2002. Following the call on Argentina’s guarantee, the World Bank’s appetite for providing PBGs dried up for almost a decade. However, the more recent experience with Western Balkan countries during 2011–14 and with Albania and Ghana in 2015, appears to have been more positive. With World Bank support through PBGs, these countries were able to meet their financing needs during adverse market conditions in the aftermath of the global financial crisis and during severe financial distress in Southern Europe. In 2013, operations policy governing PBGs was reformed, bringing such operations under OP 8.60, which covers all development policy financing (DPF).1 Today, the 1The report uses “DPF” or “development policy financing” when referring to the instrument. “DPL” or “development policy lending” was the previous name of the instrument before incorporation of guarantees under OP 8.60. The report uses “DPOs” or “development policy operations” when it refers to specific DPF operations. 1 CHAPTER 1 INTRODUCTION interest in PBGs is on the rise. Member countries are particularly attracted to the scale of financing and the market access under financial stress that PBGs have made possible. Recent Independent Evaluation Group (IEG) evaluations have generated a body of evidence and findings concerning completed PBGs. Four project performance assessment reports (PPARs) covering small Balkan economies (Serbia, the former Yugoslav Republic of Macedonia, and Montenegro) were completed in 2016.2 Evaluations of Poverty Reduction Support Credits in Ghana provided background material on the 2015 Ghana PBG. A recent review of the Albania PBG conducted by the World Bank is also relevant. The present note also draws upon IEG’s 2009 thematic evaluation reviewing all guarantees provided by the World Bank from 1990 to 2007. This learning product assembles the information and findings from these sources. As such, the report represents a step toward building a greater awareness of this important tool of development finance. Section 2 provides an overview of the instrument and its evolution. Section 3 reviews prerequisites for such operations, and how they were applied in practice. Section 4 assesses completed PBGs on the degree of access to finance and resulting financial benefits. Section 5 reviews the extent to which the reform agenda has been applied to advance the objectives of PBGs and the scope for improvement. The final section draws preliminary lessons on the potential benefits and risks of PBGs. 2The four PBGs are: Serbia Private and Financial Sector Policy-Based Guarantee (February 2011); Former Yugoslav Republic (FYR) of Macedonia Policy-Based Guarantee( November 2011); Montenegro Financial Sector Policy-Based Guarantee ( May 2012); and FYR Macedonian Public Expenditure Policy-Based Guarantee( January 2013) 2 2. The Evolution of PBGs When they were launched in 1999, PBGs specifically targeted International Bank for Reconstruction and Development (IBRD)–eligible countries with a strong track record of performance, a satisfactory structural and macroeconomic framework, and a coherent strategy for gaining access to international financial markets.3 The PBGs for Argentina (1999) and Colombia (2001) were designed and implemented under these eligibility conditions. In addition, the World Bank added the enhancement of an innovative “rolling reinstatable” mechanism by which IBRD would reactivate the guarantee if the borrower fully reimbursed IBRD following a default and a call by the lender on the guarantee. With a PBG from the World Bank, Argentina was able to borrow $1.5 billion in the international markets, with a series of six $250 million zero-coupon bonds maturing sequentially over the period 2000–04. Similarly, Colombia completed a notes issuance of $1 billion, structured with mortgage-style, semi-annual amortizations. The World Bank provided a rolling guarantee on the first two debt service payments at a time when conditions were unfavorable. Both PBGs were undertaken concurrently with other policy-based operations, including the Special Structural Adjustment Loan (SSAL, 1998) in the case of Argentina, and the Financial Sector Adjustment Loan (FSAL, 1999) in the case of Colombia. No special adjustments in the macroeconomic framework and policy priorities were needed to meet the prerequisite of PBGs. The PBGs simply replaced the second tranche of existing development policy operations, with no change in the World Bank’s exposure to the countries. When Argentina defaulted on its bonds in October 2002, the call on the World Bank’s guarantee marked a turning point in the use of PBGs. As a consequence, the World Bank made a payment of $200 million to bondholders as stipulated under the PBG. Subsequently, Argentina also failed to reimburse the World Bank within the stipulated 60-day deadline to allow for the guarantee to be reinstated. Financial markets reacted negatively to Argentina’s default, with a spillover to the Colombia PBG-backed bonds. The global financial crisis of 2007–09 and the Eurozone crisis in its aftermath created a new impetus for PBGs. As governments of the Eurozone experienced severe financial distress, with the market values of their debts suffering deep losses, major European banks which held the bulk of the troubled assets were suddenly threatened with insolvency. At the same time, more stringent regulatory requirements on capital and 3 3 CHAPTER 2 THE EVOLUTION OF PBGS liquidity under Basel III standards were taking effect. Investor sentiment turned sharply negative, even for debt instruments issued by reputable firms with a global franchise. The demand for securities issued by borrowers from emerging markets was low. Box 2.1. PBGs in the Context of Evolving Bank Policy Policy-Based Guarantees (PBGs) cover private lenders against the risk of debt service default by the sovereign government. Although they are structurally the same as Partial Credit Guarantees (PCGs), a project financing instrument, PBGs are offered for balance of payments and general budget support. PBGs can be used for any commercial debt instruments (loans, bonds) provided by any private institution. As in regular budget support loans, proceeds of the guaranteed debt can be used for any budgetary purposes. Although PBGs fully count against the World Bank’s capital, only 25 percent of an issued guarantee counts toward the World Bank’s country exposure limit. For example, with $50 million of credit available to a client, the World Bank could issue a PBG in the amount of $200 million, thereby scaling up World Bank assistance substantially without changing the World Bank’s exposure to the country. This provision creates an incentive for the use of the instrument, especially for clients with large financing needs. When they were launched in 1999, PBGs specifically targeted International Bank for Reconstruction and Development (IBRD)–eligible countries with a strong track record of performance. Specific eligibility criteria included: criterion A—The country should have a strong track record of performance and its structural, social and macroeconomic policy package should be satisfactory; criterion B—The country should have a sustainable external financing plan; and criterion C—The country should have a coherent borrowing strategy, which will enable it to become a borrower in its own name without a guarantee in the medium term. These criteria were applied in addition to development policy financing (DPF) criteria on the macroeconomic framework of DPF per OP 8.60. The operational guidance on PBGs was changed in 2013, fully incorporating PBGs under OP 8.60 which governs the World Bank’s DPF. This resulted in a number of changes, including elimination of special eligibility criteria governing PBGs and making the instrument available for International Develop[ment Association countries. The treatment of the macroeconomic framework under PBGs is the same as for DPF loans except in one detail: PBGs can be extended to those IDA-only countries that have low or modest risk of debt distress. In 2015, with a waiver from the Board, the World Bank issued a guarantee for Ghana, a country with high risk of debt distress. Source: IEG, World Bank http://siteresources.worldbank.org/INTGUARANTEES/Resources/IBRDPBG.pdf. In this environment, many borrowers, including the governments of Eastern Europe and Africa, faced an extraordinary hardship. Unable to raise funds at home or abroad, many countries turned to the World Bank and the International Monetary Fund (IMF) for assistance. This gave rise to a new burst of activities involving the use of PBGs, including four operations in the Western Balkans (one each in Serbia and Montenegro and two in FYR Macedonia) between 2011 and 2013. Another batch of three followed in 4 CHAPTER 2 THE EVOLUTION OF PBGS 2015, with PBGs extended to Albania, Angola, and Ghana. In 2016, the World Bank approved a PBG to Pakistan. PBGs to Angola and Pakistan were operational at the time of the completion of this report. Interviews with stakeholders conducted by IEG indicate growing interest in the instrument among operational staff. Thus a new generation of PBGs may be on the horizon. The overall footprint of PBGs in the World Bank’s loan portfolio has been small, despite successive waves of financial turmoil and the growing scale of financing needs. With a cumulative guarantee amount of less than $4 billion, as of July 2016 (Table 2.1), financing through PBGs pales in comparison with private capital flows to these countries and with the World Bank’s own operations. Table 2.1. World Bank Policy-Based Guarantees, 1975–2016 Status as Commitment Approval Operation title Country of June (US$ millions) date 2016 Competitiveness and Growth Pakistan 420 Active 21-June-16 Development Policy Financing Macroeconomic Stability for Ghana 400 Active 30-Jun-15 Competitiveness and Growth Credit First Fiscal Management Development Angola 200 Active 30-Jun-15 Policy Financing Operation Albania Public Finance PBG Albania 226.7 Active 27-Mar-15 FYR Macedonia Public Expenditure FYR Macedonia 201.5 Closed 8-Jan-13 PBG Montenegro Financial Sector PBG Montenegro 79.2 Closed 28-Jun-12 FYR Macedonia PBG FYR Macedonia 134.9 Closed 10-Nov-11 Private and Financial Sector PBG Serbia 400 Closed 10-Feb-11 Colombia PBG Colombia 220.3 Closed 8-Mar-01 Argentina PBG Argentina 250 Closed 16-Sep-99 Note: For active and closed projects, the commitment amount at Board approval is shown in U.S. dollars. These figures do not reflect any cancellations. The PBGs that have been reviewed and evaluated by IEG (Serbia, Montenegro, and two guarantees with FYR Macedonia) were all designed and appraised under the old framework. More recent PBGs (Albania, Ghana, Angola, and Pakistan) were prepared under OP 8.60. 5 CHAPTER 2 THE EVOLUTION OF PBGS Box 2.2. Outcome Ratings of Completed and Evaluated PBGs As of the end of FY16, IEG had validated and evaluated four PBGs. The outcome ratings are presented in the table below. Serbia’s PBG was found to be successful in providing needed financing through the country’s first commercial international borrowing, and supporting important sector specific reforms. A weak macroeconomic framework was the key reason behind the poor outcomes for Montenegro. The PBGs in FYR Macedonia were rated moderately satisfactory because of poor performance on public financial management (in the first PBG) and weaknesses in the macro framework (in the second). Outcome Rating Serbia Satisfactory FYR Macedonia 1 Moderately Satisfactory Montenegro Moderately Unsatisfactory FYR Macedonia 2 Moderately Satisfactory Source: IEG. 6 3. The Quality of the Macro-Fiscal Framework of PBGs Under the World Bank’s operational framework of DPF, PBGs are to be implemented in the context of a credible macro framework. OP 8.60 governing DPF also stipulates that, if macroeconomic instability remains a risk, the World Bank needs to require special macroeconomic and fiscal actions to be implemented as risk mitigation measures. Data from six operations implemented since 2011 and reviewed by IEG indicate that because of the guarantee element, PBGs tend to mobilize larger financing compared to development policy loans (DPLs). On average, these PBGs provided close to 2.3 percent of gross domestic product in external financing—substantially higher than under regular DPLs (table 3.1).4 This phenomenon is explained not only by the fact that the financing requirements of borrowing countries were high at that time, but also because the magnitude of funding mobilized by PBGs would not, in most cases, have been possible under regular DPOs because of the World Bank’s country exposure limit. Table 3.1. Financing Mobilized through World Bank Policy-Based Guarantees Total Total Financing PBG amount Financing as Operation title Country mobilized (US$ millions) percent of (US$ GDP millions) Macroeconomic Stability for Ghana 400 1,000 2.1 Competitiveness and Growth Credit Albania Public Finance PBG Albania 226.7 283.4 2 FYR Macedonia Public Expenditure FYR Macedonia 201.5 325 3.4 PBG Montenegro Financial Sector PBG Montenegro 79.2 132 3.2 FYR Macedonia PBG FYR Macedonia 134.9 175 1.7 Private and Financial Sector PBG Serbia 400 400 1 Source: IEG The four PBGs evaluated by IEG were deployed in a challenging macroeconomic context with important fiscal imperatives. Despite the requirement and policy guidance, 4The average size of preceding DPOs in Ghana, Albania, FYR Macedonia, Montenegro, and Serbia was 0.8 percent of GDP 7 CHAPTER 3 THE REQUIRED MACRO-FISCAL FRAMEWORK FOR PBGS the quality of the macro framework supporting PBGs varied. Table 3.2 summarizes key features of the macro framework for the PBGs for which IEG has evaluative evidence.5 Table 3.2. Macro-Framework for Selected PBGs Reviewed and Evaluated by IEG Country Macro-Fiscal Framework Serbia IMF program FYR Macedonia I IMF program Montenegro No IMF program. Substantial macro risks, no specific objectives and policy content in the program to address those risks. The PPAR found that macro eligibility criteria might have not been fully met at the time of issuance of the guarantee. FYR Macedonia II Effectively no IMF program (program was set to expire shortly after the approval of PBG). The PPAR highlighted that the availability of PBG might have postponed adjustments and led to larger fiscal deficits. Ghana IMF program, strong macro-program to mitigate significant risks. Guarantee supported Eurobond issuance for refinancing of short-term domestic debt. Source: IEG The PBG for Serbia was implemented in parallel to an IMF program with high confidence in the underlying macro framework. That was also the case for the first PBG in FYR Macedonia and the one in Ghana. The situation was different with the second PBG in FYR Macedonia and the Montenegro PBG. In both cases, IEG evaluations found that the focus on the macro framework was not adequate. In the case of FYR Macedonia, the PBG was implemented at a time when unauthorized capital expenditures and weakened fiscal control created substantial controversy. There were also legitimate concerns about some spending decisions that could undermine the fiscal framework, such as ad hoc increases in pensions. The PPAR found that the availability of a large loan through PBGs might have contributed to an expansion of the government’s spending program.6 Because the IMF program was scheduled to expire, it heightened the macro risks and put a special burden on the World Bank to ensure that the macro and fiscal framework remained adequate. The PPAR concluded that the second PBG in FYR Macedonia was not used effectively by the World Bank to secure a proper fiscal policy stance. 5 Evidence on Ghana PBGs is collected through an informal review of the macroeconomic framework of PBG as part of the PPAR on Ghana. 6 PPAR of PEPBG, FYR Macedonia, IEG, 2016. 8 CHAPTER 3 THE REQUIRED MACRO-FISCAL FRAMEWORK FOR PBGS In the case of Montenegro, IEG found that the fiscal projections underpinning the PBG were unrealistic, in contrast to independent projections available at the time, including those of the IMF.7 In contrast to Montenegro and FYR Macedonia II, the PBG in Ghana was strongly linked and complementary to an IMF program.8 In 2012, when Ghana was holding general elections and not implementing an adequate macro framework, the World Bank stopped the provision of budget support. However, in 2015, with the return of the IMF program, the World Bank resumed its adjustment lending, and introduced a PBG to help address the exceptionally large financing needs. The underlying macro framework was adequate and was endorsed by the IMF. The PBG supported the issuance of Eurobonds in 2015, with the proceeds earmarked for refinancing and buying back high- interest debt. No additional spending resulted from the large proceeds made possible by the PBG. On the contrary, the authorities appeared motivated to implement difficult adjustments because of the PBG.9 Based on evidence from these IEG PPAR evaluations, this report highlights that the requirement of an adequate macro framework for World Bank DPFs is especially important in the context of large policy loans supported by PBGs. Under a comprehensive and robust macro framework, PBGs, as a DPF instrument, can facilitate and serve as important leverage for policy reforms in support of fiscal sustainability and other important policy areas. In the absence of such a framework, PBGs may create risks of fiscal overspending as a result of the availability of large-size financing. Evidence gathered by IEG indicates that, in some contexts, the availability of large financing without an adequate macro framework may dilute the incentive of the borrower to uphold fiscal discipline. Large financing operations, especially those magnified by the addition of PBGs, should be avoided if the World Bank is not able to fully address the existing macro risks through policy lending and dialogue. In the absence of a parallel IMF program, there is a particular need to pay attention to such macroeconomic risks. It is thus imperative, for all policy-based lending, and even more so for such leveraged operations, that before granting a PBG, the World Bank ensures the integrity of the macro framework and supports additional corrective measures as needed. Moreover, 7 PPAR of PBG, Montenegro, IEG, 2016. 8“Macroeconomic Stability for Competitiveness and Growth Credit” Program Document, World Bank, 2015. 9Ghana’s case has been used only for assessing the underlying macro framework - this report does not examine the borrowing modality used by the Government under this PBG. 9 CHAPTER 3 THE REQUIRED MACRO-FISCAL FRAMEWORK FOR PBGS the World Bank is required to monitor the program up to the closing date, which is typically less than a year from approval date, typically covering only a fraction of a guarantee’s duration. It is essential to monitor the actions of the borrower during the entire period while the guarantee is in effect. This can perhaps be done as part of the World Bank’s country partnership strategy with relevant clients. For these reasons a question may be raised about whether the current operational framework of DPF adequately ensures a sound macro framework for PBGs. Eligibility criteria for PBGs prior to 2013 were more demanding, because they were built around assumptions about a downward public and external debt trajectory. That said, the Montenegro PBG illustrates that even those criteria alone could not guarantee a satisfactory macroeconomic framework. An IEG review in 2015 found that most DPOs are underpinned by an adequate macro framework, and OP 8.60 provides the necessary guidance on acceptable treatment of macroeconomic issues in World Bank DPF.10 With a stronger focus on these issues in PBGs, the World Bank should be able to address macro and fiscal risks and further promote important reforms in fiscal and debt institutions. Rigorous debt sustainability analysis would be key to ensuring quality macro and fiscal frameworks for PBGs under OP 8.60. These suggestions could be reviewed together with the overall re-appraisal of the World Bank Group’s guarantee instruments, launched in FY17. 10 “The Quality of DPOs’ Macroeconomic Framework” IEG Learning Product, 2015. 10 4. Did PBGs Improve Access to Private Finance and Credit Terms? Access to commercial finance with improved credit terms is the key motivation and an objective in all PBGs, which support not only a particular debt transaction, but also may pave the way for the regular presence of the borrower in international markets.11 The definition of “access to finance” has evolved since the introduction of PBGs in the late 90s. Whereas initially the focus of the instrument was to help clients lacking access to international markets, the emphasis has recently shifted to a broader definition of access to market, which includes the volume of private financing, maturity, and financial terms. This enabled provision of PBGs to countries that already have some level of market access.12 PBGs may be assessed on the criteria of the amount of commercial financing raised and improvements in financial terms that includes both maturity and interest rates. Though difficult to measure precisely, these criteria help capture the value of PBGs in terms of both attracting private creditors and securing advantageous terms, given the prevailing market conditions and investor sentiment. Table 4.1. Summary Terms of Commercial Loans Supported by PBGs Serbia FYR Macedonia I Montenegro FYR Macedonia II Albania Effective February,28, December 2011 July 2012 January 2013 March 2015 Date 2011 Amount EUR 293 million EUR 130 million EUR 100 million EUR 250 million EUR 250 million Coverage 100% of 76% of principal at 60% of principal at 62% of principal at 80% of principal at principal at maturity maturity maturity maturity maturity Maturity 6 years 5 years 7 years 7 years guaranteed 10 years guaranteed portion and 5 years portion and 5 the rest years the rest Interest Euro Swap plus 4.25 percent 350 basis points 3.9 percent 130 basis points 100 basis points spread over 12- spread over 12- month Euribor month Euribor 11Some PBGs had explicit “financial” objectives in the policy matrix (Serbia and FYR Macedonia), though these objectives were related to the terms of subsequent independent borrowing rather than the terms of the instruments supported by PBGs. “Modernizing the World Bank’s Operational Policy on Guarantees,” Operational Policy and 12 Country Services. 11 CHAPTER 4 DO PBGS IMPROVE ACCESS TOP PRIVATE FINANCE AND FINANCIAL TERMS? Source: IEG Effects on market access. In practice, PBGs helped secure financing that could otherwise be out of reach when market conditions were unfavorable. The four PBGs reviewed by IEG opened a pathway to private finance at a time of global financial distress. Shocked by the exceptional turmoil in major financial centers and crises in Southern Europe, investors were not receptive to small Eastern European emerging economies. The four PBGs in the Balkans met their objectives of helping governments raise the desired volume of private finance at terms that were acceptable. The guarantees secured total commercial bank credit of EUR 780 million with maturities of five to seven years. For Serbia, the PBG facilitated the country’s first external sovereign borrowing transaction and secured a relatively large loan of long maturity. The loan was extended by Société Générale in 2011 through competitive bidding. The principal amount was EUR 296.2 million, fully covered by the World Bank’s guarantee and priced at a 100 basis point spread above the EURIBOR. The maturity of six years was particularly attractive, given Serbia’s heavy reliance on more expensive short-term local market funding in the past. Following the PBG, Serbia went back several times to international markets independently between 2012 and 2015, raising $5 billion through bond issues with maturities of five to 20 years. In the case of FYR Macedonia, the first PBG helped the government regain access to international markets in 2011. Designed as a standalone PBG, the operation allowed the government to borrow EUR 130 million from Deutsche Bank and Citibank. In January 2013, FYR Macedonia received a second PBG from the World Bank to raise EUR 250 million from Deutsche Bank under somewhat more favorable market conditions than the first. The second PBG helped cover the government’s large fiscal gap. Between the two loans supported by PBGs, FYR Macedonia was able to independently tap the international loan markets in July 2012, although the amount involved was much lower than expected. The PBGs for FYR Macedonia, especially the second one, were more important in raising the volume of finance rather than opening access to markets. The experience of neighboring Montenegro underlines the critical role played by the World Bank guarantee in overcoming credit constraints during times of financial distress. Having suffered one of the deepest downturns in the aftermath of the recent global financial crisis, Montenegro was in a financial straitjacket. The EUR 60 million PBG helped the government borrow EUR 100 million from Credit Suisse to cover a large part of its budget deficits in 2012. Montenegro was able to contract a commercial loan two months prior to the approval of the PBG, but it also needed to provide gold as partial collateral and to accept a relatively high rate. 12 CHAPTER 4 DO PBGS IMPROVE ACCESS TOP PRIVATE FINANCE AND FINANCIAL TERMS? Effects on maturity. PBGs also helped extend the maturity of loans. The improved tenor, however, is mostly confined to the guaranteed portion of the loans and not to the non-guaranteed portion. This result is clear from the loans to FYR Macedonia and Montenegro, which were split into guaranteed and non-guaranteed portions. The guaranteed portion of the loans enjoyed an extended maturity. The non-guaranteed portion did not see an increase in maturity compared to earlier non-guaranteed loans. Effects on subsequent borrowing. The role of PBGs in improving the terms of credit on subsequent and non-guaranteed loans is more difficult to determine. In some cases, when a PBG was followed by an independent debt issuance, the terms were reasonable and better than those obtainable before or without the PBG. These cases, however, occurred when market conditions were benign, including the years 2014 and 2015. The role of PBGs in strengthening the borrower’s position is noticeable in the case of Serbia. In this case, there were substantial capacity development and knowledge transfers from the World Bank with respect to debt management. By contrast, FYR Macedonia and Montenegro already had established a presence in the international markets prior to PBGs. Effects on interest rates. The key contribution of the instrument is its effect on interest rates. To the extent that the World Bank is perceived to have a lower credit risk than the borrower, the guaranteed loans should carry a lower rate than independent loans as a result of better credit terms. The standard method used in the World Bank of showing these benefits is to compare the actual interest rates with hypothetical (counterfactual) rates that the borrower would have paid for a similar loan without a guarantee. For some PBGs, the World Bank provided such estimates of the reduction in interest rates as a result of World Bank guarantees. This is normally presented in program documents or implementation completion and results reports. Deriving good counterfactuals, however, is challenging and requires sophisticated market tests, which are not always possible. These analyses, while very useful, do not give the full picture of financial benefits. The difference between the counterfactual and the actual interest rate is not amenable to straightforward interpretation. How large does the difference need to be to make it a success? A higher percentage of the loan covered by the guarantee would lead to a larger reduction in interest rates. For example, the interest rate on an instrument that is 90 percent guaranteed by the World Bank should be lower compared to an instrument that is 50 percent guaranteed, if other factors are held constant. An elegant way of determining the PBG’s contribution to the borrower’s credit terms is to derive the implicit interest rate on the non-guaranteed portion of the PBG-supported loan. This is based on the premise that the actual borrowing rate of the PBG-supported 13 CHAPTER 4 DO PBGS IMPROVE ACCESS TOP PRIVATE FINANCE AND FINANCIAL TERMS? loan is the weighted average of the rate on the guaranteed portion and the rate on the non-guaranteed portion. The guaranteed portion should carry a rate that is the same or close to the borrowing cost of the World Bank. The rate on the non-guaranteed portion reflects the borrower’s creditworthiness. This approach was articulated in an Operations Policy and Country Services (OPCS) paper on enhancing the effectiveness of PBGs (Box 4.1). Box 4.1. An OPCS Metric of Improvements in Financial Terms - 2012 According to a note on PBGs provided by Operations Policy and Country Services to the Bank’s Board for discussion, in 2012, “Improved financial terms refers to the implied spread on the unguaranteed portion of the borrowing (which thereby adjusts for the effects of the protection that the [World] Bank guarantee provides) being equal to or lower than a comparable reference spread on nonguaranteed borrowing. Even with the broadened definition of improved market access, at a minimum the implied spread on the unguaranteed portion of the borrowing is expected to be equal to a comparable reference spread.” Source: World Bank, 2012. “Modernizing the World Bank’s Operational Policy on Guarantees,” Operational Policy and Country Services. URL: http://siteresources.worldbank.org/EXTRESLENDING/Resources/1-5-12-Guarantee-AP-post-CODE- clean.pdf. This implicit rate shows the sovereign risk premium associated with the debt instrument supported by PBGs. Financial benefits from PBG will fully materialize, or in other words, the value of PBG will be preserved if the interest rate of the non-guaranteed component is in line with the sovereign risk premium in the market. The difference between these two rates can point to the extent of “erosion” of the value of PBG as a result of structuring of the debt instrument, which can be driven by a number of factors, including possible additional premium demanded by the markets and other inefficiencies. 13 For the purpose of this learning product, IEG worked in cooperation with relevant World Bank units to identify the extent to which the implied interest rates for the five commercial loans supported by PBGs (four evaluated PBGs and the Albania PBG) were 13There is a hypothetical possibility that as a result of increased investor confidence, the implied interest rate can be lower than sovereign risks premium in the market. This “halo” effect of PBGs is possible especially for marginal borrowers that are not well known and monitored closely by large international lenders. For these borrowers, the World Bank can serve as an honest broker to alleviate information asymmetry between lenders and borrowers. This effect, however, is very difficult to detect. Possible changes in market perceptions resulting from PBGs are more likely to occur at the outset of the program, when markets first learn about World Bank support, and to be gradually factored in, rather than at the end of the process when the deal is signed. 14 CHAPTER 4 DO PBGS IMPROVE ACCESS TOP PRIVATE FINANCE AND FINANCIAL TERMS? different from relevant yields in the market, which would point to the extent of erosion of the value of the guarantee. (The appendix to this report illustrates the methodological approach). Overall, there was a variation across the five loans supported by PBGs. In one case, no erosion in the value of the PBG was detected. In two PBGs, the observed erosion was quite low, while in the remaining two operations, the observed erosion was higher.14 In terms of factors explaining why structuring of a debt instrument could result in a loss in the value of a guarantee and, subsequently, less than expected benefit for the borrower, the review highlights that in one case, a substantial improvement in market conditions took place in the period between the borrower’s commitment to the loan and actual signing, making the agreed interest rate less favorable compared to market conditions at the time of signing. There is also evidence that for bank-loan markets to yield pricing efficiency, competitive bidding, transparency, and due diligence on the part of the borrower and the World Bank are essential. It could be a useful exercise for PBG teams to work with clients on the decomposition and benchmarking of interest rates of debt instruments supported by PBGs to allow for better assessment of the extent of improvements in credit terms. World Bank self- evaluation of PBGs through implementation completion and results reports can serve as an appropriate platform for such analysis. Based on the cases reviewed, greater attention is also needed to the modality for raising private finance for maximizing financial benefits for the clients. The choice between bank loans and bond issuance involves several trade-offs that need to be considered in light of a particular borrower’s circumstances. Most recent PBG transactions have been through negotiated deals in the international bank loan markets. For such markets to ensure pricing efficiency, it is necessary for the World Bank and the borrower to ensure competitive bidding, transparency, and due diligence. The review also emphasizes that, in light of growing interest in PBGs, there is need for more comprehensive World Bank- wide corporate guidance on borrowing modalities and on the roles and responsibilities of relevant units in the World Bank. Such guidance could be reviewed in the light of the ongoing Bank Group-wide review of guarantee instruments15. Such a supportive governance framework for PBGs in the World Bank is essential for enhancing the overall effectiveness of this important instrument. 14 Specific PBGs are not mentioned due to market sensitivities. 15 “Steering Committee on Guarantees,” An internal World Bank document. 15 5. Reform Agenda under PBGs IEG reviews indicate that the reform content of PBGs is not materially different from that of non-guarantee DPOs, and reflect greater or lesser degrees of macroeconomic stringency, as do other DPOs. Furthermore, nothing in the World Bank’s operational policy suggests that the reform content of PBGs should be different from other policy- based operations. As emphasized by the World Bank, for DPOs backed by PBGs, access to private finance is only one side of the coin. The principal agenda is for the implementation of policy and institutional reforms. The reforms underlying all four PBGs in Balkan countries built on previous World Bank–supported operations and represented part of a broader reform agenda consistent with current country partnership strategies. For example, FYR Macedonia’s reform program, supported by its first PBG, was originally designed as the second operation of a series of programmatic DPLs. At the request of the government in early 2011, the World Bank converted DPL2 into a standalone PBG, retaining largely the same policy matrix as was envisaged for DPL1. Box 5.1. Summary of Reform Contents under Selected PBGs Envisaged reforms in Serbia under its PBG were associated with: (i) improving the business environment for the private sector, and (ii) making the financial sector more stable and efficient. The key policy objectives under the FYR Macedonia first PBG were to: (i) strengthen sustainability of public finances and functioning of labor markets; (ii) strengthen social safety nets, and (iii) improve resilience of financial sector. In many respects, FYR Macedonia’s second PBG—the Public Expenditure Policy-Based Guarantee—was, for all intents and purposes, a crisis-related assistance aimed at mitigating the impact of the renewed Euro zone turmoil and difficult financial market access. The focus of policy measures under the PBG was on improving the efficiency of public expenditures and public financial management along with strengthening health systems and improving the targeting of social assistance and social inclusion. Finally, in the case of Montenegro, FSPBG was designed to support the authorities’ efforts to strengthen the banking system and increase its resilience to possible future shocks by continuing to undertake sectoral policy reforms and system restructuring. Source: IEG According to OP8.60, the size of financial assistance in the World Bank’s policy-based operations is irrelevant for the determination of an appropriate reform agenda. The 16 CHAPTER 5 REFORM AGENDA UNDER PBGS choice of lending instruments, as well as the size of assistance, is dictated by macroeconomic considerations and by the World Bank’s risk management requirements. And the reform agenda depends on a different set of variables, including government priorities, political support, and capacity. Some stakeholders, however, find that the reform agenda for PBGs should be more demanding because they tend to give proportionally larger financial support due to increased leverage. The increase should also be used as leverage for promoting more difficult actions. Adding a PBG to a DPL offers greater financial support but may merit attaching deeper reforms. Otherwise, they may represent an opportunity lost, according to this perspective. In practice, as IEG’s reviews of PPARs and some other available information illustrate, World Bank operations have involved both of these approaches. There are cases of “performance-based” determination of the amount of financing.16 By the same token, an increase in funding could be provided for strong performance and perhaps for agreement to a stronger reform program. IEG finds that the World Bank may have missed an opportunity in some PBGs by not negotiating a deeper reform agenda when the size of financial support was significantly increased. In the case of Montenegro, for example, IEG finds that by the time the PBG was being prepared there was evidence that resolution of nonperforming loans was moving much slower than expected, and that could have been addressed through the PBG. In the second FYR Macedonia PBG, IEG finds that at the time of preparation urgent policy issues had emerged with respect to public financial management. The conversion of an existing DPL into PBG with much larger funding gave the World Bank an opportunity to address those issues more comprehensively, which, according to the PPARs, was not fully used. The policy content of PBGs should be seen as an important tool to strengthen market perceptions of debt sustainability and facilitate improvement of financing terms for clients. The demand for PBGs has been strongest when governments face severe fiscal needs, sometimes without the support of the IMF. Strong measures, including actions that the IMF may have called for, are needed to mitigate the underlying risks. With PBGs, the World Bank hopes to bring about attractive financial terms from private commercial lenders. The underlying reform program needs to be credible and capable of standing up to scrutiny by independent parties. And last, there is an element of elevated risk of financial loss to the World Bank that needs to be mitigated. Therefore, For example, downsizing of support under Tanzania’s PRCSs due to the failure to comply 16 with important triggers. 17 CHAPTER 5 REFORM AGENDA UNDER PBGS the reform agenda should be geared toward ensuring sound macro and fiscal frameworks through incorporation of fiscal reforms and other structural reforms with important implications for fiscal sustainability in the policy matrix. 18 6. Lessons This learning product offers the following lessons. PBGs represent an important instrument of development policy financing that can be used effectively to help members overcome difficult financing and reform challenges. The PBG approach brings a potentially powerful combination of client benefits, private sector participation, and financial leveraging of World Bank resources. Through a PBG, World Bank capital is tied up for a much shorter period than for a standard policy loan. The fact that only 25 percent of an issued guarantee counts toward the World Bank’s country exposure limit creates an incentive for the use of the instrument, especially for clients with large financing needs. The benefits to World Bank clients include improved market access, potential diversification of the creditor base, longer maturity, and lower interest rates. IEG’s review of PBGs found that, with World Bank support, borrowers were able to meet their financing needs during difficult market conditions. A robust macroeconomic and fiscal policy framework is essential for sustaining benefits from improved access to private finance for deficit financing. As a DPF instrument, PBGs can facilitate a range of important policy reforms in client countries. Some of the PBGs evaluated by IEG, however, would have benefited from a stronger focus on mitigating fiscal and macroeconomic risks. Evidence suggests that, in some cases, relatively large loans mobilized through PBGs might have created incentives for suboptimal fiscal choices, particularly in the absence of an IMF program. The report highlights the importance of rigorous application of the World Bank’s operational policy on the macroeconomic framework of DPF with a special focus on mitigating sovereign default risks through debt sustainability analysis. Incorporation of necessary mitigating actions in the policy matrix is especially important for PBGs. Close collaboration with IMF is likely to be valuable. The World Bank’s engagement through large DPOs backed by PBGs will also benefit from extended macroeconomic monitoring that goes beyond a program’s closing date. This can be managed in the context of the World Bank’s country partnership framework. The impact of PBGs on borrowers’ credit terms varied from one program to another. In all of the PBGs reviewed by IEG, the aggregate interest rates were lower than they would have been without guarantees. However, this fact alone may not be sufficient for assessing the effectiveness of PBGs in improving borrowers’ credit terms. Estimating the implied interest rate on the non-guaranteed portion of PBG-supported loans and the extent of possible “erosion” in the value of a guarantee is important for more comprehensive assessment of PBGs. In this respect, the review finds that the financial benefits from improved credit terms varied across the five PBGs that supported 19 CHAPTER 6 LESSONS FOR THE FUTURE commercial loans. The review therefore highlights the importance of maximizing the financial benefits from PBGs. Assessing the full extent of financial benefits using this metric in World Bank self-evaluation could further enhance understanding of the overall effectiveness of PBGs. Based on the cases reviewed, greater attention is needed to the modality for raising private finance. Whether or not to add a PBG to a DPL, as well as the choice between bank loans and bond issuance, involves several trade-offs that need to be considered in light of a particular borrower’s circumstances. Most recent PBG transactions have been through negotiated deals in the international bank loan markets. For such markets to ensure pricing efficiency, it is necessary for the World Bank and the borrower to ensure competitive bidding, transparency, and due diligence. The review also emphasizes that, in light of growing interest in PBGs, there is need for more comprehensive World Bank- wide corporate guidance on borrowing modalities and on the roles and responsibilities of relevant units in the World Bank. Such guidance could be reviewed in the light of the ongoing Bank Group-wide review of guarantee instruments. Such a supportive governance framework for PBGs in the World Bank is essential for enhancing the overall effectiveness of this important instrument. 20 Appendix A: Assessing the Implied Interest Rate of Non-guaranteed Instruments and the Value of Guarantees: A Methodological Approach Commercial loans or bonds partly guaranteed by the World Bank can be seen as a “bundle” combining two distinct credit risks. “Unbundling” of the instrument and deriving the implied interest rate on the non-guaranteed portion of the instrument would therefore allow to observe the sovereign risk premium associated with the particular transaction. It would also allow to assess the value of guarantee or the extent of losses from structuring of the instrument by comparing the implied rate with comparable rates on sovereign’s other borrowing instruments. For the purpose of this learning product IEG has developed a framework to derive upper and lower bounds of the implied interest rates associated with PBG-supported transactions. In addition, some calculations of the “value” of guarantee for commercial loans were available at the Bank based on the approach outlined in a guide for potential investors titled “Pricing partially guaranteed bonds” (Financial Solution occasional Paper 001/16, February 201617). Although somewhat different in their approaches, these two sets of estimates pointed generally to the same direction in terms of the extent of financial benefits from guarantees in individual operations. It should be noted that IEG’s approach does not depend on assumptions about sovereign yields, but comparing the results with other instruments does depend on availability of comparator instruments. Bank’s approach in deriving the actual value of guarantee is based on assumptions about sovereign risks premium. As sovereign yields for many of Bank’s clients cannot be easily observed both approaches have limitations and should be interpreted with a caution. In IEG approach, it is first assumed that the implicit interest rate for the guaranteed portion of the loan should be equal to the yields on IBRD bonds, and that there is no erosion in the value of guarantee as a result of structuring of debt instrument (markets do not require additional premium). The implied interest rate of non-guaranteed portion can be derived using the formula below. 17http://documents.worldbank.org/curated/en/723281467998238063/pdf/103282-v1-WP- Occasional-Paper-001-Partially-guaranteed-bond-valuation-final-Box394864B-PUBLIC-Volume- 1.pdf 21 APPENDIX A: ASSESSING THE IMPLIED INTEREST RATE OF NON-GUARANTEED INSTRUMENT: AN IEG APPROACH ING = (AI – WG * IIBRD)/WNG Where ING is the implied interest rate on nonguaranteed portion of the loan; AI is the aggregate interest rate on the bundle; WG and WNG are the weights of guaranteed and nonguaranteed components of the loan based on principle amount respectively derived from present values of exposure; and IIBRD – is the IBRD bond rate in euros of similar maturity. The formula captures cases when the maturity of guaranteed and nonguaranteed components is different, as in the case of the loans reviewed by IEG18. The formula above will give the lower bound of the implied interest rate of non- guaranteed portion of the instrument as markets may seek additional premium, for a range of reasons, on guaranteed component of the instrument in addition to the risk premium of the reputable guarantor. Due to the fact that the loan to Serbia supported by PBG was fully guaranteed (based on principle amount), it is the only case where the difference between IBRD bond rates and spreads on the guaranteed instrument can be observed. The difference, however, does not appear material: the interest rate on guaranteed instrument (Euro Swap rate plus 100 basis points) was largely in line with the IBRD bond rate of comparable maturity. Based on this and available other information on the extent of possible market fluctuations, IEG uses a 0.5 percentage point (50 basis points) difference to derive an upper bound for the risk premium for non-guaranteed component of the instrument with the equation below. ING = (AI – WG * (IIBRD + P)/WNG Where P is an estimate of empirically observed maximum difference between the guaranteed component of IBRD guaranteed instrument and comparable IBRD bond rates. With these two equations a range for the interest rate of non-guaranteed components of the instruments guaranteed by the World Bank can be estimated. Comparing this rates with the interest rates of comparator debt instrument would point to the extent of efficiencies in structuring of Bank-supported instruments. Identical or similar rates would indicate that there was no loss in the value of guarantee and, accordingly, financial benefits were larger. If the implied rate on the non-guaranteed portion of the loan is higher than it would indicate on possible erosion in the value of guarantee and smaller financial benefits from PBGs. 18An alternative approach could have been to derive the weights based on cash flows, but for this exercise the weights based on principle amount were found as more appropriate. 22 References Bank for International Settlements. 2003. Committee on the Global Financial System, Credit Risk Transfer, January, 2003. Bellas, D., M.G., Papaioannou, and I., Petrova. 2010. “Determinants of Emerging Market Sovereign Bond Spreads: Fundamentals vs Financial Stress”, IMF Working Paper. Comelli, F., 2012, Emerging Market Sovereign Band Spreads: Estimation and Back-testing”, IMF Working Paper, December. Dailami, M. 2010. “Sovereign Debt Distress and Corporate Spillover Impacts,” in Sovereign Debt and the Financial Crisis, Carlos A. Primo Braga and Gallina A. Vincelette, ed., World Bank, 2010 _____. 2007. “Credit Spread Determinants and Interlocking Contracts: A Clinical Study of th e Ras Gas Project” (with Robert Hauswald), Journal of Financial Economics, October 2007 Datta, S., M. Iskandar-Datta, and A. Patel. 1999. “Bank Monitoring and the Pricing of Corporate Public Debt”, Journal of Financial Economics, 51. Diamond, D. 1984. “Financial Intermediation and Delegated Monitoring”, Review of Economic Studies, 51. Drazen, A. 2000 . Political Economy in Macroeconomics, Princeton University Press. Duffie, D. 1999. Credit Swap Valuation”, Financial Analysts Journal, January -February. Fama, E., 1985, “What is Different about Banks?” Journal of Monetary Economics 15. Fung. H.G. Min-Ming Wen, and G. Zhang. 2012, How Does the Use of Credit Default Swaps Affect Firm Risk and Value? Evidence from US Life and Property/Casualty Insurance Companies” , Financial Management, Winter 2012. Gale D., and M. Hellwig. 1985. “Incentive Compatible Debt Contracts: The One-Period Problem”, Review of Economic Studies, 52. Haas, P. 1992. “Epistemic Communities and International Policy Coordination: Introduction”, International Organization, 46(1) International Monetary Fund, Independent Evaluation Office. 2003. The Role of the IMF in Argentina, 1991-2002 Issues Paper. Kiff, J., F.L. Michaud, and J. Mitchell. 2003. “An Analytical Review of Credit –Risk-Transfer Instruments”, Banque de France Financial Stability Review, June Townsend R.M, 1979. “Optimal Contracts and Competitive Markets with Costly State Verification”, Journal of Economic Theory, 21. World Bank. 2012. “Modernizing the World Bank’s Operational Policy on Guarantees,” Operational Policy and Country Services. _____. 2009. Independent Evaluation Group, The World Bank Group Guarantee Instruments 1992-2007: An Independent Evaluation, Washington DC, 2009. _____. 2006. Global Development Finance: The Development Potential of Surging Capital Flows. 23